From a9e4fb51425f680f191bb65c2778cea31bf058c7 Mon Sep 17 00:00:00 2001 From: Alyssa Rosenzweig Date: Sun, 9 Jan 2022 11:37:04 -0500 Subject: drm/panfrost: Update create_bo flags comment Update a comment stating create_bo took no flags, since it now takes a bit mask of optional flags NOEXEC and HEAP. Signed-off-by: Alyssa Rosenzweig Reviewed-by: Steven Price Signed-off-by: Steven Price Link: https://patchwork.freedesktop.org/patch/msgid/20220109163704.2564-1-alyssa.rosenzweig@collabora.com --- include/uapi/drm/panfrost_drm.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/uapi/drm') diff --git a/include/uapi/drm/panfrost_drm.h b/include/uapi/drm/panfrost_drm.h index 061e700dd06c..9e40277d8185 100644 --- a/include/uapi/drm/panfrost_drm.h +++ b/include/uapi/drm/panfrost_drm.h @@ -84,14 +84,14 @@ struct drm_panfrost_wait_bo { __s64 timeout_ns; /* absolute */ }; +/* Valid flags to pass to drm_panfrost_create_bo */ #define PANFROST_BO_NOEXEC 1 #define PANFROST_BO_HEAP 2 /** * struct drm_panfrost_create_bo - ioctl argument for creating Panfrost BOs. * - * There are currently no values for the flags argument, but it may be - * used in a future extension. + * The flags argument is a bit mask of PANFROST_BO_* flags. */ struct drm_panfrost_create_bo { __u32 size; -- cgit v1.2.3 From c4381d0ee81930097e94e55d1c23f85798ffd093 Mon Sep 17 00:00:00 2001 From: Bokun Zhang Date: Wed, 12 Jan 2022 10:34:11 -0500 Subject: drm/amdgpu: Add interface to load SRIOV cap FW - Add interface to load SRIOV cap FW. If the FW does not exist, simply skip this FW loading routine. This FW will only be loaded under SRIOV. Other driver configuration will not be affected. By adding this interface, it will make us easier to prepare SRIOV Linux guest driver for different users. - Update sysfs interface to read cap FW version. - Refactor PSP FW loading routine under SRIOV to use a unified SWITCH statement instead of using IF statement - Remove redundant amdgpu_sriov_vf() check in FW loading routine Acked-by: Monk Liu Acked-by: Guchun Chen Signed-off-by: Bokun Zhang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 14 ++++ drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 108 ++++++++++++++++++++++++++---- drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h | 9 +++ drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h | 1 + drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h | 1 + drivers/gpu/drm/amd/amdgpu/psp_v11_0.c | 4 +- drivers/gpu/drm/amd/amdgpu/psp_v3_1.c | 1 + include/uapi/drm/amdgpu_drm.h | 2 + 8 files changed, 125 insertions(+), 15 deletions(-) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 289521aafb79..85f06396d184 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -400,6 +400,10 @@ static int amdgpu_firmware_info(struct drm_amdgpu_info_firmware *fw_info, fw_info->ver = adev->psp.toc.fw_version; fw_info->feature = adev->psp.toc.feature_version; break; + case AMDGPU_INFO_FW_CAP: + fw_info->ver = adev->psp.cap_fw_version; + fw_info->feature = adev->psp.cap_feature_version; + break; default: return -EINVAL; } @@ -1617,6 +1621,16 @@ static int amdgpu_debugfs_firmware_info_show(struct seq_file *m, void *unused) seq_printf(m, "TOC feature version: %u, firmware version: 0x%08x\n", fw_info.feature, fw_info.ver); + /* CAP */ + if (adev->psp.cap_fw) { + query_fw.fw_type = AMDGPU_INFO_FW_CAP; + ret = amdgpu_firmware_info(&fw_info, &query_fw, adev); + if (ret) + return ret; + seq_printf(m, "CAP feature version: %u, firmware version: 0x%08x\n", + fw_info.feature, fw_info.ver); + } + seq_printf(m, "VBIOS version: %s\n", ctx->vbios_version); return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index dee17a0e1187..c984b5a34679 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -259,6 +259,32 @@ static bool psp_get_runtime_db_entry(struct amdgpu_device *adev, return ret; } +static int psp_init_sriov_microcode(struct psp_context *psp) +{ + struct amdgpu_device *adev = psp->adev; + int ret = 0; + + switch (adev->ip_versions[MP0_HWIP][0]) { + case IP_VERSION(9, 0, 0): + ret = psp_init_cap_microcode(psp, "vega10"); + break; + case IP_VERSION(11, 0, 9): + ret = psp_init_cap_microcode(psp, "navi12"); + break; + case IP_VERSION(11, 0, 7): + ret = psp_init_cap_microcode(psp, "sienna_cichlid"); + break; + case IP_VERSION(13, 0, 2): + ret = psp_init_ta_microcode(psp, "aldebaran"); + break; + default: + BUG(); + break; + } + + return ret; +} + static int psp_sw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; @@ -273,19 +299,13 @@ static int psp_sw_init(void *handle) ret = -ENOMEM; } - if (!amdgpu_sriov_vf(adev)) { + if (amdgpu_sriov_vf(adev)) + ret = psp_init_sriov_microcode(psp); + else ret = psp_init_microcode(psp); - if (ret) { - DRM_ERROR("Failed to load psp firmware!\n"); - return ret; - } - } else if (amdgpu_sriov_vf(adev) && - adev->ip_versions[MP0_HWIP][0] == IP_VERSION(13, 0, 2)) { - ret = psp_init_ta_microcode(psp, "aldebaran"); - if (ret) { - DRM_ERROR("Failed to initialize ta microcode!\n"); - return ret; - } + if (ret) { + DRM_ERROR("Failed to load psp firmware!\n"); + return ret; } memset(&boot_cfg_entry, 0, sizeof(boot_cfg_entry)); @@ -353,6 +373,10 @@ static int psp_sw_fini(void *handle) release_firmware(psp->ta_fw); psp->ta_fw = NULL; } + if (adev->psp.cap_fw) { + release_firmware(psp->cap_fw); + psp->cap_fw = NULL; + } if (adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 0) || adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 7)) @@ -491,7 +515,10 @@ psp_cmd_submit_buf(struct psp_context *psp, DRM_WARN("psp gfx command %s(0x%X) failed and response status is (0x%X)\n", psp_gfx_cmd_name(psp->cmd_buf_mem->cmd_id), psp->cmd_buf_mem->cmd_id, psp->cmd_buf_mem->resp.status); - if (!timeout) { + /* If we load CAP FW, PSP must return 0 under SRIOV + * also return failure in case of timeout + */ + if ((ucode && (ucode->ucode_id == AMDGPU_UCODE_ID_CAP)) || !timeout) { ret = -EINVAL; goto exit; } @@ -2051,6 +2078,9 @@ static int psp_get_fw_type(struct amdgpu_firmware_info *ucode, enum psp_gfx_fw_type *type) { switch (ucode->ucode_id) { + case AMDGPU_UCODE_ID_CAP: + *type = GFX_FW_TYPE_CAP; + break; case AMDGPU_UCODE_ID_SDMA0: *type = GFX_FW_TYPE_SDMA0; break; @@ -3217,6 +3247,58 @@ out: return err; } +int psp_init_cap_microcode(struct psp_context *psp, + const char *chip_name) +{ + struct amdgpu_device *adev = psp->adev; + char fw_name[PSP_FW_NAME_LEN]; + const struct psp_firmware_header_v1_0 *cap_hdr_v1_0; + struct amdgpu_firmware_info *info = NULL; + int err = 0; + + if (!chip_name) { + dev_err(adev->dev, "invalid chip name for cap microcode\n"); + return -EINVAL; + } + + if (!amdgpu_sriov_vf(adev)) { + dev_err(adev->dev, "cap microcode should only be loaded under SRIOV\n"); + return -EINVAL; + } + + snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_cap.bin", chip_name); + err = request_firmware(&adev->psp.cap_fw, fw_name, adev->dev); + if (err) { + dev_warn(adev->dev, "cap microcode does not exist, skip\n"); + err = 0; + goto out; + } + + err = amdgpu_ucode_validate(adev->psp.cap_fw); + if (err) { + dev_err(adev->dev, "fail to initialize cap microcode\n"); + goto out; + } + + info = &adev->firmware.ucode[AMDGPU_UCODE_ID_CAP]; + info->ucode_id = AMDGPU_UCODE_ID_CAP; + info->fw = adev->psp.cap_fw; + cap_hdr_v1_0 = (const struct psp_firmware_header_v1_0 *) + adev->psp.cap_fw->data; + adev->firmware.fw_size += ALIGN( + le32_to_cpu(cap_hdr_v1_0->header.ucode_size_bytes), PAGE_SIZE); + adev->psp.cap_fw_version = le32_to_cpu(cap_hdr_v1_0->header.ucode_version); + adev->psp.cap_feature_version = le32_to_cpu(cap_hdr_v1_0->sos.fw_version); + adev->psp.cap_ucode_size = le32_to_cpu(cap_hdr_v1_0->header.ucode_size_bytes); + + return 0; + +out: + release_firmware(adev->psp.cap_fw); + adev->psp.cap_fw = NULL; + return err; +} + static int psp_set_clockgating_state(void *handle, enum amd_clockgating_state state) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h index f29afabbff1f..ff7d533eb746 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h @@ -306,6 +306,9 @@ struct psp_context /* toc firmware */ const struct firmware *toc_fw; + /* cap firmware */ + const struct firmware *cap_fw; + /* fence buffer */ struct amdgpu_bo *fence_buf_bo; uint64_t fence_buf_mc_addr; @@ -327,6 +330,10 @@ struct psp_context const struct firmware *ta_fw; uint32_t ta_fw_version; + uint32_t cap_fw_version; + uint32_t cap_feature_version; + uint32_t cap_ucode_size; + struct ta_context asd_context; struct psp_xgmi_context xgmi_context; struct psp_ras_context ras_context; @@ -440,6 +447,8 @@ int psp_init_sos_microcode(struct psp_context *psp, const char *chip_name); int psp_init_ta_microcode(struct psp_context *psp, const char *chip_name); +int psp_init_cap_microcode(struct psp_context *psp, + const char *chip_name); int psp_get_fw_attestation_records_addr(struct psp_context *psp, uint64_t *output_ptr); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h index 7c2538db3cd5..428f4df184d0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h @@ -378,6 +378,7 @@ enum AMDGPU_UCODE_ID { AMDGPU_UCODE_ID_VCN0_RAM, AMDGPU_UCODE_ID_VCN1_RAM, AMDGPU_UCODE_ID_DMCUB, + AMDGPU_UCODE_ID_CAP, AMDGPU_UCODE_ID_MAXIMUM, }; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h index dd0dce254901..1f276ddd26e9 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h +++ b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h @@ -258,6 +258,7 @@ enum psp_gfx_fw_type { GFX_FW_TYPE_SDMA6 = 56, /* SDMA6 MI */ GFX_FW_TYPE_SDMA7 = 57, /* SDMA7 MI */ GFX_FW_TYPE_VCN1 = 58, /* VCN1 MI */ + GFX_FW_TYPE_CAP = 62, /* CAP_FW */ GFX_FW_TYPE_REG_LIST = 67, /* REG_LIST MI */ GFX_FW_TYPE_MAX }; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c index d0e76b36d4ab..9518b4394a6e 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c @@ -53,11 +53,13 @@ MODULE_FIRMWARE("amdgpu/navi14_ta.bin"); MODULE_FIRMWARE("amdgpu/navi12_sos.bin"); MODULE_FIRMWARE("amdgpu/navi12_asd.bin"); MODULE_FIRMWARE("amdgpu/navi12_ta.bin"); +MODULE_FIRMWARE("amdgpu/navi12_cap.bin"); MODULE_FIRMWARE("amdgpu/arcturus_sos.bin"); MODULE_FIRMWARE("amdgpu/arcturus_asd.bin"); MODULE_FIRMWARE("amdgpu/arcturus_ta.bin"); MODULE_FIRMWARE("amdgpu/sienna_cichlid_sos.bin"); MODULE_FIRMWARE("amdgpu/sienna_cichlid_ta.bin"); +MODULE_FIRMWARE("amdgpu/sienna_cichlid_cap.bin"); MODULE_FIRMWARE("amdgpu/navy_flounder_sos.bin"); MODULE_FIRMWARE("amdgpu/navy_flounder_ta.bin"); MODULE_FIRMWARE("amdgpu/vangogh_asd.bin"); @@ -177,8 +179,6 @@ static int psp_v11_0_init_microcode(struct psp_context *psp) err = psp_init_asd_microcode(psp, chip_name); if (err) return err; - if (amdgpu_sriov_vf(adev)) - break; snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_ta.bin", chip_name); err = request_firmware(&adev->psp.ta_fw, fw_name, adev->dev); if (err) { diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c index 1ed357cb0f49..01f3bcc62a6c 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c @@ -44,6 +44,7 @@ MODULE_FIRMWARE("amdgpu/vega10_sos.bin"); MODULE_FIRMWARE("amdgpu/vega10_asd.bin"); +MODULE_FIRMWARE("amdgpu/vega10_cap.bin"); MODULE_FIRMWARE("amdgpu/vega12_sos.bin"); MODULE_FIRMWARE("amdgpu/vega12_asd.bin"); diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index 0b94ec7b73e7..be4f9111f478 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -728,6 +728,8 @@ struct drm_amdgpu_cs_chunk_data { #define AMDGPU_INFO_FW_DMCUB 0x14 /* Subquery id: Query TOC firmware version */ #define AMDGPU_INFO_FW_TOC 0x15 + /* Subquery id: Query CAP firmware version */ + #define AMDGPU_INFO_FW_CAP 0x16 /* number of bytes moved for TTM migration */ #define AMDGPU_INFO_NUM_BYTES_MOVED 0x0f -- cgit v1.2.3 From 8cda7a4f96e435be2fd074009d69521d973d7d31 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 7 Jan 2022 17:57:33 -0500 Subject: drm/amdgpu/UAPI: add new CTX