From b98a1648d6616d288e888c6dc6dcd4fa543585b3 Mon Sep 17 00:00:00 2001 From: Victor Zhao Date: Thu, 13 Oct 2022 11:06:33 +0800 Subject: Revert "drm/amdgpu: let mode2 reset fallback to default when failure" This reverts commit dac6b80818ac2353631c5a33d140d8d5508e2957. This commit reverted the AMDGPU_SKIP_MODE2_RESET as it conflicts with the original design of reset handler. Will redesign it. Fixes: dac6b80818ac23 ("drm/amdgpu: let mode2 reset fallback to default when failure") Signed-off-by: Victor Zhao Reviewed-by: Lijo Lazar Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index ab8f970b2849..bb73fb420ffc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -5210,7 +5210,6 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, reset_context->job = job; reset_context->hive = hive; - /* * Build list of devices to reset. * In case we are in XGMI hive mode, resort the device list @@ -5337,11 +5336,8 @@ retry: /* Rest of adevs pre asic reset from XGMI hive. */ amdgpu_ras_resume(adev); } else { r = amdgpu_do_asic_reset(device_list_handle, reset_context); - if (r && r == -EAGAIN) { - set_bit(AMDGPU_SKIP_MODE2_RESET, &reset_context->flags); - adev->asic_reset_res = 0; + if (r && r == -EAGAIN) goto retry; - } if (!r && gpu_reset_for_dev_remove) goto recover_end; @@ -5777,7 +5773,6 @@ pci_ers_result_t amdgpu_pci_slot_reset(struct pci_dev *pdev) reset_context.reset_req_dev = adev; set_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); set_bit(AMDGPU_SKIP_HW_RESET, &reset_context.flags); - set_bit(AMDGPU_SKIP_MODE2_RESET, &reset_context.flags); adev->no_hw_access = true; r = amdgpu_device_pre_asic_reset(adev, &reset_context); -- cgit From b31d6ada8346574ce04656e5ce9676ec763f5144 Mon Sep 17 00:00:00 2001 From: Evan Quan Date: Thu, 29 Sep 2022 10:50:44 +0800 Subject: drm/amd/pm: disable cstate feature for gpu reset scenario Suggested by PMFW team and same as what did for gfxoff feature. This can address some Mode1Reset failures observed on SMU13.0.0. Signed-off-by: Evan Quan Reviewed-by: Hawking Zhang Reviewed-by: Lijo Lazar Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org # 6.0.x --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 8 ++++++++ drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c | 8 ++++++++ drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c | 9 +++++++++ 3 files changed, 25 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index bb73fb420ffc..e0445e8cc342 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -2928,6 +2928,14 @@ static int amdgpu_device_ip_suspend_phase1(struct amdgpu_device *adev) amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE); amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE); + /* + * Per PMFW team's suggestion, driver needs to handle gfxoff + * and df cstate features disablement for gpu reset(e.g. Mode1Reset) + * scenario. Add the missing df cstate disablement here. + */ + if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW)) + dev_warn(adev->dev, "Failed to disallow df cstate"); + for (i = adev->num_ip_blocks - 1; i >= 0; i--) { if (!adev->ip_blocks[i].status.valid) continue; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c index 445005571f76..9cd005131f56 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c @@ -2242,9 +2242,17 @@ static void arcturus_get_unique_id(struct smu_context *smu) static int arcturus_set_df_cstate(struct smu_context *smu, enum pp_df_cstate state) { + struct amdgpu_device *adev = smu->adev; uint32_t smu_version; int ret; + /* + * Arcturus does not need the cstate disablement + * prerequisite for gpu reset. + */ + if (amdgpu_in_reset(adev) || adev->in_suspend) + return 0; + ret = smu_cmn_get_smc_version(smu, NULL, &smu_version); if (ret) { dev_err(smu->adev->dev, "Failed to get smu version!\n"); 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 619aee51b123..d30ec3005ea1 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c @@ -1640,6 +1640,15 @@ static bool aldebaran_is_baco_supported(struct smu_context *smu) static int aldebaran_set_df_cstate(struct smu_context *smu, enum pp_df_cstate state) { + struct amdgpu_device *adev = smu->adev; + + /* + * Aldebaran does not need the cstate disablement + * prerequisite for gpu reset. + */ + if (amdgpu_in_reset(adev) || adev->in_suspend) + return 0; + return smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_DFCstateControl, state, NULL); } -- cgit From a340847b0214aa9b8fd9839f7b2822ccc607edab Mon Sep 17 00:00:00 2001 From: Victor Zhao Date: Thu, 13 Oct 2022 11:06:33 +0800 Subject: Revert "drm/amdgpu: let mode2 reset fallback to default when failure" This reverts commit dac6b80818ac2353631c5a33d140d8d5508e2957. This commit reverted the AMDGPU_SKIP_MODE2_RESET as it conflicts with the original design of reset handler. Will redesign it. Fixes: dac6b80818ac23 ("drm/amdgpu: let mode2 reset fallback to default when failure") Signed-off-by: Victor Zhao Reviewed-by: Lijo Lazar Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 1 - drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7 +------ drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 1 - drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 1 - drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c | 6 ------ drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h | 3 +-- drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c | 1 - drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c | 1 - drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c | 1 - 9 files changed, 2 insertions(+), 20 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index 03bbfaa51cbc..0561812aa0a4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -134,7 +134,6 @@ static void amdgpu_amdkfd_reset_work(struct work_struct *work) reset_context.method = AMD_RESET_METHOD_NONE; reset_context.reset_req_dev = adev; clear_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); - clear_bit(AMDGPU_SKIP_MODE2_RESET, &reset_context.flags); amdgpu_device_gpu_recover(adev, NULL, &reset_context); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index ab8f970b2849..bb73fb420ffc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -5210,7 +5210,6 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, reset_context->job = job; reset_context->hive = hive; - /* * Build list of devices to reset. * In case we are in XGMI hive mode, resort the device list @@ -5337,11 +5336,8 @@ retry: /* Rest of adevs pre asic reset from XGMI hive. */ amdgpu_ras_resume(adev); } else { r = amdgpu_do_asic_reset(device_list_handle, reset_context); - if (r && r == -EAGAIN) { - set_bit(AMDGPU_SKIP_MODE2_RESET, &reset_context->flags); - adev->asic_reset_res = 0; + if (r && r == -EAGAIN) goto retry; - } if (!r && gpu_reset_for_dev_remove) goto recover_end; @@ -5777,7 +5773,6 @@ pci_ers_result_t amdgpu_pci_slot_reset(struct pci_dev *pdev) reset_context.reset_req_dev = adev; set_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); set_bit(AMDGPU_SKIP_HW_RESET, &reset_context.flags); - set_bit(AMDGPU_SKIP_MODE2_RESET, &reset_context.flags); adev->no_hw_access = true; r = amdgpu_device_pre_asic_reset(adev, &reset_context); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c index 46c99331d7f1..cd968e781077 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c @@ -72,7 +72,6 @@ static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job) reset_context.method = AMD_RESET_METHOD_NONE; reset_context.reset_req_dev = adev; clear_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); - clear_bit(AMDGPU_SKIP_MODE2_RESET, &reset_context.flags); r = amdgpu_device_gpu_recover(ring->adev, job, &reset_context); if (r) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 2dad7aa9a03b..75f1402101f4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -1950,7 +1950,6 @@ static void amdgpu_ras_do_recovery(struct work_struct *work) reset_context.method = AMD_RESET_METHOD_NONE; reset_context.reset_req_dev = adev; clear_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); - clear_bit(AMDGPU_SKIP_MODE2_RESET, &reset_context.flags); amdgpu_device_gpu_recover(ras->adev, NULL, &reset_context); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c index 831fb222139c..f778466bb9db 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c @@ -74,9 +74,6 @@ int amdgpu_reset_prepare_hwcontext(struct amdgpu_device *adev, { struct amdgpu_reset_handler *reset_handler = NULL; - if (test_bit(AMDGPU_SKIP_MODE2_RESET, &reset_context->flags)) - return -ENOSYS; - if (adev->reset_cntl && adev->reset_cntl->get_reset_handler) reset_handler = adev->reset_cntl->get_reset_handler( adev->reset_cntl, reset_context); @@ -93,9 +90,6 @@ int amdgpu_reset_perform_reset(struct amdgpu_device *adev, int ret; struct amdgpu_reset_handler *reset_handler = NULL; - if (test_bit(AMDGPU_SKIP_MODE2_RESET, &reset_context->flags)) - return -ENOSYS; - if (adev->reset_cntl) reset_handler = adev->reset_cntl->get_reset_handler( adev->reset_cntl, reset_context); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h index f5318fedf2f0..f4a501ff87d9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h @@ -30,8 +30,7 @@ enum AMDGPU_RESET_FLAGS { AMDGPU_NEED_FULL_RESET = 0, AMDGPU_SKIP_HW_RESET = 1, - AMDGPU_SKIP_MODE2_RESET = 2, - AMDGPU_RESET_FOR_DEVICE_REMOVE = 3, + AMDGPU_RESET_FOR_DEVICE_REMOVE = 2, }; struct amdgpu_reset_context { diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c index a2f04b249132..12906ba74462 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c @@ -290,7 +290,6 @@ flr_done: reset_context.method = AMD_RESET_METHOD_NONE; reset_context.reset_req_dev = adev; clear_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); - clear_bit(AMDGPU_SKIP_MODE2_RESET, &reset_context.flags); amdgpu_device_gpu_recover(adev, NULL, &reset_context); } diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c index a977f0027928..e07757eea7ad 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c @@ -317,7 +317,6 @@ flr_done: reset_context.method = AMD_RESET_METHOD_NONE; reset_context.reset_req_dev = adev; clear_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); - clear_bit(AMDGPU_SKIP_MODE2_RESET, &reset_context.flags); amdgpu_device_gpu_recover(adev, NULL, &reset_context); } diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c index fd14fa9b9cd7..288c414babdf 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c @@ -529,7 +529,6 @@ static void xgpu_vi_mailbox_flr_work(struct work_struct *work) reset_context.method = AMD_RESET_METHOD_NONE; reset_context.reset_req_dev = adev; clear_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); - clear_bit(AMDGPU_SKIP_MODE2_RESET, &reset_context.flags); amdgpu_device_gpu_recover(adev, NULL, &reset_context); } -- cgit From 3059cd8c5f797ad83d2b194ae66339f5c007ca43 Mon Sep 17 00:00:00 2001 From: Evan Quan Date: Thu, 29 Sep 2022 10:50:44 +0800 Subject: drm/amd/pm: disable cstate feature for gpu reset scenario Suggested by PMFW team and same as what did for gfxoff feature. This can address some Mode1Reset failures observed on SMU13.0.0. Signed-off-by: Evan Quan Reviewed-by: Hawking Zhang Reviewed-by: Lijo Lazar Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org # 6.0.x Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 8 ++++++++ drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c | 8 ++++++++ drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c | 9 +++++++++ 3 files changed, 25 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index bb73fb420ffc..e0445e8cc342 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -2928,6 +2928,14 @@ static int amdgpu_device_ip_suspend_phase1(struct amdgpu_device *adev) amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE); amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE); + /* + * Per PMFW team's suggestion, driver needs to handle gfxoff + * and df cstate features disablement for gpu reset(e.g. Mode1Reset) + * scenario. Add the missing df cstate disablement here. + */ + if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW)) + dev_warn(adev->dev, "Failed to disallow df cstate"); + for (i = adev->num_ip_blocks - 1; i >= 0; i--) { if (!adev->ip_blocks[i].status.valid) continue; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c index 445005571f76..9cd005131f56 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c @@ -2242,9 +2242,17 @@ static void arcturus_get_unique_id(struct smu_context *smu) static int arcturus_set_df_cstate(struct smu_context *smu, enum pp_df_cstate state) { + struct amdgpu_device *adev = smu->adev; uint32_t smu_version; int ret; + /* + * Arcturus does not need the cstate disablement + * prerequisite for gpu reset. + */ + if (amdgpu_in_reset(adev) || adev->in_suspend) + return 0; + ret = smu_cmn_get_smc_version(smu, NULL, &smu_version); if (ret) { dev_err(smu->adev->dev, "Failed to get smu version!\n"); 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 619aee51b123..d30ec3005ea1 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c @@ -1640,6 +1640,15 @@ static bool aldebaran_is_baco_supported(struct smu_context *smu) static int aldebaran_set_df_cstate(struct smu_context *smu, enum pp_df_cstate state) { + struct amdgpu_device *adev = smu->adev; + + /* + * Aldebaran does not need the cstate disablement + * prerequisite for gpu reset. + */ + if (amdgpu_in_reset(adev) || adev->in_suspend) + return 0; + return smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_DFCstateControl, state, NULL); } -- cgit From 693073a04de491c82cd0e1d6bf979192375373c1 Mon Sep 17 00:00:00 2001 From: YuBiao Wang Date: Wed, 19 Oct 2022 11:36:32 +0800 Subject: drm/amdgpu: skip mes self test for gc 11.0.3 in recover Temporary disable mes self teset for gc 11.0.3 during gpu_recovery. Signed-off-by: YuBiao Wang Acked-by: Luben Tuikov Reviewed-by: Hawking Zhang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index e0445e8cc342..5b8362727226 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -5381,7 +5381,7 @@ skip_hw_reset: drm_sched_start(&ring->sched, !tmp_adev->asic_reset_res); } - if (adev->enable_mes) + if (adev->enable_mes && adev->ip_versions[GC_HWIP][0] != IP_VERSION(11, 0, 3)) amdgpu_mes_self_test(tmp_adev); if (!drm_drv_uses_atomic_modeset(adev_to_drm(tmp_adev)) && !job_signaled) { -- cgit From f543d28687480fad06b708bc6e0b0b6ec953b078 Mon Sep 17 00:00:00 2001 From: Prike Liang Date: Fri, 21 Oct 2022 10:04:40 +0800 Subject: drm/amdgpu: disallow gfxoff until GC IP blocks complete s2idle resume In the S2idle suspend/resume phase the gfxoff is keeping functional so some IP blocks will be likely to reinitialize at gfxoff entry and that will result in failing to program GC registers.Therefore, let disallow gfxoff until AMDGPU IPs reinitialized completely. Signed-off-by: Prike Liang Acked-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 5b8362727226..ddaecb2610c9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -3210,6 +3210,15 @@ static int amdgpu_device_ip_resume_phase2(struct amdgpu_device *adev) return r; } adev->ip_blocks[i].status.hw = true; + + if (adev->in_s0ix && adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) { + /* disable gfxoff for IP resume. The gfxoff will be re-enabled in + * amdgpu_device_resume() after IP resume. + */ + amdgpu_gfx_off_ctrl(adev, false); + DRM_DEBUG("will disable gfxoff for re-initializing other blocks\n"); + } + } return 0; @@ -4185,6 +4194,13 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon) /* Make sure IB tests flushed */ flush_delayed_work(&adev->delayed_init_work); + if (adev->in_s0ix) { + /* re-enable gfxoff after IP resume. This re-enables gfxoff after + * it was disabled for IP resume in amdgpu_device_ip_resume_phase2(). + */ + amdgpu_gfx_off_ctrl(adev, true); + DRM_DEBUG("will enable gfxoff for the mission mode\n"); + } if (fbcon) drm_fb_helper_set_suspend_unlocked(adev_to_drm(adev)->fb_helper, false); -- cgit From 12024b17617af9b953cf8801a6a3d3c769ebd06f Mon Sep 17 00:00:00 2001 From: wangjianli Date: Sat, 22 Oct 2022 14:01:55 +0800 Subject: amd/amdgpu: fix repeated words in comments Delete the redundant word 'the'. Signed-off-by: wangjianli Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index ddaecb2610c9..6f958603c8cc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -1568,7 +1568,7 @@ static int amdgpu_device_check_arguments(struct amdgpu_device *adev) * @pdev: pci dev pointer * @state: vga_switcheroo state * - * Callback for the switcheroo driver. Suspends or resumes the + * Callback for the switcheroo driver. Suspends or resumes * the asics before or after it is powered up using ACPI methods. */ static void amdgpu_switcheroo_set_state(struct pci_dev *pdev, -- cgit From 7863c1552632223967f561654422cd2016c2520c Mon Sep 17 00:00:00 2001 From: Mario Limonciello Date: Wed, 26 Oct 2022 14:03:55 -0500 Subject: drm/amd: Fail the suspend if resources can't be evicted MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If a system does not have swap and memory is under 100% usage, amdgpu will fail to evict resources. Currently the suspend carries on proceeding to reset the GPU: ``` [drm] evicting device resources failed [drm:amdgpu_device_ip_suspend_phase2 [amdgpu]] *ERROR* suspend of IP block failed -12 [drm] free PSP TMR buffer [TTM] Failed allocating page table [drm] evicting device resources failed amdgpu 0000:03:00.0: amdgpu: MODE1 reset amdgpu 0000:03:00.0: amdgpu: GPU mode1 reset amdgpu 0000:03:00.0: amdgpu: GPU smu mode1 reset ``` At this point if the suspend actually succeeded I think that amdgpu would have recovered because the GPU would have power cut off and restored. However the kernel fails to continue the suspend from the memory pressure and amdgpu fails to run the "resume" from the aborted suspend. ``` ACPI: PM: Preparing to enter system sleep state S3 SLUB: Unable to allocate memory on node -1, gfp=0xdc0(GFP_KERNEL|__GFP_ZERO) cache: Acpi-State, object size: 80, buffer size: 80, default order: 0, min order: 0 node 0: slabs: 22, objs: 1122, free: 0 ACPI Error: AE_NO_MEMORY, Could not update object reference count (20210730/utdelete-651) [drm:psp_hw_start [amdgpu]] *ERROR* PSP load kdb failed! [drm:psp_resume [amdgpu]] *ERROR* PSP resume failed [drm:amdgpu_device_fw_loading [amdgpu]] *ERROR* resume of IP block failed -62 amdgpu 0000:03:00.0: amdgpu: amdgpu_device_ip_resume failed (-62). PM: dpm_run_callback(): pci_pm_resume+0x0/0x100 returns -62 amdgpu 0000:03:00.0: PM: failed to resume async: error -62 ``` To avoid this series of unfortunate events, fail amdgpu's suspend when the memory eviction fails. This will let the system gracefully recover