From e43229824d5bdd8dc0c2ea5b16f79b01ed2cd843 Mon Sep 17 00:00:00 2001 From: Mario Limonciello Date: Tue, 3 Jan 2023 13:20:52 -0600 Subject: drm/amd: Use `amdgpu_ucode_request` helper for SDMA The `amdgpu_ucode_request` helper will ensure that the return code for missing firmware is -ENODEV so that early_init can fail. Reviewed-by: Alex Deucher Reviewed-by: Lijo Lazar Signed-off-by: Mario Limonciello Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c index ea5278f094c0..a6a491569022 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c @@ -154,16 +154,11 @@ int amdgpu_sdma_process_ecc_irq(struct amdgpu_device *adev, static int amdgpu_sdma_init_inst_ctx(struct amdgpu_sdma_instance *sdma_inst) { - int err = 0; uint16_t version_major; const struct common_firmware_header *header = NULL; const struct sdma_firmware_header_v1_0 *hdr; const struct sdma_firmware_header_v2_0 *hdr_v2; - err = amdgpu_ucode_validate(sdma_inst->fw); - if (err) - return err; - header = (const struct common_firmware_header *) sdma_inst->fw->data; version_major = le16_to_cpu(header->header_version_major); @@ -195,7 +190,7 @@ void amdgpu_sdma_destroy_inst_ctx(struct amdgpu_device *adev, int i; for (i = 0; i < adev->sdma.num_instances; i++) { - release_firmware(adev->sdma.instance[i].fw); + amdgpu_ucode_release(&adev->sdma.instance[i].fw); if (duplicate) break; } @@ -214,7 +209,7 @@ int amdgpu_sdma_init_microcode(struct amdgpu_device *adev, const struct sdma_firmware_header_v2_0 *sdma_hdr; uint16_t version_major; - err = request_firmware(&adev->sdma.instance[instance].fw, fw_name, adev->dev); + err = amdgpu_ucode_request(adev, &adev->sdma.instance[instance].fw, fw_name); if (err) goto out; -- cgit From 1336b4e72c4c402ca31436e4fff6c085da26057a Mon Sep 17 00:00:00 2001 From: Mario Limonciello Date: Wed, 28 Dec 2022 01:24:19 -0600 Subject: drm/amd: Convert SDMA to use `amdgpu_ucode_ip_version_decode` Simplifies the code so that all SDMA versions will get the firmware name from `amdgpu_ucode_ip_version_decode`. v2: squash in fix from Srinivasan Reviewed-by: Alex Deucher Reviewed-by: Lijo Lazar Signed-off-by: Mario Limonciello Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c | 14 +++++--- drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h | 4 +-- drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c | 47 ++------------------------- drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c | 30 ++--------------- drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c | 55 +------------------------------- drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c | 25 +-------------- 6 files changed, 18 insertions(+), 157 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c index a6a491569022..a0d10c9c0405 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c @@ -200,15 +200,21 @@ void amdgpu_sdma_destroy_inst_ctx(struct amdgpu_device *adev, } int amdgpu_sdma_init_microcode(struct amdgpu_device *adev, - char *fw_name, u32 instance, - bool duplicate) + u32 instance, bool duplicate) { struct amdgpu_firmware_info *info = NULL; const struct common_firmware_header *header = NULL; - int err = 0, i; + int err, i; const struct sdma_firmware_header_v2_0 *sdma_hdr; uint16_t version_major; - + char ucode_prefix[30]; + char fw_name[40]; + + amdgpu_ucode_ip_version_decode(adev, SDMA0_HWIP, ucode_prefix, sizeof(ucode_prefix)); + if (instance == 0) + snprintf(fw_name, sizeof(fw_name), "amdgpu/%s.bin", ucode_prefix); + else + snprintf(fw_name, sizeof(fw_name), "amdgpu/%s%d.bin", ucode_prefix, instance); err = amdgpu_ucode_request(adev, &adev->sdma.instance[instance].fw, fw_name); if (err) goto out; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h index 7d99205c2e01..2d16e6d36728 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h @@ -124,8 +124,8 @@ int amdgpu_sdma_process_ras_data_cb(struct amdgpu_device *adev, int amdgpu_sdma_process_ecc_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry); -int amdgpu_sdma_init_microcode(struct amdgpu_device *adev, - char *fw_name, u32 instance, bool duplicate); +int amdgpu_sdma_init_microcode(struct amdgpu_device *adev, u32 instance, + bool duplicate); void amdgpu_sdma_destroy_inst_ctx(struct amdgpu_device *adev, bool duplicate); void amdgpu_sdma_unset_buffer_funcs_helper(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c index 4d780e4430e7..017ae298558e 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c @@ -575,60 +575,17 @@ static void sdma_v4_0_setup_ulv(struct amdgpu_device *adev) // vega10 real chip need to use PSP to load firmware static int sdma_v4_0_init_microcode(struct amdgpu_device *adev) { - const char *chip_name; - char fw_name[30]; int ret, i; - DRM_DEBUG("\n"); - - switch (adev->ip_versions[SDMA0_HWIP][0]) { - case IP_VERSION(4, 0, 0): - chip_name = "vega10"; - break; - case IP_VERSION(4, 0, 1): - chip_name = "vega12"; - break; - case IP_VERSION(4, 2, 0): - chip_name = "vega20"; - break; - case IP_VERSION(4, 1, 0): - case IP_VERSION(4, 1, 1): - if (adev->apu_flags & AMD_APU_IS_RAVEN2) - chip_name = "raven2"; - else if (adev->apu_flags & AMD_APU_IS_PICASSO) - chip_name = "picasso"; - else - chip_name = "raven"; - break; - case IP_VERSION(4, 2, 2): - chip_name = "arcturus"; - break; - case IP_VERSION(4, 1, 2): - if (adev->apu_flags & AMD_APU_IS_RENOIR) - chip_name = "renoir"; - else - chip_name = "green_sardine"; - break; - case IP_VERSION(4, 4, 0): - chip_name = "aldebaran"; - break; - default: - BUG(); - } - for (i = 0; i < adev->sdma.num_instances; i++) { - if (i == 0) - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma.bin", chip_name); - else - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma%d.bin", chip_name, i); if (adev->ip_versions[SDMA0_HWIP][0] == IP_VERSION(4, 2, 2) || adev->ip_versions[SDMA0_HWIP][0] == IP_VERSION(4, 4, 0)) { /* Acturus & Aldebaran will leverage the same FW memory for every SDMA instance */ - ret = amdgpu_sdma_init_microcode(adev, fw_name, 0, true); + ret = amdgpu_sdma_init_microcode(adev, 0, true); break; } else { - ret = amdgpu_sdma_init_microcode(adev, fw_name, i, false); + ret = amdgpu_sdma_init_microcode(adev, i, false); if (ret) return ret; } diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c index d4d9f196db83..1941b3b7c5d9 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c @@ -237,39 +237,13 @@ static void sdma_v5_0_init_golden_registers(struct amdgpu_device *adev) // emulation only, won't work on real chip // navi10 real chip need to use PSP to load firmware static int sdma_v5_0_init_microcode(struct amdgpu_device *adev) -{ - const char *chip_name; - char fw_name[40]; - int ret, i; +{ int ret, i; if (amdgpu_sriov_vf(adev) && (adev->ip_versions[SDMA0_HWIP][0] == IP_VERSION(5, 0, 5))) return 0; - DRM_DEBUG("\n"); - - switch (adev->ip_versions[SDMA0_HWIP][0]) { - case IP_VERSION(5, 0, 0): - chip_name = "navi10"; - break; - case IP_VERSION(5, 0, 2): - chip_name = "navi14"; - break; - case IP_VERSION(5, 0, 5): - chip_name = "navi12"; - break; - case IP_VERSION(5, 0, 1): - chip_name = "cyan_skillfish2"; - break; - default: - BUG(); - } - for (i = 0; i < adev->sdma.num_instances; i++) { - if (i == 0) - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma.bin", chip_name); - else - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma1.bin", chip_name); - ret = amdgpu_sdma_init_microcode(adev, fw_name, i, false); + ret = amdgpu_sdma_init_microcode(adev, i, false); if (ret) return ret; } diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c index 65e7a710298d..8e445eb9dd49 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c @@ -89,59 +89,6 @@ static u32 sdma_v5_2_get_reg_offset(struct amdgpu_device *adev, u32 instance, u3 return base + internal_offset; } -/** - * sdma_v5_2_init_microcode - load ucode images from disk - * - * @adev: amdgpu_device pointer - * - * Use the firmware interface to load the ucode images into - * the driver (not loaded into hw). - * Returns 0 on success, error on failure. - */ - -// emulation only, won't work on real chip -// navi10 real chip need to use PSP to load firmware -static int sdma_v5_2_init_microcode(struct amdgpu_device *adev) -{ - const char *chip_name; - char fw_name[40]; - - DRM_DEBUG("\n"); - - switch (adev->ip_versions[SDMA0_HWIP][0]) { - case IP_VERSION(5, 2, 0): - chip_name = "sienna_cichlid_sdma"; - break; - case IP_VERSION(5, 2, 2): - chip_name = "navy_flounder_sdma"; - break; - case IP_VERSION(5, 2, 1): - chip_name = "vangogh_sdma"; - break; - case IP_VERSION(5, 2, 4): - chip_name = "dimgrey_cavefish_sdma"; - break; - case IP_VERSION(5, 2, 5): - chip_name = "beige_goby_sdma"; - break; - case IP_VERSION(5, 2, 3): - chip_name = "yellow_carp_sdma"; - break; - case IP_VERSION(5, 2, 6): - chip_name = "sdma_5_2_6"; - break; - case IP_VERSION(5, 2, 7): - chip_name = "sdma_5_2_7"; - break; - default: - BUG(); - } - - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s.bin", chip_name); - - return amdgpu_sdma_init_microcode(adev, fw_name, 0, true); -} - static unsigned sdma_v5_2_ring_init_cond_exec(struct amdgpu_ring *ring) { unsigned ret; @@ -1288,7 +1235,7 @@ static int sdma_v5_2_sw_init(void *handle) return r; } - r = sdma_v5_2_init_microcode(adev); + r = amdgpu_sdma_init_microcode(adev, 0, true); if (r) { DRM_ERROR("Failed to load sdma firmware!\n"); return r; diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c index 049c26a45d85..bf1fa5e8d2f9 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c @@ -78,29 +78,6 @@ static u32 sdma_v6_0_get_reg_offset(struct amdgpu_device *adev, u32 instance, u3 return base + internal_offset; } -/** - * sdma_v6_0_init_microcode - load ucode images from disk - * - * @adev: amdgpu_device pointer - * - * Use the firmware interface to load the ucode images into - * the driver (not loaded into hw). - * Returns 0 on success, error on failure. - */ -static int sdma_v6_0_init_microcode(struct amdgpu_device *adev) -{ - char fw_name[30]; - char ucode_prefix[30]; - - DRM_DEBUG("\n"); - - amdgpu_ucode_ip_version_decode(adev, SDMA0_HWIP, ucode_prefix, sizeof(ucode_prefix)); - - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s.bin", ucode_prefix); - - return amdgpu_sdma_init_microcode(adev, fw_name, 0, true); -} - static unsigned sdma_v6_0_ring_init_cond_exec(struct amdgpu_ring *ring) { unsigned ret; @@ -1260,7 +1237,7 @@ static int sdma_v6_0_sw_init(void *handle) if (r) return r; - r = sdma_v6_0_init_microcode(adev); + r = amdgpu_sdma_init_microcode(adev, 0, true); if (r) { DRM_ERROR("Failed to load sdma firmware!\n"); return r; -- cgit From 6675402a47cc9464d57ace33fb10c59f126334b8 Mon Sep 17 00:00:00 2001 From: Mario Limonciello Date: Tue, 3 Jan 2023 13:07:19 -0600 Subject: drm/amd: Make SDMA firmware load failures less noisy. When firmware is missing we get failures at every step. ``` [ 3.855086] amdgpu 0000:04:00.0: Direct firmware load for amdgpu/green_sardine_sdma.bin failed with error -2 [ 3.855087] [drm:amdgpu_sdma_init_microcode [amdgpu]] *ERROR* SDMA: Failed to init firmware "amdgpu/green_sardine_sdma.bin" [ 3.855398] [drm:sdma_v4_0_early_init [amdgpu]] *ERROR* Failed to load sdma firmware! ``` Realistically we don't need all of these, a user can tell from the first one that request_firmware emitted what happened. Drop the others. Reviewed-by: Alex Deucher Reviewed-by: Lijo Lazar Signed-off-by: Mario