diff options
216 files changed, 6134 insertions, 4810 deletions
diff --git a/Documentation/devicetree/bindings/display/msm/dpu-msm8998.yaml b/Documentation/devicetree/bindings/display/msm/dpu-msm8998.yaml index 167bc48748d7..2df64afb76e6 100644 --- a/Documentation/devicetree/bindings/display/msm/dpu-msm8998.yaml +++ b/Documentation/devicetree/bindings/display/msm/dpu-msm8998.yaml @@ -155,7 +155,7 @@ examples: #include <dt-bindings/interrupt-controller/arm-gic.h> #include <dt-bindings/power/qcom-rpmpd.h> - display-subsystem@c900000 { + mdss: display-subsystem@c900000 { compatible = "qcom,msm8998-mdss"; reg = <0x0c900000 0x1000>; reg-names = "mdss"; @@ -192,7 +192,7 @@ examples: clock-names = "iface", "bus", "mnoc", "core", "vsync"; interrupt-parent = <&mdss>; - interrupts = <0 IRQ_TYPE_LEVEL_HIGH>; + interrupts = <0>; operating-points-v2 = <&mdp_opp_table>; power-domains = <&rpmpd MSM8998_VDDMX>; diff --git a/Documentation/devicetree/bindings/display/msm/dpu-qcm2290.yaml b/Documentation/devicetree/bindings/display/msm/dpu-qcm2290.yaml index 8766b13f0c46..d31483a78eab 100644 --- a/Documentation/devicetree/bindings/display/msm/dpu-qcm2290.yaml +++ b/Documentation/devicetree/bindings/display/msm/dpu-qcm2290.yaml @@ -197,7 +197,7 @@ examples: power-domains = <&rpmpd QCM2290_VDDCX>; interrupt-parent = <&mdss>; - interrupts = <0 IRQ_TYPE_NONE>; + interrupts = <0>; ports { #address-cells = <1>; @@ -209,6 +209,7 @@ examples: remote-endpoint = <&dsi0_in>; }; }; - }; + }; }; + }; ... diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile index 5dfe08cb045e..40e2c6e2df79 100644 --- a/drivers/gpu/drm/amd/amdgpu/Makefile +++ b/drivers/gpu/drm/amd/amdgpu/Makefile @@ -53,11 +53,11 @@ amdgpu-y += amdgpu_device.o amdgpu_kms.o \ amdgpu_ucode.o amdgpu_bo_list.o amdgpu_ctx.o amdgpu_sync.o \ amdgpu_gtt_mgr.o amdgpu_preempt_mgr.o amdgpu_vram_mgr.o amdgpu_virt.o \ amdgpu_atomfirmware.o amdgpu_vf_error.o amdgpu_sched.o \ - amdgpu_debugfs.o amdgpu_ids.o amdgpu_gmc.o amdgpu_mmhub.o \ + amdgpu_debugfs.o amdgpu_ids.o amdgpu_gmc.o \ amdgpu_xgmi.o amdgpu_csa.o amdgpu_ras.o amdgpu_vm_cpu.o \ amdgpu_vm_sdma.o amdgpu_discovery.o amdgpu_ras_eeprom.o amdgpu_nbio.o \ amdgpu_umc.o smu_v11_0_i2c.o amdgpu_fru_eeprom.o amdgpu_rap.o \ - amdgpu_fw_attestation.o amdgpu_securedisplay.o amdgpu_hdp.o \ + amdgpu_fw_attestation.o amdgpu_securedisplay.o \ amdgpu_eeprom.o amdgpu_mca.o amdgpu-$(CONFIG_PROC_FS) += amdgpu_fdinfo.o diff --git a/drivers/gpu/drm/amd/amdgpu/aldebaran.c b/drivers/gpu/drm/amd/amdgpu/aldebaran.c index a545df4efce1..c6cc493a5486 100644 --- a/drivers/gpu/drm/amd/amdgpu/aldebaran.c +++ b/drivers/gpu/drm/amd/amdgpu/aldebaran.c @@ -31,6 +31,17 @@ #include "amdgpu_psp.h" #include "amdgpu_xgmi.h" +static bool aldebaran_is_mode2_default(struct amdgpu_reset_control *reset_ctl) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)reset_ctl->handle; + + if ((adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 2) && + adev->gmc.xgmi.connected_to_cpu)) + return true; + + return false; +} + static struct amdgpu_reset_handler * aldebaran_get_reset_handler(struct amdgpu_reset_control *reset_ctl, struct amdgpu_reset_context *reset_context) @@ -48,7 +59,7 @@ aldebaran_get_reset_handler(struct amdgpu_reset_control *reset_ctl, } } - if (adev->gmc.xgmi.connected_to_cpu) { + if (aldebaran_is_mode2_default(reset_ctl)) { list_for_each_entry(handler, &reset_ctl->reset_handlers, handler_list) { if (handler->reset_method == AMD_RESET_METHOD_MODE2) { @@ -136,18 +147,31 @@ static int aldebaran_mode2_perform_reset(struct amdgpu_reset_control *reset_ctl, struct amdgpu_reset_context *reset_context) { - struct amdgpu_device *tmp_adev = NULL; struct amdgpu_device *adev = (struct amdgpu_device *)reset_ctl->handle; + struct amdgpu_device *tmp_adev = NULL; + struct list_head reset_device_list; int r = 0; dev_dbg(adev->dev, "aldebaran perform hw reset\n"); - if (reset_context->hive == NULL) { + if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 2) && + reset_context->hive == NULL) { /* Wrong context, return error */ return -EINVAL; } - list_for_each_entry(tmp_adev, &reset_context->hive->device_list, - gmc.xgmi.head) { + INIT_LIST_HEAD(&reset_device_list); + if (reset_context->hive) { + list_for_each_entry (tmp_adev, + &reset_context->hive->device_list, + gmc.xgmi.head) + list_add_tail(&tmp_adev->reset_list, + &reset_device_list); + } else { + list_add_tail(&reset_context->reset_req_dev->reset_list, + &reset_device_list); + } + + list_for_each_entry (tmp_adev, &reset_device_list, reset_list) { mutex_lock(&tmp_adev->reset_cntl->reset_lock); tmp_adev->reset_cntl->active_reset = AMD_RESET_METHOD_MODE2; } @@ -155,8 +179,7 @@ aldebaran_mode2_perform_reset(struct amdgpu_reset_control *reset_ctl, * Mode2 reset doesn't need any sync between nodes in XGMI hive, instead launch * them together so that they can be completed asynchronously on multiple nodes */ - list_for_each_entry(tmp_adev, &reset_context->hive->device_list, - gmc.xgmi.head) { + list_for_each_entry (tmp_adev, &reset_device_list, reset_list) { /* For XGMI run all resets in parallel to speed up the process */ if (tmp_adev->gmc.xgmi.num_physical_nodes > 1) { if (!queue_work(system_unbound_wq, @@ -174,9 +197,7 @@ aldebaran_mode2_perform_reset(struct amdgpu_reset_control *reset_ctl, /* For XGMI wait for all resets to complete before proceed */ if (!r) { - list_for_each_entry(tmp_adev, - &reset_context->hive->device_list, - gmc.xgmi.head) { + list_for_each_entry (tmp_adev, &reset_device_list, reset_list) { if (tmp_adev->gmc.xgmi.num_physical_nodes > 1) { flush_work(&tmp_adev->reset_cntl->reset_work); r = tmp_adev->asic_reset_res; @@ -186,8 +207,7 @@ aldebaran_mode2_perform_reset(struct amdgpu_reset_control *reset_ctl, } } - list_for_each_entry(tmp_adev, &reset_context->hive->device_list, - gmc.xgmi.head) { + list_for_each_entry (tmp_adev, &reset_device_list, reset_list) { mutex_unlock(&tmp_adev->reset_cntl->reset_lock); tmp_adev->reset_cntl->active_reset = AMD_RESET_METHOD_NONE; } @@ -319,16 +339,30 @@ static int aldebaran_mode2_restore_hwcontext(struct amdgpu_reset_control *reset_ctl, struct amdgpu_reset_context *reset_context) { - int r; struct amdgpu_device *tmp_adev = NULL; + struct list_head reset_device_list; + int r; - if (reset_context->hive == NULL) { + if (reset_context->reset_req_dev->ip_versions[MP1_HWIP][0] == + IP_VERSION(13, 0, 2) && + reset_context->hive == NULL) { /* Wrong context, return error */ return -EINVAL; } - list_for_each_entry(tmp_adev, &reset_context->hive->device_list, - gmc.xgmi.head) { + INIT_LIST_HEAD(&reset_device_list); + if (reset_context->hive) { + list_for_each_entry (tmp_adev, + &reset_context->hive->device_list, + gmc.xgmi.head) + list_add_tail(&tmp_adev->reset_list, + &reset_device_list); + } else { + list_add_tail(&reset_context->reset_req_dev->reset_list, + &reset_device_list); + } + + list_for_each_entry (tmp_adev, &reset_device_list, reset_list) { dev_info(tmp_adev->dev, "GPU reset succeeded, trying to resume\n"); r = aldebaran_mode2_restore_ip(tmp_adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 37ff8cf5bbed..cdf0818088b3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -60,7 +60,6 @@ #include <drm/amdgpu_drm.h> #include <drm/drm_gem.h> #include <drm/drm_ioctl.h> -#include <drm/gpu_scheduler.h> #include <kgd_kfd_interface.h> #include "dm_pp_interface.h" @@ -233,6 +232,9 @@ extern int amdgpu_cik_support; #endif extern int amdgpu_num_kcq; +#define AMDGPU_VCNFW_LOG_SIZE (32 * 1024) +extern int amdgpu_vcnfw_log; + #define AMDGPU_VM_MAX_NUM_CTX 4096 #define AMDGPU_SG_THRESHOLD (256*1024*1024) #define AMDGPU_DEFAULT_GTT_SIZE_MB 3072ULL /* 3GB by default */ @@ -274,9 +276,6 @@ extern int amdgpu_num_kcq; #define AMDGPU_SMARTSHIFT_MIN_BIAS (-100) struct amdgpu_device; -struct amdgpu_ib; -struct amdgpu_cs_parser; -struct amdgpu_job; struct amdgpu_irq_src; struct amdgpu_fpriv; struct amdgpu_bo_va_mapping; @@ -465,20 +464,6 @@ struct amdgpu_flip_work { /* - * CP & rings. - */ - -struct amdgpu_ib { - struct amdgpu_sa_bo *sa_bo; - uint32_t length_dw; - uint64_t gpu_addr; - uint32_t *ptr; - uint32_t flags; -}; - -extern const struct drm_sched_backend_ops amdgpu_sched_ops; - -/* * file private structure */ @@ -493,79 +478,6 @@ struct amdgpu_fpriv { int amdgpu_file_to_fpriv(struct file *filp, struct amdgpu_fpriv **fpriv); -int amdgpu_ib_get(struct amdgpu_device *adev, struct amdgpu_vm *vm, - unsigned size, - enum amdgpu_ib_pool_type pool, - struct amdgpu_ib *ib); -void amdgpu_ib_free(struct amdgpu_device *adev, struct amdgpu_ib *ib, - struct dma_fence *f); -int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, - struct amdgpu_ib *ibs, struct amdgpu_job *job, - struct dma_fence **f); -int amdgpu_ib_pool_init(struct amdgpu_device *adev); -void amdgpu_ib_pool_fini(struct amdgpu_device *adev); -int amdgpu_ib_ring_tests(struct amdgpu_device *adev); - -/* - * CS. - */ -struct amdgpu_cs_chunk { - uint32_t chunk_id; - uint32_t length_dw; - void *kdata; -}; - -struct amdgpu_cs_post_dep { - struct drm_syncobj *syncobj; - struct dma_fence_chain *chain; - u64 point; -}; - -struct amdgpu_cs_parser { - struct amdgpu_device *adev; - struct drm_file *filp; - struct amdgpu_ctx *ctx; - - /* chunks */ - unsigned nchunks; - struct amdgpu_cs_chunk *chunks; - - /* scheduler job object */ - struct amdgpu_job *job; - struct drm_sched_entity *entity; - - /* buffer objects */ - struct ww_acquire_ctx ticket; - struct amdgpu_bo_list *bo_list; - struct amdgpu_mn *mn; - struct amdgpu_bo_list_entry vm_pd; - struct list_head validated; - struct dma_fence *fence; - uint64_t bytes_moved_threshold; - uint64_t bytes_moved_vis_threshold; - uint64_t bytes_moved; - uint64_t bytes_moved_vis; - - /* user fence */ - struct amdgpu_bo_list_entry uf_entry; - - unsigned num_post_deps; - struct amdgpu_cs_post_dep *post_deps; -}; - -static inline u32 amdgpu_get_ib_value(struct amdgpu_cs_parser *p, - uint32_t ib_idx, int idx) -{ - return p->job->ibs[ib_idx].ptr[idx]; -} - -static inline void amdgpu_set_ib_value(struct amdgpu_cs_parser *p, - uint32_t ib_idx, int idx, - uint32_t value) -{ - p->job->ibs[ib_idx].ptr[idx] = value; -} - /* * Writeback */ @@ -1436,10 +1348,6 @@ static inline bool amdgpu_acpi_is_s0ix_active(struct amdgpu_device *adev) { retu static inline bool amdgpu_acpi_is_s3_active(struct amdgpu_device *adev) { return false; } #endif -int amdgpu_cs_find_mapping(struct amdgpu_cs_parser *parser, - uint64_t addr, struct amdgpu_bo **bo, - struct amdgpu_bo_va_mapping **mapping); - #if defined(CONFIG_DRM_AMD_DC) int amdgpu_dm_display_resume(struct amdgpu_device *adev ); #else diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c index e9c80ce13f3e..ba21ec6b35e0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c @@ -26,6 +26,8 @@ #include "gc/gc_10_3_0_sh_mask.h" #include "oss/osssys_5_0_0_offset.h" #include "oss/osssys_5_0_0_sh_mask.h" +#include "athub/athub_2_1_0_offset.h" +#include "athub/athub_2_1_0_sh_mask.h" #include "soc15_common.h" #include "v10_structs.h" #include "nv.h" @@ -606,6 +608,18 @@ static int wave_control_execute_v10_3(struct amdgpu_device *adev, return 0; } +static bool get_atc_vmid_pasid_mapping_info_v10_3(struct amdgpu_device *adev, + uint8_t vmid, uint16_t *p_pasid) +{ + uint32_t value; + + value = RREG32(SOC15_REG_OFFSET(ATHUB, 0, mmATC_VMID0_PASID_MAPPING) + + vmid); + *p_pasid = value & ATC_VMID0_PASID_MAPPING__PASID_MASK; + + return !!(value & ATC_VMID0_PASID_MAPPING__VALID_MASK); +} + static void set_vm_context_page_table_base_v10_3(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { @@ -788,7 +802,7 @@ const struct kfd2kgd_calls gfx_v10_3_kfd2kgd = { .hqd_destroy = hqd_destroy_v10_3, .hqd_sdma_destroy = hqd_sdma_destroy_v10_3, .wave_control_execute = wave_control_execute_v10_3, - .get_atc_vmid_pasid_mapping_info = NULL, + .get_atc_vmid_pasid_mapping_info = get_atc_vmid_pasid_mapping_info_v10_3, .set_vm_context_page_table_base = set_vm_context_page_table_base_v10_3, .program_trap_handler_settings = program_trap_handler_settings_v10_3, #if 0 diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c index fa20261aa928..673078faa27a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c @@ -626,7 +626,7 @@ amdgpu_connector_fixup_lcd_native_mode(struct drm_encoder *encoder, if (mode->type & DRM_MODE_TYPE_PREFERRED) { if (mode->hdisplay != native_mode->hdisplay || mode->vdisplay != native_mode->vdisplay) - memcpy(native_mode, mode, sizeof(*mode)); + drm_mode_copy(native_mode, mode); } } @@ -635,7 +635,7 @@ amdgpu_connector_fixup_lcd_native_mode(struct drm_encoder *encoder, list_for_each_entry_safe(mode, t, &connector->probed_modes, head) { if (mode->hdisplay == native_mode->hdisplay && mode->vdisplay == native_mode->vdisplay) { - *native_mode = *mode; + drm_mode_copy(native_mode, mode); drm_mode_set_crtcinfo(native_mode, CRTC_INTERLACE_HALVE_V); DRM_DEBUG_KMS("Determined LVDS native mode details from EDID\n"); break; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index e762e45b7b85..970b065e9a6b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -32,6 +32,7 @@ #include <drm/amdgpu_drm.h> #include <drm/drm_syncobj.h> +#include "amdgpu_cs.h" #include "amdgpu.h" #include "amdgpu_trace.h" #include "amdgpu_gmc.h" @@ -339,7 +340,7 @@ static void amdgpu_cs_get_threshold_for_moves(struct amdgpu_device *adev, if (free_vram >= 128 * 1024 * 1024 || free_vram >= total_vram / 8) { s64 min_us; - /* Be more aggresive on dGPUs. Try to fill a portion of free + /* Be more aggressive on dGPUs. Try to fill a portion of free * VRAM now. */ if (!(adev->flags & AMD_IS_APU)) @@ -782,12 +783,12 @@ static int amdgpu_cs_vm_handling(struct amdgpu_cs_parser *p) memcpy(ib->ptr, kptr, chunk_ib->ib_bytes); amdgpu_bo_kunmap(aobj); - r = amdgpu_ring_parse_cs(ring, p, j); + r = amdgpu_ring_parse_cs(ring, p, p->job, ib); if (r) return r; } else { ib->ptr = (uint32_t *)kptr; - r = amdgpu_ring_patch_cs_in_place(ring, p, j); + r = amdgpu_ring_patch_cs_in_place(ring, p, p->job, ib); amdgpu_bo_kunmap(aobj); if (r) return r; @@ -1279,7 +1280,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, continue; /* - * Work around dma_resv shortcommings by wrapping up the + * Work around dma_resv shortcomings by wrapping up the * submission in a dma_fence_chain and add it as exclusive * fence. */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.h new file mode 100644 index 000000000000..30ecc4917f81 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.h @@ -0,0 +1,80 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#ifndef __AMDGPU_CS_H__ +#define __AMDGPU_CS_H__ + +#include "amdgpu_job.h" +#include "amdgpu_bo_list.h" +#include "amdgpu_ring.h" + +struct amdgpu_bo_va_mapping; + +struct amdgpu_cs_chunk { + uint32_t chunk_id; + uint32_t length_dw; + void *kdata; +}; + +struct amdgpu_cs_post_dep { + struct drm_syncobj *syncobj; + struct dma_fence_chain *chain; + u64 point; +}; + +struct amdgpu_cs_parser { + struct amdgpu_device *adev; + struct drm_file *filp; + struct amdgpu_ctx *ctx; + + /* chunks */ + unsigned nchunks; + struct amdgpu_cs_chunk *chunks; + + /* scheduler job object */ + struct amdgpu_job *job; + struct drm_sched_entity *entity; + + /* buffer objects */ + struct ww_acquire_ctx ticket; + struct amdgpu_bo_list *bo_list; + struct amdgpu_mn *mn; + struct amdgpu_bo_list_entry vm_pd; + struct list_head validated; + struct dma_fence *fence; + uint64_t bytes_moved_threshold; + uint64_t bytes_moved_vis_threshold; + uint64_t bytes_moved; + uint64_t bytes_moved_vis; + + /* user fence */ + struct amdgpu_bo_list_entry uf_entry; + + unsigned num_post_deps; + struct amdgpu_cs_post_dep *post_deps; +}; + +int amdgpu_cs_find_mapping(struct amdgpu_cs_parser *parser, + uint64_t addr, struct amdgpu_bo **bo, + struct amdgpu_bo_va_mapping **mapping); + +#endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index f522b52725e4..5981c7d9bd48 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -23,6 +23,7 @@ */ #include <drm/drm_auth.h> +#include <drm/drm_drv.h> #include "amdgpu.h" #include "amdgpu_sched.h" #include "amdgpu_ras.h" @@ -204,9 +205,15 @@ static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, u32 hw_ip, if (r) goto error_free_entity; - ctx->entities[hw_ip][ring] = entity; + /* It's not an error if we fail to install the new entity */ + if (cmpxchg(&ctx->entities[hw_ip][ring], NULL, entity)) + goto cleanup_entity; + return 0; +cleanup_entity: + drm_sched_entity_fini(&entity->entity); + error_free_entity: kfree(entity); @@ -261,9 +268,6 @@ static int amdgpu_ctx_get_stable_pstate(struct amdgpu_ctx *ctx, struct amdgpu_device *adev = ctx->adev; enum amd_dpm_forced_level current_level; - if (!ctx) - return -EINVAL; - current_level = amdgpu_dpm_get_performance_level(adev); switch (current_level) { @@ -293,9 +297,6 @@ static int amdgpu_ctx_set_stable_pstate(struct amdgpu_ctx *ctx, enum amd_dpm_forced_level level; int r; - if (!ctx) - return -EINVAL; - mutex_lock(&adev->pm.stable_pstate_ctx_lock); if (adev->pm.stable_pstate_ctx && adev->pm.stable_pstate_ctx != ctx) { r = -EBUSY; @@ -339,7 +340,7 @@ static void amdgpu_ctx_fini(struct kref *ref) { struct amdgpu_ctx *ctx = container_of(ref, struct amdgpu_ctx, refcount); struct amdgpu_device *adev = ctx->adev; - unsigned i, j; + unsigned i, j, idx; if (!adev) return; @@ -350,7 +351,12 @@ static void amdgpu_ctx_fini(struct kref *ref) ctx->entities[i][j] = NULL; } } - amdgpu_ctx_set_stable_pstate(ctx, AMDGPU_CTX_STABLE_PSTATE_NONE); + + if (drm_dev_enter(&adev->ddev, &idx)) { + amdgpu_ctx_set_stable_pstate(ctx, AMDGPU_CTX_STABLE_PSTATE_NONE); + drm_dev_exit(idx); + } + kfree(ctx); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index 426b63e4f1f6..5d04d24a0d5f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c @@ -1678,7 +1678,7 @@ static ssize_t amdgpu_reset_dump_register_list_write(struct file *f, { struct amdgpu_device *adev = (struct amdgpu_device *)file_inode(f)->i_private; char reg_offset[11]; - uint32_t *tmp; + uint32_t *new, *tmp = NULL; int ret, i = 0, len = 0; do { @@ -1689,7 +1689,12 @@ static ssize_t amdgpu_reset_dump_register_list_write(struct file *f, goto error_free; } - tmp = krealloc_array(tmp, i + 1, sizeof(uint32_t), GFP_KERNEL); + new = krealloc_array(tmp, i + 1, sizeof(uint32_t), GFP_KERNEL); + if (!new) { + ret = -ENOMEM; + goto error_free; + } + tmp = new; if (sscanf(reg_offset, "%X %n", &tmp[i], &ret) != 1) { ret = -EINVAL; goto error_free; @@ -1773,6 +1778,16 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev) amdgpu_debugfs_ring_init(adev, ring); } + for ( i = 0; i < adev->vcn.num_vcn_inst; i++) { + if (!amdgpu_vcnfw_log) + break; + + if (adev->vcn.harvest_config & (1 << i)) + continue; + + amdgpu_debugfs_vcn_fwlog_init(adev, i, &adev->vcn.inst[i]); + } + amdgpu_ras_debugfs_create_all(adev); amdgpu_rap_debugfs_init(adev); amdgpu_securedisplay_debugfs_init(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index ca854626a108..3987ecb24ef4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -80,12 +80,7 @@ MODULE_FIRMWARE("amdgpu/raven_gpu_info.bin"); MODULE_FIRMWARE("amdgpu/picasso_gpu_info.bin"); MODULE_FIRMWARE("amdgpu/raven2_gpu_info.bin"); MODULE_FIRMWARE("amdgpu/arcturus_gpu_info.bin"); -MODULE_FIRMWARE("amdgpu/renoir_gpu_info.bin"); -MODULE_FIRMWARE("amdgpu/navi10_gpu_info.bin"); -MODULE_FIRMWARE("amdgpu/navi14_gpu_info.bin"); MODULE_FIRMWARE("amdgpu/navi12_gpu_info.bin"); -MODULE_FIRMWARE("amdgpu/vangogh_gpu_info.bin"); -MODULE_FIRMWARE("amdgpu/yellow_carp_gpu_info.bin"); #define AMDGPU_RESUME_MS 2000 #define AMDGPU_MAX_RETRY_LIMIT 2 @@ -1554,7 +1549,6 @@ static int amdgpu_device_check_arguments(struct amdgpu_device *adev) amdgpu_gmc_tmz_set(adev); - amdgpu_gmc_noretry_set(adev); return 0; } @@ -1992,27 +1986,9 @@ static int amdgpu_device_parse_gpu_info_fw(struct amdgpu_device *adev) case CHIP_ARCTURUS: chip_name = "arcturus"; break; - case CHIP_RENOIR: - if (adev->apu_flags & AMD_APU_IS_RENOIR) - chip_name = "renoir"; - else - chip_name = "green_sardine"; - break; - case CHIP_NAVI10: - chip_name = "navi10"; - break; - case CHIP_NAVI14: - chip_name = "navi14"; - break; case CHIP_NAVI12: chip_name = "navi12"; break; - case CHIP_VANGOGH: - chip_name = "vangogh"; - break; - case CHIP_YELLOW_CARP: - chip_name = "yellow_carp"; - break; } snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_gpu_info.bin", chip_name); @@ -2183,8 +2159,10 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev) !pci_is_thunderbolt_attached(to_pci_dev(dev->dev))) adev->flags |= AMD_IS_PX; - parent = pci_upstream_bridge(adev->pdev); - adev->has_pr3 = parent ? pci_pr3_present(parent) : false; + if (!(adev->flags & AMD_IS_APU)) { + parent = pci_upstream_bridge(adev->pdev); + adev->has_pr3 = parent ? pci_pr3_present(parent) : false; + } amdgpu_amdkfd_device_probe(adev); @@ -3688,6 +3666,15 @@ int amdgpu_device_init(struct amdgpu_device *adev, if (amdgpu_mes && adev->asic_type >= CHIP_NAVI10) adev->enable_mes = true; + /* + * Reset domain needs to be present early, before XGMI hive discovered + * (if any) and intitialized to use reset sem and in_gpu reset flag + * early on during init and before calling to RREG32. + */ + adev->reset_domain = amdgpu_reset_create_reset_domain(SINGLE_DEVICE, "amdgpu-reset-dev"); + if (!adev->reset_domain) + return -ENOMEM; + /* detect hw virtualization here */ amdgpu_detect_virtualization(adev); @@ -3697,20 +3684,12 @@ int amdgpu_device_init(struct amdgpu_device *adev, return r; } - /* - * Reset domain needs to be present early, before XGMI hive discovered - * (if any) and intitialized to use reset sem and in_gpu reset flag - * early on during init. - */ - adev->reset_domain = amdgpu_reset_create_reset_domain(SINGLE_DEVICE ,"amdgpu-reset-dev"); - if (!adev->reset_domain) - return -ENOMEM; - /* early init functions */ r = amdgpu_device_ip_early_init(adev); if (r) return r; + amdgpu_gmc_noretry_set(adev); /* Need to get xgmi info early to decide the reset behavior*/ if (adev->gmc.xgmi.supported) { r = adev->gfxhub.funcs->get_xgmi_info(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c index cae57d3b317a..fae5c1debfad 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c @@ -691,9 +691,9 @@ static int convert_tiling_flags_to_modifier(struct amdgpu_framebuffer *afb) return -EINVAL; } - if (adev->asic_type >= CHIP_SIENNA_CICHLID) + if (adev->ip_versions[GC_HWIP][0] >= IP_VERSION(10, 3, 0)) version = AMD_FMT_MOD_TILE_VER_GFX10_RBPLUS; - else if (adev->family == AMDGPU_FAMILY_NV) + else if (adev->ip_versions[GC_HWIP][0] >= IP_VERSION(10, 0, 0)) version = AMD_FMT_MOD_TILE_VER_GFX10; else version = AMD_FMT_MOD_TILE_VER_GFX9; @@ -787,7 +787,7 @@ static int convert_tiling_flags_to_modifier(struct amdgpu_framebuffer *afb) if (adev->family >= AMDGPU_FAMILY_NV) { int extra_pipe = 0; - if (adev->asic_type >= CHIP_SIENNA_CICHLID && + if ((adev->ip_versions[GC_HWIP][0] >= IP_VERSION(10, 3, 0)) && pipes == packers && pipes > 1) extra_pipe = 1; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 2ab675123ae3..bb1c025d9001 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -100,9 +100,10 @@ * - 3.43.0 - Add device hot plug/unplug support * - 3.44.0 - DCN3 supports DCC independent block settings: !64B && 128B, 64B && 128B * - 3.45.0 - Add context ioctl stable pstate interface + * * 3.46.0 - To enable hot plug amdgpu tests in libdrm */ #define KMS_DRIVER_MAJOR 3 -#define KMS_DRIVER_MINOR 45 +#define KMS_DRIVER_MINOR 46 #define KMS_DRIVER_PATCHLEVEL 0 int amdgpu_vram_limit; @@ -177,6 +178,7 @@ int amdgpu_reset_method = -1; /* auto */ int amdgpu_num_kcq = -1; int amdgpu_smartshift_bias; int amdgpu_use_xgmi_p2p = 1; +int amdgpu_vcnfw_log; static void amdgpu_drv_delayed_reset_work_handler(struct work_struct *work); @@ -855,6 +857,13 @@ MODULE_PARM_DESC(num_kcq, "number of kernel compute queue user want to setup (8 module_param_named(num_kcq, amdgpu_num_kcq, int, 0444); /** + * DOC: vcnfw_log (int) + * Enable vcnfw log output for debugging, the default is disabled. + */ +MODULE_PARM_DESC(vcnfw_log, "Enable vcnfw log(0 = disable (default value), 1 = enable)"); +module_param_named(vcnfw_log, amdgpu_vcnfw_log, int, 0444); + +/** * DOC: smu_pptable_id (int) * Used to override pptable id. id = 0 use VBIOS pptable. * id > 0 use the soft pptable with specicfied id. diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index 52912b6bcb20..8fe939976224 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -644,13 +644,6 @@ late_fini: return r; } -void amdgpu_gfx_ras_fini(struct amdgpu_device *adev) -{ - if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX) && - adev->gfx.ras_if) - amdgpu_ras_block_late_fini(adev, adev->gfx.ras_if); -} - int amdgpu_gfx_process_ras_data_cb(struct amdgpu_device *adev, void *err_data, struct amdgpu_iv_entry *entry) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h index ccca0a85b982..dcb3c7871c73 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h @@ -387,7 +387,6 @@ bool amdgpu_gfx_is_me_queue_enabled(struct amdgpu_device *adev, int me, void amdgpu_gfx_off_ctrl(struct amdgpu_device *adev, bool enable); int amdgpu_get_gfx_off_status(struct amdgpu_device *adev, uint32_t *value); int amdgpu_gfx_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block); -void amdgpu_gfx_ras_fini(struct amdgpu_device *adev); int amdgpu_gfx_process_ras_data_cb(struct amdgpu_device *adev, void *err_data, struct amdgpu_iv_entry *entry); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 78498d1d1769..ca2cfb65f976 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -454,17 +454,7 @@ int amdgpu_gmc_ras_late_init(struct amdgpu_device *adev) void amdgpu_gmc_ras_fini(struct amdgpu_device *adev) { - if (adev->umc.ras && adev->umc.ras->ras_block.ras_fini) - adev->umc.ras->ras_block.ras_fini(adev); - if (adev->mmhub.ras && adev->mmhub.ras->ras_block.ras_fini) - adev->mmhub.ras->ras_block.ras_fini(adev); - - if (adev->gmc.xgmi.ras && adev->gmc.xgmi.ras->ras_block.ras_fini) - adev->gmc.xgmi.ras->ras_block.ras_fini(adev); - - if (adev->hdp.ras && adev->hdp.ras->ras_block.ras_fini) - adev->hdp.ras->ras_block.ras_fini(adev); } /* @@ -569,11 +559,11 @@ void amdgpu_gmc_noretry_set(struct amdgpu_device *adev) { struct amdgpu_gmc *gmc = &adev->gmc; - switch (adev->asic_type) { - case CHIP_VEGA10: - case CHIP_VEGA20: - case CHIP_ARCTURUS: - case CHIP_ALDEBARAN: + switch (adev->ip_versions[GC_HWIP][0]) { + case IP_VERSION(9, 0, 1): + case IP_VERSION(9, 4, 0): + case IP_VERSION(9, 4, 1): + case IP_VERSION(9, 4, 2): /* * noretry = 0 will cause kfd page fault tests fail * for some ASICs, so set default to 1 for these ASICs. @@ -583,7 +573,6 @@ void amdgpu_gmc_noretry_set(struct amdgpu_device *adev) else gmc->noretry = amdgpu_noretry; break; - case CHIP_RAVEN: default: /* Raven currently has issues with noretry * regardless of what we decide for other @@ -634,6 +623,13 @@ void amdgpu_gmc_get_vbios_allocations(struct amdgpu_device *adev) unsigned size; /* + * Some ASICs need to reserve a region of video memory to avoid access + * from driver + */ + adev->mman.stolen_reserved_offset = 0; + adev->mman.stolen_reserved_size = 0; + + /* * TODO: * Currently there is a bug where some memory client outside * of the driver writes to first 8M of VRAM on S3 resume, @@ -643,10 +639,25 @@ void amdgpu_gmc_get_vbios_allocations(struct amdgpu_device *adev) */ switch (adev->asic_type) { case CHIP_VEGA10: + adev->mman.keep_stolen_vga_memory = true; + /* + * VEGA10 SRIOV VF needs some firmware reserved area. + */ + if (amdgpu_sriov_vf(adev)) { + adev->mman.stolen_reserved_offset = 0x100000; + adev->mman.stolen_reserved_size = 0x600000; + } + break; case CHIP_RAVEN: case CHIP_RENOIR: adev->mman.keep_stolen_vga_memory = true; break; + case CHIP_YELLOW_CARP: + if (amdgpu_discovery == 0) { + adev->mman.stolen_reserved_offset = 0x1ffb0000; + adev->mman.stolen_reserved_size = 64 * PAGE_SIZE; + } + break; default: adev->mman.keep_stolen_vga_memory = false; break; @@ -767,25 +778,6 @@ uint64_t amdgpu_gmc_vram_cpu_pa(struct amdgpu_device *adev, struct amdgpu_bo *bo return amdgpu_bo_gpu_offset(bo) - adev->gmc.vram_start + adev->gmc.aper_base; } -void amdgpu_gmc_get_reserved_allocation(struct amdgpu_device *adev) -{ - /* Some ASICs need to reserve a region of video memory to avoid access - * from driver */ - adev->mman.stolen_reserved_offset = 0; - adev->mman.stolen_reserved_size = 0; - - switch (adev->asic_type) { - case CHIP_YELLOW_CARP: - if (amdgpu_discovery == 0) { - adev->mman.stolen_reserved_offset = 0x1ffb0000; - adev->mman.stolen_reserved_size = 64 * PAGE_SIZE; - } - break; - default: - break; - } -} - int amdgpu_gmc_vram_checking(struct amdgpu_device *adev) { struct amdgpu_bo *vram_bo = NULL; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h index 93505bb0a36c..032b0313f277 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h @@ -331,7 +331,6 @@ amdgpu_gmc_set_vm_fault_masks(struct amdgpu_device *adev, int hub_type, bool enable); void amdgpu_gmc_get_vbios_allocations(struct amdgpu_device *adev); -void amdgpu_gmc_get_reserved_allocation(struct amdgpu_device *adev); void amdgpu_gmc_init_pdb0(struct amdgpu_device *adev); uint64_t amdgpu_gmc_vram_mc2pa(struct amdgpu_device *adev, uint64_t mc_addr); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.h index aabd59aa5213..9181c7bef7c6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.h @@ -44,5 +44,4 @@ struct amdgpu_hdp { }; int amdgpu_hdp_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block); -void amdgpu_hdp_ras_fini(struct amdgpu_device *adev); #endif /* __AMDGPU_HDP_H__ */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c index bc1297dcdf97..d583766ea392 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c @@ -166,8 +166,8 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, } if ((ib->flags & AMDGPU_IB_FLAGS_SECURE) && - (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE)) { - dev_err(adev->dev, "secure submissions not supported on compute rings\n"); + (!ring->funcs->secure_submission_supported)) { + dev_err(adev->dev, "secure submissions not supported on ring <%s>\n", ring->name); return -EINVAL; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c index be48487e2ca7..92a70fb57fa3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c @@ -204,7 +204,7 @@ static int amdgpu_vmid_grab_idle(struct amdgpu_vm *vm, unsigned i; int r; - if (ring->vmid_wait && !dma_fence_is_signaled(ring->vmid_wait)) + if (!dma_fence_is_signaled(ring->vmid_wait)) return amdgpu_sync_fence(sync, ring->vmid_wait); fences = kmalloc_array(id_mgr->num_ids, sizeof(void *), GFP_KERNEL); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c index d970336d2261..67f66f2f1809 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c @@ -81,14 +81,10 @@ exit: int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs, struct amdgpu_job **job, struct amdgpu_vm *vm) { - size_t size = sizeof(struct amdgpu_job); - if (num_ibs == 0) return -EINVAL; - size += sizeof(struct amdgpu_ib) * num_ibs; - - *job = kzalloc(size, GFP_KERNEL); + *job = kzalloc(struct_size(*job, ibs, num_ibs), GFP_KERNEL); if (!*job) return -ENOMEM; @@ -98,7 +94,6 @@ int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs, */ (*job)->base.sched = &adev->rings[0]->sched; (*job)->vm = vm; - (*job)->ibs = (void *)&(*job)[1]; (*job)->num_ibs = num_ibs; amdgpu_sync_create(&(*job)->sync); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h index 9e65730193b8..d599c0540b46 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h @@ -23,6 +23,10 @@ #ifndef __AMDGPU_JOB_H__ #define __AMDGPU_JOB_H__ +#include <drm/gpu_scheduler.h> +#include "amdgpu_sync.h" +#include "amdgpu_ring.h" + /* bit set means command submit involves a preamble IB */ #define AMDGPU_PREAMBLE_IB_PRESENT (1 << 0) /* bit set means preamble IB is first presented in belonging context */ @@ -45,12 +49,10 @@ struct amdgpu_job { struct amdgpu_vm *vm; struct amdgpu_sync sync; struct amdgpu_sync sched_sync; - struct amdgpu_ib *ibs; struct dma_fence hw_fence; struct dma_fence *external_hw_fence; uint32_t preamble_status; uint32_t preemption_status; - uint32_t num_ibs; bool vm_needs_flush; uint64_t vm_pd_addr; unsigned vmid; @@ -66,6 +68,9 @@ struct amdgpu_job { /* job_run_counter >= 1 means a resubmit job */ uint32_t job_run_counter; + + uint32_t num_ibs; + struct amdgpu_ib ibs[]; }; int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c index e2607d9f5cf4..51c2a82e2fa4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c @@ -70,9 +70,3 @@ void amdgpu_mca_query_ras_error_count(struct amdgpu_device *adev, amdgpu_mca_reset_error_count(adev, mc_status_addr); } - -void amdgpu_mca_ras_fini(struct amdgpu_device *adev, - struct amdgpu_mca_ras *mca_dev) -{ - amdgpu_ras_block_late_fini(adev, mca_dev->ras_if); -}
\ No newline at end of file diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h index 15e1a1efeb4f..7ce16d16e34b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h @@ -56,7 +56,4 @@ void amdgpu_mca_query_ras_error_count(struct amdgpu_device *adev, uint64_t mc_status_addr, void *ras_error_status); -void amdgpu_mca_ras_fini(struct amdgpu_device *adev, - struct amdgpu_mca_ras *mca_dev); - #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h index 240b26d9a388..9f1540f0ebf9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h @@ -47,6 +47,5 @@ struct amdgpu_mmhub { struct amdgpu_mmhub_ras *ras; }; -void amdgpu_mmhub_ras_fini(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c index f09ad80f0772..37d779b8e4a6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c @@ -43,10 +43,3 @@ late_fini: amdgpu_ras_block_late_fini(adev, ras_block); return r; } - -void amdgpu_nbio_ras_fini(struct amdgpu_device *adev) -{ - if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__PCIE_BIF) && - adev->nbio.ras_if) - amdgpu_ras_block_late_fini(adev, adev->nbio.ras_if); -} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h index f9546c7341b8..3d13e601fc35 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h @@ -105,5 +105,4 @@ struct amdgpu_nbio { }; int amdgpu_nbio_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block); -void amdgpu_nbio_ras_fini(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index 94bfe502b55e..a6acec1a6155 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -277,7 +277,7 @@ static int psp_init_sriov_microcode(struct psp_context *psp) ret = psp_init_cap_microcode(psp, "sienna_cichlid"); break; case IP_VERSION(13, 0, 2): - ret = psp_init_ta_microcode(psp, "aldebaran"); + ret = psp_init_cap_microcode(psp, "aldebaran"); break; default: BUG(); @@ -310,6 +310,10 @@ static int psp_sw_init(void *handle) return ret; } + adev->psp.xgmi_context.supports_extended_data = + !adev->gmc.xgmi.connected_to_cpu && + adev->ip_versions[MP0_HWIP][0] == IP_VERSION(13, 0, 2); + memset(&boot_cfg_entry, 0, sizeof(boot_cfg_entry)); if (psp_get_runtime_db_entry(adev, PSP_RUNTIME_ENTRY_TYPE_BOOT_CONFIG, @@ -3008,7 +3012,6 @@ static int psp_init_sos_base_fw(struct amdgpu_device *adev) adev->psp.sos.size_bytes = le32_to_cpu(sos_hdr->sos.size_bytes); adev->psp.sos.start_addr = ucode_array_start_addr + le32_to_cpu(sos_hdr->sos.offset_bytes); - adev->psp.xgmi_context.supports_extended_data = false; } else { /* Load alternate PSP SOS FW */ sos_hdr_v1_3 = (const struct psp_firmware_header_v1_3 *)adev->psp.sos_fw->data; @@ -3023,7 +3026,6 @@ static int psp_init_sos_base_fw(struct amdgpu_device *adev) adev->psp.sos.size_bytes = le32_to_cpu(sos_hdr_v1_3->sos_aux.size_bytes); adev->psp.sos.start_addr = ucode_array_start_addr + le32_to_cpu(sos_hdr_v1_3->sos_aux.offset_bytes); - adev->psp.xgmi_context.supports_extended_data = true; } if ((adev->psp.sys.size_bytes == 0) || (adev->psp.sos.size_bytes == 0)) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index b53abc15a830..424c22a841f4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -2068,6 +2068,7 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev) mutex_init(&con->recovery_lock); INIT_WORK(&con->recovery_work, amdgpu_ras_do_recovery); atomic_set(&con->in_recovery, 0); + con->eeprom_control.bad_channel_bitmap = 0; max_eeprom_records_count = amdgpu_ras_eeprom_max_record_count(); amdgpu_ras_validate_threshold(adev, max_eeprom_records_count); @@ -2092,6 +2093,11 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev) goto free; amdgpu_dpm_send_hbm_bad_pages_num(adev, con->eeprom_control.ras_num_recs); + + if (con->update_channel_flag == true) { + amdgpu_dpm_send_hbm_bad_channel_flag(adev, con->eeprom_control.bad_channel_bitmap); + con->update_channel_flag = false; + } } #ifdef CONFIG_X86_MCE_AMD @@ -2285,6 +2291,7 @@ int amdgpu_ras_init(struct amdgpu_device *adev) goto release_con; } + con->update_channel_flag = false; con->features = 0; INIT_LIST_HEAD(&con->head); /* Might need get this flag from vbios. */ @@ -2477,6 +2484,12 @@ void amdgpu_ras_block_late_fini(struct amdgpu_device *adev, amdgpu_ras_interrupt_remove_handler(adev, ras_block); } +static void amdgpu_ras_block_late_fini_default(struct amdgpu_device *adev, + struct ras_common_if *ras_block) +{ + return amdgpu_ras_block_late_fini(adev, ras_block); +} + /* do some init work after IP late init as dependence. * and it runs in resume/gpu reset/booting up cases. */ @@ -2572,11 +2585,27 @@ int amdgpu_ras_pre_fini(struct amdgpu_device *adev) int amdgpu_ras_fini(struct amdgpu_device *adev) { struct amdgpu_ras_block_list *ras_node, *tmp; + struct amdgpu_ras_block_object *obj = NULL; struct amdgpu_ras *con = amdgpu_ras_get_context(adev); if (!adev->ras_enabled || !con) return 0; + list_for_each_entry_safe(ras_node, tmp, &adev->ras_list, node) { + if (ras_node->ras_obj) { + obj = ras_node->ras_obj; + if (amdgpu_ras_is_supported(adev, obj->ras_comm.block) && + obj->ras_fini) + obj->ras_fini(adev, &obj->ras_comm); + else + amdgpu_ras_block_late_fini_default(adev, &obj->ras_comm); + } + + /* Clear ras blocks from ras_list and free ras block list node */ + list_del(&ras_node->node); + kfree(ras_node); + } + amdgpu_ras_fs_fini(adev); amdgpu_ras_interrupt_remove_all(adev); @@ -2590,12 +2619,6 @@ int amdgpu_ras_fini(struct amdgpu_device *adev) amdgpu_ras_set_context(adev, NULL); kfree(con); - /* Clear ras blocks from ras_list and free ras block list node */ - list_for_each_entry_safe(ras_node, tmp, &adev->ras_list, node) { - list_del(&ras_node->node); - kfree(ras_node); - } - return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h index 143a83043d7c..9314fde81e68 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h @@ -374,6 +374,9 @@ struct amdgpu_ras { /* record umc error info queried from smu */ struct umc_ecc_info umc_ecc; + + /* Indicates smu whether need update bad channel info */ + bool update_channel_flag; }; struct ras_fs_data { @@ -491,7 +494,7 @@ struct amdgpu_ras_block_object { int (*ras_block_match)(struct amdgpu_ras_block_object *block_obj, enum amdgpu_ras_block block, uint32_t sub_block_index); int (*ras_late_init)(struct amdgpu_device *adev, struct ras_common_if *ras_block); - void (*ras_fini)(struct amdgpu_device *adev); + void (*ras_fini)(struct amdgpu_device *adev, struct ras_common_if *ras_block); ras_ih_cb ras_cb; const struct amdgpu_ras_block_hw_ops *hw_ops; }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c index a44f2eeed6ef..c4283987bb1e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c @@ -267,6 +267,7 @@ int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control) { struct amdgpu_device *adev = to_amdgpu_device(control); struct amdgpu_ras_eeprom_table_header *hdr = &control->tbl_hdr; + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); u8 csum; int res; @@ -287,6 +288,10 @@ int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control) amdgpu_dpm_send_hbm_bad_pages_num(adev, control->ras_num_recs); + control->bad_channel_bitmap = 0; + amdgpu_dpm_send_hbm_bad_channel_flag(adev, control->bad_channel_bitmap); + con->update_channel_flag = false; + amdgpu_ras_debugfs_set_ret_size(control); mutex_unlock(&control->ras_tbl_mutex); @@ -420,6 +425,7 @@ amdgpu_ras_eeprom_append_table(struct amdgpu_ras_eeprom_control *control, struct eeprom_table_record *record, const u32 num) { + struct amdgpu_ras *con = amdgpu_ras_get_context(to_amdgpu_device(control)); u32 a, b, i; u8 *buf, *pp; int res; @@ -431,9 +437,16 @@ amdgpu_ras_eeprom_append_table(struct amdgpu_ras_eeprom_control *control, /* Encode all of them in one go. */ pp = buf; - for (i = 0; i < num; i++, pp += RAS_TABLE_RECORD_SIZE) + for (i = 0; i < num; i++, pp += RAS_TABLE_RECORD_SIZE) { __encode_table_record_to_buf(control, &record[i], pp); + /* update bad channel bitmap */ + if (!(control->bad_channel_bitmap & (1 << record[i].mem_channel))) { + control->bad_channel_bitmap |= 1 << record[i].mem_channel; + con->update_channel_flag = true; + } + } + /* a, first record index to write into. * b, last record index to write into. * a = first index to read (fri) + number of records in the table, @@ -686,6 +699,7 @@ int amdgpu_ras_eeprom_read(struct amdgpu_ras_eeprom_control *control, const u32 num) { struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); int i, res; u8 *buf, *pp; u32 g0, g1; @@ -753,8 +767,15 @@ int amdgpu_ras_eeprom_read(struct amdgpu_ras_eeprom_control *control, /* Read up everything? Then transform. */ pp = buf; - for (i = 0; i < num; i++, pp += RAS_TABLE_RECORD_SIZE) + for (i = 0; i < num; i++, pp += RAS_TABLE_RECORD_SIZE) { __decode_table_record_from_buf(control, &record[i], pp); + + /* update bad channel bitmap */ + if (!(control->bad_channel_bitmap & (1 << record[i].mem_channel))) { + control->bad_channel_bitmap |= 1 << record[i].mem_channel; + con->update_channel_flag = true; + } + } Out: kfree(buf); mutex_unlock(&control->ras_tbl_mutex); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h index 6bb00578bfbb..54d9bfe0881d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h @@ -80,6 +80,10 @@ struct amdgpu_ras_eeprom_control { /* Protect table access via this mutex. */ struct mutex ras_tbl_mutex; + + /* Record channel info which occurred bad pages + */ + u32 bad_channel_bitmap; }; /* diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c index 248d64158721..c80af0889773 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c @@ -36,8 +36,8 @@ int amdgpu_reset_init(struct amdgpu_device *adev) { int ret = 0; - switch (adev->asic_type) { - case CHIP_ALDEBARAN: + switch (adev->ip_versions[MP1_HWIP][0]) { + case IP_VERSION(13, 0, 2): ret = aldebaran_reset_init(adev); break; default: @@ -51,8 +51,8 @@ int amdgpu_reset_fini(struct amdgpu_device *adev) { int ret = 0; - switch (adev->asic_type) { - case CHIP_ALDEBARAN: + switch (adev->ip_versions[MP1_HWIP][0]) { + case IP_VERSION(13, 0, 2): ret = aldebaran_reset_fini(adev); break; default: diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index 35bcb6dc1816..7f33ae87cb41 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -193,6 +193,7 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, adev->rings[ring->idx] = ring; ring->num_hw_submission = sched_hw_submission; ring->sched_score = sched_score; + ring->vmid_wait = dma_fence_get_stub(); r = amdgpu_fence_driver_init_ring(ring); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h index 48365da213dc..5320bb0883d8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h @@ -28,6 +28,13 @@ #include <drm/gpu_scheduler.h> #include <drm/drm_print.h> +struct amdgpu_device; +struct amdgpu_ring; +struct amdgpu_ib; +struct amdgpu_cs_parser; +struct amdgpu_job; +struct amdgpu_vm; + /* max number of rings */ #define AMDGPU_MAX_RINGS 28 #define AMDGPU_MAX_HWIP_RINGS 8 @@ -82,11 +89,13 @@ enum amdgpu_ib_pool_type { AMDGPU_IB_POOL_MAX }; -struct amdgpu_device; -struct amdgpu_ring; -struct amdgpu_ib; -struct amdgpu_cs_parser; -struct amdgpu_job; +struct amdgpu_ib { + struct amdgpu_sa_bo *sa_bo; + uint32_t length_dw; + uint64_t gpu_addr; + uint32_t *ptr; + uint32_t flags; +}; struct amdgpu_sched { u32 num_scheds; @@ -111,6 +120,8 @@ struct amdgpu_fence_driver { struct dma_fence **fences; }; +extern const struct drm_sched_backend_ops amdgpu_sched_ops; + void amdgpu_fence_driver_clear_job_fences(struct amdgpu_ring *ring); void amdgpu_fence_driver_force_completion(struct amdgpu_ring *ring); @@ -144,6 +155,7 @@ struct amdgpu_ring_funcs { u32 nop; bool support_64bit_ptrs; bool no_user_fence; + bool secure_submission_supported; unsigned vmhub; unsigned extra_dw; @@ -152,8 +164,12 @@ struct amdgpu_ring_funcs { u64 (*get_wptr)(struct amdgpu_ring *ring); void (*set_wptr)(struct amdgpu_ring *ring); /* validating and patching of IBs */ - int (*parse_cs)(struct amdgpu_cs_parser *p, uint32_t ib_idx); - int (*patch_cs_in_place)(struct amdgpu_cs_parser *p, uint32_t ib_idx); + int (*parse_cs)(struct amdgpu_cs_parser *p, + struct amdgpu_job *job, + struct amdgpu_ib *ib); + int (*patch_cs_in_place)(struct amdgpu_cs_parser *p, + struct amdgpu_job *job, + struct amdgpu_ib *ib); /* constants to calculate how many DW are needed for an emit */ unsigned emit_frame_size; unsigned emit_ib_size; @@ -253,8 +269,8 @@ struct amdgpu_ring { atomic_t *sched_score; }; -#define amdgpu_ring_parse_cs(r, p, ib) ((r)->funcs->parse_cs((p), (ib))) -#define amdgpu_ring_patch_cs_in_place(r, p, ib) ((r)->funcs->patch_cs_in_place((p), (ib))) +#define amdgpu_ring_parse_cs(r, p, job, ib) ((r)->funcs->parse_cs((p), (job), (ib))) +#define amdgpu_ring_patch_cs_in_place(r, p, job, ib) ((r)->funcs->patch_cs_in_place((p), (job), (ib))) #define amdgpu_ring_test_ring(r) (r)->funcs->test_ring((r)) #define amdgpu_ring_test_ib(r, t) (r)->funcs->test_ib((r), (t)) #define amdgpu_ring_get_rptr(r) (r)->funcs->get_rptr((r)) @@ -352,4 +368,29 @@ int amdgpu_ring_test_helper(struct amdgpu_ring *ring); void amdgpu_debugfs_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring); + +static inline u32 amdgpu_ib_get_value(struct amdgpu_ib *ib, int idx) +{ + return ib->ptr[idx]; +} + +static inline void amdgpu_ib_set_value(struct amdgpu_ib *ib, int idx, + uint32_t value) +{ + ib->ptr[idx] = value; +} + +int amdgpu_ib_get(struct amdgpu_device *adev, struct amdgpu_vm *vm, + unsigned size, + enum amdgpu_ib_pool_type pool, + struct amdgpu_ib *ib); +void amdgpu_ib_free(struct amdgpu_device *adev, struct amdgpu_ib *ib, + struct dma_fence *f); +int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, + struct amdgpu_ib *ibs, struct amdgpu_job *job, + struct dma_fence **f); +int amdgpu_ib_pool_init(struct amdgpu_device *adev); +void amdgpu_ib_pool_fini(struct amdgpu_device *adev); +int amdgpu_ib_ring_tests(struct amdgpu_device *adev); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c index 3b5c43575aa3..e1835fd4b237 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c @@ -111,13 +111,6 @@ late_fini: return r; } -void amdgpu_sdma_ras_fini(struct amdgpu_device *adev) -{ - if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA) && - adev->sdma.ras_if) - amdgpu_ras_block_late_fini(adev, adev->sdma.ras_if); -} - int amdgpu_sdma_process_ras_data_cb(struct amdgpu_device *adev, void *err_data, struct amdgpu_iv_entry *entry) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h index 8b226ffee32c..53ac3ebae8d6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h @@ -118,7 +118,6 @@ int amdgpu_sdma_get_index_from_ring(struct amdgpu_ring *ring, uint32_t *index); uint64_t amdgpu_sdma_get_csa_mc_addr(struct amdgpu_ring *ring, unsigned vmid); int amdgpu_sdma_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block); -void amdgpu_sdma_ras_fini(struct amdgpu_device *adev); int amdgpu_sdma_process_ras_data_cb(struct amdgpu_device *adev, void *err_data, struct amdgpu_iv_entry *entry); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace_points.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace_points.c index 57c6c39ba064..b96d885f6e33 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace_points.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace_points.c @@ -23,6 +23,7 @@ */ #include <drm/amdgpu_drm.h> +#include "amdgpu_cs.h" #include "amdgpu.h" #define CREATE_TRACE_POINTS diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h index 428f4df184d0..40dffbac85a0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h @@ -343,7 +343,8 @@ union amdgpu_firmware_header { * fw loading support */ enum AMDGPU_UCODE_ID { - AMDGPU_UCODE_ID_SDMA0 = 0, + AMDGPU_UCODE_ID_CAP = 0, + AMDGPU_UCODE_ID_SDMA0, AMDGPU_UCODE_ID_SDMA1, AMDGPU_UCODE_ID_SDMA2, AMDGPU_UCODE_ID_SDMA3, @@ -378,7 +379,6 @@ enum AMDGPU_UCODE_ID { AMDGPU_UCODE_ID_VCN0_RAM, AMDGPU_UCODE_ID_VCN1_RAM, AMDGPU_UCODE_ID_DMCUB, - AMDGPU_UCODE_ID_CAP, AMDGPU_UCODE_ID_MAXIMUM, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c index 9400260e3263..aad3c8b4c810 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c @@ -97,6 +97,11 @@ static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, amdgpu_ras_save_bad_pages(adev); amdgpu_dpm_send_hbm_bad_pages_num(adev, con->eeprom_control.ras_num_recs); + + if (con->update_channel_flag == true) { + amdgpu_dpm_send_hbm_bad_channel_flag(adev, con->eeprom_control.bad_channel_bitmap); + con->update_channel_flag = false; + } } if (reset) @@ -162,13 +167,6 @@ late_fini: return r; } -void amdgpu_umc_ras_fini(struct amdgpu_device *adev) -{ - if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC) && - adev->umc.ras_if) - amdgpu_ras_block_late_fini(adev, adev->umc.ras_if); -} - int amdgpu_umc_process_ecc_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h index e4b3678a6685..2ec6698aa1fe 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h @@ -73,7 +73,6 @@ struct amdgpu_umc { }; int amdgpu_umc_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block); -void amdgpu_umc_ras_fini(struct amdgpu_device *adev); int amdgpu_umc_poison_handler(struct amdgpu_device *adev, void *ras_error_status, bool reset); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c index 9cc23b220537..39c74d9fa7cc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c @@ -37,6 +37,7 @@ #include "amdgpu.h" #include "amdgpu_pm.h" #include "amdgpu_uvd.h" +#include "amdgpu_cs.h" #include "cikd.h" #include "uvd/uvd_4_2_d.h" @@ -98,7 +99,7 @@ struct amdgpu_uvd_cs_ctx { unsigned reg, count; unsigned data0, data1; unsigned idx; - unsigned ib_idx; + struct amdgpu_ib *ib; /* does the IB has a msg command */ bool has_msg_cmd; @@ -557,8 +558,8 @@ static u64 amdgpu_uvd_get_addr_from_ctx(struct amdgpu_uvd_cs_ctx *ctx) uint32_t lo, hi; uint64_t addr; - lo = amdgpu_get_ib_value(ctx->parser, ctx->ib_idx, ctx->data0); - hi = amdgpu_get_ib_value(ctx->parser, ctx->ib_idx, ctx->data1); + lo = amdgpu_ib_get_value(ctx->ib, ctx->data0); + hi = amdgpu_ib_get_value(ctx->ib, ctx->data1); addr = ((uint64_t)lo) | (((uint64_t)hi) << 32); return addr; @@ -589,7 +590,7 @@ static int amdgpu_uvd_cs_pass1(struct amdgpu_uvd_cs_ctx *ctx) if (!ctx->parser->adev->uvd.address_64_bit) { /* check if it's a message or feedback command */ - cmd = amdgpu_get_ib_value(ctx->parser, ctx->ib_idx, ctx->idx) >> 1; + cmd = amdgpu_ib_get_value(ctx->ib, ctx->idx) >> 1; if (cmd == 0x0 || cmd == 0x3) { /* yes, force it into VRAM */ uint32_t domain = AMDGPU_GEM_DOMAIN_VRAM; @@ -927,12 +928,10 @@ static int amdgpu_uvd_cs_pass2(struct amdgpu_uvd_cs_ctx *ctx) addr -= mapping->start * AMDGPU_GPU_PAGE_SIZE; start += addr; - amdgpu_set_ib_value(ctx->parser, ctx->ib_idx, ctx->data0, - lower_32_bits(start)); - amdgpu_set_ib_value(ctx->parser, ctx->ib_idx, ctx->data1, - upper_32_bits(start)); + amdgpu_ib_set_value(ctx->ib, ctx->data0, lower_32_bits(start)); + amdgpu_ib_set_value(ctx->ib, ctx->data1, upper_32_bits(start)); - cmd = amdgpu_get_ib_value(ctx->parser, ctx->ib_idx, ctx->idx) >> 1; + cmd = amdgpu_ib_get_value(ctx->ib, ctx->idx) >> 1; if (cmd < 0x4) { if ((end - start) < ctx->buf_sizes[cmd]) { DRM_ERROR("buffer (%d) to small (%d / %d)!\n", cmd, @@ -992,14 +991,13 @@ static int amdgpu_uvd_cs_pass2(struct amdgpu_uvd_cs_ctx *ctx) static int amdgpu_uvd_cs_reg(struct amdgpu_uvd_cs_ctx *ctx, int (*cb)(struct amdgpu_uvd_cs_ctx *ctx)) { - struct amdgpu_ib *ib = &ctx->parser->job->ibs[ctx->ib_idx]; int i, r; ctx->idx++; for (i = 0; i <= ctx->count; ++i) { unsigned reg = ctx->reg + i; - if (ctx->idx >= ib->length_dw) { + if (ctx->idx >= ctx->ib->length_dw) { DRM_ERROR("Register command after end of CS!\n"); return -EINVAL; } @@ -1039,11 +1037,10 @@ static int amdgpu_uvd_cs_reg(struct amdgpu_uvd_cs_ctx *ctx, static int amdgpu_uvd_cs_packets(struct amdgpu_uvd_cs_ctx *ctx, int (*cb)(struct amdgpu_uvd_cs_ctx *ctx)) { - struct amdgpu_ib *ib = &ctx->parser->job->ibs[ctx->ib_idx]; int r; - for (ctx->idx = 0 ; ctx->idx < ib->length_dw; ) { - uint32_t cmd = amdgpu_get_ib_value(ctx->parser, ctx->ib_idx, ctx->idx); + for (ctx->idx = 0 ; ctx->idx < ctx->ib->length_dw; ) { + uint32_t cmd = amdgpu_ib_get_value(ctx->ib, ctx->idx); unsigned type = CP_PACKET_GET_TYPE(cmd); switch (type) { case PACKET_TYPE0: @@ -1068,11 +1065,14 @@ static int amdgpu_uvd_cs_packets(struct amdgpu_uvd_cs_ctx *ctx, * amdgpu_uvd_ring_parse_cs - UVD command submission parser * * @parser: Command submission parser context - * @ib_idx: Which indirect buffer to use + * @job: the job to parse + * @ib: the IB to patch * * Parse the command stream, patch in addresses as necessary. */ -int amdgpu_uvd_ring_parse_cs(struct amdgpu_cs_parser *parser, uint32_t ib_idx) +int amdgpu_uvd_ring_parse_cs(struct amdgpu_cs_parser *parser, + struct amdgpu_job *job, + struct amdgpu_ib *ib) { struct amdgpu_uvd_cs_ctx ctx = {}; unsigned buf_sizes[] = { @@ -1082,10 +1082,9 @@ int amdgpu_uvd_ring_parse_cs(struct amdgpu_cs_parser *parser, uint32_t ib_idx) [0x00000003] = 2048, [0x00000004] = 0xFFFFFFFF, }; - struct amdgpu_ib *ib = &parser->job->ibs[ib_idx]; int r; - parser->job->vm = NULL; + job->vm = NULL; ib->gpu_addr = amdgpu_sa_bo_gpu_addr(ib->sa_bo); if (ib->length_dw % 16) { @@ -1096,7 +1095,7 @@ int amdgpu_uvd_ring_parse_cs(struct amdgpu_cs_parser *parser, uint32_t ib_idx) ctx.parser = parser; ctx.buf_sizes = buf_sizes; - ctx.ib_idx = ib_idx; + ctx.ib = ib; /* first round only required on chips without UVD 64 bit address support */ if (!parser->adev->uvd.address_64_bit) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.h index 76ac9699885d..9f89bb7cd60b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.h @@ -82,7 +82,9 @@ int amdgpu_uvd_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle, bool direct, struct dma_fence **fence); void amdgpu_uvd_free_handles(struct amdgpu_device *adev, struct drm_file *filp); -int amdgpu_uvd_ring_parse_cs(struct amdgpu_cs_parser *parser, uint32_t ib_idx); +int amdgpu_uvd_ring_parse_cs(struct amdgpu_cs_parser *parser, + struct amdgpu_job *job, + struct amdgpu_ib *ib); void amdgpu_uvd_ring_begin_use(struct amdgpu_ring *ring); void amdgpu_uvd_ring_end_use(struct amdgpu_ring *ring); int amdgpu_uvd_ring_test_ib(struct amdgpu_ring *ring, long timeout); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c index 344f711ad144..02cb3a12dd76 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c @@ -34,6 +34,7 @@ #include "amdgpu.h" #include "amdgpu_pm.h" #include "amdgpu_vce.h" +#include "amdgpu_cs.h" #include "cikd.h" /* 1 second timeout */ @@ -587,8 +588,7 @@ err: /** * amdgpu_vce_validate_bo - make sure not to cross 4GB boundary * - * @p: parser context - * @ib_idx: indirect buffer to use + * @ib: indirect buffer to use * @lo: address of lower dword * @hi: address of higher dword * @size: minimum size @@ -596,8 +596,9 @@ err: * * Make sure that no BO cross a 4GB boundary. */ -static int amdgpu_vce_validate_bo(struct amdgpu_cs_parser *p, uint32_t ib_idx, - int lo, int hi, unsigned size, int32_t index) +static int amdgpu_vce_validate_bo(struct amdgpu_cs_parser *p, + struct amdgpu_ib *ib, int lo, int hi, + unsigned size, int32_t index) { int64_t offset = ((uint64_t)size) * ((int64_t)index); struct ttm_operation_ctx ctx = { false, false }; @@ -607,8 +608,8 @@ static int amdgpu_vce_validate_bo(struct amdgpu_cs_parser *p, uint32_t ib_idx, uint64_t addr; int r; - addr = ((uint64_t)amdgpu_get_ib_value(p, ib_idx, lo)) | - ((uint64_t)amdgpu_get_ib_value(p, ib_idx, hi)) << 32; + addr = ((uint64_t)amdgpu_ib_get_value(ib, lo)) | + ((uint64_t)amdgpu_ib_get_value(ib, hi)) << 32; if (index >= 0) { addr += offset; fpfn = PAGE_ALIGN(offset) >> PAGE_SHIFT; @@ -638,7 +639,7 @@ static int amdgpu_vce_validate_bo(struct amdgpu_cs_parser *p, uint32_t ib_idx, * amdgpu_vce_cs_reloc - command submission relocation * * @p: parser context - * @ib_idx: indirect buffer to use + * @ib: indirect buffer to use * @lo: address of lower dword * @hi: address of higher dword * @size: minimum size @@ -646,7 +647,7 @@ static int amdgpu_vce_validate_bo(struct amdgpu_cs_parser *p, uint32_t ib_idx, * * Patch relocation inside command stream with real buffer address */ -static int amdgpu_vce_cs_reloc(struct amdgpu_cs_parser *p, uint32_t ib_idx, +static int amdgpu_vce_cs_reloc(struct amdgpu_cs_parser *p, struct amdgpu_ib *ib, int lo, int hi, unsigned size, uint32_t index) { struct amdgpu_bo_va_mapping *mapping; @@ -657,8 +658,8 @@ static int amdgpu_vce_cs_reloc(struct amdgpu_cs_parser *p, uint32_t ib_idx, if (index == 0xffffffff) index = 0; - addr = ((uint64_t)amdgpu_get_ib_value(p, ib_idx, lo)) | - ((uint64_t)amdgpu_get_ib_value(p, ib_idx, hi)) << 32; + addr = ((uint64_t)amdgpu_ib_get_value(ib, lo)) | + ((uint64_t)amdgpu_ib_get_value(ib, hi)) << 32; addr += ((uint64_t)size) * ((uint64_t)index); r = amdgpu_cs_find_mapping(p, addr, &bo, &mapping); @@ -679,8 +680,8 @@ static int amdgpu_vce_cs_reloc(struct amdgpu_cs_parser *p, uint32_t ib_idx, addr += amdgpu_bo_gpu_offset(bo); addr -= ((uint64_t)size) * ((uint64_t)index); - amdgpu_set_ib_value(p, ib_idx, lo, lower_32_bits(addr)); - amdgpu_set_ib_value(p, ib_idx, hi, upper_32_bits(addr)); + amdgpu_ib_set_value(ib, lo, lower_32_bits(addr)); + amdgpu_ib_set_value(ib, hi, upper_32_bits(addr)); return 0; } @@ -729,11 +730,13 @@ static int amdgpu_vce_validate_handle(struct amdgpu_cs_parser *p, * amdgpu_vce_ring_parse_cs - parse and validate the command stream * * @p: parser context - * @ib_idx: indirect buffer to use + * @job: the job to parse + * @ib: the IB to patch */ -int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx) +int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, + struct amdgpu_job *job, + struct amdgpu_ib *ib) { - struct amdgpu_ib *ib = &p->job->ibs[ib_idx]; unsigned fb_idx = 0, bs_idx = 0; int session_idx = -1; uint32_t destroyed = 0; @@ -744,12 +747,12 @@ int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx) unsigned idx; int i, r = 0; - p->job->vm = NULL; + job->vm = NULL; ib->gpu_addr = amdgpu_sa_bo_gpu_addr(ib->sa_bo); for (idx = 0; idx < ib->length_dw;) { - uint32_t len = amdgpu_get_ib_value(p, ib_idx, idx); - uint32_t cmd = amdgpu_get_ib_value(p, ib_idx, idx + 1); + uint32_t len = amdgpu_ib_get_value(ib, idx); + uint32_t cmd = amdgpu_ib_get_value(ib, idx + 1); if ((len < 8) || (len & 3)) { DRM_ERROR("invalid VCE command length (%d)!\n", len); @@ -759,52 +762,52 @@ int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx) switch (cmd) { case 0x00000002: /* task info */ - fb_idx = amdgpu_get_ib_value(p, ib_idx, idx + 6); - bs_idx = amdgpu_get_ib_value(p, ib_idx, idx + 7); + fb_idx = amdgpu_ib_get_value(ib, idx + 6); + bs_idx = amdgpu_ib_get_value(ib, idx + 7); break; case 0x03000001: /* encode */ - r = amdgpu_vce_validate_bo(p, ib_idx, idx + 10, - idx + 9, 0, 0); + r = amdgpu_vce_validate_bo(p, ib, idx + 10, idx + 9, + 0, 0); if (r) goto out; - r = amdgpu_vce_validate_bo(p, ib_idx, idx + 12, - idx + 11, 0, 0); + r = amdgpu_vce_validate_bo(p, ib, idx + 12, idx + 11, + 0, 0); if (r) goto out; break; case 0x05000001: /* context buffer */ - r = amdgpu_vce_validate_bo(p, ib_idx, idx + 3, - idx + 2, 0, 0); + r = amdgpu_vce_validate_bo(p, ib, idx + 3, idx + 2, + 0, 0); if (r) goto out; break; case 0x05000004: /* video bitstream buffer */ - tmp = amdgpu_get_ib_value(p, ib_idx, idx + 4); - r = amdgpu_vce_validate_bo(p, ib_idx, idx + 3, idx + 2, + tmp = amdgpu_ib_get_value(ib, idx + 4); + r = amdgpu_vce_validate_bo(p, ib, idx + 3, idx + 2, tmp, bs_idx); if (r) goto out; break; case 0x05000005: /* feedback buffer */ - r = amdgpu_vce_validate_bo(p, ib_idx, idx + 3, idx + 2, + r = amdgpu_vce_validate_bo(p, ib, idx + 3, idx + 2, 4096, fb_idx); if (r) goto out; break; case 0x0500000d: /* MV buffer */ - r = amdgpu_vce_validate_bo(p, ib_idx, idx + 3, - idx + 2, 0, 0); + r = amdgpu_vce_validate_bo(p, ib, idx + 3, idx + 2, + 0, 0); if (r) goto out; - r = amdgpu_vce_validate_bo(p, ib_idx, idx + 8, - idx + 7, 0, 0); + r = amdgpu_vce_validate_bo(p, ib, idx + 8, idx + 7, + 0, 0); if (r) goto out; break; @@ -814,12 +817,12 @@ int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx) } for (idx = 0; idx < ib->length_dw;) { - uint32_t len = amdgpu_get_ib_value(p, ib_idx, idx); - uint32_t cmd = amdgpu_get_ib_value(p, ib_idx, idx + 1); + uint32_t len = amdgpu_ib_get_value(ib, idx); + uint32_t cmd = amdgpu_ib_get_value(ib, idx + 1); switch (cmd) { case 0x00000001: /* session */ - handle = amdgpu_get_ib_value(p, ib_idx, idx + 2); + handle = amdgpu_ib_get_value(ib, idx + 2); session_idx = amdgpu_vce_validate_handle(p, handle, &allocated); if (session_idx < 0) { @@ -830,8 +833,8 @@ int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx) break; case 0x00000002: /* task info */ - fb_idx = amdgpu_get_ib_value(p, ib_idx, idx + 6); - bs_idx = amdgpu_get_ib_value(p, ib_idx, idx + 7); + fb_idx = amdgpu_ib_get_value(ib, idx + 6); + bs_idx = amdgpu_ib_get_value(ib, idx + 7); break; case 0x01000001: /* create */ @@ -846,8 +849,8 @@ int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx) goto out; } - *size = amdgpu_get_ib_value(p, ib_idx, idx + 8) * - amdgpu_get_ib_value(p, ib_idx, idx + 10) * + *size = amdgpu_ib_get_value(ib, idx + 8) * + amdgpu_ib_get_value(ib, idx + 10) * 8 * 3 / 2; break; @@ -876,12 +879,12 @@ int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx) break; case 0x03000001: /* encode */ - r = amdgpu_vce_cs_reloc(p, ib_idx, idx + 10, idx + 9, + r = amdgpu_vce_cs_reloc(p, ib, idx + 10, idx + 9, *size, 0); if (r) goto out; - r = amdgpu_vce_cs_reloc(p, ib_idx, idx + 12, idx + 11, + r = amdgpu_vce_cs_reloc(p, ib, idx + 12, idx + 11, *size / 3, 0); if (r) goto out; @@ -892,35 +895,35 @@ int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx) break; case 0x05000001: /* context buffer */ - r = amdgpu_vce_cs_reloc(p, ib_idx, idx + 3, idx + 2, + r = amdgpu_vce_cs_reloc(p, ib, idx + 3, idx + 2, *size * 2, 0); if (r) goto out; break; case 0x05000004: /* video bitstream buffer */ - tmp = amdgpu_get_ib_value(p, ib_idx, idx + 4); - r = amdgpu_vce_cs_reloc(p, ib_idx, idx + 3, idx + 2, + tmp = amdgpu_ib_get_value(ib, idx + 4); + r = amdgpu_vce_cs_reloc(p, ib, idx + 3, idx + 2, tmp, bs_idx); if (r) goto out; break; case 0x05000005: /* feedback buffer */ - r = amdgpu_vce_cs_reloc(p, ib_idx, idx + 3, idx + 2, + r = amdgpu_vce_cs_reloc(p, ib, idx + 3, idx + 2, 4096, fb_idx); if (r) goto out; break; case 0x0500000d: /* MV buffer */ - r = amdgpu_vce_cs_reloc(p, ib_idx, idx + 3, - idx + 2, *size, 0); + r = amdgpu_vce_cs_reloc(p, ib, idx + 3, + idx + 2, *size, 0); if (r) goto out; - r = amdgpu_vce_cs_reloc(p, ib_idx, idx + 8, - idx + 7, *size / 12, 0); + r = amdgpu_vce_cs_reloc(p, ib, idx + 8, + idx + 7, *size / 12, 0); if (r) goto out; break; @@ -965,11 +968,13 @@ out: * amdgpu_vce_ring_parse_cs_vm - parse the command stream in VM mode * * @p: parser context - * @ib_idx: indirect buffer to use + * @job: the job to parse + * @ib: the IB to patch */ -int amdgpu_vce_ring_parse_cs_vm(struct amdgpu_cs_parser *p, uint32_t ib_idx) +int amdgpu_vce_ring_parse_cs_vm(struct amdgpu_cs_parser *p, + struct amdgpu_job *job, + struct amdgpu_ib *ib) { - struct amdgpu_ib *ib = &p->job->ibs[ib_idx]; int session_idx = -1; uint32_t destroyed = 0; uint32_t created = 0; @@ -978,8 +983,8 @@ int amdgpu_vce_ring_parse_cs_vm(struct amdgpu_cs_parser *p, uint32_t ib_idx) int i, r = 0, idx = 0; while (idx < ib->length_dw) { - uint32_t len = amdgpu_get_ib_value(p, ib_idx, idx); - uint32_t cmd = amdgpu_get_ib_value(p, ib_idx, idx + 1); + uint32_t len = amdgpu_ib_get_value(ib, idx); + uint32_t cmd = amdgpu_ib_get_value(ib, idx + 1); if ((len < 8) || (len & 3)) { DRM_ERROR("invalid VCE command length (%d)!\n", len); @@ -989,7 +994,7 @@ int amdgpu_vce_ring_parse_cs_vm(struct amdgpu_cs_parser *p, uint32_t ib_idx) switch (cmd) { case 0x00000001: /* session */ - handle = amdgpu_get_ib_value(p, ib_idx, idx + 2); + handle = amdgpu_ib_get_value(ib, idx + 2); session_idx = amdgpu_vce_validate_handle(p, handle, &allocated); if (session_idx < 0) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h index be4a6e773c5b..ea680fc9a6c3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h @@ -59,8 +59,11 @@ int amdgpu_vce_entity_init(struct amdgpu_device *adev); int amdgpu_vce_suspend(struct amdgpu_device *adev); int amdgpu_vce_resume(struct amdgpu_device *adev); void amdgpu_vce_free_handles(struct amdgpu_device *adev, struct drm_file *filp); -int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx); -int amdgpu_vce_ring_parse_cs_vm(struct amdgpu_cs_parser *p, uint32_t ib_idx); +int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, struct amdgpu_job *job, + struct amdgpu_ib *ib); +int amdgpu_vce_ring_parse_cs_vm(struct amdgpu_cs_parser *p, + struct amdgpu_job *job, + struct amdgpu_ib *ib); void amdgpu_vce_ring_emit_ib(struct amdgpu_ring *ring, struct amdgpu_job *job, struct amdgpu_ib *ib, uint32_t flags); void amdgpu_vce_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c index 5e0dbf54d561..f99093f2ebc7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c @@ -27,6 +27,7 @@ #include <linux/firmware.h> #include <linux/module.h> #include <linux/pci.h> +#include <linux/debugfs.h> #include <drm/drm_drv.h> #include "amdgpu.h" @@ -79,6 +80,7 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev) const char *fw_name; const struct common_firmware_header *hdr; unsigned char fw_check; + unsigned int fw_shared_size, log_offset; int i, r; INIT_DELAYED_WORK(&adev->vcn.idle_work, amdgpu_vcn_idle_work_handler); @@ -226,7 +228,12 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev) bo_size = AMDGPU_VCN_STACK_SIZE + AMDGPU_VCN_CONTEXT_SIZE; if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) bo_size += AMDGPU_GPU_PAGE_ALIGN(le32_to_cpu(hdr->ucode_size_bytes) + 8); - bo_size += AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_fw_shared)); + fw_shared_size = AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_fw_shared)); + log_offset = offsetof(struct amdgpu_fw_shared, fw_log); + bo_size += fw_shared_size; + + if (amdgpu_vcnfw_log) + bo_size += AMDGPU_VCNFW_LOG_SIZE; for (i = 0; i < adev->vcn.num_vcn_inst; i++) { if (adev->vcn.harvest_config & (1 << i)) @@ -240,10 +247,18 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev) return r; } - adev->vcn.inst[i].fw_shared_cpu_addr = adev->vcn.inst[i].cpu_addr + - bo_size - AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_fw_shared)); - adev->vcn.inst[i].fw_shared_gpu_addr = adev->vcn.inst[i].gpu_addr + - bo_size - AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_fw_shared)); + adev->vcn.inst[i].fw_shared.cpu_addr = adev->vcn.inst[i].cpu_addr + + bo_size - fw_shared_size; + adev->vcn.inst[i].fw_shared.gpu_addr = adev->vcn.inst[i].gpu_addr + + bo_size - fw_shared_size; + + adev->vcn.inst[i].fw_shared.mem_size = fw_shared_size; + + if (amdgpu_vcnfw_log) { + adev->vcn.inst[i].fw_shared.cpu_addr -= AMDGPU_VCNFW_LOG_SIZE; + adev->vcn.inst[i].fw_shared.gpu_addr -= AMDGPU_VCNFW_LOG_SIZE; + adev->vcn.inst[i].fw_shared.log_offset = log_offset; + } if (adev->vcn.indirect_sram) { r = amdgpu_bo_create_kernel(adev, 64 * 2 * 4, PAGE_SIZE, @@ -979,3 +994,112 @@ void amdgpu_vcn_setup_ucode(struct amdgpu_device *adev) dev_info(adev->dev, "Will use PSP to load VCN firmware\n"); } } + +/* + * debugfs for mapping vcn firmware log buffer. + */ +#if defined(CONFIG_DEBUG_FS) +static ssize_t amdgpu_debugfs_vcn_fwlog_read(struct file *f, char __user *buf, + size_t size, loff_t *pos) +{ + struct amdgpu_vcn_inst *vcn; + void *log_buf; + volatile struct amdgpu_vcn_fwlog *plog; + unsigned int read_pos, write_pos, available, i, read_bytes = 0; + unsigned int read_num[2] = {0}; + + vcn = file_inode(f)->i_private; + if (!vcn) + return -ENODEV; + + if (!vcn->fw_shared.cpu_addr || !amdgpu_vcnfw_log) + return -EFAULT; + + log_buf = vcn->fw_shared.cpu_addr + vcn->fw_shared.mem_size; + + plog = (volatile struct amdgpu_vcn_fwlog *)log_buf; + read_pos = plog->rptr; + write_pos = plog->wptr; + + if (read_pos > AMDGPU_VCNFW_LOG_SIZE || write_pos > AMDGPU_VCNFW_LOG_SIZE) + return -EFAULT; + + if (!size || (read_pos == write_pos)) + return 0; + + if (write_pos > read_pos) { + available = write_pos - read_pos; + read_num[0] = min(size, (size_t)available); + } else { + read_num[0] = AMDGPU_VCNFW_LOG_SIZE - read_pos; + available = read_num[0] + write_pos - plog->header_size; + if (size > available) + read_num[1] = write_pos - plog->header_size; + else if (size > read_num[0]) + read_num[1] = size - read_num[0]; + else + read_num[0] = size; + } + + for (i = 0; i < 2; i++) { + if (read_num[i]) { + if (read_pos == AMDGPU_VCNFW_LOG_SIZE) + read_pos = plog->header_size; + if (read_num[i] == copy_to_user((buf + read_bytes), + (log_buf + read_pos), read_num[i])) + return -EFAULT; + + read_bytes += read_num[i]; + read_pos += read_num[i]; + } + } + + plog->rptr = read_pos; + *pos += read_bytes; + return read_bytes; +} + +static const struct file_operations amdgpu_debugfs_vcnfwlog_fops = { + .owner = THIS_MODULE, + .read = amdgpu_debugfs_vcn_fwlog_read, + .llseek = default_llseek +}; +#endif + +void amdgpu_debugfs_vcn_fwlog_init(struct amdgpu_device *adev, uint8_t i, + struct amdgpu_vcn_inst *vcn) +{ +#if defined(CONFIG_DEBUG_FS) + struct drm_minor *minor = adev_to_drm(adev)->primary; + struct dentry *root = minor->debugfs_root; + char name[32]; + + sprintf(name, "amdgpu_vcn_%d_fwlog", i); + debugfs_create_file_size(name, S_IFREG | S_IRUGO, root, vcn, + &amdgpu_debugfs_vcnfwlog_fops, + AMDGPU_VCNFW_LOG_SIZE); +#endif +} + +void amdgpu_vcn_fwlog_init(struct amdgpu_vcn_inst *vcn) +{ +#if defined(CONFIG_DEBUG_FS) + volatile uint32_t *flag = vcn->fw_shared.cpu_addr; + void *fw_log_cpu_addr = vcn->fw_shared.cpu_addr + vcn->fw_shared.mem_size; + uint64_t fw_log_gpu_addr = vcn->fw_shared.gpu_addr + vcn->fw_shared.mem_size; + volatile struct amdgpu_vcn_fwlog *log_buf = fw_log_cpu_addr; + volatile struct amdgpu_fw_shared_fw_logging *fw_log = vcn->fw_shared.cpu_addr + + vcn->fw_shared.log_offset; + *flag |= cpu_to_le32(AMDGPU_VCN_FW_LOGGING_FLAG); + fw_log->is_enabled = 1; + fw_log->addr_lo = cpu_to_le32(fw_log_gpu_addr & 0xFFFFFFFF); + fw_log->addr_hi = cpu_to_le32(fw_log_gpu_addr >> 32); + fw_log->size = cpu_to_le32(AMDGPU_VCNFW_LOG_SIZE); + + log_buf->header_size = sizeof(struct amdgpu_vcn_fwlog); + log_buf->buffer_size = AMDGPU_VCNFW_LOG_SIZE; + log_buf->rptr = log_buf->header_size; + log_buf->wptr = log_buf->header_size; + log_buf->wrapped = 0; +#endif +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h index 5d3728b027d3..e2fde88aaf5e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h @@ -158,6 +158,7 @@ #define AMDGPU_VCN_FW_SHARED_FLAG_0_RB (1 << 6) #define AMDGPU_VCN_MULTI_QUEUE_FLAG (1 << 8) #define AMDGPU_VCN_SW_RING_FLAG (1 << 9) +#define AMDGPU_VCN_FW_LOGGING_FLAG (1 << 10) #define AMDGPU_VCN_IB_FLAG_DECODE_BUFFER 0x00000001 #define AMDGPU_VCN_CMD_FLAG_MSG_BUFFER 0x00000001 @@ -205,6 +206,13 @@ struct amdgpu_vcn_reg{ unsigned scratch9; }; +struct amdgpu_vcn_fw_shared { + void *cpu_addr; + uint64_t gpu_addr; + uint32_t mem_size; + uint32_t log_offset; +}; + struct amdgpu_vcn_inst { struct amdgpu_bo *vcpu_bo; void *cpu_addr; @@ -221,8 +229,7 @@ struct amdgpu_vcn_inst { uint64_t dpg_sram_gpu_addr; uint32_t *dpg_sram_curr_addr; atomic_t dpg_enc_submission_cnt; - void *fw_shared_cpu_addr; - uint64_t fw_shared_gpu_addr; + struct amdgpu_vcn_fw_shared fw_shared; }; struct amdgpu_vcn { @@ -265,6 +272,13 @@ struct amdgpu_fw_shared_sw_ring { uint8_t padding[3]; }; +struct amdgpu_fw_shared_fw_logging { + uint8_t is_enabled; + uint32_t addr_lo; + uint32_t addr_hi; + uint32_t size; +}; + struct amdgpu_fw_shared { uint32_t present_flag_0; uint8_t pad[44]; @@ -272,6 +286,15 @@ struct amdgpu_fw_shared { uint8_t pad1[1]; struct amdgpu_fw_shared_multi_queue multi_queue; struct amdgpu_fw_shared_sw_ring sw_ring; + struct amdgpu_fw_shared_fw_logging fw_log; +}; + +struct amdgpu_vcn_fwlog { + uint32_t rptr; + uint32_t wptr; + uint32_t buffer_size; + uint32_t header_size; + uint8_t wrapped; }; struct amdgpu_vcn_decode_buffer { @@ -313,4 +336,7 @@ enum amdgpu_ring_priority_level amdgpu_vcn_get_enc_ring_prio(int ring); void amdgpu_vcn_setup_ucode(struct amdgpu_device *adev); +void amdgpu_vcn_fwlog_init(struct amdgpu_vcn_inst *vcn); +void amdgpu_debugfs_vcn_fwlog_init(struct amdgpu_device *adev, + uint8_t i, struct amdgpu_vcn_inst *vcn); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index 2cd9f1a2e5fa..fc4563cf2828 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -779,7 +779,8 @@ bool amdgpu_vm_ready(struct amdgpu_vm *vm) amdgpu_vm_eviction_lock(vm); ret = !vm->evicting; amdgpu_vm_eviction_unlock(vm); - return ret; + + return ret && list_empty(&vm->evicted); } /** diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index 91817a31f3e1..1b108d03e785 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -34,7 +34,6 @@ #include "amdgpu_reset.h" -#define smnPCS_XGMI23_PCS_ERROR_STATUS 0x11a01210 #define smnPCS_XGMI3X16_PCS_ERROR_STATUS 0x11a0020c #define smnPCS_GOPX1_PCS_ERROR_STATUS 0x12200210 @@ -69,17 +68,6 @@ static const int wafl_pcs_err_status_reg_arct[] = { smnPCS_GOPX1_0_PCS_GOPX1_PCS_ERROR_STATUS + 0x100000, }; -static const int xgmi23_pcs_err_status_reg_aldebaran[] = { - smnPCS_XGMI23_PCS_ERROR_STATUS, - smnPCS_XGMI23_PCS_ERROR_STATUS + 0x100000, - smnPCS_XGMI23_PCS_ERROR_STATUS + 0x200000, - smnPCS_XGMI23_PCS_ERROR_STATUS + 0x300000, - smnPCS_XGMI23_PCS_ERROR_STATUS + 0x400000, - smnPCS_XGMI23_PCS_ERROR_STATUS + 0x500000, - smnPCS_XGMI23_PCS_ERROR_STATUS + 0x600000, - smnPCS_XGMI23_PCS_ERROR_STATUS + 0x700000 -}; - static const int xgmi3x16_pcs_err_status_reg_aldebaran[] = { smnPCS_XGMI3X16_PCS_ERROR_STATUS, smnPCS_XGMI3X16_PCS_ERROR_STATUS + 0x100000, @@ -768,13 +756,6 @@ static int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev, struct ras_comm return amdgpu_ras_block_late_init(adev, ras_block); } -static void amdgpu_xgmi_ras_fini(struct amdgpu_device *adev) -{ - if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__XGMI_WAFL) && - adev->gmc.xgmi.ras_if) - amdgpu_ras_block_late_fini(adev, adev->gmc.xgmi.ras_if); -} - uint64_t amdgpu_xgmi_get_relative_phy_addr(struct amdgpu_device *adev, uint64_t addr) { @@ -804,9 +785,6 @@ static void amdgpu_xgmi_reset_ras_error_count(struct amdgpu_device *adev) xgmi_pcs_err_status_reg_vg20[i]); break; case CHIP_ALDEBARAN: - for (i = 0; i < ARRAY_SIZE(xgmi23_pcs_err_status_reg_aldebaran); i++) - pcs_clear_status(adev, - xgmi23_pcs_err_status_reg_aldebaran[i]); for (i = 0; i < ARRAY_SIZE(xgmi3x16_pcs_err_status_reg_aldebaran); i++) pcs_clear_status(adev, xgmi3x16_pcs_err_status_reg_aldebaran[i]); @@ -907,13 +885,6 @@ static void amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev, } break; case CHIP_ALDEBARAN: - /* check xgmi23 pcs error */ - for (i = 0; i < ARRAY_SIZE(xgmi23_pcs_err_status_reg_aldebaran); i++) { - data = RREG32_PCIE(xgmi23_pcs_err_status_reg_aldebaran[i]); - if (data) - amdgpu_xgmi_query_pcs_error_status(adev, - data, &ue_cnt, &ce_cnt, true); - } /* check xgmi3x16 pcs error */ for (i = 0; i < ARRAY_SIZE(xgmi3x16_pcs_err_status_reg_aldebaran); i++) { data = RREG32_PCIE(xgmi3x16_pcs_err_status_reg_aldebaran[i]); @@ -982,6 +953,5 @@ struct amdgpu_xgmi_ras xgmi_ras = { }, .hw_ops = &xgmi_ras_hw_ops, .ras_late_init = amdgpu_xgmi_ras_late_init, - .ras_fini = amdgpu_xgmi_ras_fini, }, }; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index 90158289cd30..f4c6accd3226 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -250,13 +250,6 @@ MODULE_FIRMWARE("amdgpu/yellow_carp_mec.bin"); MODULE_FIRMWARE("amdgpu/yellow_carp_mec2.bin"); MODULE_FIRMWARE("amdgpu/yellow_carp_rlc.bin"); -MODULE_FIRMWARE("amdgpu/cyan_skillfish_ce.bin"); -MODULE_FIRMWARE("amdgpu/cyan_skillfish_pfp.bin"); -MODULE_FIRMWARE("amdgpu/cyan_skillfish_me.bin"); -MODULE_FIRMWARE("amdgpu/cyan_skillfish_mec.bin"); -MODULE_FIRMWARE("amdgpu/cyan_skillfish_mec2.bin"); -MODULE_FIRMWARE("amdgpu/cyan_skillfish_rlc.bin"); - MODULE_FIRMWARE("amdgpu/cyan_skillfish2_ce.bin"); MODULE_FIRMWARE("amdgpu/cyan_skillfish2_pfp.bin"); MODULE_FIRMWARE("amdgpu/cyan_skillfish2_me.bin"); @@ -4043,10 +4036,7 @@ static int gfx_v10_0_init_microcode(struct amdgpu_device *adev) break; case IP_VERSION(10, 1, 3): case IP_VERSION(10, 1, 4): - if (adev->apu_flags & AMD_APU_IS_CYAN_SKILLFISH2) - chip_name = "cyan_skillfish2"; - else - chip_name = "cyan_skillfish"; + chip_name = "cyan_skillfish2"; break; case IP_VERSION(10, 3, 7): chip_name = "gc_10_3_7"; @@ -6557,6 +6547,7 @@ static void gfx_v10_0_kiq_setting(struct amdgpu_ring *ring) case IP_VERSION(10, 3, 5): case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 7): tmp = RREG32_SOC15(GC, 0, mmRLC_CP_SCHEDULERS_Sienna_Cichlid); tmp &= 0xffffff00; tmp |= (ring->me << 5) | (ring->pipe << 3) | (ring->queue); @@ -7857,6 +7848,7 @@ static void gfx_v10_0_set_safe_mode(struct amdgpu_device *adev) case IP_VERSION(10, 3, 5): case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 7): WREG32_SOC15(GC, 0, mmRLC_SAFE_MODE_Sienna_Cichlid, data); /* wait for RLC_SAFE_MODE */ @@ -7894,6 +7886,7 @@ static void gfx_v10_0_unset_safe_mode(struct amdgpu_device *adev) case IP_VERSION(10, 3, 5): case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 7): WREG32_SOC15(GC, 0, mmRLC_SAFE_MODE_Sienna_Cichlid, data); break; default: @@ -8348,6 +8341,7 @@ static void gfx_v10_cntl_power_gating(struct amdgpu_device *adev, bool enable) case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 3): case IP_VERSION(10, 3, 6): + case IP_VERSION(10, 3, 7): data = 0x4E20 & RLC_PG_DELAY_3__CGCG_ACTIVE_BEFORE_CGPG_MASK_Vangogh; WREG32_SOC15(GC, 0, mmRLC_PG_DELAY_3, data); break; @@ -8417,6 +8411,7 @@ static int gfx_v10_0_set_powergating_state(void *handle, case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 3): case IP_VERSION(10, 3, 6): + case IP_VERSION(10, 3, 7): gfx_v10_cntl_pg(adev, enable); amdgpu_gfx_off_ctrl(adev, enable); break; @@ -8445,6 +8440,7 @@ static int gfx_v10_0_set_clockgating_state(void *handle, case IP_VERSION(10, 3, 5): case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 7): gfx_v10_0_update_gfx_clock_gating(adev, state == AMD_CG_STATE_GATE); break; @@ -9381,6 +9377,7 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_gfx = { .align_mask = 0xff, .nop = PACKET3(PACKET3_NOP, 0x3FFF), .support_64bit_ptrs = true, + .secure_submission_supported = true, .vmhub = AMDGPU_GFXHUB_0, .get_rptr = gfx_v10_0_ring_get_rptr_gfx, .get_wptr = gfx_v10_0_ring_get_wptr_gfx, diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index 1997f129db9c..46d4bf27ebbb 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -2204,10 +2204,6 @@ static int gfx_v9_0_gpu_early_init(struct amdgpu_device *adev) if (!adev->gfx.ras->ras_block.ras_late_init) adev->gfx.ras->ras_block.ras_late_init = amdgpu_gfx_ras_late_init; - /* If not define special ras_fini function, use gfx default ras_fini */ - if (!adev->gfx.ras->ras_block.ras_fini) - adev->gfx.ras->ras_block.ras_fini = amdgpu_gfx_ras_fini; - /* If not defined special ras_cb function, use default ras_cb */ if (!adev->gfx.ras->ras_block.ras_cb) adev->gfx.ras->ras_block.ras_cb = amdgpu_gfx_process_ras_data_cb; @@ -2432,9 +2428,6 @@ static int gfx_v9_0_sw_fini(void *handle) int i; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (adev->gfx.ras && adev->gfx.ras->ras_block.ras_fini) - adev->gfx.ras->ras_block.ras_fini(adev); - for (i = 0; i < adev->gfx.num_gfx_rings; i++) amdgpu_ring_fini(&adev->gfx.gfx_ring[i]); for (i = 0; i < adev->gfx.num_compute_rings; i++) @@ -6872,6 +6865,7 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_gfx = { .align_mask = 0xff, .nop = PACKET3(PACKET3_NOP, 0x3FFF), .support_64bit_ptrs = true, + .secure_submission_supported = true, .vmhub = AMDGPU_GFXHUB_0, .get_rptr = gfx_v9_0_ring_get_rptr_gfx, .get_wptr = gfx_v9_0_ring_get_wptr_gfx, diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index 3dcd82b49481..3c1d440824a7 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -683,10 +683,6 @@ static void gmc_v10_0_set_umc_funcs(struct amdgpu_device *adev) if (!adev->umc.ras->ras_block.ras_late_init) adev->umc.ras->ras_block.ras_late_init = amdgpu_umc_ras_late_init; - /* If don't define special ras_fini function, use default ras_fini */ - if (!adev->umc.ras->ras_block.ras_fini) - adev->umc.ras->ras_block.ras_fini = amdgpu_umc_ras_fini; - /* If not defined special ras_cb function, use default ras_cb */ if (!adev->umc.ras->ras_block.ras_cb) adev->umc.ras->ras_block.ras_cb = amdgpu_umc_process_ras_data_cb; @@ -952,7 +948,6 @@ static int gmc_v10_0_sw_init(void *handle) return r; amdgpu_gmc_get_vbios_allocations(adev); - amdgpu_gmc_get_reserved_allocation(adev); /* Memory manager */ r = amdgpu_bo_init(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index df35f0252eea..431742eb7811 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -1243,10 +1243,6 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev) if (!adev->umc.ras->ras_block.ras_late_init) adev->umc.ras->ras_block.ras_late_init = amdgpu_umc_ras_late_init; - /* If don't define special ras_fini function, use default ras_fini */ - if (!adev->umc.ras->ras_block.ras_fini) - adev->umc.ras->ras_block.ras_fini = amdgpu_umc_ras_fini; - /* If not defined special ras_cb function, use default ras_cb */ if (!adev->umc.ras->ras_block.ras_cb) adev->umc.ras->ras_block.ras_cb = amdgpu_umc_process_ras_data_cb; @@ -1292,10 +1288,6 @@ static void gmc_v9_0_set_mmhub_ras_funcs(struct amdgpu_device *adev) adev->mmhub.ras->ras_block.ras_comm.block = AMDGPU_RAS_BLOCK__MMHUB; adev->mmhub.ras->ras_block.ras_comm.type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE; adev->mmhub.ras_if = &adev->mmhub.ras->ras_block.ras_comm; - - /* If don't define special ras_fini function, use default ras_fini */ - if (!adev->mmhub.ras->ras_block.ras_fini) - adev->mmhub.ras->ras_block.ras_fini = amdgpu_mmhub_ras_fini; } } @@ -1561,7 +1553,7 @@ static void gmc_v9_0_save_registers(struct amdgpu_device *adev) static int gmc_v9_0_sw_init(void *handle) { - int r, vram_width = 0, vram_type = 0, vram_vendor = 0; + int r, vram_width = 0, vram_type = 0, vram_vendor = 0, dma_addr_bits; struct amdgpu_device *adev = (struct amdgpu_device *)handle; adev->gfxhub.funcs->init(adev); @@ -1677,12 +1669,13 @@ static int gmc_v9_0_sw_init(void *handle) */ adev->gmc.mc_mask = 0xffffffffffffULL; /* 48 bit MC */ - r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(44)); + dma_addr_bits = adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2) ? 