and the user can try suspend again when the memory pressure is relieved. Reported-by: post@davidak.de Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2223 Signed-off-by: Mario Limonciello Reviewed-by: Alex Deucher Acked-by: Christian König Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 6f958603c8cc..ae10acede495 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -4060,15 +4060,18 @@ void amdgpu_device_fini_sw(struct amdgpu_device *adev) * at suspend time. * */ -static void amdgpu_device_evict_resources(struct amdgpu_device *adev) +static int amdgpu_device_evict_resources(struct amdgpu_device *adev) { + int ret; + /* No need to evict vram on APUs for suspend to ram or s2idle */ if ((adev->in_s3 || adev->in_s0ix) && (adev->flags & AMD_IS_APU)) - return; + return 0; - if (amdgpu_ttm_evict_resources(adev, TTM_PL_VRAM)) + ret = amdgpu_ttm_evict_resources(adev, TTM_PL_VRAM); + if (ret) DRM_WARN("evicting device resources failed\n"); - + return ret; } /* @@ -4118,7 +4121,9 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon) if (!adev->in_s0ix) amdgpu_amdkfd_suspend(adev, adev->in_runpm); - amdgpu_device_evict_resources(adev); + r = amdgpu_device_evict_resources(adev); + if (r) + return r; amdgpu_fence_driver_hw_fini(adev); -- cgit From 8a1fbb4a5ec29d9ae5f792a1d3c049c5c709851c Mon Sep 17 00:00:00 2001 From: Yiqing Yao Date: Fri, 28 Oct 2022 17:30:17 +0800 Subject: drm/amdgpu: Disable MCBP from soc21 for SRIOV [why] Start from soc21, CP does not support MCBP, so disable it. [how] Used amgpu_mcbp flag alone instead of checking if is in SRIOV to enable/disable MCBP. Only set flag to enable on asic_type prior to soc21 in SRIOV. Signed-off-by: Yiqing Yao Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 6 +++--- drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c | 4 ++++ drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 4 ++-- drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 2 +- 5 files changed, 11 insertions(+), 7 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index ae10acede495..522820eeaa59 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -2397,7 +2397,7 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) adev->ip_blocks[i].status.hw = true; /* right after GMC hw init, we create CSA */ - if (amdgpu_mcbp || amdgpu_sriov_vf(adev)) { + if (amdgpu_mcbp) { r = amdgpu_allocate_static_csa(adev, &adev->virt.csa_obj, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_CSA_SIZE); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 4e42dcb1950f..d75e0370a074 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -796,7 +796,7 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) dev_info->ids_flags = 0; if (adev->flags & AMD_IS_APU) dev_info->ids_flags |= AMDGPU_IDS_FLAGS_FUSION; - if (amdgpu_mcbp || amdgpu_sriov_vf(adev)) + if (amdgpu_mcbp) dev_info->ids_flags |= AMDGPU_IDS_FLAGS_PREEMPTION; if (amdgpu_is_tmz(adev)) dev_info->ids_flags |= AMDGPU_IDS_FLAGS_TMZ; @@ -1172,7 +1172,7 @@ int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv) goto error_vm; } - if (amdgpu_mcbp || amdgpu_sriov_vf(adev)) { + if (amdgpu_mcbp) { uint64_t csa_addr = amdgpu_csa_vaddr(adev) & AMDGPU_GMC_HOLE_MASK; r = amdgpu_map_static_csa(adev, &fpriv->vm, adev->virt.csa_obj, @@ -1236,7 +1236,7 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev, if (amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_VCE) != NULL) amdgpu_vce_free_handles(adev, file_priv); - if (amdgpu_mcbp || amdgpu_sriov_vf(adev)) { + if (amdgpu_mcbp) { /* TODO: how to handle reserve failure */ BUG_ON(amdgpu_bo_reserve(adev->virt.csa_obj, true)); amdgpu_vm_bo_del(adev, fpriv->csa_va); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c index c73abe54d974..a226a6c48fb7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c @@ -64,6 +64,10 @@ void amdgpu_virt_init_setting(struct amdgpu_device *adev) ddev->driver_features &= ~DRIVER_ATOMIC; adev->cg_flags = 0; adev->pg_flags = 0; + + /* enable mcbp for sriov asic_type before soc21 */ + amdgpu_mcbp = (adev->asic_type < CHIP_IP_DISCOVERY) ? 1 : 0; + } void amdgpu_virt_kiq_reg_write_reg_wait(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index af94ac580d3e..ff3c127cf70d 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -8489,7 +8489,7 @@ static void gfx_v10_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, control |= ib->length_dw | (vmid << 24); - if ((amdgpu_sriov_vf(ring->adev) || amdgpu_mcbp) && (ib->flags & AMDGPU_IB_FLAG_PREEMPT)) { + if (amdgpu_mcbp && (ib->flags & AMDGPU_IB_FLAG_PREEMPT)) { control |= INDIRECT_BUFFER_PRE_ENB(1); if (flags & AMDGPU_IB_PREEMPTED) @@ -8664,7 +8664,7 @@ static void gfx_v10_0_ring_emit_cntxcntl(struct amdgpu_ring *ring, { uint32_t dw2 = 0; - if (amdgpu_mcbp || amdgpu_sriov_vf(ring->adev)) + if (amdgpu_mcbp) gfx_v10_0_ring_emit_ce_meta(ring, (!amdgpu_sriov_vf(ring->adev) && flags & AMDGPU_IB_PREEMPTED) ? true : false); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index f68e13b6282c..84a76c36d9a7 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -5326,7 +5326,7 @@ static void gfx_v11_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, control |= ib->length_dw | (vmid << 24); - if ((amdgpu_sriov_vf(ring->adev) || amdgpu_mcbp) && (ib->flags & AMDGPU_IB_FLAG_PREEMPT)) { + if (amdgpu_mcbp && (ib->flags & AMDGPU_IB_FLAG_PREEMPT)) { control |= INDIRECT_BUFFER_PRE_ENB(1); if (flags & AMDGPU_IB_PREEMPTED) -- cgit From ec4927d463cbf385be212af1ade9f5a6f1474d4f Mon Sep 17 00:00:00 2001 From: Victor Zhao Date: Wed, 26 Oct 2022 11:48:50 +0800 Subject: drm/amdgpu: fix for suspend/resume sequence under sriov - clear kiq ring after suspend/resume under sriov to aviod kiq ring test failure - update irq after resume to fix kiq interrput loss Signed-off-by: Victor Zhao Acked-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 ++ drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 2 ++ drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 2 ++ drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 2 ++ drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 2 ++ 5 files changed, 10 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 522820eeaa59..5b9f992e4607 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -4197,6 +4197,8 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon) } /* Make sure IB tests flushed */ + if (amdgpu_sriov_vf(adev)) + amdgpu_irq_gpu_reset_resume_helper(adev); flush_delayed_work(&adev->delayed_init_work); if (adev->in_s0ix) { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index 7853d3ca58cf..49d34c7bbf20 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -6909,6 +6909,8 @@ static int gfx_v10_0_kiq_init_queue(struct amdgpu_ring *ring) mutex_unlock(&adev->srbm_mutex); } else { memset((void *)mqd, 0, sizeof(*mqd)); + if (amdgpu_sriov_vf(adev) && adev->in_suspend) + amdgpu_ring_clear_ring(ring); mutex_lock(&adev->srbm_mutex); nv_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0); amdgpu_ring_init_mqd(ring); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index 96b9054c790c..9447999a3a48 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -4004,6 +4004,8 @@ static int gfx_v11_0_kiq_init_queue(struct amdgpu_ring *ring) mutex_unlock(&adev->srbm_mutex); } else { memset((void *)mqd, 0, sizeof(*mqd)); + if (amdgpu_sriov_vf(adev) && adev->in_suspend) + amdgpu_ring_clear_ring(ring); mutex_lock(&adev->srbm_mutex); soc21_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0); amdgpu_ring_init_mqd(ring); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 7f0b18b0d4c4..d47135606e3e 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -4643,6 +4643,8 @@ static int gfx_v8_0_kiq_init_queue(struct amdgpu_ring *ring) memset((void *)mqd, 0, sizeof(struct vi_mqd_allocation)); ((struct vi_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF; ((struct vi_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF; + if (amdgpu_sriov_vf(adev) && adev->in_suspend) + amdgpu_ring_clear_ring(ring); mutex_lock(&adev->srbm_mutex); vi_srbm_select(adev, ring->me, ring->pipe, ring->queue, 0); gfx_v8_0_mqd_init(ring); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index ab30c18decfb..676832da75eb 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -3581,6 +3581,8 @@ static int gfx_v9_0_kiq_init_queue(struct amdgpu_ring *ring) memset((void *)mqd, 0, sizeof(struct v9_mqd_allocation)); ((struct v9_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF; ((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF; + if (amdgpu_sriov_vf(adev) && adev->in_suspend) + amdgpu_ring_clear_ring(ring); mutex_lock(&adev->srbm_mutex); soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0); gfx_v9_0_mqd_init(ring); -- cgit From 45b64fd9f7ae2cce27f85f7f0a7b1fcdd08d06b4 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Thu, 3 Nov 2022 16:14:45 +0100 Subject: drm/fb-helper: Remove unnecessary include statements Remove include statements for where it is not required (i.e., most of them). In a few places include other header files that are required by the source code. v3: * fix amdgpu include statements * fix rockchip include statements Signed-off-by: Thomas Zimmermann Reviewed-by: Javier Martinez Canillas Link: https://patchwork.freedesktop.org/patch/msgid/20221103151446.2638-23-tzimmermann@suse.de --- drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c | 1 - drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 1 + drivers/gpu/drm/amd/amdgpu/amdgpu_display.