Limonciello Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c index a0d10c9c0405..e9b78739b9ff 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c @@ -280,10 +280,8 @@ int amdgpu_sdma_init_microcode(struct amdgpu_device *adev, } out: - if (err) { - DRM_ERROR("SDMA: Failed to init firmware \"%s\"\n", fw_name); + if (err) amdgpu_sdma_destroy_inst_ctx(adev, duplicate); - } return err; } -- cgit From a57b24e170b1ffe97c4571b366c0cf1fe09e9a60 Mon Sep 17 00:00:00 2001 From: YiPeng Chai Date: Tue, 17 Jan 2023 15:44:25 +0800 Subject: drm/amdgpu: Add sdma ras function on sdma v6_0_3 Add sdma ras function on sdma v6_0_3. Signed-off-by: YiPeng Chai Reviewed-by: Tao Zhou Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c | 35 ++++++++++++++++++++++++++++++++ drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h | 1 + drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c | 24 ++++++++++++++++++++++ 3 files changed, 60 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c index e9b78739b9ff..231ca06bc9c7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c @@ -305,3 +305,38 @@ void amdgpu_sdma_unset_buffer_funcs_helper(struct amdgpu_device *adev) } } } + +int amdgpu_sdma_ras_sw_init(struct amdgpu_device *adev) +{ + int err = 0; + struct amdgpu_sdma_ras *ras = NULL; + + /* adev->sdma.ras is NULL, which means sdma does not + * support ras function, then do nothing here. + */ + if (!adev->sdma.ras) + return 0; + + ras = adev->sdma.ras; + + err = amdgpu_ras_register_ras_block(adev, &ras->ras_block); + if (err) { + dev_err(adev->dev, "Failed to register sdma ras block!\n"); + return err; + } + + strcpy(ras->ras_block.ras_comm.name, "sdma"); + ras->ras_block.ras_comm.block = AMDGPU_RAS_BLOCK__SDMA; + ras->ras_block.ras_comm.type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE; + adev->sdma.ras_if = &ras->ras_block.ras_comm; + + /* If not define special ras_late_init function, use default ras_late_init */ + if (!ras->ras_block.ras_late_init) + ras->ras_block.ras_late_init = amdgpu_sdma_ras_late_init; + + /* If not defined special ras_cb function, use default ras_cb */ + if (!ras->ras_block.ras_cb) + ras->ras_block.ras_cb = amdgpu_sdma_process_ras_data_cb; + + return 0; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h index 2d16e6d36728..fc8528812598 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h @@ -129,5 +129,6 @@ int amdgpu_sdma_init_microcode(struct amdgpu_device *adev, u32 instance, void amdgpu_sdma_destroy_inst_ctx(struct amdgpu_device *adev, bool duplicate); void amdgpu_sdma_unset_buffer_funcs_helper(struct amdgpu_device *adev); +int amdgpu_sdma_ras_sw_init(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c index 6fe292a2486b..3d36329be384 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c @@ -1211,6 +1211,24 @@ static void sdma_v6_0_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring, amdgpu_ring_emit_reg_wait(ring, reg1, mask, mask); } +static struct amdgpu_sdma_ras sdma_v6_0_3_ras = { + .ras_block = { + .ras_late_init = amdgpu_ras_block_late_init, + }, +}; + +static void sdma_v6_0_set_ras_funcs(struct amdgpu_device *adev) +{ + switch (adev->ip_versions[SDMA0_HWIP][0]) { + case IP_VERSION(6, 0, 3): + adev->sdma.ras = &sdma_v6_0_3_ras; + break; + default: + break; + } + +} + static int sdma_v6_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; @@ -1220,6 +1238,7 @@ static int sdma_v6_0_early_init(void *handle) sdma_v6_0_set_vm_pte_funcs(adev); sdma_v6_0_set_irq_funcs(adev); sdma_v6_0_set_mqd_funcs(adev); + sdma_v6_0_set_ras_funcs(adev); return 0; } @@ -1264,6 +1283,11 @@ static int sdma_v6_0_sw_init(void *handle) return r; } + if (amdgpu_sdma_ras_sw_init(adev)) { + dev_err(adev->dev, "Failed to initialize sdma ras block!\n"); + return -EINVAL; + } + return r; } -- cgit