48:44; + r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(dma_addr_bits)); if (r) { printk(KERN_WARNING "amdgpu: No suitable DMA available.\n"); return r; } - adev->need_swiotlb = drm_need_swiotlb(44); + adev->need_swiotlb = drm_need_swiotlb(dma_addr_bits); r = gmc_v9_0_mc_init(adev); if (r) diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c index 02400d97a95c..046216635262 100644 --- a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c @@ -166,7 +166,6 @@ struct amdgpu_hdp_ras hdp_v4_0_ras = { .type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE, }, .hw_ops = &hdp_v4_0_ras_hw_ops, - .ras_fini = amdgpu_hdp_ras_fini, }, }; diff --git a/drivers/gpu/drm/amd/amdgpu/mca_v3_0.c b/drivers/gpu/drm/amd/amdgpu/mca_v3_0.c index b4b36899f5c6..d4bd7d1d2649 100644 --- a/drivers/gpu/drm/amd/amdgpu/mca_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mca_v3_0.c @@ -37,11 +37,6 @@ static void mca_v3_0_mp0_query_ras_error_count(struct amdgpu_device *adev, ras_error_status); } -static void mca_v3_0_mp0_ras_fini(struct amdgpu_device *adev) -{ - amdgpu_mca_ras_fini(adev, &adev->mca.mp0); -} - static int mca_v3_0_ras_block_match(struct amdgpu_ras_block_object *block_obj, enum amdgpu_ras_block block, uint32_t sub_block_index) { @@ -71,7 +66,6 @@ struct amdgpu_mca_ras_block mca_v3_0_mp0_ras = { }, .hw_ops = &mca_v3_0_mp0_hw_ops, .ras_block_match = mca_v3_0_ras_block_match, - .ras_fini = mca_v3_0_mp0_ras_fini, }, }; @@ -83,11 +77,6 @@ static void mca_v3_0_mp1_query_ras_error_count(struct amdgpu_device *adev, ras_error_status); } -static void mca_v3_0_mp1_ras_fini(struct amdgpu_device *adev) -{ - amdgpu_mca_ras_fini(adev, &adev->mca.mp1); -} - const struct amdgpu_ras_block_hw_ops mca_v3_0_mp1_hw_ops = { .query_ras_error_count = mca_v3_0_mp1_query_ras_error_count, .query_ras_error_address = NULL, @@ -103,7 +92,6 @@ struct amdgpu_mca_ras_block mca_v3_0_mp1_ras = { }, .hw_ops = &mca_v3_0_mp1_hw_ops, .ras_block_match = mca_v3_0_ras_block_match, - .ras_fini = mca_v3_0_mp1_ras_fini, }, }; @@ -115,11 +103,6 @@ static void mca_v3_0_mpio_query_ras_error_count(struct amdgpu_device *adev, ras_error_status); } -static void mca_v3_0_mpio_ras_fini(struct amdgpu_device *adev) -{ - amdgpu_mca_ras_fini(adev, &adev->mca.mpio); -} - const struct amdgpu_ras_block_hw_ops mca_v3_0_mpio_hw_ops = { .query_ras_error_count = mca_v3_0_mpio_query_ras_error_count, .query_ras_error_address = NULL, @@ -135,7 +118,6 @@ struct amdgpu_mca_ras_block mca_v3_0_mpio_ras = { }, .hw_ops = &mca_v3_0_mpio_hw_ops, .ras_block_match = mca_v3_0_ras_block_match, - .ras_fini = mca_v3_0_mpio_ras_fini, }, }; diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c index 14768570c298..c2357e83a8c4 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c @@ -362,9 +362,24 @@ const struct nbio_hdp_flush_reg nbio_v7_4_hdp_flush_reg_ald = { static void nbio_v7_4_init_registers(struct amdgpu_device *adev) { + uint32_t baco_cntl; + if (amdgpu_sriov_vf(adev)) adev->rmmio_remap.reg_offset = SOC15_REG_OFFSET(NBIO, 0, mmBIF_BX_DEV0_EPF0_VF0_HDP_MEM_COHERENCY_FLUSH_CNTL) << 2; + + if (adev->ip_versions[NBIO_HWIP][0] == IP_VERSION(7, 4, 4) && + !amdgpu_sriov_vf(adev)) { + baco_cntl = RREG32_SOC15(NBIO, 0, mmBACO_CNTL); + if (baco_cntl & + (BACO_CNTL__BACO_DUMMY_EN_MASK | BACO_CNTL__BACO_EN_MASK)) { + baco_cntl &= ~(BACO_CNTL__BACO_DUMMY_EN_MASK | + BACO_CNTL__BACO_EN_MASK); + dev_dbg(adev->dev, "Unsetting baco dummy mode %x", + baco_cntl); + WREG32_SOC15(NBIO, 0, mmBACO_CNTL, baco_cntl); + } + } } static void nbio_v7_4_handle_ras_controller_intr_no_bifring(struct amdgpu_device *adev) @@ -671,7 +686,6 @@ struct amdgpu_nbio_ras nbio_v7_4_ras = { }, .hw_ops = &nbio_v7_4_ras_hw_ops, .ras_late_init = amdgpu_nbio_ras_late_init, - .ras_fini = amdgpu_nbio_ras_fini, }, .handle_ras_controller_intr_no_bifring = nbio_v7_4_handle_ras_controller_intr_no_bifring, .handle_ras_err_event_athub_intr_no_bifring = nbio_v7_4_handle_ras_err_event_athub_intr_no_bifring, diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c index f6aeef759ee1..e19f14c3ef59 100644 --- a/drivers/gpu/drm/amd/amdgpu/nv.c +++ b/drivers/gpu/drm/amd/amdgpu/nv.c @@ -946,10 +946,21 @@ static int nv_common_early_init(void *handle) AMD_CG_SUPPORT_GFX_3D_CGLS | AMD_CG_SUPPORT_GFX_RLC_LS | AMD_CG_SUPPORT_GFX_CP_LS | - AMD_CG_SUPPORT_GFX_FGCG; + AMD_CG_SUPPORT_GFX_FGCG | + AMD_CG_SUPPORT_MC_MGCG | + AMD_CG_SUPPORT_MC_LS | + AMD_CG_SUPPORT_SDMA_LS | + AMD_CG_SUPPORT_HDP_MGCG | + AMD_CG_SUPPORT_HDP_LS | + AMD_CG_SUPPORT_ATHUB_MGCG | + AMD_CG_SUPPORT_ATHUB_LS | + AMD_CG_SUPPORT_IH_CG | + AMD_CG_SUPPORT_VCN_MGCG | + AMD_CG_SUPPORT_JPEG_MGCG; adev->pg_flags = AMD_PG_SUPPORT_VCN | AMD_PG_SUPPORT_VCN_DPG | - AMD_PG_SUPPORT_JPEG; + AMD_PG_SUPPORT_JPEG | + AMD_PG_SUPPORT_GFX_PG; adev->external_rev_id = adev->rev_id + 0x01; break; default: diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c index 2c6070b90dcf..024f60631faf 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c @@ -31,6 +31,7 @@ MODULE_FIRMWARE("amdgpu/aldebaran_sos.bin"); MODULE_FIRMWARE("amdgpu/aldebaran_ta.bin"); +MODULE_FIRMWARE("amdgpu/aldebaran_cap.bin"); MODULE_FIRMWARE("amdgpu/yellow_carp_asd.bin"); MODULE_FIRMWARE("amdgpu/yellow_carp_toc.bin"); MODULE_FIRMWARE("amdgpu/yellow_carp_ta.bin"); diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c index 4509bd4cce2d..1d8bbcbd7a37 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c @@ -1142,6 +1142,7 @@ static const struct amdgpu_ring_funcs sdma_v2_4_ring_funcs = { .align_mask = 0xf, .nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), .support_64bit_ptrs = false, + .secure_submission_supported = true, .get_rptr = sdma_v2_4_ring_get_rptr, .get_wptr = sdma_v2_4_ring_get_wptr, .set_wptr = sdma_v2_4_ring_set_wptr, diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c index 135727b59c41..4ef4feff5649 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c @@ -1580,6 +1580,7 @@ static const struct amdgpu_ring_funcs sdma_v3_0_ring_funcs = { .align_mask = 0xf, .nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), .support_64bit_ptrs = false, + .secure_submission_supported = true, .get_rptr = sdma_v3_0_ring_get_rptr, .get_wptr = sdma_v3_0_ring_get_wptr, .set_wptr = sdma_v3_0_ring_set_wptr, diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c index e26c39fcd336..d7e8f7232364 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c @@ -1995,10 +1995,6 @@ static int sdma_v4_0_sw_fini(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; int i; - if (adev->sdma.ras && adev->sdma.ras->ras_block.hw_ops && - adev->sdma.ras->ras_block.ras_fini) - adev->sdma.ras->ras_block.ras_fini(adev); - for (i = 0; i < adev->sdma.num_instances; i++) { amdgpu_ring_fini(&adev->sdma.instance[i].ring); if (adev->sdma.has_page_queue) @@ -2418,6 +2414,7 @@ static const struct amdgpu_ring_funcs sdma_v4_0_ring_funcs = { .align_mask = 0xf, .nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), .support_64bit_ptrs = true, + .secure_submission_supported = true, .vmhub = AMDGPU_MMHUB_0, .get_rptr = sdma_v4_0_ring_get_rptr, .get_wptr = sdma_v4_0_ring_get_wptr, @@ -2454,6 +2451,7 @@ static const struct amdgpu_ring_funcs sdma_v4_0_ring_funcs_2nd_mmhub = { .align_mask = 0xf, .nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), .support_64bit_ptrs = true, + .secure_submission_supported = true, .vmhub = AMDGPU_MMHUB_1, .get_rptr = sdma_v4_0_ring_get_rptr, .get_wptr = sdma_v4_0_ring_get_wptr, @@ -2486,6 +2484,7 @@ static const struct amdgpu_ring_funcs sdma_v4_0_page_ring_funcs = { .align_mask = 0xf, .nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), .support_64bit_ptrs = true, + .secure_submission_supported = true, .vmhub = AMDGPU_MMHUB_0, .get_rptr = sdma_v4_0_ring_get_rptr, .get_wptr = sdma_v4_0_page_ring_get_wptr, @@ -2518,6 +2517,7 @@ static const struct amdgpu_ring_funcs sdma_v4_0_page_ring_funcs_2nd_mmhub = { .align_mask = 0xf, .nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), .support_64bit_ptrs = true, + .secure_submission_supported = true, .vmhub = AMDGPU_MMHUB_1, .get_rptr = sdma_v4_0_ring_get_rptr, .get_wptr = sdma_v4_0_page_ring_get_wptr, @@ -2826,10 +2826,6 @@ static void sdma_v4_0_set_ras_funcs(struct amdgpu_device *adev) if (!adev->sdma.ras->ras_block.ras_late_init) adev->sdma.ras->ras_block.ras_late_init = amdgpu_sdma_ras_late_init; - /* If don't define special ras_fini function, use default ras_fini */ - if (!adev->sdma.ras->ras_block.ras_fini) - adev->sdma.ras->ras_block.ras_fini = amdgpu_sdma_ras_fini; - /* If not defined special ras_cb function, use default ras_cb */ if (!adev->sdma.ras->ras_block.ras_cb) adev->sdma.ras->ras_block.ras_cb = amdgpu_sdma_process_ras_data_cb; diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c index 81e033549dda..a8d49c005f73 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c @@ -51,9 +51,6 @@ MODULE_FIRMWARE("amdgpu/navi14_sdma1.bin"); MODULE_FIRMWARE("amdgpu/navi12_sdma.bin"); MODULE_FIRMWARE("amdgpu/navi12_sdma1.bin"); -MODULE_FIRMWARE("amdgpu/cyan_skillfish_sdma.bin"); -MODULE_FIRMWARE("amdgpu/cyan_skillfish_sdma1.bin"); - MODULE_FIRMWARE("amdgpu/cyan_skillfish2_sdma.bin"); MODULE_FIRMWARE("amdgpu/cyan_skillfish2_sdma1.bin"); @@ -264,10 +261,7 @@ static int sdma_v5_0_init_microcode(struct amdgpu_device *adev) chip_name = "navi12"; break; case IP_VERSION(5, 0, 1): - if (adev->apu_flags & AMD_APU_IS_CYAN_SKILLFISH2) - chip_name = "cyan_skillfish2"; - else - chip_name = "cyan_skillfish"; + chip_name = "cyan_skillfish2"; break; default: BUG(); @@ -1696,6 +1690,7 @@ static const struct amdgpu_ring_funcs sdma_v5_0_ring_funcs = { .align_mask = 0xf, .nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), .support_64bit_ptrs = true, + .secure_submission_supported = true, .vmhub = AMDGPU_GFXHUB_0, .get_rptr = sdma_v5_0_ring_get_rptr, .get_wptr = sdma_v5_0_ring_get_wptr, diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c index dcc622e18d45..824eace69884 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c @@ -1687,6 +1687,7 @@ static const struct amdgpu_ring_funcs sdma_v5_2_ring_funcs = { .align_mask = 0xf, .nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), .support_64bit_ptrs = true, + .secure_submission_supported = true, .vmhub = AMDGPU_GFXHUB_0, .get_rptr = sdma_v5_2_ring_get_rptr, .get_wptr = sdma_v5_2_ring_get_wptr, diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index b7c24149a6bd..3d0251ef8d79 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -853,6 +853,10 @@ static bool soc15_need_reset_on_init(struct amdgpu_device *adev) { u32 sol_reg; + /* CP hangs in IGT reloading test on RN, reset to WA */ + if (adev->asic_type == CHIP_RENOIR) + return true; + /* Just return false for soc15 GPUs. Reset does not seem to * be necessary. */ @@ -1214,9 +1218,6 @@ static int soc15_common_sw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (adev->nbio.ras && adev->nbio.ras->ras_block.ras_fini) - adev->nbio.ras->ras_block.ras_fini(adev); - if (adev->df.funcs && adev->df.funcs->sw_fini) adev->df.funcs->sw_fini(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c index b483f03b4591..2f15b8e0f7d7 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c @@ -25,6 +25,7 @@ #include "amdgpu.h" #include "amdgpu_uvd.h" +#include "amdgpu_cs.h" #include "soc15.h" #include "soc15d.h" #include "soc15_common.h" @@ -1275,14 +1276,15 @@ static int uvd_v7_0_ring_test_ring(struct amdgpu_ring *ring) * uvd_v7_0_ring_patch_cs_in_place - Patch the IB for command submission. * * @p: the CS parser with the IBs - * @ib_idx: which IB to patch + * @job: which job this ib is in + * @ib: which IB to patch * */ static int uvd_v7_0_ring_patch_cs_in_place(struct amdgpu_cs_parser *p, - uint32_t ib_idx) + struct amdgpu_job *job, + struct amdgpu_ib *ib) { - struct amdgpu_ring *ring = to_amdgpu_ring(p->entity->rq->sched); - struct amdgpu_ib *ib = &p->job->ibs[ib_idx]; + struct amdgpu_ring *ring = to_amdgpu_ring(job->base.sched); unsigned i; /* No patching necessary for the first instance */ @@ -1290,12 +1292,12 @@ static int uvd_v7_0_ring_patch_cs_in_place(struct amdgpu_cs_parser *p, return 0; for (i = 0; i < ib->length_dw; i += 2) { - uint32_t reg = amdgpu_get_ib_value(p, ib_idx, i); + uint32_t reg = amdgpu_ib_get_value(ib, i); reg -= p->adev->reg_offset[UVD_HWIP][0][1]; reg += p->adev->reg_offset[UVD_HWIP][1][1]; - amdgpu_set_ib_value(p, ib_idx, i, reg); + amdgpu_ib_set_value(ib, i, reg); } return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c index 3799226defc0..dff54190b96c 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c @@ -148,6 +148,13 @@ static int vcn_v1_0_sw_init(void *handle) adev->vcn.pause_dpg_mode = vcn_v1_0_pause_dpg_mode; + if (amdgpu_vcnfw_log) { + volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr; + + fw_shared->present_flag_0 = 0; + amdgpu_vcn_fwlog_init(adev->vcn.inst); + } + r = jpeg_v1_0_sw_init(handle); return r; @@ -1095,13 +1102,8 @@ static int vcn_v1_0_start_dpg_mode(struct amdgpu_device *adev) static int vcn_v1_0_start(struct amdgpu_device *adev) { - int r; - - if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) - r = vcn_v1_0_start_dpg_mode(adev); - else - r = vcn_v1_0_start_spg_mode(adev); - return r; + return (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) ? + vcn_v1_0_start_dpg_mode(adev) : vcn_v1_0_start_spg_mode(adev); } /** @@ -1903,6 +1905,7 @@ static const struct amdgpu_ring_funcs vcn_v1_0_dec_ring_vm_funcs = { .align_mask = 0xf, .support_64bit_ptrs = false, .no_user_fence = true, + .secure_submission_supported = true, .vmhub = AMDGPU_MMHUB_0, .get_rptr = vcn_v1_0_dec_ring_get_rptr, .get_wptr = vcn_v1_0_dec_ring_get_wptr, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c index 313fc1b53999..7a7f35e83dd5 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c @@ -172,8 +172,12 @@ static int vcn_v2_0_sw_init(void *handle) if (r) return r; - fw_shared = adev->vcn.inst->fw_shared_cpu_addr; + fw_shared = adev->vcn.inst->fw_shared.cpu_addr; fw_shared->present_flag_0 = cpu_to_le32(AMDGPU_VCN_MULTI_QUEUE_FLAG); + + if (amdgpu_vcnfw_log) + amdgpu_vcn_fwlog_init(adev->vcn.inst); + return 0; } @@ -188,7 +192,7 @@ static int vcn_v2_0_sw_fini(void *handle) { int r, idx; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared_cpu_addr; + volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr; if (drm_dev_enter(adev_to_drm(adev), &idx)) { fw_shared->present_flag_0 = 0; @@ -364,9 +368,9 @@ static void vcn_v2_0_mc_resume(struct amdgpu_device *adev) /* non-cache window */ WREG32_SOC15(UVD, 0, mmUVD_LMI_VCPU_NC0_64BIT_BAR_LOW, - lower_32_bits(adev->vcn.inst->fw_shared_gpu_addr)); + lower_32_bits(adev->vcn.inst->fw_shared.gpu_addr)); WREG32_SOC15(UVD, 0, mmUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH, - upper_32_bits(adev->vcn.inst->fw_shared_gpu_addr)); + upper_32_bits(adev->vcn.inst->fw_shared.gpu_addr)); WREG32_SOC15(UVD, 0, mmUVD_VCPU_NONCACHE_OFFSET0, 0); WREG32_SOC15(UVD, 0, mmUVD_VCPU_NONCACHE_SIZE0, AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_fw_shared))); @@ -455,10 +459,10 @@ static void vcn_v2_0_mc_resume_dpg_mode(struct amdgpu_device *adev, bool indirec /* non-cache window */ WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET( UVD, 0, mmUVD_LMI_VCPU_NC0_64BIT_BAR_LOW), - lower_32_bits(adev->vcn.inst->fw_shared_gpu_addr), 0, indirect); + lower_32_bits(adev->vcn.inst->fw_shared.gpu_addr), 0, indirect); WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET( UVD, 0, mmUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH), - upper_32_bits(adev->vcn.inst->fw_shared_gpu_addr), 0, indirect); + upper_32_bits(adev->vcn.inst->fw_shared.gpu_addr), 0, indirect); WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET( UVD, 0, mmUVD_VCPU_NONCACHE_OFFSET0), 0, 0, indirect); WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET( @@ -784,7 +788,7 @@ static void vcn_v2_0_enable_static_power_gating(struct amdgpu_device *adev) static int vcn_v2_0_start_dpg_mode(struct amdgpu_device *adev, bool indirect) { - volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared_cpu_addr; + volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr; struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec; uint32_t rb_bufsz, tmp; @@ -921,7 +925,7 @@ static int vcn_v2_0_start_dpg_mode(struct amdgpu_device *adev, bool indirect) static int vcn_v2_0_start(struct amdgpu_device *adev) { - volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared_cpu_addr; + volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr; struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec; uint32_t rb_bufsz, tmp; uint32_t lmi_swap_cntl; @@ -1094,8 +1098,10 @@ static int vcn_v2_0_start(struct amdgpu_device *adev) static int vcn_v2_0_stop_dpg_mode(struct amdgpu_device *adev) { + struct dpg_pause_state state = {.fw_based = VCN_DPG_STATE__UNPAUSE}; uint32_t tmp; + vcn_v2_0_pause_dpg_mode(adev, 0, &state); /* Wait for power status to be 1 */ SOC15_WAIT_ON_RREG(UVD, 0, mmUVD_POWER_STATUS, 1, UVD_POWER_STATUS__UVD_POWER_STATUS_MASK); @@ -1207,7 +1213,7 @@ static int vcn_v2_0_pause_dpg_mode(struct amdgpu_device *adev, UVD_POWER_STATUS__UVD_POWER_STATUS_MASK); if (!ret_code) { - volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared_cpu_addr; + volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr; /* pause DPG */ reg_data |= UVD_DPG_PAUSE__NJ_PAUSE_DPG_REQ_MASK; WREG32_SOC15(UVD, 0, mmUVD_DPG_PAUSE, reg_data); @@ -2003,6 +2009,7 @@ static const struct amd_ip_funcs vcn_v2_0_ip_funcs = { static const struct amdgpu_ring_funcs vcn_v2_0_dec_ring_vm_funcs = { .type = AMDGPU_RING_TYPE_VCN_DEC, .align_mask = 0xf, + .secure_submission_supported = true, .vmhub = AMDGPU_MMHUB_0, .get_rptr = vcn_v2_0_dec_ring_get_rptr, .get_wptr = vcn_v2_0_dec_ring_get_wptr, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c index 44fc4c218433..1bf672966a62 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c @@ -196,8 +196,11 @@ static int vcn_v2_5_sw_init(void *handle) return r; } - fw_shared = adev->vcn.inst[j].fw_shared_cpu_addr; + fw_shared = adev->vcn.inst[j].fw_shared.cpu_addr; fw_shared->present_flag_0 = cpu_to_le32(AMDGPU_VCN_MULTI_QUEUE_FLAG); + + if (amdgpu_vcnfw_log) + amdgpu_vcn_fwlog_init(&adev->vcn.inst[i]); } if (amdgpu_sriov_vf(adev)) { @@ -229,7 +232,7 @@ static int vcn_v2_5_sw_fini(void *handle) for (i = 0; i < adev->vcn.num_vcn_inst; i++) { if (adev->vcn.harvest_config & (1 << i)) continue; - fw_shared = adev->vcn.inst[i].fw_shared_cpu_addr; + fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr; fw_shared->present_flag_0 = 0; } drm_dev_exit(idx); @@ -423,9 +426,9 @@ static void vcn_v2_5_mc_resume(struct amdgpu_device *adev) /* non-cache window */ WREG32_SOC15(VCN, i, mmUVD_LMI_VCPU_NC0_64BIT_BAR_LOW, - lower_32_bits(adev->vcn.inst[i].fw_shared_gpu_addr)); + lower_32_bits(adev->vcn.inst[i].fw_shared.gpu_addr)); WREG32_SOC15(VCN, i, mmUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH, - upper_32_bits(adev->vcn.inst[i].fw_shared_gpu_addr)); + upper_32_bits(adev->vcn.inst[i].fw_shared.gpu_addr)); WREG32_SOC15(VCN, i, mmUVD_VCPU_NONCACHE_OFFSET0, 0); WREG32_SOC15(VCN, i, mmUVD_VCPU_NONCACHE_SIZE0, AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_fw_shared))); @@ -513,10 +516,10 @@ static void vcn_v2_5_mc_resume_dpg_mode(struct amdgpu_device *adev, int inst_idx /* non-cache window */ WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( VCN, 0, mmUVD_LMI_VCPU_NC0_64BIT_BAR_LOW), - lower_32_bits(adev->vcn.inst[inst_idx].fw_shared_gpu_addr), 0, indirect); + lower_32_bits(adev->vcn.inst[inst_idx].fw_shared.gpu_addr), 0, indirect); WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( VCN, 0, mmUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH), - upper_32_bits(adev->vcn.inst[inst_idx].fw_shared_gpu_addr), 0, indirect); + upper_32_bits(adev->vcn.inst[inst_idx].fw_shared.gpu_addr), 0, indirect); WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( VCN, 0, mmUVD_VCPU_NONCACHE_OFFSET0), 0, 0, indirect); WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( @@ -757,7 +760,7 @@ static void vcn_v2_5_enable_clock_gating(struct amdgpu_device *adev) static int vcn_v2_5_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, bool indirect) { - volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared_cpu_addr; + volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr; struct amdgpu_ring *ring; uint32_t rb_bufsz, tmp; @@ -981,7 +984,7 @@ static int vcn_v2_5_start(struct amdgpu_device *adev) vcn_v2_5_mc_resume(adev); for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { - volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[i].fw_shared_cpu_addr; + volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr; if (adev->vcn.harvest_config & (1 << i)) continue; /* VCN global tiling registers */ @@ -1403,7 +1406,7 @@ static int vcn_v2_5_pause_dpg_mode(struct amdgpu_device *adev, UVD_POWER_STATUS__UVD_POWER_STATUS_MASK); if (!ret_code) { - volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared_cpu_addr; + volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr; /* pause DPG */ reg_data |= UVD_DPG_PAUSE__NJ_PAUSE_DPG_REQ_MASK; @@ -1512,6 +1515,7 @@ static void vcn_v2_5_dec_ring_set_wptr(struct amdgpu_ring *ring) static const struct amdgpu_ring_funcs vcn_v2_5_dec_ring_vm_funcs = { .type = AMDGPU_RING_TYPE_VCN_DEC, .align_mask = 0xf, + .secure_submission_supported = true, .vmhub = AMDGPU_MMHUB_1, .get_rptr = vcn_v2_5_dec_ring_get_rptr, .get_wptr = vcn_v2_5_dec_ring_get_wptr, @@ -1542,6 +1546,7 @@ static const struct amdgpu_ring_funcs vcn_v2_5_dec_ring_vm_funcs = { static const struct amdgpu_ring_funcs vcn_v2_6_dec_ring_vm_funcs = { .type = AMDGPU_RING_TYPE_VCN_DEC, .align_mask = 0xf, + .secure_submission_supported = true, .vmhub = AMDGPU_MMHUB_0, .get_rptr = vcn_v2_5_dec_ring_get_rptr, .get_wptr = vcn_v2_5_dec_ring_get_wptr, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c index da11ceba0698..c87263ed20ec 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c @@ -25,6 +25,7 @@ #include "amdgpu.h" #include "amdgpu_vcn.h" #include "amdgpu_pm.h" +#include "amdgpu_cs.h" #include "soc15.h" #include "soc15d.h" #include "vcn_v2_0.h" @@ -213,11 +214,14 @@ static int vcn_v3_0_sw_init(void *handle) return r; } - fw_shared = adev->vcn.inst[i].fw_shared_cpu_addr; + fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr; fw_shared->present_flag_0 |= cpu_to_le32(AMDGPU_VCN_SW_RING_FLAG) | cpu_to_le32(AMDGPU_VCN_MULTI_QUEUE_FLAG) | cpu_to_le32(AMDGPU_VCN_FW_SHARED_FLAG_0_RB); fw_shared->sw_ring.is_enabled = cpu_to_le32(DEC_SW_RING_ENABLED); + + if (amdgpu_vcnfw_log) + amdgpu_vcn_fwlog_init(&adev->vcn.inst[i]); } if (amdgpu_sriov_vf(adev)) { @@ -249,7 +253,7 @@ static int vcn_v3_0_sw_fini(void *handle) if (adev->vcn.harvest_config & (1 << i)) continue; - fw_shared = adev->vcn.inst[i].fw_shared_cpu_addr; + fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr; fw_shared->present_flag_0 = 0; fw_shared->sw_ring.is_enabled = false; } @@ -295,6 +299,7 @@ static int vcn_v3_0_hw_init(void *handle) ring = &adev->vcn.inst[i].ring_dec; if (amdgpu_vcn_is_disabled_vcn(adev, VCN_DECODE_RING, i)) { ring->sched.ready = false; + ring->no_scheduler = true; dev_info(adev->dev, "ring %s is disabled by hypervisor\n", ring->name); } else { ring->wptr = 0; @@ -307,6 +312,7 @@ static int vcn_v3_0_hw_init(void *handle) ring = &adev->vcn.inst[i].ring_enc[j]; if (amdgpu_vcn_is_disabled_vcn(adev, VCN_ENCODE_RING, i)) { ring->sched.ready = false; + ring->no_scheduler = true; dev_info(adev->dev, "ring %s is disabled by hypervisor\n", ring->name); } else { ring->wptr = 0; @@ -469,9 +475,9 @@ static void vcn_v3_0_mc_resume(struct amdgpu_device *adev, int inst) /* non-cache window */ WREG32_SOC15(VCN, inst, mmUVD_LMI_VCPU_NC0_64BIT_BAR_LOW, - lower_32_bits(adev->vcn.inst[inst].fw_shared_gpu_addr)); + lower_32_bits(adev->vcn.inst[inst].fw_shared.gpu_addr)); WREG32_SOC15(VCN, inst, mmUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH, - upper_32_bits(adev->vcn.inst[inst].fw_shared_gpu_addr)); + upper_32_bits(adev->vcn.inst[inst].fw_shared.gpu_addr)); WREG32_SOC15(VCN, inst, mmUVD_VCPU_NONCACHE_OFFSET0, 0); WREG32_SOC15(VCN, inst, mmUVD_VCPU_NONCACHE_SIZE0, AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_fw_shared))); @@ -558,10 +564,10 @@ static void vcn_v3_0_mc_resume_dpg_mode(struct amdgpu_device *adev, int inst_idx /* non-cache window */ WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( VCN, inst_idx, mmUVD_LMI_VCPU_NC0_64BIT_BAR_LOW), - lower_32_bits(adev->vcn.inst[inst_idx].fw_shared_gpu_addr), 0, indirect); + lower_32_bits(adev->vcn.inst[inst_idx].fw_shared.gpu_addr), 0, indirect); WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( VCN, inst_idx, mmUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH), - upper_32_bits(adev->vcn.inst[inst_idx].fw_shared_gpu_addr), 0, indirect); + upper_32_bits(adev->vcn.inst[inst_idx].fw_shared.gpu_addr), 0, indirect); WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( VCN, inst_idx, mmUVD_VCPU_NONCACHE_OFFSET0), 0, 0, indirect); WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( @@ -923,7 +929,7 @@ static void vcn_v3_0_enable_clock_gating(struct amdgpu_device *adev, int inst) static int vcn_v3_0_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, bool indirect) { - volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared_cpu_addr; + volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr; struct amdgpu_ring *ring; uint32_t rb_bufsz, tmp; @@ -1220,7 +1226,7 @@ static int vcn_v3_0_start(struct amdgpu_device *adev) tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_RPTR_WR_EN, 1); WREG32_SOC15(VCN, i, mmUVD_RBC_RB_CNTL, tmp); - fw_shared = adev->vcn.inst[i].fw_shared_cpu_addr; + fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr; fw_shared->multi_queue.decode_queue_mode |= cpu_to_le32(FW_QUEUE_RING_RESET); /* programm the RB_BASE for ring buffer */ @@ -1611,7 +1617,7 @@ static int vcn_v3_0_pause_dpg_mode(struct amdgpu_device *adev, if (adev->ip_versions[UVD_HWIP][0] != IP_VERSION(3, 0, 33)) { /* Restore */ - fw_shared = adev->vcn.inst[inst_idx].fw_shared_cpu_addr; + fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr; fw_shared->multi_queue.encode_generalpurpose_queue_mode |= cpu_to_le32(FW_QUEUE_RING_RESET); ring = &adev->vcn.inst[inst_idx].ring_enc[0]; ring->wptr = 0; @@ -1700,7 +1706,7 @@ static void vcn_v3_0_dec_ring_set_wptr(struct amdgpu_ring *ring) if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) { /*whenever update RBC_RB_WPTR, we save the wptr in shared rb.wptr and scratch2 */ - fw_shared = adev->vcn.inst[ring->me].fw_shared_cpu_addr; + fw_shared = adev->vcn.inst[ring->me].fw_shared.cpu_addr; fw_shared->rb.wptr = lower_32_bits(ring->wptr); WREG32_SOC15(VCN, ring->me, mmUVD_SCRATCH2, lower_32_bits(ring->wptr)); @@ -1780,6 +1786,7 @@ static const struct amdgpu_ring_funcs vcn_v3_0_dec_sw_ring_vm_funcs = { .type = AMDGPU_RING_TYPE_VCN_DEC, .align_mask = 0x3f, .nop = VCN_DEC_SW_CMD_NO_OP, + .secure_submission_supported = true, .vmhub = AMDGPU_MMHUB_0, .get_rptr = vcn_v3_0_dec_ring_get_rptr, .get_wptr = vcn_v3_0_dec_ring_get_wptr, @@ -1806,21 +1813,23 @@ static const struct amdgpu_ring_funcs vcn_v3_0_dec_sw_ring_vm_funcs = { .emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper, }; -static int vcn_v3_0_limit_sched(struct amdgpu_cs_parser *p) +static int vcn_v3_0_limit_sched(struct amdgpu_cs_parser *p, + struct amdgpu_job *job) { struct drm_gpu_scheduler **scheds; /* The create msg must be in the first IB submitted */ - if (atomic_read(&p->entity->fence_seq)) + if (atomic_read(&job->base.entity->fence_seq)) return -EINVAL; scheds = p->adev->gpu_sched[AMDGPU_HW_IP_VCN_DEC] [AMDGPU_RING_PRIO_DEFAULT].sched; - drm_sched_entity_modify_sched(p->entity, scheds, 1); + drm_sched_entity_modify_sched(job->base.entity, scheds, 1); return 0; } -static int vcn_v3_0_dec_msg(struct amdgpu_cs_parser *p, uint64_t addr) +static int vcn_v3_0_dec_msg(struct amdgpu_cs_parser *p, struct amdgpu_job *job, + uint64_t addr) { struct ttm_operation_ctx ctx = { false, false }; struct amdgpu_bo_va_mapping *map; @@ -1891,7 +1900,7 @@ static int vcn_v3_0_dec_msg(struct amdgpu_cs_parser *p, uint64_t addr) if (create[0] == 0x7 || create[0] == 0x10 || create[0] == 0x11) continue; - r = vcn_v3_0_limit_sched(p); + r = vcn_v3_0_limit_sched(p, job); if (r) goto out; } @@ -1902,10 +1911,10 @@ out: } static int vcn_v3_0_ring_patch_cs_in_place(struct amdgpu_cs_parser *p, - uint32_t ib_idx) + struct amdgpu_job *job, + struct amdgpu_ib *ib) { - struct amdgpu_ring *ring = to_amdgpu_ring(p->entity->rq->sched); - struct amdgpu_ib *ib = &p->job->ibs[ib_idx]; + struct amdgpu_ring *ring = to_amdgpu_ring(job->base.sched); uint32_t msg_lo = 0, msg_hi = 0; unsigned i; int r; @@ -1915,8 +1924,8 @@ static int vcn_v3_0_ring_patch_cs_in_place(struct amdgpu_cs_parser *p, return 0; for (i = 0; i < ib->length_dw; i += 2) { - uint32_t reg = amdgpu_get_ib_value(p, ib_idx, i); - uint32_t val = amdgpu_get_ib_value(p, ib_idx, i + 1); + uint32_t reg = amdgpu_ib_get_value(ib, i); + uint32_t val = amdgpu_ib_get_value(ib, i + 1); if (reg == PACKET0(p->adev->vcn.internal.data0, 0)) { msg_lo = val; @@ -1924,7 +1933,8 @@ static int vcn_v3_0_ring_patch_cs_in_place(struct amdgpu_cs_parser *p, msg_hi = val; } else if (reg == PACKET0(p->adev->vcn.internal.cmd, 0) && val == 0) { - r = vcn_v3_0_dec_msg(p, ((u64)msg_hi) << 32 | msg_lo); + r = vcn_v3_0_dec_msg(p, job, + ((u64)msg_hi) << 32 | msg_lo); if (r) return r; } @@ -1935,6 +1945,7 @@ static int vcn_v3_0_ring_patch_cs_in_place(struct amdgpu_cs_parser *p, static const struct amdgpu_ring_funcs vcn_v3_0_dec_ring_vm_funcs = { .type = AMDGPU_RING_TYPE_VCN_DEC, .align_mask = 0xf, + .secure_submission_supported = true, .vmhub = AMDGPU_MMHUB_0, .get_rptr = vcn_v3_0_dec_ring_get_rptr, .get_wptr = vcn_v3_0_dec_ring_get_wptr, diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index 59d3fe269e7c..607f65ab39ac 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -1759,14 +1759,18 @@ static int criu_checkpoint_bos(struct kfd_process *p, goto exit; } } - if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) { + if (bo_bucket->alloc_flags + & (KFD_IOC_ALLOC_MEM_FLAGS_VRAM | KFD_IOC_ALLOC_MEM_FLAGS_GTT)) { ret = criu_get_prime_handle(&dumper_bo->tbo.base, bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_WRITABLE ? DRM_RDWR : 0, &bo_bucket->dmabuf_fd); if (ret) goto exit; + } else { + bo_bucket->dmabuf_fd = KFD_INVALID_FD; } + if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) bo_bucket->offset = KFD_MMAP_TYPE_DOORBELL | KFD_MMAP_GPU_ID(pdd->dev->id); @@ -1812,7 +1816,8 @@ static int criu_checkpoint_bos(struct kfd_process *p, exit: while (ret && bo_index--) { - if (bo_buckets[bo_index].alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) + if (bo_buckets[bo_index].alloc_flags + & (KFD_IOC_ALLOC_MEM_FLAGS_VRAM | KFD_IOC_ALLOC_MEM_FLAGS_GTT)) close_fd(bo_buckets[bo_index].dmabuf_fd); } @@ -2094,6 +2099,136 @@ exit: return ret; } +static int criu_restore_memory_of_gpu(struct kfd_process_device *pdd, + struct kfd_criu_bo_bucket *bo_bucket, + struct kfd_criu_bo_priv_data *bo_priv, + struct kgd_mem **kgd_mem) +{ + int idr_handle; + int ret; + const bool criu_resume = true; + u64 offset; + + if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) { + if (bo_bucket->size != kfd_doorbell_process_slice(pdd->dev)) + return -EINVAL; + + offset = kfd_get_process_doorbells(pdd); + } else if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP) { + /* MMIO BOs need remapped bus address */ + if (bo_bucket->size != PAGE_SIZE) { + pr_err("Invalid page size\n"); + return -EINVAL; + } + offset = pdd->dev->adev->rmmio_remap.bus_addr; + if (!offset) { + pr_err("amdgpu_amdkfd_get_mmio_remap_phys_addr failed\n"); + return -ENOMEM; + } + } else if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) { + offset = bo_priv->user_addr; + } + /* Create the BO */ + ret = amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(pdd->dev->adev, bo_bucket->addr, + bo_bucket->size, pdd->drm_priv, kgd_mem, + &offset, bo_bucket->alloc_flags, criu_resume); + if (ret) { + pr_err("Could not create the BO\n"); + return ret; + } + pr_debug("New BO created: size:0x%llx addr:0x%llx offset:0x%llx\n", + bo_bucket->size, bo_bucket->addr, offset); + + /* Restore previous IDR handle */ + pr_debug("Restoring old IDR handle for the BO"); + idr_handle = idr_alloc(&pdd->alloc_idr, *kgd_mem, bo_priv->idr_handle, + bo_priv->idr_handle + 1, GFP_KERNEL); + + if (idr_handle < 0) { + pr_err("Could not allocate idr\n"); + amdgpu_amdkfd_gpuvm_free_memory_of_gpu(pdd->dev->adev, *kgd_mem, pdd->drm_priv, + NULL); + return -ENOMEM; + } + + if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) + bo_bucket->restored_offset = KFD_MMAP_TYPE_DOORBELL | KFD_MMAP_GPU_ID(pdd->dev->id); + if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP) { + bo_bucket->restored_offset = KFD_MMAP_TYPE_MMIO | KFD_MMAP_GPU_ID(pdd->dev->id); + } else if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_GTT) { + bo_bucket->restored_offset = offset; + } else if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) { + bo_bucket->restored_offset = offset; + /* Update the VRAM usage count */ + WRITE_ONCE(pdd->vram_usage, pdd->vram_usage + bo_bucket->size); + } + return 0; +} + +static int criu_restore_bo(struct kfd_process *p, + struct kfd_criu_bo_bucket *bo_bucket, + struct kfd_criu_bo_priv_data *bo_priv) +{ + struct kfd_process_device *pdd; + struct kgd_mem *kgd_mem; + int ret; + int j; + + pr_debug("Restoring BO size:0x%llx addr:0x%llx gpu_id:0x%x flags:0x%x idr_handle:0x%x\n", + bo_bucket->size, bo_bucket->addr, bo_bucket->gpu_id, bo_bucket->alloc_flags, + bo_priv->idr_handle); + + pdd = kfd_process_device_data_by_id(p, bo_bucket->gpu_id); + if (!pdd) { + pr_err("Failed to get pdd\n"); + return -ENODEV; + } + + ret = criu_restore_memory_of_gpu(pdd, bo_bucket, bo_priv, &kgd_mem); + if (ret) + return ret; + + /* now map these BOs to GPU/s */ + for (j = 0; j < p->n_pdds; j++) { + struct kfd_dev *peer; + struct kfd_process_device *peer_pdd; + + if (!bo_priv->mapped_gpuids[j]) + break; + + peer_pdd = kfd_process_device_data_by_id(p, bo_priv->mapped_gpuids[j]); + if (!peer_pdd) + return -EINVAL; + + peer = peer_pdd->dev; + + peer_pdd = kfd_bind_process_to_device(peer, p); + if (IS_ERR(peer_pdd)) + return PTR_ERR(peer_pdd); + + ret = amdgpu_amdkfd_gpuvm_map_memory_to_gpu(peer->adev, kgd_mem, peer_pdd->drm_priv, + NULL); + if (ret) { + pr_err("Failed to map to gpu %d/%d\n", j, p->n_pdds); + return ret; + } + } + + pr_debug("map memory was successful for the BO\n"); + /* create the dmabuf object and export the bo */ + if (bo_bucket->alloc_flags + & (KFD_IOC_ALLOC_MEM_FLAGS_VRAM | KFD_IOC_ALLOC_MEM_FLAGS_GTT)) { + ret = criu_get_prime_handle(&kgd_mem->bo->tbo.base, DRM_RDWR, + &bo_bucket->dmabuf_fd); + if (ret) + return ret; + } else { + bo_bucket->dmabuf_fd = KFD_INVALID_FD; + } + + return 0; +} + static int criu_restore_bos(struct kfd_process *p, struct kfd_ioctl_criu_args *args, uint64_t *priv_offset, @@ -2101,9 +2236,7 @@ static int criu_restore_bos(struct kfd_process *p, { struct kfd_criu_bo_bucket *bo_buckets = NULL; struct kfd_criu_bo_priv_data *bo_privs = NULL; - const bool criu_resume = true; - bool flush_tlbs = false; - int ret = 0, j = 0; + int ret = 0; uint32_t i = 0; if (*priv_offset + (args->num_bos * sizeof(*bo_privs)) > max_priv_data_size) @@ -2141,178 +2274,13 @@ static int criu_restore_bos(struct kfd_process *p, /* Create and map new BOs */ for (; i < args->num_bos; i++) { - struct kfd_criu_bo_bucket *bo_bucket; - struct kfd_criu_bo_priv_data *bo_priv; - struct kfd_dev *dev; - struct kfd_process_device *pdd; - struct kgd_mem *kgd_mem; - void *mem; - u64 offset; - int idr_handle; - - bo_bucket = &bo_buckets[i]; - bo_priv = &bo_privs[i]; - - pr_debug("kfd restore ioctl - bo_bucket[%d]:\n", i); - pr_debug("size = 0x%llx, bo_addr = 0x%llx bo_offset = 0x%llx\n" - "gpu_id = 0x%x alloc_flags = 0x%x\n" - "idr_handle = 0x%x\n", - bo_bucket->size, - bo_bucket->addr, - bo_bucket->offset, - bo_bucket->gpu_id, - bo_bucket->alloc_flags, - bo_priv->idr_handle); - - pdd = kfd_process_device_data_by_id(p, bo_bucket->gpu_id); - if (!pdd) { - pr_err("Failed to get pdd\n"); - ret = -ENODEV; - goto exit; - } - dev = pdd->dev; - - if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) { - pr_debug("restore ioctl: KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL\n"); - if (bo_bucket->size != kfd_doorbell_process_slice(dev)) { - ret = -EINVAL; - goto exit; - } - offset = kfd_get_process_doorbells(pdd); - } else if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP) { - /* MMIO BOs need remapped bus address */ - pr_debug("restore ioctl :KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP\n"); - if (bo_bucket->size != PAGE_SIZE) { - pr_err("Invalid page size\n"); - ret = -EINVAL; - goto exit; - } - offset = dev->adev->rmmio_remap.bus_addr; - if (!offset) { - pr_err("amdgpu_amdkfd_get_mmio_remap_phys_addr failed\n"); - ret = -ENOMEM; - goto exit; - } - } else if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) { - offset = bo_priv->user_addr; - } - /* Create the BO */ - ret = amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(dev->adev, - bo_bucket->addr, - bo_bucket->size, - pdd->drm_priv, - (struct kgd_mem **) &mem, - &offset, - bo_bucket->alloc_flags, - criu_resume); - if (ret) { - pr_err("Could not create the BO\n"); - ret = -ENOMEM; - goto exit; - } - pr_debug("New BO created: size = 0x%llx, bo_addr = 0x%llx bo_offset = 0x%llx\n", - bo_bucket->size, bo_bucket->addr, offset); - - /* Restore previuos IDR handle */ - pr_debug("Restoring old IDR handle for the BO"); - idr_handle = idr_alloc(&pdd->alloc_idr, mem, - bo_priv->idr_handle, - bo_priv->idr_handle + 1, GFP_KERNEL); - - if (idr_handle < 0) { - pr_err("Could not allocate idr\n"); - amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->adev, - (struct kgd_mem *)mem, - pdd->drm_priv, NULL); - ret = -ENOMEM; - goto exit; - } - - if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) - bo_bucket->restored_offset = KFD_MMAP_TYPE_DOORBELL | - KFD_MMAP_GPU_ID(pdd->dev->id); - if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP) { - bo_bucket->restored_offset = KFD_MMAP_TYPE_MMIO | - KFD_MMAP_GPU_ID(pdd->dev->id); - } else if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_GTT) { - bo_bucket->restored_offset = offset; - pr_debug("updating offset for GTT\n"); - } else if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) { - bo_bucket->restored_offset = offset; - /* Update the VRAM usage count */ - WRITE_ONCE(pdd->vram_usage, pdd->vram_usage + bo_bucket->size); - pr_debug("updating offset for VRAM\n"); - } - - /* now map these BOs to GPU/s */ - for (j = 0; j < p->n_pdds; j++) { - struct kfd_dev *peer; - struct kfd_process_device *peer_pdd; - bool table_freed = false; - - if (!bo_priv->mapped_gpuids[j]) - break; - - peer_pdd = kfd_process_device_data_by_id(p, bo_priv->mapped_gpuids[j]); - if (!peer_pdd) { - ret = -EINVAL; - goto exit; - } - peer = peer_pdd->dev; - - peer_pdd = kfd_bind_process_to_device(peer, p); - if (IS_ERR(peer_pdd)) { - ret = PTR_ERR(peer_pdd); - goto exit; - } - pr_debug("map mem in restore ioctl -> 0x%llx\n", - ((struct kgd_mem *)mem)->va); - ret = amdgpu_amdkfd_gpuvm_map_memory_to_gpu(peer->adev, - (struct kgd_mem *)mem, peer_pdd->drm_priv, &table_freed); - if (ret) { - pr_err("Failed to map to gpu %d/%d\n", j, p->n_pdds); - goto exit; - } - if (table_freed) - flush_tlbs = true; - } - - ret = amdgpu_amdkfd_gpuvm_sync_memory(dev->adev, - (struct kgd_mem *) mem, true); + ret = criu_restore_bo(p, &bo_buckets[i], &bo_privs[i]); if (ret) { - pr_debug("Sync memory failed, wait interrupted by user signal\n"); + pr_debug("Failed to restore BO[%d] ret%d\n", i, ret); goto exit; } - - pr_debug("map memory was successful for the BO\n"); - /* create the dmabuf object and export the bo */ - kgd_mem = (struct kgd_mem *)mem; - if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) { - ret = criu_get_prime_handle(&kgd_mem->bo->tbo.base, - DRM_RDWR, - &bo_bucket->dmabuf_fd); - if (ret) - goto exit; - } } /* done */ - if (flush_tlbs) { - /* Flush TLBs after waiting for the page table updates to complete */ - for (j = 0; j < p->n_pdds; j++) { - struct kfd_dev *peer; - struct kfd_process_device *pdd = p->pdds[j]; - struct kfd_process_device *peer_pdd; - - peer = kfd_device_by_id(pdd->dev->id); - if (WARN_ON_ONCE(!peer)) - continue; - peer_pdd = kfd_get_process_device_data(peer, p); - if (WARN_ON_ONCE(!peer_pdd)) - continue; - kfd_flush_tlb(peer_pdd, TLB_FLUSH_LEGACY); - } - } - /* Copy only the buckets back so user can read bo_buckets[N].restored_offset */ ret = copy_to_user((void __user *)args->bos, bo_buckets, @@ -2322,7 +2290,8 @@ static int criu_restore_bos(struct kfd_process *p, exit: while (ret && i--) { - if (bo_buckets[i].alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) + if (bo_buckets[i].alloc_flags + & (KFD_IOC_ALLOC_MEM_FLAGS_VRAM | KFD_IOC_ALLOC_MEM_FLAGS_GTT)) close_fd(bo_buckets[i].dmabuf_fd); } kvfree(bo_buckets); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 1cd2ea536bd0..acf4f7975850 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -500,6 +500,11 @@ static int dbgdev_wave_reset_wavefronts(struct kfd_dev *dev, struct kfd_process pr_debug("Killing all process wavefronts\n"); + if (!dev->kfd2kgd->get_atc_vmid_pasid_mapping_info) { + pr_err("no vmid pasid mapping supported \n"); + return -EOPNOTSUPP; + } + /* Scan all registers in the range ATC_VMID8_PASID_MAPPING .. * ATC_VMID15_PASID_MAPPING * to check which VMID the current process is mapped to. diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c index 7f689094be5a..3a29d857640b 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c @@ -638,6 +638,22 @@ out_oom: return r; } +/** + * svm_migrate_vma_to_ram - migrate range inside one vma from device to system + * + * @adev: amdgpu device to migrate from + * @prange: svm range structure + * @vma: vm_area_struct that range [start, end] belongs to + * @start: range start virtual address in pages + * @end: range end virtual address in pages + * + * Context: Process context, caller hold mmap read lock, prange->migrate_mutex + * + * Return: + * 0 - success with all pages migrated + * negative values - indicate error + * positive values - partial migration, number of pages not migrated + */ static long svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange, struct vm_area_struct *vma, uint64_t start, uint64_t end) @@ -709,8 +725,6 @@ out: pdd = svm_range_get_pdd_by_adev(prange, adev); if (pdd) WRITE_ONCE(pdd->page_out, pdd->page_out + cpages); - - return upages; } return r ? r : upages; } @@ -720,7 +734,7 @@ out: * @prange: range structure * @mm: process mm, use current->mm if NULL * - * Context: Process context, caller hold mmap read lock, svms lock, prange lock + * Context: Process context, caller hold mmap read lock, prange->migrate_mutex * * Return: * 0 - OK, otherwise error code @@ -759,13 +773,16 @@ int svm_migrate_vram_to_ram(struct svm_range *prange, struct mm_struct *mm) unsigned long next; vma = find_vma(mm, addr); - if (!vma || addr < vma->vm_start) + if (!vma || addr < vma->vm_start) { + pr_debug("failed to find vma for prange %p\n", prange); + r = -EFAULT; break; + } next = min(vma->vm_end, end); r = svm_migrate_vma_to_ram(adev, prange, vma, addr, next); if (r < 0) { - pr_debug("failed %ld to migrate\n", r); + pr_debug("failed %ld to migrate prange %p\n", r, prange); break; } else { upages += r; @@ -773,7 +790,7 @@ int svm_migrate_vram_to_ram(struct svm_range *prange, struct mm_struct *mm) addr = next; } - if (!upages) { + if (r >= 0 && !upages) { svm_range_vram_node_free(prange); prange->actual_loc = 0; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c index f9eafc796e70..e4beebb1c80a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c @@ -82,7 +82,8 @@ static ssize_t kfd_smi_ev_read(struct file *filep, char __user *user, struct kfd_smi_client *client = filep->private_data; unsigned char *buf; - buf = kmalloc_array(MAX_KFIFO_SIZE, sizeof(*buf), GFP_KERNEL); + size = min_t(size_t, size, MAX_KFIFO_SIZE); + buf = kmalloc(size, GFP_KERNEL); if (!buf) return -ENOMEM; @@ -96,7 +97,7 @@ static ssize_t kfd_smi_ev_read(struct file *filep, char __user *user, ret = -EAGAIN; goto ret_err; } - to_copy = min3(size, sizeof(buf), to_copy); + to_copy = min(size, to_copy); ret = kfifo_out(&client->fifo, buf, to_copy); spin_unlock(&client->lock); if (ret <= 0) { @@ -175,22 +176,29 @@ static void add_event_to_kfifo(struct kfd_dev *dev, unsigned int smi_event, rcu_read_unlock(); } -void kfd_smi_event_update_gpu_reset(struct kfd_dev *dev, bool post_reset) +__printf(3, 4) +static void kfd_smi_event_add(struct kfd_dev *dev, unsigned int event, + char *fmt, ...) { - /* - * GpuReset msg = Reset seq number (incremented for - * every reset message sent before GPU reset). - * 1 byte event + 1 byte space + 8 bytes seq num + - * 1 byte \n + 1 byte \0 = 12 - */ - char fifo_in[12]; + char fifo_in[KFD_SMI_EVENT_MSG_SIZE]; int len; - unsigned int event; + va_list args; if (list_empty(&dev->smi_clients)) return; - memset(fifo_in, 0x0, sizeof(fifo_in)); + len = snprintf(fifo_in, sizeof(fifo_in), "%x ", event); + + va_start(args, fmt); + len += vsnprintf(fifo_in + len, sizeof(fifo_in) - len, fmt, args); + va_end(args); + + add_event_to_kfifo(dev, event, fifo_in, len); +} + +void kfd_smi_event_update_gpu_reset(struct kfd_dev *dev, bool post_reset) +{ + unsigned int event; if (post_reset) { event = KFD_SMI_EVENT_GPU_POST_RESET; @@ -198,48 +206,20 @@ void kfd_smi_event_update_gpu_reset(struct kfd_dev *dev, bool post_reset) event = KFD_SMI_EVENT_GPU_PRE_RESET; ++(dev->reset_seq_num); } - - len = snprintf(fifo_in, sizeof(fifo_in), "%x %x\n", event, - dev->reset_seq_num); - - add_event_to_kfifo(dev, event, fifo_in, len); + kfd_smi_event_add(dev, event, "%x\n", dev->reset_seq_num); } void kfd_smi_event_update_thermal_throttling(struct kfd_dev *dev, uint64_t throttle_bitmask) { - /* - * ThermalThrottle msg = throttle_bitmask(8): - * thermal_interrupt_count(16): - * 1 byte event + 1 byte space + 16 byte throttle_bitmask + - * 1 byte : + 16 byte thermal_interupt_counter + 1 byte \n + - * 1 byte \0 = 37 - */ - char fifo_in[37]; - int len; - - if (list_empty(&dev->smi_clients)) - return; - - len = snprintf(fifo_in, sizeof(fifo_in), "%x %llx:%llx\n", - KFD_SMI_EVENT_THERMAL_THROTTLE, throttle_bitmask, - amdgpu_dpm_get_thermal_throttling_counter(dev->adev)); - - add_event_to_kfifo(dev, KFD_SMI_EVENT_THERMAL_THROTTLE, fifo_in, len); + kfd_smi_event_add(dev, KFD_SMI_EVENT_THERMAL_THROTTLE, "%llx:%llx\n", + throttle_bitmask, + amdgpu_dpm_get_thermal_throttling_counter(dev->adev)); } void kfd_smi_event_update_vmfault(struct kfd_dev *dev, uint16_t pasid) { struct amdgpu_task_info task_info; - /* VmFault msg = (hex)uint32_pid(8) + :(1) + task name(16) = 25 */ - /* 1 byte event + 1 byte space + 25 bytes msg + 1 byte \n + - * 1 byte \0 = 29 - */ - char fifo_in[29]; - int len; - - if (list_empty(&dev->smi_clients)) - return; memset(&task_info, 0, sizeof(struct amdgpu_task_info)); amdgpu_vm_get_task_info(dev->adev, pasid, &task_info); @@ -247,10 +227,8 @@ void kfd_smi_event_update_vmfault(struct kfd_dev *dev, uint16_t pasid) if (!task_info.pid) return; - len = snprintf(fifo_in, sizeof(fifo_in), "%x %x:%s\n", KFD_SMI_EVENT_VMFAULT, - task_info.pid, task_info.task_name); - - add_event_to_kfifo(dev, KFD_SMI_EVENT_VMFAULT, fifo_in, len); + kfd_smi_event_add(dev, KFD_SMI_EVENT_VMFAULT, "%x:%s\n", + task_info.pid, task_info.task_name); } int kfd_smi_event_open(struct kfd_dev *dev, uint32_t *fd) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c index b71d47afd243..3b8856b4cece 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c @@ -1629,6 +1629,7 @@ retry_flush_work: static void svm_range_restore_work(struct work_struct *work) { struct delayed_work *dwork = to_delayed_work(work); + struct amdkfd_process_info *process_info; struct svm_range_list *svms; struct svm_range *prange; struct kfd_process *p; @@ -1645,6 +1646,7 @@ static void svm_range_restore_work(struct work_struct *work) pr_debug("restore svm ranges\n"); p = container_of(svms, struct kfd_process, svms); + process_info = p->kgd_process_info; /* Keep mm reference when svm_range_validate_and_map ranges */ mm = get_task_mm(p->lead_thread); @@ -1653,6 +1655,7 @@ static void svm_range_restore_work(struct work_struct *work) return; } + mutex_lock(&process_info->lock); svm_range_list_lock_and_flush_work(svms, mm); mutex_lock(&svms->lock); @@ -1705,6 +1708,7 @@ static void svm_range_restore_work(struct work_struct *work) out_reschedule: mutex_unlock(&svms->lock); mmap_write_unlock(mm); + mutex_unlock(&process_info->lock); mmput(mm); /* If validation failed, reschedule another attempt */ @@ -3151,6 +3155,7 @@ static void svm_range_evict_svm_bo_worker(struct work_struct *work) struct svm_range_bo *svm_bo; struct kfd_process *p; struct mm_struct *mm; + int r = 0; svm_bo = container_of(work, struct svm_range_bo, eviction_work); if (!svm_bo_ref_unless_zero(svm_bo)) @@ -3166,7 +3171,7 @@ static void svm_range_evict_svm_bo_worker(struct work_struct *work) mmap_read_lock(mm); spin_lock(&svm_bo->list_lock); - while (!list_empty(&svm_bo->range_list)) { + while (!list_empty(&svm_bo->range_list) && !r) { struct svm_range *prange = list_first_entry(&svm_bo->range_list, struct svm_range, svm_bo_list); @@ -3180,15 +3185,18 @@ static void svm_range_evict_svm_bo_worker(struct work_struct *work) mutex_lock(&prange->migrate_mutex); do { - svm_migrate_vram_to_ram(prange, + r = svm_migrate_vram_to_ram(prange, svm_bo->eviction_fence->mm); - } while (prange->actual_loc && --retries); - WARN(prange->actual_loc, "Migration failed during eviction"); + } while (!r && prange->actual_loc && --retries); - mutex_lock(&prange->lock); - prange->svm_bo = NULL; - mutex_unlock(&prange->lock); + if (!r && prange->actual_loc) + pr_info_once("Migration failed during eviction"); + if (!prange->actual_loc) { + mutex_lock(&prange->lock); + prange->svm_bo = NULL; + mutex_unlock(&prange->lock); + } mutex_unlock(&prange->migrate_mutex); spin_lock(&svm_bo->list_lock); @@ -3197,10 +3205,11 @@ static void svm_range_evict_svm_bo_worker(struct work_struct *work) mmap_read_unlock(mm); dma_fence_signal(&svm_bo->eviction_fence->base); + /* This is the last reference to svm_bo, after svm_range_vram_node_free * has been called in svm_migrate_vram_to_ram */ - WARN_ONCE(kref_read(&svm_bo->kref) != 1, "This was not the last reference\n"); + WARN_ONCE(!r && kref_read(&svm_bo->kref) != 1, "This was not the last reference\n"); svm_range_bo_unref(svm_bo); } @@ -3209,6 +3218,7 @@ svm_range_set_attr(struct kfd_process *p, struct mm_struct *mm, uint64_t start, uint64_t size, uint32_t nattr, struct kfd_ioctl_svm_attribute *attrs) { + struct amdkfd_process_info *process_info = p->kgd_process_info; struct list_head update_list; struct list_head insert_list; struct list_head remove_list; @@ -3226,6 +3236,8 @@ svm_range_set_attr(struct kfd_process *p, struct mm_struct *mm, svms = &p->svms; + mutex_lock(&process_info->lock); + svm_range_list_lock_and_flush_work(svms, mm); r = svm_range_is_valid(p, start, size); @@ -3300,6 +3312,8 @@ out_unlock_range: mutex_unlock(&svms->lock); mmap_read_unlock(mm); out: + mutex_unlock(&process_info->lock); + pr_debug("pasid 0x%x svms 0x%p [0x%llx 0x%llx] done, r=%d\n", p->pasid, &p->svms, start, start + size - 1, r); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index e1d3db3fe8de..b30656959fd8 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -1481,6 +1481,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) case IP_VERSION(3, 1, 2): case IP_VERSION(3, 1, 3): case IP_VERSION(3, 1, 5): + case IP_VERSION(3, 1, 6): init_data.flags.gpu_vm_support = true; break; default: @@ -2633,10 +2634,13 @@ static int dm_resume(void *handle) * before the 0 streams commit. * * DC expects that link encoder assignments are *not* valid - * when committing a state, so as a workaround it needs to be - * cleared here. + * when committing a state, so as a workaround we can copy + * off of the current state. + * + * We lose the previous assignments, but we had already + * commit 0 streams anyway. */ - link_enc_cfg_init(dm->dc, dc_state); + link_enc_cfg_copy(adev->dm.dc->current_state, dc_state); if (dc_enable_dmub_notifications(adev->dm.dc)) amdgpu_dm_outbox_init(adev); @@ -6356,7 +6360,7 @@ get_highest_refresh_rate_mode(struct amdgpu_dm_connector *aconnector, } } - aconnector->freesync_vid_base = *m_pref; + drm_mode_copy(&aconnector->freesync_vid_base, m_pref); return m_pref; } @@ -6469,8 +6473,8 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, recalculate_timing = is_freesync_video_mode(&mode, aconnector); if (recalculate_timing) { freesync_mode = get_highest_refresh_rate_mode(aconnector, false); - saved_mode = mode; - mode = *freesync_mode; + drm_mode_copy(&saved_mode, &mode); + drm_mode_copy(&mode, freesync_mode); } else { decide_crtc_timing_for_drm_display_mode( &mode, preferred_mode, scale); @@ -10177,27 +10181,27 @@ static bool is_timing_unchanged_for_freesync(struct drm_crtc_state *old_crtc_state, struct drm_crtc_state *new_crtc_state) { - struct drm_display_mode old_mode, new_mode; + const struct drm_display_mode *old_mode, *new_mode; if (!old_crtc_state || !new_crtc_state) return false; - old_mode = old_crtc_state->mode; - new_mode = new_crtc_state->mode; - - if (old_mode.clock == new_mode.clock && - old_mode.hdisplay == new_mode.hdisplay && - old_mode.vdisplay == new_mode.vdisplay && - old_mode.htotal == new_mode.htotal && - old_mode.vtotal != new_mode.vtotal && - old_mode.hsync_start == new_mode.hsync_start && - old_mode.vsync_start != new_mode.vsync_start && - old_mode.hsync_end == new_mode.hsync_end && - old_mode.vsync_end != new_mode.vsync_end && - old_mode.hskew == new_mode.hskew && - old_mode.vscan == new_mode.vscan && - (old_mode.vsync_end - old_mode.vsync_start) == - (new_mode.vsync_end - new_mode.vsync_start)) + old_mode = &old_crtc_state->mode; + new_mode = &new_crtc_state->mode; + + if (old_mode->clock == new_mode->clock && + old_mode->hdisplay == new_mode->hdisplay && + old_mode->vdisplay == new_mode->vdisplay && + old_mode->htotal == new_mode->htotal && + old_mode->vtotal != new_mode->vtotal && + old_mode->hsync_start == new_mode->hsync_start && + old_mode->vsync_start != new_mode->vsync_start && + old_mode->hsync_end == new_mode->hsync_end && + old_mode->vsync_end != new_mode->vsync_end && + old_mode->hskew == new_mode->hskew && + old_mode->vscan == new_mode->vscan && + (old_mode->vsync_end - old_mode->vsync_start) == + (new_mode->vsync_end - new_mode->vsync_start)) return true; return false; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c index 777210811311..da17ece1a2c5 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c @@ -243,6 +243,7 @@ static ssize_t dp_link_settings_write(struct file *f, const char __user *buf, { struct amdgpu_dm_connector *connector = file_inode(f)->i_private; struct dc_link *link = connector->dc_link; + struct amdgpu_device *adev = drm_to_adev(connector->base.dev); struct dc *dc = (struct dc *)link->dc; struct dc_link_settings prefer_link_settings; char *wr_buf = NULL; @@ -302,6 +303,9 @@ static ssize_t dp_link_settings_write(struct file *f, const char __user *buf, if (!valid_input) { kfree(wr_buf); DRM_DEBUG_DRIVER("Invalid Input value No HW will be programmed\n"); + mutex_lock(&adev->dm.dc_lock); + dc_link_set_preferred_training_settings(dc, NULL, NULL, link, false); + mutex_unlock(&adev->dm.dc_lock); return size; } @@ -313,7 +317,9 @@ static ssize_t dp_link_settings_write(struct file *f, const char __user *buf, prefer_link_settings.lane_count = param[0]; prefer_link_settings.link_rate = param[1]; - dc_link_set_preferred_training_settings(dc, &prefer_link_settings, NULL, link, true); + mutex_lock(&adev->dm.dc_lock); + dc_link_set_preferred_training_settings(dc, &prefer_link_settings, NULL, link, false); + mutex_unlock(&adev->dm.dc_lock); kfree(wr_buf); return size; @@ -2883,7 +2889,9 @@ static ssize_t edp_ilr_write(struct file *f, const char __user *buf, kfree(wr_buf); DRM_DEBUG_DRIVER("Invalid Input value. No HW will be programmed\n"); prefer_link_settings.use_link_rate_set = false; + mutex_lock(&adev->dm.dc_lock); dc_link_set_preferred_training_settings(dc, NULL, NULL, link, false); + mutex_unlock(&adev->dm.dc_lock); return size; } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 740435ae3997..31ac1fce36f8 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -1198,11 +1198,11 @@ void pre_validate_dsc(struct drm_atomic_state *state, struct dc_state *local_dc_state = NULL; if (!is_dsc_precompute_needed(state)) { - DRM_DEBUG_DRIVER("DSC precompute is not needed.\n"); + DRM_INFO_ONCE("DSC precompute is not needed.\n"); return; } if (dm_atomic_get_state(state, dm_state_ptr)) { - DRM_DEBUG_DRIVER("dm_atomic_get_state() failed\n"); + DRM_INFO_ONCE("dm_atomic_get_state() failed\n"); return; } dm_state = *dm_state_ptr; @@ -1245,7 +1245,7 @@ void pre_validate_dsc(struct drm_atomic_state *state, } if (!pre_compute_mst_dsc_configs_for_state(state, local_dc_state, vars)) { - DRM_DEBUG_DRIVER("pre_compute_mst_dsc_configs_for_state() failed\n"); + DRM_INFO_ONCE("pre_compute_mst_dsc_configs_for_state() failed\n"); goto clean_exit; } @@ -1258,7 +1258,7 @@ void pre_validate_dsc(struct drm_atomic_state *state, if (local_dc_state->streams[i] && is_timing_changed(stream, local_dc_state->streams[i])) { - DRM_DEBUG_DRIVER("crtc[%d] needs mode_changed\n", i); + DRM_INFO_ONCE("crtc[%d] needs mode_changed\n", i); } else { int ind = find_crtc_index_in_state_by_stream(state, stream); diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table.c b/drivers/gpu/drm/amd/display/dc/bios/command_table.c index ad13e4e36d77..0e36cd800fc9 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table.c @@ -456,7 +456,7 @@ static enum bp_result transmitter_control_v2( if ((CONNECTOR_ID_DUAL_LINK_DVII == connector_id) || (CONNECTOR_ID_DUAL_LINK_DVID == connector_id)) /* on INIT this bit should be set according to the - * phisycal connector + * physical connector * Bit0: dual link connector flag * =0 connector is single link connector * =1 connector is dual link connector @@ -468,7 +468,7 @@ static enum bp_result transmitter_control_v2( cpu_to_le16((uint8_t)cntl->connector_obj_id.id); break; case TRANSMITTER_CONTROL_SET_VOLTAGE_AND_PREEMPASIS: - /* votage swing and pre-emphsis */ + /* voltage swing and pre-emphsis */ params.asMode.ucLaneSel = (uint8_t)cntl->lane_select; params.asMode.ucLaneSet = (uint8_t)cntl->lane_settings; break; @@ -2120,7 +2120,7 @@ static enum bp_result program_clock_v5( memset(¶ms, 0, sizeof(params)); if (!bp->cmd_helper->clock_source_id_to_atom( bp_params->pll_id, &atom_pll_id)) { - BREAK_TO_DEBUGGER(); /* Invalid Inpute!! */ + BREAK_TO_DEBUGGER(); /* Invalid Input!! */ return BP_RESULT_BADINPUT; } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c index 220682e45b8d..5ed6a93d1708 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c @@ -25,30 +25,23 @@ #include "dccg.h" #include "clk_mgr_internal.h" - #include "dcn30_clk_mgr_smu_msg.h" #include "dcn20/dcn20_clk_mgr.h" #include "dce100/dce_clk_mgr.h" +#include "dcn30/dcn30_clk_mgr.h" #include "reg_helper.h" #include "core_types.h" #include "dm_helpers.h" - #include "atomfirmware.h" - - #include "sienna_cichlid_ip_offset.h" #include "dcn/dcn_3_0_0_offset.h" #include "dcn/dcn_3_0_0_sh_mask.h" - #include "nbio/nbio_7_4_offset.h" - #include "dpcs/dpcs_3_0_0_offset.h" #include "dpcs/dpcs_3_0_0_sh_mask.h" - #include "mmhub/mmhub_2_0_0_offset.h" #include "mmhub/mmhub_2_0_0_sh_mask.h" -/*we don't have clk folder yet*/ -#include "dcn30/dcn30_clk_mgr.h" +#include "dcn30_smu11_driver_if.h" #undef FN #define FN(reg_name, field_name) \ @@ -83,7 +76,7 @@ static const struct clk_mgr_mask clk_mgr_mask = { /* Query SMU for all clock states for a particular clock */ -static void dcn3_init_single_clock(struct clk_mgr_internal *clk_mgr, PPCLK_e clk, unsigned int *entry_0, unsigned int *num_levels) +static void dcn3_init_single_clock(struct clk_mgr_internal *clk_mgr, uint32_t clk, unsigned int *entry_0, unsigned int *num_levels) { unsigned int i; char *entry_i = (char *)entry_0; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c index 8ecc708bcd9e..bfc960579760 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c @@ -29,6 +29,7 @@ #include "clk_mgr_internal.h" #include "reg_helper.h" #include "dalsmc.h" +#include "dcn30_smu11_driver_if.h" #define mmDAL_MSG_REG 0x1628A #define mmDAL_ARG_REG 0x16273 @@ -197,7 +198,7 @@ void dcn30_smu_transfer_wm_table_dram_2_smu(struct clk_mgr_internal *clk_mgr) } /* Returns the actual frequency that was set in MHz, 0 on failure */ -unsigned int dcn30_smu_set_hard_min_by_freq(struct clk_mgr_internal *clk_mgr, PPCLK_e clk, uint16_t freq_mhz) +unsigned int dcn30_smu_set_hard_min_by_freq(struct clk_mgr_internal *clk_mgr, uint32_t clk, uint16_t freq_mhz) { uint32_t response = 0; @@ -215,7 +216,7 @@ unsigned int dcn30_smu_set_hard_min_by_freq(struct clk_mgr_internal *clk_mgr, PP } /* Returns the actual frequency that was set in MHz, 0 on failure */ -unsigned int dcn30_smu_set_hard_max_by_freq(struct clk_mgr_internal *clk_mgr, PPCLK_e clk, uint16_t freq_mhz) +unsigned int dcn30_smu_set_hard_max_by_freq(struct clk_mgr_internal *clk_mgr, uint32_t clk, uint16_t freq_mhz) { uint32_t response = 0; @@ -246,7 +247,7 @@ unsigned int dcn30_smu_set_hard_max_by_freq(struct clk_mgr_internal *clk_mgr, PP * * Returns 0 on failure */ -unsigned int dcn30_smu_get_dpm_freq_by_index(struct clk_mgr_internal *clk_mgr, PPCLK_e clk, uint8_t dpm_level) +unsigned int dcn30_smu_get_dpm_freq_by_index(struct clk_mgr_internal *clk_mgr, uint32_t clk, uint8_t dpm_level) { uint32_t response = 0; @@ -264,7 +265,7 @@ unsigned int dcn30_smu_get_dpm_freq_by_index(struct clk_mgr_internal *clk_mgr, P } /* Returns the max DPM frequency in DC mode in MHz, 0 on failure */ -unsigned int dcn30_smu_get_dc_mode_max_dpm_freq(struct clk_mgr_internal *clk_mgr, PPCLK_e clk) +unsigned int dcn30_smu_get_dc_mode_max_dpm_freq(struct clk_mgr_internal *clk_mgr, uint32_t clk) { uint32_t response = 0; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.h index dd2640a3ce5d..ca9f5296be94 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.h +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.h @@ -28,65 +28,6 @@ #include "core_types.h" -#define SMU11_DRIVER_IF_VERSION 0x1F - -typedef enum { - PPCLK_GFXCLK = 0, - PPCLK_SOCCLK, - PPCLK_UCLK, - PPCLK_FCLK, - PPCLK_DCLK_0, - PPCLK_VCLK_0, - PPCLK_DCLK_1, - PPCLK_VCLK_1, - PPCLK_DCEFCLK, - PPCLK_DISPCLK, - PPCLK_PIXCLK, - PPCLK_PHYCLK, - PPCLK_DTBCLK, - PPCLK_COUNT, -} PPCLK_e; - -typedef struct { - uint16_t MinClock; // This is either DCEFCLK or SOCCLK (in MHz) - uint16_t MaxClock; // This is either DCEFCLK or SOCCLK (in MHz) - uint16_t MinUclk; - uint16_t MaxUclk; - - uint8_t WmSetting; - uint8_t Flags; - uint8_t Padding[2]; - -} WatermarkRowGeneric_t; - -#define NUM_WM_RANGES 4 - -typedef enum { - WM_SOCCLK = 0, - WM_DCEFCLK, - WM_COUNT, -} WM_CLOCK_e; - -typedef enum { - WATERMARKS_CLOCK_RANGE = 0, - WATERMARKS_DUMMY_PSTATE, - WATERMARKS_MALL, - WATERMARKS_COUNT, -} WATERMARKS_FLAGS_e; - -typedef struct { - // Watermarks - WatermarkRowGeneric_t WatermarkRow[WM_COUNT][NUM_WM_RANGES]; -} Watermarks_t; - -typedef struct { - Watermarks_t Watermarks; - - uint32_t MmHubPadding[8]; // SMU internal use -} WatermarksExternal_t; - -#define TABLE_WATERMARKS 1 - struct clk_mgr_internal; bool dcn30_smu_test_message(struct clk_mgr_internal *clk_mgr, uint32_t input); @@ -97,10 +38,10 @@ void dcn30_smu_set_dram_addr_high(struct clk_mgr_internal *clk_mgr, uint void dcn30_smu_set_dram_addr_low(struct clk_mgr_internal *clk_mgr, uint32_t addr_low); void dcn30_smu_transfer_wm_table_smu_2_dram(struct clk_mgr_internal *clk_mgr); void dcn30_smu_transfer_wm_table_dram_2_smu(struct clk_mgr_internal *clk_mgr); -unsigned int dcn30_smu_set_hard_min_by_freq(struct clk_mgr_internal *clk_mgr, PPCLK_e clk, uint16_t freq_mhz); -unsigned int dcn30_smu_set_hard_max_by_freq(struct clk_mgr_internal *clk_mgr, PPCLK_e clk, uint16_t freq_mhz); -unsigned int dcn30_smu_get_dpm_freq_by_index(struct clk_mgr_internal *clk_mgr, PPCLK_e clk, uint8_t dpm_level); -unsigned int dcn30_smu_get_dc_mode_max_dpm_freq(struct clk_mgr_internal *clk_mgr, PPCLK_e clk); +unsigned int dcn30_smu_set_hard_min_by_freq(struct clk_mgr_internal *clk_mgr, uint32_t clk, uint16_t freq_mhz); +unsigned int dcn30_smu_set_hard_max_by_freq(struct clk_mgr_internal *clk_mgr, uint32_t clk, uint16_t freq_mhz); +unsigned int dcn30_smu_get_dpm_freq_by_index(struct clk_mgr_internal *clk_mgr, uint32_t clk, uint8_t dpm_level); +unsigned int dcn30_smu_get_dc_mode_max_dpm_freq(struct clk_mgr_internal *clk_mgr, uint32_t clk); void dcn30_smu_set_min_deep_sleep_dcef_clk(struct clk_mgr_internal *clk_mgr, uint32_t freq_mhz); void dcn30_smu_set_num_of_displays(struct clk_mgr_internal *clk_mgr, uint32_t num_displays); void dcn30_smu_set_display_refresh_from_mall(struct clk_mgr_internal *clk_mgr, bool enable, uint8_t cache_timer_delay, uint8_t cache_timer_scale); diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_smu11_driver_if.