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 1 + drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h | 1 - drivers/gpu/drm/amd/amdgpu/dce_v10_0.c | 1 + drivers/gpu/drm/amd/amdgpu/dce_v11_0.c | 1 + drivers/gpu/drm/amd/amdgpu/dce_v6_0.c | 1 + drivers/gpu/drm/amd/amdgpu/dce_v8_0.c | 1 + drivers/gpu/drm/arm/hdlcd_crtc.c | 1 - drivers/gpu/drm/ast/ast_drv.h | 1 - drivers/gpu/drm/bridge/tc358762.c | 2 +- drivers/gpu/drm/drm_crtc_helper.c | 1 - drivers/gpu/drm/drm_gem_framebuffer_helper.c | 1 - drivers/gpu/drm/drm_probe_helper.c | 1 - drivers/gpu/drm/etnaviv/etnaviv_drv.h | 3 ++- drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h | 1 - drivers/gpu/drm/hyperv/hyperv_drm_modeset.c | 1 - drivers/gpu/drm/imx/imx-ldb.c | 2 +- drivers/gpu/drm/imx/imx-tve.c | 1 - drivers/gpu/drm/imx/parallel-display.c | 2 +- drivers/gpu/drm/kmb/kmb_plane.c | 1 - drivers/gpu/drm/mgag200/mgag200_drv.h | 1 - drivers/gpu/drm/qxl/qxl_drv.h | 1 - drivers/gpu/drm/rockchip/rockchip_drm_drv.h | 2 +- drivers/gpu/drm/rockchip/rockchip_drm_gem.c | 1 + drivers/gpu/drm/tidss/tidss_kms.c | 1 - drivers/gpu/drm/v3d/v3d_drv.c | 1 - drivers/gpu/drm/vboxvideo/vbox_main.c | 1 - drivers/gpu/drm/virtio/virtgpu_drv.h | 1 - drivers/gpu/drm/xen/xen_drm_front_gem.c | 1 - 31 files changed, 14 insertions(+), 24 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c index 491d4846fc02..e1320edfc527 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c @@ -26,7 +26,6 @@ #include #include -#include #include #include #include "amdgpu.h" diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index e0445e8cc342..cc65df9f2419 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -37,6 +37,7 @@ #include #include +#include #include #include #include diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c index dd6f9ae6fbe9..311a8ea6f065 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c @@ -39,8 +39,8 @@ #include #include #include -#include #include +#include #include #include diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index fe23e09eec98..b495eff635a3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -29,6 +29,7 @@ #include "amdgpu.h" #include #include +#include #include "amdgpu_uvd.h" #include "amdgpu_vce.h" #include "atom.h" diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h index 37322550d750..8a39300b1a84 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h @@ -36,7 +36,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c index 05051d5d2ec3..90f87b2d985b 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c @@ -21,6 +21,7 @@ * */ +#include #include #include diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c index c928bc9eb202..0352de72c886 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c @@ -21,6 +21,7 @@ * */ +#include #include #include diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c index 62315fd5a05f..07bd16e82046 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c @@ -23,6 +23,7 @@ #include +#include #include #include diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c index 87d5e4c21cb3..d73df100f2b3 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c @@ -21,6 +21,7 @@ * */ +#include #include #include diff --git a/drivers/gpu/drm/arm/hdlcd_crtc.c b/drivers/gpu/drm/arm/hdlcd_crtc.c index 7030339fa232..ddbe1dd2d44e 100644 --- a/drivers/gpu/drm/arm/hdlcd_crtc.c +++ b/drivers/gpu/drm/arm/hdlcd_crtc.c @@ -19,7 +19,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h index 74f41282444f..d51b81fea9c8 100644 --- a/drivers/gpu/drm/ast/ast_drv.h +++ b/drivers/gpu/drm/ast/ast_drv.h @@ -38,7 +38,6 @@ #include #include #include -#include #define DRIVER_AUTHOR "Dave Airlie" diff --git a/drivers/gpu/drm/bridge/tc358762.c b/drivers/gpu/drm/bridge/tc358762.c index 7f4fce1aa998..0b6a28436885 100644 --- a/drivers/gpu/drm/bridge/tc358762.c +++ b/drivers/gpu/drm/bridge/tc358762.c @@ -11,6 +11,7 @@ */ #include +#include #include #include #include @@ -19,7 +20,6 @@ #include #include -#include #include #include #include diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c index f5fb22e0d033..a209659a996c 100644 --- a/drivers/gpu/drm/drm_crtc_helper.c +++ b/drivers/gpu/drm/drm_crtc_helper.c @@ -43,7 +43,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/gpu/drm/drm_gem_framebuffer_helper.c b/drivers/gpu/drm/drm_gem_framebuffer_helper.c index e35e224e6303..e93533b86037 100644 --- a/drivers/gpu/drm/drm_gem_framebuffer_helper.c +++ b/drivers/gpu/drm/drm_gem_framebuffer_helper.c @@ -9,7 +9,6 @@ #include #include -#include #include #include #include diff --git a/drivers/gpu/drm/drm_probe_helper.c b/drivers/gpu/drm/drm_probe_helper.c index 2fc21df709bc..bcd9611dabfd 100644 --- a/drivers/gpu/drm/drm_probe_helper.c +++ b/drivers/gpu/drm/drm_probe_helper.c @@ -36,7 +36,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.h b/drivers/gpu/drm/etnaviv/etnaviv_drv.h index f32f4771dada..2bb4c25565dc 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_drv.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.h @@ -6,13 +6,14 @@ #ifndef __ETNAVIV_DRV_H__ #define __ETNAVIV_DRV_H__ +#include #include #include #include #include #include -#include +#include #include #include #include diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h index 4a0cd22c10e2..f957552c6c50 100644 --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h @@ -19,7 +19,6 @@ #include #include -#include #include struct hibmc_connector { diff --git a/drivers/gpu/drm/hyperv/hyperv_drm_modeset.c b/drivers/gpu/drm/hyperv/hyperv_drm_modeset.c index 28e732f94bf2..6c6b57298797 100644 --- a/drivers/gpu/drm/hyperv/hyperv_drm_modeset.c +++ b/drivers/gpu/drm/hyperv/hyperv_drm_modeset.c @@ -8,7 +8,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/gpu/drm/imx/imx-ldb.c b/drivers/gpu/drm/imx/imx-ldb.c index 41799011f73b..c45fc8f4744d 100644 --- a/drivers/gpu/drm/imx/imx-ldb.c +++ b/drivers/gpu/drm/imx/imx-ldb.c @@ -7,6 +7,7 @@ #include #include +#include #include #include #include @@ -23,7 +24,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/gpu/drm/imx/imx-tve.c b/drivers/gpu/drm/imx/imx-tve.c index 6b34fac3f73a..d64ebd2cf15e 100644 --- a/drivers/gpu/drm/imx/imx-tve.c +++ b/drivers/gpu/drm/imx/imx-tve.c @@ -19,7 +19,6 @@ #include #include -#include #include #include #include diff --git a/drivers/gpu/drm/imx/parallel-display.c b/drivers/gpu/drm/imx/parallel-display.c index 06723b2e9b84..0fa0b590830b 100644 --- a/drivers/gpu/drm/imx/parallel-display.c +++ b/drivers/gpu/drm/imx/parallel-display.c @@ -8,6 +8,7 @@ #include #include #include +#include #include #include @@ -16,7 +17,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/gpu/drm/kmb/kmb_plane.c b/drivers/gpu/drm/kmb/kmb_plane.c index a42f63f6f957..d172a302f902 100644 --- a/drivers/gpu/drm/kmb/kmb_plane.c +++ b/drivers/gpu/drm/kmb/kmb_plane.c @@ -9,7 +9,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/gpu/drm/mgag200/mgag200_drv.h b/drivers/gpu/drm/mgag200/mgag200_drv.h index f0c2349404b4..9e604dbb8e44 100644 --- a/drivers/gpu/drm/mgag200/mgag200_drv.h +++ b/drivers/gpu/drm/mgag200/mgag200_drv.h @@ -18,7 +18,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/gpu/drm/qxl/qxl_drv.h b/drivers/gpu/drm/qxl/qxl_drv.h index 432758ad39a3..76f060810f63 100644 --- a/drivers/gpu/drm/qxl/qxl_drv.h +++ b/drivers/gpu/drm/qxl/qxl_drv.h @@ -38,7 +38,6 @@ #include #include -#include #include #include #include diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.h b/drivers/gpu/drm/rockchip/rockchip_drm_drv.h index 1641440837af..aeb03a57240f 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.h +++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.h @@ -9,10 +9,10 @@ #ifndef _ROCKCHIP_DRM_DRV_H #define _ROCKCHIP_DRM_DRV_H -#include #include #include +#include #include #include diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c index 614e97aaac80..bf1120e0f573 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c @@ -9,6 +9,7 @@ #include #include +#include #include #include #include diff --git a/drivers/gpu/drm/tidss/tidss_kms.c b/drivers/gpu/drm/tidss/tidss_kms.c index afb2879980c6..345bcc3011e4 100644 --- a/drivers/gpu/drm/tidss/tidss_kms.c +++ b/drivers/gpu/drm/tidss/tidss_kms.c @@ -10,7 +10,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/gpu/drm/v3d/v3d_drv.c b/drivers/gpu/drm/v3d/v3d_drv.c index e8c975b81585..478f1f0f60de 100644 --- a/drivers/gpu/drm/v3d/v3d_drv.c +++ b/drivers/gpu/drm/v3d/v3d_drv.c @@ -22,7 +22,6 @@ #include #include -#include #include #include diff --git a/drivers/gpu/drm/vboxvideo/vbox_main.c b/drivers/gpu/drm/vboxvideo/vbox_main.c index c9e8b3a63c62..3b83e550f4df 100644 --- a/drivers/gpu/drm/vboxvideo/vbox_main.c +++ b/drivers/gpu/drm/vboxvideo/vbox_main.c @@ -11,7 +11,6 @@ #include #include -#include #include #include diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.h b/drivers/gpu/drm/virtio/virtgpu_drv.h index 9b98470593b0..b7a64c7dcc2c 100644 --- a/drivers/gpu/drm/virtio/virtgpu_drv.h +++ b/drivers/gpu/drm/virtio/virtgpu_drv.h @@ -35,7 +35,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/gpu/drm/xen/xen_drm_front_gem.c b/drivers/gpu/drm/xen/xen_drm_front_gem.c index e31554d7139f..4c95ebcdcc2d 100644 --- a/drivers/gpu/drm/xen/xen_drm_front_gem.c +++ b/drivers/gpu/drm/xen/xen_drm_front_gem.c @@ -12,7 +12,6 @@ #include #include -#include #include #include #include -- cgit From 25263da37693c764e198b31ab4c647934bf80e07 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 19 Jul 2022 12:13:08 -0400 Subject: drm/amdgpu: rework SR-IOV virtual display handling MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit virtual display is enabled unconditionally in SR-IOV, but without specifying the virtual_display module, the number of crtcs defaults to 0. Set a single display by default for SR-IOV if the virtual_display parameter is not set. Only enable virtual display by default on SR-IOV on asics which actually have display hardware. Acked-by: Christian König Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 ++ drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 13 +++++++++++-- drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 18 +++++++++++++++--- drivers/gpu/drm/amd/amdgpu/vi.c | 6 ++++-- 4 files changed, 32 insertions(+), 7 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 9999c18e7d8e..117902891fab 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1122,6 +1122,8 @@ void amdgpu_device_indirect_wreg64(struct amdgpu_device *adev, bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type); bool amdgpu_device_has_dc_support(struct amdgpu_device *adev); +void amdgpu_device_set_sriov_virtual_display(struct amdgpu_device *adev); + int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev, struct amdgpu_reset_context *reset_context); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 5b9f992e4607..72d803060730 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -1915,6 +1915,16 @@ static void amdgpu_device_enable_virtual_display(struct amdgpu_device *adev) } } +void amdgpu_device_set_sriov_virtual_display(struct amdgpu_device *adev) +{ + if (amdgpu_sriov_vf(adev) && !adev->enable_virtual_display) { + adev->mode_info.num_crtc = 1; + adev->enable_virtual_display = true; + DRM_INFO("virtual_display:%d, num_crtc:%d\n", + adev->enable_virtual_display, adev->mode_info.num_crtc); + } +} + /** * amdgpu_device_parse_gpu_info_fw - parse gpu info firmware * @@ -3347,8 +3357,7 @@ bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type) */ bool amdgpu_device_has_dc_support(struct amdgpu_device *adev) { - if (amdgpu_sriov_vf(adev) || - adev->enable_virtual_display || + if (adev->enable_virtual_display || (adev->harvest_ip_mask & AMD_HARVEST_IP_DMU_MASK)) return false; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index 5ea9afaaf4f1..104de4bc5177 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -1702,9 +1702,15 @@ static int amdgpu_discovery_set_smu_ip_blocks(struct amdgpu_device *adev) return 0; } +static void amdgpu_discovery_set_sriov_display(struct amdgpu_device *adev) +{ + amdgpu_device_set_sriov_virtual_display(adev); + amdgpu_device_ip_block_add(adev, &amdgpu_vkms_ip_block); +} + static int amdgpu_discovery_set_display_ip_blocks(struct amdgpu_device *adev) { - if (adev->enable_virtual_display || amdgpu_sriov_vf(adev)) { + if (adev->enable_virtual_display) { amdgpu_device_ip_block_add(adev, &amdgpu_vkms_ip_block); return 0; } @@ -1732,7 +1738,10 @@ static int amdgpu_discovery_set_display_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(3, 1, 6): case IP_VERSION(3, 2, 0): case IP_VERSION(3, 2, 1): - amdgpu_device_ip_block_add(adev, &dm_ip_block); + if (amdgpu_sriov_vf(adev)) + amdgpu_discovery_set_sriov_display(adev); + else + amdgpu_device_ip_block_add(adev, &dm_ip_block); break; default: dev_err(adev->dev, @@ -1745,7 +1754,10 @@ static int amdgpu_discovery_set_display_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(12, 0, 0): case IP_VERSION(12, 0, 1): case IP_VERSION(12, 1, 0): - amdgpu_device_ip_block_add(adev, &dm_ip_block); + if (amdgpu_sriov_vf(adev)) + amdgpu_discovery_set_sriov_display(adev); + else + amdgpu_device_ip_block_add(adev, &dm_ip_block); break; default: dev_err(adev->dev, diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c index f6ffd7c96ff9..12ef782eb478 100644 --- a/drivers/gpu/drm/amd/amdgpu/vi.c +++ b/drivers/gpu/drm/amd/amdgpu/vi.c @@ -2111,6 +2111,8 @@ void vi_set_virt_ops(struct amdgpu_device *adev) int vi_set_ip_blocks(struct amdgpu_device *adev) { + amdgpu_device_set_sriov_virtual_display(adev); + switch (adev->asic_type) { case CHIP_TOPAZ: /* topaz has no DCE, UVD, VCE */ @@ -2130,7 +2132,7 @@ int vi_set_ip_blocks(struct amdgpu_device *adev) amdgpu_device_ip_block_add(adev, &gfx_v8_0_ip_block); amdgpu_device_ip_block_add(adev, &sdma_v3_0_ip_block); amdgpu_device_ip_block_add(adev, &pp_smu_ip_block); - if (adev->enable_virtual_display || amdgpu_sriov_vf(adev)) + if (adev->enable_virtual_display) amdgpu_device_ip_block_add(adev, &amdgpu_vkms_ip_block); #if defined(CONFIG_DRM_AMD_DC) else if (amdgpu_device_has_dc_support(adev)) @@ -2150,7 +2152,7 @@ int vi_set_ip_blocks(struct amdgpu_device *adev) amdgpu_device_ip_block_add(adev, &gfx_v8_0_ip_block); amdgpu_device_ip_block_add(adev, &sdma_v3_0_ip_block); amdgpu_device_ip_block_add(adev, &pp_smu_ip_block); - if (adev->enable_virtual_display || amdgpu_sriov_vf(adev)) + if (adev->enable_virtual_display) amdgpu_device_ip_block_add(adev, &amdgpu_vkms_ip_block); #if defined(CONFIG_DRM_AMD_DC) else if (amdgpu_device_has_dc_support(adev)) -- cgit From d09ef243035b75a6d403ebfeb7e87fa20d7e25c6 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 19 Jul 2022 14:18:41 -0400 Subject: drm/amdgpu: clarify DC checks MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There are several places where we don't want to check if a particular asic could support DC, but rather, if DC is enabled. Set a flag if DC is enabled and check for that rather than if a device supports DC or not. Acked-by: Christian König Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 1 + drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 32 ++++++++++++----------- drivers/gpu/drm/amd/amdgpu/amdgpu_display.c | 4 +-- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 2 +- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 1 + drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c | 2 +- 8 files changed, 25 insertions(+), 21 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 117902891fab..5a4aaf4d9ed1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1065,6 +1065,7 @@ struct amdgpu_device { struct work_struct reset_work; bool job_hang; + bool dc_enabled; }; static inline struct amdgpu_device *drm_to_adev(struct drm_device *ddev) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c index b14800ac179e..57b5e11446c6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c @@ -847,7 +847,7 @@ int amdgpu_acpi_init(struct amdgpu_device *adev) struct amdgpu_atif *atif = &amdgpu_acpi_priv.atif; if (atif->notifications.brightness_change) { - if (amdgpu_device_has_dc_support(adev)) { + if (adev->dc_enabled) { #if defined(CONFIG_DRM_AMD_DC) struct amdgpu_display_manager *dm = &adev->dm; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index de61a85c4b02..0f16d3c09309 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c @@ -1969,7 +1969,7 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev) amdgpu_ta_if_debugfs_init(adev); #if defined(CONFIG_DRM_AMD_DC) - if (amdgpu_device_has_dc_support(adev)) + if (adev->dc_enabled) dtn_debugfs_init(adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 72d803060730..ee69cc3a5efd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -4224,25 +4224,27 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon) amdgpu_ras_resume(adev); - /* - * Most of the connector probing functions try to acquire runtime pm - * refs to ensure that the GPU is powered on when connector polling is - * performed. Since we're calling this from a runtime PM callback, - * trying to acquire rpm refs will cause us to deadlock. - * - * Since we're guaranteed to be holding the rpm lock, it's safe to - * temporarily disable the rpm helpers so this doesn't deadlock us. - */ + if (adev->mode_info.num_crtc) { + /* + * Most of the connector probing functions try to acquire runtime pm + * refs to ensure that the GPU is powered on when connector polling is + * performed. Since we're calling this from a runtime PM callback, + * trying to acquire rpm refs will cause us to deadlock. + * + * Since we're guaranteed to be holding the rpm lock, it's safe to + * temporarily disable the rpm helpers so this doesn't deadlock us. + */ #ifdef CONFIG_PM - dev->dev->power.disable_depth++; + dev->dev->power.disable_depth++; #endif - if (!amdgpu_device_has_dc_support(adev)) - drm_helper_hpd_irq_event(dev); - else - drm_kms_helper_hotplug_event(dev); + if (!adev->dc_enabled) + drm_helper_hpd_irq_event(dev); + else + drm_kms_helper_hotplug_event(dev); #ifdef CONFIG_PM - dev->dev->power.disable_depth--; + dev->dev->power.disable_depth--; #endif + } adev->in_suspend = false; if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DEV_D0)) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c index 85138157ea6a..83a271578265 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c @@ -549,7 +549,7 @@ uint32_t amdgpu_display_supported_domains(struct amdgpu_device *adev, */ if ((bo_flags & AMDGPU_GEM_CREATE_CPU_GTT_USWC) && amdgpu_bo_support_uswc(bo_flags) && - amdgpu_device_has_dc_support(adev) && + adev->dc_enabled && adev->mode_info.gpu_vm_support) domain |= AMDGPU_GEM_DOMAIN_GTT; #endif @@ -1316,7 +1316,7 @@ int amdgpu_display_modeset_create_props(struct amdgpu_device *adev) "dither", amdgpu_dither_enum_list, sz); - if (amdgpu_device_has_dc_support(adev)) { + if (adev->dc_enabled) { adev->mode_info.abm_level_property = drm_property_create_range(adev_to_drm(adev), 0, "abm level", 0, 4); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index bf2d50c8c92a..3531d9014a64 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -2471,7 +2471,7 @@ static int amdgpu_runtime_idle_check_display(struct device *dev) if (ret) return ret; - if (amdgpu_device_has_dc_support(adev)) { + if (adev->dc_enabled) { struct drm_crtc *crtc; drm_for_each_crtc(crtc, drm_dev) { 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 65715e6e3dd5..123c235b60bc 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -4592,6 +4592,7 @@ static int dm_early_init(void *handle) adev_to_drm(adev)->dev, &dev_attr_s3_debug); #endif + adev->dc_enabled = true; return 0; } diff --git a/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c index 1159ae114dd0..304190d5c9d2 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c +++ b/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c @@ -1508,7 +1508,7 @@ static void pp_pm_compute_clocks(void *handle) struct pp_hwmgr *hwmgr = handle; struct amdgpu_device *adev = hwmgr->adev; - if (!amdgpu_device_has_dc_support(adev)) { + if (!adev->dc_enabled) { amdgpu_dpm_get_active_displays(adev); adev->pm.pm_display_cfg.num_display = adev->pm.dpm.new_active_crtc_count; adev->pm.pm_display_cfg.vrefresh = amdgpu_dpm_get_vrefresh(adev); -- cgit From 220c8cc855ce7075a29926533279a8dfefdd8788 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 11 Nov 2022 12:50:38 -0500 Subject: drm/amdgpu: there is no vbios fb on devices with no display hw (v2) If we enable virtual display functionality on parts with no display hardware we can end up trying to check for and reserve the vbios FB area on devices where it doesn't exist. Check if display hardware is actually present on the hardware before trying to reserve the memory. v2: move the check into common code Acked-by: Evan Quan Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 1 + drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 41 ++++++++++++++++++++++++++++++ drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 2 +- 3 files changed, 43 insertions(+), 1 deletion(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 5a4aaf4d9ed1..1f3a4d596d0d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1298,6 +1298,7 @@ void amdgpu_device_pcie_port_wreg(struct amdgpu_device *adev, u32 reg, u32 v); struct dma_fence *amdgpu_device_switch_gang(struct amdgpu_device *adev, struct dma_fence *gang); +bool amdgpu_device_has_display_hardware(struct amdgpu_device *adev); /* atpx handler */ #if defined(CONFIG_VGA_SWITCHEROO) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index ee69cc3a5efd..31357101c704 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -6057,3 +6057,44 @@ struct dma_fence *amdgpu_device_switch_gang(struct amdgpu_device *adev, dma_fence_put(old); return NULL; } + +bool amdgpu_device_has_display_hardware(struct amdgpu_device *adev) +{ + switch (adev->asic_type) { +#ifdef CONFIG_DRM_AMDGPU_SI + case CHIP_HAINAN: +#endif + case CHIP_TOPAZ: + /* chips with no display hardware */ + return false; +#ifdef CONFIG_DRM_AMDGPU_SI + case CHIP_TAHITI: + case CHIP_PITCAIRN: + case CHIP_VERDE: + case CHIP_OLAND: +#endif +#ifdef CONFIG_DRM_AMDGPU_CIK + case CHIP_BONAIRE: + case CHIP_HAWAII: + case CHIP_KAVERI: + case CHIP_KABINI: + case CHIP_MULLINS: +#endif + case CHIP_TONGA: + case CHIP_FIJI: + case CHIP_POLARIS10: + case CHIP_POLARIS11: + case CHIP_POLARIS12: + case CHIP_VEGAM: + case CHIP_CARRIZO: + case CHIP_STONEY: + /* chips with display hardware */ + return true; + default: + /* IP discovery */ + if (!adev->ip_versions[DCE_HWIP][0] || + (adev->harvest_ip_mask & AMD_HARVEST_IP_DMU_MASK)) + return false; + return true; + } +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 9c0d9baab4e2..4365ede42855 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -657,7 +657,7 @@ void amdgpu_gmc_get_vbios_allocations(struct amdgpu_device *adev) } if (amdgpu_sriov_vf(adev) || - !amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_DCE)) { + !amdgpu_device_has_display_hardware(adev)) { size = 0; } else { size = amdgpu_gmc_get_vbios_fb_size(adev); -- cgit From d293470e10d02ce1afb59cfe43fdf1a6cdb3d726 Mon Sep 17 00:00:00 2001 From: YiPeng Chai Date: Fri, 4 Nov 2022 17:50:12 +0800 Subject: drm/amdgpu: Fixed the problem that ras error can't be queried after gpu recovery is completed Amdgpu_ras_set_error_query_ready is called at the start of amdgpu_device_gpu_recover to disable query ras error, but the code behind only enables query ras error in full reset path, but not in soft reset path, emergency restart path and skip the hardware reset path. Signed-off-by: YiPeng Chai Reviewed-by: Hawking Zhang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 31357101c704..947ed5d97942 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -5454,6 +5454,8 @@ skip_sched_resume: amdgpu_device_resume_display_audio(tmp_adev); amdgpu_device_unset_mp1_state(tmp_adev); + + amdgpu_ras_set_error_query_ready(tmp_adev, true); } recover_end: -- cgit From 06a2d7cc3f0476be4682ef90eb09a28fa3daed37 Mon Sep 17 00:00:00 2001 From: Christian König Date: Wed, 26 Oct 2022 12:26:37 +0200 Subject: drm/amdgpu: revert "implement tdr advanced mode" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This reverts commit e6c6338f393b74ac0b303d567bb918b44ae7ad75. This feature basically re-submits one job after another to figure out which one was the one causing a hang. This is obviously incompatible with gang-submit which requires that multiple jobs run at the same time. It's also absolutely not helpful to crash the hardware multiple times if a clean recovery is desired. For testing and debugging environments we should rather disable recovery alltogether to be able to inspect the state with a hw debugger. Additional to that the sw implementation is clearly buggy and causes reference count issues for the hardware fence. Signed-off-by: Christian König Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 103 ----------------------------- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 2 +- drivers/gpu/drm/scheduler/sched_main.c | 58 +++------------- include/drm/gpu_scheduler.h | 3 - 4 files changed, 10 insertions(+), 156 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 947ed5d97942..2a8166fb69f2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -5088,94 +5088,6 @@ static int amdgpu_device_suspend_display_audio(struct amdgpu_device *adev) return 0; } -static void amdgpu_device_recheck_guilty_jobs( - struct amdgpu_device *adev, struct list_head *device_list_handle, - struct amdgpu_reset_context *reset_context) -{ - int i, r = 0; - - for (i = 0; i < AMDGPU_MAX_RINGS; ++i) { - struct amdgpu_ring *ring = adev->rings[i]; - int ret = 0; - struct drm_sched_job *s_job; - - if (!ring || !ring->sched.thread) - continue; - - s_job = list_first_entry_or_null(&ring->sched.pending_list, - struct drm_sched_job, list); - if (s_job == NULL) - continue; - - /* clear job's guilty and depend the folowing step to decide the real one */ - drm_sched_reset_karma(s_job); - drm_sched_resubmit_jobs_ext(&ring->sched, 1); - - if (!s_job->s_fence->parent) { - DRM_WARN("Failed to get a HW fence for job!"); - continue; - } - - ret = dma_fence_wait_timeout(s_job->s_fence->parent, false, ring->sched.timeout); - if (ret == 0) { /* timeout */ - DRM_ERROR("Found the real bad job! ring:%s, job_id:%llx\n", - ring->sched.name, s_job->id); - - - amdgpu_fence_driver_isr_toggle(adev, true); - - /* Clear this failed job from fence array */ - amdgpu_fence_driver_clear_job_fences(ring); - - amdgpu_fence_driver_isr_toggle(adev, false); - - /* Since the job won't signal and we go for - * another resubmit drop this parent pointer - */ - dma_fence_put(s_job->s_fence->parent); - s_job->s_fence->parent = NULL; - - /* set guilty */ - drm_sched_increase_karma(s_job); - amdgpu_reset_prepare_hwcontext(adev, reset_context); -retry: - /* do hw reset */ - if (amdgpu_sriov_vf(adev)) { - amdgpu_virt_fini_data_exchange(adev); - r = amdgpu_device_reset_sriov(adev, false); - if (r) - adev->asic_reset_res = r; - } else { - clear_bit(AMDGPU_SKIP_HW_RESET, - &reset_context->flags); - r = amdgpu_do_asic_reset(device_list_handle, - reset_context); - if (r && r == -EAGAIN) - goto retry; - } - - /* - * add reset counter so that the following - * resubmitted job could flush vmid - */ - atomic_inc(&adev->gpu_reset_counter); - continue; - } - - /* got the hw fence, signal finished fence */ - atomic_dec(ring->sched.score); - dma_fence_get(&s_job->s_fence->finished); - dma_fence_signal(&s_job->s_fence->finished); - dma_fence_put(&s_job->s_fence->finished); - - /* remove node from list and free the job */ - spin_lock(&ring->sched.job_list_lock); - list_del_init(&s_job->list); - spin_unlock(&ring->sched.job_list_lock); - ring->sched.ops->free_job(s_job); - } -} - static inline void amdgpu_device_stop_pending_resets(struct amdgpu_device *adev) { struct amdgpu_ras *con = amdgpu_ras_get_context(adev); @@ -5196,7 +5108,6 @@ static inline void amdgpu_device_stop_pending_resets(struct amdgpu_device *adev) } - /** * amdgpu_device_gpu_recover - reset the asic and recover scheduler * @@ -5219,7 +5130,6 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, int i, r = 0; bool need_emergency_restart = false; bool audio_suspended = false; - int tmp_vram_lost_counter; bool gpu_reset_for_dev_remove = false; gpu_reset_for_dev_remove = @@ -5365,7 +5275,6 @@ retry: /* Rest of adevs pre asic reset from XGMI hive. */ amdgpu_device_stop_pending_resets(tmp_adev); } - tmp_vram_lost_counter = atomic_read(&((adev)->vram_lost_counter)); /* Actual ASIC resets if needed.