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_smu11_driver_if.h new file mode 100644 index 000000000000..8ea8ee57b39f --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_smu11_driver_if.h @@ -0,0 +1,74 @@ +// This is a stripped-down version of the smu11_driver_if.h file for the relevant DAL interfaces. + +#define SMU11_DRIVER_IF_VERSION 0x40 + +//Only Clks that have DPM descriptors are listed here +typedef enum { + PPCLK_GFXCLK = 0, + PPCLK_SOCCLK, + PPCLK_UCLK, + PPCLK_FCLK, + PPCLK_DCLK_0, + PPCLK_VCLK_0, + PPCLK_DCLK_1, + PPCLK_VCLK_1, + PPCLK_DCEFCLK, + PPCLK_DISPCLK, + PPCLK_PIXCLK, + PPCLK_PHYCLK, + PPCLK_DTBCLK, + PPCLK_COUNT, +} PPCLK_e; + +typedef struct { + uint16_t MinClock; // This is either DCEFCLK or SOCCLK (in MHz) + uint16_t MaxClock; // This is either DCEFCLK or SOCCLK (in MHz) + uint16_t MinUclk; + uint16_t MaxUclk; + + uint8_t WmSetting; + uint8_t Flags; + uint8_t Padding[2]; + +} WatermarkRowGeneric_t; + +#define NUM_WM_RANGES 4 + +typedef enum { + WM_SOCCLK = 0, + WM_DCEFCLK, + WM_COUNT, +} WM_CLOCK_e; + +typedef enum { + WATERMARKS_CLOCK_RANGE = 0, + WATERMARKS_DUMMY_PSTATE, + WATERMARKS_MALL, + WATERMARKS_COUNT, +} WATERMARKS_FLAGS_e; + +typedef struct { + // Watermarks + WatermarkRowGeneric_t WatermarkRow[WM_COUNT][NUM_WM_RANGES]; +} Watermarks_t; + +typedef struct { + Watermarks_t Watermarks; + + uint32_t MmHubPadding[8]; // SMU internal use +} WatermarksExternal_t; + +// Table types +#define TABLE_PPTABLE 0 +#define TABLE_WATERMARKS 1 +#define TABLE_AVFS_PSM_DEBUG 2 +#define TABLE_AVFS_FUSE_OVERRIDE 3 +#define TABLE_PMSTATUSLOG 4 +#define TABLE_SMU_METRICS 5 +#define TABLE_DRIVER_SMU_CONFIG 6 +#define TABLE_ACTIVITY_MONITOR_COEFF 7 +#define TABLE_OVERDRIVE 8 +#define TABLE_I2C_COMMANDS 9 +#define TABLE_PACE 10 +#define TABLE_ECCINFO 11 +#define TABLE_COUNT 12 diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c index 357f89b4a8e0..edda572dc570 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c @@ -131,6 +131,7 @@ static void dcn315_update_clocks(struct clk_mgr *clk_mgr_base, union display_idle_optimization_u idle_info = { 0 }; idle_info.idle_info.df_request_disabled = 1; idle_info.idle_info.phy_ref_clk_off = 1; + idle_info.idle_info.s0i2_rdy = 1; dcn315_smu_set_display_idle_optimization(clk_mgr, idle_info.data); /* update power state */ clk_mgr_base->clks.pwr_state = DCN_PWR_STATE_LOW_POWER; @@ -586,8 +587,10 @@ void dcn315_clk_mgr_construct( } clk_mgr->base.base.dprefclk_khz = 600000; - clk_mgr->base.dccg->ref_dtbclk_khz = 600000; + clk_mgr->base.base.dprefclk_khz = dcn315_smu_get_dpref_clk(&clk_mgr->base); + clk_mgr->base.dccg->ref_dtbclk_khz = clk_mgr->base.base.dprefclk_khz; dce_clock_read_ss_info(&clk_mgr->base); + clk_mgr->base.dccg->ref_dtbclk_khz = dce_adjust_dp_ref_freq_for_ss(&clk_mgr->base, clk_mgr->base.base.dprefclk_khz); clk_mgr->base.base.bw_params = &dcn315_bw_params; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_smu.c index 1420eb5097d3..880ffea2afc6 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_smu.c @@ -312,3 +312,27 @@ void dcn315_smu_transfer_wm_table_dram_2_smu(struct clk_mgr_internal *clk_mgr) VBIOSSMC_MSG_TransferTableDram2Smu, TABLE_WATERMARKS); } +int dcn315_smu_get_dpref_clk(struct clk_mgr_internal *clk_mgr) +{ + int dprefclk_get_mhz = -1; + if (clk_mgr->smu_present) { + dprefclk_get_mhz = dcn315_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_GetDprefclkFreq, + 0); + } + return (dprefclk_get_mhz * 1000); +} + +int dcn315_smu_get_smu_fclk(struct clk_mgr_internal *clk_mgr) +{ + int fclk_get_mhz = -1; + + if (clk_mgr->smu_present) { + fclk_get_mhz = dcn315_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_GetFclkFrequency, + 0); + } + return (fclk_get_mhz * 1000); +} diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_smu.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_smu.h index 96ec6bfcad22..66fa42f8dd18 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_smu.h +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_smu.h @@ -123,4 +123,6 @@ void dcn315_smu_transfer_dpm_table_smu_2_dram(struct clk_mgr_internal *clk_mgr); void dcn315_smu_transfer_wm_table_dram_2_smu(struct clk_mgr_internal *clk_mgr); void dcn315_smu_request_voltage_via_phyclk(struct clk_mgr_internal *clk_mgr, int requested_phyclk_khz); void dcn315_smu_enable_pme_wa(struct clk_mgr_internal *clk_mgr); +int dcn315_smu_get_dpref_clk(struct clk_mgr_internal *clk_mgr); +int dcn315_smu_get_smu_fclk(struct clk_mgr_internal *clk_mgr); #endif /* DAL_DC_315_SMU_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c index c940635b7a74..702d00ce7da4 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c @@ -172,8 +172,7 @@ static void dcn316_update_clocks(struct clk_mgr *clk_mgr_base, union display_idle_optimization_u idle_info = { 0 }; idle_info.idle_info.df_request_disabled = 1; idle_info.idle_info.phy_ref_clk_off = 1; - // Todo DCN316 set this to 1 if any no issue - idle_info.idle_info.s0i2_rdy = 0; + idle_info.idle_info.s0i2_rdy = 1; dcn316_smu_set_display_idle_optimization(clk_mgr, idle_info.data); /* update power state */ clk_mgr_base->clks.pwr_state = DCN_PWR_STATE_LOW_POWER; @@ -684,8 +683,11 @@ void dcn316_clk_mgr_construct( } clk_mgr->base.base.dprefclk_khz = 600000; - clk_mgr->base.dccg->ref_dtbclk_khz = 600000; + clk_mgr->base.base.dprefclk_khz = dcn316_smu_get_dpref_clk(&clk_mgr->base); + clk_mgr->base.dccg->ref_dtbclk_khz = clk_mgr->base.base.dprefclk_khz; dce_clock_read_ss_info(&clk_mgr->base); + clk_mgr->base.dccg->ref_dtbclk_khz = + dce_adjust_dp_ref_freq_for_ss(&clk_mgr->base, clk_mgr->base.base.dprefclk_khz); clk_mgr->base.base.bw_params = &dcn316_bw_params; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_smu.c index 128614dff108..dceec4b96052 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_smu.c @@ -310,3 +310,29 @@ void dcn316_smu_set_dtbclk(struct clk_mgr_internal *clk_mgr, bool enable) VBIOSSMC_MSG_SetDtbclkFreq, enable); } + +int dcn316_smu_get_dpref_clk(struct clk_mgr_internal *clk_mgr) +{ + int dprefclk_get_mhz = -1; + + if (clk_mgr->smu_present) { + dprefclk_get_mhz = dcn316_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_GetDprefclkFreq, + 0); + } + return (dprefclk_get_mhz * 1000); +} + +int dcn316_smu_get_smu_fclk(struct clk_mgr_internal *clk_mgr) +{ + int fclk_get_mhz = -1; + + if (clk_mgr->smu_present) { + fclk_get_mhz = dcn316_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_GetFclkFrequency, + 0); + } + return (fclk_get_mhz * 1000); +} diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_smu.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_smu.h index 658b36d0e107..2a7293f66515 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_smu.h +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_smu.h @@ -133,5 +133,7 @@ void dcn316_smu_transfer_wm_table_dram_2_smu(struct clk_mgr_internal *clk_mgr); void dcn316_smu_request_voltage_via_phyclk(struct clk_mgr_internal *clk_mgr, int requested_phyclk_khz); void dcn316_smu_enable_pme_wa(struct clk_mgr_internal *clk_mgr); void dcn316_smu_set_dtbclk(struct clk_mgr_internal *clk_mgr, bool enable); +int dcn316_smu_get_dpref_clk(struct clk_mgr_internal *clk_mgr); +int dcn316_smu_get_smu_fclk(struct clk_mgr_internal *clk_mgr); #endif /* DAL_DC_316_SMU_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 61e3bb99375f..f6e19efea756 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -1082,7 +1082,8 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context) break; } } - if (!should_disable && pipe_split_change) + if (!should_disable && pipe_split_change && + dc->current_state->stream_count != context->stream_count) should_disable = true; if (should_disable && old_stream) { @@ -1690,6 +1691,7 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c struct pipe_ctx *pipe; int i, k, l; struct dc_stream_state *dc_streams[MAX_STREAMS] = {0}; + struct dc_state *old_state; #if defined(CONFIG_DRM_AMD_DC_DCN) dc_z10_restore(dc); @@ -1808,10 +1810,11 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c for (i = 0; i < context->stream_count; i++) context->streams[i]->mode_changed = false; - dc_release_state(dc->current_state); - + old_state = dc->current_state; dc->current_state = context; + dc_release_state(old_state); + dc_retain_state(dc->current_state); return result; @@ -3360,6 +3363,19 @@ bool dc_is_dmcu_initialized(struct dc *dc) return false; } +bool dc_is_oem_i2c_device_present( + struct dc *dc, + size_t slave_address) +{ + if (dc->res_pool->oem_device) + return dce_i2c_oem_device_present( + dc->res_pool, + dc->res_pool->oem_device, + slave_address); + + return false; +} + bool dc_submit_i2c( struct dc *dc, uint32_t link_index, diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c index 27f3fc416448..cb87dd643180 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c @@ -50,6 +50,7 @@ #include "inc/hw/panel_cntl.h" #include "inc/link_enc_cfg.h" #include "inc/link_dpcd.h" +#include "link/link_dp_trace.h" #include "dc/dcn30/dcn30_vpg.h" @@ -730,6 +731,7 @@ static bool detect_dp(struct dc_link *link, sink_caps, audio_support); link->dpcd_caps.dongle_type = sink_caps->dongle_type; + link->dpcd_caps.is_dongle_type_one = sink_caps->is_dongle_type_one; link->dpcd_caps.dpcd_rev.raw = 0; } @@ -981,7 +983,8 @@ static bool should_verify_link_capability_destructively(struct dc_link *link, destrictive = false; } } - } + } else if (dc_is_hdmi_signal(link->local_sink->sink_signal)) + destrictive = true; return destrictive; } @@ -1180,6 +1183,9 @@ static bool detect_link_and_local_sink(struct dc_link *link, case EDID_BAD_CHECKSUM: DC_LOG_ERROR("EDID checksum invalid.\n"); break; + case EDID_PARTIAL_VALID: + DC_LOG_ERROR("Partial EDID valid, abandon invalid blocks.\n"); + break; case EDID_NO_RESPONSE: DC_LOG_ERROR("No EDID read.\n"); /* @@ -1254,6 +1260,9 @@ static bool detect_link_and_local_sink(struct dc_link *link, !sink->edid_caps.edid_hdmi) sink->sink_signal = SIGNAL_TYPE_DVI_SINGLE_LINK; + if (link->local_sink && dc_is_dp_signal(sink_caps.signal)) + dp_trace_init(link); + /* Connectivity log: detection */ for (i = 0; i < sink->dc_edid.length / DC_EDID_BLOCK_SIZE; i++) { CONN_DATA_DETECT(link, @@ -1306,6 +1315,7 @@ static bool detect_link_and_local_sink(struct dc_link *link, link->dongle_max_pix_clk = 0; dc_link_clear_dprx_states(link); + dp_trace_reset(link); } LINK_INFO("link=%d, dc_sink_in=%p is now %s prev_sink=%p edid same=%d\n", @@ -2801,6 +2811,17 @@ static bool dp_active_dongle_validate_timing( return false; } + /* Check 3D format */ + switch (timing->timing_3d_format) { + case TIMING_3D_FORMAT_NONE: + case TIMING_3D_FORMAT_FRAME_ALTERNATE: + /*Only frame alternate 3D is supported on active dongle*/ + break; + default: + /*other 3D formats are not supported due to bad infoframe translation */ + return false; + } + #if defined(CONFIG_DRM_AMD_DC_DCN) if (dongle_caps->dp_hdmi_frl_max_link_bw_in_kbps > 0) { // DP to HDMI FRL converter struct dc_crtc_timing outputTiming = *timing; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c index f1bbd918de35..1d4863763df9 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c @@ -493,6 +493,7 @@ void dal_ddc_service_i2c_query_dp_dual_mode_adaptor( sink_cap->max_hdmi_pixel_clock = max_tmds_clk * 1000; } + sink_cap->is_dongle_type_one = false; } else { if (is_valid_hdmi_signature == true) { @@ -510,6 +511,7 @@ void dal_ddc_service_i2c_query_dp_dual_mode_adaptor( "Type 1 DP-HDMI passive dongle (no signature) %dMhz: ", sink_cap->max_hdmi_pixel_clock / 1000); } + sink_cap->is_dongle_type_one = true; } return; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c index 5688b15ca9e6..351081f574cb 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c @@ -39,6 +39,7 @@ #include "dce/dmub_hw_lock_mgr.h" #include "inc/dc_link_dpia.h" #include "inc/link_enc_cfg.h" +#include "link/link_dp_trace.h" /*Travis*/ static const uint8_t DP_VGA_LVDS_CONVERTER_ID_2[] = "sivarT"; @@ -347,29 +348,6 @@ static void vendor_specific_lttpr_wa_one_start(struct dc_link *link) sizeof(vendor_lttpr_write_data)); } -static void vendor_specific_lttpr_wa_one_end( - struct dc_link *link, - uint8_t retry_count) -{ - const uint8_t vendor_lttpr_write_data[4] = {0x1, 0x50, 0x63, 0x0}; - const uint8_t offset = dp_convert_to_count( - link->dpcd_caps.lttpr_caps.phy_repeater_cnt); - uint32_t vendor_lttpr_write_address = 0xF004F; - - if (!retry_count) { - if (offset != 0xFF) - vendor_lttpr_write_address += - ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); - - /* W/A for certain LTTPR to reset their lane settings, part two of two */ - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data[0], - sizeof(vendor_lttpr_write_data)); - } -} - static void vendor_specific_lttpr_wa_one_two( struct dc_link *link, const uint8_t rate) @@ -396,9 +374,9 @@ static void vendor_specific_lttpr_wa_one_two( } } -static void vendor_specific_lttpr_wa_three( +static void dp_fixed_vs_pe_read_lane_adjust( struct dc_link *link, - union lane_adjust dpcd_lane_adjust[LANE_COUNT_DP_MAX]) + union dpcd_training_lane dpcd_lane_adjust[LANE_COUNT_DP_MAX]) { const uint8_t vendor_lttpr_write_data_vs[3] = {0x0, 0x53, 0x63}; const uint8_t vendor_lttpr_write_data_pe[3] = {0x0, 0x54, 0x63}; @@ -440,23 +418,8 @@ static void vendor_specific_lttpr_wa_three( 1); for (lane = 0; lane < LANE_COUNT_DP_MAX; lane++) { - dpcd_lane_adjust[lane].bits.VOLTAGE_SWING_LANE = (dprx_vs >> (2 * lane)) & 0x3; - dpcd_lane_adjust[lane].bits.PRE_EMPHASIS_LANE = (dprx_pe >> (2 * lane)) & 0x3; - } -} - -static void vendor_specific_lttpr_wa_three_dpcd( - struct dc_link *link, - union dpcd_training_lane dpcd_lane_adjust[LANE_COUNT_DP_MAX]) -{ - union lane_adjust lane_adjust[LANE_COUNT_DP_MAX]; - uint8_t lane = 0; - - vendor_specific_lttpr_wa_three(link, lane_adjust); - - for (lane = 0; lane < LANE_COUNT_DP_MAX; lane++) { - dpcd_lane_adjust[lane].bits.VOLTAGE_SWING_SET = lane_adjust[lane].bits.VOLTAGE_SWING_LANE; - dpcd_lane_adjust[lane].bits.PRE_EMPHASIS_SET = lane_adjust[lane].bits.PRE_EMPHASIS_LANE; + dpcd_lane_adjust[lane].bits.VOLTAGE_SWING_SET = (dprx_vs >> (2 * lane)) & 0x3; + dpcd_lane_adjust[lane].bits.PRE_EMPHASIS_SET = (dprx_pe >> (2 * lane)) & 0x3; } } @@ -505,17 +468,24 @@ static void vendor_specific_lttpr_wa_four( } } -static void vendor_specific_lttpr_wa_five( +static void dp_fixed_vs_pe_set_retimer_lane_settings( struct dc_link *link, const union dpcd_training_lane dpcd_lane_adjust[LANE_COUNT_DP_MAX], uint8_t lane_count) { - const uint32_t vendor_lttpr_write_address = 0xF004F; + const uint8_t offset = dp_convert_to_count( + link->dpcd_caps.lttpr_caps.phy_repeater_cnt); const uint8_t vendor_lttpr_write_data_reset[4] = {0x1, 0x50, 0x63, 0xFF}; + uint32_t vendor_lttpr_write_address = 0xF004F; uint8_t vendor_lttpr_write_data_vs[4] = {0x1, 0x51, 0x63, 0x0}; uint8_t vendor_lttpr_write_data_pe[4] = {0x1, 0x52, 0x63, 0x0}; uint8_t lane = 0; + if (offset != 0xFF) { + vendor_lttpr_write_address += + ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); + } + for (lane = 0; lane < lane_count; lane++) { vendor_lttpr_write_data_vs[3] |= dpcd_lane_adjust[lane].bits.VOLTAGE_SWING_SET << (2 * lane); @@ -1014,6 +984,14 @@ enum dc_status dp_get_lane_status_and_lane_adjust( (uint8_t *)(dpcd_buf), sizeof(dpcd_buf)); + if (status != DC_OK) { + DC_LOG_HW_LINK_TRAINING("%s:\n Failed to read from address 0x%X," + " keep current lane status and lane adjust unchanged", + __func__, + lane01_status_address); + return status; + } + for (lane = 0; lane < (uint32_t)(link_training_setting->link_settings.lane_count); lane++) { @@ -1154,6 +1132,9 @@ static bool perform_post_lt_adj_req_sequence( uint32_t adj_req_timer; bool req_drv_setting_changed; uint32_t lane; + union lane_status dpcd_lane_status[LANE_COUNT_DP_MAX] = {0}; + union lane_align_status_updated dpcd_lane_status_updated = {0}; + union lane_adjust dpcd_lane_adjust[LANE_COUNT_DP_MAX] = {0}; req_drv_setting_changed = false; for (adj_req_count = 0; adj_req_count < POST_LT_ADJ_REQ_LIMIT; @@ -1165,11 +1146,6 @@ static bool perform_post_lt_adj_req_sequence( adj_req_timer < POST_LT_ADJ_REQ_TIMEOUT; adj_req_timer++) { - union lane_status dpcd_lane_status[LANE_COUNT_DP_MAX]; - union lane_align_status_updated - dpcd_lane_status_updated; - union lane_adjust dpcd_lane_adjust[LANE_COUNT_DP_MAX] = { { {0} } }; - dp_get_lane_status_and_lane_adjust( link, lt_settings, @@ -1396,6 +1372,10 @@ static enum link_training_result perform_clock_recovery_sequence( retries_cr = 0; retry_count = 0; + memset(&dpcd_lane_status, '\0', sizeof(dpcd_lane_status)); + memset(&dpcd_lane_status_updated, '\0', + sizeof(dpcd_lane_status_updated)); + if (!link->ctx->dc->work_arounds.lt_early_cr_pattern) dp_set_hw_training_pattern(link, link_res, lt_settings->pattern_for_cr, offset); @@ -1407,9 +1387,6 @@ static enum link_training_result perform_clock_recovery_sequence( while ((retries_cr < LINK_TRAINING_MAX_RETRY_COUNT) && (retry_count < LINK_TRAINING_MAX_CR_RETRY)) { - memset(&dpcd_lane_status, '\0', sizeof(dpcd_lane_status)); - memset(&dpcd_lane_status_updated, '\0', - sizeof(dpcd_lane_status_updated)); /* 1. call HWSS to set lane settings*/ dp_set_hw_lane_settings( @@ -1456,13 +1433,6 @@ static enum link_training_result perform_clock_recovery_sequence( dpcd_lane_adjust, offset); - if (link->dc->debug.apply_vendor_specific_lttpr_wa && - (link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) && - link->lttpr_mode == LTTPR_MODE_TRANSPARENT) { - vendor_specific_lttpr_wa_one_end(link, retry_count); - vendor_specific_lttpr_wa_three(link, dpcd_lane_adjust); - } - /* 5. check CR done*/ if (dp_is_cr_done(lane_count, dpcd_lane_status)) return LINK_TRAINING_SUCCESS; @@ -2512,12 +2482,13 @@ static enum link_training_result dp_perform_fixed_vs_pe_training_sequence( retries_cr = 0; retry_count = 0; + memset(&dpcd_lane_status, '\0', sizeof(dpcd_lane_status)); + memset(&dpcd_lane_status_updated, '\0', + sizeof(dpcd_lane_status_updated)); + while ((retries_cr < LINK_TRAINING_MAX_RETRY_COUNT) && (retry_count < LINK_TRAINING_MAX_CR_RETRY)) { - memset(&dpcd_lane_status, '\0', sizeof(dpcd_lane_status)); - memset(&dpcd_lane_status_updated, '\0', - sizeof(dpcd_lane_status_updated)); /* 1. call HWSS to set lane settings */ dp_set_hw_lane_settings( @@ -2814,6 +2785,10 @@ bool perform_link_training_with_retries( enum link_training_result status = LINK_TRAINING_CR_FAIL_LANE0; struct dc_link_settings current_setting = *link_setting; const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); + int fail_count = 0; + + dp_trace_commit_lt_init(link); + if (dp_get_link_encoding_format(¤t_setting) == DP_8b_10b_ENCODING) /* We need to do this before the link training to ensure the idle @@ -2821,6 +2796,7 @@ bool perform_link_training_with_retries( */ link_hwss->setup_stream_encoder(pipe_ctx); + dp_trace_set_lt_start_timestamp(link, false); for (j = 0; j < attempts; ++j) { DC_LOG_HW_LINK_TRAINING("%s: Beginning link training attempt %u of %d\n", @@ -2876,10 +2852,15 @@ bool perform_link_training_with_retries( skip_video_pattern); } + dp_trace_lt_total_count_increment(link, false); + dp_trace_lt_result_update(link, status, false); + dp_trace_set_lt_end_timestamp(link, false); if (status == LINK_TRAINING_SUCCESS) return true; } + fail_count++; + dp_trace_lt_fail_count_update(link, fail_count, false); /* latest link training still fail, skip delay and keep PHY on */ if (j == (attempts - 1) && link->ep_type == DISPLAY_ENDPOINT_PHY) @@ -3347,6 +3328,8 @@ static bool dp_verify_link_cap( } else { (*fail_count)++; } + dp_trace_lt_total_count_increment(link, true); + dp_trace_lt_result_update(link, status, true); dp_disable_link_phy(link, &link_res, link->connector_signal); } while (!success && decide_fallback_link_setting(link, initial_link_settings, &cur_link_settings, status)); @@ -3378,13 +3361,16 @@ bool dp_verify_link_cap_with_retries( { int i = 0; bool success = false; + int fail_count = 0; + + dp_trace_detect_lt_init(link); if (link->link_enc && link->link_enc->features.flags.bits.DP_IS_USB_C && link->dc->debug.usbc_combo_phy_reset_wa) apply_usbc_combo_phy_reset_wa(link, known_limit_link_setting); + dp_trace_set_lt_start_timestamp(link, false); for (i = 0; i < attempts; i++) { - int fail_count = 0; enum dc_connection_type type = dc_connection_none; memset(&link->verified_link_cap, 0, @@ -3399,6 +3385,10 @@ bool dp_verify_link_cap_with_retries( } msleep(10); } + + dp_trace_lt_fail_count_update(link, fail_count, true); + dp_trace_set_lt_end_timestamp(link, true); + return success; } @@ -3488,7 +3478,8 @@ static enum dc_lane_count increase_lane_count(enum dc_lane_count lane_count) } } -static enum dc_link_rate increase_link_rate(enum dc_link_rate link_rate) +static enum dc_link_rate increase_link_rate(struct dc_link *link, + enum dc_link_rate link_rate) { switch (link_rate) { case LINK_RATE_LOW: @@ -3500,7 +3491,15 @@ static enum dc_link_rate increase_link_rate(enum dc_link_rate link_rate) case LINK_RATE_HIGH3: return LINK_RATE_UHBR10; case LINK_RATE_UHBR10: - return LINK_RATE_UHBR13_5; + /* upto DP2.x specs UHBR13.5 is the only link rate that could be + * not supported by DPRX when higher link rate is supported. + * so we treat it as a special case for code simplicity. When we + * have new specs with more link rates like this, we should + * consider a more generic solution to handle discrete link + * rate capabilities. + */ + return link->dpcd_caps.dp_128b_132b_supported_link_rates.bits.UHBR13_5 ? + LINK_RATE_UHBR13_5 : LINK_RATE_UHBR20; case LINK_RATE_UHBR13_5: return LINK_RATE_UHBR20; default: @@ -3509,12 +3508,17 @@ static enum dc_link_rate increase_link_rate(enum dc_link_rate link_rate) } static bool decide_fallback_link_setting_max_bw_policy( + struct dc_link *link, const struct dc_link_settings *max, - struct dc_link_settings *cur) + struct dc_link_settings *cur, + enum link_training_result training_result) { uint8_t cur_idx = 0, next_idx; bool found = false; + if (training_result == LINK_TRAINING_ABORT) + return false; + while (cur_idx < ARRAY_SIZE(dp_lt_fallbacks)) /* find current index */ if (dp_lt_fallbacks[cur_idx].lane_count == cur->lane_count && @@ -3527,11 +3531,22 @@ static bool decide_fallback_link_setting_max_bw_policy( while (next_idx < ARRAY_SIZE(dp_lt_fallbacks)) /* find next index */ - if (dp_lt_fallbacks[next_idx].lane_count <= max->lane_count && - dp_lt_fallbacks[next_idx].link_rate <= max->link_rate) - break; - else + if (dp_lt_fallbacks[next_idx].lane_count > max->lane_count || + dp_lt_fallbacks[next_idx].link_rate > max->link_rate) + next_idx++; + else if (dp_lt_fallbacks[next_idx].link_rate == LINK_RATE_UHBR13_5 && + link->dpcd_caps.dp_128b_132b_supported_link_rates.bits.UHBR13_5 == 0) + /* upto DP2.x specs UHBR13.5 is the only link rate that + * could be not supported by DPRX when higher link rate + * is supported. so we treat it as a special case for + * code simplicity. When we have new specs with more + * link rates like this, we should consider a more + * generic solution to handle discrete link rate + * capabilities. + */ next_idx++; + else + break; if (next_idx < ARRAY_SIZE(dp_lt_fallbacks)) { cur->lane_count = dp_lt_fallbacks[next_idx].lane_count; @@ -3560,8 +3575,8 @@ static bool decide_fallback_link_setting( return false; if (dp_get_link_encoding_format(&initial_link_settings) == DP_128b_132b_ENCODING || link->dc->debug.force_dp2_lt_fallback_method) - return decide_fallback_link_setting_max_bw_policy(&initial_link_settings, - current_link_setting); + return decide_fallback_link_setting_max_bw_policy(link, &initial_link_settings, + current_link_setting, training_result); switch (training_result) { case LINK_TRAINING_CR_FAIL_LANE0: @@ -3716,7 +3731,7 @@ static bool decide_dp_link_settings(struct dc_link *link, struct dc_link_setting current_link_setting.lane_count); } else { current_link_setting.link_rate = - increase_link_rate( + increase_link_rate(link, current_link_setting.link_rate); current_link_setting.lane_count = initial_link_setting.lane_count; @@ -3831,7 +3846,7 @@ static bool decide_edp_link_settings_with_dsc(struct dc_link *link, /* minimize lane */ if (current_link_setting.link_rate < max_link_rate) { current_link_setting.link_rate = - increase_link_rate( + increase_link_rate(link, current_link_setting.link_rate); } else { if (current_link_setting.lane_count < @@ -3852,7 +3867,7 @@ static bool decide_edp_link_settings_with_dsc(struct dc_link *link, current_link_setting.lane_count); } else { current_link_setting.link_rate = - increase_link_rate( + increase_link_rate(link, current_link_setting.link_rate); current_link_setting.lane_count = initial_link_setting.lane_count; @@ -4129,7 +4144,7 @@ static void dp_test_send_phy_test_pattern(struct dc_link *link) if (link->dc->debug.apply_vendor_specific_lttpr_wa && (link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) && link->lttpr_mode == LTTPR_MODE_TRANSPARENT) - vendor_specific_lttpr_wa_three_dpcd( + dp_fixed_vs_pe_read_lane_adjust( link, link_training_settings.dpcd_lane_settings); @@ -4665,6 +4680,8 @@ bool dc_link_handle_hpd_rx_irq(struct dc_link *link, union hpd_irq_data *out_hpd status = false; if (out_link_loss) *out_link_loss = true; + + dp_trace_link_loss_increment(link); } if (link->type == dc_connection_sst_branch && @@ -5101,6 +5118,7 @@ bool dp_retrieve_lttpr_cap(struct dc_link *link) sizeof(lttpr_dpcd_data)); if (status != DC_OK) { DC_LOG_DP2("%s: Read LTTPR caps data failed.\n", __func__); + link->lttpr_mode = LTTPR_MODE_NON_LTTPR; return false; } @@ -5989,15 +6007,14 @@ bool dc_link_dp_set_test_pattern( if (link->dc->debug.apply_vendor_specific_lttpr_wa && (link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) && link->lttpr_mode == LTTPR_MODE_TRANSPARENT) { - dpcd_set_lane_settings(link, p_link_settings, DPRX); - vendor_specific_lttpr_wa_five( + dp_fixed_vs_pe_set_retimer_lane_settings( link, p_link_settings->dpcd_lane_settings, p_link_settings->link_settings.lane_count); } else { dp_set_hw_lane_settings(link, &pipe_ctx->link_res, p_link_settings, DPRX); - dpcd_set_lane_settings(link, p_link_settings, DPRX); } + dpcd_set_lane_settings(link, p_link_settings, DPRX); } /* Blank stream if running test pattern */ diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c index 047c626a4a34..42da7f430113 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c @@ -272,6 +272,13 @@ void link_enc_cfg_init( state->res_ctx.link_enc_cfg_ctx.mode = LINK_ENC_CFG_STEADY; } +void link_enc_cfg_copy(const struct dc_state *src_ctx, struct dc_state *dst_ctx) +{ + memcpy(&dst_ctx->res_ctx.link_enc_cfg_ctx, + &src_ctx->res_ctx.link_enc_cfg_ctx, + sizeof(dst_ctx->res_ctx.link_enc_cfg_ctx)); +} + void link_enc_cfg_link_encs_assign( struct dc *dc, struct dc_state *state, diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index cc8e60ec35c6..7af153434e9e 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -2335,6 +2335,9 @@ void dc_resource_state_construct( bool dc_resource_is_dsc_encoding_supported(const struct dc *dc) { + if (dc->res_pool == NULL) + return false; + return dc->res_pool->res_cap->num_dsc > 0; } @@ -2615,6 +2618,8 @@ static void set_avi_info_frame( hdmi_info.bits.YQ0_YQ1 = YYC_QUANTIZATION_LIMITED_RANGE; ///VIC + if (pipe_ctx->stream->timing.hdmi_vic != 0) + vic = 0; format = stream->timing.timing_3d_format; /*todo, add 3DStereo support*/ if (format != TIMING_3D_FORMAT_NONE) { diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c index dc5fd27b031a..c4e871f358ab 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c @@ -718,6 +718,20 @@ enum dc_status dc_stream_add_dsc_to_resource(struct dc *dc, } } +struct pipe_ctx *dc_stream_get_pipe_ctx(struct dc_stream_state *stream) +{ + int i = 0; + + for (i = 0; i < MAX_PIPES; i++) { + struct pipe_ctx *pipe = &stream->ctx->dc->current_state->res_ctx.pipe_ctx[i]; + + if (pipe->stream == stream) + return pipe; + } + + return NULL; +} + void dc_stream_log(const struct dc *dc, const struct dc_stream_state *stream) { DC_LOG_DC( diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index 55d43d642b38..4ffab7bb1098 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -47,7 +47,7 @@ struct aux_payload; struct set_config_cmd_payload; struct dmub_notification; -#define DC_VER "3.2.174" +#define DC_VER "3.2.177" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -354,6 +354,7 @@ enum dc_psr_power_opts { psr_power_opt_invalid = 0x0, psr_power_opt_smu_opt_static_screen = 0x1, psr_power_opt_z10_static_screen = 0x10, + psr_power_opt_ds_disable_allow = 0x100, }; enum dcc_option { @@ -710,6 +711,7 @@ struct dc_debug_options { #endif bool apply_vendor_specific_lttpr_wa; bool ignore_dpref_ss; + uint8_t psr_power_use_phy_fsm; }; struct gpu_info_soc_bounding_box_v1_0; @@ -1204,6 +1206,7 @@ struct dpcd_caps { /* dongle type (DP converter, CV smart dongle) */ enum display_dongle_type dongle_type; + bool is_dongle_type_one; /* branch device or sink device */ bool is_branch_dev; /* Dongle's downstream count. */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h index ce6e8d013459..aa818bf840eb 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_link.h +++ b/drivers/gpu/drm/amd/display/dc/dc_link.h @@ -76,6 +76,28 @@ struct link_trace { struct time_stamp time_stamp; }; +struct dp_trace_lt_counts { + unsigned int total; + unsigned int fail; +}; + +struct dp_trace_lt { + struct dp_trace_lt_counts counts; + struct dp_trace_timestamps { + unsigned long long start; + unsigned long long end; + } timestamps; + enum link_training_result result; + bool is_logged; +}; + +struct dp_trace { + struct dp_trace_lt detect_lt_trace; + struct dp_trace_lt commit_lt_trace; + unsigned int link_loss_count; + bool is_initialized; +}; + /* PSR feature flags */ struct psr_settings { bool psr_feature_enabled; // PSR is supported by sink @@ -121,6 +143,8 @@ struct dc_link { bool edp_sink_present; + struct dp_trace dp_trace; + /* caps is the same as reported_link_cap. link_traing use * reported_link_cap. Will clean up. TODO */ @@ -442,6 +466,11 @@ const struct dc_link_settings *dc_link_get_link_cap( void dc_link_overwrite_extended_receiver_cap( struct dc_link *link); +bool dc_is_oem_i2c_device_present( + struct dc *dc, + size_t slave_address +); + bool dc_submit_i2c( struct dc *dc, uint32_t link_index, @@ -470,4 +499,16 @@ void dc_link_clear_dprx_states(struct dc_link *link); struct gpio *get_hpd_gpio(struct dc_bios *dcb, struct graphics_object_id link_id, struct gpio_service *gpio_service); +void dp_trace_reset(struct dc_link *link); +bool dc_dp_trace_is_initialized(struct dc_link *link); +unsigned long long dc_dp_trace_get_lt_end_timestamp(struct dc_link *link, + bool in_detection); +void dc_dp_trace_set_is_logged_flag(struct dc_link *link, + bool in_detection, + bool is_logged); +bool dc_dp_trace_is_logged(struct dc_link *link, + bool in_detection); +struct dp_trace_lt_counts *dc_dp_trace_get_lt_counts(struct dc_link *link, + bool in_detection); +unsigned int dc_dp_trace_get_link_loss_count(struct dc_link *link); #endif /* DC_LINK_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h index f631b61abedd..99a750f561f8 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_stream.h +++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h @@ -524,4 +524,6 @@ bool dc_stream_get_crtc_position(struct dc *dc, unsigned int *v_pos, unsigned int *nom_v_pos); +struct pipe_ctx *dc_stream_get_pipe_ctx(struct dc_stream_state *stream); + #endif /* DC_STREAM_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index 48859d5fc172..2ba9f528c0fe 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -138,6 +138,7 @@ enum dc_edid_status { EDID_BAD_CHECKSUM, EDID_THE_SAME, EDID_FALL_BACK, + EDID_PARTIAL_VALID, }; enum act_return_status { diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c index 74b05b3aef08..8e814000db62 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c @@ -413,7 +413,8 @@ static bool acquire( return false; if (!acquire_engine(engine)) { - dal_ddc_close(ddc); + engine->ddc = ddc; + release_engine(engine); return false; } diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c index 1435d7bc1f21..07359eb89efc 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c @@ -450,6 +450,8 @@ void dce_clock_read_ss_info(struct dce_clk_mgr *clk_mgr_dce) clk_mgr_dce->dprefclk_ss_percentage = info.spread_spectrum_percentage; } + if (clk_mgr_dce->base.ctx->dc->debug.ignore_dpref_ss) + clk_mgr_dce->dprefclk_ss_percentage = 0; } } } diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c index 2c7eb982eabc..cc5128e67daf 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c @@ -971,6 +971,98 @@ static bool dce112_program_pix_clk( return true; } +#if defined(CONFIG_DRM_AMD_DC_DCN) +static bool dcn31_program_pix_clk( + struct clock_source *clock_source, + struct pixel_clk_params *pix_clk_params, + struct pll_settings *pll_settings) +{ + struct dce110_clk_src *clk_src = TO_DCE110_CLK_SRC(clock_source); + unsigned int inst = pix_clk_params->controller_id - CONTROLLER_ID_D0; + unsigned int dp_dto_ref_khz = clock_source->ctx->dc->clk_mgr->dprefclk_khz; + const struct pixel_rate_range_table_entry *e = + look_up_in_video_optimized_rate_tlb(pix_clk_params->requested_pix_clk_100hz / 10); + struct bp_pixel_clock_parameters bp_pc_params = {0}; + enum transmitter_color_depth bp_pc_colour_depth = TRANSMITTER_COLOR_DEPTH_24; + // For these signal types Driver to program DP_DTO without calling VBIOS Command table + if (dc_is_dp_signal(pix_clk_params->signal_type)) { + if (e) { + /* Set DTO values: phase = target clock, modulo = reference clock*/ + REG_WRITE(PHASE[inst], e->target_pixel_rate_khz * e->mult_factor); + REG_WRITE(MODULO[inst], dp_dto_ref_khz * e->div_factor); + } else { + /* Set DTO values: phase = target clock, modulo = reference clock*/ + REG_WRITE(PHASE[inst], pll_settings->actual_pix_clk_100hz * 100); + REG_WRITE(MODULO[inst], dp_dto_ref_khz * 1000); + } + REG_UPDATE(PIXEL_RATE_CNTL[inst], DP_DTO0_ENABLE, 1); + } else { + if (IS_FPGA_MAXIMUS_DC(clock_source->ctx->dce_environment)) { + unsigned int inst = pix_clk_params->controller_id - CONTROLLER_ID_D0; + unsigned dp_dto_ref_100hz = 7000000; + unsigned clock_100hz = pll_settings->actual_pix_clk_100hz; + + /* Set DTO values: phase = target clock, modulo = reference clock */ + REG_WRITE(PHASE[inst], clock_100hz); + REG_WRITE(MODULO[inst], dp_dto_ref_100hz); + + /* Enable DTO */ + REG_UPDATE(PIXEL_RATE_CNTL[inst], DP_DTO0_ENABLE, 1); + return true; + } + + /*ATOMBIOS expects pixel rate adjusted by deep color ratio)*/ + bp_pc_params.controller_id = pix_clk_params->controller_id; + bp_pc_params.pll_id = clock_source->id; + bp_pc_params.target_pixel_clock_100hz = pll_settings->actual_pix_clk_100hz; + bp_pc_params.encoder_object_id = pix_clk_params->encoder_object_id; + bp_pc_params.signal_type = pix_clk_params->signal_type; + + // Make sure we send the correct color depth to DMUB for HDMI + if (pix_clk_params->signal_type == SIGNAL_TYPE_HDMI_TYPE_A) { + switch (pix_clk_params->color_depth) { + case COLOR_DEPTH_888: + bp_pc_colour_depth = TRANSMITTER_COLOR_DEPTH_24; + break; + case COLOR_DEPTH_101010: + bp_pc_colour_depth = TRANSMITTER_COLOR_DEPTH_30; + break; + case COLOR_DEPTH_121212: + bp_pc_colour_depth = TRANSMITTER_COLOR_DEPTH_36; + break; + case COLOR_DEPTH_161616: + bp_pc_colour_depth = TRANSMITTER_COLOR_DEPTH_48; + break; + default: + bp_pc_colour_depth = TRANSMITTER_COLOR_DEPTH_24; + break; + } + bp_pc_params.color_depth = bp_pc_colour_depth; + } + + if (clock_source->id != CLOCK_SOURCE_ID_DP_DTO) { + bp_pc_params.flags.SET_GENLOCK_REF_DIV_SRC = + pll_settings->use_external_clk; + bp_pc_params.flags.SET_XTALIN_REF_SRC = + !pll_settings->use_external_clk; + if (pix_clk_params->flags.SUPPORT_YCBCR420) { + bp_pc_params.flags.SUPPORT_YUV_420 = 1; + } + } + if (clk_src->bios->funcs->set_pixel_clock( + clk_src->bios, &bp_pc_params) != BP_RESULT_OK) + return false; + /* Resync deep color DTO */ + if (clock_source->id != CLOCK_SOURCE_ID_DP_DTO) + dce112_program_pixel_clk_resync(clk_src, + pix_clk_params->signal_type, + pix_clk_params->color_depth, + pix_clk_params->flags.SUPPORT_YCBCR420); + } + + return true; +} +#endif static bool dce110_clock_source_power_down( struct clock_source *clk_src) @@ -1205,6 +1297,13 @@ static const struct clock_source_funcs dcn3_clk_src_funcs = { .get_pix_clk_dividers = dcn3_get_pix_clk_dividers, .get_pixel_clk_frequency_100hz = get_pixel_clk_frequency_100hz }; + +static const struct clock_source_funcs dcn31_clk_src_funcs = { + .cs_power_down = dce110_clock_source_power_down, + .program_pix_clk = dcn31_program_pix_clk, + .get_pix_clk_dividers = dcn3_get_pix_clk_dividers, + .get_pixel_clk_frequency_100hz = get_pixel_clk_frequency_100hz +}; #endif /*****************************************/ /* Constructor */ @@ -1610,6 +1709,24 @@ bool dcn3_clk_src_construct( #endif #if defined(CONFIG_DRM_AMD_DC_DCN) +bool dcn31_clk_src_construct( + struct dce110_clk_src *clk_src, + struct dc_context *ctx, + struct dc_bios *bios, + enum clock_source_id id, + const struct dce110_clk_src_regs *regs, + const struct dce110_clk_src_shift *cs_shift, + const struct dce110_clk_src_mask *cs_mask) +{ + bool ret = dce112_clk_src_construct(clk_src, ctx, bios, id, regs, cs_shift, cs_mask); + + clk_src->base.funcs = &dcn31_clk_src_funcs; + + return ret; +} +#endif + +#if defined(CONFIG_DRM_AMD_DC_DCN) bool dcn301_clk_src_construct( struct dce110_clk_src *clk_src, struct dc_context *ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h index 692fa23ca02b..069de7649c8c 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h @@ -292,6 +292,15 @@ bool dcn301_clk_src_construct( const struct dce110_clk_src_regs *regs, const struct dce110_clk_src_shift *cs_shift, const struct dce110_clk_src_mask *cs_mask); + +bool dcn31_clk_src_construct( + struct dce110_clk_src *clk_src, + struct dc_context *ctx, + struct dc_bios *bios, + enum clock_source_id id, + const struct dce110_clk_src_regs *regs, + const struct dce110_clk_src_shift *cs_shift, + const struct dce110_clk_src_mask *cs_mask); #endif /* this table is use to find *1.001 and /1.001 pixel rates from non-precise pixel rate */ diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c.c b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c.c index dd41736bb5c4..f5cd2392fc5f 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c.c @@ -25,6 +25,32 @@ #include "dce_i2c.h" #include "reg_helper.h" +bool dce_i2c_oem_device_present( + struct resource_pool *pool, + struct ddc_service *ddc, + size_t slave_address +) +{ + struct dc *dc = ddc->ctx->dc; + struct dc_bios *dcb = dc->ctx->dc_bios; + struct graphics_object_id id = {0}; + struct graphics_object_i2c_info i2c_info; + + if (!dc->ctx->dc_bios->fw_info.oem_i2c_present) + return false; + + id.id = dc->ctx->dc_bios->fw_info.oem_i2c_obj_id; + id.enum_id = 0; + id.type = OBJECT_TYPE_GENERIC; + if (dcb->funcs->get_i2c_info(dcb, id, &i2c_info) != BP_RESULT_OK) + return false; + + if (i2c_info.i2c_slave_address != slave_address) + return false; + + return true; +} + bool dce_i2c_submit_command( struct resource_pool *pool, struct ddc *ddc, diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c.h b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c.h index a171c5cd8439..535fd58de450 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c.h @@ -30,6 +30,12 @@ #include "dce_i2c_hw.h" #include "dce_i2c_sw.h" +bool dce_i2c_oem_device_present( + struct resource_pool *pool, + struct ddc_service *ddc, + size_t slave_address +); + bool dce_i2c_submit_command( struct resource_pool *pool, struct ddc *ddc, diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c index 8bd265b40847..312c68172689 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c @@ -320,6 +320,7 @@ static bool dmub_psr_copy_settings(struct dmub_psr *dmub, copy_settings_data->otg_inst = 0; // Misc + copy_settings_data->use_phy_fsm = link->ctx->dc->debug.psr_power_use_phy_fsm; copy_settings_data->psr_level = psr_context->psr_level.u32all; copy_settings_data->smu_optimizations_en = psr_context->allow_smu_optimizations; copy_settings_data->multi_disp_optimizations_en = psr_context->allow_multi_disp_optimizations; diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c index 8378b80e8517..248602c15f3a 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c @@ -1509,6 +1509,31 @@ static enum dc_status apply_single_controller_ctx_to_hw( if (!pipe_ctx->stream->apply_seamless_boot_optimization && dc->config.use_pipe_ctx_sync_logic) check_syncd_pipes_for_disabled_master_pipe(dc, context, pipe_ctx->pipe_idx); + pipe_ctx->stream_res.opp->funcs->opp_program_fmt( + pipe_ctx->stream_res.opp, + &stream->bit_depth_params, + &stream->clamping); + + pipe_ctx->stream_res.opp->funcs->opp_set_dyn_expansion( + pipe_ctx->stream_res.opp, + COLOR_SPACE_YCBCR601, + stream->timing.display_color_depth, + stream->signal); + + while (odm_pipe) { + odm_pipe->stream_res.opp->funcs->opp_set_dyn_expansion( + odm_pipe->stream_res.opp, + COLOR_SPACE_YCBCR601, + stream->timing.display_color_depth, + stream->signal); + + odm_pipe->stream_res.opp->funcs->opp_program_fmt( + odm_pipe->stream_res.opp, + &stream->bit_depth_params, + &stream->clamping); + odm_pipe = odm_pipe->next_odm_pipe; + } + /* DCN3.1 FPGA Workaround * Need to enable HPO DP Stream Encoder before setting OTG master enable. * To do so, move calling function enable_stream_timing to only be done AFTER calling @@ -1548,30 +1573,6 @@ static enum dc_status apply_single_controller_ctx_to_hw( if (dc_is_dp_signal(pipe_ctx->stream->signal)) dp_source_sequence_trace(link, DPCD_SOURCE_SEQ_AFTER_CONNECT_DIG_FE_OTG); - pipe_ctx->stream_res.opp->funcs->opp_set_dyn_expansion( - pipe_ctx->stream_res.opp, - COLOR_SPACE_YCBCR601, - stream->timing.display_color_depth, - stream->signal); - - pipe_ctx->stream_res.opp->funcs->opp_program_fmt( - pipe_ctx->stream_res.opp, - &stream->bit_depth_params, - &stream->clamping); - while (odm_pipe) { - odm_pipe->stream_res.opp->funcs->opp_set_dyn_expansion( - odm_pipe->stream_res.opp, - COLOR_SPACE_YCBCR601, - stream->timing.display_color_depth, - stream->signal); - - odm_pipe->stream_res.opp->funcs->opp_program_fmt( - odm_pipe->stream_res.opp, - &stream->bit_depth_params, - &stream->clamping); - odm_pipe = odm_pipe->next_odm_pipe; - } - if (!stream->dpms_off) core_link_enable_stream(context, pipe_ctx); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c index f4f423d0b8c3..80595d7f060c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c @@ -940,6 +940,7 @@ static const struct hubbub_funcs hubbub1_funcs = { .program_watermarks = hubbub1_program_watermarks, .is_allow_self_refresh_enabled = hubbub1_is_allow_self_refresh_enabled, .allow_self_refresh_control = hubbub1_allow_self_refresh_control, + .verify_allow_pstate_change_high = hubbub1_verify_allow_pstate_change_high, }; void hubbub1_construct(struct hubbub *hubbub, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c index ea185c877323..fbff6beb78be 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c @@ -1311,6 +1311,20 @@ void hubp1_set_flip_int(struct hubp *hubp) return; } +/** + * hubp1_wait_pipe_read_start - wait for hubp ret path starting read. + * + * @hubp: hubp struct reference. + */ +void hubp1_wait_pipe_read_start(struct hubp *hubp) +{ + struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp); + + REG_WAIT(HUBPRET_READ_LINE_STATUS, + PIPE_READ_VBLANK, 0, + 1, 1000); +} + void hubp1_init(struct hubp *hubp) { //do nothing @@ -1345,6 +1359,7 @@ static const struct hubp_funcs dcn10_hubp_funcs = { .hubp_soft_reset = hubp1_soft_reset, .hubp_in_blank = hubp1_in_blank, .hubp_set_flip_int = hubp1_set_flip_int, + .hubp_wait_pipe_read_start = hubp1_wait_pipe_read_start, }; /*****************************************/ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h index 9cb8c383d673..0b17c2993ca5 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h @@ -76,6 +76,7 @@ SRI(DCSURF_SURFACE_CONTROL, HUBPREQ, id),\ SRI(DCSURF_SURFACE_FLIP_INTERRUPT, HUBPREQ, id),\ SRI(HUBPRET_CONTROL, HUBPRET, id),\ + SRI(HUBPRET_READ_LINE_STATUS, HUBPRET, id),\ SRI(DCN_EXPANSION_MODE, HUBPREQ, id),\ SRI(DCHUBP_REQ_SIZE_CONFIG, HUBP, id),\ SRI(DCHUBP_REQ_SIZE_CONFIG_C, HUBP, id),\ @@ -186,6 +187,7 @@ uint32_t DCSURF_SURFACE_CONTROL; \ uint32_t DCSURF_SURFACE_FLIP_INTERRUPT; \ uint32_t HUBPRET_CONTROL; \ + uint32_t HUBPRET_READ_LINE_STATUS; \ uint32_t DCN_EXPANSION_MODE; \ uint32_t DCHUBP_REQ_SIZE_CONFIG; \ uint32_t DCHUBP_REQ_SIZE_CONFIG_C; \ @@ -338,6 +340,7 @@ HUBP_SF(HUBPRET0_HUBPRET_CONTROL, DET_BUF_PLANE1_BASE_ADDRESS, mask_sh),\ HUBP_SF(HUBPRET0_HUBPRET_CONTROL, CROSSBAR_SRC_CB_B, mask_sh),\ HUBP_SF(HUBPRET0_HUBPRET_CONTROL, CROSSBAR_SRC_CR_R, mask_sh),\ + HUBP_SF(HUBPRET0_HUBPRET_READ_LINE_STATUS, PIPE_READ_VBLANK, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_EXPANSION_MODE, DRQ_EXPANSION_MODE, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_EXPANSION_MODE, PRQ_EXPANSION_MODE, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_EXPANSION_MODE, MRQ_EXPANSION_MODE, mask_sh),\ @@ -538,6 +541,7 @@ type DET_BUF_PLANE1_BASE_ADDRESS;\ type CROSSBAR_SRC_CB_B;\ type CROSSBAR_SRC_CR_R;\ + type PIPE_READ_VBLANK;\ type DRQ_EXPANSION_MODE;\ type PRQ_EXPANSION_MODE;\ type MRQ_EXPANSION_MODE;\ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c index 559aa45f27e7..c3e141c19a77 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c @@ -107,7 +107,7 @@ void dcn10_lock_all_pipes(struct dc *dc, * (un)locking. Also skip if pipe is disabled. */ if (pipe_ctx->top_pipe || - !pipe_ctx->stream || !pipe_ctx->plane_state || + !pipe_ctx->stream || !tg->funcs->is_tg_enabled(tg)) continue; @@ -1112,9 +1112,13 @@ static bool dcn10_hw_wa_force_recovery(struct dc *dc) void dcn10_verify_allow_pstate_change_high(struct dc *dc) { + struct hubbub *hubbub = dc->res_pool->hubbub; static bool should_log_hw_state; /* prevent hw state log by default */ - if (!hubbub1_verify_allow_pstate_change_high(dc->res_pool->hubbub)) { + if (!hubbub->funcs->verify_allow_pstate_change_high) + return; + + if (!hubbub->funcs->verify_allow_pstate_change_high(hubbub)) { int i = 0; if (should_log_hw_state) @@ -1123,8 +1127,8 @@ void dcn10_verify_allow_pstate_change_high(struct dc *dc) TRACE_DC_PIPE_STATE(pipe_ctx, i, MAX_PIPES); BREAK_TO_DEBUGGER(); if (dcn10_hw_wa_force_recovery(dc)) { - /*check again*/ - if (!hubbub1_verify_allow_pstate_change_high(dc->res_pool->hubbub)) + /*check again*/ + if (!hubbub->funcs->verify_allow_pstate_change_high(hubbub)) BREAK_TO_DEBUGGER(); } } @@ -3093,7 +3097,8 @@ static void dcn10_config_stereo_parameters( flags->PROGRAM_STEREO = 1; flags->PROGRAM_POLARITY = 1; - if (timing_3d_format == TIMING_3D_FORMAT_INBAND_FA || + if (timing_3d_format == TIMING_3D_FORMAT_FRAME_ALTERNATE || + timing_3d_format == TIMING_3D_FORMAT_INBAND_FA || timing_3d_format == TIMING_3D_FORMAT_DP_HDMI_INBAND_FA || timing_3d_format == TIMING_3D_FORMAT_SIDEBAND_FA) { enum display_dongle_type dongle = \ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c index 858b72149897..4048908dd265 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c @@ -70,68 +70,6 @@ #include "dce/dce_aux.h" #include "dce/dce_i2c.h" -const struct _vcs_dpi_ip_params_st dcn1_0_ip = { - .rob_buffer_size_kbytes = 64, - .det_buffer_size_kbytes = 164, - .dpte_buffer_size_in_pte_reqs_luma = 42, - .dpp_output_buffer_pixels = 2560, - .opp_output_buffer_lines = 1, - .pixel_chunk_size_kbytes = 8, - .pte_enable = 1, - .pte_chunk_size_kbytes = 2, - .meta_chunk_size_kbytes = 2, - .writeback_chunk_size_kbytes = 2, - .line_buffer_size_bits = 589824, - .max_line_buffer_lines = 12, - .IsLineBufferBppFixed = 0, - .LineBufferFixedBpp = -1, - .writeback_luma_buffer_size_kbytes = 12, - .writeback_chroma_buffer_size_kbytes = 8, - .max_num_dpp = 4, - .max_num_wb = 2, - .max_dchub_pscl_bw_pix_per_clk = 4, - .max_pscl_lb_bw_pix_per_clk = 2, - .max_lb_vscl_bw_pix_per_clk = 4, - .max_vscl_hscl_bw_pix_per_clk = 4, - .max_hscl_ratio = 4, - .max_vscl_ratio = 4, - .hscl_mults = 4, - .vscl_mults = 4, - .max_hscl_taps = 8, - .max_vscl_taps = 8, - .dispclk_ramp_margin_percent = 1, - .underscan_factor = 1.10, - .min_vblank_lines = 14, - .dppclk_delay_subtotal = 90, - .dispclk_delay_subtotal = 42, - .dcfclk_cstate_latency = 10, - .max_inter_dcn_tile_repeaters = 8, - .can_vstartup_lines_exceed_vsync_plus_back_porch_lines_minus_one = 0, - .bug_forcing_LC_req_same_size_fixed = 0, -}; - -const struct _vcs_dpi_soc_bounding_box_st dcn1_0_soc = { - .sr_exit_time_us = 9.0, - .sr_enter_plus_exit_time_us = 11.0, - .urgent_latency_us = 4.0, - .writeback_latency_us = 12.0, - .ideal_dram_bw_after_urgent_percent = 80.0, - .max_request_size_bytes = 256, - .downspread_percent = 0.5, - .dram_page_open_time_ns = 50.0, - .dram_rw_turnaround_time_ns = 17.5, - .dram_return_buffer_per_channel_bytes = 8192, - .round_trip_ping_latency_dcfclk_cycles = 128, - .urgent_out_of_order_return_per_channel_bytes = 256, - .channel_interleave_bytes = 256, - .num_banks = 8, - .num_chans = 2, - .vmm_page_size_bytes = 4096, - .dram_clock_change_latency_us = 17.0, - .writeback_dram_clock_change_latency_us = 23.0, - .return_bus_width_bytes = 64, -}; - #ifndef mmDP0_DP_DPHY_INTERNAL_CTRL #define mmDP0_DP_DPHY_INTERNAL_CTRL 0x210f #define mmDP0_DP_DPHY_INTERNAL_CTRL_BASE_IDX 2 @@ -925,6 +863,21 @@ static struct dce_hwseq *dcn10_hwseq_create( hws->wa.DEGVIDCN10_253 = true; hws->wa.false_optc_underflow = true; hws->wa.DEGVIDCN10_254 = true; + + if ((ctx->asic_id.chip_family == FAMILY_RV) && + ASICREV_IS_RAVEN2(ctx->asic_id.hw_internal_rev)) + switch (ctx->asic_id.pci_revision_id) { + case PRID_POLLOCK_94: + case PRID_POLLOCK_95: + case PRID_POLLOCK_E9: + case PRID_POLLOCK_EA: + case PRID_POLLOCK_EB: + hws->wa.wait_hubpret_read_start_during_mpo_transition = true; + break; + default: + hws->wa.wait_hubpret_read_start_during_mpo_transition = false; + break; + } } return hws; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.h index 633025ccb870..bf8e33cd8147 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.h @@ -27,6 +27,7 @@ #define __DC_RESOURCE_DCN10_H__ #include "core_types.h" +#include "dml/dcn10/dcn10_fpu.h" #define TO_DCN10_RES_POOL(pool)\ container_of(pool, struct dcn10_resource_pool, base) @@ -35,6 +36,9 @@ struct dc; struct resource_pool; struct _vcs_dpi_display_pipe_params_st; +extern struct _vcs_dpi_ip_params_st dcn1_0_ip; +extern struct _vcs_dpi_soc_bounding_box_st dcn1_0_soc; + struct dcn10_resource_pool { struct resource_pool base; }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile index 5fcaf78334ff..abaed2121feb 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile @@ -9,31 +9,6 @@ DCN20 = dcn20_resource.o dcn20_init.o dcn20_hwseq.o dcn20_dpp.o dcn20_dpp_cm.o d DCN20 += dcn20_dsc.o -ifdef CONFIG_X86 -CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -msse -endif - -ifdef CONFIG_PPC64 -CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -maltivec -endif - -ifdef CONFIG_CC_IS_GCC -ifeq ($(call cc-ifversion, -lt, 0701, y), y) -IS_OLD_GCC = 1 -endif -endif - -ifdef CONFIG_X86 -ifdef IS_OLD_GCC -# Stack alignment mismatch, proceed with caution. -# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3 -# (8B stack alignment). -CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -mpreferred-stack-boundary=4 -else -CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -msse2 -endif -endif - AMD_DAL_DCN20 = $(addprefix $(AMDDALPATH)/dc/dcn20/,$(DCN20)) AMD_DISPLAY_FILES += $(AMD_DAL_DCN20) diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c index dc1752e9f461..a665af19f201 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c @@ -707,17 +707,6 @@ bool hubp2_program_surface_flip_and_addr( REG_UPDATE(VMID_SETTINGS_0, VMID, address->vmid); - if (address->type == PLN_ADDR_TYPE_GRPH_STEREO) { - REG_UPDATE(DCSURF_FLIP_CONTROL, SURFACE_FLIP_MODE_FOR_STEREOSYNC, 0x1); - REG_UPDATE(DCSURF_FLIP_CONTROL, SURFACE_FLIP_IN_STEREOSYNC, 0x1); - - } else { - // turn off stereo if not in stereo - REG_UPDATE(DCSURF_FLIP_CONTROL, SURFACE_FLIP_MODE_FOR_STEREOSYNC, 0x0); - REG_UPDATE(DCSURF_FLIP_CONTROL, SURFACE_FLIP_IN_STEREOSYNC, 0x0); - } - - /* HW automatically latch rest of address register on write to * DCSURF_PRIMARY_SURFACE_ADDRESS if SURFACE_UPDATE_LOCK is not used @@ -942,10 +931,6 @@ void hubp2_set_blank_regs(struct hubp *hubp, bool blank) struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); uint32_t blank_en = blank ? 1 : 0; - REG_UPDATE_2(DCHUBP_CNTL, - HUBP_BLANK_EN, blank_en, - HUBP_TTU_DISABLE, blank_en); - if (blank) { uint32_t reg_val = REG_READ(DCHUBP_CNTL); @@ -958,9 +943,13 @@ void hubp2_set_blank_regs(struct hubp *hubp, bool blank) */ REG_WAIT(DCHUBP_CNTL, HUBP_NO_OUTSTANDING_REQ, 1, - 1, 200); + 1, 100000); } } + + REG_UPDATE_2(DCHUBP_CNTL, + HUBP_BLANK_EN, blank_en, + HUBP_TTU_DISABLE, 0); } void hubp2_cursor_set_position( diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c index 1ef880fed776..ab910deed481 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c @@ -1739,6 +1739,16 @@ void dcn20_program_front_end_for_ctx( || pipe->stream->update_flags.raw) && hws->funcs.program_all_writeback_pipes_in_tree) hws->funcs.program_all_writeback_pipes_in_tree(dc, pipe->stream, context); + + /* Avoid underflow by check of pipe line read when adding 2nd plane. */ + if (hws->wa.wait_hubpret_read_start_during_mpo_transition && + !pipe->top_pipe && + pipe->stream && + pipe->plane_res.hubp->funcs->hubp_wait_pipe_read_start && + dc->current_state->stream_status[0].plane_count == 1 && + context->stream_status[0].plane_count > 1) { + pipe->plane_res.hubp->funcs->hubp_wait_pipe_read_start(pipe->plane_res.hubp); + } } } diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c index dfe2e1c25a26..d473708d5399 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c @@ -63,7 +63,6 @@ #include "dcn20_dccg.h" #include "dcn20_vmid.h" #include "dc_link_ddc.h" -#include "dc_link_dp.h" #include "dce/dce_panel_cntl.h" #include "navi10_ip_offset.h" @@ -93,367 +92,6 @@ #define DC_LOGGER_INIT(logger) -struct _vcs_dpi_ip_params_st dcn2_0_ip = { - .odm_capable = 1, - .gpuvm_enable = 0, - .hostvm_enable = 0, - .gpuvm_max_page_table_levels = 4, - .hostvm_max_page_table_levels = 4, - .hostvm_cached_page_table_levels = 0, - .pte_group_size_bytes = 2048, - .num_dsc = 6, - .rob_buffer_size_kbytes = 168, - .det_buffer_size_kbytes = 164, - .dpte_buffer_size_in_pte_reqs_luma = 84, - .pde_proc_buffer_size_64k_reqs = 48, - .dpp_output_buffer_pixels = 2560, - .opp_output_buffer_lines = 1, - .pixel_chunk_size_kbytes = 8, - .pte_chunk_size_kbytes = 2, - .meta_chunk_size_kbytes = 2, - .writeback_chunk_size_kbytes = 2, - .line_buffer_size_bits = 789504, - .is_line_buffer_bpp_fixed = 0, - .line_buffer_fixed_bpp = 0, - .dcc_supported = true, - .max_line_buffer_lines = 12, - .writeback_luma_buffer_size_kbytes = 12, - .writeback_chroma_buffer_size_kbytes = 8, - .writeback_chroma_line_buffer_width_pixels = 4, - .writeback_max_hscl_ratio = 1, - .writeback_max_vscl_ratio = 1, - .writeback_min_hscl_ratio = 1, - .writeback_min_vscl_ratio = 1, - .writeback_max_hscl_taps = 12, - .writeback_max_vscl_taps = 12, - .writeback_line_buffer_luma_buffer_size = 0, - .writeback_line_buffer_chroma_buffer_size = 14643, - .cursor_buffer_size = 8, - .cursor_chunk_size = 2, - .max_num_otg = 6, - .max_num_dpp = 6, - .max_num_wb = 1, - .max_dchub_pscl_bw_pix_per_clk = 4, - .max_pscl_lb_bw_pix_per_clk = 2, - .max_lb_vscl_bw_pix_per_clk = 4, - .max_vscl_hscl_bw_pix_per_clk = 4, - .max_hscl_ratio = 8, - .max_vscl_ratio = 8, - .hscl_mults = 4, - .vscl_mults = 4, - .max_hscl_taps = 8, - .max_vscl_taps = 8, - .dispclk_ramp_margin_percent = 1, - .underscan_factor = 1.10, - .min_vblank_lines = 32, // - .dppclk_delay_subtotal = 77, // - .dppclk_delay_scl_lb_only = 16, - .dppclk_delay_scl = 50, - .dppclk_delay_cnvc_formatter = 8, - .dppclk_delay_cnvc_cursor = 6, - .dispclk_delay_subtotal = 87, // - .dcfclk_cstate_latency = 10, // SRExitTime - .max_inter_dcn_tile_repeaters = 8, - .xfc_supported = true, - .xfc_fill_bw_overhead_percent = 10.0, - .xfc_fill_constant_bytes = 0, - .number_of_cursors = 1, -}; - -static struct _vcs_dpi_ip_params_st dcn2_0_nv14_ip = { - .odm_capable = 1, - .gpuvm_enable = 0, - .hostvm_enable = 0, - .gpuvm_max_page_table_levels = 4, - .hostvm_max_page_table_levels = 4, - .hostvm_cached_page_table_levels = 0, - .num_dsc = 5, - .rob_buffer_size_kbytes = 168, - .det_buffer_size_kbytes = 164, - .dpte_buffer_size_in_pte_reqs_luma = 84, - .dpte_buffer_size_in_pte_reqs_chroma = 42,//todo - .dpp_output_buffer_pixels = 2560, - .opp_output_buffer_lines = 1, - .pixel_chunk_size_kbytes = 8, - .pte_enable = 1, - .max_page_table_levels = 4, - .pte_chunk_size_kbytes = 2, - .meta_chunk_size_kbytes = 2, - .writeback_chunk_size_kbytes = 2, - .line_buffer_size_bits = 789504, - .is_line_buffer_bpp_fixed = 0, - .line_buffer_fixed_bpp = 0, - .dcc_supported = true, - .max_line_buffer_lines = 12, - .writeback_luma_buffer_size_kbytes = 12, - .writeback_chroma_buffer_size_kbytes = 8, - .writeback_chroma_line_buffer_width_pixels = 4, - .writeback_max_hscl_ratio = 1, - .writeback_max_vscl_ratio = 1, - .writeback_min_hscl_ratio = 1, - .writeback_min_vscl_ratio = 1, - .writeback_max_hscl_taps = 12, - .writeback_max_vscl_taps = 12, - .writeback_line_buffer_luma_buffer_size = 0, - .writeback_line_buffer_chroma_buffer_size = 14643, - .cursor_buffer_size = 8, - .cursor_chunk_size = 2, - .max_num_otg = 5, - .max_num_dpp = 5, - .max_num_wb = 1, - .max_dchub_pscl_bw_pix_per_clk = 4, - .max_pscl_lb_bw_pix_per_clk = 2, - .max_lb_vscl_bw_pix_per_clk = 4, - .max_vscl_hscl_bw_pix_per_clk = 4, - .max_hscl_ratio = 8, - .max_vscl_ratio = 8, - .hscl_mults = 4, - .vscl_mults = 4, - .max_hscl_taps = 8, - .max_vscl_taps = 8, - .dispclk_ramp_margin_percent = 1, - .underscan_factor = 1.10, - .min_vblank_lines = 32, // - .dppclk_delay_subtotal = 77, // - .dppclk_delay_scl_lb_only = 16, - .dppclk_delay_scl = 50, - .dppclk_delay_cnvc_formatter = 8, - .dppclk_delay_cnvc_cursor = 6, - .dispclk_delay_subtotal = 87, // - .dcfclk_cstate_latency = 10, // SRExitTime - .max_inter_dcn_tile_repeaters = 8, - .xfc_supported = true, - .xfc_fill_bw_overhead_percent = 10.0, - .xfc_fill_constant_bytes = 0, - .ptoi_supported = 0, - .number_of_cursors = 1, -}; - -static struct _vcs_dpi_soc_bounding_box_st dcn2_0_soc = { - /* Defaults that get patched on driver load from firmware. */ - .clock_limits = { - { - .state = 0, - .dcfclk_mhz = 560.0, - .fabricclk_mhz = 560.0, - .dispclk_mhz = 513.0, - .dppclk_mhz = 513.0, - .phyclk_mhz = 540.0, - .socclk_mhz = 560.0, - .dscclk_mhz = 171.0, - .dram_speed_mts = 8960.0, - }, - { - .state = 1, - .dcfclk_mhz = 694.0, - .fabricclk_mhz = 694.0, - .dispclk_mhz = 642.0, - .dppclk_mhz = 642.0, - .phyclk_mhz = 600.0, - .socclk_mhz = 694.0, - .dscclk_mhz = 214.0, - .dram_speed_mts = 11104.0, - }, - { - .state = 2, - .dcfclk_mhz = 875.0, - .fabricclk_mhz = 875.0, - .dispclk_mhz = 734.0, - .dppclk_mhz = 734.0, - .phyclk_mhz = 810.0, - .socclk_mhz = 875.0, - .dscclk_mhz = 245.0, - .dram_speed_mts = 14000.0, - }, - { - .state = 3, - .dcfclk_mhz = 1000.0, - .fabricclk_mhz = 1000.0, - .dispclk_mhz = 1100.0, - .dppclk_mhz = 1100.0, - .phyclk_mhz = 810.0, - .socclk_mhz = 1000.0, - .dscclk_mhz = 367.0, - .dram_speed_mts = 16000.0, - }, - { - .state = 4, - .dcfclk_mhz = 1200.0, - .fabricclk_mhz = 1200.0, - .dispclk_mhz = 1284.0, - .dppclk_mhz = 1284.0, - .phyclk_mhz = 810.0, - .socclk_mhz = 1200.0, - .dscclk_mhz = 428.0, - .dram_speed_mts = 16000.0, - }, - /*Extra state, no dispclk ramping*/ - { - .state = 5, - .dcfclk_mhz = 1200.0, - .fabricclk_mhz = 1200.0, - .dispclk_mhz = 1284.0, - .dppclk_mhz = 1284.0, - .phyclk_mhz = 810.0, - .socclk_mhz = 1200.0, - .dscclk_mhz = 428.0, - .dram_speed_mts = 16000.0, - }, - }, - .num_states = 5, - .sr_exit_time_us = 8.6, - .sr_enter_plus_exit_time_us = 10.9, - .urgent_latency_us = 4.0, - .urgent_latency_pixel_data_only_us = 4.0, - .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, - .urgent_latency_vm_data_only_us = 4.0, - .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, - .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, - .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, - .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 40.0, - .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 40.0, - .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, - .max_avg_sdp_bw_use_normal_percent = 40.0, - .max_avg_dram_bw_use_normal_percent = 40.0, - .writeback_latency_us = 12.0, - .ideal_dram_bw_after_urgent_percent = 40.0, - .max_request_size_bytes = 256, - .dram_channel_width_bytes = 2, - .fabric_datapath_to_dcn_data_return_bytes = 64, - .dcn_downspread_percent = 0.5, - .downspread_percent = 0.38, - .dram_page_open_time_ns = 50.0, - .dram_rw_turnaround_time_ns = 17.5, - .dram_return_buffer_per_channel_bytes = 8192, - .round_trip_ping_latency_dcfclk_cycles = 131, - .urgent_out_of_order_return_per_channel_bytes = 256, - .channel_interleave_bytes = 256, - .num_banks = 8, - .num_chans = 16, - .vmm_page_size_bytes = 4096, - .dram_clock_change_latency_us = 404.0, - .dummy_pstate_latency_us = 5.0, - .writeback_dram_clock_change_latency_us = 23.0, - .return_bus_width_bytes = 64, - .dispclk_dppclk_vco_speed_mhz = 3850, - .xfc_bus_transport_time_us = 20, - .xfc_xbuf_latency_tolerance_us = 4, - .use_urgent_burst_bw = 0 -}; - -static struct _vcs_dpi_soc_bounding_box_st dcn2_0_nv14_soc = { - .clock_limits = { - { - .state = 0, - .dcfclk_mhz = 560.0, - .fabricclk_mhz = 560.0, - .dispclk_mhz = 513.0, - .dppclk_mhz = 513.0, - .phyclk_mhz = 540.0, - .socclk_mhz = 560.0, - .dscclk_mhz = 171.0, - .dram_speed_mts = 8960.0, - }, - { - .state = 1, - .dcfclk_mhz = 694.0, - .fabricclk_mhz = 694.0, - .dispclk_mhz = 642.0, - .dppclk_mhz = 642.0, - .phyclk_mhz = 600.0, - .socclk_mhz = 694.0, - .dscclk_mhz = 214.0, - .dram_speed_mts = 11104.0, - }, - { - .state = 2, - .dcfclk_mhz = 875.0, - .fabricclk_mhz = 875.0, - .dispclk_mhz = 734.0, - .dppclk_mhz = 734.0, - .phyclk_mhz = 810.0, - .socclk_mhz = 875.0, - .dscclk_mhz = 245.0, - .dram_speed_mts = 14000.0, - }, - { - .state = 3, - .dcfclk_mhz = 1000.0, - .fabricclk_mhz = 1000.0, - .dispclk_mhz = 1100.0, - .dppclk_mhz = 1100.0, - .phyclk_mhz = 810.0, - .socclk_mhz = 1000.0, - .dscclk_mhz = 367.0, - .dram_speed_mts = 16000.0, - }, - { - .state = 4, - .dcfclk_mhz = 1200.0, - .fabricclk_mhz = 1200.0, - .dispclk_mhz = 1284.0, - .dppclk_mhz = 1284.0, - .phyclk_mhz = 810.0, - .socclk_mhz = 1200.0, - .dscclk_mhz = 428.0, - .dram_speed_mts = 16000.0, - }, - /*Extra state, no dispclk ramping*/ - { - .state = 5, - .dcfclk_mhz = 1200.0, - .fabricclk_mhz = 1200.0, - .dispclk_mhz = 1284.0, - .dppclk_mhz = 1284.0, - .phyclk_mhz = 810.0, - .socclk_mhz = 1200.0, - .dscclk_mhz = 428.0, - .dram_speed_mts = 16000.0, - }, - }, - .num_states = 5, - .sr_exit_time_us = 11.6, - .sr_enter_plus_exit_time_us = 13.9, - .urgent_latency_us = 4.0, - .urgent_latency_pixel_data_only_us = 4.0, - .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, - .urgent_latency_vm_data_only_us = 4.0, - .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, - .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, - .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, - .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 40.0, - .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 40.0, - .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, - .max_avg_sdp_bw_use_normal_percent = 40.0, - .max_avg_dram_bw_use_normal_percent = 40.0, - .writeback_latency_us = 12.0, - .ideal_dram_bw_after_urgent_percent = 40.0, - .max_request_size_bytes = 256, - .dram_channel_width_bytes = 2, - .fabric_datapath_to_dcn_data_return_bytes = 64, - .dcn_downspread_percent = 0.5, - .downspread_percent = 0.38, - .dram_page_open_time_ns = 50.0, - .dram_rw_turnaround_time_ns = 17.5, - .dram_return_buffer_per_channel_bytes = 8192, - .round_trip_ping_latency_dcfclk_cycles = 131, - .urgent_out_of_order_return_per_channel_bytes = 256, - .channel_interleave_bytes = 256, - .num_banks = 8, - .num_chans = 8, - .vmm_page_size_bytes = 4096, - .dram_clock_change_latency_us = 404.0, - .dummy_pstate_latency_us = 5.0, - .writeback_dram_clock_change_latency_us = 23.0, - .return_bus_width_bytes = 64, - .dispclk_dppclk_vco_speed_mhz = 3850, - .xfc_bus_transport_time_us = 20, - .xfc_xbuf_latency_tolerance_us = 4, - .use_urgent_burst_bw = 0 -}; - -static struct _vcs_dpi_soc_bounding_box_st dcn2_0_nv12_soc = { 0 }; - #ifndef mmDP0_DP_DPHY_INTERNAL_CTRL #define mmDP0_DP_DPHY_INTERNAL_CTRL 0x210f #define mmDP0_DP_DPHY_INTERNAL_CTRL_BASE_IDX 2 @@ -1810,69 +1448,6 @@ enum dc_status dcn20_remove_stream_from_ctx(struct dc *dc, struct dc_state *new_ return result; } - -static void swizzle_to_dml_params( - enum swizzle_mode_values swizzle, - unsigned int *sw_mode) -{ - switch (swizzle) { - case DC_SW_LINEAR: - *sw_mode = dm_sw_linear; - break; - case DC_SW_4KB_S: - *sw_mode = dm_sw_4kb_s; - break; - case DC_SW_4KB_S_X: - *sw_mode = dm_sw_4kb_s_x; - break; - case DC_SW_4KB_D: - *sw_mode = dm_sw_4kb_d; - break; - case DC_SW_4KB_D_X: - *sw_mode = dm_sw_4kb_d_x; - break; - case DC_SW_64KB_S: - *sw_mode = dm_sw_64kb_s; - break; - case DC_SW_64KB_S_X: - *sw_mode = dm_sw_64kb_s_x; - break; - case DC_SW_64KB_S_T: - *sw_mode = dm_sw_64kb_s_t; - break; - case DC_SW_64KB_D: - *sw_mode = dm_sw_64kb_d; - break; - case DC_SW_64KB_D_X: - *sw_mode = dm_sw_64kb_d_x; - break; - case DC_SW_64KB_D_T: - *sw_mode = dm_sw_64kb_d_t; - break; - case DC_SW_64KB_R_X: - *sw_mode = dm_sw_64kb_r_x; - break; - case DC_SW_VAR_S: - *sw_mode = dm_sw_var_s; - break; - case DC_SW_VAR_S_X: - *sw_mode = dm_sw_var_s_x; - break; - case DC_SW_VAR_D: - *sw_mode = dm_sw_var_d; - break; - case DC_SW_VAR_D_X: - *sw_mode = dm_sw_var_d_x; - break; - case DC_SW_VAR_R_X: - *sw_mode = dm_sw_var_r_x; - break; - default: - ASSERT(0); /* Not supported */ - break; - } -} - bool dcn20_split_stream_for_odm( const struct dc *dc, struct resource_context *res_ctx, @@ -1988,394 +1563,6 @@ void dcn20_split_stream_for_mpc( ASSERT(primary_pipe->plane_state); } -int dcn20_populate_dml_pipes_from_context( - struct dc *dc, - struct dc_state *context, - display_e2e_pipe_params_st *pipes, - bool fast_validate) -{ - int pipe_cnt, i; - bool synchronized_vblank = true; - struct resource_context *res_ctx = &context->res_ctx; - - for (i = 0, pipe_cnt = -1; i < dc->res_pool->pipe_count; i++) { - if (!res_ctx->pipe_ctx[i].stream) - continue; - - if (pipe_cnt < 0) { - pipe_cnt = i; - continue; - } - - if (res_ctx->pipe_ctx[pipe_cnt].stream == res_ctx->pipe_ctx[i].stream) - continue; - - if (dc->debug.disable_timing_sync || - (!resource_are_streams_timing_synchronizable( - res_ctx->pipe_ctx[pipe_cnt].stream, - res_ctx->pipe_ctx[i].stream) && - !resource_are_vblanks_synchronizable( - res_ctx->pipe_ctx[pipe_cnt].stream, - res_ctx->pipe_ctx[i].stream))) { - synchronized_vblank = false; - break; - } - } - - for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { - struct dc_crtc_timing *timing = &res_ctx->pipe_ctx[i].stream->timing; - unsigned int v_total; - unsigned int front_porch; - int output_bpc; - struct audio_check aud_check = {0}; - - if (!res_ctx->pipe_ctx[i].stream) - continue; - - v_total = timing->v_total; - front_porch = timing->v_front_porch; - - /* todo: - pipes[pipe_cnt].pipe.src.dynamic_metadata_enable = 0; - pipes[pipe_cnt].pipe.src.dcc = 0; - pipes[pipe_cnt].pipe.src.vm = 0;*/ - - pipes[pipe_cnt].clks_cfg.refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000.0; - - pipes[pipe_cnt].dout.dsc_enable = res_ctx->pipe_ctx[i].stream->timing.flags.DSC; - /* todo: rotation?*/ - pipes[pipe_cnt].dout.dsc_slices = res_ctx->pipe_ctx[i].stream->timing.dsc_cfg.num_slices_h; - if (res_ctx->pipe_ctx[i].stream->use_dynamic_meta) { - pipes[pipe_cnt].pipe.src.dynamic_metadata_enable = true; - /* 1/2 vblank */ - pipes[pipe_cnt].pipe.src.dynamic_metadata_lines_before_active = - (v_total - timing->v_addressable - - timing->v_border_top - timing->v_border_bottom) / 2; - /* 36 bytes dp, 32 hdmi */ - pipes[pipe_cnt].pipe.src.dynamic_metadata_xmit_bytes = - dc_is_dp_signal(res_ctx->pipe_ctx[i].stream->signal) ? 36 : 32; - } - pipes[pipe_cnt].pipe.src.dcc = false; - pipes[pipe_cnt].pipe.src.dcc_rate = 1; - pipes[pipe_cnt].pipe.dest.synchronized_vblank_all_planes = synchronized_vblank; - pipes[pipe_cnt].pipe.dest.hblank_start = timing->h_total - timing->h_front_porch; - pipes[pipe_cnt].pipe.dest.hblank_end = pipes[pipe_cnt].pipe.dest.hblank_start - - timing->h_addressable - - timing->h_border_left - - timing->h_border_right; - pipes[pipe_cnt].pipe.dest.vblank_start = v_total - front_porch; - pipes[pipe_cnt].pipe.dest.vblank_end = pipes[pipe_cnt].pipe.dest.vblank_start - - timing->v_addressable - - timing->v_border_top - - timing->v_border_bottom; - pipes[pipe_cnt].pipe.dest.htotal = timing->h_total; - pipes[pipe_cnt].pipe.dest.vtotal = v_total; - pipes[pipe_cnt].pipe.dest.hactive = - timing->h_addressable + timing->h_border_left + timing->h_border_right; - pipes[pipe_cnt].pipe.dest.vactive = - timing->v_addressable + timing->v_border_top + timing->v_border_bottom; - pipes[pipe_cnt].pipe.dest.interlaced = timing->flags.INTERLACE; - pipes[pipe_cnt].pipe.dest.pixel_rate_mhz = timing->pix_clk_100hz/10000.0; - if (timing->timing_3d_format == TIMING_3D_FORMAT_HW_FRAME_PACKING) - pipes[pipe_cnt].pipe.dest.pixel_rate_mhz *= 2; - pipes[pipe_cnt].pipe.dest.otg_inst = res_ctx->pipe_ctx[i].stream_res.tg->inst; - pipes[pipe_cnt].dout.dp_lanes = 4; - pipes[pipe_cnt].dout.is_virtual = 0; - pipes[pipe_cnt].pipe.dest.vtotal_min = res_ctx->pipe_ctx[i].stream->adjust.v_total_min; - pipes[pipe_cnt].pipe.dest.vtotal_max = res_ctx->pipe_ctx[i].stream->adjust.v_total_max; - switch (get_num_odm_splits(&res_ctx->pipe_ctx[i])) { - case 1: - pipes[pipe_cnt].pipe.dest.odm_combine = dm_odm_combine_mode_2to1; - break; - case 3: - pipes[pipe_cnt].pipe.dest.odm_combine = dm_odm_combine_mode_4to1; - break; - default: - pipes[pipe_cnt].pipe.dest.odm_combine = dm_odm_combine_mode_disabled; - } - pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].pipe_idx; - if (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state - == res_ctx->pipe_ctx[i].plane_state) { - struct pipe_ctx *first_pipe = res_ctx->pipe_ctx[i].top_pipe; - int split_idx = 0; - - while (first_pipe->top_pipe && first_pipe->top_pipe->plane_state - == res_ctx->pipe_ctx[i].plane_state) { - first_pipe = first_pipe->top_pipe; - split_idx++; - } - /* Treat 4to1 mpc combine as an mpo of 2 2-to-1 combines */ - if (split_idx == 0) - pipes[pipe_cnt].pipe.src.hsplit_grp = first_pipe->pipe_idx; - else if (split_idx == 1) - pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].pipe_idx; - else if (split_idx == 2) - pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].top_pipe->pipe_idx; - } else if (res_ctx->pipe_ctx[i].prev_odm_pipe) { - struct pipe_ctx *first_pipe = res_ctx->pipe_ctx[i].prev_odm_pipe; - - while (first_pipe->prev_odm_pipe) - first_pipe = first_pipe->prev_odm_pipe; - pipes[pipe_cnt].pipe.src.hsplit_grp = first_pipe->pipe_idx; - } - - switch (res_ctx->pipe_ctx[i].stream->signal) { - case SIGNAL_TYPE_DISPLAY_PORT_MST: - case SIGNAL_TYPE_DISPLAY_PORT: - pipes[pipe_cnt].dout.output_type = dm_dp; - break; - case SIGNAL_TYPE_EDP: - pipes[pipe_cnt].dout.output_type = dm_edp; - break; - case SIGNAL_TYPE_HDMI_TYPE_A: - case SIGNAL_TYPE_DVI_SINGLE_LINK: - case SIGNAL_TYPE_DVI_DUAL_LINK: - pipes[pipe_cnt].dout.output_type = dm_hdmi; - break; - default: - /* In case there is no signal, set dp with 4 lanes to allow max config */ - pipes[pipe_cnt].dout.is_virtual = 1; - pipes[pipe_cnt].dout.output_type = dm_dp; - pipes[pipe_cnt].dout.dp_lanes = 4; - } - - switch (res_ctx->pipe_ctx[i].stream->timing.display_color_depth) { - case COLOR_DEPTH_666: - output_bpc = 6; - break; - case COLOR_DEPTH_888: - output_bpc = 8; - break; - case COLOR_DEPTH_101010: - output_bpc = 10; - break; - case COLOR_DEPTH_121212: - output_bpc = 12; - break; - case COLOR_DEPTH_141414: - output_bpc = 14; - break; - case COLOR_DEPTH_161616: - output_bpc = 16; - break; - case COLOR_DEPTH_999: - output_bpc = 9; - break; - case COLOR_DEPTH_111111: - output_bpc = 11; - break; - default: - output_bpc = 8; - break; - } - - switch (res_ctx->pipe_ctx[i].stream->timing.pixel_encoding) { - case PIXEL_ENCODING_RGB: - case PIXEL_ENCODING_YCBCR444: - pipes[pipe_cnt].dout.output_format = dm_444; - pipes[pipe_cnt].dout.output_bpp = output_bpc * 3; - break; - case PIXEL_ENCODING_YCBCR420: - pipes[pipe_cnt].dout.output_format = dm_420; - pipes[pipe_cnt].dout.output_bpp = (output_bpc * 3.0) / 2; - break; - case PIXEL_ENCODING_YCBCR422: - if (res_ctx->pipe_ctx[i].stream->timing.flags.DSC && - !res_ctx->pipe_ctx[i].stream->timing.dsc_cfg.ycbcr422_simple) - pipes[pipe_cnt].dout.output_format = dm_n422; - else - pipes[pipe_cnt].dout.output_format = dm_s422; - pipes[pipe_cnt].dout.output_bpp = output_bpc * 2; - break; - default: - pipes[pipe_cnt].dout.output_format = dm_444; - pipes[pipe_cnt].dout.output_bpp = output_bpc * 3; - } - - if (res_ctx->pipe_ctx[i].stream->timing.flags.DSC) - pipes[pipe_cnt].dout.output_bpp = res_ctx->pipe_ctx[i].stream->timing.dsc_cfg.bits_per_pixel / 16.0; - - /* todo: default max for now, until there is logic reflecting this in dc*/ - pipes[pipe_cnt].dout.dsc_input_bpc = 12; - /*fill up the audio sample rate (unit in kHz)*/ - get_audio_check(&res_ctx->pipe_ctx[i].stream->audio_info, &aud_check); - pipes[pipe_cnt].dout.max_audio_sample_rate = aud_check.max_audiosample_rate / 1000; - /* - * For graphic plane, cursor number is 1, nv12 is 0 - * bw calculations due to cursor on/off - */ - if (res_ctx->pipe_ctx[i].plane_state && - res_ctx->pipe_ctx[i].plane_state->address.type == PLN_ADDR_TYPE_VIDEO_PROGRESSIVE) - pipes[pipe_cnt].pipe.src.num_cursors = 0; - else - pipes[pipe_cnt].pipe.src.num_cursors = dc->dml.ip.number_of_cursors; - - pipes[pipe_cnt].pipe.src.cur0_src_width = 256; - pipes[pipe_cnt].pipe.src.cur0_bpp = dm_cur_32bit; - - if (!res_ctx->pipe_ctx[i].plane_state) { - pipes[pipe_cnt].pipe.src.is_hsplit = pipes[pipe_cnt].pipe.dest.odm_combine != dm_odm_combine_mode_disabled; - pipes[pipe_cnt].pipe.src.source_scan = dm_horz; - pipes[pipe_cnt].pipe.src.sw_mode = dm_sw_4kb_s; - pipes[pipe_cnt].pipe.src.macro_tile_size = dm_64k_tile; - pipes[pipe_cnt].pipe.src.viewport_width = timing->h_addressable; - if (pipes[pipe_cnt].pipe.src.viewport_width > 1920) - pipes[pipe_cnt].pipe.src.viewport_width = 1920; - pipes[pipe_cnt].pipe.src.viewport_height = timing->v_addressable; - if (pipes[pipe_cnt].pipe.src.viewport_height > 1080) - pipes[pipe_cnt].pipe.src.viewport_height = 1080; - pipes[pipe_cnt].pipe.src.surface_height_y = pipes[pipe_cnt].pipe.src.viewport_height; - pipes[pipe_cnt].pipe.src.surface_width_y = pipes[pipe_cnt].pipe.src.viewport_width; - pipes[pipe_cnt].pipe.src.surface_height_c = pipes[pipe_cnt].pipe.src.viewport_height; - pipes[pipe_cnt].pipe.src.surface_width_c = pipes[pipe_cnt].pipe.src.viewport_width; - pipes[pipe_cnt].pipe.src.data_pitch = ((pipes[pipe_cnt].pipe.src.viewport_width + 255) / 256) * 256; - pipes[pipe_cnt].pipe.src.source_format = dm_444_32; - pipes[pipe_cnt].pipe.dest.recout_width = pipes[pipe_cnt].pipe.src.viewport_width; /*vp_width/hratio*/ - pipes[pipe_cnt].pipe.dest.recout_height = pipes[pipe_cnt].pipe.src.viewport_height; /*vp_height/vratio*/ - pipes[pipe_cnt].pipe.dest.full_recout_width = pipes[pipe_cnt].pipe.dest.recout_width; /*when is_hsplit != 1*/ - pipes[pipe_cnt].pipe.dest.full_recout_height = pipes[pipe_cnt].pipe.dest.recout_height; /*when is_hsplit != 1*/ - pipes[pipe_cnt].pipe.scale_ratio_depth.lb_depth = dm_lb_16; - pipes[pipe_cnt].pipe.scale_ratio_depth.hscl_ratio = 1.0; - pipes[pipe_cnt].pipe.scale_ratio_depth.vscl_ratio = 1.0; - pipes[pipe_cnt].pipe.scale_ratio_depth.scl_enable = 0; /*Lb only or Full scl*/ - pipes[pipe_cnt].pipe.scale_taps.htaps = 1; - pipes[pipe_cnt].pipe.scale_taps.vtaps = 1; - pipes[pipe_cnt].pipe.dest.vtotal_min = v_total; - pipes[pipe_cnt].pipe.dest.vtotal_max = v_total; - - if (pipes[pipe_cnt].pipe.dest.odm_combine == dm_odm_combine_mode_2to1) { - pipes[pipe_cnt].pipe.src.viewport_width /= 2; - pipes[pipe_cnt].pipe.dest.recout_width /= 2; - } else if (pipes[pipe_cnt].pipe.dest.odm_combine == dm_odm_combine_mode_4to1) { - pipes[pipe_cnt].pipe.src.viewport_width /= 4; - pipes[pipe_cnt].pipe.dest.recout_width /= 4; - } - } else { - struct dc_plane_state *pln = res_ctx->pipe_ctx[i].plane_state; - struct scaler_data *scl = &res_ctx->pipe_ctx[i].plane_res.scl_data; - - pipes[pipe_cnt].pipe.src.immediate_flip = pln->flip_immediate; - pipes[pipe_cnt].pipe.src.is_hsplit = (res_ctx->pipe_ctx[i].bottom_pipe && res_ctx->pipe_ctx[i].bottom_pipe->plane_state == pln) - || (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state == pln) - || pipes[pipe_cnt].pipe.dest.odm_combine != dm_odm_combine_mode_disabled; - - /* stereo is not split */ - if (pln->stereo_format == PLANE_STEREO_FORMAT_SIDE_BY_SIDE || - pln->stereo_format == PLANE_STEREO_FORMAT_TOP_AND_BOTTOM) { - pipes[pipe_cnt].pipe.src.is_hsplit = false; - pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].pipe_idx; - } - - pipes[pipe_cnt].pipe.src.source_scan = pln->rotation == ROTATION_ANGLE_90 - || pln->rotation == ROTATION_ANGLE_270 ? dm_vert : dm_horz; - pipes[pipe_cnt].pipe.src.viewport_y_y = scl->viewport.y; - pipes[pipe_cnt].pipe.src.viewport_y_c = scl->viewport_c.y; - pipes[pipe_cnt].pipe.src.viewport_width = scl->viewport.width; - pipes[pipe_cnt].pipe.src.viewport_width_c = scl->viewport_c.width; - pipes[pipe_cnt].pipe.src.viewport_height = scl->viewport.height; - pipes[pipe_cnt].pipe.src.viewport_height_c = scl->viewport_c.height; - pipes[pipe_cnt].pipe.src.viewport_width_max = pln->src_rect.width; - pipes[pipe_cnt].pipe.src.viewport_height_max = pln->src_rect.height; - pipes[pipe_cnt].pipe.src.surface_width_y = pln->plane_size.surface_size.width; - pipes[pipe_cnt].pipe.src.surface_height_y = pln->plane_size.surface_size.height; - pipes[pipe_cnt].pipe.src.surface_width_c = pln->plane_size.chroma_size.width; - pipes[pipe_cnt].pipe.src.surface_height_c = pln->plane_size.chroma_size.height; - if (pln->format == SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA - || pln->format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN) { - pipes[pipe_cnt].pipe.src.data_pitch = pln->plane_size.surface_pitch; - pipes[pipe_cnt].pipe.src.data_pitch_c = pln->plane_size.chroma_pitch; - pipes[pipe_cnt].pipe.src.meta_pitch = pln->dcc.meta_pitch; - pipes[pipe_cnt].pipe.src.meta_pitch_c = pln->dcc.meta_pitch_c; - } else { - pipes[pipe_cnt].pipe.src.data_pitch = pln->plane_size.surface_pitch; - pipes[pipe_cnt].pipe.src.meta_pitch = pln->dcc.meta_pitch; - } - pipes[pipe_cnt].pipe.src.dcc = pln->dcc.enable; - pipes[pipe_cnt].pipe.dest.recout_width = scl->recout.width; - pipes[pipe_cnt].pipe.dest.recout_height = scl->recout.height; - pipes[pipe_cnt].pipe.dest.full_recout_height = scl->recout.height; - pipes[pipe_cnt].pipe.dest.full_recout_width = scl->recout.width; - if (pipes[pipe_cnt].pipe.dest.odm_combine == dm_odm_combine_mode_2to1) - pipes[pipe_cnt].pipe.dest.full_recout_width *= 2; - else if (pipes[pipe_cnt].pipe.dest.odm_combine == dm_odm_combine_mode_4to1) - pipes[pipe_cnt].pipe.dest.full_recout_width *= 4; - else { - struct pipe_ctx *split_pipe = res_ctx->pipe_ctx[i].bottom_pipe; - - while (split_pipe && split_pipe->plane_state == pln) { - pipes[pipe_cnt].pipe.dest.full_recout_width += split_pipe->plane_res.scl_data.recout.width; - split_pipe = split_pipe->bottom_pipe; - } - split_pipe = res_ctx->pipe_ctx[i].top_pipe; - while (split_pipe && split_pipe->plane_state == pln) { - pipes[pipe_cnt].pipe.dest.full_recout_width += split_pipe->plane_res.scl_data.recout.width; - split_pipe = split_pipe->top_pipe; - } - } - - pipes[pipe_cnt].pipe.scale_ratio_depth.lb_depth = dm_lb_16; - pipes[pipe_cnt].pipe.scale_ratio_depth.hscl_ratio = (double) scl->ratios.horz.value / (1ULL<<32); - pipes[pipe_cnt].pipe.scale_ratio_depth.hscl_ratio_c = (double) scl->ratios.horz_c.value / (1ULL<<32); - pipes[pipe_cnt].pipe.scale_ratio_depth.vscl_ratio = (double) scl->ratios.vert.value / (1ULL<<32); - pipes[pipe_cnt].pipe.scale_ratio_depth.vscl_ratio_c = (double) scl->ratios.vert_c.value / (1ULL<<32); - pipes[pipe_cnt].pipe.scale_ratio_depth.scl_enable = - scl->ratios.vert.value != dc_fixpt_one.value - || scl->ratios.horz.value != dc_fixpt_one.value - || scl->ratios.vert_c.value != dc_fixpt_one.value - || scl->ratios.horz_c.value != dc_fixpt_one.value /*Lb only or Full scl*/ - || dc->debug.always_scale; /*support always scale*/ - pipes[pipe_cnt].pipe.scale_taps.htaps = scl->taps.h_taps; - pipes[pipe_cnt].pipe.scale_taps.htaps_c = scl->taps.h_taps_c; - pipes[pipe_cnt].pipe.scale_taps.vtaps = scl->taps.v_taps; - pipes[pipe_cnt].pipe.scale_taps.vtaps_c = scl->taps.v_taps_c; - - pipes[pipe_cnt].pipe.src.macro_tile_size = - swizzle_mode_to_macro_tile_size(pln->tiling_info.gfx9.swizzle); - swizzle_to_dml_params(pln->tiling_info.gfx9.swizzle, - &pipes[pipe_cnt].pipe.src.sw_mode); - - switch (pln->format) { - case SURFACE_PIXEL_FORMAT_VIDEO_420_YCbCr: - case SURFACE_PIXEL_FORMAT_VIDEO_420_YCrCb: - pipes[pipe_cnt].pipe.src.source_format = dm_420_8; - break; - case SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCbCr: - case SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCrCb: - pipes[pipe_cnt].pipe.src.source_format = dm_420_10; - break; - case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616: - case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616: - case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616F: - case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F: - pipes[pipe_cnt].pipe.src.source_format = dm_444_64; - break; - case SURFACE_PIXEL_FORMAT_GRPH_ARGB1555: - case SURFACE_PIXEL_FORMAT_GRPH_RGB565: - pipes[pipe_cnt].pipe.src.source_format = dm_444_16; - break; - case SURFACE_PIXEL_FORMAT_GRPH_PALETA_256_COLORS: - pipes[pipe_cnt].pipe.src.source_format = dm_444_8; - break; - case SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA: - pipes[pipe_cnt].pipe.src.source_format = dm_rgbe_alpha; - break; - default: - pipes[pipe_cnt].pipe.src.source_format = dm_444_32; - break; - } - } - - pipe_cnt++; - } - - /* populate writeback information */ - DC_FP_START(); - dc->res_pool->funcs->populate_dml_writeback_from_context(dc, res_ctx, pipes); - DC_FP_END(); - - return pipe_cnt; -} - unsigned int dcn20_calc_max_scaled_time( unsigned int time_per_pixel, enum mmhubbub_wbif_mode mode, @@ -2413,7 +1600,7 @@ void dcn20_set_mcif_arb_params( { enum mmhubbub_wbif_mode wbif_mode; struct mcif_arb_params *wb_arb_params; - int i, j, k, dwb_pipe; + int i, j, dwb_pipe; /* Writeback MCIF_WB arbitration parameters */ dwb_pipe = 0; @@ -2437,11 +1624,10 @@ void dcn20_set_mcif_arb_params( } else wbif_mode = PACKED_444; - for (k = 0; k < sizeof(wb_arb_params->cli_watermark)/sizeof(wb_arb_params->cli_watermark[0]); k++) { - wb_arb_params->cli_watermark[k] = get_wm_writeback_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - wb_arb_params->pstate_watermark[k] = get_wm_writeback_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - } - wb_arb_params->time_per_pixel = 16.0 * 1000 / (context->res_ctx.pipe_ctx[i].stream->phy_pix_clk / 1000); /* 4 bit fraction, ms */ + DC_FP_START(); + dcn20_fpu_set_wb_arb_params(wb_arb_params, context, pipes, pipe_cnt, i); + DC_FP_END(); + wb_arb_params->slice_lines = 32; wb_arb_params->arbitration_slice = 2; wb_arb_params->max_scaled_time = dcn20_calc_max_scaled_time(wb_arb_params->time_per_pixel, @@ -2808,8 +1994,11 @@ int dcn20_validate_apply_pipe_split_flags( } /* Adjust dppclk when split is forced, do not bother with dispclk */ - if (split[i] != 0 && v->NoOfDPP[vlevel][max_mpc_comb][pipe_idx] == 1) - v->RequiredDPPCLK[vlevel][max_mpc_comb][pipe_idx] /= 2; + if (split[i] != 0 && v->NoOfDPP[vlevel][max_mpc_comb][pipe_idx] == 1) { + DC_FP_START(); + dcn20_fpu_adjust_dppclk(v, vlevel, max_mpc_comb, pipe_idx, false); + DC_FP_END(); + } pipe_idx++; } @@ -2835,7 +2024,9 @@ bool dcn20_fast_validate_bw( dcn20_merge_pipes_for_validate(dc, context); + DC_FP_START(); pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); + DC_FP_END(); *pipe_cnt_out = pipe_cnt; @@ -2892,7 +2083,9 @@ bool dcn20_fast_validate_bw( hsplit_pipe = dcn20_find_secondary_pipe(dc, &context->res_ctx, dc->res_pool, pipe); ASSERT(hsplit_pipe); if (!hsplit_pipe) { - context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] *= 2; + DC_FP_START(); + dcn20_fpu_adjust_dppclk(&context->bw_ctx.dml.vba, vlevel, context->bw_ctx.dml.vba.maxMpcComb, pipe_idx, true); + DC_FP_END(); continue; } if (context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx]) { @@ -2934,369 +2127,6 @@ validate_out: return out; } -static void dcn20_calculate_wm( - struct dc *dc, struct dc_state *context, - display_e2e_pipe_params_st *pipes, - int *out_pipe_cnt, - int *pipe_split_from, - int vlevel, - bool fast_validate) -{ - int pipe_cnt, i, pipe_idx; - - for (i = 0, pipe_idx = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { - if (!context->res_ctx.pipe_ctx[i].stream) - continue; - - pipes[pipe_cnt].clks_cfg.refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000.0; - pipes[pipe_cnt].clks_cfg.dispclk_mhz = context->bw_ctx.dml.vba.RequiredDISPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb]; - - if (pipe_split_from[i] < 0) { - pipes[pipe_cnt].clks_cfg.dppclk_mhz = - context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx]; - if (context->bw_ctx.dml.vba.BlendingAndTiming[pipe_idx] == pipe_idx) - pipes[pipe_cnt].pipe.dest.odm_combine = - context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx]; - else - pipes[pipe_cnt].pipe.dest.odm_combine = 0; - pipe_idx++; - } else { - pipes[pipe_cnt].clks_cfg.dppclk_mhz = - context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_split_from[i]]; - if (context->bw_ctx.dml.vba.BlendingAndTiming[pipe_split_from[i]] == pipe_split_from[i]) - pipes[pipe_cnt].pipe.dest.odm_combine = - context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_split_from[i]]; - else - pipes[pipe_cnt].pipe.dest.odm_combine = 0; - } - - if (dc->config.forced_clocks) { - pipes[pipe_cnt].clks_cfg.dispclk_mhz = context->bw_ctx.dml.soc.clock_limits[0].dispclk_mhz; - pipes[pipe_cnt].clks_cfg.dppclk_mhz = context->bw_ctx.dml.soc.clock_limits[0].dppclk_mhz; - } - if (dc->debug.min_disp_clk_khz > pipes[pipe_cnt].clks_cfg.dispclk_mhz * 1000) - pipes[pipe_cnt].clks_cfg.dispclk_mhz = dc->debug.min_disp_clk_khz / 1000.0; - if (dc->debug.min_dpp_clk_khz > pipes[pipe_cnt].clks_cfg.dppclk_mhz * 1000) - pipes[pipe_cnt].clks_cfg.dppclk_mhz = dc->debug.min_dpp_clk_khz / 1000.0; - - pipe_cnt++; - } - - if (pipe_cnt != pipe_idx) { - if (dc->res_pool->funcs->populate_dml_pipes) - pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, - context, pipes, fast_validate); - else - pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, - context, pipes, fast_validate); - } - - *out_pipe_cnt = pipe_cnt; - - pipes[0].clks_cfg.voltage = vlevel; - pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[vlevel].dcfclk_mhz; - pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[vlevel].socclk_mhz; - - /* only pipe 0 is read for voltage and dcf/soc clocks */ - if (vlevel < 1) { - pipes[0].clks_cfg.voltage = 1; - pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[1].dcfclk_mhz; - pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[1].socclk_mhz; - } - context->bw_ctx.bw.dcn.watermarks.b.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - - if (vlevel < 2) { - pipes[0].clks_cfg.voltage = 2; - pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[2].dcfclk_mhz; - pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[2].socclk_mhz; - } - context->bw_ctx.bw.dcn.watermarks.c.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - - if (vlevel < 3) { - pipes[0].clks_cfg.voltage = 3; - pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[2].dcfclk_mhz; - pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[2].socclk_mhz; - } - context->bw_ctx.bw.dcn.watermarks.d.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - - pipes[0].clks_cfg.voltage = vlevel; - pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[vlevel].dcfclk_mhz; - pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[vlevel].socclk_mhz; - context->bw_ctx.bw.dcn.watermarks.a.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.a.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; -} - -static bool is_dtbclk_required(struct dc *dc, struct dc_state *context) -{ - int i; - for (i = 0; i < dc->res_pool->pipe_count; i++) { - if (!context->res_ctx.pipe_ctx[i].stream) - continue; - if (is_dp_128b_132b_signal(&context->res_ctx.pipe_ctx[i])) - return true; - } - return false; -} - -static enum dcn_zstate_support_state decide_zstate_support(struct dc *dc, struct dc_state *context) -{ - int plane_count; - int i; - - plane_count = 0; - for (i = 0; i < dc->res_pool->pipe_count; i++) { - if (context->res_ctx.pipe_ctx[i].plane_state) - plane_count++; - } - - /* - * Zstate is allowed in following scenarios: - * 1. Single eDP with PSR enabled - * 2. 0 planes (No memory requests) - * 3. Single eDP without PSR but > 5ms stutter period - */ - if (plane_count == 0) - return DCN_ZSTATE_SUPPORT_ALLOW; - else if (context->stream_count == 1 && context->streams[0]->signal == SIGNAL_TYPE_EDP) { - struct dc_link *link = context->streams[0]->sink->link; - - /* zstate only supported on PWRSEQ0 */ - if (link->link_index != 0) - return DCN_ZSTATE_SUPPORT_DISALLOW; - - if (context->bw_ctx.dml.vba.StutterPeriod > 5000.0) - return DCN_ZSTATE_SUPPORT_ALLOW; - else if (link->psr_settings.psr_version == DC_PSR_VERSION_1 && !dc->debug.disable_psr) - return DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY; - else - return DCN_ZSTATE_SUPPORT_DISALLOW; - } else - return DCN_ZSTATE_SUPPORT_DISALLOW; -} - -void dcn20_calculate_dlg_params( - struct dc *dc, struct dc_state *context, - display_e2e_pipe_params_st *pipes, - int pipe_cnt, - int vlevel) -{ - int i, pipe_idx; - - /* Writeback MCIF_WB arbitration parameters */ - dc->res_pool->funcs->set_mcif_arb_params(dc, context, pipes, pipe_cnt); - - context->bw_ctx.bw.dcn.clk.dispclk_khz = context->bw_ctx.dml.vba.DISPCLK * 1000; - context->bw_ctx.bw.dcn.clk.dcfclk_khz = context->bw_ctx.dml.vba.DCFCLK * 1000; - context->bw_ctx.bw.dcn.clk.socclk_khz = context->bw_ctx.dml.vba.SOCCLK * 1000; - context->bw_ctx.bw.dcn.clk.dramclk_khz = context->bw_ctx.dml.vba.DRAMSpeed * 1000 / 16; - - if (dc->debug.min_dram_clk_khz > context->bw_ctx.bw.dcn.clk.dramclk_khz) - context->bw_ctx.bw.dcn.clk.dramclk_khz = dc->debug.min_dram_clk_khz; - - context->bw_ctx.bw.dcn.clk.dcfclk_deep_sleep_khz = context->bw_ctx.dml.vba.DCFCLKDeepSleep * 1000; - context->bw_ctx.bw.dcn.clk.fclk_khz = context->bw_ctx.dml.vba.FabricClock * 1000; - context->bw_ctx.bw.dcn.clk.p_state_change_support = - context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] - != dm_dram_clock_change_unsupported; - context->bw_ctx.bw.dcn.clk.dppclk_khz = 0; - - context->bw_ctx.bw.dcn.clk.zstate_support = decide_zstate_support(dc, context); - - context->bw_ctx.bw.dcn.clk.dtbclk_en = is_dtbclk_required(dc, context); - - if (context->bw_ctx.bw.dcn.clk.dispclk_khz < dc->debug.min_disp_clk_khz) - context->bw_ctx.bw.dcn.clk.dispclk_khz = dc->debug.min_disp_clk_khz; - - for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { - if (!context->res_ctx.pipe_ctx[i].stream) - continue; - pipes[pipe_idx].pipe.dest.vstartup_start = get_vstartup(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); - pipes[pipe_idx].pipe.dest.vupdate_offset = get_vupdate_offset(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); - pipes[pipe_idx].pipe.dest.vupdate_width = get_vupdate_width(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); - pipes[pipe_idx].pipe.dest.vready_offset = get_vready_offset(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); - context->res_ctx.pipe_ctx[i].det_buffer_size_kb = context->bw_ctx.dml.ip.det_buffer_size_kbytes; - context->res_ctx.pipe_ctx[i].unbounded_req = pipes[pipe_idx].pipe.src.unbounded_req_mode; - - if (context->bw_ctx.bw.dcn.clk.dppclk_khz < pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000) - context->bw_ctx.bw.dcn.clk.dppclk_khz = pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000; - context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz = - pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000; - context->res_ctx.pipe_ctx[i].pipe_dlg_param = pipes[pipe_idx].pipe.dest; - pipe_idx++; - } - /*save a original dppclock copy*/ - context->bw_ctx.bw.dcn.clk.bw_dppclk_khz = context->bw_ctx.bw.dcn.clk.dppclk_khz; - context->bw_ctx.bw.dcn.clk.bw_dispclk_khz = context->bw_ctx.bw.dcn.clk.dispclk_khz; - context->bw_ctx.bw.dcn.clk.max_supported_dppclk_khz = context->bw_ctx.dml.soc.clock_limits[vlevel].dppclk_mhz * 1000; - context->bw_ctx.bw.dcn.clk.max_supported_dispclk_khz = context->bw_ctx.dml.soc.clock_limits[vlevel].dispclk_mhz * 1000; - - context->bw_ctx.bw.dcn.compbuf_size_kb = context->bw_ctx.dml.ip.config_return_buffer_size_in_kbytes - - context->bw_ctx.dml.ip.det_buffer_size_kbytes * pipe_idx; - - for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { - bool cstate_en = context->bw_ctx.dml.vba.PrefetchMode[vlevel][context->bw_ctx.dml.vba.maxMpcComb] != 2; - - if (!context->res_ctx.pipe_ctx[i].stream) - continue; - - if (dc->ctx->dce_version == DCN_VERSION_2_01) - cstate_en = false; - - context->bw_ctx.dml.funcs.rq_dlg_get_dlg_reg(&context->bw_ctx.dml, - &context->res_ctx.pipe_ctx[i].dlg_regs, - &context->res_ctx.pipe_ctx[i].ttu_regs, - pipes, - pipe_cnt, - pipe_idx, - cstate_en, - context->bw_ctx.bw.dcn.clk.p_state_change_support, - false, false, true); - - context->bw_ctx.dml.funcs.rq_dlg_get_rq_reg(&context->bw_ctx.dml, - &context->res_ctx.pipe_ctx[i].rq_regs, - &pipes[pipe_idx].pipe); - pipe_idx++; - } -} - -static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *context, - bool fast_validate) -{ - bool out = false; - - BW_VAL_TRACE_SETUP(); - - int vlevel = 0; - int pipe_split_from[MAX_PIPES]; - int pipe_cnt = 0; - display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_ATOMIC); - DC_LOGGER_INIT(dc->ctx->logger); - - BW_VAL_TRACE_COUNT(); - - out = dcn20_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel, fast_validate); - - if (pipe_cnt == 0) - goto validate_out; - - if (!out) - goto validate_fail; - - BW_VAL_TRACE_END_VOLTAGE_LEVEL(); - - if (fast_validate) { - BW_VAL_TRACE_SKIP(fast); - goto validate_out; - } - - dcn20_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, fast_validate); - dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); - - BW_VAL_TRACE_END_WATERMARKS(); - - goto validate_out; - -validate_fail: - DC_LOG_WARNING("Mode Validation Warning: %s failed validation.\n", - dml_get_status_message(context->bw_ctx.dml.vba.ValidationStatus[context->bw_ctx.dml.vba.soc.num_states])); - - BW_VAL_TRACE_SKIP(fail); - out = false; - -validate_out: - kfree(pipes); - - BW_VAL_TRACE_FINISH(); - - return out; -} - -/* - * This must be noinline to ensure anything that deals with FP registers - * is contained within this call; previously our compiling with hard-float - * would result in fp instructions being emitted outside of the boundaries - * of the DC_FP_START/END macros, which makes sense as the compiler has no - * idea about what is wrapped and what is not - * - * This is largely just a workaround to avoid breakage introduced with 5.6, - * ideally all fp-using code should be moved into its own file, only that - * should be compiled with hard-float, and all code exported from there - * should be strictly wrapped with DC_FP_START/END - */ -static noinline bool dcn20_validate_bandwidth_fp(struct dc *dc, - struct dc_state *context, bool fast_validate) -{ - bool voltage_supported = false; - bool full_pstate_supported = false; - bool dummy_pstate_supported = false; - double p_state_latency_us; - - p_state_latency_us = context->bw_ctx.dml.soc.dram_clock_change_latency_us; - context->bw_ctx.dml.soc.disable_dram_clock_change_vactive_support = - dc->debug.disable_dram_clock_change_vactive_support; - context->bw_ctx.dml.soc.allow_dram_clock_one_display_vactive = - dc->debug.enable_dram_clock_change_one_display_vactive; - - /*Unsafe due to current pipe merge and split logic*/ - ASSERT(context != dc->current_state); - - if (fast_validate) { - return dcn20_validate_bandwidth_internal(dc, context, true); - } - - // Best case, we support full UCLK switch latency - voltage_supported = dcn20_validate_bandwidth_internal(dc, context, false); - full_pstate_supported = context->bw_ctx.bw.dcn.clk.p_state_change_support; - - if (context->bw_ctx.dml.soc.dummy_pstate_latency_us == 0 || - (voltage_supported && full_pstate_supported)) { - context->bw_ctx.bw.dcn.clk.p_state_change_support = full_pstate_supported; - goto restore_dml_state; - } - - // Fallback: Try to only support G6 temperature read latency - context->bw_ctx.dml.soc.dram_clock_change_latency_us = context->bw_ctx.dml.soc.dummy_pstate_latency_us; - - voltage_supported = dcn20_validate_bandwidth_internal(dc, context, false); - dummy_pstate_supported = context->bw_ctx.bw.dcn.clk.p_state_change_support; - - if (voltage_supported && (dummy_pstate_supported || !(context->stream_count))) { - context->bw_ctx.bw.dcn.clk.p_state_change_support = false; - goto restore_dml_state; - } - - // ERROR: fallback is supposed to always work. - ASSERT(false); - -restore_dml_state: - context->bw_ctx.dml.soc.dram_clock_change_latency_us = p_state_latency_us; - return voltage_supported; -} - bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, bool fast_validate) { @@ -3464,170 +2294,6 @@ static void dcn20_pp_smu_destroy(struct pp_smu_funcs **pp_smu) } } -void dcn20_cap_soc_clocks( - struct _vcs_dpi_soc_bounding_box_st *bb, - struct pp_smu_nv_clock_table max_clocks) -{ - int i; - - // First pass - cap all clocks higher than the reported max - for (i = 0; i < bb->num_states; i++) { - if ((bb->clock_limits[i].dcfclk_mhz > (max_clocks.dcfClockInKhz / 1000)) - && max_clocks.dcfClockInKhz != 0) - bb->clock_limits[i].dcfclk_mhz = (max_clocks.dcfClockInKhz / 1000); - - if ((bb->clock_limits[i].dram_speed_mts > (max_clocks.uClockInKhz / 1000) * 16) - && max_clocks.uClockInKhz != 0) - bb->clock_limits[i].dram_speed_mts = (max_clocks.uClockInKhz / 1000) * 16; - - if ((bb->clock_limits[i].fabricclk_mhz > (max_clocks.fabricClockInKhz / 1000)) - && max_clocks.fabricClockInKhz != 0) - bb->clock_limits[i].fabricclk_mhz = (max_clocks.fabricClockInKhz / 1000); - - if ((bb->clock_limits[i].dispclk_mhz > (max_clocks.displayClockInKhz / 1000)) - && max_clocks.displayClockInKhz != 0) - bb->clock_limits[i].dispclk_mhz = (max_clocks.displayClockInKhz / 1000); - - if ((bb->clock_limits[i].dppclk_mhz > (max_clocks.dppClockInKhz / 1000)) - && max_clocks.dppClockInKhz != 0) - bb->clock_limits[i].dppclk_mhz = (max_clocks.dppClockInKhz / 1000); - - if ((bb->clock_limits[i].phyclk_mhz > (max_clocks.phyClockInKhz / 1000)) - && max_clocks.phyClockInKhz != 0) - bb->clock_limits[i].phyclk_mhz = (max_clocks.phyClockInKhz / 1000); - - if ((bb->clock_limits[i].socclk_mhz > (max_clocks.socClockInKhz / 1000)) - && max_clocks.socClockInKhz != 0) - bb->clock_limits[i].socclk_mhz = (max_clocks.socClockInKhz / 1000); - - if ((bb->clock_limits[i].dscclk_mhz > (max_clocks.dscClockInKhz / 1000)) - && max_clocks.dscClockInKhz != 0) - bb->clock_limits[i].dscclk_mhz = (max_clocks.dscClockInKhz / 1000); - } - - // Second pass - remove all duplicate clock states - for (i = bb->num_states - 1; i > 1; i--) { - bool duplicate = true; - - if (bb->clock_limits[i-1].dcfclk_mhz != bb->clock_limits[i].dcfclk_mhz) - duplicate = false; - if (bb->clock_limits[i-1].dispclk_mhz != bb->clock_limits[i].dispclk_mhz) - duplicate = false; - if (bb->clock_limits[i-1].dppclk_mhz != bb->clock_limits[i].dppclk_mhz) - duplicate = false; - if (bb->clock_limits[i-1].dram_speed_mts != bb->clock_limits[i].dram_speed_mts) - duplicate = false; - if (bb->clock_limits[i-1].dscclk_mhz != bb->clock_limits[i].dscclk_mhz) - duplicate = false; - if (bb->clock_limits[i-1].fabricclk_mhz != bb->clock_limits[i].fabricclk_mhz) - duplicate = false; - if (bb->clock_limits[i-1].phyclk_mhz != bb->clock_limits[i].phyclk_mhz) - duplicate = false; - if (bb->clock_limits[i-1].socclk_mhz != bb->clock_limits[i].socclk_mhz) - duplicate = false; - - if (duplicate) - bb->num_states--; - } -} - -void dcn20_update_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb, - struct pp_smu_nv_clock_table *max_clocks, unsigned int *uclk_states, unsigned int num_states) -{ - struct _vcs_dpi_voltage_scaling_st calculated_states[DC__VOLTAGE_STATES]; - int i; - int num_calculated_states = 0; - int min_dcfclk = 0; - - if (num_states == 0) - return; - - memset(calculated_states, 0, sizeof(calculated_states)); - - if (dc->bb_overrides.min_dcfclk_mhz > 0) - min_dcfclk = dc->bb_overrides.min_dcfclk_mhz; - else { - if (ASICREV_IS_NAVI12_P(dc->ctx->asic_id.hw_internal_rev)) - min_dcfclk = 310; - else - // Accounting for SOC/DCF relationship, we can go as high as - // 506Mhz in Vmin. - min_dcfclk = 506; - } - - for (i = 0; i < num_states; i++) { - int min_fclk_required_by_uclk; - calculated_states[i].state = i; - calculated_states[i].dram_speed_mts = uclk_states[i] * 16 / 1000; - - // FCLK:UCLK ratio is 1.08 - min_fclk_required_by_uclk = div_u64(((unsigned long long)uclk_states[i]) * 1080, - 1000000); - - calculated_states[i].fabricclk_mhz = (min_fclk_required_by_uclk < min_dcfclk) ? - min_dcfclk : min_fclk_required_by_uclk; - - calculated_states[i].socclk_mhz = (calculated_states[i].fabricclk_mhz > max_clocks->socClockInKhz / 1000) ? - max_clocks->socClockInKhz / 1000 : calculated_states[i].fabricclk_mhz; - - calculated_states[i].dcfclk_mhz = (calculated_states[i].fabricclk_mhz > max_clocks->dcfClockInKhz / 1000) ? - max_clocks->dcfClockInKhz / 1000 : calculated_states[i].fabricclk_mhz; - - calculated_states[i].dispclk_mhz = max_clocks->displayClockInKhz / 1000; - calculated_states[i].dppclk_mhz = max_clocks->displayClockInKhz / 1000; - calculated_states[i].dscclk_mhz = max_clocks->displayClockInKhz / (1000 * 3); - - calculated_states[i].phyclk_mhz = max_clocks->phyClockInKhz / 1000; - - num_calculated_states++; - } - - calculated_states[num_calculated_states - 1].socclk_mhz = max_clocks->socClockInKhz / 1000; - calculated_states[num_calculated_states - 1].fabricclk_mhz = max_clocks->socClockInKhz / 1000; - calculated_states[num_calculated_states - 1].dcfclk_mhz = max_clocks->dcfClockInKhz / 1000; - - memcpy(bb->clock_limits, calculated_states, sizeof(bb->clock_limits)); - bb->num_states = num_calculated_states; - - // Duplicate the last state, DML always an extra state identical to max state to work - memcpy(&bb->clock_limits[num_calculated_states], &bb->clock_limits[num_calculated_states - 1], sizeof(struct _vcs_dpi_voltage_scaling_st)); - bb->clock_limits[num_calculated_states].state = bb->num_states; -} - -void dcn20_patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb) -{ - if ((int)(bb->sr_exit_time_us * 1000) != dc->bb_overrides.sr_exit_time_ns - && dc->bb_overrides.sr_exit_time_ns) { - bb->sr_exit_time_us = dc->bb_overrides.sr_exit_time_ns / 1000.0; - } - - if ((int)(bb->sr_enter_plus_exit_time_us * 1000) - != dc->bb_overrides.sr_enter_plus_exit_time_ns - && dc->bb_overrides.sr_enter_plus_exit_time_ns) { - bb->sr_enter_plus_exit_time_us = - dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; - } - - if ((int)(bb->urgent_latency_us * 1000) != dc->bb_overrides.urgent_latency_ns - && dc->bb_overrides.urgent_latency_ns) { - bb->urgent_latency_us = dc->bb_overrides.urgent_latency_ns / 1000.0; - } - - if ((int)(bb->dram_clock_change_latency_us * 1000) - != dc->bb_overrides.dram_clock_change_latency_ns - && dc->bb_overrides.dram_clock_change_latency_ns) { - bb->dram_clock_change_latency_us = - dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; - } - - if ((int)(bb->dummy_pstate_latency_us * 1000) - != dc->bb_overrides.dummy_clock_change_latency_ns - && dc->bb_overrides.dummy_clock_change_latency_ns) { - bb->dummy_pstate_latency_us = - dc->bb_overrides.dummy_clock_change_latency_ns / 1000.0; - } -} - static struct _vcs_dpi_soc_bounding_box_st *get_asic_rev_soc_bb( uint32_t hw_internal_rev) { @@ -3910,8 +2576,9 @@ static bool dcn20_resource_construct( ranges.reader_wm_sets[i].wm_inst = i; ranges.reader_wm_sets[i].min_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; ranges.reader_wm_sets[i].max_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; - ranges.reader_wm_sets[i].min_fill_clk_mhz = (i > 0) ? (loaded_bb->clock_limits[i - 1].dram_speed_mts / 16) + 1 : 0; - ranges.reader_wm_sets[i].max_fill_clk_mhz = loaded_bb->clock_limits[i].dram_speed_mts / 16; + DC_FP_START(); + dcn20_fpu_set_wm_ranges(i, &ranges, loaded_bb); + DC_FP_END(); ranges.num_reader_wm_sets = i + 1; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h index 6ec8ff45f0f7..7cbe1e9daa36 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h @@ -27,6 +27,7 @@ #define __DC_RESOURCE_DCN20_H__ #include "core_types.h" +#include "dml/dcn20/dcn20_fpu.h" #define TO_DCN20_RES_POOL(pool)\ container_of(pool, struct dcn20_resource_pool, base) @@ -35,6 +36,12 @@ struct dc; struct resource_pool; struct _vcs_dpi_display_pipe_params_st; +extern struct _vcs_dpi_ip_params_st dcn2_0_ip; +extern struct _vcs_dpi_ip_params_st dcn2_0_nv14_ip; +extern struct _vcs_dpi_soc_bounding_box_st dcn2_0_soc; +extern struct _vcs_dpi_soc_bounding_box_st dcn2_0_nv14_soc; +extern struct _vcs_dpi_soc_bounding_box_st dcn2_0_nv12_soc; + struct dcn20_resource_pool { struct resource_pool base; }; @@ -49,11 +56,7 @@ unsigned int dcn20_calc_max_scaled_time( unsigned int time_per_pixel, enum mmhubbub_wbif_mode mode, unsigned int urgent_watermark); -int dcn20_populate_dml_pipes_from_context( - struct dc *dc, - struct dc_state *context, - display_e2e_pipe_params_st *pipes, - bool fast_validate); + struct pipe_ctx *dcn20_acquire_idle_pipe_for_layer( struct dc_state *state, const struct resource_pool *pool, @@ -79,7 +82,6 @@ struct dpp *dcn20_dpp_create( struct input_pixel_processor *dcn20_ipp_create( struct dc_context *ctx, uint32_t inst); - struct output_pixel_processor *dcn20_opp_create( struct dc_context *ctx, uint32_t inst); @@ -96,11 +98,6 @@ struct display_stream_compressor *dcn20_dsc_create( struct dc_context *ctx, uint32_t inst); void dcn20_dsc_destroy(struct display_stream_compressor **dsc); -void dcn20_cap_soc_clocks( - struct _vcs_dpi_soc_bounding_box_st *bb, - struct pp_smu_nv_clock_table max_clocks); -void dcn20_update_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb, - struct pp_smu_nv_clock_table *max_clocks, unsigned int *uclk_states, unsigned int num_states); struct hubp *dcn20_hubp_create( struct dc_context *ctx, uint32_t inst); @@ -158,11 +155,6 @@ bool dcn20_fast_validate_bw( int *pipe_split_from, int *vlevel_out, bool fast_validate); -void dcn20_calculate_dlg_params( - struct dc *dc, struct dc_state *context, - display_e2e_pipe_params_st *pipes, - int pipe_cnt, - int vlevel); enum dc_status dcn20_build_mapped_resource(const struct dc *dc, struct dc_state *context, struct dc_stream_state *stream); enum dc_status dcn20_add_stream_to_ctx(struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *dc_stream); @@ -170,12 +162,5 @@ enum dc_status dcn20_add_dsc_to_stream_resource(struct dc *dc, struct dc_state * enum dc_status dcn20_remove_stream_from_ctx(struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *dc_stream); enum dc_status dcn20_patch_unknown_plane_state(struct dc_plane_state *plane_state); -void dcn20_patch_bounding_box( - struct dc *dc, - struct _vcs_dpi_soc_bounding_box_st *bb); -void dcn20_cap_soc_clocks( - struct _vcs_dpi_soc_bounding_box_st *bb, - struct pp_smu_nv_clock_table max_clocks); - #endif /* __DC_RESOURCE_DCN20_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/Makefile b/drivers/gpu/drm/amd/display/dc/dcn21/Makefile index bb8c95141082..0dc06e428999 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn21/Makefile @@ -5,31 +5,6 @@ DCN21 = dcn21_init.o dcn21_hubp.o dcn21_hubbub.o dcn21_resource.o \ dcn21_hwseq.o dcn21_link_encoder.o dcn21_dccg.o -ifdef CONFIG_X86 -CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -msse -endif - -ifdef CONFIG_PPC64 -CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -maltivec -endif - -ifdef CONFIG_CC_IS_GCC -ifeq ($(call cc-ifversion, -lt, 0701, y), y) -IS_OLD_GCC = 1 -endif -endif - -ifdef CONFIG_X86 -ifdef IS_OLD_GCC -# Stack alignment mismatch, proceed with caution. -# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3 -# (8B stack alignment). -CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o += -mpreferred-stack-boundary=4 -else -CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o += -msse2 -endif -endif - AMD_DAL_DCN21 = $(addprefix $(AMDDALPATH)/dc/dcn21/,$(DCN21)) AMD_DISPLAY_FILES += $(AMD_DAL_DCN21) diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c index e5cc6bf45743..612732656772 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c @@ -34,6 +34,7 @@ #include "resource.h" #include "include/irq_service_interface.h" #include "dcn20/dcn20_resource.h" +#include "dcn21/dcn21_resource.h" #include "dml/dcn20/dcn20_fpu.h" @@ -89,230 +90,6 @@ #include "dce/dmub_psr.h" #include "dce/dmub_abm.h" -#define DC_LOGGER_INIT(logger) - - -struct _vcs_dpi_ip_params_st dcn2_1_ip = { - .odm_capable = 1, - .gpuvm_enable = 1, - .hostvm_enable = 1, - .gpuvm_max_page_table_levels = 1, - .hostvm_max_page_table_levels = 4, - .hostvm_cached_page_table_levels = 2, - .num_dsc = 3, - .rob_buffer_size_kbytes = 168, - .det_buffer_size_kbytes = 164, - .dpte_buffer_size_in_pte_reqs_luma = 44, - .dpte_buffer_size_in_pte_reqs_chroma = 42,//todo - .dpp_output_buffer_pixels = 2560, - .opp_output_buffer_lines = 1, - .pixel_chunk_size_kbytes = 8, - .pte_enable = 1, - .max_page_table_levels = 4, - .pte_chunk_size_kbytes = 2, - .meta_chunk_size_kbytes = 2, - .min_meta_chunk_size_bytes = 256, - .writeback_chunk_size_kbytes = 2, - .line_buffer_size_bits = 789504, - .is_line_buffer_bpp_fixed = 0, - .line_buffer_fixed_bpp = 0, - .dcc_supported = true, - .max_line_buffer_lines = 12, - .writeback_luma_buffer_size_kbytes = 12, - .writeback_chroma_buffer_size_kbytes = 8, - .writeback_chroma_line_buffer_width_pixels = 4, - .writeback_max_hscl_ratio = 1, - .writeback_max_vscl_ratio = 1, - .writeback_min_hscl_ratio = 1, - .writeback_min_vscl_ratio = 1, - .writeback_max_hscl_taps = 12, - .writeback_max_vscl_taps = 12, - .writeback_line_buffer_luma_buffer_size = 0, - .writeback_line_buffer_chroma_buffer_size = 14643, - .cursor_buffer_size = 8, - .cursor_chunk_size = 2, - .max_num_otg = 4, - .max_num_dpp = 4, - .max_num_wb = 1, - .max_dchub_pscl_bw_pix_per_clk = 4, - .max_pscl_lb_bw_pix_per_clk = 2, - .max_lb_vscl_bw_pix_per_clk = 4, - .max_vscl_hscl_bw_pix_per_clk = 4, - .max_hscl_ratio = 4, - .max_vscl_ratio = 4, - .hscl_mults = 4, - .vscl_mults = 4, - .max_hscl_taps = 8, - .max_vscl_taps = 8, - .dispclk_ramp_margin_percent = 1, - .underscan_factor = 1.10, - .min_vblank_lines = 32, // - .dppclk_delay_subtotal = 77, // - .dppclk_delay_scl_lb_only = 16, - .dppclk_delay_scl = 50, - .dppclk_delay_cnvc_formatter = 8, - .dppclk_delay_cnvc_cursor = 6, - .dispclk_delay_subtotal = 87, // - .dcfclk_cstate_latency = 10, // SRExitTime - .max_inter_dcn_tile_repeaters = 8, - - .xfc_supported = false, - .xfc_fill_bw_overhead_percent = 10.0, - .xfc_fill_constant_bytes = 0, - .ptoi_supported = 0, - .number_of_cursors = 1, -}; - -struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { - .clock_limits = { - { - .state = 0, - .dcfclk_mhz = 400.0, - .fabricclk_mhz = 400.0, - .dispclk_mhz = 600.0, - .dppclk_mhz = 400.00, - .phyclk_mhz = 600.0, - .socclk_mhz = 278.0, - .dscclk_mhz = 205.67, - .dram_speed_mts = 1600.0, - }, - { - .state = 1, - .dcfclk_mhz = 464.52, - .fabricclk_mhz = 800.0, - .dispclk_mhz = 654.55, - .dppclk_mhz = 626.09, - .phyclk_mhz = 600.0, - .socclk_mhz = 278.0, - .dscclk_mhz = 205.67, - .dram_speed_mts = 1600.0, - }, - { - .state = 2, - .dcfclk_mhz = 514.29, - .fabricclk_mhz = 933.0, - .dispclk_mhz = 757.89, - .dppclk_mhz = 685.71, - .phyclk_mhz = 600.0, - .socclk_mhz = 278.0, - .dscclk_mhz = 287.67, - .dram_speed_mts = 1866.0, - }, - { - .state = 3, - .dcfclk_mhz = 576.00, - .fabricclk_mhz = 1067.0, - .dispclk_mhz = 847.06, - .dppclk_mhz = 757.89, - .phyclk_mhz = 600.0, - .socclk_mhz = 715.0, - .dscclk_mhz = 318.334, - .dram_speed_mts = 2134.0, - }, - { - .state = 4, - .dcfclk_mhz = 626.09, - .fabricclk_mhz = 1200.0, - .dispclk_mhz = 900.00, - .dppclk_mhz = 847.06, - .phyclk_mhz = 810.0, - .socclk_mhz = 953.0, - .dscclk_mhz = 489.0, - .dram_speed_mts = 2400.0, - }, - { - .state = 5, - .dcfclk_mhz = 685.71, - .fabricclk_mhz = 1333.0, - .dispclk_mhz = 1028.57, - .dppclk_mhz = 960.00, - .phyclk_mhz = 810.0, - .socclk_mhz = 278.0, - .dscclk_mhz = 287.67, - .dram_speed_mts = 2666.0, - }, - { - .state = 6, - .dcfclk_mhz = 757.89, - .fabricclk_mhz = 1467.0, - .dispclk_mhz = 1107.69, - .dppclk_mhz = 1028.57, - .phyclk_mhz = 810.0, - .socclk_mhz = 715.0, - .dscclk_mhz = 318.334, - .dram_speed_mts = 3200.0, - }, - { - .state = 7, - .dcfclk_mhz = 847.06, - .fabricclk_mhz = 1600.0, - .dispclk_mhz = 1395.0, - .dppclk_mhz = 1285.00, - .phyclk_mhz = 1325.0, - .socclk_mhz = 953.0, - .dscclk_mhz = 489.0, - .dram_speed_mts = 4266.0, - }, - /*Extra state, no dispclk ramping*/ - { - .state = 8, - .dcfclk_mhz = 847.06, - .fabricclk_mhz = 1600.0, - .dispclk_mhz = 1395.0, - .dppclk_mhz = 1285.0, - .phyclk_mhz = 1325.0, - .socclk_mhz = 953.0, - .dscclk_mhz = 489.0, - .dram_speed_mts = 4266.0, - }, - - }, - - .sr_exit_time_us = 12.5, - .sr_enter_plus_exit_time_us = 17.0, - .urgent_latency_us = 4.0, - .urgent_latency_pixel_data_only_us = 4.0, - .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, - .urgent_latency_vm_data_only_us = 4.0, - .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, - .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, - .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, - .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 80.0, - .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 75.0, - .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, - .max_avg_sdp_bw_use_normal_percent = 60.0, - .max_avg_dram_bw_use_normal_percent = 100.0, - .writeback_latency_us = 12.0, - .max_request_size_bytes = 256, - .dram_channel_width_bytes = 4, - .fabric_datapath_to_dcn_data_return_bytes = 32, - .dcn_downspread_percent = 0.5, - .downspread_percent = 0.38, - .dram_page_open_time_ns = 50.0, - .dram_rw_turnaround_time_ns = 17.5, - .dram_return_buffer_per_channel_bytes = 8192, - .round_trip_ping_latency_dcfclk_cycles = 128, - .urgent_out_of_order_return_per_channel_bytes = 4096, - .channel_interleave_bytes = 256, - .num_banks = 8, - .num_chans = 4, - .vmm_page_size_bytes = 4096, - .dram_clock_change_latency_us = 23.84, - .return_bus_width_bytes = 64, - .dispclk_dppclk_vco_speed_mhz = 3600, - .xfc_bus_transport_time_us = 4, - .xfc_xbuf_latency_tolerance_us = 4, - .use_urgent_burst_bw = 1, - .num_states = 8 -}; - -#ifndef MAX -#define MAX(X, Y) ((X) > (Y) ? (X) : (Y)) -#endif -#ifndef MIN -#define MIN(X, Y) ((X) < (Y) ? (X) : (Y)) -#endif - /* begin ********************* * macros to expend register list macro defined in HW object header file */ @@ -705,12 +482,6 @@ static const struct dcn10_stream_encoder_mask se_mask = { static void dcn21_pp_smu_destroy(struct pp_smu_funcs **pp_smu); -static int dcn21_populate_dml_pipes_from_context( - struct dc *dc, - struct dc_state *context, - display_e2e_pipe_params_st *pipes, - bool fast_validate); - static struct input_pixel_processor *dcn21_ipp_create( struct dc_context *ctx, uint32_t inst) { @@ -1029,163 +800,13 @@ static void dcn21_resource_destruct(struct dcn21_resource_pool *pool) dcn21_pp_smu_destroy(&pool->base.pp_smu); } - -static void calculate_wm_set_for_vlevel( - int vlevel, - struct wm_range_table_entry *table_entry, - struct dcn_watermarks *wm_set, - struct display_mode_lib *dml, - display_e2e_pipe_params_st *pipes, - int pipe_cnt) -{ - double dram_clock_change_latency_cached = dml->soc.dram_clock_change_latency_us; - - ASSERT(vlevel < dml->soc.num_states); - /* only pipe 0 is read for voltage and dcf/soc clocks */ - pipes[0].clks_cfg.voltage = vlevel; - pipes[0].clks_cfg.dcfclk_mhz = dml->soc.clock_limits[vlevel].dcfclk_mhz; - pipes[0].clks_cfg.socclk_mhz = dml->soc.clock_limits[vlevel].socclk_mhz; - - dml->soc.dram_clock_change_latency_us = table_entry->pstate_latency_us; - dml->soc.sr_exit_time_us = table_entry->sr_exit_time_us; - dml->soc.sr_enter_plus_exit_time_us = table_entry->sr_enter_plus_exit_time_us; - - wm_set->urgent_ns = get_wm_urgent(dml, pipes, pipe_cnt) * 1000; - wm_set->cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(dml, pipes, pipe_cnt) * 1000; - wm_set->cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(dml, pipes, pipe_cnt) * 1000; - wm_set->cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(dml, pipes, pipe_cnt) * 1000; - wm_set->pte_meta_urgent_ns = get_wm_memory_trip(dml, pipes, pipe_cnt) * 1000; - wm_set->frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(dml, pipes, pipe_cnt) * 1000; - wm_set->frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(dml, pipes, pipe_cnt) * 1000; - wm_set->urgent_latency_ns = get_urgent_latency(dml, pipes, pipe_cnt) * 1000; - dml->soc.dram_clock_change_latency_us = dram_clock_change_latency_cached; - -} - -static void patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb) -{ - int i; - - if (dc->bb_overrides.sr_exit_time_ns) { - for (i = 0; i < WM_SET_COUNT; i++) { - dc->clk_mgr->bw_params->wm_table.entries[i].sr_exit_time_us = - dc->bb_overrides.sr_exit_time_ns / 1000.0; - } - } - - if (dc->bb_overrides.sr_enter_plus_exit_time_ns) { - for (i = 0; i < WM_SET_COUNT; i++) { - dc->clk_mgr->bw_params->wm_table.entries[i].sr_enter_plus_exit_time_us = - dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; - } - } - - if (dc->bb_overrides.urgent_latency_ns) { - bb->urgent_latency_us = dc->bb_overrides.urgent_latency_ns / 1000.0; - } - - if (dc->bb_overrides.dram_clock_change_latency_ns) { - for (i = 0; i < WM_SET_COUNT; i++) { - dc->clk_mgr->bw_params->wm_table.entries[i].pstate_latency_us = - dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; - } - } -} - -static void dcn21_calculate_wm( - struct dc *dc, struct dc_state *context, - display_e2e_pipe_params_st *pipes, - int *out_pipe_cnt, - int *pipe_split_from, - int vlevel_req, - bool fast_validate) -{ - int pipe_cnt, i, pipe_idx; - int vlevel, vlevel_max; - struct wm_range_table_entry *table_entry; - struct clk_bw_params *bw_params = dc->clk_mgr->bw_params; - - ASSERT(bw_params); - - patch_bounding_box(dc, &context->bw_ctx.dml.soc); - - for (i = 0, pipe_idx = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { - if (!context->res_ctx.pipe_ctx[i].stream) - continue; - - pipes[pipe_cnt].clks_cfg.refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000.0; - pipes[pipe_cnt].clks_cfg.dispclk_mhz = context->bw_ctx.dml.vba.RequiredDISPCLK[vlevel_req][context->bw_ctx.dml.vba.maxMpcComb]; - - if (pipe_split_from[i] < 0) { - pipes[pipe_cnt].clks_cfg.dppclk_mhz = - context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel_req][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx]; - if (context->bw_ctx.dml.vba.BlendingAndTiming[pipe_idx] == pipe_idx) - pipes[pipe_cnt].pipe.dest.odm_combine = - context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel_req][pipe_idx]; - else - pipes[pipe_cnt].pipe.dest.odm_combine = 0; - pipe_idx++; - } else { - pipes[pipe_cnt].clks_cfg.dppclk_mhz = - context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel_req][context->bw_ctx.dml.vba.maxMpcComb][pipe_split_from[i]]; - if (context->bw_ctx.dml.vba.BlendingAndTiming[pipe_split_from[i]] == pipe_split_from[i]) - pipes[pipe_cnt].pipe.dest.odm_combine = - context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel_req][pipe_split_from[i]]; - else - pipes[pipe_cnt].pipe.dest.odm_combine = 0; - } - pipe_cnt++; - } - - if (pipe_cnt != pipe_idx) { - if (dc->res_pool->funcs->populate_dml_pipes) - pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, - context, pipes, fast_validate); - else - pipe_cnt = dcn21_populate_dml_pipes_from_context(dc, - context, pipes, fast_validate); - } - - *out_pipe_cnt = pipe_cnt; - - vlevel_max = bw_params->clk_table.num_entries - 1; - - - /* WM Set D */ - table_entry = &bw_params->wm_table.entries[WM_D]; - if (table_entry->wm_type == WM_TYPE_RETRAINING) - vlevel = 0; - else - vlevel = vlevel_max; - calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.d, - &context->bw_ctx.dml, pipes, pipe_cnt); - /* WM Set C */ - table_entry = &bw_params->wm_table.entries[WM_C]; - vlevel = MIN(MAX(vlevel_req, 3), vlevel_max); - calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.c, - &context->bw_ctx.dml, pipes, pipe_cnt); - /* WM Set B */ - table_entry = &bw_params->wm_table.entries[WM_B]; - vlevel = MIN(MAX(vlevel_req, 2), vlevel_max); - calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.b, - &context->bw_ctx.dml, pipes, pipe_cnt); - - /* WM Set A */ - table_entry = &bw_params->wm_table.entries[WM_A]; - vlevel = MIN(vlevel_req, vlevel_max); - calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.a, - &context->bw_ctx.dml, pipes, pipe_cnt); -} - - -static bool dcn21_fast_validate_bw( - struct dc *dc, - struct dc_state *context, - display_e2e_pipe_params_st *pipes, - int *pipe_cnt_out, - int *pipe_split_from, - int *vlevel_out, - bool fast_validate) +bool dcn21_fast_validate_bw(struct dc *dc, + struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int *pipe_cnt_out, + int *pipe_split_from, + int *vlevel_out, + bool fast_validate) { bool out = false; int split[MAX_PIPES] = { 0 }; @@ -1197,7 +818,9 @@ static bool dcn21_fast_validate_bw( dcn20_merge_pipes_for_validate(dc, context); + DC_FP_START(); pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); + DC_FP_END(); *pipe_cnt_out = pipe_cnt; @@ -1287,7 +910,9 @@ static bool dcn21_fast_validate_bw( hsplit_pipe = dcn20_find_secondary_pipe(dc, &context->res_ctx, dc->res_pool, pipe); ASSERT(hsplit_pipe); if (!hsplit_pipe) { - context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] *= 2; + DC_FP_START(); + dcn20_fpu_adjust_dppclk(&context->bw_ctx.dml.vba, vlevel, context->bw_ctx.dml.vba.maxMpcComb, pipe_idx, true); + DC_FP_END(); continue; } if (context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx]) { @@ -1329,61 +954,6 @@ validate_out: return out; } -static noinline bool dcn21_validate_bandwidth_fp(struct dc *dc, - struct dc_state *context, bool fast_validate) -{ - bool out = false; - - BW_VAL_TRACE_SETUP(); - - int vlevel = 0; - int pipe_split_from[MAX_PIPES]; - int pipe_cnt = 0; - display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_ATOMIC); - DC_LOGGER_INIT(dc->ctx->logger); - - BW_VAL_TRACE_COUNT(); - - /*Unsafe due to current pipe merge and split logic*/ - ASSERT(context != dc->current_state); - - out = dcn21_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel, fast_validate); - - if (pipe_cnt == 0) - goto validate_out; - - if (!out) - goto validate_fail; - - BW_VAL_TRACE_END_VOLTAGE_LEVEL(); - - if (fast_validate) { - BW_VAL_TRACE_SKIP(fast); - goto validate_out; - } - - dcn21_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, fast_validate); - dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); - - BW_VAL_TRACE_END_WATERMARKS(); - - goto validate_out; - -validate_fail: - DC_LOG_WARNING("Mode Validation Warning: %s failed validation.\n", - dml_get_status_message(context->bw_ctx.dml.vba.ValidationStatus[context->bw_ctx.dml.vba.soc.num_states])); - - BW_VAL_TRACE_SKIP(fail); - out = false; - -validate_out: - kfree(pipes); - - BW_VAL_TRACE_FINISH(); - - return out; -} - /* * Some of the functions further below use the FPU, so we need to wrap this * with DC_FP_START()/DC_FP_END(). Use the same approach as for @@ -1558,94 +1128,6 @@ static struct display_stream_compressor *dcn21_dsc_create(struct dc_context *ctx return &dsc->base; } -static struct _vcs_dpi_voltage_scaling_st construct_low_pstate_lvl(struct clk_limit_table *clk_table, unsigned int high_voltage_lvl) -{ - struct _vcs_dpi_voltage_scaling_st low_pstate_lvl; - int i; - - low_pstate_lvl.state = 1; - low_pstate_lvl.dcfclk_mhz = clk_table->entries[0].dcfclk_mhz; - low_pstate_lvl.fabricclk_mhz = clk_table->entries[0].fclk_mhz; - low_pstate_lvl.socclk_mhz = clk_table->entries[0].socclk_mhz; - low_pstate_lvl.dram_speed_mts = clk_table->entries[0].memclk_mhz * 2; - - low_pstate_lvl.dispclk_mhz = dcn2_1_soc.clock_limits[high_voltage_lvl].dispclk_mhz; - low_pstate_lvl.dppclk_mhz = dcn2_1_soc.clock_limits[high_voltage_lvl].dppclk_mhz; - low_pstate_lvl.dram_bw_per_chan_gbps = dcn2_1_soc.clock_limits[high_voltage_lvl].dram_bw_per_chan_gbps; - low_pstate_lvl.dscclk_mhz = dcn2_1_soc.clock_limits[high_voltage_lvl].dscclk_mhz; - low_pstate_lvl.dtbclk_mhz = dcn2_1_soc.clock_limits[high_voltage_lvl].dtbclk_mhz; - low_pstate_lvl.phyclk_d18_mhz = dcn2_1_soc.clock_limits[high_voltage_lvl].phyclk_d18_mhz; - low_pstate_lvl.phyclk_mhz = dcn2_1_soc.clock_limits[high_voltage_lvl].phyclk_mhz; - - for (i = clk_table->num_entries; i > 1; i--) - clk_table->entries[i] = clk_table->entries[i-1]; - clk_table->entries[1] = clk_table->entries[0]; - clk_table->num_entries++; - - return low_pstate_lvl; -} - -static void update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) -{ - struct dcn21_resource_pool *pool = TO_DCN21_RES_POOL(dc->res_pool); - struct clk_limit_table *clk_table = &bw_params->clk_table; - struct _vcs_dpi_voltage_scaling_st clock_limits[DC__VOLTAGE_STATES]; - unsigned int i, closest_clk_lvl = 0, k = 0; - int j; - - dcn2_1_ip.max_num_otg = pool->base.res_cap->num_timing_generator; - dcn2_1_ip.max_num_dpp = pool->base.pipe_count; - dcn2_1_soc.num_chans = bw_params->num_channels; - - ASSERT(clk_table->num_entries); - /* Copy dcn2_1_soc.clock_limits to clock_limits to avoid copying over null states later */ - for (i = 0; i < dcn2_1_soc.num_states + 1; i++) { - clock_limits[i] = dcn2_1_soc.clock_limits[i]; - } - - for (i = 0; i < clk_table->num_entries; i++) { - /* loop backwards*/ - for (closest_clk_lvl = 0, j = dcn2_1_soc.num_states - 1; j >= 0; j--) { - if ((unsigned int) dcn2_1_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) { - closest_clk_lvl = j; - break; - } - } - - /* clk_table[1] is reserved for min DF PState. skip here to fill in later. */ - if (i == 1) - k++; - - clock_limits[k].state = k; - clock_limits[k].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz; - clock_limits[k].fabricclk_mhz = clk_table->entries[i].fclk_mhz; - clock_limits[k].socclk_mhz = clk_table->entries[i].socclk_mhz; - clock_limits[k].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2; - - clock_limits[k].dispclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dispclk_mhz; - clock_limits[k].dppclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dppclk_mhz; - clock_limits[k].dram_bw_per_chan_gbps = dcn2_1_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps; - clock_limits[k].dscclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dscclk_mhz; - clock_limits[k].dtbclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dtbclk_mhz; - clock_limits[k].phyclk_d18_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz; - clock_limits[k].phyclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].phyclk_mhz; - - k++; - } - for (i = 0; i < clk_table->num_entries + 1; i++) - dcn2_1_soc.clock_limits[i] = clock_limits[i]; - if (clk_table->num_entries) { - dcn2_1_soc.num_states = clk_table->num_entries + 1; - /* fill in min DF PState */ - dcn2_1_soc.clock_limits[1] = construct_low_pstate_lvl(clk_table, closest_clk_lvl); - /* duplicate last level */ - dcn2_1_soc.clock_limits[dcn2_1_soc.num_states] = dcn2_1_soc.clock_limits[dcn2_1_soc.num_states - 1]; - dcn2_1_soc.clock_limits[dcn2_1_soc.num_states].state = dcn2_1_soc.num_states; - } - - dml_init_instance(&dc->dml, &dcn2_1_soc, &dcn2_1_ip, DML_PROJECT_DCN21); -} - static struct pp_smu_funcs *dcn21_pp_smu_create(struct dc_context *ctx) { struct pp_smu_funcs *pp_smu = kzalloc(sizeof(*pp_smu), GFP_KERNEL); @@ -1896,24 +1378,6 @@ static uint32_t read_pipe_fuses(struct dc_context *ctx) return value; } -static int dcn21_populate_dml_pipes_from_context( - struct dc *dc, - struct dc_state *context, - display_e2e_pipe_params_st *pipes, - bool fast_validate) -{ - uint32_t pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); - int i; - - for (i = 0; i < pipe_cnt; i++) { - - pipes[i].pipe.src.hostvm = dc->res_pool->hubbub->riommu_active; - pipes[i].pipe.src.gpuvm = 1; - } - - return pipe_cnt; -} - static enum dc_status dcn21_patch_unknown_plane_state(struct dc_plane_state *plane_state) { enum dc_status result = DC_OK; @@ -1941,7 +1405,7 @@ static const struct resource_funcs dcn21_res_pool_funcs = { .patch_unknown_plane_state = dcn21_patch_unknown_plane_state, .set_mcif_arb_params = dcn20_set_mcif_arb_params, .find_first_free_match_stream_enc_for_link = dcn10_find_first_free_match_stream_enc_for_link, - .update_bw_bounding_box = update_bw_bounding_box + .update_bw_bounding_box = dcn21_update_bw_bounding_box, }; static bool dcn21_resource_construct( diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.h b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.h index a27355171bca..f7ecc002c2f7 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.h @@ -35,11 +35,22 @@ struct dc; struct resource_pool; struct _vcs_dpi_display_pipe_params_st; +extern struct _vcs_dpi_ip_params_st dcn2_1_ip; +extern struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc; + struct dcn21_resource_pool { struct resource_pool base; }; struct resource_pool *dcn21_create_resource_pool( const struct dc_init_data *init_data, struct dc *dc); +bool dcn21_fast_validate_bw( + struct dc *dc, + struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int *pipe_cnt_out, + int *pipe_split_from, + int *vlevel_out, + bool fast_validate); #endif /* _DCN21_RESOURCE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp_cm.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp_cm.c index 387eec616162..87dbeca18984 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp_cm.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp_cm.c @@ -408,6 +408,6 @@ void dpp3_cm_set_gamut_remap( gamut_mode = 1; //follow dcn2 approach for now - using only coefficient set A - program_gamut_remap(dpp, arr_reg_val, GAMUT_REMAP_COEFF); + program_gamut_remap(dpp, arr_reg_val, gamut_mode); } } diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubbub.c index f4414de96acc..152c9c5733f1 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubbub.c @@ -448,6 +448,7 @@ static const struct hubbub_funcs hubbub30_funcs = { .program_watermarks = hubbub3_program_watermarks, .allow_self_refresh_control = hubbub1_allow_self_refresh_control, .is_allow_self_refresh_enabled = hubbub1_is_allow_self_refresh_enabled, + .verify_allow_pstate_change_high = hubbub1_verify_allow_pstate_change_high, .force_wm_propagate_to_pipes = hubbub3_force_wm_propagate_to_pipes, .force_pstate_change_control = hubbub3_force_pstate_change_control, .init_watermarks = hubbub3_init_watermarks, diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c index f10f7a0ca02a..e6a62cc75139 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c @@ -1462,7 +1462,9 @@ int dcn30_populate_dml_pipes_from_context( int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; + DC_FP_START(); dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + DC_FP_END(); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { if (!res_ctx->pipe_ctx[i].stream) @@ -1731,7 +1733,10 @@ static bool init_soc_bounding_box(struct dc *dc, loaded_ip->max_num_otg = pool->base.res_cap->num_timing_generator; loaded_ip->max_num_dpp = pool->base.pipe_count; loaded_ip->clamp_min_dcfclk = dc->config.clamp_min_dcfclk; + + DC_FP_START(); dcn20_patch_bounding_box(dc, loaded_bb); + DC_FP_END(); if (dc->ctx->dc_bios->funcs->get_soc_bb_info) { struct bp_soc_bb_info bb_info = {0}; @@ -2261,7 +2266,9 @@ static noinline void dcn30_calculate_wm_and_dlg_fp( pipe_idx++; } + DC_FP_START(); dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); + DC_FP_END(); if (!pstate_en) /* Restore full p-state latency */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_hubbub.c index 1e3bd2e9cdcc..a046664e2031 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_hubbub.c @@ -60,6 +60,7 @@ static const struct hubbub_funcs hubbub301_funcs = { .program_watermarks = hubbub3_program_watermarks, .allow_self_refresh_control = hubbub1_allow_self_refresh_control, .is_allow_self_refresh_enabled = hubbub1_is_allow_self_refresh_enabled, + .verify_allow_pstate_change_high = hubbub1_verify_allow_pstate_change_high, .force_wm_propagate_to_pipes = hubbub3_force_wm_propagate_to_pipes, .force_pstate_change_control = hubbub3_force_pstate_change_control, .hubbub_read_state = hubbub2_read_state, diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/Makefile b/drivers/gpu/drm/amd/display/dc/dcn303/Makefile index 6f7a1f2b49f0..8702e0b7fda3 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn303/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn303/Makefile @@ -8,32 +8,6 @@ DCN3_03 = dcn303_init.o dcn303_hwseq.o dcn303_resource.o -ifdef CONFIG_X86 -CFLAGS_$(AMDDALPATH)/dc/dcn303/dcn303_resource.o := -msse -endif - -ifdef CONFIG_PPC64 -CFLAGS_$(AMDDALPATH)/dc/dcn303/dcn303_resource.o := -mhard-float -maltivec -endif - -ifdef CONFIG_CC_IS_GCC -ifeq ($(call cc-ifversion, -lt, 0701, y), y) -IS_OLD_GCC = 1 -endif -CFLAGS_$(AMDDALPATH)/dc/dcn303/dcn303_resource.o += -mhard-float -endif - -ifdef CONFIG_X86 -ifdef IS_OLD_GCC -# Stack alignment mismatch, proceed with caution. -# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3 -# (8B stack alignment). -CFLAGS_$(AMDDALPATH)/dc/dcn303/dcn303_resource.o += -mpreferred-stack-boundary=4 -else -CFLAGS_$(AMDDALPATH)/dc/dcn303/dcn303_resource.o += -msse2 -endif -endif - AMD_DAL_DCN3_03 = $(addprefix $(AMDDALPATH)/dc/dcn303/,$(DCN3_03)) AMD_DISPLAY_FILES += $(AMD_DAL_DCN3_03) diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c index 36649716e991..4fcbc0502808 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c @@ -52,140 +52,10 @@ #include "dpcs/dpcs_3_0_3_sh_mask.h" #include "nbio/nbio_2_3_offset.h" -#define DC_LOGGER_INIT(logger) +#include "dml/dcn303/dcn303_fpu.h" -struct _vcs_dpi_ip_params_st dcn3_03_ip = { - .use_min_dcfclk = 0, - .clamp_min_dcfclk = 0, - .odm_capable = 1, - .gpuvm_enable = 1, - .hostvm_enable = 0, - .gpuvm_max_page_table_levels = 4, - .hostvm_max_page_table_levels = 4, - .hostvm_cached_page_table_levels = 0, - .pte_group_size_bytes = 2048, - .num_dsc = 2, - .rob_buffer_size_kbytes = 184, - .det_buffer_size_kbytes = 184, - .dpte_buffer_size_in_pte_reqs_luma = 64, - .dpte_buffer_size_in_pte_reqs_chroma = 34, - .pde_proc_buffer_size_64k_reqs = 48, - .dpp_output_buffer_pixels = 2560, - .opp_output_buffer_lines = 1, - .pixel_chunk_size_kbytes = 8, - .pte_enable = 1, - .max_page_table_levels = 2, - .pte_chunk_size_kbytes = 2, // ? - .meta_chunk_size_kbytes = 2, - .writeback_chunk_size_kbytes = 8, - .line_buffer_size_bits = 789504, - .is_line_buffer_bpp_fixed = 0, // ? - .line_buffer_fixed_bpp = 0, // ? - .dcc_supported = true, - .writeback_interface_buffer_size_kbytes = 90, - .writeback_line_buffer_buffer_size = 0, - .max_line_buffer_lines = 12, - .writeback_luma_buffer_size_kbytes = 12, // writeback_line_buffer_buffer_size = 656640 - .writeback_chroma_buffer_size_kbytes = 8, - .writeback_chroma_line_buffer_width_pixels = 4, - .writeback_max_hscl_ratio = 1, - .writeback_max_vscl_ratio = 1, - .writeback_min_hscl_ratio = 1, - .writeback_min_vscl_ratio = 1, - .writeback_max_hscl_taps = 1, - .writeback_max_vscl_taps = 1, - .writeback_line_buffer_luma_buffer_size = 0, - .writeback_line_buffer_chroma_buffer_size = 14643, - .cursor_buffer_size = 8, - .cursor_chunk_size = 2, - .max_num_otg = 2, - .max_num_dpp = 2, - .max_num_wb = 1, - .max_dchub_pscl_bw_pix_per_clk = 4, - .max_pscl_lb_bw_pix_per_clk = 2, - .max_lb_vscl_bw_pix_per_clk = 4, - .max_vscl_hscl_bw_pix_per_clk = 4, - .max_hscl_ratio = 6, - .max_vscl_ratio = 6, - .hscl_mults = 4, - .vscl_mults = 4, - .max_hscl_taps = 8, - .max_vscl_taps = 8, - .dispclk_ramp_margin_percent = 1, - .underscan_factor = 1.11, - .min_vblank_lines = 32, - .dppclk_delay_subtotal = 46, - .dynamic_metadata_vm_enabled = true, - .dppclk_delay_scl_lb_only = 16, - .dppclk_delay_scl = 50, - .dppclk_delay_cnvc_formatter = 27, - .dppclk_delay_cnvc_cursor = 6, - .dispclk_delay_subtotal = 119, - .dcfclk_cstate_latency = 5.2, // SRExitTime - .max_inter_dcn_tile_repeaters = 8, - .max_num_hdmi_frl_outputs = 1, - .odm_combine_4to1_supported = false, - .xfc_supported = false, - .xfc_fill_bw_overhead_percent = 10.0, - .xfc_fill_constant_bytes = 0, - .gfx7_compat_tiling_supported = 0, - .number_of_cursors = 1, -}; - -struct _vcs_dpi_soc_bounding_box_st dcn3_03_soc = { - .clock_limits = { - { - .state = 0, - .dispclk_mhz = 1217.0, - .dppclk_mhz = 1217.0, - .phyclk_mhz = 810.0, - .phyclk_d18_mhz = 667.0, - .dscclk_mhz = 405.6, - }, - }, +#define DC_LOGGER_INIT(logger) - .min_dcfclk = 500.0, /* TODO: set this to actual min DCFCLK */ - .num_states = 1, - .sr_exit_time_us = 35.5, - .sr_enter_plus_exit_time_us = 40, - .urgent_latency_us = 4.0, - .urgent_latency_pixel_data_only_us = 4.0, - .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, - .urgent_latency_vm_data_only_us = 4.0, - .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, - .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, - .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, - .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 80.0, - .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0, - .