*/ /* Host driver will handle XGMI hive reset for SRIOV */ if (amdgpu_sriov_vf(adev)) { @@ -5390,18 +5299,6 @@ skip_hw_reset: /* Post ASIC reset for all devs .*/ list_for_each_entry(tmp_adev, device_list_handle, reset_list) { - /* - * Sometimes a later bad compute job can block a good gfx job as gfx - * and compute ring share internal GC HW mutually. We add an additional - * guilty jobs recheck step to find the real guilty job, it synchronously - * submits and pends for the first job being signaled. If it gets timeout, - * we identify it as a real guilty job. - */ - if (amdgpu_gpu_recovery == 2 && - !(tmp_vram_lost_counter < atomic_read(&adev->vram_lost_counter))) - amdgpu_device_recheck_guilty_jobs( - tmp_adev, device_list_handle, reset_context); - for (i = 0; i < AMDGPU_MAX_RINGS; ++i) { struct amdgpu_ring *ring = tmp_adev->rings[i]; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 8281d37220cf..72b77b3ac4fb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -533,7 +533,7 @@ module_param_named(compute_multipipe, amdgpu_compute_multipipe, int, 0444); * DOC: gpu_recovery (int) * Set to enable GPU recovery mechanism (1 = enable, 0 = disable). The default is -1 (auto, disabled except SRIOV). */ -MODULE_PARM_DESC(gpu_recovery, "Enable GPU recovery mechanism, (2 = advanced tdr mode, 1 = enable, 0 = disable, -1 = auto)"); +MODULE_PARM_DESC(gpu_recovery, "Enable GPU recovery mechanism, (1 = enable, 0 = disable, -1 = auto)"); module_param_named(gpu_recovery, amdgpu_gpu_recovery, int, 0444); /** diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c index d0ff9e11cb69..0150ac7fbf40 100644 --- a/drivers/gpu/drm/scheduler/sched_main.c +++ b/drivers/gpu/drm/scheduler/sched_main.c @@ -443,27 +443,6 @@ static void drm_sched_job_timedout(struct work_struct *work) } } - /** - * drm_sched_increase_karma - Update sched_entity guilty flag - * - * @bad: The job guilty of time out - * - * Increment on every hang caused by the 'bad' job. If this exceeds the hang - * limit of the scheduler then the respective sched entity is marked guilty and - * jobs from it will not be scheduled further - */ -void drm_sched_increase_karma(struct drm_sched_job *bad) -{ - drm_sched_increase_karma_ext(bad, 1); -} -EXPORT_SYMBOL(drm_sched_increase_karma); - -void drm_sched_reset_karma(struct drm_sched_job *bad) -{ - drm_sched_increase_karma_ext(bad, 0); -} -EXPORT_SYMBOL(drm_sched_reset_karma); - /** * drm_sched_stop - stop the scheduler * @@ -604,32 +583,15 @@ EXPORT_SYMBOL(drm_sched_start); * */ void drm_sched_resubmit_jobs(struct drm_gpu_scheduler *sched) -{ - drm_sched_resubmit_jobs_ext(sched, INT_MAX); -} -EXPORT_SYMBOL(drm_sched_resubmit_jobs); - -/** - * drm_sched_resubmit_jobs_ext - helper to relunch certain number of jobs from mirror ring list - * - * @sched: scheduler instance - * @max: job numbers to relaunch - * - */ -void drm_sched_resubmit_jobs_ext(struct drm_gpu_scheduler *sched, int max) { struct drm_sched_job *s_job, *tmp; uint64_t guilty_context; bool found_guilty = false; struct dma_fence *fence; - int i = 0; list_for_each_entry_safe(s_job, tmp, &sched->pending_list, list) { struct drm_sched_fence *s_fence = s_job->s_fence; - if (i >= max) - break; - if (!found_guilty && atomic_read(&s_job->karma) > sched->hang_limit) { found_guilty = true; guilty_context = s_job->s_fence->scheduled.context; @@ -639,7 +601,6 @@ void drm_sched_resubmit_jobs_ext(struct drm_gpu_scheduler *sched, int max) dma_fence_set_error(&s_fence->finished, -ECANCELED); fence = sched->ops->run_job(s_job); - i++; if (IS_ERR_OR_NULL(fence)) { if (IS_ERR(fence)) @@ -655,7 +616,7 @@ void drm_sched_resubmit_jobs_ext(struct drm_gpu_scheduler *sched, int max) } } } -EXPORT_SYMBOL(drm_sched_resubmit_jobs_ext); +EXPORT_SYMBOL(drm_sched_resubmit_jobs); /** * drm_sched_job_init - init a scheduler job @@ -1172,13 +1133,15 @@ void drm_sched_fini(struct drm_gpu_scheduler *sched) EXPORT_SYMBOL(drm_sched_fini); /** - * drm_sched_increase_karma_ext - Update sched_entity guilty flag + * drm_sched_increase_karma - Update sched_entity guilty flag * * @bad: The job guilty of time out - * @type: type for increase/reset karma * + * Increment on every hang caused by the 'bad' job. If this exceeds the hang + * limit of the scheduler then the respective sched entity is marked guilty and + * jobs from it will not be scheduled further */ -void drm_sched_increase_karma_ext(struct drm_sched_job *bad, int type) +void drm_sched_increase_karma(struct drm_sched_job *bad) { int i; struct drm_sched_entity *tmp; @@ -1190,10 +1153,7 @@ void drm_sched_increase_karma_ext(struct drm_sched_job *bad, int type) * corrupt but keep in mind that kernel jobs always considered good. */ if (bad->s_priority != DRM_SCHED_PRIORITY_KERNEL) { - if (type == 0) - atomic_set(&bad->karma, 0); - else if (type == 1) - atomic_inc(&bad->karma); + atomic_inc(&bad->karma); for (i = DRM_SCHED_PRIORITY_MIN; i < DRM_SCHED_PRIORITY_KERNEL; i++) { @@ -1204,7 +1164,7 @@ void drm_sched_increase_karma_ext(struct drm_sched_job *bad, int type) if (bad->s_fence->scheduled.context == entity->fence_context) { if (entity->guilty) - atomic_set(entity->guilty, type); + atomic_set(entity->guilty, 1); break; } } @@ -1214,4 +1174,4 @@ void drm_sched_increase_karma_ext(struct drm_sched_job *bad, int type) } } } -EXPORT_SYMBOL(drm_sched_increase_karma_ext); +EXPORT_SYMBOL(drm_sched_increase_karma); diff --git a/include/drm/gpu_scheduler.h b/include/drm/gpu_scheduler.h index ca11716d084a..0168ff469ae0 100644 --- a/include/drm/gpu_scheduler.h +++ b/include/drm/gpu_scheduler.h @@ -528,10 +528,7 @@ void drm_sched_wakeup(struct drm_gpu_scheduler *sched); void drm_sched_stop(struct drm_gpu_scheduler *sched, struct drm_sched_job *bad); void drm_sched_start(struct drm_gpu_scheduler *sched, bool full_recovery); void drm_sched_resubmit_jobs(struct drm_gpu_scheduler *sched); -void drm_sched_resubmit_jobs_ext(struct drm_gpu_scheduler *sched, int max); void drm_sched_increase_karma(struct drm_sched_job *bad); -void drm_sched_reset_karma(struct drm_sched_job *bad); -void drm_sched_increase_karma_ext(struct drm_sched_job *bad, int type); bool drm_sched_dependency_optimized(struct dma_fence* fence, struct drm_sched_entity *entity); void drm_sched_fault(struct drm_gpu_scheduler *sched); -- cgit From 6868a2c46560670efc0d1f2b446cc57edcaf960d Mon Sep 17 00:00:00 2001 From: Christian König Date: Wed, 26 Oct 2022 12:57:44 +0200 Subject: drm/amdgpu: stop resubmitting jobs for GPU reset v2 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Re-submitting IBs by the kernel has many problems because pre- requisite state is not automatically re-created as well. In other words neither binary semaphores nor things like ring buffer pointers are in the state they should be when the hardware starts to work on the IBs again. Additional to that even after more than 5 years of developing this feature it is still not stable and we have massively problems getting the reference counts right. As discussed with user space developers this behavior is not helpful in the first place. For graphics and multimedia workloads it makes much more sense to either completely re-create the context or at least re-submitting the IBs from userspace. For compute use cases re-submitting is also not very helpful since userspace must rely on the accuracy of the result. Because of this we stop this practice and instead just properly note that the fence submission was canceled. The only use case we keep the re-submission for now is SRIOV and function level resets. v2: as suggested by Sshaoyun stop resubmitting jobs even for SRIOV Signed-off-by: Christian König Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 2a8166fb69f2..62b508053eb1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -5305,11 +5305,7 @@ skip_hw_reset: if (!ring || !ring->sched.thread) continue; - /* No point to resubmit jobs if we didn't HW reset*/ - if (!tmp_adev->asic_reset_res && !job_signaled) - drm_sched_resubmit_jobs(&ring->sched); - - drm_sched_start(&ring->sched, !tmp_adev->asic_reset_res); + drm_sched_start(&ring->sched, true); } if (adev->enable_mes && adev->ip_versions[GC_HWIP][0] != IP_VERSION(11, 0, 3)) -- cgit From 0788a47e7cec7ebdcb1ad8912754b8b8b06ee915 Mon Sep 17 00:00:00 2001 From: Christian König Date: Wed, 26 Oct 2022 12:48:24 +0200 Subject: drm/amdgpu: stop resubmittting jobs in amdgpu_pci_resume MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The state