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, - .max_avg_sdp_bw_use_normal_percent = 60.0, - .max_avg_dram_bw_use_normal_percent = 40.0, - .writeback_latency_us = 12.0, - .max_request_size_bytes = 256, - .fabric_datapath_to_dcn_data_return_bytes = 64, - .dcn_downspread_percent = 0.5, - .downspread_percent = 0.38, - .dram_page_open_time_ns = 50.0, - .dram_rw_turnaround_time_ns = 17.5, - .dram_return_buffer_per_channel_bytes = 8192, - .round_trip_ping_latency_dcfclk_cycles = 156, - .urgent_out_of_order_return_per_channel_bytes = 4096, - .channel_interleave_bytes = 256, - .num_banks = 8, - .gpuvm_min_page_size_bytes = 4096, - .hostvm_min_page_size_bytes = 4096, - .dram_clock_change_latency_us = 404, - .dummy_pstate_latency_us = 5, - .writeback_dram_clock_change_latency_us = 23.0, - .return_bus_width_bytes = 64, - .dispclk_dppclk_vco_speed_mhz = 3650, - .xfc_bus_transport_time_us = 20, // ? - .xfc_xbuf_latency_tolerance_us = 4, // ? - .use_urgent_burst_bw = 1, // ? - .do_urgent_latency_adjustment = true, - .urgent_latency_adjustment_fabric_clock_component_us = 1.0, - .urgent_latency_adjustment_fabric_clock_reference_mhz = 1000, -}; static const struct dc_debug_options debug_defaults_drv = { .disable_dmcu = true, @@ -1031,24 +901,18 @@ static bool init_soc_bounding_box(struct dc *dc, struct resource_pool *pool) loaded_ip->max_num_otg = pool->pipe_count; loaded_ip->max_num_dpp = pool->pipe_count; loaded_ip->clamp_min_dcfclk = dc->config.clamp_min_dcfclk; + DC_FP_START(); dcn20_patch_bounding_box(dc, loaded_bb); + DC_FP_END(); if (dc->ctx->dc_bios->funcs->get_soc_bb_info) { struct bp_soc_bb_info bb_info = { 0 }; if (dc->ctx->dc_bios->funcs->get_soc_bb_info( dc->ctx->dc_bios, &bb_info) == BP_RESULT_OK) { - if (bb_info.dram_clock_change_latency_100ns > 0) - dcn3_03_soc.dram_clock_change_latency_us = - bb_info.dram_clock_change_latency_100ns * 10; - - if (bb_info.dram_sr_enter_exit_latency_100ns > 0) - dcn3_03_soc.sr_enter_plus_exit_time_us = - bb_info.dram_sr_enter_exit_latency_100ns * 10; - - if (bb_info.dram_sr_exit_latency_100ns > 0) - dcn3_03_soc.sr_exit_time_us = - bb_info.dram_sr_exit_latency_100ns * 10; + DC_FP_START(); + dcn303_fpu_init_soc_bounding_box(bb_info); + DC_FP_END(); } } @@ -1186,183 +1050,12 @@ static void dcn303_destroy_resource_pool(struct resource_pool **pool) *pool = NULL; } -static void dcn303_get_optimal_dcfclk_fclk_for_uclk(unsigned int uclk_mts, - unsigned int *optimal_dcfclk, - unsigned int *optimal_fclk) -{ - double bw_from_dram, bw_from_dram1, bw_from_dram2; - - bw_from_dram1 = uclk_mts * dcn3_03_soc.num_chans * - dcn3_03_soc.dram_channel_width_bytes * (dcn3_03_soc.max_avg_dram_bw_use_normal_percent / 100); - bw_from_dram2 = uclk_mts * dcn3_03_soc.num_chans * - dcn3_03_soc.dram_channel_width_bytes * (dcn3_03_soc.max_avg_sdp_bw_use_normal_percent / 100); - - bw_from_dram = (bw_from_dram1 < bw_from_dram2) ? bw_from_dram1 : bw_from_dram2; - - if (optimal_fclk) - *optimal_fclk = bw_from_dram / - (dcn3_03_soc.fabric_datapath_to_dcn_data_return_bytes * - (dcn3_03_soc.max_avg_sdp_bw_use_normal_percent / 100)); - - if (optimal_dcfclk) - *optimal_dcfclk = bw_from_dram / - (dcn3_03_soc.return_bus_width_bytes * (dcn3_03_soc.max_avg_sdp_bw_use_normal_percent / 100)); -} void dcn303_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) { - unsigned int i, j; - unsigned int num_states = 0; - - unsigned int dcfclk_mhz[DC__VOLTAGE_STATES] = {0}; - unsigned int dram_speed_mts[DC__VOLTAGE_STATES] = {0}; - unsigned int optimal_uclk_for_dcfclk_sta_targets[DC__VOLTAGE_STATES] = {0}; - unsigned int optimal_dcfclk_for_uclk[DC__VOLTAGE_STATES] = {0}; - - unsigned int dcfclk_sta_targets[DC__VOLTAGE_STATES] = {694, 875, 1000, 1200}; - unsigned int num_dcfclk_sta_targets = 4; - unsigned int num_uclk_states; - - - if (dc->ctx->dc_bios->vram_info.num_chans) - dcn3_03_soc.num_chans = dc->ctx->dc_bios->vram_info.num_chans; - - if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) - dcn3_03_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; - - dcn3_03_soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0; - dc->dml.soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0; - - if (bw_params->clk_table.entries[0].memclk_mhz) { - int max_dcfclk_mhz = 0, max_dispclk_mhz = 0, max_dppclk_mhz = 0, max_phyclk_mhz = 0; - - for (i = 0; i < MAX_NUM_DPM_LVL; i++) { - if (bw_params->clk_table.entries[i].dcfclk_mhz > max_dcfclk_mhz) - max_dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz; - if (bw_params->clk_table.entries[i].dispclk_mhz > max_dispclk_mhz) - max_dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz; - if (bw_params->clk_table.entries[i].dppclk_mhz > max_dppclk_mhz) - max_dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz; - if (bw_params->clk_table.entries[i].phyclk_mhz > max_phyclk_mhz) - max_phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz; - } - if (!max_dcfclk_mhz) - max_dcfclk_mhz = dcn3_03_soc.clock_limits[0].dcfclk_mhz; - if (!max_dispclk_mhz) - max_dispclk_mhz = dcn3_03_soc.clock_limits[0].dispclk_mhz; - if (!max_dppclk_mhz) - max_dppclk_mhz = dcn3_03_soc.clock_limits[0].dppclk_mhz; - if (!max_phyclk_mhz) - max_phyclk_mhz = dcn3_03_soc.clock_limits[0].phyclk_mhz; - - if (max_dcfclk_mhz > dcfclk_sta_targets[num_dcfclk_sta_targets-1]) { - dcfclk_sta_targets[num_dcfclk_sta_targets] = max_dcfclk_mhz; - num_dcfclk_sta_targets++; - } else if (max_dcfclk_mhz < dcfclk_sta_targets[num_dcfclk_sta_targets-1]) { - for (i = 0; i < num_dcfclk_sta_targets; i++) { - if (dcfclk_sta_targets[i] > max_dcfclk_mhz) { - dcfclk_sta_targets[i] = max_dcfclk_mhz; - break; - } - } - /* Update size of array since we "removed" duplicates */ - num_dcfclk_sta_targets = i + 1; - } - - num_uclk_states = bw_params->clk_table.num_entries; - - /* Calculate optimal dcfclk for each uclk */ - for (i = 0; i < num_uclk_states; i++) { - dcn303_get_optimal_dcfclk_fclk_for_uclk(bw_params->clk_table.entries[i].memclk_mhz * 16, - &optimal_dcfclk_for_uclk[i], NULL); - if (optimal_dcfclk_for_uclk[i] < bw_params->clk_table.entries[0].dcfclk_mhz) - optimal_dcfclk_for_uclk[i] = bw_params->clk_table.entries[0].dcfclk_mhz; - } - - /* Calculate optimal uclk for each dcfclk sta target */ - for (i = 0; i < num_dcfclk_sta_targets; i++) { - for (j = 0; j < num_uclk_states; j++) { - if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) { - optimal_uclk_for_dcfclk_sta_targets[i] = - bw_params->clk_table.entries[j].memclk_mhz * 16; - break; - } - } - } - - i = 0; - j = 0; - /* create the final dcfclk and uclk table */ - while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) { - if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) { - dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; - dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; - } else { - if (j < num_uclk_states && optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) { - dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j]; - dram_speed_mts[num_states++] = - bw_params->clk_table.entries[j++].memclk_mhz * 16; - } else { - j = num_uclk_states; - } - } - } - - while (i < num_dcfclk_sta_targets && num_states < DC__VOLTAGE_STATES) { - dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; - dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; - } - - while (j < num_uclk_states && num_states < DC__VOLTAGE_STATES && - optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) { - dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j]; - dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16; - } - - dcn3_03_soc.num_states = num_states; - for (i = 0; i < dcn3_03_soc.num_states; i++) { - dcn3_03_soc.clock_limits[i].state = i; - dcn3_03_soc.clock_limits[i].dcfclk_mhz = dcfclk_mhz[i]; - dcn3_03_soc.clock_limits[i].fabricclk_mhz = dcfclk_mhz[i]; - dcn3_03_soc.clock_limits[i].dram_speed_mts = dram_speed_mts[i]; - - /* Fill all states with max values of all other clocks */ - dcn3_03_soc.clock_limits[i].dispclk_mhz = max_dispclk_mhz; - dcn3_03_soc.clock_limits[i].dppclk_mhz = max_dppclk_mhz; - dcn3_03_soc.clock_limits[i].phyclk_mhz = max_phyclk_mhz; - /* Populate from bw_params for DTBCLK, SOCCLK */ - if (!bw_params->clk_table.entries[i].dtbclk_mhz && i > 0) - dcn3_03_soc.clock_limits[i].dtbclk_mhz = dcn3_03_soc.clock_limits[i-1].dtbclk_mhz; - else - dcn3_03_soc.clock_limits[i].dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz; - if (!bw_params->clk_table.entries[i].socclk_mhz && i > 0) - dcn3_03_soc.clock_limits[i].socclk_mhz = dcn3_03_soc.clock_limits[i-1].socclk_mhz; - else - dcn3_03_soc.clock_limits[i].socclk_mhz = bw_params->clk_table.entries[i].socclk_mhz; - /* These clocks cannot come from bw_params, always fill from dcn3_03_soc[1] */ - /* FCLK, PHYCLK_D18, DSCCLK */ - dcn3_03_soc.clock_limits[i].phyclk_d18_mhz = dcn3_03_soc.clock_limits[0].phyclk_d18_mhz; - dcn3_03_soc.clock_limits[i].dscclk_mhz = dcn3_03_soc.clock_limits[0].dscclk_mhz; - } - - // WA: patch strobe modes to compensate for DCN303 BW issue - if (dcn3_03_soc.num_chans <= 4) { - for (i = 0; i < dcn3_03_soc.num_states; i++) { - if (dcn3_03_soc.clock_limits[i].dram_speed_mts > 1700) - break; - - if (dcn3_03_soc.clock_limits[i].dram_speed_mts >= 1500) { - dcn3_03_soc.clock_limits[i].dcfclk_mhz = 100; - dcn3_03_soc.clock_limits[i].fabricclk_mhz = 100; - } - } - } - - /* re-init DML with updated bb */ - dml_init_instance(&dc->dml, &dcn3_03_soc, &dcn3_03_ip, DML_PROJECT_DCN30); - if (dc->current_state) - dml_init_instance(&dc->current_state->bw_ctx.dml, &dcn3_03_soc, &dcn3_03_ip, DML_PROJECT_DCN30); - } + DC_FP_START(); + dcn303_fpu_update_bw_bounding_box(dc, bw_params); + DC_FP_END(); } static struct resource_funcs dcn303_res_pool_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.h b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.h index 5b590c169763..9c7d79540900 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.h @@ -10,6 +10,9 @@ #include "core_types.h" +extern struct _vcs_dpi_ip_params_st dcn3_03_ip; +extern struct _vcs_dpi_soc_bounding_box_st dcn3_03_soc; + struct resource_pool *dcn303_create_resource_pool(const struct dc_init_data *init_data, struct dc *dc); void dcn303_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params); diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c index 8b9b1a5309ba..d94fd1010deb 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c @@ -240,18 +240,9 @@ static void enc31_hw_init(struct link_encoder *enc) // 100MHz -> 0x32 // 48MHz -> 0x18 -#ifdef CLEANUP_FIXME - /*from display_init*/ - REG_WRITE(RDPCSTX_DEBUG_CONFIG, 0); -#endif - // Set TMDS_CTL0 to 1. This is a legacy setting. REG_UPDATE(TMDS_CTL_BITS, TMDS_CTL0, 1); - /*HW default is 5*/ - REG_UPDATE(RDPCSTX_CNTL, - RDPCS_TX_FIFO_RD_START_DELAY, 4); - dcn10_aux_initialize(enc10); } diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hubbub.c index 5e3bcaf12cac..3e6d6ebd199e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hubbub.c @@ -949,6 +949,65 @@ static void hubbub31_get_dchub_ref_freq(struct hubbub *hubbub, } } +static bool hubbub31_verify_allow_pstate_change_high(struct hubbub *hubbub) +{ + struct dcn20_hubbub *hubbub2 = TO_DCN20_HUBBUB(hubbub); + + /* + * Pstate latency is ~20us so if we wait over 40us and pstate allow + * still not asserted, we are probably stuck and going to hang + */ + const unsigned int pstate_wait_timeout_us = 100; + const unsigned int pstate_wait_expected_timeout_us = 40; + + static unsigned int max_sampled_pstate_wait_us; /* data collection */ + static bool forced_pstate_allow; /* help with revert wa */ + + unsigned int debug_data = 0; + unsigned int i; + + if (forced_pstate_allow) { + /* we hacked to force pstate allow to prevent hang last time + * we verify_allow_pstate_change_high. so disable force + * here so we can check status + */ + REG_UPDATE_2(DCHUBBUB_ARB_DRAM_STATE_CNTL, + DCHUBBUB_ARB_ALLOW_PSTATE_CHANGE_FORCE_VALUE, 0, + DCHUBBUB_ARB_ALLOW_PSTATE_CHANGE_FORCE_ENABLE, 0); + forced_pstate_allow = false; + } + + REG_WRITE(DCHUBBUB_TEST_DEBUG_INDEX, hubbub2->debug_test_index_pstate); + + for (i = 0; i < pstate_wait_timeout_us; i++) { + debug_data = REG_READ(DCHUBBUB_TEST_DEBUG_DATA); + + /* Debug bit is specific to ASIC. */ + if (debug_data & (1 << 26)) { + if (i > pstate_wait_expected_timeout_us) + DC_LOG_WARNING("pstate took longer than expected ~%dus\n", i); + return true; + } + if (max_sampled_pstate_wait_us < i) + max_sampled_pstate_wait_us = i; + + udelay(1); + } + + /* force pstate allow to prevent system hang + * and break to debugger to investigate + */ + REG_UPDATE_2(DCHUBBUB_ARB_DRAM_STATE_CNTL, + DCHUBBUB_ARB_ALLOW_PSTATE_CHANGE_FORCE_VALUE, 1, + DCHUBBUB_ARB_ALLOW_PSTATE_CHANGE_FORCE_ENABLE, 1); + forced_pstate_allow = true; + + DC_LOG_WARNING("pstate TEST_DEBUG_DATA: 0x%X\n", + debug_data); + + return false; +} + static const struct hubbub_funcs hubbub31_funcs = { .update_dchub = hubbub2_update_dchub, .init_dchub_sys_ctx = hubbub31_init_dchub_sys_ctx, @@ -961,6 +1020,7 @@ static const struct hubbub_funcs hubbub31_funcs = { .program_watermarks = hubbub31_program_watermarks, .allow_self_refresh_control = hubbub1_allow_self_refresh_control, .is_allow_self_refresh_enabled = hubbub1_is_allow_self_refresh_enabled, + .verify_allow_pstate_change_high = hubbub31_verify_allow_pstate_change_high, .program_det_size = dcn31_program_det_size, .program_compbuf_size = dcn31_program_compbuf_size, .init_crb = dcn31_init_crb, diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c index 83ece02380a8..11ea9d13e312 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c @@ -83,7 +83,8 @@ static uint32_t dcn31_panel_cntl_hw_init(struct panel_cntl *panel_cntl) cmd.panel_cntl.data.bl_pwm_period_cntl = panel_cntl->stored_backlight_registers.BL_PWM_PERIOD_CNTL; cmd.panel_cntl.data.bl_pwm_ref_div1 = panel_cntl->stored_backlight_registers.LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV; - + cmd.panel_cntl.data.bl_pwm_ref_div2 = + panel_cntl->stored_backlight_registers.PANEL_PWRSEQ_REF_DIV2; if (!dc_dmub_srv_cmd_with_reply_data(dc_dmub_srv, &cmd)) return 0; @@ -92,6 +93,8 @@ static uint32_t dcn31_panel_cntl_hw_init(struct panel_cntl *panel_cntl) panel_cntl->stored_backlight_registers.BL_PWM_PERIOD_CNTL = cmd.panel_cntl.data.bl_pwm_period_cntl; panel_cntl->stored_backlight_registers.LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV = cmd.panel_cntl.data.bl_pwm_ref_div1; + panel_cntl->stored_backlight_registers.PANEL_PWRSEQ_REF_DIV2 = + cmd.panel_cntl.data.bl_pwm_ref_div2; return cmd.panel_cntl.data.current_backlight; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c index 660e96a7fe7f..89b7b6b7254a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c @@ -1011,7 +1011,7 @@ static const struct dc_debug_options debug_defaults_drv = { .max_downscale_src_width = 4096,/*upto true 4K*/ .disable_pplib_wm_range = false, .scl_reset_length10 = true, - .sanity_checks = false, + .sanity_checks = true, .underflow_assert_delay_us = 0xFFFFFFFF, .dwb_fi_phase = -1, // -1 = disable, .dmub_command_table = true, @@ -1777,7 +1777,7 @@ static bool is_dual_plane(enum surface_pixel_format format) return format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN || format == SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA; } -static int dcn31_populate_dml_pipes_from_context( +int dcn31_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, bool fast_validate) @@ -1787,7 +1787,9 @@ static int dcn31_populate_dml_pipes_from_context( struct pipe_ctx *pipe; bool upscaled = false; + DC_FP_START(); dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + DC_FP_END(); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { struct dc_crtc_timing *timing; @@ -1810,6 +1812,7 @@ static int dcn31_populate_dml_pipes_from_context( pipes[pipe_cnt].pipe.src.immediate_flip = true; pipes[pipe_cnt].pipe.src.unbounded_req_mode = false; + pipes[pipe_cnt].pipe.src.hostvm = dc->res_pool->hubbub->riommu_active; pipes[pipe_cnt].pipe.src.gpuvm = true; pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_luma = 0; pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_chroma = 0; @@ -1998,7 +2001,9 @@ static void dcn31_calculate_wm_and_dlg_fp( pipe_idx++; } + DC_FP_START(); dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); + DC_FP_END(); } void dcn31_calculate_wm_and_dlg( @@ -2068,7 +2073,7 @@ static struct dc_cap_funcs cap_funcs = { .get_dcc_compression_cap = dcn20_get_dcc_compression_cap }; -static void dcn31_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) +void dcn31_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) { struct clk_limit_table *clk_table = &bw_params->clk_table; struct _vcs_dpi_voltage_scaling_st clock_limits[DC__VOLTAGE_STATES]; @@ -2176,7 +2181,7 @@ static struct clock_source *dcn30_clock_source_create( if (!clk_src) return NULL; - if (dcn3_clk_src_construct(clk_src, ctx, bios, id, + if (dcn31_clk_src_construct(clk_src, ctx, bios, id, regs, &cs_shift, &cs_mask)) { clk_src->base.dp_clk_src = dp_clk_src; return &clk_src->base; diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.h b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.h index a513363b3326..4b7ab21ea15b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.h @@ -43,6 +43,11 @@ void dcn31_calculate_wm_and_dlg( display_e2e_pipe_params_st *pipes, int pipe_cnt, int vlevel); +int dcn31_populate_dml_pipes_from_context( + struct dc *dc, struct dc_state *context, + display_e2e_pipe_params_st *pipes, + bool fast_validate); +void dcn31_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params); void dcn31_update_soc_for_wm_a(struct dc *dc, struct dc_state *context); struct resource_pool *dcn31_create_resource_pool( diff --git a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c index a71073482881..06adb77c206b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c @@ -1035,6 +1035,7 @@ static const struct dc_debug_options debug_defaults_drv = { }, .optimize_edp_link_rate = true, .enable_sw_cntl_psr = true, + .psr_power_use_phy_fsm = 0, }; static const struct dc_debug_options debug_defaults_diags = { @@ -1759,7 +1760,7 @@ static struct clock_source *dcn31_clock_source_create( if (!clk_src) return NULL; - if (dcn3_clk_src_construct(clk_src, ctx, bios, id, + if (dcn31_clk_src_construct(clk_src, ctx, bios, id, regs, &cs_shift, &cs_mask)) { clk_src->base.dp_clk_src = dp_clk_src; return &clk_src->base; @@ -1784,7 +1785,9 @@ static int dcn315_populate_dml_pipes_from_context( struct pipe_ctx *pipe; const int max_usable_det = context->bw_ctx.dml.ip.config_return_buffer_size_in_kbytes - DCN3_15_MIN_COMPBUF_SIZE_KB; + DC_FP_START(); dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + DC_FP_END(); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { struct dc_crtc_timing *timing; @@ -1962,29 +1965,6 @@ static struct resource_funcs dcn315_res_pool_funcs = { .patch_unknown_plane_state = dcn20_patch_unknown_plane_state, }; -static struct clock_source *dcn30_clock_source_create( - struct dc_context *ctx, - struct dc_bios *bios, - enum clock_source_id id, - const struct dce110_clk_src_regs *regs, - bool dp_clk_src) -{ - struct dce110_clk_src *clk_src = - kzalloc(sizeof(struct dce110_clk_src), GFP_KERNEL); - - if (!clk_src) - return NULL; - - if (dcn3_clk_src_construct(clk_src, ctx, bios, id, - regs, &cs_shift, &cs_mask)) { - clk_src->base.dp_clk_src = dp_clk_src; - return &clk_src->base; - } - - BREAK_TO_DEBUGGER(); - return NULL; -} - static bool dcn315_resource_construct( uint8_t num_virtual_links, struct dc *dc, @@ -2090,23 +2070,23 @@ static bool dcn315_resource_construct( /* Clock Sources for Pixel Clock*/ pool->base.clock_sources[DCN31_CLK_SRC_PLL0] = - dcn30_clock_source_create(ctx, ctx->dc_bios, + dcn31_clock_source_create(ctx, ctx->dc_bios, CLOCK_SOURCE_COMBO_PHY_PLL0, &clk_src_regs[0], false); pool->base.clock_sources[DCN31_CLK_SRC_PLL1] = - dcn30_clock_source_create(ctx, ctx->dc_bios, + dcn31_clock_source_create(ctx, ctx->dc_bios, CLOCK_SOURCE_COMBO_PHY_PLL1, &clk_src_regs[1], false); pool->base.clock_sources[DCN31_CLK_SRC_PLL2] = - dcn30_clock_source_create(ctx, ctx->dc_bios, + dcn31_clock_source_create(ctx, ctx->dc_bios, CLOCK_SOURCE_COMBO_PHY_PLL2, &clk_src_regs[2], false); pool->base.clock_sources[DCN31_CLK_SRC_PLL3] = - dcn30_clock_source_create(ctx, ctx->dc_bios, + dcn31_clock_source_create(ctx, ctx->dc_bios, CLOCK_SOURCE_COMBO_PHY_PLL3, &clk_src_regs[3], false); pool->base.clock_sources[DCN31_CLK_SRC_PLL4] = - dcn30_clock_source_create(ctx, ctx->dc_bios, + dcn31_clock_source_create(ctx, ctx->dc_bios, CLOCK_SOURCE_COMBO_PHY_PLL4, &clk_src_regs[4], false); diff --git a/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c b/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c index 72d581c6092f..8decc3ccf8ca 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c @@ -1759,7 +1759,7 @@ static struct clock_source *dcn31_clock_source_create( if (!clk_src) return NULL; - if (dcn3_clk_src_construct(clk_src, ctx, bios, id, + if (dcn31_clk_src_construct(clk_src, ctx, bios, id, regs, &cs_shift, &cs_mask)) { clk_src->base.dp_clk_src = dp_clk_src; return &clk_src->base; @@ -1786,7 +1786,9 @@ static int dcn316_populate_dml_pipes_from_context( struct pipe_ctx *pipe; const int max_usable_det = context->bw_ctx.dml.ip.config_return_buffer_size_in_kbytes - DCN3_16_MIN_COMPBUF_SIZE_KB; + DC_FP_START(); dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + DC_FP_END(); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { struct dc_crtc_timing *timing; @@ -1964,29 +1966,6 @@ static struct resource_funcs dcn316_res_pool_funcs = { .patch_unknown_plane_state = dcn20_patch_unknown_plane_state, }; -static struct clock_source *dcn30_clock_source_create( - struct dc_context *ctx, - struct dc_bios *bios, - enum clock_source_id id, - const struct dce110_clk_src_regs *regs, - bool dp_clk_src) -{ - struct dce110_clk_src *clk_src = - kzalloc(sizeof(struct dce110_clk_src), GFP_KERNEL); - - if (!clk_src) - return NULL; - - if (dcn3_clk_src_construct(clk_src, ctx, bios, id, - regs, &cs_shift, &cs_mask)) { - clk_src->base.dp_clk_src = dp_clk_src; - return &clk_src->base; - } - - BREAK_TO_DEBUGGER(); - return NULL; -} - static bool dcn316_resource_construct( uint8_t num_virtual_links, struct dc *dc, @@ -2092,23 +2071,23 @@ static bool dcn316_resource_construct( /* Clock Sources for Pixel Clock*/ pool->base.clock_sources[DCN31_CLK_SRC_PLL0] = - dcn30_clock_source_create(ctx, ctx->dc_bios, + dcn31_clock_source_create(ctx, ctx->dc_bios, CLOCK_SOURCE_COMBO_PHY_PLL0, &clk_src_regs[0], false); pool->base.clock_sources[DCN31_CLK_SRC_PLL1] = - dcn30_clock_source_create(ctx, ctx->dc_bios, + dcn31_clock_source_create(ctx, ctx->dc_bios, CLOCK_SOURCE_COMBO_PHY_PLL1, &clk_src_regs[1], false); pool->base.clock_sources[DCN31_CLK_SRC_PLL2] = - dcn30_clock_source_create(ctx, ctx->dc_bios, + dcn31_clock_source_create(ctx, ctx->dc_bios, CLOCK_SOURCE_COMBO_PHY_PLL2, &clk_src_regs[2], false); pool->base.clock_sources[DCN31_CLK_SRC_PLL3] = - dcn30_clock_source_create(ctx, ctx->dc_bios, + dcn31_clock_source_create(ctx, ctx->dc_bios, CLOCK_SOURCE_COMBO_PHY_PLL3, &clk_src_regs[3], false); pool->base.clock_sources[DCN31_CLK_SRC_PLL4] = - dcn30_clock_source_create(ctx, ctx->dc_bios, + dcn31_clock_source_create(ctx, ctx->dc_bios, CLOCK_SOURCE_COMBO_PHY_PLL4, &clk_src_regs[4], false); diff --git a/drivers/gpu/drm/amd/display/dc/dm_helpers.h b/drivers/gpu/drm/amd/display/dc/dm_helpers.h index b71b5fb894e2..31109db02e93 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_helpers.h +++ b/drivers/gpu/drm/amd/display/dc/dm_helpers.h @@ -184,4 +184,7 @@ int dm_helpers_dmub_set_config_sync(struct dc_context *ctx, const struct dc_link *link, struct set_config_cmd_payload *payload, enum set_config_status *operation_result); + +enum dc_edid_status dm_helpers_get_sbios_edid(struct dc_link *link, struct dc_edid *edid); + #endif /* __DM_HELPERS__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile index b16c492593e2..28978ce62f87 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile @@ -58,6 +58,7 @@ CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_lib.o := $(dml_ccflags) ifdef CONFIG_DRM_AMD_DC_DCN CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_vba.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn10/dcn10_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/dcn20_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20.o := $(dml_ccflags) @@ -72,6 +73,7 @@ CFLAGS_$(AMDDALPATH)/dc/dml/dcn31/display_mode_vba_31.o := $(dml_ccflags) $(fram CFLAGS_$(AMDDALPATH)/dc/dml/dcn31/display_rq_dlg_calc_31.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn301/dcn301_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn302/dcn302_fpu.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn303/dcn303_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dsc/rc_calc_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_lib.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/calcs/dcn_calcs.o := $(dml_ccflags) @@ -105,6 +107,7 @@ DML = calcs/dce_calcs.o calcs/custom_float.o calcs/bw_fixed.o ifdef CONFIG_DRM_AMD_DC_DCN DML += display_mode_lib.o display_rq_dlg_helpers.o dml1_display_rq_dlg_calc.o +DML += dcn10/dcn10_fpu.o DML += dcn20/dcn20_fpu.o DML += display_mode_vba.o dcn20/display_rq_dlg_calc_20.o dcn20/display_mode_vba_20.o DML += dcn20/display_rq_dlg_calc_20v2.o dcn20/display_mode_vba_20v2.o @@ -113,6 +116,7 @@ DML += dcn30/display_mode_vba_30.o dcn30/display_rq_dlg_calc_30.o DML += dcn31/display_mode_vba_31.o dcn31/display_rq_dlg_calc_31.o DML += dcn301/dcn301_fpu.o DML += dcn302/dcn302_fpu.o +DML += dcn303/dcn303_fpu.o DML += dsc/rc_calc_fpu.o DML += calcs/dcn_calcs.o calcs/dcn_calc_math.o calcs/dcn_calc_auto.o endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn10/dcn10_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn10/dcn10_fpu.c new file mode 100644 index 000000000000..99644d896222 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn10/dcn10_fpu.c @@ -0,0 +1,123 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright 2021 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "dcn10/dcn10_resource.h" + +#include "dcn10_fpu.h" + +/** + * DOC: DCN10 FPU manipulation Overview + * + * The DCN architecture relies on FPU operations, which require special + * compilation flags and the use of kernel_fpu_begin/end functions; ideally, we + * want to avoid spreading FPU access across multiple files. With this idea in + * mind, this file aims to centralize DCN10 functions that require FPU access + * in a single place. Code in this file follows the following code pattern: + * + * 1. Functions that use FPU operations should be isolated in static functions. + * 2. The FPU functions should have the noinline attribute to ensure anything + * that deals with FP register is contained within this call. + * 3. All function that needs to be accessed outside this file requires a + * public interface that not uses any FPU reference. + * 4. Developers **must not** use DC_FP_START/END in this file, but they need + * to ensure that the caller invokes it before access any function available + * in this file. For this reason, public functions in this file must invoke + * dc_assert_fp_enabled(); + * + * Let's expand a little bit more the idea in the code pattern. To fully + * isolate FPU operations in a single place, we must avoid situations where + * compilers spill FP values to registers due to FP enable in a specific C + * file. Note that even if we isolate all FPU functions in a single file and + * call its interface from other files, the compiler might enable the use of + * FPU before we call DC_FP_START. Nevertheless, it is the programmer's + * responsibility to invoke DC_FP_START/END in the correct place. To highlight + * situations where developers forgot to use the FP protection before calling + * the DC FPU interface functions, we introduce a helper that checks if the + * function is invoked under FP protection. If not, it will trigger a kernel + * warning. + */ + +struct _vcs_dpi_ip_params_st dcn1_0_ip = { + .rob_buffer_size_kbytes = 64, + .det_buffer_size_kbytes = 164, + .dpte_buffer_size_in_pte_reqs_luma = 42, + .dpp_output_buffer_pixels = 2560, + .opp_output_buffer_lines = 1, + .pixel_chunk_size_kbytes = 8, + .pte_enable = 1, + .pte_chunk_size_kbytes = 2, + .meta_chunk_size_kbytes = 2, + .writeback_chunk_size_kbytes = 2, + .line_buffer_size_bits = 589824, + .max_line_buffer_lines = 12, + .IsLineBufferBppFixed = 0, + .LineBufferFixedBpp = -1, + .writeback_luma_buffer_size_kbytes = 12, + .writeback_chroma_buffer_size_kbytes = 8, + .max_num_dpp = 4, + .max_num_wb = 2, + .max_dchub_pscl_bw_pix_per_clk = 4, + .max_pscl_lb_bw_pix_per_clk = 2, + .max_lb_vscl_bw_pix_per_clk = 4, + .max_vscl_hscl_bw_pix_per_clk = 4, + .max_hscl_ratio = 4, + .max_vscl_ratio = 4, + .hscl_mults = 4, + .vscl_mults = 4, + .max_hscl_taps = 8, + .max_vscl_taps = 8, + .dispclk_ramp_margin_percent = 1, + .underscan_factor = 1.10, + .min_vblank_lines = 14, + .dppclk_delay_subtotal = 90, + .dispclk_delay_subtotal = 42, + .dcfclk_cstate_latency = 10, + .max_inter_dcn_tile_repeaters = 8, + .can_vstartup_lines_exceed_vsync_plus_back_porch_lines_minus_one = 0, + .bug_forcing_LC_req_same_size_fixed = 0, +}; + +struct _vcs_dpi_soc_bounding_box_st dcn1_0_soc = { + .sr_exit_time_us = 9.0, + .sr_enter_plus_exit_time_us = 11.0, + .urgent_latency_us = 4.0, + .writeback_latency_us = 12.0, + .ideal_dram_bw_after_urgent_percent = 80.0, + .max_request_size_bytes = 256, + .downspread_percent = 0.5, + .dram_page_open_time_ns = 50.0, + .dram_rw_turnaround_time_ns = 17.5, + .dram_return_buffer_per_channel_bytes = 8192, + .round_trip_ping_latency_dcfclk_cycles = 128, + .urgent_out_of_order_return_per_channel_bytes = 256, + .channel_interleave_bytes = 256, + .num_banks = 8, + .num_chans = 2, + .vmm_page_size_bytes = 4096, + .dram_clock_change_latency_us = 17.0, + .writeback_dram_clock_change_latency_us = 23.0, + .return_bus_width_bytes = 64, +}; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.c b/drivers/gpu/drm/amd/display/dc/dml/dcn10/dcn10_fpu.h index b7fbc114a175..e74ed4b4ce5b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn10/dcn10_fpu.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: MIT */ /* * Copyright 2021 Advanced Micro Devices, Inc. * @@ -19,14 +20,11 @@ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. * + * Authors: AMD + * */ -#include "amdgpu.h" -#include "amdgpu_ras.h" +#ifndef __DCN10_FPU_H__ +#define __DCN10_FPU_H__ -void amdgpu_hdp_ras_fini(struct amdgpu_device *adev) -{ - if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__HDP) && - adev->hdp.ras_if) - amdgpu_ras_block_late_fini(adev, adev->hdp.ras_if); -} +#endif /* __DCN20_FPU_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c index d590dc917363..2f6122153bdb 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c @@ -25,9 +25,23 @@ */ #include "resource.h" +#include "clk_mgr.h" +#include "dc_link_dp.h" +#include "dchubbub.h" +#include "dcn20/dcn20_resource.h" +#include "dcn21/dcn21_resource.h" #include "dcn20_fpu.h" +#define DC_LOGGER_INIT(logger) + +#ifndef MAX +#define MAX(X, Y) ((X) > (Y) ? (X) : (Y)) +#endif +#ifndef MIN +#define MIN(X, Y) ((X) < (Y) ? (X) : (Y)) +#endif + /** * DOC: DCN2x FPU manipulation Overview * @@ -61,6 +75,581 @@ * warning. */ +struct _vcs_dpi_ip_params_st dcn2_0_ip = { + .odm_capable = 1, + .gpuvm_enable = 0, + .hostvm_enable = 0, + .gpuvm_max_page_table_levels = 4, + .hostvm_max_page_table_levels = 4, + .hostvm_cached_page_table_levels = 0, + .pte_group_size_bytes = 2048, + .num_dsc = 6, + .rob_buffer_size_kbytes = 168, + .det_buffer_size_kbytes = 164, + .dpte_buffer_size_in_pte_reqs_luma = 84, + .pde_proc_buffer_size_64k_reqs = 48, + .dpp_output_buffer_pixels = 2560, + .opp_output_buffer_lines = 1, + .pixel_chunk_size_kbytes = 8, + .pte_chunk_size_kbytes = 2, + .meta_chunk_size_kbytes = 2, + .writeback_chunk_size_kbytes = 2, + .line_buffer_size_bits = 789504, + .is_line_buffer_bpp_fixed = 0, + .line_buffer_fixed_bpp = 0, + .dcc_supported = true, + .max_line_buffer_lines = 12, + .writeback_luma_buffer_size_kbytes = 12, + .writeback_chroma_buffer_size_kbytes = 8, + .writeback_chroma_line_buffer_width_pixels = 4, + .writeback_max_hscl_ratio = 1, + .writeback_max_vscl_ratio = 1, + .writeback_min_hscl_ratio = 1, + .writeback_min_vscl_ratio = 1, + .writeback_max_hscl_taps = 12, + .writeback_max_vscl_taps = 12, + .writeback_line_buffer_luma_buffer_size = 0, + .writeback_line_buffer_chroma_buffer_size = 14643, + .cursor_buffer_size = 8, + .cursor_chunk_size = 2, + .max_num_otg = 6, + .max_num_dpp = 6, + .max_num_wb = 1, + .max_dchub_pscl_bw_pix_per_clk = 4, + .max_pscl_lb_bw_pix_per_clk = 2, + .max_lb_vscl_bw_pix_per_clk = 4, + .max_vscl_hscl_bw_pix_per_clk = 4, + .max_hscl_ratio = 8, + .max_vscl_ratio = 8, + .hscl_mults = 4, + .vscl_mults = 4, + .max_hscl_taps = 8, + .max_vscl_taps = 8, + .dispclk_ramp_margin_percent = 1, + .underscan_factor = 1.10, + .min_vblank_lines = 32, // + .dppclk_delay_subtotal = 77, // + .dppclk_delay_scl_lb_only = 16, + .dppclk_delay_scl = 50, + .dppclk_delay_cnvc_formatter = 8, + .dppclk_delay_cnvc_cursor = 6, + .dispclk_delay_subtotal = 87, // + .dcfclk_cstate_latency = 10, // SRExitTime + .max_inter_dcn_tile_repeaters = 8, + .xfc_supported = true, + .xfc_fill_bw_overhead_percent = 10.0, + .xfc_fill_constant_bytes = 0, + .number_of_cursors = 1, +}; + +struct _vcs_dpi_ip_params_st dcn2_0_nv14_ip = { + .odm_capable = 1, + .gpuvm_enable = 0, + .hostvm_enable = 0, + .gpuvm_max_page_table_levels = 4, + .hostvm_max_page_table_levels = 4, + .hostvm_cached_page_table_levels = 0, + .num_dsc = 5, + .rob_buffer_size_kbytes = 168, + .det_buffer_size_kbytes = 164, + .dpte_buffer_size_in_pte_reqs_luma = 84, + .dpte_buffer_size_in_pte_reqs_chroma = 42,//todo + .dpp_output_buffer_pixels = 2560, + .opp_output_buffer_lines = 1, + .pixel_chunk_size_kbytes = 8, + .pte_enable = 1, + .max_page_table_levels = 4, + .pte_chunk_size_kbytes = 2, + .meta_chunk_size_kbytes = 2, + .writeback_chunk_size_kbytes = 2, + .line_buffer_size_bits = 789504, + .is_line_buffer_bpp_fixed = 0, + .line_buffer_fixed_bpp = 0, + .dcc_supported = true, + .max_line_buffer_lines = 12, + .writeback_luma_buffer_size_kbytes = 12, + .writeback_chroma_buffer_size_kbytes = 8, + .writeback_chroma_line_buffer_width_pixels = 4, + .writeback_max_hscl_ratio = 1, + .writeback_max_vscl_ratio = 1, + .writeback_min_hscl_ratio = 1, + .writeback_min_vscl_ratio = 1, + .writeback_max_hscl_taps = 12, + .writeback_max_vscl_taps = 12, + .writeback_line_buffer_luma_buffer_size = 0, + .writeback_line_buffer_chroma_buffer_size = 14643, + .cursor_buffer_size = 8, + .cursor_chunk_size = 2, + .max_num_otg = 5, + .max_num_dpp = 5, + .max_num_wb = 1, + .max_dchub_pscl_bw_pix_per_clk = 4, + .max_pscl_lb_bw_pix_per_clk = 2, + .max_lb_vscl_bw_pix_per_clk = 4, + .max_vscl_hscl_bw_pix_per_clk = 4, + .max_hscl_ratio = 8, + .max_vscl_ratio = 8, + .hscl_mults = 4, + .vscl_mults = 4, + .max_hscl_taps = 8, + .max_vscl_taps = 8, + .dispclk_ramp_margin_percent = 1, + .underscan_factor = 1.10, + .min_vblank_lines = 32, // + .dppclk_delay_subtotal = 77, // + .dppclk_delay_scl_lb_only = 16, + .dppclk_delay_scl = 50, + .dppclk_delay_cnvc_formatter = 8, + .dppclk_delay_cnvc_cursor = 6, + .dispclk_delay_subtotal = 87, // + .dcfclk_cstate_latency = 10, // SRExitTime + .max_inter_dcn_tile_repeaters = 8, + .xfc_supported = true, + .xfc_fill_bw_overhead_percent = 10.0, + .xfc_fill_constant_bytes = 0, + .ptoi_supported = 0, + .number_of_cursors = 1, +}; + +struct _vcs_dpi_soc_bounding_box_st dcn2_0_soc = { + /* Defaults that get patched on driver load from firmware. */ + .clock_limits = { + { + .state = 0, + .dcfclk_mhz = 560.0, + .fabricclk_mhz = 560.0, + .dispclk_mhz = 513.0, + .dppclk_mhz = 513.0, + .phyclk_mhz = 540.0, + .socclk_mhz = 560.0, + .dscclk_mhz = 171.0, + .dram_speed_mts = 8960.0, + }, + { + .state = 1, + .dcfclk_mhz = 694.0, + .fabricclk_mhz = 694.0, + .dispclk_mhz = 642.0, + .dppclk_mhz = 642.0, + .phyclk_mhz = 600.0, + .socclk_mhz = 694.0, + .dscclk_mhz = 214.0, + .dram_speed_mts = 11104.0, + }, + { + .state = 2, + .dcfclk_mhz = 875.0, + .fabricclk_mhz = 875.0, + .dispclk_mhz = 734.0, + .dppclk_mhz = 734.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 875.0, + .dscclk_mhz = 245.0, + .dram_speed_mts = 14000.0, + }, + { + .state = 3, + .dcfclk_mhz = 1000.0, + .fabricclk_mhz = 1000.0, + .dispclk_mhz = 1100.0, + .dppclk_mhz = 1100.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1000.0, + .dscclk_mhz = 367.0, + .dram_speed_mts = 16000.0, + }, + { + .state = 4, + .dcfclk_mhz = 1200.0, + .fabricclk_mhz = 1200.0, + .dispclk_mhz = 1284.0, + .dppclk_mhz = 1284.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1200.0, + .dscclk_mhz = 428.0, + .dram_speed_mts = 16000.0, + }, + /*Extra state, no dispclk ramping*/ + { + .state = 5, + .dcfclk_mhz = 1200.0, + .fabricclk_mhz = 1200.0, + .dispclk_mhz = 1284.0, + .dppclk_mhz = 1284.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1200.0, + .dscclk_mhz = 428.0, + .dram_speed_mts = 16000.0, + }, + }, + .num_states = 5, + .sr_exit_time_us = 8.6, + .sr_enter_plus_exit_time_us = 10.9, + .urgent_latency_us = 4.0, + .urgent_latency_pixel_data_only_us = 4.0, + .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, + .urgent_latency_vm_data_only_us = 4.0, + .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, + .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, + .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 40.0, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 40.0, + .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, + .max_avg_sdp_bw_use_normal_percent = 40.0, + .max_avg_dram_bw_use_normal_percent = 40.0, + .writeback_latency_us = 12.0, + .ideal_dram_bw_after_urgent_percent = 40.0, + .max_request_size_bytes = 256, + .dram_channel_width_bytes = 2, + .fabric_datapath_to_dcn_data_return_bytes = 64, + .dcn_downspread_percent = 0.5, + .downspread_percent = 0.38, + .dram_page_open_time_ns = 50.0, + .dram_rw_turnaround_time_ns = 17.5, + .dram_return_buffer_per_channel_bytes = 8192, + .round_trip_ping_latency_dcfclk_cycles = 131, + .urgent_out_of_order_return_per_channel_bytes = 256, + .channel_interleave_bytes = 256, + .num_banks = 8, + .num_chans = 16, + .vmm_page_size_bytes = 4096, + .dram_clock_change_latency_us = 404.0, + .dummy_pstate_latency_us = 5.0, + .writeback_dram_clock_change_latency_us = 23.0, + .return_bus_width_bytes = 64, + .dispclk_dppclk_vco_speed_mhz = 3850, + .xfc_bus_transport_time_us = 20, + .xfc_xbuf_latency_tolerance_us = 4, + .use_urgent_burst_bw = 0 +}; + +struct _vcs_dpi_soc_bounding_box_st dcn2_0_nv14_soc = { + .clock_limits = { + { + .state = 0, + .dcfclk_mhz = 560.0, + .fabricclk_mhz = 560.0, + .dispclk_mhz = 513.0, + .dppclk_mhz = 513.0, + .phyclk_mhz = 540.0, + .socclk_mhz = 560.0, + .dscclk_mhz = 171.0, + .dram_speed_mts = 8960.0, + }, + { + .state = 1, + .dcfclk_mhz = 694.0, + .fabricclk_mhz = 694.0, + .dispclk_mhz = 642.0, + .dppclk_mhz = 642.0, + .phyclk_mhz = 600.0, + .socclk_mhz = 694.0, + .dscclk_mhz = 214.0, + .dram_speed_mts = 11104.0, + }, + { + .state = 2, + .dcfclk_mhz = 875.0, + .fabricclk_mhz = 875.0, + .dispclk_mhz = 734.0, + .dppclk_mhz = 734.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 875.0, + .dscclk_mhz = 245.0, + .dram_speed_mts = 14000.0, + }, + { + .state = 3, + .dcfclk_mhz = 1000.0, + .fabricclk_mhz = 1000.0, + .dispclk_mhz = 1100.0, + .dppclk_mhz = 1100.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1000.0, + .dscclk_mhz = 367.0, + .dram_speed_mts = 16000.0, + }, + { + .state = 4, + .dcfclk_mhz = 1200.0, + .fabricclk_mhz = 1200.0, + .dispclk_mhz = 1284.0, + .dppclk_mhz = 1284.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1200.0, + .dscclk_mhz = 428.0, + .dram_speed_mts = 16000.0, + }, + /*Extra state, no dispclk ramping*/ + { + .state = 5, + .dcfclk_mhz = 1200.0, + .fabricclk_mhz = 1200.0, + .dispclk_mhz = 1284.0, + .dppclk_mhz = 1284.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1200.0, + .dscclk_mhz = 428.0, + .dram_speed_mts = 16000.0, + }, + }, + .num_states = 5, + .sr_exit_time_us = 11.6, + .sr_enter_plus_exit_time_us = 13.9, + .urgent_latency_us = 4.0, + .urgent_latency_pixel_data_only_us = 4.0, + .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, + .urgent_latency_vm_data_only_us = 4.0, + .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, + .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, + .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 40.0, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 40.0, + .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, + .max_avg_sdp_bw_use_normal_percent = 40.0, + .max_avg_dram_bw_use_normal_percent = 40.0, + .writeback_latency_us = 12.0, + .ideal_dram_bw_after_urgent_percent = 40.0, + .max_request_size_bytes = 256, + .dram_channel_width_bytes = 2, + .fabric_datapath_to_dcn_data_return_bytes = 64, + .dcn_downspread_percent = 0.5, + .downspread_percent = 0.38, + .dram_page_open_time_ns = 50.0, + .dram_rw_turnaround_time_ns = 17.5, + .dram_return_buffer_per_channel_bytes = 8192, + .round_trip_ping_latency_dcfclk_cycles = 131, + .urgent_out_of_order_return_per_channel_bytes = 256, + .channel_interleave_bytes = 256, + .num_banks = 8, + .num_chans = 8, + .vmm_page_size_bytes = 4096, + .dram_clock_change_latency_us = 404.0, + .dummy_pstate_latency_us = 5.0, + .writeback_dram_clock_change_latency_us = 23.0, + .return_bus_width_bytes = 64, + .dispclk_dppclk_vco_speed_mhz = 3850, + .xfc_bus_transport_time_us = 20, + .xfc_xbuf_latency_tolerance_us = 4, + .use_urgent_burst_bw = 0 +}; + +struct _vcs_dpi_soc_bounding_box_st dcn2_0_nv12_soc = { 0 }; + +struct _vcs_dpi_ip_params_st dcn2_1_ip = { + .odm_capable = 1, + .gpuvm_enable = 1, + .hostvm_enable = 1, + .gpuvm_max_page_table_levels = 1, + .hostvm_max_page_table_levels = 4, + .hostvm_cached_page_table_levels = 2, + .num_dsc = 3, + .rob_buffer_size_kbytes = 168, + .det_buffer_size_kbytes = 164, + .dpte_buffer_size_in_pte_reqs_luma = 44, + .dpte_buffer_size_in_pte_reqs_chroma = 42,//todo + .dpp_output_buffer_pixels = 2560, + .opp_output_buffer_lines = 1, + .pixel_chunk_size_kbytes = 8, + .pte_enable = 1, + .max_page_table_levels = 4, + .pte_chunk_size_kbytes = 2, + .meta_chunk_size_kbytes = 2, + .min_meta_chunk_size_bytes = 256, + .writeback_chunk_size_kbytes = 2, + .line_buffer_size_bits = 789504, + .is_line_buffer_bpp_fixed = 0, + .line_buffer_fixed_bpp = 0, + .dcc_supported = true, + .max_line_buffer_lines = 12, + .writeback_luma_buffer_size_kbytes = 12, + .writeback_chroma_buffer_size_kbytes = 8, + .writeback_chroma_line_buffer_width_pixels = 4, + .writeback_max_hscl_ratio = 1, + .writeback_max_vscl_ratio = 1, + .writeback_min_hscl_ratio = 1, + .writeback_min_vscl_ratio = 1, + .writeback_max_hscl_taps = 12, + .writeback_max_vscl_taps = 12, + .writeback_line_buffer_luma_buffer_size = 0, + .writeback_line_buffer_chroma_buffer_size = 14643, + .cursor_buffer_size = 8, + .cursor_chunk_size = 2, + .max_num_otg = 4, + .max_num_dpp = 4, + .max_num_wb = 1, + .max_dchub_pscl_bw_pix_per_clk = 4, + .max_pscl_lb_bw_pix_per_clk = 2, + .max_lb_vscl_bw_pix_per_clk = 4, + .max_vscl_hscl_bw_pix_per_clk = 4, + .max_hscl_ratio = 4, + .max_vscl_ratio = 4, + .hscl_mults = 4, + .vscl_mults = 4, + .max_hscl_taps = 8, + .max_vscl_taps = 8, + .dispclk_ramp_margin_percent = 1, + .underscan_factor = 1.10, + .min_vblank_lines = 32, // + .dppclk_delay_subtotal = 77, // + .dppclk_delay_scl_lb_only = 16, + .dppclk_delay_scl = 50, + .dppclk_delay_cnvc_formatter = 8, + .dppclk_delay_cnvc_cursor = 6, + .dispclk_delay_subtotal = 87, // + .dcfclk_cstate_latency = 10, // SRExitTime + .max_inter_dcn_tile_repeaters = 8, + + .xfc_supported = false, + .xfc_fill_bw_overhead_percent = 10.0, + .xfc_fill_constant_bytes = 0, + .ptoi_supported = 0, + .number_of_cursors = 1, +}; + +struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { + .clock_limits = { + { + .state = 0, + .dcfclk_mhz = 400.0, + .fabricclk_mhz = 400.0, + .dispclk_mhz = 600.0, + .dppclk_mhz = 400.00, + .phyclk_mhz = 600.0, + .socclk_mhz = 278.0, + .dscclk_mhz = 205.67, + .dram_speed_mts = 1600.0, + }, + { + .state = 1, + .dcfclk_mhz = 464.52, + .fabricclk_mhz = 800.0, + .dispclk_mhz = 654.55, + .dppclk_mhz = 626.09, + .phyclk_mhz = 600.0, + .socclk_mhz = 278.0, + .dscclk_mhz = 205.67, + .dram_speed_mts = 1600.0, + }, + { + .state = 2, + .dcfclk_mhz = 514.29, + .fabricclk_mhz = 933.0, + .dispclk_mhz = 757.89, + .dppclk_mhz = 685.71, + .phyclk_mhz = 600.0, + .socclk_mhz = 278.0, + .dscclk_mhz = 287.67, + .dram_speed_mts = 1866.0, + }, + { + .state = 3, + .dcfclk_mhz = 576.00, + .fabricclk_mhz = 1067.0, + .dispclk_mhz = 847.06, + .dppclk_mhz = 757.89, + .phyclk_mhz = 600.0, + .socclk_mhz = 715.0, + .dscclk_mhz = 318.334, + .dram_speed_mts = 2134.0, + }, + { + .state = 4, + .dcfclk_mhz = 626.09, + .fabricclk_mhz = 1200.0, + .dispclk_mhz = 900.00, + .dppclk_mhz = 847.06, + .phyclk_mhz = 810.0, + .socclk_mhz = 953.0, + .dscclk_mhz = 489.0, + .dram_speed_mts = 2400.0, + }, + { + .state = 5, + .dcfclk_mhz = 685.71, + .fabricclk_mhz = 1333.0, + .dispclk_mhz = 1028.57, + .dppclk_mhz = 960.00, + .phyclk_mhz = 810.0, + .socclk_mhz = 278.0, + .dscclk_mhz = 287.67, + .dram_speed_mts = 2666.0, + }, + { + .state = 6, + .dcfclk_mhz = 757.89, + .fabricclk_mhz = 1467.0, + .dispclk_mhz = 1107.69, + .dppclk_mhz = 1028.57, + .phyclk_mhz = 810.0, + .socclk_mhz = 715.0, + .dscclk_mhz = 318.334, + .dram_speed_mts = 3200.0, + }, + { + .state = 7, + .dcfclk_mhz = 847.06, + .fabricclk_mhz = 1600.0, + .dispclk_mhz = 1395.0, + .dppclk_mhz = 1285.00, + .phyclk_mhz = 1325.0, + .socclk_mhz = 953.0, + .dscclk_mhz = 489.0, + .dram_speed_mts = 4266.0, + }, + /*Extra state, no dispclk ramping*/ + { + .state = 8, + .dcfclk_mhz = 847.06, + .fabricclk_mhz = 1600.0, + .dispclk_mhz = 1395.0, + .dppclk_mhz = 1285.0, + .phyclk_mhz = 1325.0, + .socclk_mhz = 953.0, + .dscclk_mhz = 489.0, + .dram_speed_mts = 4266.0, + }, + + }, + + .sr_exit_time_us = 12.5, + .sr_enter_plus_exit_time_us = 17.0, + .urgent_latency_us = 4.0, + .urgent_latency_pixel_data_only_us = 4.0, + .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, + .urgent_latency_vm_data_only_us = 4.0, + .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, + .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, + .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 80.0, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 75.0, + .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, + .max_avg_sdp_bw_use_normal_percent = 60.0, + .max_avg_dram_bw_use_normal_percent = 100.0, + .writeback_latency_us = 12.0, + .max_request_size_bytes = 256, + .dram_channel_width_bytes = 4, + .fabric_datapath_to_dcn_data_return_bytes = 32, + .dcn_downspread_percent = 0.5, + .downspread_percent = 0.38, + .dram_page_open_time_ns = 50.0, + .dram_rw_turnaround_time_ns = 17.5, + .dram_return_buffer_per_channel_bytes = 8192, + .round_trip_ping_latency_dcfclk_cycles = 128, + .urgent_out_of_order_return_per_channel_bytes = 4096, + .channel_interleave_bytes = 256, + .num_banks = 8, + .num_chans = 4, + .vmm_page_size_bytes = 4096, + .dram_clock_change_latency_us = 23.84, + .return_bus_width_bytes = 64, + .dispclk_dppclk_vco_speed_mhz = 3600, + .xfc_bus_transport_time_us = 4, + .xfc_xbuf_latency_tolerance_us = 4, + .use_urgent_burst_bw = 1, + .num_states = 8 +}; + void dcn20_populate_dml_writeback_from_context(struct dc *dc, struct resource_context *res_ctx, display_e2e_pipe_params_st *pipes) @@ -100,3 +689,1340 @@ void dcn20_populate_dml_writeback_from_context(struct dc *dc, pipe_cnt++; } } + +void dcn20_fpu_set_wb_arb_params(struct mcif_arb_params *wb_arb_params, + struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int pipe_cnt, int i) +{ + int k; + + dc_assert_fp_enabled(); + + for (k = 0; k < sizeof(wb_arb_params->cli_watermark)/sizeof(wb_arb_params->cli_watermark[0]); k++) { + wb_arb_params->cli_watermark[k] = get_wm_writeback_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + wb_arb_params->pstate_watermark[k] = get_wm_writeback_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + } + wb_arb_params->time_per_pixel = 16.0 * 1000 / (context->res_ctx.pipe_ctx[i].stream->phy_pix_clk / 1000); /* 4 bit fraction, ms */ +} + +static bool is_dtbclk_required(struct dc *dc, struct dc_state *context) +{ + int i; + for (i = 0; i < dc->res_pool->pipe_count; i++) { + if (!context->res_ctx.pipe_ctx[i].stream) + continue; + if (is_dp_128b_132b_signal(&context->res_ctx.pipe_ctx[i])) + return true; + } + return false; +} + +static enum dcn_zstate_support_state decide_zstate_support(struct dc *dc, struct dc_state *context) +{ + int plane_count; + int i; + + plane_count = 0; + for (i = 0; i < dc->res_pool->pipe_count; i++) { + if (context->res_ctx.pipe_ctx[i].plane_state) + plane_count++; + } + + /* + * Z9 and Z10 allowed cases: + * 1. 0 Planes enabled + * 2. single eDP, on link 0, 1 plane and stutter period > 5ms + * Z10 only cases: + * 1. single eDP, on link 0, 1 plane and stutter period >= 5ms + * Zstate not allowed cases: + * 1. Everything else + */ + if (plane_count == 0) + return DCN_ZSTATE_SUPPORT_ALLOW; + else if (context->stream_count == 1 && context->streams[0]->signal == SIGNAL_TYPE_EDP) { + struct dc_link *link = context->streams[0]->sink->link; + struct dc_stream_status *stream_status = &context->stream_status[0]; + + /* zstate only supported on PWRSEQ0 and when there's <2 planes*/ + if (link->link_index != 0 || stream_status->plane_count > 1) + return DCN_ZSTATE_SUPPORT_DISALLOW; + + if (context->bw_ctx.dml.vba.StutterPeriod > 5000.0) + return DCN_ZSTATE_SUPPORT_ALLOW; + else if (link->psr_settings.psr_version == DC_PSR_VERSION_1 && !dc->debug.disable_psr) + return DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY; + else + return DCN_ZSTATE_SUPPORT_DISALLOW; + } else + return DCN_ZSTATE_SUPPORT_DISALLOW; +} + +void dcn20_calculate_dlg_params( + struct dc *dc, struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int pipe_cnt, + int vlevel) +{ + int i, pipe_idx; + + dc_assert_fp_enabled(); + + /* Writeback MCIF_WB arbitration parameters */ + dc->res_pool->funcs->set_mcif_arb_params(dc, context, pipes, pipe_cnt); + + context->bw_ctx.bw.dcn.clk.dispclk_khz = context->bw_ctx.dml.vba.DISPCLK * 1000; + context->bw_ctx.bw.dcn.clk.dcfclk_khz = context->bw_ctx.dml.vba.DCFCLK * 1000; + context->bw_ctx.bw.dcn.clk.socclk_khz = context->bw_ctx.dml.vba.SOCCLK * 1000; + context->bw_ctx.bw.dcn.clk.dramclk_khz = context->bw_ctx.dml.vba.DRAMSpeed * 1000 / 16; + + if (dc->debug.min_dram_clk_khz > context->bw_ctx.bw.dcn.clk.dramclk_khz) + context->bw_ctx.bw.dcn.clk.dramclk_khz = dc->debug.min_dram_clk_khz; + + context->bw_ctx.bw.dcn.clk.dcfclk_deep_sleep_khz = context->bw_ctx.dml.vba.DCFCLKDeepSleep * 1000; + context->bw_ctx.bw.dcn.clk.fclk_khz = context->bw_ctx.dml.vba.FabricClock * 1000; + context->bw_ctx.bw.dcn.clk.p_state_change_support = + context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] + != dm_dram_clock_change_unsupported; + context->bw_ctx.bw.dcn.clk.dppclk_khz = 0; + + context->bw_ctx.bw.dcn.clk.zstate_support = decide_zstate_support(dc, context); + + context->bw_ctx.bw.dcn.clk.dtbclk_en = is_dtbclk_required(dc, context); + + if (context->bw_ctx.bw.dcn.clk.dispclk_khz < dc->debug.min_disp_clk_khz) + context->bw_ctx.bw.dcn.clk.dispclk_khz = dc->debug.min_disp_clk_khz; + + for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { + if (!context->res_ctx.pipe_ctx[i].stream) + continue; + pipes[pipe_idx].pipe.dest.vstartup_start = get_vstartup(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); + pipes[pipe_idx].pipe.dest.vupdate_offset = get_vupdate_offset(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); + pipes[pipe_idx].pipe.dest.vupdate_width = get_vupdate_width(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); + pipes[pipe_idx].pipe.dest.vready_offset = get_vready_offset(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); + context->res_ctx.pipe_ctx[i].det_buffer_size_kb = context->bw_ctx.dml.ip.det_buffer_size_kbytes; + context->res_ctx.pipe_ctx[i].unbounded_req = pipes[pipe_idx].pipe.src.unbounded_req_mode; + + if (context->bw_ctx.bw.dcn.clk.dppclk_khz < pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000) + context->bw_ctx.bw.dcn.clk.dppclk_khz = pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000; + context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz = + pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000; + context->res_ctx.pipe_ctx[i].pipe_dlg_param = pipes[pipe_idx].pipe.dest; + pipe_idx++; + } + /*save a original dppclock copy*/ + context->bw_ctx.bw.dcn.clk.bw_dppclk_khz = context->bw_ctx.bw.dcn.clk.dppclk_khz; + context->bw_ctx.bw.dcn.clk.bw_dispclk_khz = context->bw_ctx.bw.dcn.clk.dispclk_khz; + context->bw_ctx.bw.dcn.clk.max_supported_dppclk_khz = context->bw_ctx.dml.soc.clock_limits[vlevel].dppclk_mhz * 1000; + context->bw_ctx.bw.dcn.clk.max_supported_dispclk_khz = context->bw_ctx.dml.soc.clock_limits[vlevel].dispclk_mhz * 1000; + + context->bw_ctx.bw.dcn.compbuf_size_kb = context->bw_ctx.dml.ip.config_return_buffer_size_in_kbytes + - context->bw_ctx.dml.ip.det_buffer_size_kbytes * pipe_idx; + + for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { + bool cstate_en = context->bw_ctx.dml.vba.PrefetchMode[vlevel][context->bw_ctx.dml.vba.maxMpcComb] != 2; + + if (!context->res_ctx.pipe_ctx[i].stream) + continue; + + if (dc->ctx->dce_version == DCN_VERSION_2_01) + cstate_en = false; + + context->bw_ctx.dml.funcs.rq_dlg_get_dlg_reg(&context->bw_ctx.dml, + &context->res_ctx.pipe_ctx[i].dlg_regs, + &context->res_ctx.pipe_ctx[i].ttu_regs, + pipes, + pipe_cnt, + pipe_idx, + cstate_en, + context->bw_ctx.bw.dcn.clk.p_state_change_support, + false, false, true); + + context->bw_ctx.dml.funcs.rq_dlg_get_rq_reg(&context->bw_ctx.dml, + &context->res_ctx.pipe_ctx[i].rq_regs, + &pipes[pipe_idx].pipe); + pipe_idx++; + } +} + +static void swizzle_to_dml_params( + enum swizzle_mode_values swizzle, + unsigned int *sw_mode) +{ + switch (swizzle) { + case DC_SW_LINEAR: + *sw_mode = dm_sw_linear; + break; + case DC_SW_4KB_S: + *sw_mode = dm_sw_4kb_s; + break; + case DC_SW_4KB_S_X: + *sw_mode = dm_sw_4kb_s_x; + break; + case DC_SW_4KB_D: + *sw_mode = dm_sw_4kb_d; + break; + case DC_SW_4KB_D_X: + *sw_mode = dm_sw_4kb_d_x; + break; + case DC_SW_64KB_S: + *sw_mode = dm_sw_64kb_s; + break; + case DC_SW_64KB_S_X: + *sw_mode = dm_sw_64kb_s_x; + break; + case DC_SW_64KB_S_T: + *sw_mode = dm_sw_64kb_s_t; + break; + case DC_SW_64KB_D: + *sw_mode = dm_sw_64kb_d; + break; + case DC_SW_64KB_D_X: + *sw_mode = dm_sw_64kb_d_x; + break; + case DC_SW_64KB_D_T: + *sw_mode = dm_sw_64kb_d_t; + break; + case DC_SW_64KB_R_X: + *sw_mode = dm_sw_64kb_r_x; + break; + case DC_SW_VAR_S: + *sw_mode = dm_sw_var_s; + break; + case DC_SW_VAR_S_X: + *sw_mode = dm_sw_var_s_x; + break; + case DC_SW_VAR_D: + *sw_mode = dm_sw_var_d; + break; + case DC_SW_VAR_D_X: + *sw_mode = dm_sw_var_d_x; + break; + case DC_SW_VAR_R_X: + *sw_mode = dm_sw_var_r_x; + break; + default: + ASSERT(0); /* Not supported */ + break; + } +} + +int dcn20_populate_dml_pipes_from_context( + struct dc *dc, + struct dc_state *context, + display_e2e_pipe_params_st *pipes, + bool fast_validate) +{ + int pipe_cnt, i; + bool synchronized_vblank = true; + struct resource_context *res_ctx = &context->res_ctx; + + dc_assert_fp_enabled(); + + for (i = 0, pipe_cnt = -1; i < dc->res_pool->pipe_count; i++) { + if (!res_ctx->pipe_ctx[i].stream) + continue; + + if (pipe_cnt < 0) { + pipe_cnt = i; + continue; + } + + if (res_ctx->pipe_ctx[pipe_cnt].stream == res_ctx->pipe_ctx[i].stream) + continue; + + if (dc->debug.disable_timing_sync || + (!resource_are_streams_timing_synchronizable( + res_ctx->pipe_ctx[pipe_cnt].stream, + res_ctx->pipe_ctx[i].stream) && + !resource_are_vblanks_synchronizable( + res_ctx->pipe_ctx[pipe_cnt].stream, + res_ctx->pipe_ctx[i].stream))) { + synchronized_vblank = false; + break; + } + } + + for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { + struct dc_crtc_timing *timing = &res_ctx->pipe_ctx[i].stream->timing; + unsigned int v_total; + unsigned int front_porch; + int output_bpc; + struct audio_check aud_check = {0}; + + if (!res_ctx->pipe_ctx[i].stream) + continue; + + v_total = timing->v_total; + front_porch = timing->v_front_porch; + + /* todo: + pipes[pipe_cnt].pipe.src.dynamic_metadata_enable = 0; + pipes[pipe_cnt].pipe.src.dcc = 0; + pipes[pipe_cnt].pipe.src.vm = 0;*/ + + pipes[pipe_cnt].clks_cfg.refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000.0; + + pipes[pipe_cnt].dout.dsc_enable = res_ctx->pipe_ctx[i].stream->timing.flags.DSC; + /* todo: rotation?*/ + pipes[pipe_cnt].dout.dsc_slices = res_ctx->pipe_ctx[i].stream->timing.dsc_cfg.num_slices_h; + if (res_ctx->pipe_ctx[i].stream->use_dynamic_meta) { + pipes[pipe_cnt].pipe.src.dynamic_metadata_enable = true; + /* 1/2 vblank */ + pipes[pipe_cnt].pipe.src.dynamic_metadata_lines_before_active = + (v_total - timing->v_addressable + - timing->v_border_top - timing->v_border_bottom) / 2; + /* 36 bytes dp, 32 hdmi */ + pipes[pipe_cnt].pipe.src.dynamic_metadata_xmit_bytes = + dc_is_dp_signal(res_ctx->pipe_ctx[i].stream->signal) ? 36 : 32; + } + pipes[pipe_cnt].pipe.src.dcc = false; + pipes[pipe_cnt].pipe.src.dcc_rate = 1; + pipes[pipe_cnt].pipe.dest.synchronized_vblank_all_planes = synchronized_vblank; + pipes[pipe_cnt].pipe.dest.hblank_start = timing->h_total - timing->h_front_porch; + pipes[pipe_cnt].pipe.dest.hblank_end = pipes[pipe_cnt].pipe.dest.hblank_start + - timing->h_addressable + - timing->h_border_left + - timing->h_border_right; + pipes[pipe_cnt].pipe.dest.vblank_start = v_total - front_porch; + pipes[pipe_cnt].pipe.dest.vblank_end = pipes[pipe_cnt].pipe.dest.vblank_start + - timing->v_addressable + - timing->v_border_top + - timing->v_border_bottom; + pipes[pipe_cnt].pipe.dest.htotal = timing->h_total; + pipes[pipe_cnt].pipe.dest.vtotal = v_total; + pipes[pipe_cnt].pipe.dest.hactive = + timing->h_addressable + timing->h_border_left + timing->h_border_right; + pipes[pipe_cnt].pipe.dest.vactive = + timing->v_addressable + timing->v_border_top + timing->v_border_bottom; + pipes[pipe_cnt].pipe.dest.interlaced = timing->flags.INTERLACE; + pipes[pipe_cnt].pipe.dest.pixel_rate_mhz = timing->pix_clk_100hz/10000.0; + if (timing->timing_3d_format == TIMING_3D_FORMAT_HW_FRAME_PACKING) + pipes[pipe_cnt].pipe.dest.pixel_rate_mhz *= 2; + pipes[pipe_cnt].pipe.dest.otg_inst = res_ctx->pipe_ctx[i].stream_res.tg->inst; + pipes[pipe_cnt].dout.dp_lanes = 4; + pipes[pipe_cnt].dout.is_virtual = 0; + pipes[pipe_cnt].pipe.dest.vtotal_min = res_ctx->pipe_ctx[i].stream->adjust.v_total_min; + pipes[pipe_cnt].pipe.dest.vtotal_max = res_ctx->pipe_ctx[i].stream->adjust.v_total_max; + switch (get_num_odm_splits(&res_ctx->pipe_ctx[i])) { + case 1: + pipes[pipe_cnt].pipe.dest.odm_combine = dm_odm_combine_mode_2to1; + break; + case 3: + pipes[pipe_cnt].pipe.dest.odm_combine = dm_odm_combine_mode_4to1; + break; + default: + pipes[pipe_cnt].pipe.dest.odm_combine = dm_odm_combine_mode_disabled; + } + pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].pipe_idx; + if (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state + == res_ctx->pipe_ctx[i].plane_state) { + struct pipe_ctx *first_pipe = res_ctx->pipe_ctx[i].top_pipe; + int split_idx = 0; + + while (first_pipe->top_pipe && first_pipe->top_pipe->plane_state + == res_ctx->pipe_ctx[i].plane_state) { + first_pipe = first_pipe->top_pipe; + split_idx++; + } + /* Treat 4to1 mpc combine as an mpo of 2 2-to-1 combines */ + if (split_idx == 0) + pipes[pipe_cnt].pipe.src.hsplit_grp = first_pipe->pipe_idx; + else if (split_idx == 1) + pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].pipe_idx; + else if (split_idx == 2) + pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].top_pipe->pipe_idx; + } else if (res_ctx->pipe_ctx[i].prev_odm_pipe) { + struct pipe_ctx *first_pipe = res_ctx->pipe_ctx[i].prev_odm_pipe; + + while (first_pipe->prev_odm_pipe) + first_pipe = first_pipe->prev_odm_pipe; + pipes[pipe_cnt].pipe.src.hsplit_grp = first_pipe->pipe_idx; + } + + switch (res_ctx->pipe_ctx[i].stream->signal) { + case SIGNAL_TYPE_DISPLAY_PORT_MST: + case SIGNAL_TYPE_DISPLAY_PORT: + pipes[pipe_cnt].dout.output_type = dm_dp; + break; + case SIGNAL_TYPE_EDP: + pipes[pipe_cnt].dout.output_type = dm_edp; + break; + case SIGNAL_TYPE_HDMI_TYPE_A: + case SIGNAL_TYPE_DVI_SINGLE_LINK: + case SIGNAL_TYPE_DVI_DUAL_LINK: + pipes[pipe_cnt].dout.output_type = dm_hdmi; + break; + default: + /* In case there is no signal, set dp with 4 lanes to allow max config */ + pipes[pipe_cnt].dout.is_virtual = 1; + pipes[pipe_cnt].dout.output_type = dm_dp; + pipes[pipe_cnt].dout.dp_lanes = 4; + } + + switch (res_ctx->pipe_ctx[i].stream->timing.display_color_depth) { + case COLOR_DEPTH_666: + output_bpc = 6; + break; + case COLOR_DEPTH_888: + output_bpc = 8; + break; + case COLOR_DEPTH_101010: + output_bpc = 10; + break; + case COLOR_DEPTH_121212: + output_bpc = 12; + break; + case COLOR_DEPTH_141414: + output_bpc = 14; + break; + case COLOR_DEPTH_161616: + output_bpc = 16; + break; + case COLOR_DEPTH_999: + output_bpc = 9; + break; + case COLOR_DEPTH_111111: + output_bpc = 11; + break; + default: + output_bpc = 8; + break; + } + + switch (res_ctx->pipe_ctx[i].stream->timing.pixel_encoding) { + case PIXEL_ENCODING_RGB: + case PIXEL_ENCODING_YCBCR444: + pipes[pipe_cnt].dout.output_format = dm_444; + pipes[pipe_cnt].dout.output_bpp = output_bpc * 3; + break; + case PIXEL_ENCODING_YCBCR420: + pipes[pipe_cnt].dout.output_format = dm_420; + pipes[pipe_cnt].dout.output_bpp = (output_bpc * 3.0) / 2; + break; + case PIXEL_ENCODING_YCBCR422: + if (res_ctx->pipe_ctx[i].stream->timing.flags.DSC && + !res_ctx->pipe_ctx[i].stream->timing.dsc_cfg.ycbcr422_simple) + pipes[pipe_cnt].dout.output_format = dm_n422; + else + pipes[pipe_cnt].dout.output_format = dm_s422; + pipes[pipe_cnt].dout.output_bpp = output_bpc * 2; + break; + default: + pipes[pipe_cnt].dout.output_format = dm_444; + pipes[pipe_cnt].dout.output_bpp = output_bpc * 3; + } + + if (res_ctx->pipe_ctx[i].stream->timing.flags.DSC) + pipes[pipe_cnt].dout.output_bpp = res_ctx->pipe_ctx[i].stream->timing.dsc_cfg.bits_per_pixel / 16.0; + + /* todo: default max for now, until there is logic reflecting this in dc*/ + pipes[pipe_cnt].dout.dsc_input_bpc = 12; + /*fill up the audio sample rate (unit in kHz)*/ + get_audio_check(&res_ctx->pipe_ctx[i].stream->audio_info, &aud_check); + pipes[pipe_cnt].dout.max_audio_sample_rate = aud_check.max_audiosample_rate / 1000; + /* + * For graphic plane, cursor number is 1, nv12 is 0 + * bw calculations due to cursor on/off + */ + if (res_ctx->pipe_ctx[i].plane_state && + res_ctx->pipe_ctx[i].plane_state->address.type == PLN_ADDR_TYPE_VIDEO_PROGRESSIVE) + pipes[pipe_cnt].pipe.src.num_cursors = 0; + else + pipes[pipe_cnt].pipe.src.num_cursors = dc->dml.ip.number_of_cursors; + + pipes[pipe_cnt].pipe.src.cur0_src_width = 256; + pipes[pipe_cnt].pipe.src.cur0_bpp = dm_cur_32bit; + + if (!res_ctx->pipe_ctx[i].plane_state) { + pipes[pipe_cnt].pipe.src.is_hsplit = pipes[pipe_cnt].pipe.dest.odm_combine != dm_odm_combine_mode_disabled; + pipes[pipe_cnt].pipe.src.source_scan = dm_horz; + pipes[pipe_cnt].pipe.src.sw_mode = dm_sw_4kb_s; + pipes[pipe_cnt].pipe.src.macro_tile_size = dm_64k_tile; + pipes[pipe_cnt].pipe.src.viewport_width = timing->h_addressable; + if (pipes[pipe_cnt].pipe.src.viewport_width > 1920) + pipes[pipe_cnt].pipe.src.viewport_width = 1920; + pipes[pipe_cnt].pipe.src.viewport_height = timing->v_addressable; + if (pipes[pipe_cnt].pipe.src.viewport_height > 1080) + pipes[pipe_cnt].pipe.src.viewport_height = 1080; + pipes[pipe_cnt].pipe.src.surface_height_y = pipes[pipe_cnt].pipe.src.viewport_height; + pipes[pipe_cnt].pipe.src.surface_width_y = pipes[pipe_cnt].pipe.src.viewport_width; + pipes[pipe_cnt].pipe.src.surface_height_c = pipes[pipe_cnt].pipe.src.viewport_height; + pipes[pipe_cnt].pipe.src.surface_width_c = pipes[pipe_cnt].pipe.src.viewport_width; + pipes[pipe_cnt].pipe.src.data_pitch = ((pipes[pipe_cnt].pipe.src.viewport_width + 255) / 256) * 256; + pipes[pipe_cnt].pipe.src.source_format = dm_444_32; + pipes[pipe_cnt].pipe.dest.recout_width = pipes[pipe_cnt].pipe.src.viewport_width; /*vp_width/hratio*/ + pipes[pipe_cnt].pipe.dest.recout_height = pipes[pipe_cnt].pipe.src.viewport_height; /*vp_height/vratio*/ + pipes[pipe_cnt].pipe.dest.full_recout_width = pipes[pipe_cnt].pipe.dest.recout_width; /*when is_hsplit != 1*/ + pipes[pipe_cnt].pipe.dest.full_recout_height = pipes[pipe_cnt].pipe.dest.recout_height; /*when is_hsplit != 1*/ + pipes[pipe_cnt].pipe.scale_ratio_depth.lb_depth = dm_lb_16; + pipes[pipe_cnt].pipe.scale_ratio_depth.hscl_ratio = 1.0; + pipes[pipe_cnt].pipe.scale_ratio_depth.vscl_ratio = 1.0; + pipes[pipe_cnt].pipe.scale_ratio_depth.scl_enable = 0; /*Lb only or Full scl*/ + pipes[pipe_cnt].pipe.scale_taps.htaps = 1; + pipes[pipe_cnt].pipe.scale_taps.vtaps = 1; + pipes[pipe_cnt].pipe.dest.vtotal_min = v_total; + pipes[pipe_cnt].pipe.dest.vtotal_max = v_total; + + if (pipes[pipe_cnt].pipe.dest.odm_combine == dm_odm_combine_mode_2to1) { + pipes[pipe_cnt].pipe.src.viewport_width /= 2; + pipes[pipe_cnt].pipe.dest.recout_width /= 2; + } else if (pipes[pipe_cnt].pipe.dest.odm_combine == dm_odm_combine_mode_4to1) { + pipes[pipe_cnt].pipe.src.viewport_width /= 4; + pipes[pipe_cnt].pipe.dest.recout_width /= 4; + } + } else { + struct dc_plane_state *pln = res_ctx->pipe_ctx[i].plane_state; + struct scaler_data *scl = &res_ctx->pipe_ctx[i].plane_res.scl_data; + + pipes[pipe_cnt].pipe.src.immediate_flip = pln->flip_immediate; + pipes[pipe_cnt].pipe.src.is_hsplit = (res_ctx->pipe_ctx[i].bottom_pipe && res_ctx->pipe_ctx[i].bottom_pipe->plane_state == pln) + || (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state == pln) + || pipes[pipe_cnt].pipe.dest.odm_combine != dm_odm_combine_mode_disabled; + + /* stereo is not split */ + if (pln->stereo_format == PLANE_STEREO_FORMAT_SIDE_BY_SIDE || + pln->stereo_format == PLANE_STEREO_FORMAT_TOP_AND_BOTTOM) { + pipes[pipe_cnt].pipe.src.is_hsplit = false; + pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].pipe_idx; + } + + pipes[pipe_cnt].pipe.src.source_scan = pln->rotation == ROTATION_ANGLE_90 + || pln->rotation == ROTATION_ANGLE_270 ? dm_vert : dm_horz; + pipes[pipe_cnt].pipe.src.viewport_y_y = scl->viewport.y; + pipes[pipe_cnt].pipe.src.viewport_y_c = scl->viewport_c.y; + pipes[pipe_cnt].pipe.src.viewport_width = scl->viewport.width; + pipes[pipe_cnt].pipe.src.viewport_width_c = scl->viewport_c.width; + pipes[pipe_cnt].pipe.src.viewport_height = scl->viewport.height; + pipes[pipe_cnt].pipe.src.viewport_height_c = scl->viewport_c.height; + pipes[pipe_cnt].pipe.src.viewport_width_max = pln->src_rect.width; + pipes[pipe_cnt].pipe.src.viewport_height_max = pln->src_rect.height; + pipes[pipe_cnt].pipe.src.surface_width_y = pln->plane_size.surface_size.width; + pipes[pipe_cnt].pipe.src.surface_height_y = pln->plane_size.surface_size.height; + pipes[pipe_cnt].pipe.src.surface_width_c = pln->plane_size.chroma_size.width; + pipes[pipe_cnt].pipe.src.surface_height_c = pln->plane_size.chroma_size.height; + if (pln->format == SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA + || pln->format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN) { + pipes[pipe_cnt].pipe.src.data_pitch = pln->plane_size.surface_pitch; + pipes[pipe_cnt].pipe.src.data_pitch_c = pln->plane_size.chroma_pitch; + pipes[pipe_cnt].pipe.src.meta_pitch = pln->dcc.meta_pitch; + pipes[pipe_cnt].pipe.src.meta_pitch_c = pln->dcc.meta_pitch_c; + } else { + pipes[pipe_cnt].pipe.src.data_pitch = pln->plane_size.surface_pitch; + pipes[pipe_cnt].pipe.src.meta_pitch = pln->dcc.meta_pitch; + } + pipes[pipe_cnt].pipe.src.dcc = pln->dcc.enable; + pipes[pipe_cnt].pipe.dest.recout_width = scl->recout.width; + pipes[pipe_cnt].pipe.dest.recout_height = scl->recout.height; + pipes[pipe_cnt].pipe.dest.full_recout_height = scl->recout.height; + pipes[pipe_cnt].pipe.dest.full_recout_width = scl->recout.width; + if (pipes[pipe_cnt].pipe.dest.odm_combine == dm_odm_combine_mode_2to1) + pipes[pipe_cnt].pipe.dest.full_recout_width *= 2; + else if (pipes[pipe_cnt].pipe.dest.odm_combine == dm_odm_combine_mode_4to1) + pipes[pipe_cnt].pipe.dest.full_recout_width *= 4; + else { + struct pipe_ctx *split_pipe = res_ctx->pipe_ctx[i].bottom_pipe; + + while (split_pipe && split_pipe->plane_state == pln) { + pipes[pipe_cnt].pipe.dest.full_recout_width += split_pipe->plane_res.scl_data.recout.width; + split_pipe = split_pipe->bottom_pipe; + } + split_pipe = res_ctx->pipe_ctx[i].top_pipe; + while (split_pipe && split_pipe->plane_state == pln) { + pipes[pipe_cnt].pipe.dest.full_recout_width += split_pipe->plane_res.scl_data.recout.width; + split_pipe = split_pipe->top_pipe; + } + } + + pipes[pipe_cnt].pipe.scale_ratio_depth.lb_depth = dm_lb_16; + pipes[pipe_cnt].pipe.scale_ratio_depth.hscl_ratio = (double) scl->ratios.horz.value / (1ULL<<32); + pipes[pipe_cnt].pipe.scale_ratio_depth.hscl_ratio_c = (double) scl->ratios.horz_c.value / (1ULL<<32); + pipes[pipe_cnt].pipe.scale_ratio_depth.vscl_ratio = (double) scl->ratios.vert.value / (1ULL<<32); + pipes[pipe_cnt].pipe.scale_ratio_depth.vscl_ratio_c = (double) scl->ratios.vert_c.value / (1ULL<<32); + pipes[pipe_cnt].pipe.scale_ratio_depth.scl_enable = + scl->ratios.vert.value != dc_fixpt_one.value + || scl->ratios.horz.value != dc_fixpt_one.value + || scl->ratios.vert_c.value != dc_fixpt_one.value + || scl->ratios.horz_c.value != dc_fixpt_one.value /*Lb only or Full scl*/ + || dc->debug.always_scale; /*support always scale*/ + pipes[pipe_cnt].pipe.scale_taps.htaps = scl->taps.h_taps; + pipes[pipe_cnt].pipe.scale_taps.htaps_c = scl->taps.h_taps_c; + pipes[pipe_cnt].pipe.scale_taps.vtaps = scl->taps.v_taps; + pipes[pipe_cnt].pipe.scale_taps.vtaps_c = scl->taps.v_taps_c; + + pipes[pipe_cnt].pipe.src.macro_tile_size = + swizzle_mode_to_macro_tile_size(pln->tiling_info.gfx9.swizzle); + swizzle_to_dml_params(pln->tiling_info.gfx9.swizzle, + &pipes[pipe_cnt].pipe.src.sw_mode); + + switch (pln->format) { + case SURFACE_PIXEL_FORMAT_VIDEO_420_YCbCr: + case SURFACE_PIXEL_FORMAT_VIDEO_420_YCrCb: + pipes[pipe_cnt].pipe.src.source_format = dm_420_8; + break; + case SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCbCr: + case SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCrCb: + pipes[pipe_cnt].pipe.src.source_format = dm_420_10; + break; + case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616: + case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616: + case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616F: + case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F: + pipes[pipe_cnt].pipe.src.source_format = dm_444_64; + break; + case SURFACE_PIXEL_FORMAT_GRPH_ARGB1555: + case SURFACE_PIXEL_FORMAT_GRPH_RGB565: + pipes[pipe_cnt].pipe.src.source_format = dm_444_16; + break; + case SURFACE_PIXEL_FORMAT_GRPH_PALETA_256_COLORS: + pipes[pipe_cnt].pipe.src.source_format = dm_444_8; + break; + case SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA: + pipes[pipe_cnt].pipe.src.source_format = dm_rgbe_alpha; + break; + default: + pipes[pipe_cnt].pipe.src.source_format = dm_444_32; + break; + } + } + + pipe_cnt++; + } + + /* populate writeback information */ + DC_FP_START(); + dc->res_pool->funcs->populate_dml_writeback_from_context(dc, res_ctx, pipes); + DC_FP_END(); + + return pipe_cnt; +} + +void dcn20_calculate_wm( + struct dc *dc, struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int *out_pipe_cnt, + int *pipe_split_from, + int vlevel, + bool fast_validate) +{ + int pipe_cnt, i, pipe_idx; + + dc_assert_fp_enabled(); + + for (i = 0, pipe_idx = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { + if (!context->res_ctx.pipe_ctx[i].stream) + continue; + + pipes[pipe_cnt].clks_cfg.refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000.0; + pipes[pipe_cnt].clks_cfg.dispclk_mhz = context->bw_ctx.dml.vba.RequiredDISPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb]; + + if (pipe_split_from[i] < 0) { + pipes[pipe_cnt].clks_cfg.dppclk_mhz = + context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx]; + if (context->bw_ctx.dml.vba.BlendingAndTiming[pipe_idx] == pipe_idx) + pipes[pipe_cnt].pipe.dest.odm_combine = + context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx]; + else + pipes[pipe_cnt].pipe.dest.odm_combine = 0; + pipe_idx++; + } else { + pipes[pipe_cnt].clks_cfg.dppclk_mhz = + context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_split_from[i]]; + if (context->bw_ctx.dml.vba.BlendingAndTiming[pipe_split_from[i]] == pipe_split_from[i]) + pipes[pipe_cnt].pipe.dest.odm_combine = + context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_split_from[i]]; + else + pipes[pipe_cnt].pipe.dest.odm_combine = 0; + } + + if (dc->config.forced_clocks) { + pipes[pipe_cnt].clks_cfg.dispclk_mhz = context->bw_ctx.dml.soc.clock_limits[0].dispclk_mhz; + pipes[pipe_cnt].clks_cfg.dppclk_mhz = context->bw_ctx.dml.soc.clock_limits[0].dppclk_mhz; + } + if (dc->debug.min_disp_clk_khz > pipes[pipe_cnt].clks_cfg.dispclk_mhz * 1000) + pipes[pipe_cnt].clks_cfg.dispclk_mhz = dc->debug.min_disp_clk_khz / 1000.0; + if (dc->debug.min_dpp_clk_khz > pipes[pipe_cnt].clks_cfg.dppclk_mhz * 1000) + pipes[pipe_cnt].clks_cfg.dppclk_mhz = dc->debug.min_dpp_clk_khz / 1000.0; + + pipe_cnt++; + } + + if (pipe_cnt != pipe_idx) { + if (dc->res_pool->funcs->populate_dml_pipes) + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, + context, pipes, fast_validate); + else + pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, + context, pipes, fast_validate); + } + + *out_pipe_cnt = pipe_cnt; + + pipes[0].clks_cfg.voltage = vlevel; + pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[vlevel].dcfclk_mhz; + pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[vlevel].socclk_mhz; + + /* only pipe 0 is read for voltage and dcf/soc clocks */ + if (vlevel < 1) { + pipes[0].clks_cfg.voltage = 1; + pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[1].dcfclk_mhz; + pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[1].socclk_mhz; + } + context->bw_ctx.bw.dcn.watermarks.b.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + + if (vlevel < 2) { + pipes[0].clks_cfg.voltage = 2; + pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[2].dcfclk_mhz; + pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[2].socclk_mhz; + } + context->bw_ctx.bw.dcn.watermarks.c.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + + if (vlevel < 3) { + pipes[0].clks_cfg.voltage = 3; + pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[2].dcfclk_mhz; + pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[2].socclk_mhz; + } + context->bw_ctx.bw.dcn.watermarks.d.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + + pipes[0].clks_cfg.voltage = vlevel; + pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[vlevel].dcfclk_mhz; + pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[vlevel].socclk_mhz; + context->bw_ctx.bw.dcn.watermarks.a.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.a.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; +} + +void dcn20_update_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb, + struct pp_smu_nv_clock_table *max_clocks, unsigned int *uclk_states, unsigned int num_states) +{ + struct _vcs_dpi_voltage_scaling_st calculated_states[DC__VOLTAGE_STATES]; + int i; + int num_calculated_states = 0; + int min_dcfclk = 0; + + dc_assert_fp_enabled(); + + if (num_states == 0) + return; + + memset(calculated_states, 0, sizeof(calculated_states)); + + if (dc->bb_overrides.min_dcfclk_mhz > 0) + min_dcfclk = dc->bb_overrides.min_dcfclk_mhz; + else { + if (ASICREV_IS_NAVI12_P(dc->ctx->asic_id.hw_internal_rev)) + min_dcfclk = 310; + else + // Accounting for SOC/DCF relationship, we can go as high as + // 506Mhz in Vmin. + min_dcfclk = 506; + } + + for (i = 0; i < num_states; i++) { + int min_fclk_required_by_uclk; + calculated_states[i].state = i; + calculated_states[i].dram_speed_mts = uclk_states[i] * 16 / 1000; + + // FCLK:UCLK ratio is 1.08 + min_fclk_required_by_uclk = div_u64(((unsigned long long)uclk_states[i]) * 1080, + 1000000); + + calculated_states[i].fabricclk_mhz = (min_fclk_required_by_uclk < min_dcfclk) ? + min_dcfclk : min_fclk_required_by_uclk; + + calculated_states[i].socclk_mhz = (calculated_states[i].fabricclk_mhz > max_clocks->socClockInKhz / 1000) ? + max_clocks->socClockInKhz / 1000 : calculated_states[i].fabricclk_mhz; + + calculated_states[i].dcfclk_mhz = (calculated_states[i].fabricclk_mhz > max_clocks->dcfClockInKhz / 1000) ? + max_clocks->dcfClockInKhz / 1000 : calculated_states[i].fabricclk_mhz; + + calculated_states[i].dispclk_mhz = max_clocks->displayClockInKhz / 1000; + calculated_states[i].dppclk_mhz = max_clocks->displayClockInKhz / 1000; + calculated_states[i].dscclk_mhz = max_clocks->displayClockInKhz / (1000 * 3); + + calculated_states[i].phyclk_mhz = max_clocks->phyClockInKhz / 1000; + + num_calculated_states++; + } + + calculated_states[num_calculated_states - 1].socclk_mhz = max_clocks->socClockInKhz / 1000; + calculated_states[num_calculated_states - 1].fabricclk_mhz = max_clocks->socClockInKhz / 1000; + calculated_states[num_calculated_states - 1].dcfclk_mhz = max_clocks->dcfClockInKhz / 1000; + + memcpy(bb->clock_limits, calculated_states, sizeof(bb->clock_limits)); + bb->num_states = num_calculated_states; + + // Duplicate the last state, DML always an extra state identical to max state to work + memcpy(&bb->clock_limits[num_calculated_states], &bb->clock_limits[num_calculated_states - 1], sizeof(struct _vcs_dpi_voltage_scaling_st)); + bb->clock_limits[num_calculated_states].state = bb->num_states; +} + +void dcn20_cap_soc_clocks( + struct _vcs_dpi_soc_bounding_box_st *bb, + struct pp_smu_nv_clock_table max_clocks) +{ + int i; + + dc_assert_fp_enabled(); + + // First pass - cap all clocks higher than the reported max + for (i = 0; i < bb->num_states; i++) { + if ((bb->clock_limits[i].dcfclk_mhz > (max_clocks.dcfClockInKhz / 1000)) + && max_clocks.dcfClockInKhz != 0) + bb->clock_limits[i].dcfclk_mhz = (max_clocks.dcfClockInKhz / 1000); + + if ((bb->clock_limits[i].dram_speed_mts > (max_clocks.uClockInKhz / 1000) * 16) + && max_clocks.uClockInKhz != 0) + bb->clock_limits[i].dram_speed_mts = (max_clocks.uClockInKhz / 1000) * 16; + + if ((bb->clock_limits[i].fabricclk_mhz > (max_clocks.fabricClockInKhz / 1000)) + && max_clocks.fabricClockInKhz != 0) + bb->clock_limits[i].fabricclk_mhz = (max_clocks.fabricClockInKhz / 1000); + + if ((bb->clock_limits[i].dispclk_mhz > (max_clocks.displayClockInKhz / 1000)) + && max_clocks.displayClockInKhz != 0) + bb->clock_limits[i].dispclk_mhz = (max_clocks.displayClockInKhz / 1000); + + if ((bb->clock_limits[i].dppclk_mhz > (max_clocks.dppClockInKhz / 1000)) + && max_clocks.dppClockInKhz != 0) + bb->clock_limits[i].dppclk_mhz = (max_clocks.dppClockInKhz / 1000); + + if ((bb->clock_limits[i].phyclk_mhz > (max_clocks.phyClockInKhz / 1000)) + && max_clocks.phyClockInKhz != 0) + bb->clock_limits[i].phyclk_mhz = (max_clocks.phyClockInKhz / 1000); + + if ((bb->clock_limits[i].socclk_mhz > (max_clocks.socClockInKhz / 1000)) + && max_clocks.socClockInKhz != 0) + bb->clock_limits[i].socclk_mhz = (max_clocks.socClockInKhz / 1000); + + if ((bb->clock_limits[i].dscclk_mhz > (max_clocks.dscClockInKhz / 1000)) + && max_clocks.dscClockInKhz != 0) + bb->clock_limits[i].dscclk_mhz = (max_clocks.dscClockInKhz / 1000); + } + + // Second pass - remove all duplicate clock states + for (i = bb->num_states - 1; i > 1; i--) { + bool duplicate = true; + + if (bb->clock_limits[i-1].dcfclk_mhz != bb->clock_limits[i].dcfclk_mhz) + duplicate = false; + if (bb->clock_limits[i-1].dispclk_mhz != bb->clock_limits[i].dispclk_mhz) + duplicate = false; + if (bb->clock_limits[i-1].dppclk_mhz != bb->clock_limits[i].dppclk_mhz) + duplicate = false; + if (bb->clock_limits[i-1].dram_speed_mts != bb->clock_limits[i].dram_speed_mts) + duplicate = false; + if (bb->clock_limits[i-1].dscclk_mhz != bb->clock_limits[i].dscclk_mhz) + duplicate = false; + if (bb->clock_limits[i-1].fabricclk_mhz != bb->clock_limits[i].fabricclk_mhz) + duplicate = false; + if (bb->clock_limits[i-1].phyclk_mhz != bb->clock_limits[i].phyclk_mhz) + duplicate = false; + if (bb->clock_limits[i-1].socclk_mhz != bb->clock_limits[i].socclk_mhz) + duplicate = false; + + if (duplicate) + bb->num_states--; + } +} + +void dcn20_patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb) +{ + dc_assert_fp_enabled(); + + if ((int)(bb->sr_exit_time_us * 1000) != dc->bb_overrides.sr_exit_time_ns + && dc->bb_overrides.sr_exit_time_ns) { + bb->sr_exit_time_us = dc->bb_overrides.sr_exit_time_ns / 1000.0; + } + + if ((int)(bb->sr_enter_plus_exit_time_us * 1000) + != dc->bb_overrides.sr_enter_plus_exit_time_ns + && dc->bb_overrides.sr_enter_plus_exit_time_ns) { + bb->sr_enter_plus_exit_time_us = + dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; + } + + if ((int)(bb->urgent_latency_us * 1000) != dc->bb_overrides.urgent_latency_ns + && dc->bb_overrides.urgent_latency_ns) { + bb->urgent_latency_us = dc->bb_overrides.urgent_latency_ns / 1000.0; + } + + if ((int)(bb->dram_clock_change_latency_us * 1000) + != dc->bb_overrides.dram_clock_change_latency_ns + && dc->bb_overrides.dram_clock_change_latency_ns) { + bb->dram_clock_change_latency_us = + dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; + } + + if ((int)(bb->dummy_pstate_latency_us * 1000) + != dc->bb_overrides.dummy_clock_change_latency_ns + && dc->bb_overrides.dummy_clock_change_latency_ns) { + bb->dummy_pstate_latency_us = + dc->bb_overrides.dummy_clock_change_latency_ns / 1000.0; + } +} + +static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *context, + bool fast_validate) +{ + bool out = false; + + BW_VAL_TRACE_SETUP(); + + int vlevel = 0; + int pipe_split_from[MAX_PIPES]; + int pipe_cnt = 0; + display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_ATOMIC); + DC_LOGGER_INIT(dc->ctx->logger); + + BW_VAL_TRACE_COUNT(); + + out = dcn20_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel, fast_validate); + + if (pipe_cnt == 0) + goto validate_out; + + if (!out) + goto validate_fail; + + BW_VAL_TRACE_END_VOLTAGE_LEVEL(); + + if (fast_validate) { + BW_VAL_TRACE_SKIP(fast); + goto validate_out; + } + + dcn20_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, fast_validate); + dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); + + BW_VAL_TRACE_END_WATERMARKS(); + + goto validate_out; + +validate_fail: + DC_LOG_WARNING("Mode Validation Warning: %s failed validation.\n", + dml_get_status_message(context->bw_ctx.dml.vba.ValidationStatus[context->bw_ctx.dml.vba.soc.num_states])); + + BW_VAL_TRACE_SKIP(fail); + out = false; + +validate_out: + kfree(pipes); + + BW_VAL_TRACE_FINISH(); + + return out; +} + +bool dcn20_validate_bandwidth_fp(struct dc *dc, + struct dc_state *context, + bool fast_validate) +{ + bool voltage_supported = false; + bool full_pstate_supported = false; + bool dummy_pstate_supported = false; + double p_state_latency_us; + + dc_assert_fp_enabled(); + + p_state_latency_us = context->bw_ctx.dml.soc.dram_clock_change_latency_us; + context->bw_ctx.dml.soc.disable_dram_clock_change_vactive_support = + dc->debug.disable_dram_clock_change_vactive_support; + context->bw_ctx.dml.soc.allow_dram_clock_one_display_vactive = + dc->debug.enable_dram_clock_change_one_display_vactive; + + /*Unsafe due to current pipe merge and split logic*/ + ASSERT(context != dc->current_state); + + if (fast_validate) { + return dcn20_validate_bandwidth_internal(dc, context, true); + } + + // Best case, we support full UCLK switch latency + voltage_supported = dcn20_validate_bandwidth_internal(dc, context, false); + full_pstate_supported = context->bw_ctx.bw.dcn.clk.p_state_change_support; + + if (context->bw_ctx.dml.soc.dummy_pstate_latency_us == 0 || + (voltage_supported && full_pstate_supported)) { + context->bw_ctx.bw.dcn.clk.p_state_change_support = full_pstate_supported; + goto restore_dml_state; + } + + // Fallback: Try to only support G6 temperature read latency + context->bw_ctx.dml.soc.dram_clock_change_latency_us = context->bw_ctx.dml.soc.dummy_pstate_latency_us; + + voltage_supported = dcn20_validate_bandwidth_internal(dc, context, false); + dummy_pstate_supported = context->bw_ctx.bw.dcn.clk.p_state_change_support; + + if (voltage_supported && (dummy_pstate_supported || !(context->stream_count))) { + context->bw_ctx.bw.dcn.clk.p_state_change_support = false; + goto restore_dml_state; + } + + // ERROR: fallback is supposed to always work. + ASSERT(false); + +restore_dml_state: + context->bw_ctx.dml.soc.dram_clock_change_latency_us = p_state_latency_us; + return voltage_supported; +} + +void dcn20_fpu_set_wm_ranges(int i, + struct pp_smu_wm_range_sets *ranges, + struct _vcs_dpi_soc_bounding_box_st *loaded_bb) +{ + dc_assert_fp_enabled(); + + ranges->reader_wm_sets[i].min_fill_clk_mhz = (i > 0) ? (loaded_bb->clock_limits[i - 1].dram_speed_mts / 16) + 1 : 0; + ranges->reader_wm_sets[i].max_fill_clk_mhz = loaded_bb->clock_limits[i].dram_speed_mts / 16; +} + +void dcn20_fpu_adjust_dppclk(struct vba_vars_st *v, + int vlevel, + int max_mpc_comb, + int pipe_idx, + bool is_validating_bw) +{ + dc_assert_fp_enabled(); + + if (is_validating_bw) + v->RequiredDPPCLK[vlevel][max_mpc_comb][pipe_idx] *= 2; + else + v->RequiredDPPCLK[vlevel][max_mpc_comb][pipe_idx] /= 2; +} + +int dcn21_populate_dml_pipes_from_context(struct dc *dc, + struct dc_state *context, + display_e2e_pipe_params_st *pipes, + bool fast_validate) +{ + uint32_t pipe_cnt; + int i; + + dc_assert_fp_enabled(); + + pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + + for (i = 0; i < pipe_cnt; i++) { + + pipes[i].pipe.src.hostvm = dc->res_pool->hubbub->riommu_active; + pipes[i].pipe.src.gpuvm = 1; + } + + return pipe_cnt; +} + +static void patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb) +{ + int i; + + if (dc->bb_overrides.sr_exit_time_ns) { + for (i = 0; i < WM_SET_COUNT; i++) { + dc->clk_mgr->bw_params->wm_table.entries[i].sr_exit_time_us = + dc->bb_overrides.sr_exit_time_ns / 1000.0; + } + } + + if (dc->bb_overrides.sr_enter_plus_exit_time_ns) { + for (i = 0; i < WM_SET_COUNT; i++) { + dc->clk_mgr->bw_params->wm_table.entries[i].sr_enter_plus_exit_time_us = + dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; + } + } + + if (dc->bb_overrides.urgent_latency_ns) { + bb->urgent_latency_us = dc->bb_overrides.urgent_latency_ns / 1000.0; + } + + if (dc->bb_overrides.dram_clock_change_latency_ns) { + for (i = 0; i < WM_SET_COUNT; i++) { + dc->clk_mgr->bw_params->wm_table.entries[i].pstate_latency_us = + dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; + } + } +} + +static void calculate_wm_set_for_vlevel(int vlevel, + struct wm_range_table_entry *table_entry, + struct dcn_watermarks *wm_set, + struct display_mode_lib *dml, + display_e2e_pipe_params_st *pipes, + int pipe_cnt) +{ + double dram_clock_change_latency_cached = dml->soc.dram_clock_change_latency_us; + + ASSERT(vlevel < dml->soc.num_states); + /* only pipe 0 is read for voltage and dcf/soc clocks */ + pipes[0].clks_cfg.voltage = vlevel; + pipes[0].clks_cfg.dcfclk_mhz = dml->soc.clock_limits[vlevel].dcfclk_mhz; + pipes[0].clks_cfg.socclk_mhz = dml->soc.clock_limits[vlevel].socclk_mhz; + + dml->soc.dram_clock_change_latency_us = table_entry->pstate_latency_us; + dml->soc.sr_exit_time_us = table_entry->sr_exit_time_us; + dml->soc.sr_enter_plus_exit_time_us = table_entry->sr_enter_plus_exit_time_us; + + wm_set->urgent_ns = get_wm_urgent(dml, pipes, pipe_cnt) * 1000; + wm_set->cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(dml, pipes, pipe_cnt) * 1000; + wm_set->cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(dml, pipes, pipe_cnt) * 1000; + wm_set->cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(dml, pipes, pipe_cnt) * 1000; + wm_set->pte_meta_urgent_ns = get_wm_memory_trip(dml, pipes, pipe_cnt) * 1000; + wm_set->frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(dml, pipes, pipe_cnt) * 1000; + wm_set->frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(dml, pipes, pipe_cnt) * 1000; + wm_set->urgent_latency_ns = get_urgent_latency(dml, pipes, pipe_cnt) * 1000; + dml->soc.dram_clock_change_latency_us = dram_clock_change_latency_cached; +} + +static void dcn21_calculate_wm(struct dc *dc, struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int *out_pipe_cnt, + int *pipe_split_from, + int vlevel_req, + bool fast_validate) +{ + int pipe_cnt, i, pipe_idx; + int vlevel, vlevel_max; + struct wm_range_table_entry *table_entry; + struct clk_bw_params *bw_params = dc->clk_mgr->bw_params; + + ASSERT(bw_params); + + patch_bounding_box(dc, &context->bw_ctx.dml.soc); + + for (i = 0, pipe_idx = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { + if (!context->res_ctx.pipe_ctx[i].stream) + continue; + + pipes[pipe_cnt].clks_cfg.refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000.0; + pipes[pipe_cnt].clks_cfg.dispclk_mhz = context->bw_ctx.dml.vba.RequiredDISPCLK[vlevel_req][context->bw_ctx.dml.vba.maxMpcComb]; + + if (pipe_split_from[i] < 0) { + pipes[pipe_cnt].clks_cfg.dppclk_mhz = + context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel_req][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx]; + if (context->bw_ctx.dml.vba.BlendingAndTiming[pipe_idx] == pipe_idx) + pipes[pipe_cnt].pipe.dest.odm_combine = + context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel_req][pipe_idx]; + else + pipes[pipe_cnt].pipe.dest.odm_combine = 0; + pipe_idx++; + } else { + pipes[pipe_cnt].clks_cfg.dppclk_mhz = + context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel_req][context->bw_ctx.dml.vba.maxMpcComb][pipe_split_from[i]]; + if (context->bw_ctx.dml.vba.BlendingAndTiming[pipe_split_from[i]] == pipe_split_from[i]) + pipes[pipe_cnt].pipe.dest.odm_combine = + context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel_req][pipe_split_from[i]]; + else + pipes[pipe_cnt].pipe.dest.odm_combine = 0; + } + pipe_cnt++; + } + + if (pipe_cnt != pipe_idx) { + if (dc->res_pool->funcs->populate_dml_pipes) + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, + context, pipes, fast_validate); + else + pipe_cnt = dcn21_populate_dml_pipes_from_context(dc, + context, pipes, fast_validate); + } + + *out_pipe_cnt = pipe_cnt; + + vlevel_max = bw_params->clk_table.num_entries - 1; + + + /* WM Set D */ + table_entry = &bw_params->wm_table.entries[WM_D]; + if (table_entry->wm_type == WM_TYPE_RETRAINING) + vlevel = 0; + else + vlevel = vlevel_max; + calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.d, + &context->bw_ctx.dml, pipes, pipe_cnt); + /* WM Set C */ + table_entry = &bw_params->wm_table.entries[WM_C]; + vlevel = MIN(MAX(vlevel_req, 3), vlevel_max); + calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.c, + &context->bw_ctx.dml, pipes, pipe_cnt); + /* WM Set B */ + table_entry = &bw_params->wm_table.entries[WM_B]; + vlevel = MIN(MAX(vlevel_req, 2), vlevel_max); + calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.b, + &context->bw_ctx.dml, pipes, pipe_cnt); + + /* WM Set A */ + table_entry = &bw_params->wm_table.entries[WM_A]; + vlevel = MIN(vlevel_req, vlevel_max); + calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.a, + &context->bw_ctx.dml, pipes, pipe_cnt); +} + +bool dcn21_validate_bandwidth_fp(struct dc *dc, + struct dc_state *context, + bool fast_validate) +{ + bool out = false; + + BW_VAL_TRACE_SETUP(); + + int vlevel = 0; + int pipe_split_from[MAX_PIPES]; + int pipe_cnt = 0; + display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_ATOMIC); + DC_LOGGER_INIT(dc->ctx->logger); + + BW_VAL_TRACE_COUNT(); + + dc_assert_fp_enabled(); + + /*Unsafe due to current pipe merge and split logic*/ + ASSERT(context != dc->current_state); + + out = dcn21_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel, fast_validate); + + if (pipe_cnt == 0) + goto validate_out; + + if (!out) + goto validate_fail; + + BW_VAL_TRACE_END_VOLTAGE_LEVEL(); + + if (fast_validate) { + BW_VAL_TRACE_SKIP(fast); + goto validate_out; + } + + dcn21_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, fast_validate); + dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); + + BW_VAL_TRACE_END_WATERMARKS(); + + goto validate_out; + +validate_fail: + DC_LOG_WARNING("Mode Validation Warning: %s failed validation.\n", + dml_get_status_message(context->bw_ctx.dml.vba.ValidationStatus[context->bw_ctx.dml.vba.soc.num_states])); + + BW_VAL_TRACE_SKIP(fail); + out = false; + +validate_out: + kfree(pipes); + + BW_VAL_TRACE_FINISH(); + + return out; +} + +static struct _vcs_dpi_voltage_scaling_st construct_low_pstate_lvl(struct clk_limit_table *clk_table, unsigned int high_voltage_lvl) +{ + struct _vcs_dpi_voltage_scaling_st low_pstate_lvl; + int i; + + low_pstate_lvl.state = 1; + low_pstate_lvl.dcfclk_mhz = clk_table->entries[0].dcfclk_mhz; + low_pstate_lvl.fabricclk_mhz = clk_table->entries[0].fclk_mhz; + low_pstate_lvl.socclk_mhz = clk_table->entries[0].socclk_mhz; + low_pstate_lvl.dram_speed_mts = clk_table->entries[0].memclk_mhz * 2; + + low_pstate_lvl.dispclk_mhz = dcn2_1_soc.clock_limits[high_voltage_lvl].dispclk_mhz; + low_pstate_lvl.dppclk_mhz = dcn2_1_soc.clock_limits[high_voltage_lvl].dppclk_mhz; + low_pstate_lvl.dram_bw_per_chan_gbps = dcn2_1_soc.clock_limits[high_voltage_lvl].dram_bw_per_chan_gbps; + low_pstate_lvl.dscclk_mhz = dcn2_1_soc.clock_limits[high_voltage_lvl].dscclk_mhz; + low_pstate_lvl.dtbclk_mhz = dcn2_1_soc.clock_limits[high_voltage_lvl].dtbclk_mhz; + low_pstate_lvl.phyclk_d18_mhz = dcn2_1_soc.clock_limits[high_voltage_lvl].phyclk_d18_mhz; + low_pstate_lvl.phyclk_mhz = dcn2_1_soc.clock_limits[high_voltage_lvl].phyclk_mhz; + + for (i = clk_table->num_entries; i > 1; i--) + clk_table->entries[i] = clk_table->entries[i-1]; + clk_table->entries[1] = clk_table->entries[0]; + clk_table->num_entries++; + + return low_pstate_lvl; +} + +void dcn21_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) +{ + struct dcn21_resource_pool *pool = TO_DCN21_RES_POOL(dc->res_pool); + struct clk_limit_table *clk_table = &bw_params->clk_table; + struct _vcs_dpi_voltage_scaling_st clock_limits[DC__VOLTAGE_STATES]; + unsigned int i, closest_clk_lvl = 0, k = 0; + int j; + + dc_assert_fp_enabled(); + + dcn2_1_ip.max_num_otg = pool->base.res_cap->num_timing_generator; + dcn2_1_ip.max_num_dpp = pool->base.pipe_count; + dcn2_1_soc.num_chans = bw_params->num_channels; + + ASSERT(clk_table->num_entries); + /* Copy dcn2_1_soc.clock_limits to clock_limits to avoid copying over null states later */ + for (i = 0; i < dcn2_1_soc.num_states + 1; i++) { + clock_limits[i] = dcn2_1_soc.clock_limits[i]; + } + + for (i = 0; i < clk_table->num_entries; i++) { + /* loop backwards*/ + for (closest_clk_lvl = 0, j = dcn2_1_soc.num_states - 1; j >= 0; j--) { + if ((unsigned int) dcn2_1_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) { + closest_clk_lvl = j; + break; + } + } + + /* clk_table[1] is reserved for min DF PState. skip here to fill in later. */ + if (i == 1) + k++; + + clock_limits[k].state = k; + clock_limits[k].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz; + clock_limits[k].fabricclk_mhz = clk_table->entries[i].fclk_mhz; + clock_limits[k].socclk_mhz = clk_table->entries[i].socclk_mhz; + clock_limits[k].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2; + + clock_limits[k].dispclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dispclk_mhz; + clock_limits[k].dppclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dppclk_mhz; + clock_limits[k].dram_bw_per_chan_gbps = dcn2_1_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps; + clock_limits[k].dscclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dscclk_mhz; + clock_limits[k].dtbclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dtbclk_mhz; + clock_limits[k].phyclk_d18_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz; + clock_limits[k].phyclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].phyclk_mhz; + + k++; + } + for (i = 0; i < clk_table->num_entries + 1; i++) + dcn2_1_soc.clock_limits[i] = clock_limits[i]; + if (clk_table->num_entries) { + dcn2_1_soc.num_states = clk_table->num_entries + 1; + /* fill in min DF PState */ + dcn2_1_soc.clock_limits[1] = construct_low_pstate_lvl(clk_table, closest_clk_lvl); + /* duplicate last level */ + dcn2_1_soc.clock_limits[dcn2_1_soc.num_states] = dcn2_1_soc.clock_limits[dcn2_1_soc.num_states - 1]; + dcn2_1_soc.clock_limits[dcn2_1_soc.num_states].state = dcn2_1_soc.num_states; + } + + dml_init_instance(&dc->dml, &dcn2_1_soc, &dcn2_1_ip, DML_PROJECT_DCN21); +} diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.h index 36f26126d574..aa892193e485 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.h @@ -23,6 +23,7 @@ * Authors: AMD * */ +#include "core_types.h" #ifndef __DCN20_FPU_H__ #define __DCN20_FPU_H__ @@ -31,4 +32,54 @@ void dcn20_populate_dml_writeback_from_context(struct dc *dc, struct resource_context *res_ctx, display_e2e_pipe_params_st *pipes); +void dcn20_fpu_set_wb_arb_params(struct mcif_arb_params *wb_arb_params, + struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int pipe_cnt, int i); +void dcn20_calculate_dlg_params(struct dc *dc, + struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int pipe_cnt, + int vlevel); +int dcn20_populate_dml_pipes_from_context(struct dc *dc, + struct dc_state *context, + display_e2e_pipe_params_st *pipes, + bool fast_validate); +void dcn20_calculate_wm(struct dc *dc, + struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int *out_pipe_cnt, + int *pipe_split_from, + int vlevel, + bool fast_validate); +void dcn20_cap_soc_clocks(struct _vcs_dpi_soc_bounding_box_st *bb, + struct pp_smu_nv_clock_table max_clocks); +void dcn20_update_bounding_box(struct dc *dc, + struct _vcs_dpi_soc_bounding_box_st *bb, + struct pp_smu_nv_clock_table *max_clocks, + unsigned int *uclk_states, + unsigned int num_states); +void dcn20_patch_bounding_box(struct dc *dc, + struct _vcs_dpi_soc_bounding_box_st *bb); +bool dcn20_validate_bandwidth_fp(struct dc *dc, + struct dc_state *context, + bool fast_validate); +void dcn20_fpu_set_wm_ranges(int i, + struct pp_smu_wm_range_sets *ranges, + struct _vcs_dpi_soc_bounding_box_st *loaded_bb); +void dcn20_fpu_adjust_dppclk(struct vba_vars_st *v, + int vlevel, + int max_mpc_comb, + int pipe_idx, + bool is_validating_bw); + +int dcn21_populate_dml_pipes_from_context(struct dc *dc, + struct dc_state *context, + display_e2e_pipe_params_st *pipes, + bool fast_validate); +bool dcn21_validate_bandwidth_fp(struct dc *dc, + struct dc_state *context, + bool fast_validate); +void dcn21_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params); + #endif /* __DCN20_FPU_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c new file mode 100644 index 000000000000..8fb14baed208 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c @@ -0,0 +1,362 @@ +/* + * Copyright 2019-2021 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#include "resource.h" +#include "clk_mgr.h" +#include "dcn20/dcn20_resource.h" +#include "dcn303/dcn303_resource.h" + +#include "dml/dcn20/dcn20_fpu.h" +#include "dcn303_fpu.h" + +struct _vcs_dpi_ip_params_st dcn3_03_ip = { + .use_min_dcfclk = 0, + .clamp_min_dcfclk = 0, + .odm_capable = 1, + .gpuvm_enable = 1, + .hostvm_enable = 0, + .gpuvm_max_page_table_levels = 4, + .hostvm_max_page_table_levels = 4, + .hostvm_cached_page_table_levels = 0, + .pte_group_size_bytes = 2048, + .num_dsc = 2, + .rob_buffer_size_kbytes = 184, + .det_buffer_size_kbytes = 184, + .dpte_buffer_size_in_pte_reqs_luma = 64, + .dpte_buffer_size_in_pte_reqs_chroma = 34, + .pde_proc_buffer_size_64k_reqs = 48, + .dpp_output_buffer_pixels = 2560, + .opp_output_buffer_lines = 1, + .pixel_chunk_size_kbytes = 8, + .pte_enable = 1, + .max_page_table_levels = 2, + .pte_chunk_size_kbytes = 2, // ? + .meta_chunk_size_kbytes = 2, + .writeback_chunk_size_kbytes = 8, + .line_buffer_size_bits = 789504, + .is_line_buffer_bpp_fixed = 0, // ? + .line_buffer_fixed_bpp = 0, // ? + .dcc_supported = true, + .writeback_interface_buffer_size_kbytes = 90, + .writeback_line_buffer_buffer_size = 0, + .max_line_buffer_lines = 12, + .writeback_luma_buffer_size_kbytes = 12, // writeback_line_buffer_buffer_size = 656640 + .writeback_chroma_buffer_size_kbytes = 8, + .writeback_chroma_line_buffer_width_pixels = 4, + .writeback_max_hscl_ratio = 1, + .writeback_max_vscl_ratio = 1, + .writeback_min_hscl_ratio = 1, + .writeback_min_vscl_ratio = 1, + .writeback_max_hscl_taps = 1, + .writeback_max_vscl_taps = 1, + .writeback_line_buffer_luma_buffer_size = 0, + .writeback_line_buffer_chroma_buffer_size = 14643, + .cursor_buffer_size = 8, + .cursor_chunk_size = 2, + .max_num_otg = 2, + .max_num_dpp = 2, + .max_num_wb = 1, + .max_dchub_pscl_bw_pix_per_clk = 4, + .max_pscl_lb_bw_pix_per_clk = 2, + .max_lb_vscl_bw_pix_per_clk = 4, + .max_vscl_hscl_bw_pix_per_clk = 4, + .max_hscl_ratio = 6, + .max_vscl_ratio = 6, + .hscl_mults = 4, + .vscl_mults = 4, + .max_hscl_taps = 8, + .max_vscl_taps = 8, + .dispclk_ramp_margin_percent = 1, + .underscan_factor = 1.11, + .min_vblank_lines = 32, + .dppclk_delay_subtotal = 46, + .dynamic_metadata_vm_enabled = true, + .dppclk_delay_scl_lb_only = 16, + .dppclk_delay_scl = 50, + .dppclk_delay_cnvc_formatter = 27, + .dppclk_delay_cnvc_cursor = 6, + .dispclk_delay_subtotal = 119, + .dcfclk_cstate_latency = 5.2, // SRExitTime + .max_inter_dcn_tile_repeaters = 8, + .max_num_hdmi_frl_outputs = 1, + .odm_combine_4to1_supported = false, + + .xfc_supported = false, + .xfc_fill_bw_overhead_percent = 10.0, + .xfc_fill_constant_bytes = 0, + .gfx7_compat_tiling_supported = 0, + .number_of_cursors = 1, +}; + +struct _vcs_dpi_soc_bounding_box_st dcn3_03_soc = { + .clock_limits = { + { + .state = 0, + .dispclk_mhz = 562.0, + .dppclk_mhz = 300.0, + .phyclk_mhz = 300.0, + .phyclk_d18_mhz = 667.0, + .dscclk_mhz = 405.6, + }, + }, + + .min_dcfclk = 500.0, /* TODO: set this to actual min DCFCLK */ + .num_states = 1, + .sr_exit_time_us = 35.5, + .sr_enter_plus_exit_time_us = 40, + .urgent_latency_us = 4.0, + .urgent_latency_pixel_data_only_us = 4.0, + .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, + .urgent_latency_vm_data_only_us = 4.0, + .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, + .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, + .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 80.0, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0, + .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, + .max_avg_sdp_bw_use_normal_percent = 60.0, + .max_avg_dram_bw_use_normal_percent = 40.0, + .writeback_latency_us = 12.0, + .max_request_size_bytes = 256, + .fabric_datapath_to_dcn_data_return_bytes = 64, + .dcn_downspread_percent = 0.5, + .downspread_percent = 0.38, + .dram_page_open_time_ns = 50.0, + .dram_rw_turnaround_time_ns = 17.5, + .dram_return_buffer_per_channel_bytes = 8192, + .round_trip_ping_latency_dcfclk_cycles = 156, + .urgent_out_of_order_return_per_channel_bytes = 4096, + .channel_interleave_bytes = 256, + .num_banks = 8, + .gpuvm_min_page_size_bytes = 4096, + .hostvm_min_page_size_bytes = 4096, + .dram_clock_change_latency_us = 404, + .dummy_pstate_latency_us = 5, + .writeback_dram_clock_change_latency_us = 23.0, + .return_bus_width_bytes = 64, + .dispclk_dppclk_vco_speed_mhz = 3650, + .xfc_bus_transport_time_us = 20, // ? + .xfc_xbuf_latency_tolerance_us = 4, // ? + .use_urgent_burst_bw = 1, // ? + .do_urgent_latency_adjustment = true, + .urgent_latency_adjustment_fabric_clock_component_us = 1.0, + .urgent_latency_adjustment_fabric_clock_reference_mhz = 1000, +}; + +static void dcn303_get_optimal_dcfclk_fclk_for_uclk(unsigned int uclk_mts, + unsigned int *optimal_dcfclk, + unsigned int *optimal_fclk) +{ + double bw_from_dram, bw_from_dram1, bw_from_dram2; + + bw_from_dram1 = uclk_mts * dcn3_03_soc.num_chans * + dcn3_03_soc.dram_channel_width_bytes * (dcn3_03_soc.max_avg_dram_bw_use_normal_percent / 100); + bw_from_dram2 = uclk_mts * dcn3_03_soc.num_chans * + dcn3_03_soc.dram_channel_width_bytes * (dcn3_03_soc.max_avg_sdp_bw_use_normal_percent / 100); + + bw_from_dram = (bw_from_dram1 < bw_from_dram2) ? bw_from_dram1 : bw_from_dram2; + + if (optimal_fclk) + *optimal_fclk = bw_from_dram / + (dcn3_03_soc.fabric_datapath_to_dcn_data_return_bytes * + (dcn3_03_soc.max_avg_sdp_bw_use_normal_percent / 100)); + + if (optimal_dcfclk) + *optimal_dcfclk = bw_from_dram / + (dcn3_03_soc.return_bus_width_bytes * (dcn3_03_soc.max_avg_sdp_bw_use_normal_percent / 100)); +} + + +void dcn303_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) +{ + unsigned int i, j; + unsigned int num_states = 0; + + unsigned int dcfclk_mhz[DC__VOLTAGE_STATES] = {0}; + unsigned int dram_speed_mts[DC__VOLTAGE_STATES] = {0}; + unsigned int optimal_uclk_for_dcfclk_sta_targets[DC__VOLTAGE_STATES] = {0}; + unsigned int optimal_dcfclk_for_uclk[DC__VOLTAGE_STATES] = {0}; + + unsigned int dcfclk_sta_targets[DC__VOLTAGE_STATES] = {694, 875, 1000, 1200}; + unsigned int num_dcfclk_sta_targets = 4; + unsigned int num_uclk_states; + + dc_assert_fp_enabled(); + + if (dc->ctx->dc_bios->vram_info.num_chans) + dcn3_03_soc.num_chans = dc->ctx->dc_bios->vram_info.num_chans; + + if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) + dcn3_03_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; + + dcn3_03_soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0; + dc->dml.soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0; + + if (bw_params->clk_table.entries[0].memclk_mhz) { + int max_dcfclk_mhz = 0, max_dispclk_mhz = 0, max_dppclk_mhz = 0, max_phyclk_mhz = 0; + + for (i = 0; i < MAX_NUM_DPM_LVL; i++) { + if (bw_params->clk_table.entries[i].dcfclk_mhz > max_dcfclk_mhz) + max_dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz; + if (bw_params->clk_table.entries[i].dispclk_mhz > max_dispclk_mhz) + max_dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz; + if (bw_params->clk_table.entries[i].dppclk_mhz > max_dppclk_mhz) + max_dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz; + if (bw_params->clk_table.entries[i].phyclk_mhz > max_phyclk_mhz) + max_phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz; + } + if (!max_dcfclk_mhz) + max_dcfclk_mhz = dcn3_03_soc.clock_limits[0].dcfclk_mhz; + if (!max_dispclk_mhz) + max_dispclk_mhz = dcn3_03_soc.clock_limits[0].dispclk_mhz; + if (!max_dppclk_mhz) + max_dppclk_mhz = dcn3_03_soc.clock_limits[0].dppclk_mhz; + if (!max_phyclk_mhz) + max_phyclk_mhz = dcn3_03_soc.clock_limits[0].phyclk_mhz; + + if (max_dcfclk_mhz > dcfclk_sta_targets[num_dcfclk_sta_targets-1]) { + dcfclk_sta_targets[num_dcfclk_sta_targets] = max_dcfclk_mhz; + num_dcfclk_sta_targets++; + } else if (max_dcfclk_mhz < dcfclk_sta_targets[num_dcfclk_sta_targets-1]) { + for (i = 0; i < num_dcfclk_sta_targets; i++) { + if (dcfclk_sta_targets[i] > max_dcfclk_mhz) { + dcfclk_sta_targets[i] = max_dcfclk_mhz; + break; + } + } + /* Update size of array since we "removed" duplicates */ + num_dcfclk_sta_targets = i + 1; + } + + num_uclk_states = bw_params->clk_table.num_entries; + + /* Calculate optimal dcfclk for each uclk */ + for (i = 0; i < num_uclk_states; i++) { + dcn303_get_optimal_dcfclk_fclk_for_uclk(bw_params->clk_table.entries[i].memclk_mhz * 16, + &optimal_dcfclk_for_uclk[i], NULL); + if (optimal_dcfclk_for_uclk[i] < bw_params->clk_table.entries[0].dcfclk_mhz) + optimal_dcfclk_for_uclk[i] = bw_params->clk_table.entries[0].dcfclk_mhz; + } + + /* Calculate optimal uclk for each dcfclk sta target */ + for (i = 0; i < num_dcfclk_sta_targets; i++) { + for (j = 0; j < num_uclk_states; j++) { + if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) { + optimal_uclk_for_dcfclk_sta_targets[i] = + bw_params->clk_table.entries[j].memclk_mhz * 16; + break; + } + } + } + + i = 0; + j = 0; + /* create the final dcfclk and uclk table */ + while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) { + if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) { + dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; + dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; + } else { + if (j < num_uclk_states && optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) { + dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j]; + dram_speed_mts[num_states++] = + bw_params->clk_table.entries[j++].memclk_mhz * 16; + } else { + j = num_uclk_states; + } + } + } + + while (i < num_dcfclk_sta_targets && num_states < DC__VOLTAGE_STATES) { + dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; + dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; + } + + while (j < num_uclk_states && num_states < DC__VOLTAGE_STATES && + optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) { + dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j]; + dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16; + } + + dcn3_03_soc.num_states = num_states; + for (i = 0; i < dcn3_03_soc.num_states; i++) { + dcn3_03_soc.clock_limits[i].state = i; + dcn3_03_soc.clock_limits[i].dcfclk_mhz = dcfclk_mhz[i]; + dcn3_03_soc.clock_limits[i].fabricclk_mhz = dcfclk_mhz[i]; + dcn3_03_soc.clock_limits[i].dram_speed_mts = dram_speed_mts[i]; + + /* Fill all states with max values of all other clocks */ + dcn3_03_soc.clock_limits[i].dispclk_mhz = max_dispclk_mhz; + dcn3_03_soc.clock_limits[i].dppclk_mhz = max_dppclk_mhz; + dcn3_03_soc.clock_limits[i].phyclk_mhz = max_phyclk_mhz; + /* Populate from bw_params for DTBCLK, SOCCLK */ + if (!bw_params->clk_table.entries[i].dtbclk_mhz && i > 0) + dcn3_03_soc.clock_limits[i].dtbclk_mhz = dcn3_03_soc.clock_limits[i-1].dtbclk_mhz; + else + dcn3_03_soc.clock_limits[i].dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz; + if (!bw_params->clk_table.entries[i].socclk_mhz && i > 0) + dcn3_03_soc.clock_limits[i].socclk_mhz = dcn3_03_soc.clock_limits[i-1].socclk_mhz; + else + dcn3_03_soc.clock_limits[i].socclk_mhz = bw_params->clk_table.entries[i].socclk_mhz; + /* These clocks cannot come from bw_params, always fill from dcn3_03_soc[1] */ + /* FCLK, PHYCLK_D18, DSCCLK */ + dcn3_03_soc.clock_limits[i].phyclk_d18_mhz = dcn3_03_soc.clock_limits[0].phyclk_d18_mhz; + dcn3_03_soc.clock_limits[i].dscclk_mhz = dcn3_03_soc.clock_limits[0].dscclk_mhz; + } + + if (dcn3_03_soc.num_chans <= 4) { + for (i = 0; i < dcn3_03_soc.num_states; i++) { + if (dcn3_03_soc.clock_limits[i].dram_speed_mts > 1700) + break; + + if (dcn3_03_soc.clock_limits[i].dram_speed_mts >= 1500) { + dcn3_03_soc.clock_limits[i].dcfclk_mhz = 100; + dcn3_03_soc.clock_limits[i].fabricclk_mhz = 100; + } + } + } + + /* re-init DML with updated bb */ + dml_init_instance(&dc->dml, &dcn3_03_soc, &dcn3_03_ip, DML_PROJECT_DCN30); + if (dc->current_state) + dml_init_instance(&dc->current_state->bw_ctx.dml, &dcn3_03_soc, &dcn3_03_ip, DML_PROJECT_DCN30); + } +} + +void dcn303_fpu_init_soc_bounding_box(struct bp_soc_bb_info bb_info) +{ + dc_assert_fp_enabled(); + + if (bb_info.dram_clock_change_latency_100ns > 0) + dcn3_03_soc.dram_clock_change_latency_us = + bb_info.dram_clock_change_latency_100ns * 10; + + if (bb_info.dram_sr_enter_exit_latency_100ns > 0) + dcn3_03_soc.sr_enter_plus_exit_time_us = + bb_info.dram_sr_enter_exit_latency_100ns * 10; + + if (bb_info.dram_sr_exit_latency_100ns > 0) + dcn3_03_soc.sr_exit_time_us = + bb_info.dram_sr_exit_latency_100ns * 10; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c b/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.h index 42413813765a..92ec833fa528 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.h @@ -1,5 +1,5 @@ /* - * Copyright 2019 Advanced Micro Devices, Inc. + * Copyright 2019-2021 Advanced Micro Devices, Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -19,14 +19,14 @@ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. * + * Authors: AMD + * */ -#include "amdgpu.h" -#include "amdgpu_ras.h" +#ifndef __DCN303_FPU_H__ +#define __DCN303_FPU_H__ + +void dcn303_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params); +void dcn303_fpu_init_soc_bounding_box(struct bp_soc_bb_info bb_info); -void amdgpu_mmhub_ras_fini(struct amdgpu_device *adev) -{ - if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__MMHUB) && - adev->mmhub.ras_if) - amdgpu_ras_block_late_fini(adev, adev->mmhub.ras_if); -} +#endif /* __DCN303_FPU_H__*/ diff --git a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c index dae8e489c8cf..778c206f754d 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c @@ -647,7 +647,9 @@ enum gpio_result dal_ddc_set_config( void dal_ddc_close( struct ddc *ddc) { - dal_gpio_close(ddc->pin_clock); - dal_gpio_close(ddc->pin_data); + if (ddc != NULL) { + dal_gpio_close(ddc->pin_clock); + dal_gpio_close(ddc->pin_data); + } } diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h index 713f5558f5e1..9195dec294c2 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h @@ -154,6 +154,8 @@ struct hubbub_funcs { bool (*is_allow_self_refresh_enabled)(struct hubbub *hubbub); void (*allow_self_refresh_control)(struct hubbub *hubbub, bool allow); + bool (*verify_allow_pstate_change_high)(struct hubbub *hubbub); + void (*apply_DEDCN21_147_wa)(struct hubbub *hubbub); void (*force_wm_propagate_to_pipes)(struct hubbub *hubbub); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h index 2c031586f4e6..e45b7993c5c5 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h @@ -195,6 +195,7 @@ struct hubp_funcs { void (*hubp_set_flip_int)(struct hubp *hubp); + void (*hubp_wait_pipe_read_start)(struct hubp *hubp); }; #endif diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h b/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h index 48eac622c6a0..24af9d80b937 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h @@ -39,6 +39,7 @@ struct panel_cntl_backlight_registers { unsigned int BL_PWM_CNTL2; unsigned int BL_PWM_PERIOD_CNTL; unsigned int LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV; + unsigned int PANEL_PWRSEQ_REF_DIV2; }; struct panel_cntl_funcs { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h index 280c8764b636..8c2f190c4712 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h @@ -42,6 +42,7 @@ struct dce_hwseq_wa { bool DEGVIDCN21; bool disallow_self_refresh_during_multi_plane_transition; bool dp_hpo_and_otg_sequence; + bool wait_hubpret_read_start_during_mpo_transition; }; struct hwseq_wa_state { diff --git a/drivers/gpu/drm/amd/display/dc/inc/link_enc_cfg.h b/drivers/gpu/drm/amd/display/dc/inc/link_enc_cfg.h index c2f08514a1d9..c6f6baa6e677 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/link_enc_cfg.h +++ b/drivers/gpu/drm/amd/display/dc/inc/link_enc_cfg.h @@ -40,6 +40,11 @@ void link_enc_cfg_init( struct dc_state *state); /* + * Copies a link encoder assignment from another state. + */ +void link_enc_cfg_copy(const struct dc_state *src_ctx, struct dc_state *dst_ctx); + +/* * Algorithm for assigning available DIG link encoders to streams. * * Update link_enc_assignments table and link_enc_avail list accordingly in diff --git a/drivers/gpu/drm/amd/display/dc/link/Makefile b/drivers/gpu/drm/amd/display/dc/link/Makefile index c4a69ba7cb62..054c2a727eb2 100644 --- a/drivers/gpu/drm/amd/display/dc/link/Makefile +++ b/drivers/gpu/drm/amd/display/dc/link/Makefile @@ -23,7 +23,7 @@ # It abstracts the control and status of back end pipe such as DIO, HPO, DPIA, # PHY, HPD, DDC and etc). -LINK = link_hwss_dio.o link_hwss_dpia.o link_hwss_hpo_dp.o +LINK = link_hwss_dio.o link_hwss_dpia.o link_hwss_hpo_dp.o link_dp_trace.o AMD_DAL_LINK = $(addprefix $(AMDDALPATH)/dc/link/,$(LINK)) diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dp_trace.c b/drivers/gpu/drm/amd/display/dc/link/link_dp_trace.c new file mode 100644 index 000000000000..e7047391934b --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/link/link_dp_trace.c @@ -0,0 +1,146 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#include "dc_link.h" +#include "link_dp_trace.h" + +void dp_trace_init(struct dc_link *link) +{ + memset(&link->dp_trace, 0, sizeof(link->dp_trace)); + link->dp_trace.is_initialized = true; +} + +void dp_trace_reset(struct dc_link *link) +{ + memset(&link->dp_trace, 0, sizeof(link->dp_trace)); +} + +bool dc_dp_trace_is_initialized(struct dc_link *link) +{ + return link->dp_trace.is_initialized; +} + +void dp_trace_detect_lt_init(struct dc_link *link) +{ + memset(&link->dp_trace.detect_lt_trace, 0, sizeof(link->dp_trace.detect_lt_trace)); +} + +void dp_trace_commit_lt_init(struct dc_link *link) +{ + memset(&link->dp_trace.commit_lt_trace, 0, sizeof(link->dp_trace.commit_lt_trace)); +} + +void dp_trace_link_loss_increment(struct dc_link *link) +{ + link->dp_trace.link_loss_count++; +} + +void dp_trace_lt_fail_count_update(struct dc_link *link, + unsigned int fail_count, + bool in_detection) +{ + if (in_detection) + link->dp_trace.detect_lt_trace.counts.fail = fail_count; + else + link->dp_trace.commit_lt_trace.counts.fail = fail_count; +} + +void dp_trace_lt_total_count_increment(struct dc_link *link, + bool in_detection) +{ + if (in_detection) + link->dp_trace.detect_lt_trace.counts.total++; + else + link->dp_trace.commit_lt_trace.counts.total++; +} + +void dc_dp_trace_set_is_logged_flag(struct dc_link *link, + bool in_detection, + bool is_logged) +{ + if (in_detection) + link->dp_trace.detect_lt_trace.is_logged = is_logged; + else + link->dp_trace.commit_lt_trace.is_logged = is_logged; +} + +bool dc_dp_trace_is_logged(struct dc_link *link, + bool in_detection) +{ + if (in_detection) + return link->dp_trace.detect_lt_trace.is_logged; + else + return link->dp_trace.commit_lt_trace.is_logged; +} + +void dp_trace_lt_result_update(struct dc_link *link, + enum link_training_result result, + bool in_detection) +{ + if (in_detection) + link->dp_trace.detect_lt_trace.result = result; + else + link->dp_trace.commit_lt_trace.result = result; +} + +void dp_trace_set_lt_start_timestamp(struct dc_link *link, + bool in_detection) +{ + if (in_detection) + link->dp_trace.detect_lt_trace.timestamps.start = dm_get_timestamp(link->dc->ctx); + else + link->dp_trace.commit_lt_trace.timestamps.start = dm_get_timestamp(link->dc->ctx); +} + +void dp_trace_set_lt_end_timestamp(struct dc_link *link, + bool in_detection) +{ + if (in_detection) + link->dp_trace.detect_lt_trace.timestamps.end = dm_get_timestamp(link->dc->ctx); + else + link->dp_trace.commit_lt_trace.timestamps.end = dm_get_timestamp(link->dc->ctx); +} + +unsigned long long dc_dp_trace_get_lt_end_timestamp(struct dc_link *link, + bool in_detection) +{ + if (in_detection) + return link->dp_trace.detect_lt_trace.timestamps.end; + else + return link->dp_trace.commit_lt_trace.timestamps.end; +} + +struct dp_trace_lt_counts *dc_dp_trace_get_lt_counts(struct dc_link *link, + bool in_detection) +{ + if (in_detection) + return &link->dp_trace.detect_lt_trace.counts; + else + return &link->dp_trace.commit_lt_trace.counts; +} + +unsigned int dc_dp_trace_get_link_loss_count(struct dc_link *link) +{ + return link->dp_trace.link_loss_count; +} diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dp_trace.h b/drivers/gpu/drm/amd/display/dc/link/link_dp_trace.h new file mode 100644 index 000000000000..702f97c6ead0 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/link/link_dp_trace.h @@ -0,0 +1,57 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#ifndef __LINK_DP_TRACE_H__ +#define __LINK_DP_TRACE_H__ + +void dp_trace_init(struct dc_link *link); +void dp_trace_reset(struct dc_link *link); +bool dc_dp_trace_is_initialized(struct dc_link *link); +void dp_trace_detect_lt_init(struct dc_link *link); +void dp_trace_commit_lt_init(struct dc_link *link); +void dp_trace_link_loss_increment(struct dc_link *link); +void dp_trace_lt_fail_count_update(struct dc_link *link, + unsigned int fail_count, + bool in_detection); +void dp_trace_lt_total_count_increment(struct dc_link *link, + bool in_detection); +void dc_dp_trace_set_is_logged_flag(struct dc_link *link, + bool in_detection, + bool is_logged); +bool dc_dp_trace_is_logged(struct dc_link *link, + bool in_detection); +void dp_trace_lt_result_update(struct dc_link *link, + enum link_training_result result, + bool in_detection); +void dp_trace_set_lt_start_timestamp(struct dc_link *link, + bool in_detection); +void dp_trace_set_lt_end_timestamp(struct dc_link *link, + bool in_detection); +unsigned long long dc_dp_trace_get_lt_end_timestamp(struct dc_link *link, + bool in_detection); +struct dp_trace_lt_counts *dc_dp_trace_get_lt_counts(struct dc_link *link, + bool in_detection); +unsigned int dc_dp_trace_get_link_loss_count(struct dc_link *link); + +#endif /* __LINK_DP_TRACE_H__ */ diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h index 34fb148474cc..71214c7a60fc 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h @@ -46,10 +46,10 @@ /* Firmware versioning. */ #ifdef DMUB_EXPOSE_VERSION -#define DMUB_FW_VERSION_GIT_HASH 0x1422ef84 +#define DMUB_FW_VERSION_GIT_HASH 0x929554ba #define DMUB_FW_VERSION_MAJOR 0 #define DMUB_FW_VERSION_MINOR 0 -#define DMUB_FW_VERSION_REVISION 104 +#define DMUB_FW_VERSION_REVISION 108 #define DMUB_FW_VERSION_TEST 0 #define DMUB_FW_VERSION_VBIOS 0 #define DMUB_FW_VERSION_HOTFIX 0 @@ -1450,6 +1450,80 @@ enum dmub_cmd_mall_type { DMUB_CMD__MALL_ACTION_NO_DF_REQ = 3, }; +/** + * PHY Link rate for DP. + */ +enum phy_link_rate { + /** + * not supported. + */ + PHY_RATE_UNKNOWN = 0, + /** + * Rate_1 (RBR) - 1.62 Gbps/Lane + */ + PHY_RATE_162 = 1, + /** + * Rate_2 - 2.16 Gbps/Lane + */ + PHY_RATE_216 = 2, + /** + * Rate_3 - 2.43 Gbps/Lane + */ + PHY_RATE_243 = 3, + /** + * Rate_4 (HBR) - 2.70 Gbps/Lane + */ + PHY_RATE_270 = 4, + /** + * Rate_5 (RBR2)- 3.24 Gbps/Lane + */ + PHY_RATE_324 = 5, + /** + * Rate_6 - 4.32 Gbps/Lane + */ + PHY_RATE_432 = 6, + /** + * Rate_7 (HBR2)- 5.40 Gbps/Lane + */ + PHY_RATE_540 = 7, + /** + * Rate_8 (HBR3)- 8.10 Gbps/Lane + */ + PHY_RATE_810 = 8, + /** + * UHBR10 - 10.0 Gbps/Lane + */ + PHY_RATE_1000 = 9, + /** + * UHBR13.5 - 13.5 Gbps/Lane + */ + PHY_RATE_1350 = 10, + /** + * UHBR10 - 20.0 Gbps/Lane + */ + PHY_RATE_2000 = 11, +}; + +/** + * enum dmub_phy_fsm_state - PHY FSM states. + * PHY FSM state to transit to during PSR enable/disable. + */ +enum dmub_phy_fsm_state { + DMUB_PHY_FSM_POWER_UP_DEFAULT = 0, + DMUB_PHY_FSM_RESET, + DMUB_PHY_FSM_RESET_RELEASED, + DMUB_PHY_FSM_SRAM_LOAD_DONE, + DMUB_PHY_FSM_INITIALIZED, + DMUB_PHY_FSM_CALIBRATED, + DMUB_PHY_FSM_CALIBRATED_LP, + DMUB_PHY_FSM_CALIBRATED_PG, + DMUB_PHY_FSM_POWER_DOWN, + DMUB_PHY_FSM_PLL_EN, + DMUB_PHY_FSM_TX_EN, + DMUB_PHY_FSM_FAST_LP, +}; + + /** * Data passed from driver to FW in a DMUB_CMD__PSR_COPY_SETTINGS command. @@ -1560,10 +1634,14 @@ struct dmub_cmd_psr_copy_settings_data { * DSC enable status in driver */ uint8_t dsc_enable_status; + /* + * Use FSM state for PSR power up/down + */ + uint8_t use_phy_fsm; /** - * Explicit padding to 3 byte boundary. + * Explicit padding to 2 byte boundary. */ - uint8_t pad3[3]; + uint8_t pad3[2]; }; /** @@ -1694,9 +1772,16 @@ struct dmub_cmd_psr_force_static_data { */ uint8_t panel_inst; /** - * Explicit padding to 4 byte boundary. + * Phy state to enter. + * Values to use are defined in dmub_phy_fsm_state */ - uint8_t pad[2]; + uint8_t phy_fsm_state; + /** + * Phy rate for DP - RBR/HBR/HBR2/HBR3. + * Set this using enum phy_link_rate. + * This does not support HDMI/DP2 for now. + */ + uint8_t phy_rate; }; /** @@ -2373,6 +2458,9 @@ struct dmub_cmd_panel_cntl_data { uint32_t bl_pwm_ref_div1; /* in/out */ uint8_t is_backlight_on : 1; /* in/out */ uint8_t is_powered_on : 1; /* in/out */ + uint8_t padding[3]; + uint32_t bl_pwm_ref_div2; /* in/out */ + uint8_t reserved[4]; }; /** diff --git a/drivers/gpu/drm/amd/display/include/ddc_service_types.h b/drivers/gpu/drm/amd/display/include/ddc_service_types.h index f561e213bf98..f883d87791fe 100644 --- a/drivers/gpu/drm/amd/display/include/ddc_service_types.h +++ b/drivers/gpu/drm/amd/display/include/ddc_service_types.h @@ -64,6 +64,7 @@ enum ddc_service_type { struct display_sink_capability { /* dongle type (DP converter, CV smart dongle) */ enum display_dongle_type dongle_type; + bool is_dongle_type_one; /********************************************************** capabilities going INTO SINK DEVICE (stream capabilities) diff --git a/drivers/gpu/drm/amd/display/include/grph_object_ctrl_defs.h b/drivers/gpu/drm/amd/display/include/grph_object_ctrl_defs.h index dd974c428d23..bc96d0211360 100644 --- a/drivers/gpu/drm/amd/display/include/grph_object_ctrl_defs.h +++ b/drivers/gpu/drm/amd/display/include/grph_object_ctrl_defs.h @@ -419,28 +419,6 @@ struct integrated_info { struct edp_info edp2_info; }; -/** -* Power source ids. -*/ -enum power_source { - POWER_SOURCE_AC = 0, - POWER_SOURCE_DC, - POWER_SOURCE_LIMITED_POWER, - POWER_SOURCE_LIMITED_POWER_2, - POWER_SOURCE_MAX -}; - -struct bios_event_info { - uint32_t thermal_state; - uint32_t backlight_level; - enum power_source powerSource; - bool has_thermal_state_changed; - bool has_power_source_changed; - bool has_forced_mode_changed; - bool forced_mode; - bool backlight_changed; -}; - /* * DFS-bypass flag */ diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c index 1d63f1e8884c..89fbee568be4 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c @@ -507,6 +507,18 @@ int amdgpu_dpm_send_hbm_bad_pages_num(struct amdgpu_device *adev, uint32_t size) return ret; } +int amdgpu_dpm_send_hbm_bad_channel_flag(struct amdgpu_device *adev, uint32_t size) +{ + struct smu_context *smu = adev->powerplay.pp_handle; + int ret = 0; + + mutex_lock(&adev->pm.mutex); + ret = smu_send_hbm_bad_channel_flag(smu, size); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + int amdgpu_dpm_get_dpm_freq_range(struct amdgpu_device *adev, enum pp_clock_type type, uint32_t *min, @@ -631,11 +643,16 @@ int amdgpu_dpm_get_ecc_info(struct amdgpu_device *adev, void *umc_ecc) { struct smu_context *smu = adev->powerplay.pp_handle; + int ret = 0; if (!is_support_sw_smu(adev)) return -EOPNOTSUPP; - return smu_get_ecc_info(smu, umc_ecc); + mutex_lock(&adev->pm.mutex); + ret = smu_get_ecc_info(smu, umc_ecc); + mutex_unlock(&adev->pm.mutex); + + return ret; } struct amd_vce_state *amdgpu_dpm_get_vce_clock_state(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c index cbbbd4079249..5cd67ddf8495 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c @@ -1870,7 +1870,7 @@ static ssize_t amdgpu_set_smartshift_bias(struct device *dev, amdgpu_smartshift_bias = bias; r = count; - /* TODO: upadte bias level with SMU message */ + /* TODO: update bias level with SMU message */ out: pm_runtime_mark_last_busy(ddev->dev); diff --git a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h index ddfa55b59d02..3e78b3057277 100644 --- a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h +++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h @@ -412,6 +412,7 @@ void amdgpu_dpm_enable_jpeg(struct amdgpu_device *adev, bool enable); int amdgpu_pm_load_smu_firmware(struct amdgpu_device *adev, uint32_t *smu_version); int amdgpu_dpm_handle_passthrough_sbr(struct amdgpu_device *adev, bool enable); int amdgpu_dpm_send_hbm_bad_pages_num(struct amdgpu_device *adev, uint32_t size); +int amdgpu_dpm_send_hbm_bad_channel_flag(struct amdgpu_device *adev, uint32_t size); int amdgpu_dpm_get_dpm_freq_range(struct amdgpu_device *adev, enum pp_clock_type type, uint32_t *min, diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_hwmgr.c index 3f040be0d158..37324f2009ca 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_hwmgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_hwmgr.c @@ -3095,7 +3095,7 @@ static int vega10_get_pp_table_entry_callback_func(struct pp_hwmgr *hwmgr, void *pp_table, uint32_t classification_flag) { ATOM_Vega10_GFXCLK_Dependency_Record_V2 *patom_record_V2; - struct vega10_power_state *vega10_power_state = + struct vega10_power_state *vega10_ps = cast_phw_vega10_power_state(&(power_state->hardware)); struct vega10_performance_level *performance_level; ATOM_Vega10_State *state_entry = (ATOM_Vega10_State *)state; @@ -3145,17 +3145,17 @@ static int vega10_get_pp_table_entry_callback_func(struct pp_hwmgr *hwmgr, power_state->temperatures.min = 0; power_state->temperatures.max = 0; - performance_level = &(vega10_power_state->performance_levels - [vega10_power_state->performance_level_count++]); + performance_level = &(vega10_ps->performance_levels + [vega10_ps->performance_level_count++]); PP_ASSERT_WITH_CODE( - (vega10_power_state->performance_level_count < + (vega10_ps->performance_level_count < NUM_GFXCLK_DPM_LEVELS), "Performance levels exceeds SMC limit!", return -1); PP_ASSERT_WITH_CODE( - (vega10_power_state->performance_level_count <= + (vega10_ps->performance_level_count <= hwmgr->platform_descriptor. hardwareActivityPerformanceLevels), "Performance levels exceeds Driver limit!", @@ -3169,8 +3169,8 @@ static int vega10_get_pp_table_entry_callback_func(struct pp_hwmgr *hwmgr, performance_level->mem_clock = mclk_dep_table->entries [state_entry->ucMemClockIndexLow].ulMemClk; - performance_level = &(vega10_power_state->performance_levels - [vega10_power_state->performance_level_count++]); + performance_level = &(vega10_ps->performance_levels + [vega10_ps->performance_level_count++]); performance_level->soc_clock = socclk_dep_table->entries [state_entry->ucSocClockIndexHigh].ulClk; if (gfxclk_dep_table->ucRevId == 0) { @@ -3201,11 +3201,11 @@ static int vega10_get_pp_table_entry(struct pp_hwmgr *hwmgr, unsigned long entry_index, struct pp_power_state *state) { int result; - struct vega10_power_state *ps; + struct vega10_power_state *vega10_ps; state->hardware.magic = PhwVega10_Magic; - ps = cast_phw_vega10_power_state(&state->hardware); + vega10_ps = cast_phw_vega10_power_state(&state->hardware); result = vega10_get_powerplay_table_entry(hwmgr, entry_index, state, vega10_get_pp_table_entry_callback_func); @@ -3218,10 +3218,10 @@ static int vega10_get_pp_table_entry(struct pp_hwmgr *hwmgr, */ /* set DC compatible flag if this state supports DC */ if (!state->validation.disallowOnDC) - ps->dc_compatible = true; + vega10_ps->dc_compatible = true; - ps->uvd_clks.vclk = state->uvd_clocks.VCLK; - ps->uvd_clks.dclk = state->uvd_clocks.DCLK; + vega10_ps->uvd_clks.vclk = state->uvd_clocks.VCLK; + vega10_ps->uvd_clks.dclk = state->uvd_clocks.DCLK; return 0; } @@ -4823,33 +4823,41 @@ static int vega10_check_states_equal(struct pp_hwmgr *hwmgr, const struct pp_hw_power_state *pstate1, const struct pp_hw_power_state *pstate2, bool *equal) { - const struct vega10_power_state *psa; - const struct vega10_power_state *psb; + const struct vega10_power_state *vega10_psa; + const struct vega10_power_state *vega10_psb; int i; if (pstate1 == NULL || pstate2 == NULL || equal == NULL) return -EINVAL; - psa = cast_const_phw_vega10_power_state(pstate1); - psb = cast_const_phw_vega10_power_state(pstate2); - /* If the two states don't even have the same number of performance levels they cannot be the same state. */ - if (psa->performance_level_count != psb->performance_level_count) { + vega10_psa = cast_const_phw_vega10_power_state(pstate1); + vega10_psb = cast_const_phw_vega10_power_state(pstate2); + + /* If the two states don't even have the same number of performance levels + * they cannot be the same state. + */ + if (vega10_psa->performance_level_count != vega10_psb->performance_level_count) { *equal = false; return 0; } - for (i = 0; i < psa->performance_level_count; i++) { - if (!vega10_are_power_levels_equal(&(psa->performance_levels[i]), &(psb->performance_levels[i]))) { - /* If we have found even one performance level pair that is different the states are different. */ + for (i = 0; i < vega10_psa->performance_level_count; i++) { + if (!vega10_are_power_levels_equal(&(vega10_psa->performance_levels[i]), + &(vega10_psb->performance_levels[i]))) { + /* If we have found even one performance level pair + * that is different the states are different. + */ *equal = false; return 0; } } /* If all performance levels are the same try to use the UVD clocks to break the tie.*/ - *equal = ((psa->uvd_clks.vclk == psb->uvd_clks.vclk) && (psa->uvd_clks.dclk == psb->uvd_clks.dclk)); - *equal &= ((psa->vce_clks.evclk == psb->vce_clks.evclk) && (psa->vce_clks.ecclk == psb->vce_clks.ecclk)); - *equal &= (psa->sclk_threshold == psb->sclk_threshold); + *equal = ((vega10_psa->uvd_clks.vclk == vega10_psb->uvd_clks.vclk) && + (vega10_psa->uvd_clks.dclk == vega10_psb->uvd_clks.dclk)); + *equal &= ((vega10_psa->vce_clks.evclk == vega10_psb->vce_clks.evclk) && + (vega10_psa->vce_clks.ecclk == vega10_psb->vce_clks.ecclk)); + *equal &= (vega10_psa->sclk_threshold == vega10_psb->sclk_threshold); return 0; } @@ -5444,19 +5452,19 @@ static int vega10_get_performance_level(struct pp_hwmgr *hwmgr, const struct pp_ PHM_PerformanceLevelDesignation designation, uint32_t index, PHM_PerformanceLevel *level) { - const struct vega10_power_state *ps; + const struct vega10_power_state *vega10_ps; uint32_t i; if (level == NULL || hwmgr == NULL || state == NULL) return -EINVAL; - ps = cast_const_phw_vega10_power_state(state); + vega10_ps = cast_const_phw_vega10_power_state(state); - i = index > ps->performance_level_count - 1 ? - ps->performance_level_count - 1 : index; + i = index > vega10_ps->performance_level_count - 1 ? + vega10_ps->performance_level_count - 1 : index; - level->coreClock = ps->performance_levels[i].gfx_clock; - level->memory_clock = ps->performance_levels[i].mem_clock; + level->coreClock = vega10_ps->performance_levels[i].gfx_clock; + level->memory_clock = vega10_ps->performance_levels[i].mem_clock; return 0; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c index 7e79a67bb8ef..f1544755d8b4 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c @@ -3052,3 +3052,13 @@ int smu_send_hbm_bad_pages_num(struct smu_context *smu, uint32_t size) return ret; } + +int smu_send_hbm_bad_channel_flag(struct smu_context *smu, uint32_t size) +{ + int ret = 0; + + if (smu->ppt_funcs && smu->ppt_funcs->send_hbm_bad_channel_flag) + ret = smu->ppt_funcs->send_hbm_bad_channel_flag(smu, size); + + return ret; +} diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h index fbef3ab8d487..ef57b6089c69 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h @@ -1292,6 +1292,12 @@ struct pptable_funcs { * @set_config_table: Apply the input DriverSmuConfig table settings. */ int (*set_config_table)(struct smu_context *smu, struct config_table_setting *table); + + /** + * @sned_hbm_bad_channel_flag: message SMU to update bad channel info + * of SMUBUS table. + */ + int (*send_hbm_bad_channel_flag)(struct smu_context *smu, uint32_t size); }; typedef enum { @@ -1428,5 +1434,6 @@ int smu_get_ecc_info(struct smu_context *smu, void *umc_ecc); int smu_stb_collect_info(struct smu_context *smu, void *buff, uint32_t size); void amdgpu_smu_stb_debug_fs_init(struct amdgpu_device *adev); int smu_send_hbm_bad_pages_num(struct smu_context *smu, uint32_t size); +int smu_send_hbm_bad_channel_flag(struct smu_context *smu, uint32_t size); #endif #endif diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/aldebaran_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/aldebaran_ppsmc.h index ab66a4b9e438..0f498baf6838 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/aldebaran_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/aldebaran_ppsmc.h @@ -103,7 +103,8 @@ #define PPSMC_MSG_GfxDriverResetRecovery 0x42 #define PPSMC_MSG_BoardPowerCalibration 0x43 #define PPSMC_MSG_HeavySBR 0x45 -#define PPSMC_Message_Count 0x46 +#define PPSMC_MSG_SetBadHBMPagesRetiredFlagsPerChannel 0x46 +#define PPSMC_Message_Count 0x47 //PPSMC Reset Types diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu11_driver_if_sienna_cichlid.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu11_driver_if_sienna_cichlid.h index b253be602cc2..3e4a314ef925 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu11_driver_if_sienna_cichlid.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu11_driver_if_sienna_cichlid.h @@ -1481,9 +1481,67 @@ typedef struct { } SmuMetrics_V2_t; typedef struct { + uint32_t CurrClock[PPCLK_COUNT]; + + uint16_t AverageGfxclkFrequencyPreDs; + uint16_t AverageGfxclkFrequencyPostDs; + uint16_t AverageFclkFrequencyPreDs; + uint16_t AverageFclkFrequencyPostDs; + uint16_t AverageUclkFrequencyPreDs; + uint16_t AverageUclkFrequencyPostDs; + + + uint16_t AverageGfxActivity; + uint16_t AverageUclkActivity; + uint8_t CurrSocVoltageOffset; + uint8_t CurrGfxVoltageOffset; + uint8_t CurrMemVidOffset; + uint8_t Padding8; + uint16_t AverageSocketPower; + uint16_t TemperatureEdge; + uint16_t TemperatureHotspot; + uint16_t TemperatureMem; + uint16_t TemperatureVrGfx; + uint16_t TemperatureVrMem0; + uint16_t TemperatureVrMem1; + uint16_t TemperatureVrSoc; + uint16_t TemperatureLiquid0; + uint16_t TemperatureLiquid1; + uint16_t TemperaturePlx; + uint16_t Padding16; + uint32_t AccCnt; + uint8_t ThrottlingPercentage[THROTTLER_COUNT]; + + + uint8_t LinkDpmLevel; + uint8_t CurrFanPwm; + uint16_t CurrFanSpeed; + + //BACO metrics, PMFW-1721 + //metrics for D3hot entry/exit and driver ARM msgs + uint8_t D3HotEntryCountPerMode[D3HOT_SEQUENCE_COUNT]; + uint8_t D3HotExitCountPerMode[D3HOT_SEQUENCE_COUNT]; + uint8_t ArmMsgReceivedCountPerMode[D3HOT_SEQUENCE_COUNT]; + + //PMFW-4362 + uint32_t EnergyAccumulator; + uint16_t AverageVclk0Frequency; + uint16_t AverageDclk0Frequency; + uint16_t AverageVclk1Frequency; + uint16_t AverageDclk1Frequency; + uint16_t VcnUsagePercentage0; + uint16_t VcnUsagePercentage1; + uint8_t PcieRate; + uint8_t PcieWidth; + uint16_t AverageGfxclkFrequencyTarget; + +} SmuMetrics_V3_t; + +typedef struct { union { SmuMetrics_t SmuMetrics; SmuMetrics_V2_t SmuMetrics_V2; + SmuMetrics_V3_t SmuMetrics_V3; }; uint32_t Spare[1]; diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h index d787c3b9fc52..9f6f306eeca0 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h @@ -232,7 +232,8 @@ __SMU_DUMMY_MAP(ForceGfxVid), \ __SMU_DUMMY_MAP(Spare0), \ __SMU_DUMMY_MAP(UnforceGfxVid), \ - __SMU_DUMMY_MAP(HeavySBR), + __SMU_DUMMY_MAP(HeavySBR), \ + __SMU_DUMMY_MAP(SetBadHBMPagesRetiredFlagsPerChannel), #undef __SMU_DUMMY_MAP #define __SMU_DUMMY_MAP(type) SMU_MSG_##type diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c index d9d634ce9575..38f04836c82f 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c @@ -554,6 +554,11 @@ static uint32_t sienna_cichlid_get_throttler_status_locked(struct smu_context *s int i; if ((smu->adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 7)) && + (smu->smc_fw_version >= 0x3A4900)) { + for (i = 0; i < THROTTLER_COUNT; i++) + throttler_status |= + (metrics_ext->SmuMetrics_V3.ThrottlingPercentage[i] ? 1U << i : 0); + } else if ((smu->adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 7)) && (smu->smc_fw_version >= 0x3A4300)) { for (i = 0; i < THROTTLER_COUNT; i++) throttler_status |= @@ -574,11 +579,20 @@ static int sienna_cichlid_get_smu_metrics_data(struct smu_context *smu, &(((SmuMetricsExternal_t *)(smu_table->metrics_table))->SmuMetrics); SmuMetrics_V2_t *metrics_v2 = &(((SmuMetricsExternal_t *)(smu_table->metrics_table))->SmuMetrics_V2); - bool use_metrics_v2 = ((smu->adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 7)) && - (smu->smc_fw_version >= 0x3A4300)) ? true : false; + SmuMetrics_V3_t *metrics_v3 = + &(((SmuMetricsExternal_t *)(smu_table->metrics_table))->SmuMetrics_V3); + bool use_metrics_v2 = false; + bool use_metrics_v3 = false; uint16_t average_gfx_activity; int ret = 0; + if ((smu->adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 7)) && + (smu->smc_fw_version >= 0x3A4900)) + use_metrics_v3 = true; + else if ((smu->adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 7)) && + (smu->smc_fw_version >= 0x3A4300)) + use_metrics_v2 = true; + ret = smu_cmn_get_metrics_table(smu, NULL, false); @@ -587,96 +601,119 @@ static int sienna_cichlid_get_smu_metrics_data(struct smu_context *smu, switch (member) { case METRICS_CURR_GFXCLK: - *value = use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_GFXCLK] : + *value = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_GFXCLK] : + use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_GFXCLK] : metrics->CurrClock[PPCLK_GFXCLK]; break; case METRICS_CURR_SOCCLK: - *value = use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_SOCCLK] : + *value = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_SOCCLK] : + use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_SOCCLK] : metrics->CurrClock[PPCLK_SOCCLK]; break; case METRICS_CURR_UCLK: - *value = use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_UCLK] : + *value = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_UCLK] : + use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_UCLK] : metrics->CurrClock[PPCLK_UCLK]; break; case METRICS_CURR_VCLK: - *value = use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_VCLK_0] : + *value = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_VCLK_0] : + use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_VCLK_0] : metrics->CurrClock[PPCLK_VCLK_0]; break; case METRICS_CURR_VCLK1: - *value = use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_VCLK_1] : + *value = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_VCLK_1] : + use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_VCLK_1] : metrics->CurrClock[PPCLK_VCLK_1]; break; case METRICS_CURR_DCLK: - *value = use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_DCLK_0] : + *value = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_DCLK_0] : + use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_DCLK_0] : metrics->CurrClock[PPCLK_DCLK_0]; break; case METRICS_CURR_DCLK1: - *value = use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_DCLK_1] : + *value = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_DCLK_1] : + use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_DCLK_1] : metrics->CurrClock[PPCLK_DCLK_1]; break; case METRICS_CURR_DCEFCLK: - *value = use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_DCEFCLK] : + *value = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_DCEFCLK] : + use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_DCEFCLK] : metrics->CurrClock[PPCLK_DCEFCLK]; break; case METRICS_CURR_FCLK: - *value = use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_FCLK] : + *value = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_FCLK] : + use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_FCLK] : metrics->CurrClock[PPCLK_FCLK]; break; case METRICS_AVERAGE_GFXCLK: - average_gfx_activity = use_metrics_v2 ? metrics_v2->AverageGfxActivity : + average_gfx_activity = use_metrics_v3 ? metrics_v3->AverageGfxActivity : + use_metrics_v2 ? metrics_v2->AverageGfxActivity : metrics->AverageGfxActivity; if (average_gfx_activity <= SMU_11_0_7_GFX_BUSY_THRESHOLD) - *value = use_metrics_v2 ? metrics_v2->AverageGfxclkFrequencyPostDs : + *value = use_metrics_v3 ? metrics_v3->AverageGfxclkFrequencyPostDs : + use_metrics_v2 ? metrics_v2->AverageGfxclkFrequencyPostDs : metrics->AverageGfxclkFrequencyPostDs; else - *value = use_metrics_v2 ? metrics_v2->AverageGfxclkFrequencyPreDs : + *value = use_metrics_v3 ? metrics_v3->AverageGfxclkFrequencyPreDs : + use_metrics_v2 ? metrics_v2->AverageGfxclkFrequencyPreDs : metrics->AverageGfxclkFrequencyPreDs; break; case METRICS_AVERAGE_FCLK: - *value = use_metrics_v2 ? metrics_v2->AverageFclkFrequencyPostDs : + *value = use_metrics_v3 ? metrics_v3->AverageFclkFrequencyPostDs : + use_metrics_v2 ? metrics_v2->AverageFclkFrequencyPostDs : metrics->AverageFclkFrequencyPostDs; break; case METRICS_AVERAGE_UCLK: - *value = use_metrics_v2 ? metrics_v2->AverageUclkFrequencyPostDs : + *value = use_metrics_v3 ? metrics_v3->AverageUclkFrequencyPostDs : + use_metrics_v2 ? metrics_v2->AverageUclkFrequencyPostDs : metrics->AverageUclkFrequencyPostDs; break; case METRICS_AVERAGE_GFXACTIVITY: - *value = use_metrics_v2 ? metrics_v2->AverageGfxActivity : + *value = use_metrics_v3 ? metrics_v3->AverageGfxActivity : + use_metrics_v2 ? metrics_v2->AverageGfxActivity : metrics->AverageGfxActivity; break; case METRICS_AVERAGE_MEMACTIVITY: - *value = use_metrics_v2 ? metrics_v2->AverageUclkActivity : + *value = use_metrics_v3 ? metrics_v3->AverageUclkActivity : + use_metrics_v2 ? metrics_v2->AverageUclkActivity : metrics->AverageUclkActivity; break; case METRICS_AVERAGE_SOCKETPOWER: - *value = use_metrics_v2 ? metrics_v2->AverageSocketPower << 8 : + *value = use_metrics_v3 ? metrics_v3->AverageSocketPower << 8 : + use_metrics_v2 ? metrics_v2->AverageSocketPower << 8 : metrics->AverageSocketPower << 8; break; case METRICS_TEMPERATURE_EDGE: - *value = (use_metrics_v2 ? metrics_v2->TemperatureEdge : metrics->TemperatureEdge) * - SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; + *value = (use_metrics_v3 ? metrics_v3->TemperatureEdge : + use_metrics_v2 ? metrics_v2->TemperatureEdge : + metrics->TemperatureEdge) * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; break; case METRICS_TEMPERATURE_HOTSPOT: - *value = (use_metrics_v2 ? metrics_v2->TemperatureHotspot : metrics->TemperatureHotspot) * - SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; + *value = (use_metrics_v3 ? metrics_v3->TemperatureHotspot : + use_metrics_v2 ? metrics_v2->TemperatureHotspot : + metrics->TemperatureHotspot) * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; break; case METRICS_TEMPERATURE_MEM: - *value = (use_metrics_v2 ? metrics_v2->TemperatureMem : metrics->TemperatureMem) * - SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; + *value = (use_metrics_v3 ? metrics_v3->TemperatureMem : + use_metrics_v2 ? metrics_v2->TemperatureMem : + metrics->TemperatureMem) * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; break; case METRICS_TEMPERATURE_VRGFX: - *value = (use_metrics_v2 ? metrics_v2->TemperatureVrGfx : metrics->TemperatureVrGfx) * - SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; + *value = (use_metrics_v3 ? metrics_v3->TemperatureVrGfx : + use_metrics_v2 ? metrics_v2->TemperatureVrGfx : + metrics->TemperatureVrGfx) * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; break; case METRICS_TEMPERATURE_VRSOC: - *value = (use_metrics_v2 ? metrics_v2->TemperatureVrSoc : metrics->TemperatureVrSoc) * - SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; + *value = (use_metrics_v3 ? metrics_v3->TemperatureVrSoc : + use_metrics_v2 ? metrics_v2->TemperatureVrSoc : + metrics->TemperatureVrSoc) * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; break; case METRICS_THROTTLER_STATUS: *value = sienna_cichlid_get_throttler_status_locked(smu); break; case METRICS_CURR_FANSPEED: - *value = use_metrics_v2 ? metrics_v2->CurrFanSpeed : metrics->CurrFanSpeed; + *value = use_metrics_v3 ? metrics_v3->CurrFanSpeed : + use_metrics_v2 ? metrics_v2->CurrFanSpeed : metrics->CurrFanSpeed; break; default: *value = UINT_MAX; @@ -3656,12 +3693,22 @@ static ssize_t sienna_cichlid_get_gpu_metrics(struct smu_context *smu, &(metrics_external.SmuMetrics); SmuMetrics_V2_t *metrics_v2 = &(metrics_external.SmuMetrics_V2); + SmuMetrics_V3_t *metrics_v3 = + &(metrics_external.SmuMetrics_V3); struct amdgpu_device *adev = smu->adev; - bool use_metrics_v2 = ((adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 7)) && - (smu->smc_fw_version >= 0x3A4300)) ? true : false; + bool use_metrics_v2 = false; + bool use_metrics_v3 = false; uint16_t average_gfx_activity; int ret = 0; + if ((adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 7)) && + (smu->smc_fw_version >= 0x3A4900)) + use_metrics_v3 = true; + else if ((adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 7)) && + (smu->smc_fw_version >= 0x3A4300)) + use_metrics_v2 = true; + + ret = smu_cmn_get_metrics_table(smu, &metrics_external, true); @@ -3670,29 +3717,30 @@ static ssize_t sienna_cichlid_get_gpu_metrics(struct smu_context *smu, smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 3); - gpu_metrics->temperature_edge = + gpu_metrics->temperature_edge = use_metrics_v3 ? metrics_v3->TemperatureEdge : use_metrics_v2 ? metrics_v2->TemperatureEdge : metrics->TemperatureEdge; - gpu_metrics->temperature_hotspot = + gpu_metrics->temperature_hotspot = use_metrics_v3 ? metrics_v3->TemperatureHotspot : use_metrics_v2 ? metrics_v2->TemperatureHotspot : metrics->TemperatureHotspot; - gpu_metrics->temperature_mem = + gpu_metrics->temperature_mem = use_metrics_v3 ? metrics_v3->TemperatureMem : use_metrics_v2 ? metrics_v2->TemperatureMem : metrics->TemperatureMem; - gpu_metrics->temperature_vrgfx = + gpu_metrics->temperature_vrgfx = use_metrics_v3 ? metrics_v3->TemperatureVrGfx : use_metrics_v2 ? metrics_v2->TemperatureVrGfx : metrics->TemperatureVrGfx; - gpu_metrics->temperature_vrsoc = + gpu_metrics->temperature_vrsoc = use_metrics_v3 ? metrics_v3->TemperatureVrSoc : use_metrics_v2 ? metrics_v2->TemperatureVrSoc : metrics->TemperatureVrSoc; - gpu_metrics->temperature_vrmem = + gpu_metrics->temperature_vrmem = use_metrics_v3 ? metrics_v3->TemperatureVrMem0 : use_metrics_v2 ? metrics_v2->TemperatureVrMem0 : metrics->TemperatureVrMem0; - gpu_metrics->average_gfx_activity = + gpu_metrics->average_gfx_activity = use_metrics_v3 ? metrics_v3->AverageGfxActivity : use_metrics_v2 ? metrics_v2->AverageGfxActivity : metrics->AverageGfxActivity; - gpu_metrics->average_umc_activity = + gpu_metrics->average_umc_activity = use_metrics_v3 ? metrics_v3->AverageUclkActivity : use_metrics_v2 ? metrics_v2->AverageUclkActivity : metrics->AverageUclkActivity; - gpu_metrics->average_mm_activity = + gpu_metrics->average_mm_activity = use_metrics_v3 ? + (metrics_v3->VcnUsagePercentage0 + metrics_v3->VcnUsagePercentage1) / 2 : use_metrics_v2 ? metrics_v2->VcnActivityPercentage : metrics->VcnActivityPercentage; - gpu_metrics->average_socket_power = + gpu_metrics->average_socket_power = use_metrics_v3 ? metrics_v3->AverageSocketPower : use_metrics_v2 ? metrics_v2->AverageSocketPower : metrics->AverageSocketPower; - gpu_metrics->energy_accumulator = + gpu_metrics->energy_accumulator = use_metrics_v3 ? metrics_v3->EnergyAccumulator : use_metrics_v2 ? metrics_v2->EnergyAccumulator : metrics->EnergyAccumulator; if (metrics->CurrGfxVoltageOffset) @@ -3705,37 +3753,45 @@ static ssize_t sienna_cichlid_get_gpu_metrics(struct smu_context *smu, gpu_metrics->voltage_soc = (155000 - 625 * metrics->CurrSocVoltageOffset) / 100; - average_gfx_activity = use_metrics_v2 ? metrics_v2->AverageGfxActivity : metrics->AverageGfxActivity; + average_gfx_activity = use_metrics_v3 ? metrics_v3->AverageGfxActivity : + use_metrics_v2 ? metrics_v2->AverageGfxActivity : metrics->AverageGfxActivity; if (average_gfx_activity <= SMU_11_0_7_GFX_BUSY_THRESHOLD) gpu_metrics->average_gfxclk_frequency = - use_metrics_v2 ? metrics_v2->AverageGfxclkFrequencyPostDs : metrics->AverageGfxclkFrequencyPostDs; + use_metrics_v3 ? metrics_v3->AverageGfxclkFrequencyPostDs : + use_metrics_v2 ? metrics_v2->AverageGfxclkFrequencyPostDs : + metrics->AverageGfxclkFrequencyPostDs; else gpu_metrics->average_gfxclk_frequency = - use_metrics_v2 ? metrics_v2->AverageGfxclkFrequencyPreDs : metrics->AverageGfxclkFrequencyPreDs; + use_metrics_v3 ? metrics_v3->AverageGfxclkFrequencyPreDs : + use_metrics_v2 ? metrics_v2->AverageGfxclkFrequencyPreDs : + metrics->AverageGfxclkFrequencyPreDs; + gpu_metrics->average_uclk_frequency = - use_metrics_v2 ? metrics_v2->AverageUclkFrequencyPostDs : metrics->AverageUclkFrequencyPostDs; - gpu_metrics->average_vclk0_frequency = + use_metrics_v3 ? metrics_v3->AverageUclkFrequencyPostDs : + use_metrics_v2 ? metrics_v2->AverageUclkFrequencyPostDs : + metrics->AverageUclkFrequencyPostDs; + gpu_metrics->average_vclk0_frequency = use_metrics_v3 ? metrics_v3->AverageVclk0Frequency : use_metrics_v2 ? metrics_v2->AverageVclk0Frequency : metrics->AverageVclk0Frequency; - gpu_metrics->average_dclk0_frequency = + gpu_metrics->average_dclk0_frequency = use_metrics_v3 ? metrics_v3->AverageDclk0Frequency : use_metrics_v2 ? metrics_v2->AverageDclk0Frequency : metrics->AverageDclk0Frequency; - gpu_metrics->average_vclk1_frequency = + gpu_metrics->average_vclk1_frequency = use_metrics_v3 ? metrics_v3->AverageVclk1Frequency : use_metrics_v2 ? metrics_v2->AverageVclk1Frequency : metrics->AverageVclk1Frequency; - gpu_metrics->average_dclk1_frequency = + gpu_metrics->average_dclk1_frequency = use_metrics_v3 ? metrics_v3->AverageDclk1Frequency : use_metrics_v2 ? metrics_v2->AverageDclk1Frequency : metrics->AverageDclk1Frequency; - gpu_metrics->current_gfxclk = + gpu_metrics->current_gfxclk = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_GFXCLK] : use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_GFXCLK] : metrics->CurrClock[PPCLK_GFXCLK]; - gpu_metrics->current_socclk = + gpu_metrics->current_socclk = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_SOCCLK] : use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_SOCCLK] : metrics->CurrClock[PPCLK_SOCCLK]; - gpu_metrics->current_uclk = + gpu_metrics->current_uclk = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_UCLK] : use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_UCLK] : metrics->CurrClock[PPCLK_UCLK]; - gpu_metrics->current_vclk0 = + gpu_metrics->current_vclk0 = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_VCLK_0] : use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_VCLK_0] : metrics->CurrClock[PPCLK_VCLK_0]; - gpu_metrics->current_dclk0 = + gpu_metrics->current_dclk0 = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_DCLK_0] : use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_DCLK_0] : metrics->CurrClock[PPCLK_DCLK_0]; - gpu_metrics->current_vclk1 = + gpu_metrics->current_vclk1 = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_VCLK_1] : use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_VCLK_1] : metrics->CurrClock[PPCLK_VCLK_1]; - gpu_metrics->current_dclk1 = + gpu_metrics->current_dclk1 = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_DCLK_1] : use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_DCLK_1] : metrics->CurrClock[PPCLK_DCLK_1]; gpu_metrics->throttle_status = sienna_cichlid_get_throttler_status_locked(smu); @@ -3743,12 +3799,15 @@ static ssize_t sienna_cichlid_get_gpu_metrics(struct smu_context *smu, smu_cmn_get_indep_throttler_status(gpu_metrics->throttle_status, sienna_cichlid_throttler_map); - gpu_metrics->current_fan_speed = use_metrics_v2 ? metrics_v2->CurrFanSpeed : metrics->CurrFanSpeed; + gpu_metrics->current_fan_speed = use_metrics_v3 ? metrics_v3->CurrFanSpeed : + use_metrics_v2 ? metrics_v2->CurrFanSpeed : metrics->CurrFanSpeed; if (((adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 7)) && smu->smc_fw_version > 0x003A1E00) || ((adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 11)) && smu->smc_fw_version > 0x00410400)) { - gpu_metrics->pcie_link_width = use_metrics_v2 ? metrics_v2->PcieWidth : metrics->PcieWidth; - gpu_metrics->pcie_link_speed = link_speed[use_metrics_v2 ? metrics_v2->PcieRate : metrics->PcieRate]; + gpu_metrics->pcie_link_width = use_metrics_v3 ? metrics_v3->PcieWidth : + use_metrics_v2 ? metrics_v2->PcieWidth : metrics->PcieWidth; + gpu_metrics->pcie_link_speed = link_speed[use_metrics_v3 ? metrics_v3->PcieRate : + use_metrics_v2 ? metrics_v2->PcieRate : metrics->PcieRate]; } else { gpu_metrics->pcie_link_width = smu_v11_0_get_current_pcie_link_width(smu); diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c index 890acc4e2cb8..cd81f848d45a 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c @@ -82,6 +82,12 @@ */ #define SUPPORT_ECCTABLE_SMU_VERSION 0x00442a00 +/* + * SMU support BAD CHENNEL info MSG since version 68.51.00, + * use this to check ECCTALE feature whether support + */ +#define SUPPORT_BAD_CHANNEL_INFO_MSG_VERSION 0x00443300 + static const struct smu_temperature_range smu13_thermal_policy[] = { {-273150, 99000, 99000, -273150, 99000, 99000, -273150, 99000, 99000}, @@ -140,6 +146,7 @@ static const struct cmn2asic_msg_mapping aldebaran_message_map[SMU_MSG_MAX_COUNT MSG_MAP(GfxDriverResetRecovery, PPSMC_MSG_GfxDriverResetRecovery, 0), MSG_MAP(BoardPowerCalibration, PPSMC_MSG_BoardPowerCalibration, 0), MSG_MAP(HeavySBR, PPSMC_MSG_HeavySBR, 0), + MSG_MAP(SetBadHBMPagesRetiredFlagsPerChannel, PPSMC_MSG_SetBadHBMPagesRetiredFlagsPerChannel, 0), }; static const struct cmn2asic_mapping aldebaran_clk_map[SMU_CLK_COUNT] = { @@ -1997,6 +2004,41 @@ static int aldebaran_smu_send_hbm_bad_page_num(struct smu_context *smu, return ret; } +static int aldebaran_check_bad_channel_info_support(struct smu_context *smu) +{ + uint32_t if_version = 0xff, smu_version = 0xff; + int ret = 0; + + ret = smu_cmn_get_smc_version(smu, &if_version, &smu_version); + if (ret) { + /* return not support if failed get smu_version */ + ret = -EOPNOTSUPP; + } + + if (smu_version < SUPPORT_BAD_CHANNEL_INFO_MSG_VERSION) + ret = -EOPNOTSUPP; + + return ret; +} + +static int aldebaran_send_hbm_bad_channel_flag(struct smu_context *smu, + uint32_t size) +{ + int ret = 0; + + ret = aldebaran_check_bad_channel_info_support(smu); + if (ret) + return ret; + + /* message SMU to update the bad channel info on SMUBUS */ + ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetBadHBMPagesRetiredFlagsPerChannel, size, NULL); + if (ret) + dev_err(smu->adev->dev, "[%s] failed to message SMU to update HBM bad channel info\n", + __func__); + + return ret; +} + static const struct pptable_funcs aldebaran_ppt_funcs = { /* init dpm */ .get_allowed_feature_mask = aldebaran_get_allowed_feature_mask, @@ -2062,6 +2104,7 @@ static const struct pptable_funcs aldebaran_ppt_funcs = { .i2c_fini = aldebaran_i2c_control_fini, .send_hbm_bad_pages_num = aldebaran_smu_send_hbm_bad_page_num, .get_ecc_info = aldebaran_get_ecc_info, + .send_hbm_bad_channel_flag = aldebaran_send_hbm_bad_channel_flag, }; void aldebaran_set_ppt_funcs(struct smu_context *smu) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c index 4e557f4f7c4d..cf09e30bdfe0 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c @@ -752,6 +752,7 @@ int smu_v13_0_gfx_off_control(struct smu_context *smu, bool enable) case IP_VERSION(13, 0, 1): case IP_VERSION(13, 0, 3): case IP_VERSION(13, 0, 5): + case IP_VERSION(13, 0, 8): if (!