of VRAM is unreliable due to a PCI event like AER, link reset or DPC. Signed-off-by: Christian König Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 62b508053eb1..0f5cbc70682a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -5760,8 +5760,6 @@ void amdgpu_pci_resume(struct pci_dev *pdev) if (!ring || !ring->sched.thread) continue; - - drm_sched_resubmit_jobs(&ring->sched); drm_sched_start(&ring->sched, true); } -- cgit From 1a11a65d5395ccdcd07f19a75da82a3d74c368dd Mon Sep 17 00:00:00 2001 From: YiPeng Chai Date: Tue, 8 Nov 2022 17:11:18 +0800 Subject: drm/amdgpu: Enable mode-1 reset for RAS recovery in fatal error mode The patch is enabling mode-1 reset for RAS recovery in fatal error mode. Signed-off-by: YiPeng Chai Reviewed-by: Hawking Zhang Reviewed-by: Tao Zhou Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 ++++ drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 7 ++++++- 2 files changed, 10 insertions(+), 1 deletion(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 0f5cbc70682a..3d5d5d49cfab 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -4593,6 +4593,10 @@ bool amdgpu_device_should_recover_gpu(struct amdgpu_device *adev) if (amdgpu_gpu_recovery == 0) goto disabled; + /* Skip soft reset check in fatal error mode */ + if (!amdgpu_ras_is_poison_mode_supported(adev)) + return true; + if (!amdgpu_device_ip_check_soft_reset(adev)) { dev_info(adev->dev,"Timeout, but no hardware hang detected.\n"); return false; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 693bce07eb46..8fca3cc273e8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -1948,7 +1948,12 @@ static void amdgpu_ras_do_recovery(struct work_struct *work) reset_context.method = AMD_RESET_METHOD_NONE; reset_context.reset_req_dev = adev; - clear_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); + + /* Perform full reset in fatal error mode */ + if (!amdgpu_ras_is_poison_mode_supported(ras->adev)) + set_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); + else + clear_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); amdgpu_device_gpu_recover(ras->adev, NULL, &reset_context); } -- cgit From b85e285e3d6352b02947fc1b72303673dfacb0aa Mon Sep 17 00:00:00 2001 From: Yang Yingliang Date: Thu, 17 Nov 2022 23:00:03 +0800 Subject: drm/amdgpu: fix pci device refcount leak As comment of pci_get_domain_bus_and_slot() says, it returns a pci device with refcount increment, when finish using it, the caller must decrement the reference count by calling pci_dev_put(). So before returning from amdgpu_device_resume|suspend_display_audio(), pci_dev_put() is called to avoid refcount leak. Fixes: 3f12acc8d6d4 ("drm/amdgpu: put the audio codec into suspend state before gpu reset V3") Reviewed-by: Evan Quan Signed-off-by: Yang Yingliang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index b2b1c66bfe39..3a6b24a9e3a1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -5045,6 +5045,8 @@ static void amdgpu_device_resume_display_audio(struct amdgpu_device *adev) pm_runtime_enable(&(p->dev)); pm_runtime_resume(&(p->dev)); } + + pci_dev_put(p); } static int amdgpu_device_suspend_display_audio(struct amdgpu_device *adev) @@ -5083,6 +5085,7 @@ static int amdgpu_device_suspend_display_audio(struct amdgpu_device *adev) if (expires < ktime_get_mono_fast_ns()) { dev_warn(adev->dev, "failed to suspend display audio\n"); + pci_dev_put(p); /* TODO: abort the succeeding gpu reset? */ return -ETIMEDOUT; } @@ -5090,6 +5093,7 @@ static int amdgpu_device_suspend_display_audio(struct amdgpu_device *adev) pm_runtime_disable(&(p->dev)); + pci_dev_put(p); return 0; } -- cgit From 3c22c1ead6b2e6a9c0f2eeef143948f5d701dd08 Mon Sep 17 00:00:00 2001 From: Shikang Fan Date: Fri, 18 Nov 2022 17:35:33 +0800 Subject: drm/amdgpu: fix for suspend/resume kiq fence fallback under sriov - in device_resume, sriov configure interrupt should be in full access, so release_full_gpu should be done after kfd_resume. - remove the previous workaround solution for sriov. Fixes: ec4927d463cb ("drm/amdgpu: fix for suspend/resume sequence under sriov") Signed-off-by: Shikang Fan Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 23 ++++++++++++----------- 1 file changed, 12 insertions(+), 11 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 3a6b24a9e3a1..9f7741871bb4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -4181,21 +4181,15 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon) r = amdgpu_device_ip_resume(adev); - /* no matter what r is, always need to properly release full GPU */ - if (amdgpu_sriov_vf(adev)) { - amdgpu_virt_init_data_exchange(adev); - amdgpu_virt_release_full_gpu(adev, true); - } - if (r) { dev_err(adev->dev, "amdgpu_device_ip_resume failed (%d).\n", r); - return r; + goto exit; } amdgpu_fence_driver_hw_init(adev); r = amdgpu_device_ip_late_init(adev); if (r) - return r; + goto exit; queue_delayed_work(system_wq, &adev->delayed_init_work, msecs_to_jiffies(AMDGPU_RESUME_MS)); @@ -4203,12 +4197,19 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon) if (!adev->in_s0ix) { r = amdgpu_amdkfd_resume(adev, adev->in_runpm); if (r) - return r; + goto exit; + } + +exit: + if (amdgpu_sriov_vf(adev)) { + amdgpu_virt_init_data_exchange(adev); + amdgpu_virt_release_full_gpu(adev, true); } + if (r) + return r; + /* Make sure IB tests flushed */ - if (amdgpu_sriov_vf(adev)) - amdgpu_irq_gpu_reset_resume_helper(adev); flush_delayed_work(&adev->delayed_init_work); if (adev->in_s0ix) { -- cgit From dfd0287bd3920e132a8dae2a0ec3d92eaff5f2dd Mon Sep 17 00:00:00 2001 From: Liang He Date: Tue, 22 Nov 2022 12:28:49 +0800 Subject: drm/amdgpu: Fix potential double free and null pointer dereference In amdgpu_get_xgmi_hive(), we should not call kfree() after kobject_put() as the PUT will call kfree(). In amdgpu_device_ip_init(), we need to check the returned *hive* which can be NULL before we dereference it. Signed-off-by: Liang He Reviewed-by: Luben Tuikov Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 5 +++++ drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c | 2 -- 2 files changed, 5 insertions(+), 2 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 9f7741871bb4..cfa411c12072 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -2473,6 +2473,11 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) if (!amdgpu_sriov_vf(adev)) { struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev); + if (WARN_ON(!hive)) { + r = -ENOENT; + goto init_failed; + } + if (!hive->reset_domain || !amdgpu_reset_get_reset_domain(hive->reset_domain)) { r = -ENOENT; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index 47159e9a0884..4b9e7b050ccd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -386,7 +386,6 @@ struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev) if (ret) { dev_err(adev->dev, "XGMI: failed initializing kobject for xgmi hive\n"); kobject_put(&hive->kobj); - kfree(hive); hive = NULL; goto pro_end; } @@ -410,7 +409,6 @@ struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev) dev_err(adev->dev, "XGMI: failed initializing reset domain for xgmi hive\n"); ret = -ENOMEM; kobject_put(&hive->kobj); - kfree(hive); hive = NULL; goto pro_end; } -- cgit From 47ea20762bb7875a62e10433a3cd5d34e9133f47 Mon Sep 17 00:00:00 2001 From: Shikang Fan Date: Thu, 8 Dec 2022 19:53:14 +0800 Subject: drm/amdgpu: Add an extra evict_resource call during device_suspend. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - evict_resource is taking too long causing sriov full access mode timeout. So, add an extra evict_resource in the beginning as an early evict. Signed-off-by: Shikang Fan Reviewed-by: Christian König Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index cfa411c12072..64660a41d53c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -4112,6 +4112,11 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon) adev->in_suspend = true; + /* Evict the majority of BOs before grabbing the full access */ + r = amdgpu_device_evict_resources(adev); + if (r) + return r; + if (amdgpu_sriov_vf(adev)) { amdgpu_virt_fini_data_exchange(adev); r = amdgpu_virt_request_full_gpu(adev, false); -- cgit From afa6646b1c5d3affd541f76bd7476e4b835a9174 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 16 Dec 2022 11:42:20 -0500 Subject: drm/amdgpu: skip MES for S0ix as well since it's part of GFX It's also part of gfxoff. Cc: stable@vger.kernel.org # 6.0, 6.1 Reviewed-by: Mario Limonciello Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_device.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 64660a41d53c..afe6af9c0138 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -3016,14 +3016,15 @@ static int amdgpu_device_ip_suspend_phase2(struct amdgpu_device *adev) continue; } - /* skip suspend of gfx and psp for S0ix + /* skip suspend of gfx/mes and psp for S0ix * gfx is in gfxoff state, so on resume it will exit gfxoff just * like at runtime. PSP is also part of the always on hardware * so no need to suspend it. */ if (adev->in_s0ix && (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_PSP || - adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GFX)) + adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GFX || + adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_MES)) continue; /* XXX handle errors */ -- cgit