(adev->pm.pp_feature & PP_GFXOFF_MASK)) return 0; if (enable) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c index 590a6ed12d54..b8d0c70ff668 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c @@ -160,15 +160,21 @@ static void __smu_cmn_reg_print_error(struct smu_context *smu, { struct amdgpu_device *adev = smu->adev; const char *message = smu_get_message_name(smu, msg); + u32 msg_idx, prm; switch (reg_c2pmsg_90) { case SMU_RESP_NONE: { - u32 msg_idx = RREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_66); - u32 prm = RREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_82); + if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 5)) { + msg_idx = RREG32_SOC15(MP1, 0, mmMP1_C2PMSG_2); + prm = RREG32_SOC15(MP1, 0, mmMP1_C2PMSG_34); + } else { + msg_idx = RREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_66); + prm = RREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_82); + } dev_err_ratelimited(adev->dev, "SMU: I'm not done with your previous command: SMN_C2PMSG_66:0x%08X SMN_C2PMSG_82:0x%08X", msg_idx, prm); - } + } break; case SMU_RESP_OK: /* The SMU executed the command. It completed with a @@ -297,7 +303,6 @@ int smu_cmn_send_msg_without_waiting(struct smu_context *smu, reg = __smu_cmn_poll_stat(smu); res = __smu_cmn_reg2errno(smu, reg); if (reg == SMU_RESP_NONE || - reg == SMU_RESP_BUSY_OTHER || res == -EREMOTEIO) goto Out; __smu_cmn_send_msg(smu, msg_index, param); @@ -391,7 +396,6 @@ int smu_cmn_send_smc_msg_with_param(struct smu_context *smu, reg = __smu_cmn_poll_stat(smu); res = __smu_cmn_reg2errno(smu, reg); if (reg == SMU_RESP_NONE || - reg == SMU_RESP_BUSY_OTHER || res == -EREMOTEIO) { __smu_cmn_reg_print_error(smu, reg, index, param, msg); goto Out; diff --git a/drivers/gpu/drm/exynos/exynos7_drm_decon.c b/drivers/gpu/drm/exynos/exynos7_drm_decon.c index 12571ac45540..c04264f70ad1 100644 --- a/drivers/gpu/drm/exynos/exynos7_drm_decon.c +++ b/drivers/gpu/drm/exynos/exynos7_drm_decon.c @@ -678,7 +678,6 @@ static int decon_probe(struct platform_device *pdev) struct device *dev = &pdev->dev; struct decon_context *ctx; struct device_node *i80_if_timings; - struct resource *res; int ret; if (!dev->of_node) @@ -728,16 +727,11 @@ static int decon_probe(struct platform_device *pdev) goto err_iounmap; } - res = platform_get_resource_byname(pdev, IORESOURCE_IRQ, - ctx->i80_if ? "lcd_sys" : "vsync"); - if (!res) { - dev_err(dev, "irq request failed.\n"); - ret = -ENXIO; + ret = platform_get_irq_byname(pdev, ctx->i80_if ? "lcd_sys" : "vsync"); + if (ret < 0) goto err_iounmap; - } - ret = devm_request_irq(dev, res->start, decon_irq_handler, - 0, "drm_decon", ctx); + ret = devm_request_irq(dev, ret, decon_irq_handler, 0, "drm_decon", ctx); if (ret) { dev_err(dev, "irq request failed.\n"); goto err_iounmap; diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c index b7d0a4aead0a..334862d422e2 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c +++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c @@ -1335,8 +1335,10 @@ static int exynos_dsi_register_te_irq(struct exynos_dsi *dsi, int ret; int te_gpio_irq; - dsi->te_gpio = devm_gpiod_get_optional(dsi->dev, "te", GPIOD_IN); - if (IS_ERR(dsi->te_gpio)) { + dsi->te_gpio = gpiod_get_optional(panel, "te", GPIOD_IN); + if (!dsi->te_gpio) { + return 0; + } else if (IS_ERR(dsi->te_gpio)) { dev_err(dsi->dev, "gpio request failed with %ld\n", PTR_ERR(dsi->te_gpio)); return PTR_ERR(dsi->te_gpio); diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c index 023f54ee61a8..0ee32e4b1e43 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c @@ -1267,7 +1267,6 @@ static int fimc_probe(struct platform_device *pdev) struct exynos_drm_ipp_formats *formats; struct device *dev = &pdev->dev; struct fimc_context *ctx; - struct resource *res; int ret; int i, j, num_limits, num_formats; @@ -1330,14 +1329,12 @@ static int fimc_probe(struct platform_device *pdev) return PTR_ERR(ctx->regs); /* resource irq */ - res = platform_get_resource(pdev, IORESOURCE_IRQ, 0); - if (!res) { - dev_err(dev, "failed to request irq resource.\n"); - return -ENOENT; - } + ret = platform_get_irq(pdev, 0); + if (ret < 0) + return ret; - ret = devm_request_irq(dev, res->start, fimc_irq_handler, - 0, dev_name(dev), ctx); + ret = devm_request_irq(dev, ret, fimc_irq_handler, + 0, dev_name(dev), ctx); if (ret < 0) { dev_err(dev, "failed to request irq.\n"); return ret; diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c index c735e53939d8..d5720fab510c 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c @@ -109,6 +109,7 @@ struct fimd_driver_data { unsigned int has_dp_clk:1; unsigned int has_hw_trigger:1; unsigned int has_trigger_per_te:1; + unsigned int has_bgr_support:1; }; static struct fimd_driver_data s3c64xx_fimd_driver_data = { @@ -138,6 +139,7 @@ static struct fimd_driver_data exynos4_fimd_driver_data = { .lcdblk_bypass_shift = 1, .has_shadowcon = 1, .has_vtsel = 1, + .has_bgr_support = 1, }; static struct fimd_driver_data exynos5_fimd_driver_data = { @@ -149,6 +151,7 @@ static struct fimd_driver_data exynos5_fimd_driver_data = { .has_vidoutcon = 1, .has_vtsel = 1, .has_dp_clk = 1, + .has_bgr_support = 1, }; static struct fimd_driver_data exynos5420_fimd_driver_data = { @@ -162,6 +165,7 @@ static struct fimd_driver_data exynos5420_fimd_driver_data = { .has_vtsel = 1, .has_mic_bypass = 1, .has_dp_clk = 1, + .has_bgr_support = 1, }; struct fimd_context { @@ -226,6 +230,18 @@ static const uint32_t fimd_formats[] = { DRM_FORMAT_ARGB8888, }; +static const uint32_t fimd_extended_formats[] = { + DRM_FORMAT_C8, + DRM_FORMAT_XRGB1555, + DRM_FORMAT_XBGR1555, + DRM_FORMAT_RGB565, + DRM_FORMAT_BGR565, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_XBGR8888, + DRM_FORMAT_ARGB8888, + DRM_FORMAT_ABGR8888, +}; + static const unsigned int capabilities[WINDOWS_NR] = { 0, EXYNOS_DRM_PLANE_CAP_WIN_BLEND | EXYNOS_DRM_PLANE_CAP_PIX_BLEND, @@ -673,21 +689,25 @@ static void fimd_win_set_pixfmt(struct fimd_context *ctx, unsigned int win, val |= WINCONx_BYTSWP; break; case DRM_FORMAT_XRGB1555: + case DRM_FORMAT_XBGR1555: val |= WINCON0_BPPMODE_16BPP_1555; val |= WINCONx_HAWSWP; val |= WINCONx_BURSTLEN_16WORD; break; case DRM_FORMAT_RGB565: + case DRM_FORMAT_BGR565: val |= WINCON0_BPPMODE_16BPP_565; val |= WINCONx_HAWSWP; val |= WINCONx_BURSTLEN_16WORD; break; case DRM_FORMAT_XRGB8888: + case DRM_FORMAT_XBGR8888: val |= WINCON0_BPPMODE_24BPP_888; val |= WINCONx_WSWP; val |= WINCONx_BURSTLEN_16WORD; break; case DRM_FORMAT_ARGB8888: + case DRM_FORMAT_ABGR8888: default: val |= WINCON1_BPPMODE_25BPP_A1888; val |= WINCONx_WSWP; @@ -695,6 +715,18 @@ static void fimd_win_set_pixfmt(struct fimd_context *ctx, unsigned int win, break; } + switch (pixel_format) { + case DRM_FORMAT_XBGR1555: + case DRM_FORMAT_XBGR8888: + case DRM_FORMAT_ABGR8888: + case DRM_FORMAT_BGR565: + writel(WIN_RGB_ORDER_REVERSE, ctx->regs + WIN_RGB_ORDER(win)); + break; + default: + writel(WIN_RGB_ORDER_FORWARD, ctx->regs + WIN_RGB_ORDER(win)); + break; + } + /* * Setting dma-burst to 16Word causes permanent tearing for very small * buffers, e.g. cursor buffer. Burst Mode switching which based on @@ -1074,8 +1106,14 @@ static int fimd_bind(struct device *dev, struct device *master, void *data) ctx->drm_dev = drm_dev; for (i = 0; i < WINDOWS_NR; i++) { - ctx->configs[i].pixel_formats = fimd_formats; - ctx->configs[i].num_pixel_formats = ARRAY_SIZE(fimd_formats); + if (ctx->driver_data->has_bgr_support) { + ctx->configs[i].pixel_formats = fimd_extended_formats; + ctx->configs[i].num_pixel_formats = ARRAY_SIZE(fimd_extended_formats); + } else { + ctx->configs[i].pixel_formats = fimd_formats; + ctx->configs[i].num_pixel_formats = ARRAY_SIZE(fimd_formats); + } + ctx->configs[i].zpos = i; ctx->configs[i].type = fimd_win_types[i]; ctx->configs[i].capabilities = capabilities[i]; @@ -1133,7 +1171,6 @@ static int fimd_probe(struct platform_device *pdev) struct device *dev = &pdev->dev; struct fimd_context *ctx; struct device_node *i80_if_timings; - struct resource *res; int ret; if (!dev->of_node) @@ -1206,15 +1243,11 @@ static int fimd_probe(struct platform_device *pdev) if (IS_ERR(ctx->regs)) return PTR_ERR(ctx->regs); - res = platform_get_resource_byname(pdev, IORESOURCE_IRQ, - ctx->i80_if ? "lcd_sys" : "vsync"); - if (!res) { - dev_err(dev, "irq request failed.\n"); - return -ENXIO; - } + ret = platform_get_irq_byname(pdev, ctx->i80_if ? "lcd_sys" : "vsync"); + if (ret < 0) + return ret; - ret = devm_request_irq(dev, res->start, fimd_irq_handler, - 0, "drm_fimd", ctx); + ret = devm_request_irq(dev, ret, fimd_irq_handler, 0, "drm_fimd", ctx); if (ret) { dev_err(dev, "irq request failed.\n"); return ret; diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c index 166a80262896..964dceb28c1e 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c @@ -1220,7 +1220,6 @@ static int gsc_probe(struct platform_device *pdev) struct gsc_driverdata *driver_data; struct exynos_drm_ipp_formats *formats; struct gsc_context *ctx; - struct resource *res; int num_formats, ret, i, j; ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL); @@ -1275,13 +1274,10 @@ static int gsc_probe(struct platform_device *pdev) return PTR_ERR(ctx->regs); /* resource irq */ - res = platform_get_resource(pdev, IORESOURCE_IRQ, 0); - if (!res) { - dev_err(dev, "failed to request irq resource.\n"); - return -ENOENT; - } + ctx->irq = platform_get_irq(pdev, 0); + if (ctx->irq < 0) + return ctx->irq; - ctx->irq = res->start; ret = devm_request_irq(dev, ctx->irq, gsc_irq_handler, 0, dev_name(dev), ctx); if (ret < 0) { diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c index 41c54f1f60bc..e5204be86093 100644 --- a/drivers/gpu/drm/exynos/exynos_mixer.c +++ b/drivers/gpu/drm/exynos/exynos_mixer.c @@ -809,19 +809,17 @@ static int mixer_resources_init(struct mixer_context *mixer_ctx) return -ENXIO; } - res = platform_get_resource(mixer_ctx->pdev, IORESOURCE_IRQ, 0); - if (res == NULL) { - dev_err(dev, "get interrupt resource failed.\n"); - return -ENXIO; - } + ret = platform_get_irq(mixer_ctx->pdev, 0); + if (ret < 0) + return ret; + mixer_ctx->irq = ret; - ret = devm_request_irq(dev, res->start, mixer_irq_handler, - 0, "drm_mixer", mixer_ctx); + ret = devm_request_irq(dev, mixer_ctx->irq, mixer_irq_handler, + 0, "drm_mixer", mixer_ctx); if (ret) { dev_err(dev, "request interrupt failed.\n"); return ret; } - mixer_ctx->irq = res->start; return 0; } diff --git a/drivers/gpu/drm/msm/adreno/a2xx.xml.h b/drivers/gpu/drm/msm/adreno/a2xx.xml.h index 4ff529576093..afa6023346c4 100644 --- a/drivers/gpu/drm/msm/adreno/a2xx.xml.h +++ b/drivers/gpu/drm/msm/adreno/a2xx.xml.h @@ -8,19 +8,19 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno.xml ( 594 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a2xx.xml ( 90810 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_common.xml ( 14386 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_pm4.xml ( 67699 bytes, from 2021-05-31 20:21:57) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a3xx.xml ( 84226 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a4xx.xml ( 112551 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a5xx.xml ( 150713 bytes, from 2021-06-10 22:34:02) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a6xx.xml ( 180049 bytes, from 2021-06-02 21:44:19) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a6xx_gmu.xml ( 11331 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/ocmem.xml ( 1773 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_control_regs.xml ( 6038 bytes, from 2021-05-27 20:22:36) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_pipe_regs.xml ( 2924 bytes, from 2021-05-27 20:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno.xml ( 594 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a2xx.xml ( 90810 bytes, from 2021-06-21 15:24:24) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_common.xml ( 14609 bytes, from 2021-11-24 23:05:10) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_pm4.xml ( 69086 bytes, from 2022-03-03 16:41:33) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a3xx.xml ( 84231 bytes, from 2021-11-24 23:05:10) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a4xx.xml ( 113358 bytes, from 2022-01-31 23:06:21) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a5xx.xml ( 149512 bytes, from 2022-01-31 23:06:21) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a6xx.xml ( 184954 bytes, from 2022-03-03 16:41:33) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a6xx_gmu.xml ( 11331 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/ocmem.xml ( 1773 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_control_regs.xml ( 6038 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_pipe_regs.xml ( 2924 bytes, from 2021-07-22 15:21:56) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) diff --git a/drivers/gpu/drm/msm/adreno/a2xx_gpu.c b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c index 22e8295a5e2b..6c9a747eb4ad 100644 --- a/drivers/gpu/drm/msm/adreno/a2xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c @@ -471,6 +471,7 @@ static u32 a2xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) static const struct adreno_gpu_funcs funcs = { .base = { .get_param = adreno_get_param, + .set_param = adreno_set_param, .hw_init = a2xx_hw_init, .pm_suspend = msm_gpu_pm_suspend, .pm_resume = msm_gpu_pm_resume, diff --git a/drivers/gpu/drm/msm/adreno/a3xx.xml.h b/drivers/gpu/drm/msm/adreno/a3xx.xml.h index e106b65abe26..520ae3f375a1 100644 --- a/drivers/gpu/drm/msm/adreno/a3xx.xml.h +++ b/drivers/gpu/drm/msm/adreno/a3xx.xml.h @@ -8,19 +8,19 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno.xml ( 594 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a2xx.xml ( 90810 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_common.xml ( 14386 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_pm4.xml ( 67699 bytes, from 2021-05-31 20:21:57) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a3xx.xml ( 84226 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a4xx.xml ( 112551 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a5xx.xml ( 150713 bytes, from 2021-06-10 22:34:02) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a6xx.xml ( 180049 bytes, from 2021-06-02 21:44:19) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a6xx_gmu.xml ( 11331 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/ocmem.xml ( 1773 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_control_regs.xml ( 6038 bytes, from 2021-05-27 20:22:36) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_pipe_regs.xml ( 2924 bytes, from 2021-05-27 20:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno.xml ( 594 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a2xx.xml ( 90810 bytes, from 2021-06-21 15:24:24) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_common.xml ( 14609 bytes, from 2021-11-24 23:05:10) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_pm4.xml ( 69086 bytes, from 2022-03-03 16:41:33) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a3xx.xml ( 84231 bytes, from 2021-11-24 23:05:10) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a4xx.xml ( 113358 bytes, from 2022-01-31 23:06:21) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a5xx.xml ( 149512 bytes, from 2022-01-31 23:06:21) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a6xx.xml ( 184954 bytes, from 2022-03-03 16:41:33) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a6xx_gmu.xml ( 11331 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/ocmem.xml ( 1773 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_control_regs.xml ( 6038 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_pipe_regs.xml ( 2924 bytes, from 2021-07-22 15:21:56) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) @@ -1466,7 +1466,7 @@ static inline uint32_t A3XX_RB_COPY_DEST_INFO_ENDIAN(enum adreno_rb_surface_endi #define REG_A3XX_RB_DEPTH_CONTROL 0x00002100 #define A3XX_RB_DEPTH_CONTROL_FRAG_WRITES_Z 0x00000001 -#define A3XX_RB_DEPTH_CONTROL_Z_ENABLE 0x00000002 +#define A3XX_RB_DEPTH_CONTROL_Z_TEST_ENABLE 0x00000002 #define A3XX_RB_DEPTH_CONTROL_Z_WRITE_ENABLE 0x00000004 #define A3XX_RB_DEPTH_CONTROL_EARLY_Z_DISABLE 0x00000008 #define A3XX_RB_DEPTH_CONTROL_ZFUNC__MASK 0x00000070 @@ -1476,7 +1476,7 @@ static inline uint32_t A3XX_RB_DEPTH_CONTROL_ZFUNC(enum adreno_compare_func val) return ((val) << A3XX_RB_DEPTH_CONTROL_ZFUNC__SHIFT) & A3XX_RB_DEPTH_CONTROL_ZFUNC__MASK; } #define A3XX_RB_DEPTH_CONTROL_Z_CLAMP_ENABLE 0x00000080 -#define A3XX_RB_DEPTH_CONTROL_Z_TEST_ENABLE 0x80000000 +#define A3XX_RB_DEPTH_CONTROL_Z_READ_ENABLE 0x80000000 #define REG_A3XX_RB_DEPTH_CLEAR 0x00002101 diff --git a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c index 2e481e2692ba..0ab0e1dd8bbb 100644 --- a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c @@ -486,6 +486,7 @@ static u32 a3xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) static const struct adreno_gpu_funcs funcs = { .base = { .get_param = adreno_get_param, + .set_param = adreno_set_param, .hw_init = a3xx_hw_init, .pm_suspend = msm_gpu_pm_suspend, .pm_resume = msm_gpu_pm_resume, diff --git a/drivers/gpu/drm/msm/adreno/a4xx.xml.h b/drivers/gpu/drm/msm/adreno/a4xx.xml.h index b26ede96ae2a..7e5c21015d10 100644 --- a/drivers/gpu/drm/msm/adreno/a4xx.xml.h +++ b/drivers/gpu/drm/msm/adreno/a4xx.xml.h @@ -8,21 +8,21 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno.xml ( 594 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a2xx.xml ( 90810 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_common.xml ( 14386 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_pm4.xml ( 67699 bytes, from 2021-05-31 20:21:57) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a3xx.xml ( 84226 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a4xx.xml ( 112551 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a5xx.xml ( 150713 bytes, from 2021-06-10 22:34:02) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a6xx.xml ( 180049 bytes, from 2021-06-02 21:44:19) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a6xx_gmu.xml ( 11331 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/ocmem.xml ( 1773 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_control_regs.xml ( 6038 bytes, from 2021-05-27 20:22:36) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_pipe_regs.xml ( 2924 bytes, from 2021-05-27 20:18:13) - -Copyright (C) 2013-2021 by the following authors: +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno.xml ( 594 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a2xx.xml ( 90810 bytes, from 2021-06-21 15:24:24) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_common.xml ( 14609 bytes, from 2021-11-24 23:05:10) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_pm4.xml ( 69086 bytes, from 2022-03-03 16:41:33) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a3xx.xml ( 84231 bytes, from 2021-11-24 23:05:10) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a4xx.xml ( 113358 bytes, from 2022-01-31 23:06:21) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a5xx.xml ( 149512 bytes, from 2022-01-31 23:06:21) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a6xx.xml ( 184954 bytes, from 2022-03-03 16:41:33) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a6xx_gmu.xml ( 11331 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/ocmem.xml ( 1773 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_control_regs.xml ( 6038 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_pipe_regs.xml ( 2924 bytes, from 2021-07-22 15:21:56) + +Copyright (C) 2013-2022 by the following authors: - Rob Clark <[email protected]> (robclark) - Ilia Mirkin <[email protected]> (imirkin) @@ -837,6 +837,7 @@ enum a4xx_tex_type { A4XX_TEX_2D = 1, A4XX_TEX_CUBE = 2, A4XX_TEX_3D = 3, + A4XX_TEX_BUFFER = 4, }; #define A4XX_CGC_HLSQ_EARLY_CYC__MASK 0x00700000 @@ -1360,7 +1361,7 @@ static inline uint32_t A4XX_RB_FS_OUTPUT_REG_MRT(uint32_t val) #define REG_A4XX_RB_DEPTH_CONTROL 0x00002101 #define A4XX_RB_DEPTH_CONTROL_FRAG_WRITES_Z 0x00000001 -#define A4XX_RB_DEPTH_CONTROL_Z_ENABLE 0x00000002 +#define A4XX_RB_DEPTH_CONTROL_Z_TEST_ENABLE 0x00000002 #define A4XX_RB_DEPTH_CONTROL_Z_WRITE_ENABLE 0x00000004 #define A4XX_RB_DEPTH_CONTROL_ZFUNC__MASK 0x00000070 #define A4XX_RB_DEPTH_CONTROL_ZFUNC__SHIFT 4 @@ -1371,7 +1372,7 @@ static inline uint32_t A4XX_RB_DEPTH_CONTROL_ZFUNC(enum adreno_compare_func val) #define A4XX_RB_DEPTH_CONTROL_Z_CLAMP_ENABLE 0x00000080 #define A4XX_RB_DEPTH_CONTROL_EARLY_Z_DISABLE 0x00010000 #define A4XX_RB_DEPTH_CONTROL_FORCE_FRAGZ_TO_FS 0x00020000 -#define A4XX_RB_DEPTH_CONTROL_Z_TEST_ENABLE 0x80000000 +#define A4XX_RB_DEPTH_CONTROL_Z_READ_ENABLE 0x80000000 #define REG_A4XX_RB_DEPTH_CLEAR 0x00002102 @@ -2541,6 +2542,8 @@ static inline uint32_t A4XX_SP_FS_MRT_REG_REGID(uint32_t val) return ((val) << A4XX_SP_FS_MRT_REG_REGID__SHIFT) & A4XX_SP_FS_MRT_REG_REGID__MASK; } #define A4XX_SP_FS_MRT_REG_HALF_PRECISION 0x00000100 +#define A4XX_SP_FS_MRT_REG_COLOR_SINT 0x00000400 +#define A4XX_SP_FS_MRT_REG_COLOR_UINT 0x00000800 #define A4XX_SP_FS_MRT_REG_MRTFORMAT__MASK 0x0003f000 #define A4XX_SP_FS_MRT_REG_MRTFORMAT__SHIFT 12 static inline uint32_t A4XX_SP_FS_MRT_REG_MRTFORMAT(enum a4xx_color_fmt val) @@ -2550,6 +2553,40 @@ static inline uint32_t A4XX_SP_FS_MRT_REG_MRTFORMAT(enum a4xx_color_fmt val) #define A4XX_SP_FS_MRT_REG_COLOR_SRGB 0x00040000 #define REG_A4XX_SP_CS_CTRL_REG0 0x00002300 +#define A4XX_SP_CS_CTRL_REG0_THREADMODE__MASK 0x00000001 +#define A4XX_SP_CS_CTRL_REG0_THREADMODE__SHIFT 0 +static inline uint32_t A4XX_SP_CS_CTRL_REG0_THREADMODE(enum a3xx_threadmode val) +{ + return ((val) << A4XX_SP_CS_CTRL_REG0_THREADMODE__SHIFT) & A4XX_SP_CS_CTRL_REG0_THREADMODE__MASK; +} +#define A4XX_SP_CS_CTRL_REG0_VARYING 0x00000002 +#define A4XX_SP_CS_CTRL_REG0_CACHEINVALID 0x00000004 +#define A4XX_SP_CS_CTRL_REG0_HALFREGFOOTPRINT__MASK 0x000003f0 +#define A4XX_SP_CS_CTRL_REG0_HALFREGFOOTPRINT__SHIFT 4 +static inline uint32_t A4XX_SP_CS_CTRL_REG0_HALFREGFOOTPRINT(uint32_t val) +{ + return ((val) << A4XX_SP_CS_CTRL_REG0_HALFREGFOOTPRINT__SHIFT) & A4XX_SP_CS_CTRL_REG0_HALFREGFOOTPRINT__MASK; +} +#define A4XX_SP_CS_CTRL_REG0_FULLREGFOOTPRINT__MASK 0x0000fc00 +#define A4XX_SP_CS_CTRL_REG0_FULLREGFOOTPRINT__SHIFT 10 +static inline uint32_t A4XX_SP_CS_CTRL_REG0_FULLREGFOOTPRINT(uint32_t val) +{ + return ((val) << A4XX_SP_CS_CTRL_REG0_FULLREGFOOTPRINT__SHIFT) & A4XX_SP_CS_CTRL_REG0_FULLREGFOOTPRINT__MASK; +} +#define A4XX_SP_CS_CTRL_REG0_INOUTREGOVERLAP__MASK 0x000c0000 +#define A4XX_SP_CS_CTRL_REG0_INOUTREGOVERLAP__SHIFT 18 +static inline uint32_t A4XX_SP_CS_CTRL_REG0_INOUTREGOVERLAP(uint32_t val) +{ + return ((val) << A4XX_SP_CS_CTRL_REG0_INOUTREGOVERLAP__SHIFT) & A4XX_SP_CS_CTRL_REG0_INOUTREGOVERLAP__MASK; +} +#define A4XX_SP_CS_CTRL_REG0_THREADSIZE__MASK 0x00100000 +#define A4XX_SP_CS_CTRL_REG0_THREADSIZE__SHIFT 20 +static inline uint32_t A4XX_SP_CS_CTRL_REG0_THREADSIZE(enum a3xx_threadsize val) +{ + return ((val) << A4XX_SP_CS_CTRL_REG0_THREADSIZE__SHIFT) & A4XX_SP_CS_CTRL_REG0_THREADSIZE__MASK; +} +#define A4XX_SP_CS_CTRL_REG0_SUPERTHREADMODE 0x00200000 +#define A4XX_SP_CS_CTRL_REG0_PIXLODENABLE 0x00400000 #define REG_A4XX_SP_CS_OBJ_OFFSET_REG 0x00002301 @@ -3795,12 +3832,18 @@ static inline uint32_t A4XX_HLSQ_CL_NDRANGE_5_SIZE_Z(uint32_t val) #define REG_A4XX_HLSQ_CL_NDRANGE_6 0x000023d3 #define REG_A4XX_HLSQ_CL_CONTROL_0 0x000023d4 -#define A4XX_HLSQ_CL_CONTROL_0_WGIDCONSTID__MASK 0x000000ff +#define A4XX_HLSQ_CL_CONTROL_0_WGIDCONSTID__MASK 0x00000fff #define A4XX_HLSQ_CL_CONTROL_0_WGIDCONSTID__SHIFT 0 static inline uint32_t A4XX_HLSQ_CL_CONTROL_0_WGIDCONSTID(uint32_t val) { return ((val) << A4XX_HLSQ_CL_CONTROL_0_WGIDCONSTID__SHIFT) & A4XX_HLSQ_CL_CONTROL_0_WGIDCONSTID__MASK; } +#define A4XX_HLSQ_CL_CONTROL_0_KERNELDIMCONSTID__MASK 0x00fff000 +#define A4XX_HLSQ_CL_CONTROL_0_KERNELDIMCONSTID__SHIFT 12 +static inline uint32_t A4XX_HLSQ_CL_CONTROL_0_KERNELDIMCONSTID(uint32_t val) +{ + return ((val) << A4XX_HLSQ_CL_CONTROL_0_KERNELDIMCONSTID__SHIFT) & A4XX_HLSQ_CL_CONTROL_0_KERNELDIMCONSTID__MASK; +} #define A4XX_HLSQ_CL_CONTROL_0_LOCALIDREGID__MASK 0xff000000 #define A4XX_HLSQ_CL_CONTROL_0_LOCALIDREGID__SHIFT 24 static inline uint32_t A4XX_HLSQ_CL_CONTROL_0_LOCALIDREGID(uint32_t val) @@ -3809,8 +3852,32 @@ static inline uint32_t A4XX_HLSQ_CL_CONTROL_0_LOCALIDREGID(uint32_t val) } #define REG_A4XX_HLSQ_CL_CONTROL_1 0x000023d5 +#define A4XX_HLSQ_CL_CONTROL_1_UNK0CONSTID__MASK 0x00000fff +#define A4XX_HLSQ_CL_CONTROL_1_UNK0CONSTID__SHIFT 0 +static inline uint32_t A4XX_HLSQ_CL_CONTROL_1_UNK0CONSTID(uint32_t val) +{ + return ((val) << A4XX_HLSQ_CL_CONTROL_1_UNK0CONSTID__SHIFT) & A4XX_HLSQ_CL_CONTROL_1_UNK0CONSTID__MASK; +} +#define A4XX_HLSQ_CL_CONTROL_1_WORKGROUPSIZECONSTID__MASK 0x00fff000 +#define A4XX_HLSQ_CL_CONTROL_1_WORKGROUPSIZECONSTID__SHIFT 12 +static inline uint32_t A4XX_HLSQ_CL_CONTROL_1_WORKGROUPSIZECONSTID(uint32_t val) +{ + return ((val) << A4XX_HLSQ_CL_CONTROL_1_WORKGROUPSIZECONSTID__SHIFT) & A4XX_HLSQ_CL_CONTROL_1_WORKGROUPSIZECONSTID__MASK; +} #define REG_A4XX_HLSQ_CL_KERNEL_CONST 0x000023d6 +#define A4XX_HLSQ_CL_KERNEL_CONST_UNK0CONSTID__MASK 0x00000fff +#define A4XX_HLSQ_CL_KERNEL_CONST_UNK0CONSTID__SHIFT 0 +static inline uint32_t A4XX_HLSQ_CL_KERNEL_CONST_UNK0CONSTID(uint32_t val) +{ + return ((val) << A4XX_HLSQ_CL_KERNEL_CONST_UNK0CONSTID__SHIFT) & A4XX_HLSQ_CL_KERNEL_CONST_UNK0CONSTID__MASK; +} +#define A4XX_HLSQ_CL_KERNEL_CONST_NUMWGCONSTID__MASK 0x00fff000 +#define A4XX_HLSQ_CL_KERNEL_CONST_NUMWGCONSTID__SHIFT 12 +static inline uint32_t A4XX_HLSQ_CL_KERNEL_CONST_NUMWGCONSTID(uint32_t val) +{ + return ((val) << A4XX_HLSQ_CL_KERNEL_CONST_NUMWGCONSTID__SHIFT) & A4XX_HLSQ_CL_KERNEL_CONST_NUMWGCONSTID__MASK; +} #define REG_A4XX_HLSQ_CL_KERNEL_GROUP_X 0x000023d7 @@ -3819,6 +3886,12 @@ static inline uint32_t A4XX_HLSQ_CL_CONTROL_0_LOCALIDREGID(uint32_t val) #define REG_A4XX_HLSQ_CL_KERNEL_GROUP_Z 0x000023d9 #define REG_A4XX_HLSQ_CL_WG_OFFSET 0x000023da +#define A4XX_HLSQ_CL_WG_OFFSET_UNK0CONSTID__MASK 0x00000fff +#define A4XX_HLSQ_CL_WG_OFFSET_UNK0CONSTID__SHIFT 0 +static inline uint32_t A4XX_HLSQ_CL_WG_OFFSET_UNK0CONSTID(uint32_t val) +{ + return ((val) << A4XX_HLSQ_CL_WG_OFFSET_UNK0CONSTID__SHIFT) & A4XX_HLSQ_CL_WG_OFFSET_UNK0CONSTID__MASK; +} #define REG_A4XX_HLSQ_UPDATE_CONTROL 0x000023db @@ -4130,7 +4203,7 @@ static inline uint32_t A4XX_TEX_CONST_0_FMT(enum a4xx_tex_fmt val) { return ((val) << A4XX_TEX_CONST_0_FMT__SHIFT) & A4XX_TEX_CONST_0_FMT__MASK; } -#define A4XX_TEX_CONST_0_TYPE__MASK 0x60000000 +#define A4XX_TEX_CONST_0_TYPE__MASK 0xe0000000 #define A4XX_TEX_CONST_0_TYPE__SHIFT 29 static inline uint32_t A4XX_TEX_CONST_0_TYPE(enum a4xx_tex_type val) { @@ -4158,6 +4231,7 @@ static inline uint32_t A4XX_TEX_CONST_2_PITCHALIGN(uint32_t val) { return ((val) << A4XX_TEX_CONST_2_PITCHALIGN__SHIFT) & A4XX_TEX_CONST_2_PITCHALIGN__MASK; } +#define A4XX_TEX_CONST_2_BUFFER 0x00000040 #define A4XX_TEX_CONST_2_PITCH__MASK 0x3ffffe00 #define A4XX_TEX_CONST_2_PITCH__SHIFT 9 static inline uint32_t A4XX_TEX_CONST_2_PITCH(uint32_t val) diff --git a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c index c5524d6e8705..0c6b2a6d0b4c 100644 --- a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c @@ -621,6 +621,7 @@ static u32 a4xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) static const struct adreno_gpu_funcs funcs = { .base = { .get_param = adreno_get_param, + .set_param = adreno_set_param, .hw_init = a4xx_hw_init, .pm_suspend = a4xx_pm_suspend, .pm_resume = a4xx_pm_resume, diff --git a/drivers/gpu/drm/msm/adreno/a5xx.xml.h b/drivers/gpu/drm/msm/adreno/a5xx.xml.h index 1e575ca1f270..2505b4e43ca0 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx.xml.h +++ b/drivers/gpu/drm/msm/adreno/a5xx.xml.h @@ -8,21 +8,21 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno.xml ( 594 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a2xx.xml ( 90810 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_common.xml ( 14386 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_pm4.xml ( 67699 bytes, from 2021-05-31 20:21:57) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a3xx.xml ( 84226 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a4xx.xml ( 112551 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a5xx.xml ( 150713 bytes, from 2021-06-10 22:34:02) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a6xx.xml ( 180049 bytes, from 2021-06-02 21:44:19) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a6xx_gmu.xml ( 11331 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/ocmem.xml ( 1773 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_control_regs.xml ( 6038 bytes, from 2021-05-27 20:22:36) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_pipe_regs.xml ( 2924 bytes, from 2021-05-27 20:18:13) - -Copyright (C) 2013-2021 by the following authors: +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno.xml ( 594 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a2xx.xml ( 90810 bytes, from 2021-06-21 15:24:24) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_common.xml ( 14609 bytes, from 2021-11-24 23:05:10) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_pm4.xml ( 69086 bytes, from 2022-03-03 16:41:33) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a3xx.xml ( 84231 bytes, from 2021-11-24 23:05:10) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a4xx.xml ( 113358 bytes, from 2022-01-31 23:06:21) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a5xx.xml ( 149512 bytes, from 2022-01-31 23:06:21) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a6xx.xml ( 184954 bytes, from 2022-03-03 16:41:33) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a6xx_gmu.xml ( 11331 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/ocmem.xml ( 1773 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_control_regs.xml ( 6038 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_pipe_regs.xml ( 2924 bytes, from 2021-07-22 15:21:56) + +Copyright (C) 2013-2022 by the following authors: - Rob Clark <[email protected]> (robclark) - Ilia Mirkin <[email protected]> (imirkin) @@ -872,6 +872,7 @@ enum a5xx_tex_type { A5XX_TEX_2D = 1, A5XX_TEX_CUBE = 2, A5XX_TEX_3D = 3, + A5XX_TEX_BUFFER = 4, }; #define A5XX_INT0_RBBM_GPU_IDLE 0x00000001 @@ -2830,7 +2831,9 @@ static inline uint32_t A5XX_GRAS_VS_CL_CNTL_CULL_MASK(uint32_t val) #define A5XX_GRAS_CNTL_IJ_PERSP_PIXEL 0x00000001 #define A5XX_GRAS_CNTL_IJ_PERSP_CENTROID 0x00000002 #define A5XX_GRAS_CNTL_IJ_PERSP_SAMPLE 0x00000004 -#define A5XX_GRAS_CNTL_SIZE 0x00000008 +#define A5XX_GRAS_CNTL_IJ_LINEAR_PIXEL 0x00000008 +#define A5XX_GRAS_CNTL_IJ_LINEAR_CENTROID 0x00000010 +#define A5XX_GRAS_CNTL_IJ_LINEAR_SAMPLE 0x00000020 #define A5XX_GRAS_CNTL_COORD_MASK__MASK 0x000003c0 #define A5XX_GRAS_CNTL_COORD_MASK__SHIFT 6 static inline uint32_t A5XX_GRAS_CNTL_COORD_MASK(uint32_t val) @@ -2911,7 +2914,12 @@ static inline uint32_t A5XX_GRAS_SU_CNTL_LINEHALFWIDTH(float val) return ((((int32_t)(val * 4.0))) << A5XX_GRAS_SU_CNTL_LINEHALFWIDTH__SHIFT) & A5XX_GRAS_SU_CNTL_LINEHALFWIDTH__MASK; } #define A5XX_GRAS_SU_CNTL_POLY_OFFSET 0x00000800 -#define A5XX_GRAS_SU_CNTL_MSAA_ENABLE 0x00002000 +#define A5XX_GRAS_SU_CNTL_LINE_MODE__MASK 0x00002000 +#define A5XX_GRAS_SU_CNTL_LINE_MODE__SHIFT 13 +static inline uint32_t A5XX_GRAS_SU_CNTL_LINE_MODE(enum a5xx_line_mode val) +{ + return ((val) << A5XX_GRAS_SU_CNTL_LINE_MODE__SHIFT) & A5XX_GRAS_SU_CNTL_LINE_MODE__MASK; +} #define REG_A5XX_GRAS_SU_POINT_MINMAX 0x0000e091 #define A5XX_GRAS_SU_POINT_MINMAX_MIN__MASK 0x0000ffff @@ -3166,7 +3174,9 @@ static inline uint32_t A5XX_RB_DEST_MSAA_CNTL_SAMPLES(enum a3xx_msaa_samples val #define A5XX_RB_RENDER_CONTROL0_IJ_PERSP_PIXEL 0x00000001 #define A5XX_RB_RENDER_CONTROL0_IJ_PERSP_CENTROID 0x00000002 #define A5XX_RB_RENDER_CONTROL0_IJ_PERSP_SAMPLE 0x00000004 -#define A5XX_RB_RENDER_CONTROL0_SIZE 0x00000008 +#define A5XX_RB_RENDER_CONTROL0_IJ_LINEAR_PIXEL 0x00000008 +#define A5XX_RB_RENDER_CONTROL0_IJ_LINEAR_CENTROID 0x00000010 +#define A5XX_RB_RENDER_CONTROL0_IJ_LINEAR_SAMPLE 0x00000020 #define A5XX_RB_RENDER_CONTROL0_COORD_MASK__MASK 0x000003c0 #define A5XX_RB_RENDER_CONTROL0_COORD_MASK__SHIFT 6 static inline uint32_t A5XX_RB_RENDER_CONTROL0_COORD_MASK(uint32_t val) @@ -3490,7 +3500,7 @@ static inline uint32_t A5XX_RB_BLEND_CNTL_SAMPLE_MASK(uint32_t val) #define A5XX_RB_DEPTH_PLANE_CNTL_UNK1 0x00000002 #define REG_A5XX_RB_DEPTH_CNTL 0x0000e1b1 -#define A5XX_RB_DEPTH_CNTL_Z_ENABLE 0x00000001 +#define A5XX_RB_DEPTH_CNTL_Z_TEST_ENABLE 0x00000001 #define A5XX_RB_DEPTH_CNTL_Z_WRITE_ENABLE 0x00000002 #define A5XX_RB_DEPTH_CNTL_ZFUNC__MASK 0x0000001c #define A5XX_RB_DEPTH_CNTL_ZFUNC__SHIFT 2 @@ -3498,7 +3508,7 @@ static inline uint32_t A5XX_RB_DEPTH_CNTL_ZFUNC(enum adreno_compare_func val) { return ((val) << A5XX_RB_DEPTH_CNTL_ZFUNC__SHIFT) & A5XX_RB_DEPTH_CNTL_ZFUNC__MASK; } -#define A5XX_RB_DEPTH_CNTL_Z_TEST_ENABLE 0x00000040 +#define A5XX_RB_DEPTH_CNTL_Z_READ_ENABLE 0x00000040 #define REG_A5XX_RB_DEPTH_BUFFER_INFO 0x0000e1b2 #define A5XX_RB_DEPTH_BUFFER_INFO_DEPTH_FORMAT__MASK 0x00000007 @@ -5034,6 +5044,7 @@ static inline uint32_t A5XX_RB_2D_SRC_INFO_COLOR_SWAP(enum a3xx_color_swap val) return ((val) << A5XX_RB_2D_SRC_INFO_COLOR_SWAP__SHIFT) & A5XX_RB_2D_SRC_INFO_COLOR_SWAP__MASK; } #define A5XX_RB_2D_SRC_INFO_FLAGS 0x00001000 +#define A5XX_RB_2D_SRC_INFO_SRGB 0x00002000 #define REG_A5XX_RB_2D_SRC_LO 0x00002108 @@ -5073,6 +5084,7 @@ static inline uint32_t A5XX_RB_2D_DST_INFO_COLOR_SWAP(enum a3xx_color_swap val) return ((val) << A5XX_RB_2D_DST_INFO_COLOR_SWAP__SHIFT) & A5XX_RB_2D_DST_INFO_COLOR_SWAP__MASK; } #define A5XX_RB_2D_DST_INFO_FLAGS 0x00001000 +#define A5XX_RB_2D_DST_INFO_SRGB 0x00002000 #define REG_A5XX_RB_2D_DST_LO 0x00002111 @@ -5138,6 +5150,7 @@ static inline uint32_t A5XX_GRAS_2D_SRC_INFO_COLOR_SWAP(enum a3xx_color_swap val return ((val) << A5XX_GRAS_2D_SRC_INFO_COLOR_SWAP__SHIFT) & A5XX_GRAS_2D_SRC_INFO_COLOR_SWAP__MASK; } #define A5XX_GRAS_2D_SRC_INFO_FLAGS 0x00001000 +#define A5XX_GRAS_2D_SRC_INFO_SRGB 0x00002000 #define REG_A5XX_GRAS_2D_DST_INFO 0x00002182 #define A5XX_GRAS_2D_DST_INFO_COLOR_FORMAT__MASK 0x000000ff @@ -5159,10 +5172,7 @@ static inline uint32_t A5XX_GRAS_2D_DST_INFO_COLOR_SWAP(enum a3xx_color_swap val return ((val) << A5XX_GRAS_2D_DST_INFO_COLOR_SWAP__SHIFT) & A5XX_GRAS_2D_DST_INFO_COLOR_SWAP__MASK; } #define A5XX_GRAS_2D_DST_INFO_FLAGS 0x00001000 - -#define REG_A5XX_UNKNOWN_2100 0x00002100 - -#define REG_A5XX_UNKNOWN_2180 0x00002180 +#define A5XX_GRAS_2D_DST_INFO_SRGB 0x00002000 #define REG_A5XX_UNKNOWN_2184 0x00002184 @@ -5316,7 +5326,7 @@ static inline uint32_t A5XX_TEX_CONST_1_HEIGHT(uint32_t val) } #define REG_A5XX_TEX_CONST_2 0x00000002 -#define A5XX_TEX_CONST_2_UNK4 0x00000010 +#define A5XX_TEX_CONST_2_BUFFER 0x00000010 #define A5XX_TEX_CONST_2_PITCHALIGN__MASK 0x0000000f #define A5XX_TEX_CONST_2_PITCHALIGN__SHIFT 0 static inline uint32_t A5XX_TEX_CONST_2_PITCHALIGN(uint32_t val) @@ -5329,13 +5339,12 @@ static inline uint32_t A5XX_TEX_CONST_2_PITCH(uint32_t val) { return ((val) << A5XX_TEX_CONST_2_PITCH__SHIFT) & A5XX_TEX_CONST_2_PITCH__MASK; } -#define A5XX_TEX_CONST_2_TYPE__MASK 0x60000000 +#define A5XX_TEX_CONST_2_TYPE__MASK 0xe0000000 #define A5XX_TEX_CONST_2_TYPE__SHIFT 29 static inline uint32_t A5XX_TEX_CONST_2_TYPE(enum a5xx_tex_type val) { return ((val) << A5XX_TEX_CONST_2_TYPE__SHIFT) & A5XX_TEX_CONST_2_TYPE__MASK; } -#define A5XX_TEX_CONST_2_UNK31 0x80000000 #define REG_A5XX_TEX_CONST_3 0x00000003 #define A5XX_TEX_CONST_3_ARRAY_PITCH__MASK 0x00003fff diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c index 3d28fcf841a6..407f50a15faa 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c @@ -1700,6 +1700,7 @@ static uint32_t a5xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) static const struct adreno_gpu_funcs funcs = { .base = { .get_param = adreno_get_param, + .set_param = adreno_set_param, .hw_init = a5xx_hw_init, .pm_suspend = a5xx_pm_suspend, .pm_resume = a5xx_pm_resume, diff --git a/drivers/gpu/drm/msm/adreno/a6xx.xml.h b/drivers/gpu/drm/msm/adreno/a6xx.xml.h index a3cb3d988ba4..b03e2c413ab1 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx.xml.h +++ b/drivers/gpu/drm/msm/adreno/a6xx.xml.h @@ -8,21 +8,21 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno.xml ( 594 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a2xx.xml ( 90810 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_common.xml ( 14386 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_pm4.xml ( 67699 bytes, from 2021-05-31 20:21:57) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a3xx.xml ( 84226 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a4xx.xml ( 112551 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a5xx.xml ( 150713 bytes, from 2021-06-10 22:34:02) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a6xx.xml ( 180049 bytes, from 2021-06-02 21:44:19) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a6xx_gmu.xml ( 11331 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/ocmem.xml ( 1773 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_control_regs.xml ( 6038 bytes, from 2021-05-27 20:22:36) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_pipe_regs.xml ( 2924 bytes, from 2021-05-27 20:18:13) - -Copyright (C) 2013-2021 by the following authors: +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno.xml ( 594 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a2xx.xml ( 90810 bytes, from 2021-06-21 15:24:24) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_common.xml ( 14609 bytes, from 2021-11-24 23:05:10) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_pm4.xml ( 69086 bytes, from 2022-03-03 16:41:33) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a3xx.xml ( 84231 bytes, from 2021-11-24 23:05:10) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a4xx.xml ( 113358 bytes, from 2022-01-31 23:06:21) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a5xx.xml ( 149512 bytes, from 2022-01-31 23:06:21) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a6xx.xml ( 184954 bytes, from 2022-03-03 16:41:33) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a6xx_gmu.xml ( 11331 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/ocmem.xml ( 1773 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_control_regs.xml ( 6038 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_pipe_regs.xml ( 2924 bytes, from 2021-07-22 15:21:56) + +Copyright (C) 2013-2022 by the following authors: - Rob Clark <[email protected]> (robclark) - Ilia Mirkin <[email protected]> (imirkin) @@ -132,9 +132,21 @@ enum a6xx_format { FMT6_G8R8B8R8_422_UNORM = 140, FMT6_R8G8R8B8_422_UNORM = 141, FMT6_R8_G8B8_2PLANE_420_UNORM = 142, + FMT6_NV21 = 143, FMT6_R8_G8_B8_3PLANE_420_UNORM = 144, FMT6_Z24_UNORM_S8_UINT_AS_R8G8B8A8 = 145, - FMT6_8_PLANE_UNORM = 148, + FMT6_NV12_Y = 148, + FMT6_NV12_UV = 149, + FMT6_NV12_VU = 150, + FMT6_NV12_4R = 151, + FMT6_NV12_4R_Y = 152, + FMT6_NV12_4R_UV = 153, + FMT6_P010 = 154, + FMT6_P010_Y = 155, + FMT6_P010_UV = 156, + FMT6_TP10 = 157, + FMT6_TP10_Y = 158, + FMT6_TP10_UV = 159, FMT6_Z24_UNORM_S8_UINT = 160, FMT6_ETC2_RG11_UNORM = 171, FMT6_ETC2_RG11_SNORM = 172, @@ -885,6 +897,44 @@ enum a6xx_ztest_mode { A6XX_EARLY_LRZ_LATE_Z = 2, }; +enum a6xx_sequenced_thread_dist { + DIST_SCREEN_COORD = 0, + DIST_ALL_TO_RB0 = 1, +}; + +enum a6xx_single_prim_mode { + NO_FLUSH = 0, + FLUSH_PER_OVERLAP_AND_OVERWRITE = 1, + FLUSH_PER_OVERLAP = 3, +}; + +enum a6xx_raster_mode { + TYPE_TILED = 0, + TYPE_WRITER = 1, +}; + +enum a6xx_raster_direction { + LR_TB = 0, + RL_TB = 1, + LR_BT = 2, + RB_BT = 3, +}; + +enum a6xx_render_mode { + RENDERING_PASS = 0, + BINNING_PASS = 1, +}; + +enum a6xx_buffers_location { + BUFFERS_IN_GMEM = 0, + BUFFERS_IN_SYSMEM = 3, +}; + +enum a6xx_fragcoord_sample_mode { + FRAGCOORD_CENTER = 0, + FRAGCOORD_SAMPLE = 3, +}; + enum a6xx_rotation { ROTATE_0 = 0, ROTATE_90 = 1, @@ -912,6 +962,10 @@ enum a6xx_threadsize { THREAD128 = 1, }; +enum a6xx_isam_mode { + ISAMMODE_GL = 2, +}; + enum a6xx_tex_filter { A6XX_TEX_NEAREST = 0, A6XX_TEX_LINEAR = 1, @@ -955,6 +1009,7 @@ enum a6xx_tex_type { A6XX_TEX_2D = 1, A6XX_TEX_CUBE = 2, A6XX_TEX_3D = 3, + A6XX_TEX_BUFFER = 4, }; #define A6XX_RBBM_INT_0_MASK_RBBM_GPU_IDLE 0x00000001 @@ -1292,6 +1347,10 @@ static inline uint32_t REG_A6XX_RBBM_PERFCTR_RBBM_SEL(uint32_t i0) { return 0x00 #define REG_A6XX_RBBM_PERFCTR_GPU_BUSY_MASKED 0x0000050b +#define REG_A6XX_RBBM_PERFCTR_SRAM_INIT_CMD 0x0000050e + +#define REG_A6XX_RBBM_PERFCTR_SRAM_INIT_STATUS 0x0000050f + #define REG_A6XX_RBBM_ISDB_CNT 0x00000533 #define REG_A6XX_RBBM_PRIMCTR_0_LO 0x00000540 @@ -1934,6 +1993,8 @@ static inline uint32_t A6XX_VBIF_TEST_BUS2_CTRL1_DATA_SEL(uint32_t val) #define REG_A6XX_GBIF_PWR_CNT_HIGH2 0x00003cd1 +#define REG_A6XX_VSC_DBG_ECO_CNTL 0x00000c00 + #define REG_A6XX_VSC_BIN_SIZE 0x00000c02 #define A6XX_VSC_BIN_SIZE_WIDTH__MASK 0x000000ff #define A6XX_VSC_BIN_SIZE_WIDTH__SHIFT 0 @@ -2076,9 +2137,9 @@ static inline uint32_t A6XX_GRAS_GS_CL_CNTL_CULL_MASK(uint32_t val) #define A6XX_GRAS_CNTL_IJ_PERSP_PIXEL 0x00000001 #define A6XX_GRAS_CNTL_IJ_PERSP_CENTROID 0x00000002 #define A6XX_GRAS_CNTL_IJ_PERSP_SAMPLE 0x00000004 -#define A6XX_GRAS_CNTL_SIZE 0x00000008 -#define A6XX_GRAS_CNTL_UNK4 0x00000010 -#define A6XX_GRAS_CNTL_SIZE_PERSAMP 0x00000020 +#define A6XX_GRAS_CNTL_IJ_LINEAR_PIXEL 0x00000008 +#define A6XX_GRAS_CNTL_IJ_LINEAR_CENTROID 0x00000010 +#define A6XX_GRAS_CNTL_IJ_LINEAR_SAMPLE 0x00000020 #define A6XX_GRAS_CNTL_COORD_MASK__MASK 0x000003c0 #define A6XX_GRAS_CNTL_COORD_MASK__SHIFT 6 static inline uint32_t A6XX_GRAS_CNTL_COORD_MASK(uint32_t val) @@ -2185,7 +2246,12 @@ static inline uint32_t A6XX_GRAS_SU_CNTL_UNK12(uint32_t val) { return ((val) << A6XX_GRAS_SU_CNTL_UNK12__SHIFT) & A6XX_GRAS_SU_CNTL_UNK12__MASK; } -#define A6XX_GRAS_SU_CNTL_MSAA_ENABLE 0x00002000 +#define A6XX_GRAS_SU_CNTL_LINE_MODE__MASK 0x00002000 +#define A6XX_GRAS_SU_CNTL_LINE_MODE__SHIFT 13 +static inline uint32_t A6XX_GRAS_SU_CNTL_LINE_MODE(enum a5xx_line_mode val) +{ + return ((val) << A6XX_GRAS_SU_CNTL_LINE_MODE__SHIFT) & A6XX_GRAS_SU_CNTL_LINE_MODE__MASK; +} #define A6XX_GRAS_SU_CNTL_UNK15__MASK 0x00018000 #define A6XX_GRAS_SU_CNTL_UNK15__SHIFT 15 static inline uint32_t A6XX_GRAS_SU_CNTL_UNK15(uint32_t val) @@ -2269,9 +2335,25 @@ static inline uint32_t A6XX_GRAS_SU_DEPTH_BUFFER_INFO_UNK3(uint32_t val) return ((val) << A6XX_GRAS_SU_DEPTH_BUFFER_INFO_UNK3__SHIFT) & A6XX_GRAS_SU_DEPTH_BUFFER_INFO_UNK3__MASK; } -#define REG_A6XX_GRAS_UNKNOWN_8099 0x00008099 +#define REG_A6XX_GRAS_SU_CONSERVATIVE_RAS_CNTL 0x00008099 +#define A6XX_GRAS_SU_CONSERVATIVE_RAS_CNTL_CONSERVATIVERASEN 0x00000001 +#define A6XX_GRAS_SU_CONSERVATIVE_RAS_CNTL_SHIFTAMOUNT__MASK 0x00000006 +#define A6XX_GRAS_SU_CONSERVATIVE_RAS_CNTL_SHIFTAMOUNT__SHIFT 1 +static inline uint32_t A6XX_GRAS_SU_CONSERVATIVE_RAS_CNTL_SHIFTAMOUNT(uint32_t val) +{ + return ((val) << A6XX_GRAS_SU_CONSERVATIVE_RAS_CNTL_SHIFTAMOUNT__SHIFT) & A6XX_GRAS_SU_CONSERVATIVE_RAS_CNTL_SHIFTAMOUNT__MASK; +} +#define A6XX_GRAS_SU_CONSERVATIVE_RAS_CNTL_INNERCONSERVATIVERASEN 0x00000008 +#define A6XX_GRAS_SU_CONSERVATIVE_RAS_CNTL_UNK4__MASK 0x00000030 +#define A6XX_GRAS_SU_CONSERVATIVE_RAS_CNTL_UNK4__SHIFT 4 +static inline uint32_t A6XX_GRAS_SU_CONSERVATIVE_RAS_CNTL_UNK4(uint32_t val) +{ + return ((val) << A6XX_GRAS_SU_CONSERVATIVE_RAS_CNTL_UNK4__SHIFT) & A6XX_GRAS_SU_CONSERVATIVE_RAS_CNTL_UNK4__MASK; +} -#define REG_A6XX_GRAS_UNKNOWN_809A 0x0000809a +#define REG_A6XX_GRAS_SU_PATH_RENDERING_CNTL 0x0000809a +#define A6XX_GRAS_SU_PATH_RENDERING_CNTL_UNK0 0x00000001 +#define A6XX_GRAS_SU_PATH_RENDERING_CNTL_LINELENGTHEN 0x00000002 #define REG_A6XX_GRAS_VS_LAYER_CNTL 0x0000809b #define A6XX_GRAS_VS_LAYER_CNTL_WRITES_LAYER 0x00000001 @@ -2285,7 +2367,44 @@ static inline uint32_t A6XX_GRAS_SU_DEPTH_BUFFER_INFO_UNK3(uint32_t val) #define A6XX_GRAS_DS_LAYER_CNTL_WRITES_LAYER 0x00000001 #define A6XX_GRAS_DS_LAYER_CNTL_WRITES_VIEW 0x00000002 -#define REG_A6XX_GRAS_UNKNOWN_80A0 0x000080a0 +#define REG_A6XX_GRAS_SC_CNTL 0x000080a0 +#define A6XX_GRAS_SC_CNTL_CCUSINGLECACHELINESIZE__MASK 0x00000007 +#define A6XX_GRAS_SC_CNTL_CCUSINGLECACHELINESIZE__SHIFT 0 +static inline uint32_t A6XX_GRAS_SC_CNTL_CCUSINGLECACHELINESIZE(uint32_t val) +{ + return ((val) << A6XX_GRAS_SC_CNTL_CCUSINGLECACHELINESIZE__SHIFT) & A6XX_GRAS_SC_CNTL_CCUSINGLECACHELINESIZE__MASK; +} +#define A6XX_GRAS_SC_CNTL_SINGLE_PRIM_MODE__MASK 0x00000018 +#define A6XX_GRAS_SC_CNTL_SINGLE_PRIM_MODE__SHIFT 3 +static inline uint32_t A6XX_GRAS_SC_CNTL_SINGLE_PRIM_MODE(enum a6xx_single_prim_mode val) +{ + return ((val) << A6XX_GRAS_SC_CNTL_SINGLE_PRIM_MODE__SHIFT) & A6XX_GRAS_SC_CNTL_SINGLE_PRIM_MODE__MASK; +} +#define A6XX_GRAS_SC_CNTL_RASTER_MODE__MASK 0x00000020 +#define A6XX_GRAS_SC_CNTL_RASTER_MODE__SHIFT 5 +static inline uint32_t A6XX_GRAS_SC_CNTL_RASTER_MODE(enum a6xx_raster_mode val) +{ + return ((val) << A6XX_GRAS_SC_CNTL_RASTER_MODE__SHIFT) & A6XX_GRAS_SC_CNTL_RASTER_MODE__MASK; +} +#define A6XX_GRAS_SC_CNTL_RASTER_DIRECTION__MASK 0x000000c0 +#define A6XX_GRAS_SC_CNTL_RASTER_DIRECTION__SHIFT 6 +static inline uint32_t A6XX_GRAS_SC_CNTL_RASTER_DIRECTION(enum a6xx_raster_direction val) +{ + return ((val) << A6XX_GRAS_SC_CNTL_RASTER_DIRECTION__SHIFT) & A6XX_GRAS_SC_CNTL_RASTER_DIRECTION__MASK; +} +#define A6XX_GRAS_SC_CNTL_SEQUENCED_THREAD_DISTRIBUTION__MASK 0x00000100 +#define A6XX_GRAS_SC_CNTL_SEQUENCED_THREAD_DISTRIBUTION__SHIFT 8 +static inline uint32_t A6XX_GRAS_SC_CNTL_SEQUENCED_THREAD_DISTRIBUTION(enum a6xx_sequenced_thread_dist val) +{ + return ((val) << A6XX_GRAS_SC_CNTL_SEQUENCED_THREAD_DISTRIBUTION__SHIFT) & A6XX_GRAS_SC_CNTL_SEQUENCED_THREAD_DISTRIBUTION__MASK; +} +#define A6XX_GRAS_SC_CNTL_UNK9__MASK 0x00000e00 +#define A6XX_GRAS_SC_CNTL_UNK9__SHIFT 9 +static inline uint32_t A6XX_GRAS_SC_CNTL_UNK9(uint32_t val) +{ + return ((val) << A6XX_GRAS_SC_CNTL_UNK9__SHIFT) & A6XX_GRAS_SC_CNTL_UNK9__MASK; +} +#define A6XX_GRAS_SC_CNTL_EARLYVIZOUTEN 0x00001000 #define REG_A6XX_GRAS_BIN_CONTROL 0x000080a1 #define A6XX_GRAS_BIN_CONTROL_BINW__MASK 0x0000003f @@ -2300,25 +2419,30 @@ static inline uint32_t A6XX_GRAS_BIN_CONTROL_BINH(uint32_t val) { return ((val >> 4) << A6XX_GRAS_BIN_CONTROL_BINH__SHIFT) & A6XX_GRAS_BIN_CONTROL_BINH__MASK; } -#define A6XX_GRAS_BIN_CONTROL_BINNING_PASS 0x00040000 -#define A6XX_GRAS_BIN_CONTROL_UNK19__MASK 0x00080000 -#define A6XX_GRAS_BIN_CONTROL_UNK19__SHIFT 19 -static inline uint32_t A6XX_GRAS_BIN_CONTROL_UNK19(uint32_t val) +#define A6XX_GRAS_BIN_CONTROL_RENDER_MODE__MASK 0x001c0000 +#define A6XX_GRAS_BIN_CONTROL_RENDER_MODE__SHIFT 18 +static inline uint32_t A6XX_GRAS_BIN_CONTROL_RENDER_MODE(enum a6xx_render_mode val) +{ + return ((val) << A6XX_GRAS_BIN_CONTROL_RENDER_MODE__SHIFT) & A6XX_GRAS_BIN_CONTROL_RENDER_MODE__MASK; +} +#define A6XX_GRAS_BIN_CONTROL_FORCE_LRZ_WRITE_DIS 0x00200000 +#define A6XX_GRAS_BIN_CONTROL_BUFFERS_LOCATION__MASK 0x00c00000 +#define A6XX_GRAS_BIN_CONTROL_BUFFERS_LOCATION__SHIFT 22 +static inline uint32_t A6XX_GRAS_BIN_CONTROL_BUFFERS_LOCATION(enum a6xx_buffers_location val) { - return ((val) << A6XX_GRAS_BIN_CONTROL_UNK19__SHIFT) & A6XX_GRAS_BIN_CONTROL_UNK19__MASK; + return ((val) << A6XX_GRAS_BIN_CONTROL_BUFFERS_LOCATION__SHIFT) & A6XX_GRAS_BIN_CONTROL_BUFFERS_LOCATION__MASK; } -#define A6XX_GRAS_BIN_CONTROL_UNK20__MASK 0x00100000 -#define A6XX_GRAS_BIN_CONTROL_UNK20__SHIFT 20 -static inline uint32_t A6XX_GRAS_BIN_CONTROL_UNK20(uint32_t val) +#define A6XX_GRAS_BIN_CONTROL_LRZ_FEEDBACK_ZMODE_MASK__MASK 0x07000000 +#define A6XX_GRAS_BIN_CONTROL_LRZ_FEEDBACK_ZMODE_MASK__SHIFT 24 +static inline uint32_t A6XX_GRAS_BIN_CONTROL_LRZ_FEEDBACK_ZMODE_MASK(uint32_t val) { - return ((val) << A6XX_GRAS_BIN_CONTROL_UNK20__SHIFT) & A6XX_GRAS_BIN_CONTROL_UNK20__MASK; + return ((val) << A6XX_GRAS_BIN_CONTROL_LRZ_FEEDBACK_ZMODE_MASK__SHIFT) & A6XX_GRAS_BIN_CONTROL_LRZ_FEEDBACK_ZMODE_MASK__MASK; } -#define A6XX_GRAS_BIN_CONTROL_USE_VIZ 0x00200000 -#define A6XX_GRAS_BIN_CONTROL_UNK22__MASK 0x0fc00000 -#define A6XX_GRAS_BIN_CONTROL_UNK22__SHIFT 22 -static inline uint32_t A6XX_GRAS_BIN_CONTROL_UNK22(uint32_t val) +#define A6XX_GRAS_BIN_CONTROL_UNK27__MASK 0x08000000 +#define A6XX_GRAS_BIN_CONTROL_UNK27__SHIFT 27 +static inline uint32_t A6XX_GRAS_BIN_CONTROL_UNK27(uint32_t val) { - return ((val) << A6XX_GRAS_BIN_CONTROL_UNK22__SHIFT) & A6XX_GRAS_BIN_CONTROL_UNK22__MASK; + return ((val) << A6XX_GRAS_BIN_CONTROL_UNK27__SHIFT) & A6XX_GRAS_BIN_CONTROL_UNK27__MASK; } #define REG_A6XX_GRAS_RAS_MSAA_CNTL 0x000080a2 @@ -2558,14 +2682,21 @@ static inline uint32_t A6XX_GRAS_LRZ_CNTL_UNK6(uint32_t val) return ((val) << A6XX_GRAS_LRZ_CNTL_UNK6__SHIFT) & A6XX_GRAS_LRZ_CNTL_UNK6__MASK; } -#define REG_A6XX_GRAS_UNKNOWN_8101 0x00008101 +#define REG_A6XX_GRAS_LRZ_PS_INPUT_CNTL 0x00008101 +#define A6XX_GRAS_LRZ_PS_INPUT_CNTL_SAMPLEID 0x00000001 +#define A6XX_GRAS_LRZ_PS_INPUT_CNTL_FRAGCOORDSAMPLEMODE__MASK 0x00000006 +#define A6XX_GRAS_LRZ_PS_INPUT_CNTL_FRAGCOORDSAMPLEMODE__SHIFT 1 +static inline uint32_t A6XX_GRAS_LRZ_PS_INPUT_CNTL_FRAGCOORDSAMPLEMODE(enum a6xx_fragcoord_sample_mode val) +{ + return ((val) << A6XX_GRAS_LRZ_PS_INPUT_CNTL_FRAGCOORDSAMPLEMODE__SHIFT) & A6XX_GRAS_LRZ_PS_INPUT_CNTL_FRAGCOORDSAMPLEMODE__MASK; +} -#define REG_A6XX_GRAS_2D_BLIT_INFO 0x00008102 -#define A6XX_GRAS_2D_BLIT_INFO_COLOR_FORMAT__MASK 0x000000ff -#define A6XX_GRAS_2D_BLIT_INFO_COLOR_FORMAT__SHIFT 0 -static inline uint32_t A6XX_GRAS_2D_BLIT_INFO_COLOR_FORMAT(enum a6xx_format val) +#define REG_A6XX_GRAS_LRZ_MRT_BUF_INFO_0 0x00008102 +#define A6XX_GRAS_LRZ_MRT_BUF_INFO_0_COLOR_FORMAT__MASK 0x000000ff +#define A6XX_GRAS_LRZ_MRT_BUF_INFO_0_COLOR_FORMAT__SHIFT 0 +static inline uint32_t A6XX_GRAS_LRZ_MRT_BUF_INFO_0_COLOR_FORMAT(enum a6xx_format val) { - return ((val) << A6XX_GRAS_2D_BLIT_INFO_COLOR_FORMAT__SHIFT) & A6XX_GRAS_2D_BLIT_INFO_COLOR_FORMAT__MASK; + return ((val) << A6XX_GRAS_LRZ_MRT_BUF_INFO_0_COLOR_FORMAT__SHIFT) & A6XX_GRAS_LRZ_MRT_BUF_INFO_0_COLOR_FORMAT__MASK; } #define REG_A6XX_GRAS_LRZ_BUFFER_BASE 0x00008103 @@ -2630,11 +2761,12 @@ static inline uint32_t A6XX_GRAS_2D_BLIT_CNTL_ROTATE(enum a6xx_rotation val) { return ((val) << A6XX_GRAS_2D_BLIT_CNTL_ROTATE__SHIFT) & A6XX_GRAS_2D_BLIT_CNTL_ROTATE__MASK; } -#define A6XX_GRAS_2D_BLIT_CNTL_UNK3__MASK 0x00000078 -#define A6XX_GRAS_2D_BLIT_CNTL_UNK3__SHIFT 3 -static inline uint32_t A6XX_GRAS_2D_BLIT_CNTL_UNK3(uint32_t val) +#define A6XX_GRAS_2D_BLIT_CNTL_OVERWRITEEN 0x00000008 +#define A6XX_GRAS_2D_BLIT_CNTL_UNK4__MASK 0x00000070 +#define A6XX_GRAS_2D_BLIT_CNTL_UNK4__SHIFT 4 +static inline uint32_t A6XX_GRAS_2D_BLIT_CNTL_UNK4(uint32_t val) { - return ((val) << A6XX_GRAS_2D_BLIT_CNTL_UNK3__SHIFT) & A6XX_GRAS_2D_BLIT_CNTL_UNK3__MASK; + return ((val) << A6XX_GRAS_2D_BLIT_CNTL_UNK4__SHIFT) & A6XX_GRAS_2D_BLIT_CNTL_UNK4__MASK; } #define A6XX_GRAS_2D_BLIT_CNTL_SOLID_COLOR 0x00000080 #define A6XX_GRAS_2D_BLIT_CNTL_COLOR_FORMAT__MASK 0x0000ff00 @@ -2663,11 +2795,11 @@ static inline uint32_t A6XX_GRAS_2D_BLIT_CNTL_IFMT(enum a6xx_2d_ifmt val) { return ((val) << A6XX_GRAS_2D_BLIT_CNTL_IFMT__SHIFT) & A6XX_GRAS_2D_BLIT_CNTL_IFMT__MASK; } -#define A6XX_GRAS_2D_BLIT_CNTL_UNK29__MASK 0x20000000 -#define A6XX_GRAS_2D_BLIT_CNTL_UNK29__SHIFT 29 -static inline uint32_t A6XX_GRAS_2D_BLIT_CNTL_UNK29(uint32_t val) +#define A6XX_GRAS_2D_BLIT_CNTL_RASTER_MODE__MASK 0x20000000 +#define A6XX_GRAS_2D_BLIT_CNTL_RASTER_MODE__SHIFT 29 +static inline uint32_t A6XX_GRAS_2D_BLIT_CNTL_RASTER_MODE(enum a6xx_raster_mode val) { - return ((val) << A6XX_GRAS_2D_BLIT_CNTL_UNK29__SHIFT) & A6XX_GRAS_2D_BLIT_CNTL_UNK29__MASK; + return ((val) << A6XX_GRAS_2D_BLIT_CNTL_RASTER_MODE__SHIFT) & A6XX_GRAS_2D_BLIT_CNTL_RASTER_MODE__MASK; } #define REG_A6XX_GRAS_2D_SRC_TL_X 0x00008401 @@ -2740,7 +2872,9 @@ static inline uint32_t A6XX_GRAS_2D_RESOLVE_CNTL_2_Y(uint32_t val) return ((val) << A6XX_GRAS_2D_RESOLVE_CNTL_2_Y__SHIFT) & A6XX_GRAS_2D_RESOLVE_CNTL_2_Y__MASK; } -#define REG_A6XX_GRAS_UNKNOWN_8600 0x00008600 +#define REG_A6XX_GRAS_DBG_ECO_CNTL 0x00008600 +#define A6XX_GRAS_DBG_ECO_CNTL_UNK7 0x00000080 +#define A6XX_GRAS_DBG_ECO_CNTL_LRZCACHELOCKDIS 0x00000800 #define REG_A6XX_GRAS_ADDR_MODE_CNTL 0x00008601 @@ -2763,43 +2897,55 @@ static inline uint32_t A6XX_RB_BIN_CONTROL_BINH(uint32_t val) { return ((val >> 4) << A6XX_RB_BIN_CONTROL_BINH__SHIFT) & A6XX_RB_BIN_CONTROL_BINH__MASK; } -#define A6XX_RB_BIN_CONTROL_BINNING_PASS 0x00040000 -#define A6XX_RB_BIN_CONTROL_UNK19__MASK 0x00080000 -#define A6XX_RB_BIN_CONTROL_UNK19__SHIFT 19 -static inline uint32_t A6XX_RB_BIN_CONTROL_UNK19(uint32_t val) +#define A6XX_RB_BIN_CONTROL_RENDER_MODE__MASK 0x001c0000 +#define A6XX_RB_BIN_CONTROL_RENDER_MODE__SHIFT 18 +static inline uint32_t A6XX_RB_BIN_CONTROL_RENDER_MODE(enum a6xx_render_mode val) { - return ((val) << A6XX_RB_BIN_CONTROL_UNK19__SHIFT) & A6XX_RB_BIN_CONTROL_UNK19__MASK; + return ((val) << A6XX_RB_BIN_CONTROL_RENDER_MODE__SHIFT) & A6XX_RB_BIN_CONTROL_RENDER_MODE__MASK; } -#define A6XX_RB_BIN_CONTROL_UNK20__MASK 0x00100000 -#define A6XX_RB_BIN_CONTROL_UNK20__SHIFT 20 -static inline uint32_t A6XX_RB_BIN_CONTROL_UNK20(uint32_t val) +#define A6XX_RB_BIN_CONTROL_FORCE_LRZ_WRITE_DIS 0x00200000 +#define A6XX_RB_BIN_CONTROL_BUFFERS_LOCATION__MASK 0x00c00000 +#define A6XX_RB_BIN_CONTROL_BUFFERS_LOCATION__SHIFT 22 +static inline uint32_t A6XX_RB_BIN_CONTROL_BUFFERS_LOCATION(enum a6xx_buffers_location val) { - return ((val) << A6XX_RB_BIN_CONTROL_UNK20__SHIFT) & A6XX_RB_BIN_CONTROL_UNK20__MASK; + return ((val) << A6XX_RB_BIN_CONTROL_BUFFERS_LOCATION__SHIFT) & A6XX_RB_BIN_CONTROL_BUFFERS_LOCATION__MASK; } -#define A6XX_RB_BIN_CONTROL_USE_VIZ 0x00200000 -#define A6XX_RB_BIN_CONTROL_UNK22__MASK 0x07c00000 -#define A6XX_RB_BIN_CONTROL_UNK22__SHIFT 22 -static inline uint32_t A6XX_RB_BIN_CONTROL_UNK22(uint32_t val) +#define A6XX_RB_BIN_CONTROL_LRZ_FEEDBACK_ZMODE_MASK__MASK 0x07000000 +#define A6XX_RB_BIN_CONTROL_LRZ_FEEDBACK_ZMODE_MASK__SHIFT 24 +static inline uint32_t A6XX_RB_BIN_CONTROL_LRZ_FEEDBACK_ZMODE_MASK(uint32_t val) { - return ((val) << A6XX_RB_BIN_CONTROL_UNK22__SHIFT) & A6XX_RB_BIN_CONTROL_UNK22__MASK; + return ((val) << A6XX_RB_BIN_CONTROL_LRZ_FEEDBACK_ZMODE_MASK__SHIFT) & A6XX_RB_BIN_CONTROL_LRZ_FEEDBACK_ZMODE_MASK__MASK; } #define REG_A6XX_RB_RENDER_CNTL 0x00008801 -#define A6XX_RB_RENDER_CNTL_UNK3 0x00000008 -#define A6XX_RB_RENDER_CNTL_UNK4 0x00000010 -#define A6XX_RB_RENDER_CNTL_UNK5__MASK 0x00000060 -#define A6XX_RB_RENDER_CNTL_UNK5__SHIFT 5 -static inline uint32_t A6XX_RB_RENDER_CNTL_UNK5(uint32_t val) +#define A6XX_RB_RENDER_CNTL_CCUSINGLECACHELINESIZE__MASK 0x00000038 +#define A6XX_RB_RENDER_CNTL_CCUSINGLECACHELINESIZE__SHIFT 3 +static inline uint32_t A6XX_RB_RENDER_CNTL_CCUSINGLECACHELINESIZE(uint32_t val) { - return ((val) << A6XX_RB_RENDER_CNTL_UNK5__SHIFT) & A6XX_RB_RENDER_CNTL_UNK5__MASK; + return ((val) << A6XX_RB_RENDER_CNTL_CCUSINGLECACHELINESIZE__SHIFT) & A6XX_RB_RENDER_CNTL_CCUSINGLECACHELINESIZE__MASK; } +#define A6XX_RB_RENDER_CNTL_EARLYVIZOUTEN 0x00000040 #define A6XX_RB_RENDER_CNTL_BINNING 0x00000080 -#define A6XX_RB_RENDER_CNTL_UNK8__MASK 0x00001f00 +#define A6XX_RB_RENDER_CNTL_UNK8__MASK 0x00000700 #define A6XX_RB_RENDER_CNTL_UNK8__SHIFT 8 static inline uint32_t A6XX_RB_RENDER_CNTL_UNK8(uint32_t val) { return ((val) << A6XX_RB_RENDER_CNTL_UNK8__SHIFT) & A6XX_RB_RENDER_CNTL_UNK8__MASK; } +#define A6XX_RB_RENDER_CNTL_RASTER_MODE__MASK 0x00000100 +#define A6XX_RB_RENDER_CNTL_RASTER_MODE__SHIFT 8 +static inline uint32_t A6XX_RB_RENDER_CNTL_RASTER_MODE(enum a6xx_raster_mode val) +{ + return ((val) << A6XX_RB_RENDER_CNTL_RASTER_MODE__SHIFT) & A6XX_RB_RENDER_CNTL_RASTER_MODE__MASK; +} +#define A6XX_RB_RENDER_CNTL_RASTER_DIRECTION__MASK 0x00000600 +#define A6XX_RB_RENDER_CNTL_RASTER_DIRECTION__SHIFT 9 +static inline uint32_t A6XX_RB_RENDER_CNTL_RASTER_DIRECTION(enum a6xx_raster_direction val) +{ + return ((val) << A6XX_RB_RENDER_CNTL_RASTER_DIRECTION__SHIFT) & A6XX_RB_RENDER_CNTL_RASTER_DIRECTION__MASK; +} +#define A6XX_RB_RENDER_CNTL_CONSERVATIVERASEN 0x00000800 +#define A6XX_RB_RENDER_CNTL_INNERCONSERVATIVERASEN 0x00001000 #define A6XX_RB_RENDER_CNTL_FLAG_DEPTH 0x00004000 #define A6XX_RB_RENDER_CNTL_FLAG_MRTS__MASK 0x00ff0000 #define A6XX_RB_RENDER_CNTL_FLAG_MRTS__SHIFT 16 @@ -2945,9 +3091,9 @@ static inline uint32_t A6XX_RB_SAMPLE_LOCATION_1_SAMPLE_3_Y(float val) #define A6XX_RB_RENDER_CONTROL0_IJ_PERSP_PIXEL 0x00000001 #define A6XX_RB_RENDER_CONTROL0_IJ_PERSP_CENTROID 0x00000002 #define A6XX_RB_RENDER_CONTROL0_IJ_PERSP_SAMPLE 0x00000004 -#define A6XX_RB_RENDER_CONTROL0_SIZE 0x00000008 -#define A6XX_RB_RENDER_CONTROL0_UNK4 0x00000010 -#define A6XX_RB_RENDER_CONTROL0_SIZE_PERSAMP 0x00000020 +#define A6XX_RB_RENDER_CONTROL0_IJ_LINEAR_PIXEL 0x00000008 +#define A6XX_RB_RENDER_CONTROL0_IJ_LINEAR_CENTROID 0x00000010 +#define A6XX_RB_RENDER_CONTROL0_IJ_LINEAR_SAMPLE 0x00000020 #define A6XX_RB_RENDER_CONTROL0_COORD_MASK__MASK 0x000003c0 #define A6XX_RB_RENDER_CONTROL0_COORD_MASK__SHIFT 6 static inline uint32_t A6XX_RB_RENDER_CONTROL0_COORD_MASK(uint32_t val) @@ -2961,11 +3107,15 @@ static inline uint32_t A6XX_RB_RENDER_CONTROL0_COORD_MASK(uint32_t val) #define A6XX_RB_RENDER_CONTROL1_UNK1 0x00000002 #define A6XX_RB_RENDER_CONTROL1_FACENESS 0x00000004 #define A6XX_RB_RENDER_CONTROL1_SAMPLEID 0x00000008 -#define A6XX_RB_RENDER_CONTROL1_UNK4 0x00000010 -#define A6XX_RB_RENDER_CONTROL1_UNK5 0x00000020 +#define A6XX_RB_RENDER_CONTROL1_FRAGCOORDSAMPLEMODE__MASK 0x00000030 +#define A6XX_RB_RENDER_CONTROL1_FRAGCOORDSAMPLEMODE__SHIFT 4 +static inline uint32_t A6XX_RB_RENDER_CONTROL1_FRAGCOORDSAMPLEMODE(enum a6xx_fragcoord_sample_mode val) +{ + return ((val) << A6XX_RB_RENDER_CONTROL1_FRAGCOORDSAMPLEMODE__SHIFT) & A6XX_RB_RENDER_CONTROL1_FRAGCOORDSAMPLEMODE__MASK; +} #define A6XX_RB_RENDER_CONTROL1_SIZE 0x00000040 -#define A6XX_RB_RENDER_CONTROL1_UNK7 0x00000080 -#define A6XX_RB_RENDER_CONTROL1_UNK8 0x00000100 +#define A6XX_RB_RENDER_CONTROL1_LINELENGTHEN 0x00000080 +#define A6XX_RB_RENDER_CONTROL1_FOVEATION 0x00000100 #define REG_A6XX_RB_FS_OUTPUT_CNTL0 0x0000880b #define A6XX_RB_FS_OUTPUT_CNTL0_DUAL_COLOR_IN_ENABLE 0x00000001 @@ -3299,7 +3449,7 @@ static inline uint32_t A6XX_RB_DEPTH_PLANE_CNTL_Z_MODE(enum a6xx_ztest_mode val) } #define REG_A6XX_RB_DEPTH_CNTL 0x00008871 -#define A6XX_RB_DEPTH_CNTL_Z_ENABLE 0x00000001 +#define A6XX_RB_DEPTH_CNTL_Z_TEST_ENABLE 0x00000001 #define A6XX_RB_DEPTH_CNTL_Z_WRITE_ENABLE 0x00000002 #define A6XX_RB_DEPTH_CNTL_ZFUNC__MASK 0x0000001c #define A6XX_RB_DEPTH_CNTL_ZFUNC__SHIFT 2 @@ -3308,7 +3458,7 @@ static inline uint32_t A6XX_RB_DEPTH_CNTL_ZFUNC(enum adreno_compare_func val) return ((val) << A6XX_RB_DEPTH_CNTL_ZFUNC__SHIFT) & A6XX_RB_DEPTH_CNTL_ZFUNC__MASK; } #define A6XX_RB_DEPTH_CNTL_Z_CLAMP_ENABLE 0x00000020 -#define A6XX_RB_DEPTH_CNTL_Z_TEST_ENABLE 0x00000040 +#define A6XX_RB_DEPTH_CNTL_Z_READ_ENABLE 0x00000040 #define A6XX_RB_DEPTH_CNTL_Z_BOUNDS_ENABLE 0x00000080 #define REG_A6XX_RB_DEPTH_BUFFER_INFO 0x00008872 @@ -3819,6 +3969,14 @@ static inline uint32_t A6XX_RB_SAMPLE_COUNT_ADDR(uint32_t val) return ((val) << A6XX_RB_SAMPLE_COUNT_ADDR__SHIFT) & A6XX_RB_SAMPLE_COUNT_ADDR__MASK; } +#define REG_A6XX_RB_UNKNOWN_8A00 0x00008a00 + +#define REG_A6XX_RB_UNKNOWN_8A10 0x00008a10 + +#define REG_A6XX_RB_UNKNOWN_8A20 0x00008a20 + +#define REG_A6XX_RB_UNKNOWN_8A30 0x00008a30 + #define REG_A6XX_RB_2D_BLIT_CNTL 0x00008c00 #define A6XX_RB_2D_BLIT_CNTL_ROTATE__MASK 0x00000007 #define A6XX_RB_2D_BLIT_CNTL_ROTATE__SHIFT 0 @@ -3826,11 +3984,12 @@ static inline uint32_t A6XX_RB_2D_BLIT_CNTL_ROTATE(enum a6xx_rotation val) { return ((val) << A6XX_RB_2D_BLIT_CNTL_ROTATE__SHIFT) & A6XX_RB_2D_BLIT_CNTL_ROTATE__MASK; } -#define A6XX_RB_2D_BLIT_CNTL_UNK3__MASK 0x00000078 -#define A6XX_RB_2D_BLIT_CNTL_UNK3__SHIFT 3 -static inline uint32_t A6XX_RB_2D_BLIT_CNTL_UNK3(uint32_t val) +#define A6XX_RB_2D_BLIT_CNTL_OVERWRITEEN 0x00000008 +#define A6XX_RB_2D_BLIT_CNTL_UNK4__MASK 0x00000070 +#define A6XX_RB_2D_BLIT_CNTL_UNK4__SHIFT 4 +static inline uint32_t A6XX_RB_2D_BLIT_CNTL_UNK4(uint32_t val) { - return ((val) << A6XX_RB_2D_BLIT_CNTL_UNK3__SHIFT) & A6XX_RB_2D_BLIT_CNTL_UNK3__MASK; + return ((val) << A6XX_RB_2D_BLIT_CNTL_UNK4__SHIFT) & A6XX_RB_2D_BLIT_CNTL_UNK4__MASK; } #define A6XX_RB_2D_BLIT_CNTL_SOLID_COLOR 0x00000080 #define A6XX_RB_2D_BLIT_CNTL_COLOR_FORMAT__MASK 0x0000ff00 @@ -3859,11 +4018,11 @@ static inline uint32_t A6XX_RB_2D_BLIT_CNTL_IFMT(enum a6xx_2d_ifmt val) { return ((val) << A6XX_RB_2D_BLIT_CNTL_IFMT__SHIFT) & A6XX_RB_2D_BLIT_CNTL_IFMT__MASK; } -#define A6XX_RB_2D_BLIT_CNTL_UNK29__MASK 0x20000000 -#define A6XX_RB_2D_BLIT_CNTL_UNK29__SHIFT 29 -static inline uint32_t A6XX_RB_2D_BLIT_CNTL_UNK29(uint32_t val) +#define A6XX_RB_2D_BLIT_CNTL_RASTER_MODE__MASK 0x20000000 +#define A6XX_RB_2D_BLIT_CNTL_RASTER_MODE__SHIFT 29 +static inline uint32_t A6XX_RB_2D_BLIT_CNTL_RASTER_MODE(enum a6xx_raster_mode val) { - return ((val) << A6XX_RB_2D_BLIT_CNTL_UNK29__SHIFT) & A6XX_RB_2D_BLIT_CNTL_UNK29__MASK; + return ((val) << A6XX_RB_2D_BLIT_CNTL_RASTER_MODE__SHIFT) & A6XX_RB_2D_BLIT_CNTL_RASTER_MODE__MASK; } #define REG_A6XX_RB_2D_UNKNOWN_8C01 0x00008c01 @@ -3997,11 +4156,17 @@ static inline uint32_t A6XX_RB_2D_DST_FLAGS_PLANE_PITCH(uint32_t val) #define REG_A6XX_RB_ADDR_MODE_CNTL 0x00008e05 #define REG_A6XX_RB_CCU_CNTL 0x00008e07 -#define A6XX_RB_CCU_CNTL_OFFSET__MASK 0xff800000 -#define A6XX_RB_CCU_CNTL_OFFSET__SHIFT 23 -static inline uint32_t A6XX_RB_CCU_CNTL_OFFSET(uint32_t val) +#define A6XX_RB_CCU_CNTL_COLOR_OFFSET__MASK 0xff800000 +#define A6XX_RB_CCU_CNTL_COLOR_OFFSET__SHIFT 23 +static inline uint32_t A6XX_RB_CCU_CNTL_COLOR_OFFSET(uint32_t val) { - return ((val >> 12) << A6XX_RB_CCU_CNTL_OFFSET__SHIFT) & A6XX_RB_CCU_CNTL_OFFSET__MASK; + return ((val >> 12) << A6XX_RB_CCU_CNTL_COLOR_OFFSET__SHIFT) & A6XX_RB_CCU_CNTL_COLOR_OFFSET__MASK; +} +#define A6XX_RB_CCU_CNTL_DEPTH_OFFSET__MASK 0x001ff000 +#define A6XX_RB_CCU_CNTL_DEPTH_OFFSET__SHIFT 12 +static inline uint32_t A6XX_RB_CCU_CNTL_DEPTH_OFFSET(uint32_t val) +{ + return ((val >> 12) << A6XX_RB_CCU_CNTL_DEPTH_OFFSET__SHIFT) & A6XX_RB_CCU_CNTL_DEPTH_OFFSET__MASK; } #define A6XX_RB_CCU_CNTL_GMEM 0x00400000 #define A6XX_RB_CCU_CNTL_UNK2 0x00000004 @@ -4052,7 +4217,13 @@ static inline uint32_t A6XX_RB_UNKNOWN_8E51(uint32_t val) return ((val) << A6XX_RB_UNKNOWN_8E51__SHIFT) & A6XX_RB_UNKNOWN_8E51__MASK; } -#define REG_A6XX_VPC_UNKNOWN_9100 0x00009100 +#define REG_A6XX_VPC_GS_PARAM 0x00009100 +#define A6XX_VPC_GS_PARAM_LINELENGTHLOC__MASK 0x000000ff +#define A6XX_VPC_GS_PARAM_LINELENGTHLOC__SHIFT 0 +static inline uint32_t A6XX_VPC_GS_PARAM_LINELENGTHLOC(uint32_t val) +{ + return ((val) << A6XX_VPC_GS_PARAM_LINELENGTHLOC__SHIFT) & A6XX_VPC_GS_PARAM_LINELENGTHLOC__MASK; +} #define REG_A6XX_VPC_VS_CLIP_CNTL 0x00009101 #define A6XX_VPC_VS_CLIP_CNTL_CLIP_MASK__MASK 0x000000ff @@ -4448,10 +4619,16 @@ static inline uint32_t A6XX_PC_TESS_CNTL_OUTPUT(enum a6xx_tess_output val) #define REG_A6XX_PC_MODE_CNTL 0x00009804 -#define REG_A6XX_PC_UNKNOWN_9805 0x00009805 +#define REG_A6XX_PC_POWER_CNTL 0x00009805 #define REG_A6XX_PC_PRIMID_PASSTHRU 0x00009806 +#define REG_A6XX_PC_SO_STREAM_CNTL 0x00009808 +#define A6XX_PC_SO_STREAM_CNTL_STREAM_ENABLE 0x00008000 + +#define REG_A6XX_PC_DGEN_SU_CONSERVATIVE_RAS_CNTL 0x0000980a +#define A6XX_PC_DGEN_SU_CONSERVATIVE_RAS_CNTL_CONSERVATIVERASEN 0x00000001 + #define REG_A6XX_PC_DRAW_CMD 0x00009840 #define A6XX_PC_DRAW_CMD_STATE_ID__MASK 0x000000ff #define A6XX_PC_DRAW_CMD_STATE_ID__SHIFT 0 @@ -4543,7 +4720,23 @@ static inline uint32_t A6XX_PC_GS_OUT_CNTL_CLIP_MASK(uint32_t val) return ((val) << A6XX_PC_GS_OUT_CNTL_CLIP_MASK__SHIFT) & A6XX_PC_GS_OUT_CNTL_CLIP_MASK__MASK; } -#define REG_A6XX_PC_PRIMITIVE_CNTL_3 0x00009b03 +#define REG_A6XX_PC_HS_OUT_CNTL 0x00009b03 +#define A6XX_PC_HS_OUT_CNTL_STRIDE_IN_VPC__MASK 0x000000ff +#define A6XX_PC_HS_OUT_CNTL_STRIDE_IN_VPC__SHIFT 0 +static inline uint32_t A6XX_PC_HS_OUT_CNTL_STRIDE_IN_VPC(uint32_t val) +{ + return ((val) << A6XX_PC_HS_OUT_CNTL_STRIDE_IN_VPC__SHIFT) & A6XX_PC_HS_OUT_CNTL_STRIDE_IN_VPC__MASK; +} +#define A6XX_PC_HS_OUT_CNTL_PSIZE 0x00000100 +#define A6XX_PC_HS_OUT_CNTL_LAYER 0x00000200 +#define A6XX_PC_HS_OUT_CNTL_VIEW 0x00000400 +#define A6XX_PC_HS_OUT_CNTL_PRIMITIVE_ID 0x00000800 +#define A6XX_PC_HS_OUT_CNTL_CLIP_MASK__MASK 0x00ff0000 +#define A6XX_PC_HS_OUT_CNTL_CLIP_MASK__SHIFT 16 +static inline uint32_t A6XX_PC_HS_OUT_CNTL_CLIP_MASK(uint32_t val) +{ + return ((val) << A6XX_PC_HS_OUT_CNTL_CLIP_MASK__SHIFT) & A6XX_PC_HS_OUT_CNTL_CLIP_MASK__MASK; +} #define REG_A6XX_PC_DS_OUT_CNTL 0x00009b04 #define A6XX_PC_DS_OUT_CNTL_STRIDE_IN_VPC__MASK 0x000000ff @@ -4576,6 +4769,7 @@ static inline uint32_t A6XX_PC_PRIMITIVE_CNTL_5_GS_INVOCATIONS(uint32_t val) { return ((val) << A6XX_PC_PRIMITIVE_CNTL_5_GS_INVOCATIONS__SHIFT) & A6XX_PC_PRIMITIVE_CNTL_5_GS_INVOCATIONS__MASK; } +#define A6XX_PC_PRIMITIVE_CNTL_5_LINELENGTHEN 0x00008000 #define A6XX_PC_PRIMITIVE_CNTL_5_GS_OUTPUT__MASK 0x00030000 #define A6XX_PC_PRIMITIVE_CNTL_5_GS_OUTPUT__SHIFT 16 static inline uint32_t A6XX_PC_PRIMITIVE_CNTL_5_GS_OUTPUT(enum a6xx_tess_output val) @@ -4763,11 +4957,11 @@ static inline uint32_t A6XX_VFD_CONTROL_1_REGID4VIEWID(uint32_t val) } #define REG_A6XX_VFD_CONTROL_2 0x0000a002 -#define A6XX_VFD_CONTROL_2_REGID_HSPATCHID__MASK 0x000000ff -#define A6XX_VFD_CONTROL_2_REGID_HSPATCHID__SHIFT 0 -static inline uint32_t A6XX_VFD_CONTROL_2_REGID_HSPATCHID(uint32_t val) +#define A6XX_VFD_CONTROL_2_REGID_HSRELPATCHID__MASK 0x000000ff +#define A6XX_VFD_CONTROL_2_REGID_HSRELPATCHID__SHIFT 0 +static inline uint32_t A6XX_VFD_CONTROL_2_REGID_HSRELPATCHID(uint32_t val) { - return ((val) << A6XX_VFD_CONTROL_2_REGID_HSPATCHID__SHIFT) & A6XX_VFD_CONTROL_2_REGID_HSPATCHID__MASK; + return ((val) << A6XX_VFD_CONTROL_2_REGID_HSRELPATCHID__SHIFT) & A6XX_VFD_CONTROL_2_REGID_HSRELPATCHID__MASK; } #define A6XX_VFD_CONTROL_2_REGID_INVOCATIONID__MASK 0x0000ff00 #define A6XX_VFD_CONTROL_2_REGID_INVOCATIONID__SHIFT 8 @@ -4777,17 +4971,17 @@ static inline uint32_t A6XX_VFD_CONTROL_2_REGID_INVOCATIONID(uint32_t val) } #define REG_A6XX_VFD_CONTROL_3 0x0000a003 -#define A6XX_VFD_CONTROL_3_UNK0__MASK 0x000000ff -#define A6XX_VFD_CONTROL_3_UNK0__SHIFT 0 -static inline uint32_t A6XX_VFD_CONTROL_3_UNK0(uint32_t val) +#define A6XX_VFD_CONTROL_3_REGID_DSPRIMID__MASK 0x000000ff +#define A6XX_VFD_CONTROL_3_REGID_DSPRIMID__SHIFT 0 +static inline uint32_t A6XX_VFD_CONTROL_3_REGID_DSPRIMID(uint32_t val) { - return ((val) << A6XX_VFD_CONTROL_3_UNK0__SHIFT) & A6XX_VFD_CONTROL_3_UNK0__MASK; + return ((val) << A6XX_VFD_CONTROL_3_REGID_DSPRIMID__SHIFT) & A6XX_VFD_CONTROL_3_REGID_DSPRIMID__MASK; } -#define A6XX_VFD_CONTROL_3_REGID_DSPATCHID__MASK 0x0000ff00 -#define A6XX_VFD_CONTROL_3_REGID_DSPATCHID__SHIFT 8 -static inline uint32_t A6XX_VFD_CONTROL_3_REGID_DSPATCHID(uint32_t val) +#define A6XX_VFD_CONTROL_3_REGID_DSRELPATCHID__MASK 0x0000ff00 +#define A6XX_VFD_CONTROL_3_REGID_DSRELPATCHID__SHIFT 8 +static inline uint32_t A6XX_VFD_CONTROL_3_REGID_DSRELPATCHID(uint32_t val) { - return ((val) << A6XX_VFD_CONTROL_3_REGID_DSPATCHID__SHIFT) & A6XX_VFD_CONTROL_3_REGID_DSPATCHID__MASK; + return ((val) << A6XX_VFD_CONTROL_3_REGID_DSRELPATCHID__SHIFT) & A6XX_VFD_CONTROL_3_REGID_DSRELPATCHID__MASK; } #define A6XX_VFD_CONTROL_3_REGID_TESSX__MASK 0x00ff0000 #define A6XX_VFD_CONTROL_3_REGID_TESSX__SHIFT 16 @@ -4828,9 +5022,12 @@ static inline uint32_t A6XX_VFD_CONTROL_5_UNK8(uint32_t val) #define A6XX_VFD_CONTROL_6_PRIMID_PASSTHRU 0x00000001 #define REG_A6XX_VFD_MODE_CNTL 0x0000a007 -#define A6XX_VFD_MODE_CNTL_BINNING_PASS 0x00000001 -#define A6XX_VFD_MODE_CNTL_UNK1 0x00000002 -#define A6XX_VFD_MODE_CNTL_UNK2 0x00000004 +#define A6XX_VFD_MODE_CNTL_RENDER_MODE__MASK 0x00000007 +#define A6XX_VFD_MODE_CNTL_RENDER_MODE__SHIFT 0 +static inline uint32_t A6XX_VFD_MODE_CNTL_RENDER_MODE(enum a6xx_render_mode val) +{ + return ((val) << A6XX_VFD_MODE_CNTL_RENDER_MODE__SHIFT) & A6XX_VFD_MODE_CNTL_RENDER_MODE__MASK; +} #define REG_A6XX_VFD_MULTIVIEW_CNTL 0x0000a008 #define A6XX_VFD_MULTIVIEW_CNTL_ENABLE 0x00000001 @@ -4913,7 +5110,7 @@ static inline uint32_t A6XX_VFD_DEST_CNTL_INSTR_REGID(uint32_t val) return ((val) << A6XX_VFD_DEST_CNTL_INSTR_REGID__SHIFT) & A6XX_VFD_DEST_CNTL_INSTR_REGID__MASK; } -#define REG_A6XX_SP_UNKNOWN_A0F8 0x0000a0f8 +#define REG_A6XX_VFD_POWER_CNTL 0x0000a0f8 #define REG_A6XX_VFD_ADDR_MODE_CNTL 0x0000a601 @@ -5091,11 +5288,11 @@ static inline uint32_t A6XX_SP_VS_CONFIG_NIBO(uint32_t val) #define REG_A6XX_SP_VS_INSTRLEN 0x0000a824 #define REG_A6XX_SP_VS_PVT_MEM_HW_STACK_OFFSET 0x0000a825 -#define A6XX_SP_VS_PVT_MEM_HW_STACK_OFFSET__MASK 0x0007ffff -#define A6XX_SP_VS_PVT_MEM_HW_STACK_OFFSET__SHIFT 0 -static inline uint32_t A6XX_SP_VS_PVT_MEM_HW_STACK_OFFSET(uint32_t val) +#define A6XX_SP_VS_PVT_MEM_HW_STACK_OFFSET_OFFSET__MASK 0x0007ffff +#define A6XX_SP_VS_PVT_MEM_HW_STACK_OFFSET_OFFSET__SHIFT 0 +static inline uint32_t A6XX_SP_VS_PVT_MEM_HW_STACK_OFFSET_OFFSET(uint32_t val) { - return ((val >> 11) << A6XX_SP_VS_PVT_MEM_HW_STACK_OFFSET__SHIFT) & A6XX_SP_VS_PVT_MEM_HW_STACK_OFFSET__MASK; + return ((val >> 11) << A6XX_SP_VS_PVT_MEM_HW_STACK_OFFSET_OFFSET__SHIFT) & A6XX_SP_VS_PVT_MEM_HW_STACK_OFFSET_OFFSET__MASK; } #define REG_A6XX_SP_HS_CTRL_REG0 0x0000a830 @@ -5201,11 +5398,11 @@ static inline uint32_t A6XX_SP_HS_CONFIG_NIBO(uint32_t val) #define REG_A6XX_SP_HS_INSTRLEN 0x0000a83c #define REG_A6XX_SP_HS_PVT_MEM_HW_STACK_OFFSET 0x0000a83d -#define A6XX_SP_HS_PVT_MEM_HW_STACK_OFFSET__MASK 0x0007ffff -#define A6XX_SP_HS_PVT_MEM_HW_STACK_OFFSET__SHIFT 0 -static inline uint32_t A6XX_SP_HS_PVT_MEM_HW_STACK_OFFSET(uint32_t val) +#define A6XX_SP_HS_PVT_MEM_HW_STACK_OFFSET_OFFSET__MASK 0x0007ffff +#define A6XX_SP_HS_PVT_MEM_HW_STACK_OFFSET_OFFSET__SHIFT 0 +static inline uint32_t A6XX_SP_HS_PVT_MEM_HW_STACK_OFFSET_OFFSET(uint32_t val) { - return ((val >> 11) << A6XX_SP_HS_PVT_MEM_HW_STACK_OFFSET__SHIFT) & A6XX_SP_HS_PVT_MEM_HW_STACK_OFFSET__MASK; + return ((val >> 11) << A6XX_SP_HS_PVT_MEM_HW_STACK_OFFSET_OFFSET__SHIFT) & A6XX_SP_HS_PVT_MEM_HW_STACK_OFFSET_OFFSET__MASK; } #define REG_A6XX_SP_DS_CTRL_REG0 0x0000a840 @@ -5379,11 +5576,11 @@ static inline uint32_t A6XX_SP_DS_CONFIG_NIBO(uint32_t val) #define REG_A6XX_SP_DS_INSTRLEN 0x0000a864 #define REG_A6XX_SP_DS_PVT_MEM_HW_STACK_OFFSET 0x0000a865 -#define A6XX_SP_DS_PVT_MEM_HW_STACK_OFFSET__MASK 0x0007ffff -#define A6XX_SP_DS_PVT_MEM_HW_STACK_OFFSET__SHIFT 0 -static inline uint32_t A6XX_SP_DS_PVT_MEM_HW_STACK_OFFSET(uint32_t val) +#define A6XX_SP_DS_PVT_MEM_HW_STACK_OFFSET_OFFSET__MASK 0x0007ffff +#define A6XX_SP_DS_PVT_MEM_HW_STACK_OFFSET_OFFSET__SHIFT 0 +static inline uint32_t A6XX_SP_DS_PVT_MEM_HW_STACK_OFFSET_OFFSET(uint32_t val) { - return ((val >> 11) << A6XX_SP_DS_PVT_MEM_HW_STACK_OFFSET__SHIFT) & A6XX_SP_DS_PVT_MEM_HW_STACK_OFFSET__MASK; + return ((val >> 11) << A6XX_SP_DS_PVT_MEM_HW_STACK_OFFSET_OFFSET__SHIFT) & A6XX_SP_DS_PVT_MEM_HW_STACK_OFFSET_OFFSET__MASK; } #define REG_A6XX_SP_GS_CTRL_REG0 0x0000a870 @@ -5559,11 +5756,11 @@ static inline uint32_t A6XX_SP_GS_CONFIG_NIBO(uint32_t val) #define REG_A6XX_SP_GS_INSTRLEN 0x0000a895 #define REG_A6XX_SP_GS_PVT_MEM_HW_STACK_OFFSET 0x0000a896 -#define A6XX_SP_GS_PVT_MEM_HW_STACK_OFFSET__MASK 0x0007ffff -#define A6XX_SP_GS_PVT_MEM_HW_STACK_OFFSET__SHIFT 0 -static inline uint32_t A6XX_SP_GS_PVT_MEM_HW_STACK_OFFSET(uint32_t val) +#define A6XX_SP_GS_PVT_MEM_HW_STACK_OFFSET_OFFSET__MASK 0x0007ffff +#define A6XX_SP_GS_PVT_MEM_HW_STACK_OFFSET_OFFSET__SHIFT 0 +static inline uint32_t A6XX_SP_GS_PVT_MEM_HW_STACK_OFFSET_OFFSET(uint32_t val) { - return ((val >> 11) << A6XX_SP_GS_PVT_MEM_HW_STACK_OFFSET__SHIFT) & A6XX_SP_GS_PVT_MEM_HW_STACK_OFFSET__MASK; + return ((val >> 11) << A6XX_SP_GS_PVT_MEM_HW_STACK_OFFSET_OFFSET__SHIFT) & A6XX_SP_GS_PVT_MEM_HW_STACK_OFFSET_OFFSET__MASK; } #define REG_A6XX_SP_VS_TEX_SAMP 0x0000a8a0 @@ -5926,11 +6123,11 @@ static inline uint32_t A6XX_SP_FS_BINDLESS_PREFETCH_CMD_TEX_ID(uint32_t val) #define REG_A6XX_SP_UNKNOWN_A9A8 0x0000a9a8 #define REG_A6XX_SP_FS_PVT_MEM_HW_STACK_OFFSET 0x0000a9a9 -#define A6XX_SP_FS_PVT_MEM_HW_STACK_OFFSET__MASK 0x0007ffff -#define A6XX_SP_FS_PVT_MEM_HW_STACK_OFFSET__SHIFT 0 -static inline uint32_t A6XX_SP_FS_PVT_MEM_HW_STACK_OFFSET(uint32_t val) +#define A6XX_SP_FS_PVT_MEM_HW_STACK_OFFSET_OFFSET__MASK 0x0007ffff +#define A6XX_SP_FS_PVT_MEM_HW_STACK_OFFSET_OFFSET__SHIFT 0 +static inline uint32_t A6XX_SP_FS_PVT_MEM_HW_STACK_OFFSET_OFFSET(uint32_t val) { - return ((val >> 11) << A6XX_SP_FS_PVT_MEM_HW_STACK_OFFSET__SHIFT) & A6XX_SP_FS_PVT_MEM_HW_STACK_OFFSET__MASK; + return ((val >> 11) << A6XX_SP_FS_PVT_MEM_HW_STACK_OFFSET_OFFSET__SHIFT) & A6XX_SP_FS_PVT_MEM_HW_STACK_OFFSET_OFFSET__MASK; } #define REG_A6XX_SP_CS_CTRL_REG0 0x0000a9b0 @@ -6053,12 +6250,54 @@ static inline uint32_t A6XX_SP_CS_CONFIG_NIBO(uint32_t val) #define REG_A6XX_SP_CS_INSTRLEN 0x0000a9bc #define REG_A6XX_SP_CS_PVT_MEM_HW_STACK_OFFSET 0x0000a9bd -#define A6XX_SP_CS_PVT_MEM_HW_STACK_OFFSET__MASK 0x0007ffff -#define A6XX_SP_CS_PVT_MEM_HW_STACK_OFFSET__SHIFT 0 -static inline uint32_t A6XX_SP_CS_PVT_MEM_HW_STACK_OFFSET(uint32_t val) +#define A6XX_SP_CS_PVT_MEM_HW_STACK_OFFSET_OFFSET__MASK 0x0007ffff +#define A6XX_SP_CS_PVT_MEM_HW_STACK_OFFSET_OFFSET__SHIFT 0 +static inline uint32_t A6XX_SP_CS_PVT_MEM_HW_STACK_OFFSET_OFFSET(uint32_t val) +{ + return ((val >> 11) << A6XX_SP_CS_PVT_MEM_HW_STACK_OFFSET_OFFSET__SHIFT) & A6XX_SP_CS_PVT_MEM_HW_STACK_OFFSET_OFFSET__MASK; +} + +#define REG_A6XX_SP_CS_CNTL_0 0x0000a9c2 +#define A6XX_SP_CS_CNTL_0_WGIDCONSTID__MASK 0x000000ff +#define A6XX_SP_CS_CNTL_0_WGIDCONSTID__SHIFT 0 +static inline uint32_t A6XX_SP_CS_CNTL_0_WGIDCONSTID(uint32_t val) +{ + return ((val) << A6XX_SP_CS_CNTL_0_WGIDCONSTID__SHIFT) & A6XX_SP_CS_CNTL_0_WGIDCONSTID__MASK; +} +#define A6XX_SP_CS_CNTL_0_WGSIZECONSTID__MASK 0x0000ff00 +#define A6XX_SP_CS_CNTL_0_WGSIZECONSTID__SHIFT 8 +static inline uint32_t A6XX_SP_CS_CNTL_0_WGSIZECONSTID(uint32_t val) +{ + return ((val) << A6XX_SP_CS_CNTL_0_WGSIZECONSTID__SHIFT) & A6XX_SP_CS_CNTL_0_WGSIZECONSTID__MASK; +} +#define A6XX_SP_CS_CNTL_0_WGOFFSETCONSTID__MASK 0x00ff0000 +#define A6XX_SP_CS_CNTL_0_WGOFFSETCONSTID__SHIFT 16 +static inline uint32_t A6XX_SP_CS_CNTL_0_WGOFFSETCONSTID(uint32_t val) { - return ((val >> 11) << A6XX_SP_CS_PVT_MEM_HW_STACK_OFFSET__SHIFT) & A6XX_SP_CS_PVT_MEM_HW_STACK_OFFSET__MASK; + return ((val) << A6XX_SP_CS_CNTL_0_WGOFFSETCONSTID__SHIFT) & A6XX_SP_CS_CNTL_0_WGOFFSETCONSTID__MASK; } +#define A6XX_SP_CS_CNTL_0_LOCALIDREGID__MASK 0xff000000 +#define A6XX_SP_CS_CNTL_0_LOCALIDREGID__SHIFT 24 +static inline uint32_t A6XX_SP_CS_CNTL_0_LOCALIDREGID(uint32_t val) +{ + return ((val) << A6XX_SP_CS_CNTL_0_LOCALIDREGID__SHIFT) & A6XX_SP_CS_CNTL_0_LOCALIDREGID__MASK; +} + +#define REG_A6XX_SP_CS_CNTL_1 0x0000a9c3 +#define A6XX_SP_CS_CNTL_1_LINEARLOCALIDREGID__MASK 0x000000ff +#define A6XX_SP_CS_CNTL_1_LINEARLOCALIDREGID__SHIFT 0 +static inline uint32_t A6XX_SP_CS_CNTL_1_LINEARLOCALIDREGID(uint32_t val) +{ + return ((val) << A6XX_SP_CS_CNTL_1_LINEARLOCALIDREGID__SHIFT) & A6XX_SP_CS_CNTL_1_LINEARLOCALIDREGID__MASK; +} +#define A6XX_SP_CS_CNTL_1_SINGLE_SP_CORE 0x00000100 +#define A6XX_SP_CS_CNTL_1_THREADSIZE__MASK 0x00000200 +#define A6XX_SP_CS_CNTL_1_THREADSIZE__SHIFT 9 +static inline uint32_t A6XX_SP_CS_CNTL_1_THREADSIZE(enum a6xx_threadsize val) +{ + return ((val) << A6XX_SP_CS_CNTL_1_THREADSIZE__SHIFT) & A6XX_SP_CS_CNTL_1_THREADSIZE__MASK; +} +#define A6XX_SP_CS_CNTL_1_THREADSIZE_SCALAR 0x00000400 #define REG_A6XX_SP_FS_TEX_SAMP 0x0000a9e0 #define A6XX_SP_FS_TEX_SAMP__MASK 0xffffffff @@ -6108,8 +6347,12 @@ static inline uint32_t A6XX_SP_CS_IBO(uint32_t val) #define REG_A6XX_SP_MODE_CONTROL 0x0000ab00 #define A6XX_SP_MODE_CONTROL_CONSTANT_DEMOTION_ENABLE 0x00000001 -#define A6XX_SP_MODE_CONTROL_UNK1 0x00000002 -#define A6XX_SP_MODE_CONTROL_UNK2 0x00000004 +#define A6XX_SP_MODE_CONTROL_ISAMMODE__MASK 0x00000006 +#define A6XX_SP_MODE_CONTROL_ISAMMODE__SHIFT 1 +static inline uint32_t A6XX_SP_MODE_CONTROL_ISAMMODE(enum a6xx_isam_mode val) +{ + return ((val) << A6XX_SP_MODE_CONTROL_ISAMMODE__SHIFT) & A6XX_SP_MODE_CONTROL_ISAMMODE__MASK; +} #define A6XX_SP_MODE_CONTROL_SHARED_CONSTS_ENABLE 0x00000008 #define REG_A6XX_SP_FS_CONFIG 0x0000ab04 @@ -6177,7 +6420,7 @@ static inline uint32_t A6XX_SP_2D_DST_FORMAT_MASK(uint32_t val) #define REG_A6XX_SP_NC_MODE_CNTL 0x0000ae02 -#define REG_A6XX_SP_UNKNOWN_AE03 0x0000ae03 +#define REG_A6XX_SP_CHICKEN_BITS 0x0000ae03 #define REG_A6XX_SP_FLOAT_CNTL 0x0000ae04 #define A6XX_SP_FLOAT_CNTL_F16_NO_INF 0x00000008 @@ -6192,6 +6435,8 @@ static inline uint32_t A6XX_SP_2D_DST_FORMAT_MASK(uint32_t val) static inline uint32_t REG_A6XX_SP_PERFCTR_SP_SEL(uint32_t i0) { return 0x0000ae10 + 0x1*i0; } +#define REG_A6XX_SP_CONTEXT_SWITCH_GFX_PREEMPTION_SAFE_MODE 0x0000be22 + #define REG_A6XX_SP_PS_TP_BORDER_COLOR_BASE_ADDR 0x0000b180 #define A6XX_SP_PS_TP_BORDER_COLOR_BASE_ADDR__MASK 0xffffffff #define A6XX_SP_PS_TP_BORDER_COLOR_BASE_ADDR__SHIFT 0 @@ -6357,7 +6602,19 @@ static inline uint32_t A6XX_SP_TP_WINDOW_OFFSET_Y(uint32_t val) return ((val) << A6XX_SP_TP_WINDOW_OFFSET_Y__SHIFT) & A6XX_SP_TP_WINDOW_OFFSET_Y__MASK; } -#define REG_A6XX_SP_TP_UNKNOWN_B309 0x0000b309 +#define REG_A6XX_SP_TP_MODE_CNTL 0x0000b309 +#define A6XX_SP_TP_MODE_CNTL_ISAMMODE__MASK 0x00000003 +#define A6XX_SP_TP_MODE_CNTL_ISAMMODE__SHIFT 0 +static inline uint32_t A6XX_SP_TP_MODE_CNTL_ISAMMODE(enum a6xx_isam_mode val) +{ + return ((val) << A6XX_SP_TP_MODE_CNTL_ISAMMODE__SHIFT) & A6XX_SP_TP_MODE_CNTL_ISAMMODE__MASK; +} +#define A6XX_SP_TP_MODE_CNTL_UNK3__MASK 0x000000fc +#define A6XX_SP_TP_MODE_CNTL_UNK3__SHIFT 2 +static inline uint32_t A6XX_SP_TP_MODE_CNTL_UNK3(uint32_t val) +{ + return ((val) << A6XX_SP_TP_MODE_CNTL_UNK3__SHIFT) & A6XX_SP_TP_MODE_CNTL_UNK3__MASK; +} #define REG_A6XX_SP_PS_2D_SRC_INFO 0x0000b4c0 #define A6XX_SP_PS_2D_SRC_INFO_COLOR_FORMAT__MASK 0x000000ff @@ -6499,7 +6756,7 @@ static inline uint32_t A6XX_SP_WINDOW_OFFSET_Y(uint32_t val) return ((val) << A6XX_SP_WINDOW_OFFSET_Y__SHIFT) & A6XX_SP_WINDOW_OFFSET_Y__MASK; } -#define REG_A6XX_TPL1_UNKNOWN_B600 0x0000b600 +#define REG_A6XX_TPL1_DBG_ECO_CNTL 0x0000b600 #define REG_A6XX_TPL1_ADDR_MODE_CNTL 0x0000b601 @@ -6687,17 +6944,17 @@ static inline uint32_t A6XX_HLSQ_CONTROL_4_REG_ZWCOORDREGID(uint32_t val) } #define REG_A6XX_HLSQ_CONTROL_5_REG 0x0000b986 -#define A6XX_HLSQ_CONTROL_5_REG_UNK0__MASK 0x000000ff -#define A6XX_HLSQ_CONTROL_5_REG_UNK0__SHIFT 0 -static inline uint32_t A6XX_HLSQ_CONTROL_5_REG_UNK0(uint32_t val) +#define A6XX_HLSQ_CONTROL_5_REG_LINELENGTHREGID__MASK 0x000000ff +#define A6XX_HLSQ_CONTROL_5_REG_LINELENGTHREGID__SHIFT 0 +static inline uint32_t A6XX_HLSQ_CONTROL_5_REG_LINELENGTHREGID(uint32_t val) { - return ((val) << A6XX_HLSQ_CONTROL_5_REG_UNK0__SHIFT) & A6XX_HLSQ_CONTROL_5_REG_UNK0__MASK; + return ((val) << A6XX_HLSQ_CONTROL_5_REG_LINELENGTHREGID__SHIFT) & A6XX_HLSQ_CONTROL_5_REG_LINELENGTHREGID__MASK; } -#define A6XX_HLSQ_CONTROL_5_REG_UNK8__MASK 0x0000ff00 -#define A6XX_HLSQ_CONTROL_5_REG_UNK8__SHIFT 8 -static inline uint32_t A6XX_HLSQ_CONTROL_5_REG_UNK8(uint32_t val) +#define A6XX_HLSQ_CONTROL_5_REG_FOVEATIONQUALITYREGID__MASK 0x0000ff00 +#define A6XX_HLSQ_CONTROL_5_REG_FOVEATIONQUALITYREGID__SHIFT 8 +static inline uint32_t A6XX_HLSQ_CONTROL_5_REG_FOVEATIONQUALITYREGID(uint32_t val) { - return ((val) << A6XX_HLSQ_CONTROL_5_REG_UNK8__SHIFT) & A6XX_HLSQ_CONTROL_5_REG_UNK8__MASK; + return ((val) << A6XX_HLSQ_CONTROL_5_REG_FOVEATIONQUALITYREGID__SHIFT) & A6XX_HLSQ_CONTROL_5_REG_FOVEATIONQUALITYREGID__MASK; } #define REG_A6XX_HLSQ_CS_CNTL 0x0000b987 @@ -6847,6 +7104,16 @@ static inline uint32_t REG_A6XX_HLSQ_CS_BINDLESS_BASE(uint32_t i0) { return 0x00 static inline uint32_t REG_A6XX_HLSQ_CS_BINDLESS_BASE_ADDR(uint32_t i0) { return 0x0000b9c0 + 0x2*i0; } +#define REG_A6XX_HLSQ_CS_UNKNOWN_B9D0 0x0000b9d0 +#define A6XX_HLSQ_CS_UNKNOWN_B9D0_SHARED_SIZE__MASK 0x0000001f +#define A6XX_HLSQ_CS_UNKNOWN_B9D0_SHARED_SIZE__SHIFT 0 +static inline uint32_t A6XX_HLSQ_CS_UNKNOWN_B9D0_SHARED_SIZE(uint32_t val) +{ + return ((val) << A6XX_HLSQ_CS_UNKNOWN_B9D0_SHARED_SIZE__SHIFT) & A6XX_HLSQ_CS_UNKNOWN_B9D0_SHARED_SIZE__MASK; +} +#define A6XX_HLSQ_CS_UNKNOWN_B9D0_UNK5 0x00000020 +#define A6XX_HLSQ_CS_UNKNOWN_B9D0_UNK6 0x00000040 + #define REG_A6XX_HLSQ_DRAW_CMD 0x0000bb00 #define A6XX_HLSQ_DRAW_CMD_STATE_ID__MASK 0x000000ff #define A6XX_HLSQ_DRAW_CMD_STATE_ID__SHIFT 0 @@ -6943,6 +7210,8 @@ static inline uint32_t A6XX_HLSQ_2D_EVENT_CMD_EVENT(enum vgt_event_type val) static inline uint32_t REG_A6XX_HLSQ_PERFCTR_HLSQ_SEL(uint32_t i0) { return 0x0000be10 + 0x1*i0; } +#define REG_A6XX_HLSQ_CONTEXT_SWITCH_GFX_PREEMPTION_SAFE_MODE 0x0000be22 + #define REG_A6XX_CP_EVENT_START 0x0000d600 #define A6XX_CP_EVENT_START_STATE_ID__MASK 0x000000ff #define A6XX_CP_EVENT_START_STATE_ID__SHIFT 0 @@ -7021,7 +7290,7 @@ static inline uint32_t A6XX_TEX_SAMP_0_LOD_BIAS(float val) } #define REG_A6XX_TEX_SAMP_1 0x00000001 -#define A6XX_TEX_SAMP_1_UNK0 0x00000001 +#define A6XX_TEX_SAMP_1_CLAMPENABLE 0x00000001 #define A6XX_TEX_SAMP_1_COMPARE_FUNC__MASK 0x0000000e #define A6XX_TEX_SAMP_1_COMPARE_FUNC__SHIFT 1 static inline uint32_t A6XX_TEX_SAMP_1_COMPARE_FUNC(enum adreno_compare_func val) @@ -7135,7 +7404,7 @@ static inline uint32_t A6XX_TEX_CONST_1_HEIGHT(uint32_t val) } #define REG_A6XX_TEX_CONST_2 0x00000002 -#define A6XX_TEX_CONST_2_UNK4 0x00000010 +#define A6XX_TEX_CONST_2_BUFFER 0x00000010 #define A6XX_TEX_CONST_2_PITCHALIGN__MASK 0x0000000f #define A6XX_TEX_CONST_2_PITCHALIGN__SHIFT 0 static inline uint32_t A6XX_TEX_CONST_2_PITCHALIGN(uint32_t val) @@ -7148,13 +7417,12 @@ static inline uint32_t A6XX_TEX_CONST_2_PITCH(uint32_t val) { return ((val) << A6XX_TEX_CONST_2_PITCH__SHIFT) & A6XX_TEX_CONST_2_PITCH__MASK; } -#define A6XX_TEX_CONST_2_TYPE__MASK 0x60000000 +#define A6XX_TEX_CONST_2_TYPE__MASK 0xe0000000 #define A6XX_TEX_CONST_2_TYPE__SHIFT 29 static inline uint32_t A6XX_TEX_CONST_2_TYPE(enum a6xx_tex_type val) { return ((val) << A6XX_TEX_CONST_2_TYPE__SHIFT) & A6XX_TEX_CONST_2_TYPE__MASK; } -#define A6XX_TEX_CONST_2_UNK31 0x80000000 #define REG_A6XX_TEX_CONST_3 0x00000003 #define A6XX_TEX_CONST_3_ARRAY_PITCH__MASK 0x00003fff @@ -7256,104 +7524,6 @@ static inline uint32_t A6XX_TEX_CONST_10_FLAG_BUFFER_LOGH(uint32_t val) #define REG_A6XX_TEX_CONST_15 0x0000000f -#define REG_A6XX_IBO_0 0x00000000 -#define A6XX_IBO_0_TILE_MODE__MASK 0x00000003 -#define A6XX_IBO_0_TILE_MODE__SHIFT 0 -static inline uint32_t A6XX_IBO_0_TILE_MODE(enum a6xx_tile_mode val) -{ - return ((val) << A6XX_IBO_0_TILE_MODE__SHIFT) & A6XX_IBO_0_TILE_MODE__MASK; -} -#define A6XX_IBO_0_FMT__MASK 0x3fc00000 -#define A6XX_IBO_0_FMT__SHIFT 22 -static inline uint32_t A6XX_IBO_0_FMT(enum a6xx_format val) -{ - return ((val) << A6XX_IBO_0_FMT__SHIFT) & A6XX_IBO_0_FMT__MASK; -} - -#define REG_A6XX_IBO_1 0x00000001 -#define A6XX_IBO_1_WIDTH__MASK 0x00007fff -#define A6XX_IBO_1_WIDTH__SHIFT 0 -static inline uint32_t A6XX_IBO_1_WIDTH(uint32_t val) -{ - return ((val) << A6XX_IBO_1_WIDTH__SHIFT) & A6XX_IBO_1_WIDTH__MASK; -} -#define A6XX_IBO_1_HEIGHT__MASK 0x3fff8000 -#define A6XX_IBO_1_HEIGHT__SHIFT 15 -static inline uint32_t A6XX_IBO_1_HEIGHT(uint32_t val) -{ - return ((val) << A6XX_IBO_1_HEIGHT__SHIFT) & A6XX_IBO_1_HEIGHT__MASK; -} - -#define REG_A6XX_IBO_2 0x00000002 -#define A6XX_IBO_2_UNK4 0x00000010 -#define A6XX_IBO_2_PITCH__MASK 0x1fffff80 -#define A6XX_IBO_2_PITCH__SHIFT 7 -static inline uint32_t A6XX_IBO_2_PITCH(uint32_t val) -{ - return ((val) << A6XX_IBO_2_PITCH__SHIFT) & A6XX_IBO_2_PITCH__MASK; -} -#define A6XX_IBO_2_TYPE__MASK 0x60000000 -#define A6XX_IBO_2_TYPE__SHIFT 29 -static inline uint32_t A6XX_IBO_2_TYPE(enum a6xx_tex_type val) -{ - return ((val) << A6XX_IBO_2_TYPE__SHIFT) & A6XX_IBO_2_TYPE__MASK; -} -#define A6XX_IBO_2_UNK31 0x80000000 - -#define REG_A6XX_IBO_3 0x00000003 -#define A6XX_IBO_3_ARRAY_PITCH__MASK 0x00003fff -#define A6XX_IBO_3_ARRAY_PITCH__SHIFT 0 -static inline uint32_t A6XX_IBO_3_ARRAY_PITCH(uint32_t val) -{ - return ((val >> 12) << A6XX_IBO_3_ARRAY_PITCH__SHIFT) & A6XX_IBO_3_ARRAY_PITCH__MASK; -} -#define A6XX_IBO_3_UNK27 0x08000000 -#define A6XX_IBO_3_FLAG 0x10000000 - -#define REG_A6XX_IBO_4 0x00000004 -#define A6XX_IBO_4_BASE_LO__MASK 0xffffffff -#define A6XX_IBO_4_BASE_LO__SHIFT 0 -static inline uint32_t A6XX_IBO_4_BASE_LO(uint32_t val) -{ - return ((val) << A6XX_IBO_4_BASE_LO__SHIFT) & A6XX_IBO_4_BASE_LO__MASK; -} - -#define REG_A6XX_IBO_5 0x00000005 -#define A6XX_IBO_5_BASE_HI__MASK 0x0001ffff -#define A6XX_IBO_5_BASE_HI__SHIFT 0 -static inline uint32_t A6XX_IBO_5_BASE_HI(uint32_t val) -{ - return ((val) << A6XX_IBO_5_BASE_HI__SHIFT) & A6XX_IBO_5_BASE_HI__MASK; -} -#define A6XX_IBO_5_DEPTH__MASK 0x3ffe0000 -#define A6XX_IBO_5_DEPTH__SHIFT 17 -static inline uint32_t A6XX_IBO_5_DEPTH(uint32_t val) -{ - return ((val) << A6XX_IBO_5_DEPTH__SHIFT) & A6XX_IBO_5_DEPTH__MASK; -} - -#define REG_A6XX_IBO_6 0x00000006 - -#define REG_A6XX_IBO_7 0x00000007 - -#define REG_A6XX_IBO_8 0x00000008 - -#define REG_A6XX_IBO_9 0x00000009 -#define A6XX_IBO_9_FLAG_BUFFER_ARRAY_PITCH__MASK 0x0001ffff -#define A6XX_IBO_9_FLAG_BUFFER_ARRAY_PITCH__SHIFT 0 -static inline uint32_t A6XX_IBO_9_FLAG_BUFFER_ARRAY_PITCH(uint32_t val) -{ - return ((val >> 4) << A6XX_IBO_9_FLAG_BUFFER_ARRAY_PITCH__SHIFT) & A6XX_IBO_9_FLAG_BUFFER_ARRAY_PITCH__MASK; -} - -#define REG_A6XX_IBO_10 0x0000000a -#define A6XX_IBO_10_FLAG_BUFFER_PITCH__MASK 0x0000007f -#define A6XX_IBO_10_FLAG_BUFFER_PITCH__SHIFT 0 -static inline uint32_t A6XX_IBO_10_FLAG_BUFFER_PITCH(uint32_t val) -{ - return ((val >> 6) << A6XX_IBO_10_FLAG_BUFFER_PITCH__SHIFT) & A6XX_IBO_10_FLAG_BUFFER_PITCH__MASK; -} - #define REG_A6XX_UBO_0 0x00000000 #define A6XX_UBO_0_BASE_LO__MASK 0xffffffff #define A6XX_UBO_0_BASE_LO__SHIFT 0 diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.xml.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.xml.h index 811589215834..4a3230978c0e 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.xml.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.xml.h @@ -8,19 +8,19 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno.xml ( 594 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a2xx.xml ( 90810 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_common.xml ( 14386 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_pm4.xml ( 67699 bytes, from 2021-05-31 20:21:57) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a3xx.xml ( 84226 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a4xx.xml ( 112551 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a5xx.xml ( 150713 bytes, from 2021-06-10 22:34:02) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a6xx.xml ( 180049 bytes, from 2021-06-02 21:44:19) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a6xx_gmu.xml ( 11331 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/ocmem.xml ( 1773 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_control_regs.xml ( 6038 bytes, from 2021-05-27 20:22:36) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_pipe_regs.xml ( 2924 bytes, from 2021-05-27 20:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno.xml ( 594 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a2xx.xml ( 90810 bytes, from 2021-06-21 15:24:24) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_common.xml ( 14609 bytes, from 2021-11-24 23:05:10) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_pm4.xml ( 69086 bytes, from 2022-03-03 16:41:33) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a3xx.xml ( 84231 bytes, from 2021-11-24 23:05:10) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a4xx.xml ( 113358 bytes, from 2022-01-31 23:06:21) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a5xx.xml ( 149512 bytes, from 2022-01-31 23:06:21) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a6xx.xml ( 184954 bytes, from 2022-03-03 16:41:33) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a6xx_gmu.xml ( 11331 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/ocmem.xml ( 1773 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_control_regs.xml ( 6038 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_pipe_regs.xml ( 2924 bytes, from 2021-07-22 15:21:56) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c index 7d23c741db4a..83c31b2ad865 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c @@ -101,6 +101,7 @@ static void get_stats_counter(struct msm_ringbuffer *ring, u32 counter, static void a6xx_set_pagetable(struct a6xx_gpu *a6xx_gpu, struct msm_ringbuffer *ring, struct msm_file_private *ctx) { + bool sysprof = refcount_read(&a6xx_gpu->base.base.sysprof_active) > 1; phys_addr_t ttbr; u32 asid; u64 memptr = rbmemptr(ring, ttbr0); @@ -111,6 +112,15 @@ static void a6xx_set_pagetable(struct a6xx_gpu *a6xx_gpu, if (msm_iommu_pagetable_params(ctx->aspace->mmu, &ttbr, &asid)) return; + if (!sysprof) { + /* Turn off protected mode to write to special registers */ + OUT_PKT7(ring, CP_SET_PROTECTED_MODE, 1); + OUT_RING(ring, 0); + + OUT_PKT4(ring, REG_A6XX_RBBM_PERFCTR_SRAM_INIT_CMD, 1); + OUT_RING(ring, 1); + } + /* Execute the table update */ OUT_PKT7(ring, CP_SMMU_TABLE_UPDATE, 4); OUT_RING(ring, CP_SMMU_TABLE_UPDATE_0_TTBR0_LO(lower_32_bits(ttbr))); @@ -137,6 +147,25 @@ static void a6xx_set_pagetable(struct a6xx_gpu *a6xx_gpu, OUT_PKT7(ring, CP_EVENT_WRITE, 1); OUT_RING(ring, 0x31); + + if (!sysprof) { + /* + * Wait for SRAM clear after the pgtable update, so the + * two can happen in parallel: + */ + OUT_PKT7(ring, CP_WAIT_REG_MEM, 6); + OUT_RING(ring, CP_WAIT_REG_MEM_0_FUNCTION(WRITE_EQ)); + OUT_RING(ring, CP_WAIT_REG_MEM_1_POLL_ADDR_LO( + REG_A6XX_RBBM_PERFCTR_SRAM_INIT_STATUS)); + OUT_RING(ring, CP_WAIT_REG_MEM_2_POLL_ADDR_HI(0)); + OUT_RING(ring, CP_WAIT_REG_MEM_3_REF(0x1)); + OUT_RING(ring, CP_WAIT_REG_MEM_4_MASK(0x1)); + OUT_RING(ring, CP_WAIT_REG_MEM_5_DELAY_LOOP_CYCLES(0)); + + /* Re-enable protected mode: */ + OUT_PKT7(ring, CP_SET_PROTECTED_MODE, 1); + OUT_RING(ring, 1); + } } static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) @@ -654,19 +683,23 @@ static void a6xx_set_cp_protect(struct msm_gpu *gpu) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); const u32 *regs = a6xx_protect; - unsigned i, count = ARRAY_SIZE(a6xx_protect), count_max = 32; - - BUILD_BUG_ON(ARRAY_SIZE(a6xx_protect) > 32); - BUILD_BUG_ON(ARRAY_SIZE(a650_protect) > 48); + unsigned i, count, count_max; if (adreno_is_a650(adreno_gpu)) { regs = a650_protect; count = ARRAY_SIZE(a650_protect); count_max = 48; + BUILD_BUG_ON(ARRAY_SIZE(a650_protect) > 48); } else if (adreno_is_a660_family(adreno_gpu)) { regs = a660_protect; count = ARRAY_SIZE(a660_protect); count_max = 48; + BUILD_BUG_ON(ARRAY_SIZE(a660_protect) > 48); + } else { + regs = a6xx_protect; + count = ARRAY_SIZE(a6xx_protect); + count_max = 32; + BUILD_BUG_ON(ARRAY_SIZE(a6xx_protect) > 32); } /* @@ -1800,6 +1833,7 @@ done: static const struct adreno_gpu_funcs funcs = { .base = { .get_param = adreno_get_param, + .set_param = adreno_set_param, .hw_init = a6xx_hw_init, .pm_suspend = a6xx_pm_suspend, .pm_resume = a6xx_pm_resume, diff --git a/drivers/gpu/drm/msm/adreno/adreno_common.xml.h b/drivers/gpu/drm/msm/adreno/adreno_common.xml.h index c9389d9fb599..abb037ccc02b 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_common.xml.h +++ b/drivers/gpu/drm/msm/adreno/adreno_common.xml.h @@ -8,19 +8,19 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno.xml ( 594 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a2xx.xml ( 90810 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_common.xml ( 14386 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_pm4.xml ( 67699 bytes, from 2021-05-31 20:21:57) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a3xx.xml ( 84226 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a4xx.xml ( 112551 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a5xx.xml ( 150713 bytes, from 2021-06-10 22:34:02) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a6xx.xml ( 180049 bytes, from 2021-06-02 21:44:19) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a6xx_gmu.xml ( 11331 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/ocmem.xml ( 1773 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_control_regs.xml ( 6038 bytes, from 2021-05-27 20:22:36) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_pipe_regs.xml ( 2924 bytes, from 2021-05-27 20:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno.xml ( 594 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a2xx.xml ( 90810 bytes, from 2021-06-21 15:24:24) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_common.xml ( 14609 bytes, from 2021-11-24 23:05:10) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_pm4.xml ( 69086 bytes, from 2022-03-03 16:41:33) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a3xx.xml ( 84231 bytes, from 2021-11-24 23:05:10) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a4xx.xml ( 113358 bytes, from 2022-01-31 23:06:21) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a5xx.xml ( 149512 bytes, from 2022-01-31 23:06:21) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a6xx.xml ( 184954 bytes, from 2022-03-03 16:41:33) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a6xx_gmu.xml ( 11331 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/ocmem.xml ( 1773 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_control_regs.xml ( 6038 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_pipe_regs.xml ( 2924 bytes, from 2021-07-22 15:21:56) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) @@ -205,6 +205,11 @@ enum a5xx_address_mode { ADDR_64B = 1, }; +enum a5xx_line_mode { + BRESENHAM = 0, + RECTANGULAR = 1, +}; + #define REG_AXXX_CP_RB_BASE 0x000001c0 #define REG_AXXX_CP_RB_CNTL 0x000001c1 diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index 15c8997b7251..9efc84929be0 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -244,10 +244,10 @@ int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx, *value = !adreno_is_a650_family(adreno_gpu) ? 0x100000 : 0; return 0; case MSM_PARAM_CHIP_ID: - *value = (uint64_t) adreno_gpu->rev.patchid | - (uint64_t) (adreno_gpu->rev.minor << 8) | - (uint64_t) (adreno_gpu->rev.major << 16) | - (uint64_t) (adreno_gpu->rev.core << 24); + *value = (uint64_t)adreno_gpu->rev.patchid | + ((uint64_t)adreno_gpu->rev.minor << 8) | + ((uint64_t)adreno_gpu->rev.major << 16) | + ((uint64_t)adreno_gpu->rev.core << 24); if (!adreno_gpu->info->revn) *value |= ((uint64_t) adreno_gpu->speedbin) << 32; return 0; @@ -283,6 +283,20 @@ int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx, } } +int adreno_set_param(struct msm_gpu *gpu, struct msm_file_private *ctx, + uint32_t param, uint64_t value) +{ + switch (param) { + case MSM_PARAM_SYSPROF: + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + return msm_file_private_set_sysprof(ctx, gpu, value); + default: + DBG("%s: invalid param: %u", gpu->name, param); + return -EINVAL; + } +} + const struct firmware * adreno_request_fw(struct adreno_gpu *adreno_gpu, const char *fwname) { diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h index b1ee453d627d..0490c5fbb780 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h @@ -282,6 +282,8 @@ static inline int adreno_is_a650_family(struct adreno_gpu *gpu) int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx, uint32_t param, uint64_t *value); +int adreno_set_param(struct msm_gpu *gpu, struct msm_file_private *ctx, + uint32_t param, uint64_t value); const struct firmware *adreno_request_fw(struct adreno_gpu *adreno_gpu, const char *fwname); struct drm_gem_object *adreno_fw_create_bo(struct msm_gpu *gpu, diff --git a/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h b/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h index e832ae4b937a..7aecf920f9b9 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h +++ b/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h @@ -8,21 +8,21 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno.xml ( 594 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a2xx.xml ( 90810 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_common.xml ( 14386 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_pm4.xml ( 67699 bytes, from 2021-05-31 20:21:57) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a3xx.xml ( 84226 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a4xx.xml ( 112551 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a5xx.xml ( 150713 bytes, from 2021-06-10 22:34:02) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a6xx.xml ( 180049 bytes, from 2021-06-02 21:44:19) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/a6xx_gmu.xml ( 11331 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/ocmem.xml ( 1773 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_control_regs.xml ( 6038 bytes, from 2021-05-27 20:22:36) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/adreno/adreno_pipe_regs.xml ( 2924 bytes, from 2021-05-27 20:18:13) - -Copyright (C) 2013-2021 by the following authors: +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno.xml ( 594 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a2xx.xml ( 90810 bytes, from 2021-06-21 15:24:24) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_common.xml ( 14609 bytes, from 2021-11-24 23:05:10) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_pm4.xml ( 69086 bytes, from 2022-03-03 16:41:33) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a3xx.xml ( 84231 bytes, from 2021-11-24 23:05:10) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a4xx.xml ( 113358 bytes, from 2022-01-31 23:06:21) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a5xx.xml ( 149512 bytes, from 2022-01-31 23:06:21) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a6xx.xml ( 184954 bytes, from 2022-03-03 16:41:33) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/a6xx_gmu.xml ( 11331 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/ocmem.xml ( 1773 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_control_regs.xml ( 6038 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/adreno/adreno_pipe_regs.xml ( 2924 bytes, from 2021-07-22 15:21:56) + +Copyright (C) 2013-2022 by the following authors: - Rob Clark <[email protected]> (robclark) - Ilia Mirkin <[email protected]> (imirkin) @@ -299,6 +299,8 @@ enum adreno_pm4_type3_packets { CP_SET_BIN_DATA5_OFFSET = 46, CP_SET_CTXSWITCH_IB = 85, CP_REG_WRITE = 109, + CP_START_BIN = 80, + CP_END_BIN = 81, }; enum adreno_state_block { @@ -438,7 +440,7 @@ enum cp_blit_cmd { BLIT_OP_SCALE = 3, }; -enum a6xx_render_mode { +enum a6xx_marker { RM6_BYPASS = 1, RM6_BINNING = 2, RM6_GMEM = 4, @@ -2150,13 +2152,13 @@ static inline uint32_t A5XX_CP_EXEC_CS_INDIRECT_3_LOCALSIZEZ(uint32_t val) #define REG_A6XX_CP_SET_MARKER_0 0x00000000 #define A6XX_CP_SET_MARKER_0_MODE__MASK 0x000001ff #define A6XX_CP_SET_MARKER_0_MODE__SHIFT 0 -static inline uint32_t A6XX_CP_SET_MARKER_0_MODE(enum a6xx_render_mode val) +static inline uint32_t A6XX_CP_SET_MARKER_0_MODE(enum a6xx_marker val) { return ((val) << A6XX_CP_SET_MARKER_0_MODE__SHIFT) & A6XX_CP_SET_MARKER_0_MODE__MASK; } #define A6XX_CP_SET_MARKER_0_MARKER__MASK 0x0000000f #define A6XX_CP_SET_MARKER_0_MARKER__SHIFT 0 -static inline uint32_t A6XX_CP_SET_MARKER_0_MARKER(enum a6xx_render_mode val) +static inline uint32_t A6XX_CP_SET_MARKER_0_MARKER(enum a6xx_marker val) { return ((val) << A6XX_CP_SET_MARKER_0_MARKER__SHIFT) & A6XX_CP_SET_MARKER_0_MARKER__MASK; } @@ -2351,5 +2353,13 @@ static inline uint32_t CP_SMMU_TABLE_UPDATE_3_CONTEXTBANK(uint32_t val) return ((val) << CP_SMMU_TABLE_UPDATE_3_CONTEXTBANK__SHIFT) & CP_SMMU_TABLE_UPDATE_3_CONTEXTBANK__MASK; } +#define REG_CP_START_BIN_BIN_COUNT 0x00000000 + +#define REG_CP_START_BIN_PREFIX_ADDR 0x00000001 + +#define REG_CP_START_BIN_PREFIX_DWORDS 0x00000003 + +#define REG_CP_START_BIN_BODY_DWORDS 0x00000004 + #endif /* ADRENO_PM4_XML */ diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4.xml.h b/drivers/gpu/drm/msm/disp/mdp4/mdp4.xml.h index c82d1bd09359..a2b6422948ec 100644 --- a/drivers/gpu/drm/msm/disp/mdp4/mdp4.xml.h +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4.xml.h @@ -8,25 +8,24 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/msm.xml ( 981 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi.xml ( 15291 bytes, from 2021-06-15 22:36:13) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 10953 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_5nm.xml ( 10900 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-02-18 16:45:44) +- /home/robclark/tmp/mesa/src/freedreno/registers/msm.xml ( 944 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi.xml ( 17560 bytes, from 2021-09-16 22:37:02) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 11007 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-01-30 18:25:22) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5.xml.h b/drivers/gpu/drm/msm/disp/mdp5/mdp5.xml.h index 0e21cf3f9e2e..86fc44b518cb 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5.xml.h +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5.xml.h @@ -8,25 +8,24 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/msm.xml ( 981 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi.xml ( 15291 bytes, from 2021-06-15 22:36:13) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 10953 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_5nm.xml ( 10900 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-02-18 16:45:44) +- /home/robclark/tmp/mesa/src/freedreno/registers/msm.xml ( 944 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi.xml ( 17560 bytes, from 2021-09-16 22:37:02) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 11007 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-01-30 18:25:22) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) diff --git a/drivers/gpu/drm/msm/disp/mdp_common.xml.h b/drivers/gpu/drm/msm/disp/mdp_common.xml.h index bb0066a63732..be759106b621 100644 --- a/drivers/gpu/drm/msm/disp/mdp_common.xml.h +++ b/drivers/gpu/drm/msm/disp/mdp_common.xml.h @@ -8,25 +8,24 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/msm.xml ( 981 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi.xml ( 15291 bytes, from 2021-06-15 22:36:13) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 10953 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_5nm.xml ( 10900 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-02-18 16:45:44) +- /home/robclark/tmp/mesa/src/freedreno/registers/msm.xml ( 944 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi.xml ( 17560 bytes, from 2021-09-16 22:37:02) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 11007 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-01-30 18:25:22) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) diff --git a/drivers/gpu/drm/msm/dsi/dsi.xml.h b/drivers/gpu/drm/msm/dsi/dsi.xml.h index 49b551ad1bff..4dee6f0bdda6 100644 --- a/drivers/gpu/drm/msm/dsi/dsi.xml.h +++ b/drivers/gpu/drm/msm/dsi/dsi.xml.h @@ -8,25 +8,24 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/msm.xml ( 981 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi.xml ( 15291 bytes, from 2021-06-15 22:36:13) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 10953 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_5nm.xml ( 10900 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-02-18 16:45:44) +- /home/robclark/tmp/mesa/src/freedreno/registers/msm.xml ( 944 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi.xml ( 17560 bytes, from 2021-09-16 22:37:02) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 11007 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-01-30 18:25:22) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) diff --git a/drivers/gpu/drm/msm/dsi/dsi_phy_10nm.xml.h b/drivers/gpu/drm/msm/dsi/dsi_phy_10nm.xml.h index 8872d77027dd..8b1be69ccf89 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_phy_10nm.xml.h +++ b/drivers/gpu/drm/msm/dsi/dsi_phy_10nm.xml.h @@ -8,25 +8,24 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/msm.xml ( 981 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi.xml ( 15291 bytes, from 2021-06-15 22:36:13) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 10953 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_5nm.xml ( 10900 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-02-18 16:45:44) +- /home/robclark/tmp/mesa/src/freedreno/registers/msm.xml ( 944 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi.xml ( 17560 bytes, from 2021-09-16 22:37:02) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 11007 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-01-30 18:25:22) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) diff --git a/drivers/gpu/drm/msm/dsi/dsi_phy_14nm.xml.h b/drivers/gpu/drm/msm/dsi/dsi_phy_14nm.xml.h index 0d9a5ccfb808..515f1fa605bf 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_phy_14nm.xml.h +++ b/drivers/gpu/drm/msm/dsi/dsi_phy_14nm.xml.h @@ -8,25 +8,24 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/msm.xml ( 981 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi.xml ( 15291 bytes, from 2021-06-15 22:36:13) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 10953 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_5nm.xml ( 10900 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-02-18 16:45:44) +- /home/robclark/tmp/mesa/src/freedreno/registers/msm.xml ( 944 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi.xml ( 17560 bytes, from 2021-09-16 22:37:02) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 11007 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-01-30 18:25:22) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) diff --git a/drivers/gpu/drm/msm/dsi/dsi_phy_20nm.xml.h b/drivers/gpu/drm/msm/dsi/dsi_phy_20nm.xml.h index 15c7a5852c7a..81e4622eb358 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_phy_20nm.xml.h +++ b/drivers/gpu/drm/msm/dsi/dsi_phy_20nm.xml.h @@ -8,25 +8,24 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/msm.xml ( 981 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi.xml ( 15291 bytes, from 2021-06-15 22:36:13) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 10953 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_5nm.xml ( 10900 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-02-18 16:45:44) +- /home/robclark/tmp/mesa/src/freedreno/registers/msm.xml ( 944 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi.xml ( 17560 bytes, from 2021-09-16 22:37:02) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 11007 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-01-30 18:25:22) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) diff --git a/drivers/gpu/drm/msm/dsi/dsi_phy_28nm.xml.h b/drivers/gpu/drm/msm/dsi/dsi_phy_28nm.xml.h index 5148f4d5b182..8c7db35c12c8 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_phy_28nm.xml.h +++ b/drivers/gpu/drm/msm/dsi/dsi_phy_28nm.xml.h @@ -8,25 +8,24 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/msm.xml ( 981 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi.xml ( 15291 bytes, from 2021-06-15 22:36:13) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 10953 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_5nm.xml ( 10900 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-02-18 16:45:44) +- /home/robclark/tmp/mesa/src/freedreno/registers/msm.xml ( 944 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi.xml ( 17560 bytes, from 2021-09-16 22:37:02) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 11007 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-01-30 18:25:22) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) diff --git a/drivers/gpu/drm/msm/dsi/dsi_phy_28nm_8960.xml.h b/drivers/gpu/drm/msm/dsi/dsi_phy_28nm_8960.xml.h index 446fe9ffc9c4..44eeca31a811 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_phy_28nm_8960.xml.h +++ b/drivers/gpu/drm/msm/dsi/dsi_phy_28nm_8960.xml.h @@ -8,25 +8,24 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/msm.xml ( 981 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi.xml ( 15291 bytes, from 2021-06-15 22:36:13) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 10953 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_5nm.xml ( 10900 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-02-18 16:45:44) +- /home/robclark/tmp/mesa/src/freedreno/registers/msm.xml ( 944 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi.xml ( 17560 bytes, from 2021-09-16 22:37:02) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 11007 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-01-30 18:25:22) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) diff --git a/drivers/gpu/drm/msm/dsi/dsi_phy_5nm.xml.h b/drivers/gpu/drm/msm/dsi/dsi_phy_5nm.xml.h deleted file mode 100644 index 404c890a29f1..000000000000 --- a/drivers/gpu/drm/msm/dsi/dsi_phy_5nm.xml.h +++ /dev/null @@ -1,480 +0,0 @@ -#ifndef DSI_PHY_5NM_XML -#define DSI_PHY_5NM_XML - -/* Autogenerated file, DO NOT EDIT manually! - -This file was generated by the rules-ng-ng headergen tool in this git repository: -http://github.com/freedreno/envytools/ -git clone https://github.com/freedreno/envytools.git - -The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/msm.xml ( 981 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi.xml ( 15291 bytes, from 2021-06-15 22:36:13) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 10953 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_5nm.xml ( 10900 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-02-18 16:45:44) - -Copyright (C) 2013-2021 by the following authors: -- Rob Clark <[email protected]> (robclark) -- Ilia Mirkin <[email protected]> (imirkin) - -Permission is hereby granted, free of charge, to any person obtaining -a copy of this software and associated documentation files (the -"Software"), to deal in the Software without restriction, including -without limitation the rights to use, copy, modify, merge, publish, -distribute, sublicense, and/or sell copies of the Software, and to -permit persons to whom the Software is furnished to do so, subject to -the following conditions: - -The above copyright notice and this permission notice (including the -next paragraph) shall be included in all copies or substantial -portions of the Software. - -THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, -EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF -MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. -IN NO EVENT SHALL THE COPYRIGHT OWNER(S) AND/OR ITS SUPPLIERS BE -LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION -OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION -WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. -*/ - - -#define REG_DSI_5nm_PHY_CMN_REVISION_ID0 0x00000000 - -#define REG_DSI_5nm_PHY_CMN_REVISION_ID1 0x00000004 - -#define REG_DSI_5nm_PHY_CMN_REVISION_ID2 0x00000008 - -#define REG_DSI_5nm_PHY_CMN_REVISION_ID3 0x0000000c - -#define REG_DSI_5nm_PHY_CMN_CLK_CFG0 0x00000010 - -#define REG_DSI_5nm_PHY_CMN_CLK_CFG1 0x00000014 - -#define REG_DSI_5nm_PHY_CMN_GLBL_CTRL 0x00000018 - -#define REG_DSI_5nm_PHY_CMN_RBUF_CTRL 0x0000001c - -#define REG_DSI_5nm_PHY_CMN_VREG_CTRL_0 0x00000020 - -#define REG_DSI_5nm_PHY_CMN_CTRL_0 0x00000024 - -#define REG_DSI_5nm_PHY_CMN_CTRL_1 0x00000028 - -#define REG_DSI_5nm_PHY_CMN_CTRL_2 0x0000002c - -#define REG_DSI_5nm_PHY_CMN_CTRL_3 0x00000030 - -#define REG_DSI_5nm_PHY_CMN_LANE_CFG0 0x00000034 - -#define REG_DSI_5nm_PHY_CMN_LANE_CFG1 0x00000038 - -#define REG_DSI_5nm_PHY_CMN_PLL_CNTRL 0x0000003c - -#define REG_DSI_5nm_PHY_CMN_DPHY_SOT 0x00000040 - -#define REG_DSI_5nm_PHY_CMN_LANE_CTRL0 0x000000a0 - -#define REG_DSI_5nm_PHY_CMN_LANE_CTRL1 0x000000a4 - -#define REG_DSI_5nm_PHY_CMN_LANE_CTRL2 0x000000a8 - -#define REG_DSI_5nm_PHY_CMN_LANE_CTRL3 0x000000ac - -#define REG_DSI_5nm_PHY_CMN_LANE_CTRL4 0x000000b0 - -#define REG_DSI_5nm_PHY_CMN_TIMING_CTRL_0 0x000000b4 - -#define REG_DSI_5nm_PHY_CMN_TIMING_CTRL_1 0x000000b8 - -#define REG_DSI_5nm_PHY_CMN_TIMING_CTRL_2 0x000000bc - -#define REG_DSI_5nm_PHY_CMN_TIMING_CTRL_3 0x000000c0 - -#define REG_DSI_5nm_PHY_CMN_TIMING_CTRL_4 0x000000c4 - -#define REG_DSI_5nm_PHY_CMN_TIMING_CTRL_5 0x000000c8 - -#define REG_DSI_5nm_PHY_CMN_TIMING_CTRL_6 0x000000cc - -#define REG_DSI_5nm_PHY_CMN_TIMING_CTRL_7 0x000000d0 - -#define REG_DSI_5nm_PHY_CMN_TIMING_CTRL_8 0x000000d4 - -#define REG_DSI_5nm_PHY_CMN_TIMING_CTRL_9 0x000000d8 - -#define REG_DSI_5nm_PHY_CMN_TIMING_CTRL_10 0x000000dc - -#define REG_DSI_5nm_PHY_CMN_TIMING_CTRL_11 0x000000e0 - -#define REG_DSI_5nm_PHY_CMN_TIMING_CTRL_12 0x000000e4 - -#define REG_DSI_5nm_PHY_CMN_TIMING_CTRL_13 0x000000e8 - -#define REG_DSI_5nm_PHY_CMN_GLBL_HSTX_STR_CTRL_0 0x000000ec - -#define REG_DSI_5nm_PHY_CMN_GLBL_HSTX_STR_CTRL_1 0x000000f0 - -#define REG_DSI_5nm_PHY_CMN_GLBL_RESCODE_OFFSET_TOP_CTRL 0x000000f4 - -#define REG_DSI_5nm_PHY_CMN_GLBL_RESCODE_OFFSET_BOT_CTRL 0x000000f8 - -#define REG_DSI_5nm_PHY_CMN_GLBL_RESCODE_OFFSET_MID_CTRL 0x000000fc - -#define REG_DSI_5nm_PHY_CMN_GLBL_LPTX_STR_CTRL 0x00000100 - -#define REG_DSI_5nm_PHY_CMN_GLBL_PEMPH_CTRL_0 0x00000104 - -#define REG_DSI_5nm_PHY_CMN_GLBL_PEMPH_CTRL_1 0x00000108 - -#define REG_DSI_5nm_PHY_CMN_GLBL_STR_SWI_CAL_SEL_CTRL 0x0000010c - -#define REG_DSI_5nm_PHY_CMN_VREG_CTRL_1 0x00000110 - -#define REG_DSI_5nm_PHY_CMN_CTRL_4 0x00000114 - -#define REG_DSI_5nm_PHY_CMN_PHY_STATUS 0x00000140 - -#define REG_DSI_5nm_PHY_CMN_LANE_STATUS0 0x00000148 - -#define REG_DSI_5nm_PHY_CMN_LANE_STATUS1 0x0000014c - -static inline uint32_t REG_DSI_5nm_PHY_LN(uint32_t i0) { return 0x00000000 + 0x80*i0; } - -static inline uint32_t REG_DSI_5nm_PHY_LN_CFG0(uint32_t i0) { return 0x00000000 + 0x80*i0; } - -static inline uint32_t REG_DSI_5nm_PHY_LN_CFG1(uint32_t i0) { return 0x00000004 + 0x80*i0; } - -static inline uint32_t REG_DSI_5nm_PHY_LN_CFG2(uint32_t i0) { return 0x00000008 + 0x80*i0; } - -static inline uint32_t REG_DSI_5nm_PHY_LN_TEST_DATAPATH(uint32_t i0) { return 0x0000000c + 0x80*i0; } - -static inline uint32_t REG_DSI_5nm_PHY_LN_PIN_SWAP(uint32_t i0) { return 0x00000010 + 0x80*i0; } - -static inline uint32_t REG_DSI_5nm_PHY_LN_LPRX_CTRL(uint32_t i0) { return 0x00000014 + 0x80*i0; } - -static inline uint32_t REG_DSI_5nm_PHY_LN_TX_DCTRL(uint32_t i0) { return 0x00000018 + 0x80*i0; } - -#define REG_DSI_5nm_PHY_PLL_ANALOG_CONTROLS_ONE 0x00000000 - -#define REG_DSI_5nm_PHY_PLL_ANALOG_CONTROLS_TWO 0x00000004 - -#define REG_DSI_5nm_PHY_PLL_INT_LOOP_SETTINGS 0x00000008 - -#define REG_DSI_5nm_PHY_PLL_INT_LOOP_SETTINGS_TWO 0x0000000c - -#define REG_DSI_5nm_PHY_PLL_ANALOG_CONTROLS_THREE 0x00000010 - -#define REG_DSI_5nm_PHY_PLL_ANALOG_CONTROLS_FOUR 0x00000014 - -#define REG_DSI_5nm_PHY_PLL_ANALOG_CONTROLS_FIVE 0x00000018 - -#define REG_DSI_5nm_PHY_PLL_INT_LOOP_CONTROLS 0x0000001c - -#define REG_DSI_5nm_PHY_PLL_DSM_DIVIDER 0x00000020 - -#define REG_DSI_5nm_PHY_PLL_FEEDBACK_DIVIDER 0x00000024 - -#define REG_DSI_5nm_PHY_PLL_SYSTEM_MUXES 0x00000028 - -#define REG_DSI_5nm_PHY_PLL_FREQ_UPDATE_CONTROL_OVERRIDES 0x0000002c - -#define REG_DSI_5nm_PHY_PLL_CMODE 0x00000030 - -#define REG_DSI_5nm_PHY_PLL_PSM_CTRL 0x00000034 - -#define REG_DSI_5nm_PHY_PLL_RSM_CTRL 0x00000038 - -#define REG_DSI_5nm_PHY_PLL_VCO_TUNE_MAP 0x0000003c - -#define REG_DSI_5nm_PHY_PLL_PLL_CNTRL 0x00000040 - -#define REG_DSI_5nm_PHY_PLL_CALIBRATION_SETTINGS 0x00000044 - -#define REG_DSI_5nm_PHY_PLL_BAND_SEL_CAL_TIMER_LOW 0x00000048 - -#define REG_DSI_5nm_PHY_PLL_BAND_SEL_CAL_TIMER_HIGH 0x0000004c - -#define REG_DSI_5nm_PHY_PLL_BAND_SEL_CAL_SETTINGS 0x00000050 - -#define REG_DSI_5nm_PHY_PLL_BAND_SEL_MIN 0x00000054 - -#define REG_DSI_5nm_PHY_PLL_BAND_SEL_MAX 0x00000058 - -#define REG_DSI_5nm_PHY_PLL_BAND_SEL_PFILT 0x0000005c - -#define REG_DSI_5nm_PHY_PLL_BAND_SEL_IFILT 0x00000060 - -#define REG_DSI_5nm_PHY_PLL_BAND_SEL_CAL_SETTINGS_TWO 0x00000064 - -#define REG_DSI_5nm_PHY_PLL_BAND_SEL_CAL_SETTINGS_THREE 0x00000068 - -#define REG_DSI_5nm_PHY_PLL_BAND_SEL_CAL_SETTINGS_FOUR 0x0000006c - -#define REG_DSI_5nm_PHY_PLL_BAND_SEL_ICODE_HIGH 0x00000070 - -#define REG_DSI_5nm_PHY_PLL_BAND_SEL_ICODE_LOW 0x00000074 - -#define REG_DSI_5nm_PHY_PLL_FREQ_DETECT_SETTINGS_ONE 0x00000078 - -#define REG_DSI_5nm_PHY_PLL_FREQ_DETECT_THRESH 0x0000007c - -#define REG_DSI_5nm_PHY_PLL_FREQ_DET_REFCLK_HIGH 0x00000080 - -#define REG_DSI_5nm_PHY_PLL_FREQ_DET_REFCLK_LOW 0x00000084 - -#define REG_DSI_5nm_PHY_PLL_FREQ_DET_PLLCLK_HIGH 0x00000088 - -#define REG_DSI_5nm_PHY_PLL_FREQ_DET_PLLCLK_LOW 0x0000008c - -#define REG_DSI_5nm_PHY_PLL_PFILT 0x00000090 - -#define REG_DSI_5nm_PHY_PLL_IFILT 0x00000094 - -#define REG_DSI_5nm_PHY_PLL_PLL_GAIN 0x00000098 - -#define REG_DSI_5nm_PHY_PLL_ICODE_LOW 0x0000009c - -#define REG_DSI_5nm_PHY_PLL_ICODE_HIGH 0x000000a0 - -#define REG_DSI_5nm_PHY_PLL_LOCKDET 0x000000a4 - -#define REG_DSI_5nm_PHY_PLL_OUTDIV 0x000000a8 - -#define REG_DSI_5nm_PHY_PLL_FASTLOCK_CONTROL 0x000000ac - -#define REG_DSI_5nm_PHY_PLL_PASS_OUT_OVERRIDE_ONE 0x000000b0 - -#define REG_DSI_5nm_PHY_PLL_PASS_OUT_OVERRIDE_TWO 0x000000b4 - -#define REG_DSI_5nm_PHY_PLL_CORE_OVERRIDE 0x000000b8 - -#define REG_DSI_5nm_PHY_PLL_CORE_INPUT_OVERRIDE 0x000000bc - -#define REG_DSI_5nm_PHY_PLL_RATE_CHANGE 0x000000c0 - -#define REG_DSI_5nm_PHY_PLL_PLL_DIGITAL_TIMERS 0x000000c4 - -#define REG_DSI_5nm_PHY_PLL_PLL_DIGITAL_TIMERS_TWO 0x000000c8 - -#define REG_DSI_5nm_PHY_PLL_DECIMAL_DIV_START 0x000000cc - -#define REG_DSI_5nm_PHY_PLL_FRAC_DIV_START_LOW 0x000000d0 - -#define REG_DSI_5nm_PHY_PLL_FRAC_DIV_START_MID 0x000000d4 - -#define REG_DSI_5nm_PHY_PLL_FRAC_DIV_START_HIGH 0x000000d8 - -#define REG_DSI_5nm_PHY_PLL_DEC_FRAC_MUXES 0x000000dc - -#define REG_DSI_5nm_PHY_PLL_DECIMAL_DIV_START_1 0x000000e0 - -#define REG_DSI_5nm_PHY_PLL_FRAC_DIV_START_LOW_1 0x000000e4 - -#define REG_DSI_5nm_PHY_PLL_FRAC_DIV_START_MID_1 0x000000e8 - -#define REG_DSI_5nm_PHY_PLL_FRAC_DIV_START_HIGH_1 0x000000ec - -#define REG_DSI_5nm_PHY_PLL_DECIMAL_DIV_START_2 0x000000f0 - -#define REG_DSI_5nm_PHY_PLL_FRAC_DIV_START_LOW_2 0x000000f4 - -#define REG_DSI_5nm_PHY_PLL_FRAC_DIV_START_MID_2 0x000000f8 - -#define REG_DSI_5nm_PHY_PLL_FRAC_DIV_START_HIGH_2 0x000000fc - -#define REG_DSI_5nm_PHY_PLL_MASH_CONTROL 0x00000100 - -#define REG_DSI_5nm_PHY_PLL_SSC_STEPSIZE_LOW 0x00000104 - -#define REG_DSI_5nm_PHY_PLL_SSC_STEPSIZE_HIGH 0x00000108 - -#define REG_DSI_5nm_PHY_PLL_SSC_DIV_PER_LOW 0x0000010c - -#define REG_DSI_5nm_PHY_PLL_SSC_DIV_PER_HIGH 0x00000110 - -#define REG_DSI_5nm_PHY_PLL_SSC_ADJPER_LOW 0x00000114 - -#define REG_DSI_5nm_PHY_PLL_SSC_ADJPER_HIGH 0x00000118 - -#define REG_DSI_5nm_PHY_PLL_SSC_MUX_CONTROL 0x0000011c - -#define REG_DSI_5nm_PHY_PLL_SSC_STEPSIZE_LOW_1 0x00000120 - -#define REG_DSI_5nm_PHY_PLL_SSC_STEPSIZE_HIGH_1 0x00000124 - -#define REG_DSI_5nm_PHY_PLL_SSC_DIV_PER_LOW_1 0x00000128 - -#define REG_DSI_5nm_PHY_PLL_SSC_DIV_PER_HIGH_1 0x0000012c - -#define REG_DSI_5nm_PHY_PLL_SSC_ADJPER_LOW_1 0x00000130 - -#define REG_DSI_5nm_PHY_PLL_SSC_ADJPER_HIGH_1 0x00000134 - -#define REG_DSI_5nm_PHY_PLL_SSC_STEPSIZE_LOW_2 0x00000138 - -#define REG_DSI_5nm_PHY_PLL_SSC_STEPSIZE_HIGH_2 0x0000013c - -#define REG_DSI_5nm_PHY_PLL_SSC_DIV_PER_LOW_2 0x00000140 - -#define REG_DSI_5nm_PHY_PLL_SSC_DIV_PER_HIGH_2 0x00000144 - -#define REG_DSI_5nm_PHY_PLL_SSC_ADJPER_LOW_2 0x00000148 - -#define REG_DSI_5nm_PHY_PLL_SSC_ADJPER_HIGH_2 0x0000014c - -#define REG_DSI_5nm_PHY_PLL_SSC_CONTROL 0x00000150 - -#define REG_DSI_5nm_PHY_PLL_PLL_OUTDIV_RATE 0x00000154 - -#define REG_DSI_5nm_PHY_PLL_PLL_LOCKDET_RATE_1 0x00000158 - -#define REG_DSI_5nm_PHY_PLL_PLL_LOCKDET_RATE_2 0x0000015c - -#define REG_DSI_5nm_PHY_PLL_PLL_PROP_GAIN_RATE_1 0x00000160 - -#define REG_DSI_5nm_PHY_PLL_PLL_PROP_GAIN_RATE_2 0x00000164 - -#define REG_DSI_5nm_PHY_PLL_PLL_BAND_SEL_RATE_1 0x00000168 - -#define REG_DSI_5nm_PHY_PLL_PLL_BAND_SEL_RATE_2 0x0000016c - -#define REG_DSI_5nm_PHY_PLL_PLL_INT_GAIN_IFILT_BAND_1 0x00000170 - -#define REG_DSI_5nm_PHY_PLL_PLL_INT_GAIN_IFILT_BAND_2 0x00000174 - -#define REG_DSI_5nm_PHY_PLL_PLL_FL_INT_GAIN_PFILT_BAND_1 0x00000178 - -#define REG_DSI_5nm_PHY_PLL_PLL_FL_INT_GAIN_PFILT_BAND_2 0x0000017c - -#define REG_DSI_5nm_PHY_PLL_PLL_FASTLOCK_EN_BAND 0x00000180 - -#define REG_DSI_5nm_PHY_PLL_FREQ_TUNE_ACCUM_INIT_MID 0x00000184 - -#define REG_DSI_5nm_PHY_PLL_FREQ_TUNE_ACCUM_INIT_HIGH 0x00000188 - -#define REG_DSI_5nm_PHY_PLL_FREQ_TUNE_ACCUM_INIT_MUX 0x0000018c - -#define REG_DSI_5nm_PHY_PLL_PLL_LOCK_OVERRIDE 0x00000190 - -#define REG_DSI_5nm_PHY_PLL_PLL_LOCK_DELAY 0x00000194 - -#define REG_DSI_5nm_PHY_PLL_PLL_LOCK_MIN_DELAY 0x00000198 - -#define REG_DSI_5nm_PHY_PLL_CLOCK_INVERTERS 0x0000019c - -#define REG_DSI_5nm_PHY_PLL_SPARE_AND_JPC_OVERRIDES 0x000001a0 - -#define REG_DSI_5nm_PHY_PLL_BIAS_CONTROL_1 0x000001a4 - -#define REG_DSI_5nm_PHY_PLL_BIAS_CONTROL_2 0x000001a8 - -#define REG_DSI_5nm_PHY_PLL_ALOG_OBSV_BUS_CTRL_1 0x000001ac - -#define REG_DSI_5nm_PHY_PLL_COMMON_STATUS_ONE 0x000001b0 - -#define REG_DSI_5nm_PHY_PLL_COMMON_STATUS_TWO 0x000001b4 - -#define REG_DSI_5nm_PHY_PLL_BAND_SEL_CAL 0x000001b8 - -#define REG_DSI_5nm_PHY_PLL_ICODE_ACCUM_STATUS_LOW 0x000001bc - -#define REG_DSI_5nm_PHY_PLL_ICODE_ACCUM_STATUS_HIGH 0x000001c0 - -#define REG_DSI_5nm_PHY_PLL_FD_OUT_LOW 0x000001c4 - -#define REG_DSI_5nm_PHY_PLL_FD_OUT_HIGH 0x000001c8 - -#define REG_DSI_5nm_PHY_PLL_ALOG_OBSV_BUS_STATUS_1 0x000001cc - -#define REG_DSI_5nm_PHY_PLL_PLL_MISC_CONFIG 0x000001d0 - -#define REG_DSI_5nm_PHY_PLL_FLL_CONFIG 0x000001d4 - -#define REG_DSI_5nm_PHY_PLL_FLL_FREQ_ACQ_TIME 0x000001d8 - -#define REG_DSI_5nm_PHY_PLL_FLL_CODE0 0x000001dc - -#define REG_DSI_5nm_PHY_PLL_FLL_CODE1 0x000001e0 - -#define REG_DSI_5nm_PHY_PLL_FLL_GAIN0 0x000001e4 - -#define REG_DSI_5nm_PHY_PLL_FLL_GAIN1 0x000001e8 - -#define REG_DSI_5nm_PHY_PLL_SW_RESET 0x000001ec - -#define REG_DSI_5nm_PHY_PLL_FAST_PWRUP 0x000001f0 - -#define REG_DSI_5nm_PHY_PLL_LOCKTIME0 0x000001f4 - -#define REG_DSI_5nm_PHY_PLL_LOCKTIME1 0x000001f8 - -#define REG_DSI_5nm_PHY_PLL_DEBUG_BUS_SEL 0x000001fc - -#define REG_DSI_5nm_PHY_PLL_DEBUG_BUS0 0x00000200 - -#define REG_DSI_5nm_PHY_PLL_DEBUG_BUS1 0x00000204 - -#define REG_DSI_5nm_PHY_PLL_DEBUG_BUS2 0x00000208 - -#define REG_DSI_5nm_PHY_PLL_DEBUG_BUS3 0x0000020c - -#define REG_DSI_5nm_PHY_PLL_ANALOG_FLL_CONTROL_OVERRIDES 0x00000210 - -#define REG_DSI_5nm_PHY_PLL_VCO_CONFIG 0x00000214 - -#define REG_DSI_5nm_PHY_PLL_VCO_CAL_CODE1_MODE0_STATUS 0x00000218 - -#define REG_DSI_5nm_PHY_PLL_VCO_CAL_CODE1_MODE1_STATUS 0x0000021c - -#define REG_DSI_5nm_PHY_PLL_RESET_SM_STATUS 0x00000220 - -#define REG_DSI_5nm_PHY_PLL_TDC_OFFSET 0x00000224 - -#define REG_DSI_5nm_PHY_PLL_PS3_PWRDOWN_CONTROLS 0x00000228 - -#define REG_DSI_5nm_PHY_PLL_PS4_PWRDOWN_CONTROLS 0x0000022c - -#define REG_DSI_5nm_PHY_PLL_PLL_RST_CONTROLS 0x00000230 - -#define REG_DSI_5nm_PHY_PLL_GEAR_BAND_SELECT_CONTROLS 0x00000234 - -#define REG_DSI_5nm_PHY_PLL_PSM_CLK_CONTROLS 0x00000238 - -#define REG_DSI_5nm_PHY_PLL_SYSTEM_MUXES_2 0x0000023c - -#define REG_DSI_5nm_PHY_PLL_VCO_CONFIG_1 0x00000240 - -#define REG_DSI_5nm_PHY_PLL_VCO_CONFIG_2 0x00000244 - -#define REG_DSI_5nm_PHY_PLL_CLOCK_INVERTERS_1 0x00000248 - -#define REG_DSI_5nm_PHY_PLL_CLOCK_INVERTERS_2 0x0000024c - -#define REG_DSI_5nm_PHY_PLL_CMODE_1 0x00000250 - -#define REG_DSI_5nm_PHY_PLL_CMODE_2 0x00000254 - -#define REG_DSI_5nm_PHY_PLL_ANALOG_CONTROLS_FIVE_1 0x00000258 - -#define REG_DSI_5nm_PHY_PLL_ANALOG_CONTROLS_FIVE_2 0x0000025c - -#define REG_DSI_5nm_PHY_PLL_PERF_OPTIMIZE 0x00000260 - - -#endif /* DSI_PHY_5NM_XML */ diff --git a/drivers/gpu/drm/msm/dsi/dsi_phy_7nm.xml.h b/drivers/gpu/drm/msm/dsi/dsi_phy_7nm.xml.h index 782f6b332baf..5bc061797003 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_phy_7nm.xml.h +++ b/drivers/gpu/drm/msm/dsi/dsi_phy_7nm.xml.h @@ -8,27 +8,26 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/msm.xml ( 981 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi.xml ( 15291 bytes, from 2021-06-15 22:36:13) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 10953 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_5nm.xml ( 10900 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-02-18 16:45:44) - -Copyright (C) 2013-2021 by the following authors: +- /home/robclark/tmp/mesa/src/freedreno/registers/msm.xml ( 944 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi.xml ( 17560 bytes, from 2021-09-16 22:37:02) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 11007 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-01-30 18:25:22) + +Copyright (C) 2013-2022 by the following authors: - Rob Clark <[email protected]> (robclark) - Ilia Mirkin <[email protected]> (imirkin) @@ -156,6 +155,8 @@ WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. #define REG_DSI_7nm_PHY_CMN_LANE_STATUS1 0x0000014c +#define REG_DSI_7nm_PHY_CMN_GLBL_DIGTOP_SPARE10 0x000001ac + static inline uint32_t REG_DSI_7nm_PHY_LN(uint32_t i0) { return 0x00000000 + 0x80*i0; } static inline uint32_t REG_DSI_7nm_PHY_LN_CFG0(uint32_t i0) { return 0x00000000 + 0x80*i0; } diff --git a/drivers/gpu/drm/msm/dsi/mmss_cc.xml.h b/drivers/gpu/drm/msm/dsi/mmss_cc.xml.h index fbb6a7a6b940..03bc322d0487 100644 --- a/drivers/gpu/drm/msm/dsi/mmss_cc.xml.h +++ b/drivers/gpu/drm/msm/dsi/mmss_cc.xml.h @@ -8,25 +8,24 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/msm.xml ( 981 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi.xml ( 15291 bytes, from 2021-06-15 22:36:13) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 10953 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_5nm.xml ( 10900 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-02-18 16:45:44) +- /home/robclark/tmp/mesa/src/freedreno/registers/msm.xml ( 944 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi.xml ( 17560 bytes, from 2021-09-16 22:37:02) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 11007 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-01-30 18:25:22) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) diff --git a/drivers/gpu/drm/msm/dsi/sfpb.xml.h b/drivers/gpu/drm/msm/dsi/sfpb.xml.h index 1f5e9e9f9803..2ae711cbec36 100644 --- a/drivers/gpu/drm/msm/dsi/sfpb.xml.h +++ b/drivers/gpu/drm/msm/dsi/sfpb.xml.h @@ -8,25 +8,24 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/msm.xml ( 981 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi.xml ( 15291 bytes, from 2021-06-15 22:36:13) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 10953 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_5nm.xml ( 10900 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-02-18 16:45:44) +- /home/robclark/tmp/mesa/src/freedreno/registers/msm.xml ( 944 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi.xml ( 17560 bytes, from 2021-09-16 22:37:02) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 11007 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-01-30 18:25:22) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.xml.h b/drivers/gpu/drm/msm/hdmi/hdmi.xml.h index a1cede6cfd02..c9eb26df67c0 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi.xml.h +++ b/drivers/gpu/drm/msm/hdmi/hdmi.xml.h @@ -8,25 +8,24 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/msm.xml ( 981 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi.xml ( 15291 bytes, from 2021-06-15 22:36:13) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 10953 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_5nm.xml ( 10900 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-02-18 16:45:44) +- /home/robclark/tmp/mesa/src/freedreno/registers/msm.xml ( 944 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi.xml ( 17560 bytes, from 2021-09-16 22:37:02) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 11007 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-01-30 18:25:22) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) diff --git a/drivers/gpu/drm/msm/hdmi/qfprom.xml.h b/drivers/gpu/drm/msm/hdmi/qfprom.xml.h index 4bf34cea1c89..3edc698c4df5 100644 --- a/drivers/gpu/drm/msm/hdmi/qfprom.xml.h +++ b/drivers/gpu/drm/msm/hdmi/qfprom.xml.h @@ -8,25 +8,24 @@ http://github.com/freedreno/envytools/ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: -- /home/robclark/src/mesa/mesa/src/freedreno/registers/msm.xml ( 981 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi.xml ( 15291 bytes, from 2021-06-15 22:36:13) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-06-05 21:37:42) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 10953 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/dsi_phy_5nm.xml ( 10900 bytes, from 2021-05-21 19:18:08) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-02-18 16:45:44) -- /home/robclark/src/mesa/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-02-18 16:45:44) +- /home/robclark/tmp/mesa/src/freedreno/registers/msm.xml ( 944 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/freedreno_copyright.xml ( 1572 bytes, from 2020-12-31 19:26:32) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp4.xml ( 20912 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp_common.xml ( 2849 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/mdp/mdp5.xml ( 37461 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi.xml ( 17560 bytes, from 2021-09-16 22:37:02) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_v2.xml ( 3236 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm_8960.xml ( 4935 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_28nm.xml ( 7004 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_20nm.xml ( 3712 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_14nm.xml ( 5381 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_10nm.xml ( 4499 bytes, from 2021-07-22 15:21:56) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/dsi_phy_7nm.xml ( 11007 bytes, from 2022-03-03 01:18:13) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/sfpb.xml ( 602 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/dsi/mmss_cc.xml ( 1686 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/qfprom.xml ( 600 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/hdmi/hdmi.xml ( 41874 bytes, from 2021-01-30 18:25:22) +- /home/robclark/tmp/mesa/src/freedreno/registers/edp/edp.xml ( 10416 bytes, from 2021-01-30 18:25:22) Copyright (C) 2013-2021 by the following authors: - Rob Clark <[email protected]> (robclark) diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index c09c23a80c81..61c81af23ba7 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -41,9 +41,10 @@ * - 1.6.0 - Syncobj support * - 1.7.0 - Add MSM_PARAM_SUSPENDS to access suspend count * - 1.8.0 - Add MSM_BO_CACHED_COHERENT for supported GPUs (a6xx) + * - 1.9.0 - Add MSM_SUBMIT_FENCE_SN_IN */ #define MSM_VERSION_MAJOR 1 -#define MSM_VERSION_MINOR 8 +#define MSM_VERSION_MINOR 9 #define MSM_VERSION_PATCHLEVEL 0 static const struct drm_mode_config_funcs mode_config_funcs = { @@ -561,8 +562,16 @@ static void context_close(struct msm_file_private *ctx) static void msm_postclose(struct drm_device *dev, struct drm_file *file) { + struct msm_drm_private *priv = dev->dev_private; struct msm_file_private *ctx = file->driver_priv; + /* + * It is not possible to set sysprof param to non-zero if gpu + * is not initialized: + */ + if (priv->gpu) + msm_file_private_set_sysprof(ctx, priv->gpu, 0); + context_close(ctx); } @@ -616,6 +625,25 @@ static int msm_ioctl_get_param(struct drm_device *dev, void *data, args->param, &args->value); } +static int msm_ioctl_set_param(struct drm_device *dev, void *data, + struct drm_file *file) +{ + struct msm_drm_private *priv = dev->dev_private; + struct drm_msm_param *args = data; + struct msm_gpu *gpu; + + if (args->pipe != MSM_PIPE_3D0) + return -EINVAL; + + gpu = priv->gpu; + + if (!gpu) + return -ENXIO; + + return gpu->funcs->set_param(gpu, file->driver_priv, + args->param, args->value); +} + static int msm_ioctl_gem_new(struct drm_device *dev, void *data, struct drm_file *file) { @@ -901,6 +929,7 @@ static int msm_ioctl_submitqueue_close(struct drm_device *dev, void *data, static const struct drm_ioctl_desc msm_ioctls[] = { DRM_IOCTL_DEF_DRV(MSM_GET_PARAM, msm_ioctl_get_param, DRM_RENDER_ALLOW), + DRM_IOCTL_DEF_DRV(MSM_SET_PARAM, msm_ioctl_set_param, DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(MSM_GEM_NEW, msm_ioctl_gem_new, DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(MSM_GEM_INFO, msm_ioctl_gem_info, DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(MSM_GEM_CPU_PREP, msm_ioctl_gem_cpu_prep, DRM_RENDER_ALLOW), diff --git a/drivers/gpu/drm/msm/msm_fb.c b/drivers/gpu/drm/msm/msm_fb.c index 96b379a08327..7137492fe78e 100644 --- a/drivers/gpu/drm/msm/msm_fb.c +++ b/drivers/gpu/drm/msm/msm_fb.c @@ -37,7 +37,7 @@ static int msm_framebuffer_dirtyfb(struct drm_framebuffer *fb, /* If this fb is not used on any display requiring pixel data to be * flushed, then skip dirtyfb */ - if (refcount_read(&msm_fb->dirtyfb) == 0) + if (refcount_read(&msm_fb->dirtyfb) == 1) return 0; return drm_atomic_helper_dirtyfb(fb, file_priv, flags, color, @@ -221,6 +221,8 @@ static struct drm_framebuffer *msm_framebuffer_init(struct drm_device *dev, goto fail; } + refcount_set(&msm_fb->dirtyfb, 1); + drm_dbg_state(dev, "create: FB ID: %d (%p)", fb->base.id, fb); return fb; diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c index 6cfa984dee6a..c6d60c8d286d 100644 --- a/drivers/gpu/drm/msm/msm_gem_submit.c +++ b/drivers/gpu/drm/msm/msm_gem_submit.c @@ -872,16 +872,46 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, submit->nr_cmds = i; + /* + * If using userspace provided seqno fence, validate that the id + * is available before arming sched job. Since access to fence_idr + * is serialized on the queue lock, the slot should be still avail + * after the job is armed + */ + if ((args->flags & MSM_SUBMIT_FENCE_SN_IN) && + idr_find(&queue->fence_idr, args->fence)) { + ret = -EINVAL; + goto out; + } + drm_sched_job_arm(&submit->base); submit->user_fence = dma_fence_get(&submit->base.s_fence->finished); - /* - * Allocate an id which can be used by WAIT_FENCE ioctl to map back - * to the underlying fence. - */ - submit->fence_id = idr_alloc_cyclic(&queue->fence_idr, - submit->user_fence, 1, INT_MAX, GFP_KERNEL); + if (args->flags & MSM_SUBMIT_FENCE_SN_IN) { + /* + * Userspace has assigned the seqno fence that it wants + * us to use. It is an error to pick a fence sequence + * number that is not available. + */ + submit->fence_id = args->fence; + ret = idr_alloc_u32(&queue->fence_idr, submit->user_fence, + &submit->fence_id, submit->fence_id, + GFP_KERNEL); + /* + * We've already validated that the fence_id slot is valid, + * so if idr_alloc_u32 failed, it is a kernel bug + */ + WARN_ON(ret); + } else { + /* + * Allocate an id which can be used by WAIT_FENCE ioctl to map + * back to the underlying fence. + */ + submit->fence_id = idr_alloc_cyclic(&queue->fence_idr, + submit->user_fence, 1, + INT_MAX, GFP_KERNEL); + } if (submit->fence_id < 0) { ret = submit->fence_id = 0; submit->fence_id = 0; diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index bacdabbaad96..faf0c242874e 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -959,6 +959,8 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev, gpu->nr_rings = nr_rings; + refcount_set(&gpu->sysprof_active, 1); + return 0; fail: diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h index c99627fc99dd..02419f2ca2bc 100644 --- a/drivers/gpu/drm/msm/msm_gpu.h +++ b/drivers/gpu/drm/msm/msm_gpu.h @@ -44,6 +44,8 @@ struct msm_gpu_config { struct msm_gpu_funcs { int (*get_param)(struct msm_gpu *gpu, struct msm_file_private *ctx, uint32_t param, uint64_t *value); + int (*set_param)(struct msm_gpu *gpu, struct msm_file_private *ctx, + uint32_t param, uint64_t value); int (*hw_init)(struct msm_gpu *gpu); int (*pm_suspend)(struct msm_gpu *gpu); int (*pm_resume)(struct msm_gpu *gpu); @@ -158,6 +160,13 @@ struct msm_gpu { int nr_rings; /** + * sysprof_active: + * + * The count of contexts that have enabled system profiling. + */ + refcount_t sysprof_active; + + /** * cur_ctx_seqno: * * The ctx->seqno value of the last context to submit rendering, @@ -328,6 +337,24 @@ struct msm_file_private { int seqno; /** + * sysprof: + * + * The value of MSM_PARAM_SYSPROF set by userspace. This is + * intended to be used by system profiling tools like Mesa's + * pps-producer (perfetto), and restricted to CAP_SYS_ADMIN. + * + * Setting a value of 1 will preserve performance counters across + * context switches. Setting a value of 2 will in addition + * suppress suspend. (Performance counters lose state across + * power collapse, which is undesirable for profiling in some + * cases.) + * + * The value automatically reverts to zero when the drm device + * file is closed. + */ + int sysprof; + + /** * entities: * * Table of per-priority-level sched entities used by submitqueues @@ -523,6 +550,8 @@ void msm_submitqueue_close(struct msm_file_private *ctx); void msm_submitqueue_destroy(struct kref *kref); +int msm_file_private_set_sysprof(struct msm_file_private *ctx, + struct msm_gpu *gpu, int sysprof); void __msm_file_private_destroy(struct kref *kref); static inline void msm_file_private_put(struct msm_file_private *ctx) diff --git a/drivers/gpu/drm/msm/msm_gpu_devfreq.c b/drivers/gpu/drm/msm/msm_gpu_devfreq.c index 9bf319be11f6..12641616acd3 100644 --- a/drivers/gpu/drm/msm/msm_gpu_devfreq.c +++ b/drivers/gpu/drm/msm/msm_gpu_devfreq.c @@ -83,6 +83,12 @@ static struct devfreq_dev_profile msm_devfreq_profile = { static void msm_devfreq_boost_work(struct kthread_work *work); static void msm_devfreq_idle_work(struct kthread_work *work); +static bool has_devfreq(struct msm_gpu *gpu) +{ + struct msm_gpu_devfreq *df = &gpu->devfreq; + return !!df->devfreq; +} + void msm_devfreq_init(struct msm_gpu *gpu) { struct msm_gpu_devfreq *df = &gpu->devfreq; @@ -149,6 +155,9 @@ void msm_devfreq_cleanup(struct msm_gpu *gpu) { struct msm_gpu_devfreq *df = &gpu->devfreq; + if (!has_devfreq(gpu)) + return; + devfreq_cooling_unregister(gpu->cooling); dev_pm_qos_remove_request(&df->boost_freq); dev_pm_qos_remove_request(&df->idle_freq); @@ -156,16 +165,24 @@ void msm_devfreq_cleanup(struct msm_gpu *gpu) void msm_devfreq_resume(struct msm_gpu *gpu) { - gpu->devfreq.busy_cycles = 0; - gpu->devfreq.time = ktime_get(); + struct msm_gpu_devfreq *df = &gpu->devfreq; - devfreq_resume_device(gpu->devfreq.devfreq); + if (!has_devfreq(gpu)) + return; + + df->busy_cycles = 0; + df->time = ktime_get(); + + devfreq_resume_device(df->devfreq); } void msm_devfreq_suspend(struct msm_gpu *gpu) { struct msm_gpu_devfreq *df = &gpu->devfreq; + if (!has_devfreq(gpu)) + return; + devfreq_suspend_device(df->devfreq); cancel_idle_work(df); @@ -185,6 +202,9 @@ void msm_devfreq_boost(struct msm_gpu *gpu, unsigned factor) struct msm_gpu_devfreq *df = &gpu->devfreq; uint64_t freq; + if (!has_devfreq(gpu)) + return; + freq = get_freq(gpu); freq *= factor; @@ -207,7 +227,7 @@ void msm_devfreq_active(struct msm_gpu *gpu) struct devfreq_dev_status status; unsigned int idle_time; - if (!df->devfreq) + if (!has_devfreq(gpu)) return; /* @@ -253,7 +273,7 @@ void msm_devfreq_idle(struct msm_gpu *gpu) { struct msm_gpu_devfreq *df = &gpu->devfreq; - if (!df->devfreq) + if (!has_devfreq(gpu)) return; msm_hrtimer_queue_work(&df->idle_work, ms_to_ktime(1), diff --git a/drivers/gpu/drm/msm/msm_submitqueue.c b/drivers/gpu/drm/msm/msm_submitqueue.c index 7cb158bcbcf6..79b6ccd6ce64 100644 --- a/drivers/gpu/drm/msm/msm_submitqueue.c +++ b/drivers/gpu/drm/msm/msm_submitqueue.c @@ -7,6 +7,45 @@ #include "msm_gpu.h" +int msm_file_private_set_sysprof(struct msm_file_private *ctx, + struct msm_gpu *gpu, int sysprof) +{ + /* + * Since pm_runtime and sysprof_active are both refcounts, we + * call apply the new value first, and then unwind the previous + * value + */ + + switch (sysprof) { + default: + return -EINVAL; + case 2: + pm_runtime_get_sync(&gpu->pdev->dev); + fallthrough; + case 1: + refcount_inc(&gpu->sysprof_active); + fallthrough; + case 0: + break; + } + + /* unwind old value: */ + switch (ctx->sysprof) { + case 2: + pm_runtime_put_autosuspend(&gpu->pdev->dev); + fallthrough; + case 1: + refcount_dec(&gpu->sysprof_active); + fallthrough; + case 0: + break; + } + + ctx->sysprof = sysprof; + + return 0; +} + void __msm_file_private_destroy(struct kref *kref) { struct msm_file_private *ctx = container_of(kref, diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index a7925a8290b2..0cb1345c6ba4 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -777,7 +777,7 @@ static void radeon_fixup_lvds_native_mode(struct drm_encoder *encoder, if (mode->type & DRM_MODE_TYPE_PREFERRED) { if (mode->hdisplay != native_mode->hdisplay || mode->vdisplay != native_mode->vdisplay) - memcpy(native_mode, mode, sizeof(*mode)); + drm_mode_copy(native_mode, mode); } } @@ -786,7 +786,7 @@ static void radeon_fixup_lvds_native_mode(struct drm_encoder *encoder, list_for_each_entry_safe(mode, t, &connector->probed_modes, head) { if (mode->hdisplay == native_mode->hdisplay && mode->vdisplay == native_mode->vdisplay) { - *native_mode = *mode; + drm_mode_copy(native_mode, mode); drm_mode_set_crtcinfo(native_mode, CRTC_INTERLACE_HALVE_V); DRM_DEBUG_KMS("Determined LVDS native mode details from EDID\n"); break; diff --git a/include/uapi/drm/msm_drm.h b/include/uapi/drm/msm_drm.h index 6b8fffc28a50..07efc8033492 100644 --- a/include/uapi/drm/msm_drm.h +++ b/include/uapi/drm/msm_drm.h @@ -67,16 +67,21 @@ struct drm_msm_timespec { __s64 tv_nsec; /* nanoseconds */ }; -#define MSM_PARAM_GPU_ID 0x01 -#define MSM_PARAM_GMEM_SIZE 0x02 -#define MSM_PARAM_CHIP_ID 0x03 -#define MSM_PARAM_MAX_FREQ 0x04 -#define MSM_PARAM_TIMESTAMP 0x05 -#define MSM_PARAM_GMEM_BASE 0x06 -#define MSM_PARAM_PRIORITIES 0x07 /* The # of priority levels */ -#define MSM_PARAM_PP_PGTABLE 0x08 /* => 1 for per-process pagetables, else 0 */ -#define MSM_PARAM_FAULTS 0x09 -#define MSM_PARAM_SUSPENDS 0x0a +/* Below "RO" indicates a read-only param, "WO" indicates write-only, and + * "RW" indicates a param that can be both read (GET_PARAM) and written + * (SET_PARAM) + */ +#define MSM_PARAM_GPU_ID 0x01 /* RO */ +#define MSM_PARAM_GMEM_SIZE 0x02 /* RO */ +#define MSM_PARAM_CHIP_ID 0x03 /* RO */ +#define MSM_PARAM_MAX_FREQ 0x04 /* RO */ +#define MSM_PARAM_TIMESTAMP 0x05 /* RO */ +#define MSM_PARAM_GMEM_BASE 0x06 /* RO */ +#define MSM_PARAM_PRIORITIES 0x07 /* RO: The # of priority levels */ +#define MSM_PARAM_PP_PGTABLE 0x08 /* RO: Deprecated, always returns zero */ +#define MSM_PARAM_FAULTS 0x09 /* RO */ +#define MSM_PARAM_SUSPENDS 0x0a /* RO */ +#define MSM_PARAM_SYSPROF 0x0b /* WO: 1 preserves perfcntrs, 2 also disables suspend */ /* For backwards compat. The original support for preemption was based on * a single ring per priority level so # of priority levels equals the # @@ -227,6 +232,7 @@ struct drm_msm_gem_submit_bo { #define MSM_SUBMIT_SUDO 0x10000000 /* run submitted cmds from RB */ #define MSM_SUBMIT_SYNCOBJ_IN 0x08000000 /* enable input syncobj */ #define MSM_SUBMIT_SYNCOBJ_OUT 0x04000000 /* enable output syncobj */ +#define MSM_SUBMIT_FENCE_SN_IN 0x02000000 /* userspace passes in seqno fence */ #define MSM_SUBMIT_FLAGS ( \ MSM_SUBMIT_NO_IMPLICIT | \ MSM_SUBMIT_FENCE_FD_IN | \ @@ -234,6 +240,7 @@ struct drm_msm_gem_submit_bo { MSM_SUBMIT_SUDO | \ MSM_SUBMIT_SYNCOBJ_IN | \ MSM_SUBMIT_SYNCOBJ_OUT | \ + MSM_SUBMIT_FENCE_SN_IN | \ 0) #define MSM_SUBMIT_SYNCOBJ_RESET 0x00000001 /* Reset syncobj after wait. */ @@ -253,7 +260,7 @@ struct drm_msm_gem_submit_syncobj { */ struct drm_msm_gem_submit { __u32 flags; /* MSM_PIPE_x | MSM_SUBMIT_x */ - __u32 fence; /* out */ + __u32 fence; /* out (or in with MSM_SUBMIT_FENCE_SN_IN flag) */ __u32 nr_bos; /* in, number of submit_bo's */ __u32 nr_cmds; /* in, number of submit_cmd's */ __u64 bos; /* in, ptr to array of submit_bo's */ @@ -333,9 +340,7 @@ struct drm_msm_submitqueue_query { }; #define DRM_MSM_GET_PARAM 0x00 -/* placeholder: #define DRM_MSM_SET_PARAM 0x01 - */ #define DRM_MSM_GEM_NEW 0x02 #define DRM_MSM_GEM_INFO 0x03 #define DRM_MSM_GEM_CPU_PREP 0x04 @@ -351,6 +356,7 @@ struct drm_msm_submitqueue_query { #define DRM_MSM_SUBMITQUEUE_QUERY 0x0C #define DRM_IOCTL_MSM_GET_PARAM DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_GET_PARAM, struct drm_msm_param) +#define DRM_IOCTL_MSM_SET_PARAM DRM_IOW (DRM_COMMAND_BASE + DRM_MSM_SET_PARAM, struct drm_msm_param) #define DRM_IOCTL_MSM_GEM_NEW DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_GEM_NEW, struct drm_msm_gem_new) #define DRM_IOCTL_MSM_GEM_INFO DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_GEM_INFO, struct drm_msm_gem_info) #define DRM_IOCTL_MSM_GEM_CPU_PREP DRM_IOW (DRM_COMMAND_BASE + DRM_MSM_GEM_CPU_PREP, struct drm_msm_gem_cpu_prep) diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h index baec5a41de3e..42975e940758 100644 --- a/include/uapi/linux/kfd_ioctl.h +++ b/include/uapi/linux/kfd_ioctl.h @@ -33,9 +33,10 @@ * - 1.5 - Add SVM API * - 1.6 - Query clear flags in SVM get_attr API * - 1.7 - Checkpoint Restore (CRIU) API + * - 1.8 - CRIU - Support for SDMA transfers with GTT BOs */ #define KFD_IOCTL_MAJOR_VERSION 1 -#define KFD_IOCTL_MINOR_VERSION 7 +#define KFD_IOCTL_MINOR_VERSION 8 struct kfd_ioctl_get_version_args { __u32 major_version; /* from KFD */ @@ -195,6 +196,8 @@ struct kfd_ioctl_dbg_wave_control_args { __u32 buf_size_in_bytes; /*including gpu_id and buf_size */ }; +#define KFD_INVALID_FD 0xffffffff + /* Matching HSA_EVENTTYPE */ #define KFD_IOC_EVENT_SIGNAL 0 #define KFD_IOC_EVENT_NODECHANGE 1 @@ -463,6 +466,7 @@ enum kfd_smi_event { }; #define KFD_SMI_EVENT_MASK_FROM_INDEX(i) (1ULL << ((i) - 1)) +#define KFD_SMI_EVENT_MSG_SIZE 96 struct kfd_ioctl_smi_events_args { __u32 gpuid; /* to KFD */ diff --git a/include/video/samsung_fimd.h b/include/video/samsung_fimd.h index c4a93ce1de48..e6966d187591 100644 --- a/include/video/samsung_fimd.h +++ b/include/video/samsung_fimd.h @@ -476,6 +476,10 @@ * 1111 -none- -none- -none- -none- -none- */ +#define WIN_RGB_ORDER(_win) (0x2020 + ((_win) * 4)) +#define WIN_RGB_ORDER_FORWARD (0 << 11) +#define WIN_RGB_ORDER_REVERSE (1 << 11) + /* FIMD Version 8 register offset definitions */ #define FIMD_V8_VIDTCON0 0x20010 #define FIMD_V8_VIDTCON1 0x20014 |