OP to get/set stable pstates Add a new CTX ioctl operation to set stable pstates for profiling. When creating traces for tools like RGP or using SPM or doing performance profiling, it's required to enable a special stable profiling power state on the GPU. These profiling states set fixed clocks and disable certain other power features like powergating which may impact the results. Historically, these profiling pstates were enabled via sysfs, but this adds an interface to enable it via the CTX ioctl from the application. Since the power state is global only one application can set it at a time, so if multiple applications try and use it only the first will get it, the ioctl will return -EBUSY for others. The sysfs interface will override whatever has been set by this interface. Mesa MR: https://gitlab.freedesktop.org/mesa/drm/-/merge_requests/207 v2: don't default r = 0; v3: rebase on Evan's PM cleanup Reviewed-by: Evan Quan Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 129 ++++++++++++++++++++++++++++- drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h | 1 + drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 1 + drivers/gpu/drm/amd/pm/amdgpu_pm.c | 5 ++ drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h | 3 + include/uapi/drm/amdgpu_drm.h | 17 +++- 6 files changed, 153 insertions(+), 3 deletions(-) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index 468003583b2a..2c929fa40379 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -237,6 +237,7 @@ static int amdgpu_ctx_init(struct amdgpu_device *adev, ctx->vram_lost_counter = atomic_read(&adev->vram_lost_counter); ctx->init_priority = priority; ctx->override_priority = AMDGPU_CTX_PRIORITY_UNSET; + ctx->stable_pstate = AMDGPU_CTX_STABLE_PSTATE_NONE; return 0; } @@ -255,6 +256,86 @@ static void amdgpu_ctx_fini_entity(struct amdgpu_ctx_entity *entity) kfree(entity); } +static int amdgpu_ctx_get_stable_pstate(struct amdgpu_ctx *ctx, + u32 *stable_pstate) +{ + struct amdgpu_device *adev = ctx->adev; + enum amd_dpm_forced_level current_level; + + if (!ctx) + return -EINVAL; + + current_level = amdgpu_dpm_get_performance_level(adev); + + switch (current_level) { + case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD: + *stable_pstate = AMDGPU_CTX_STABLE_PSTATE_STANDARD; + break; + case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK: + *stable_pstate = AMDGPU_CTX_STABLE_PSTATE_MIN_SCLK; + break; + case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK: + *stable_pstate = AMDGPU_CTX_STABLE_PSTATE_MIN_MCLK; + break; + case AMD_DPM_FORCED_LEVEL_PROFILE_PEAK: + *stable_pstate = AMDGPU_CTX_STABLE_PSTATE_PEAK; + break; + default: + *stable_pstate = AMDGPU_CTX_STABLE_PSTATE_NONE; + break; + } + return 0; +} + +static int amdgpu_ctx_set_stable_pstate(struct amdgpu_ctx *ctx, + u32 stable_pstate) +{ + struct amdgpu_device *adev = ctx->adev; + enum amd_dpm_forced_level level; + int r; + + if (!ctx) + return -EINVAL; + + mutex_lock(&adev->pm.stable_pstate_ctx_lock); + if (adev->pm.stable_pstate_ctx && adev->pm.stable_pstate_ctx != ctx) { + r = -EBUSY; + goto done; + } + + switch (stable_pstate) { + case AMDGPU_CTX_STABLE_PSTATE_NONE: + level = AMD_DPM_FORCED_LEVEL_AUTO; + break; + case AMDGPU_CTX_STABLE_PSTATE_STANDARD: + level = AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD; + break; + case AMDGPU_CTX_STABLE_PSTATE_MIN_SCLK: + level = AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK; + break; + case AMDGPU_CTX_STABLE_PSTATE_MIN_MCLK: + level = AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK; + break; + case AMDGPU_CTX_STABLE_PSTATE_PEAK: + level = AMD_DPM_FORCED_LEVEL_PROFILE_PEAK; + break; + default: + r = -EINVAL; + goto done; + } + + r = amdgpu_dpm_force_performance_level(adev, level); + + if (level == AMD_DPM_FORCED_LEVEL_AUTO) + adev->pm.stable_pstate_ctx = NULL; + else + adev->pm.stable_pstate_ctx = ctx; +done: + mutex_unlock(&adev->pm.stable_pstate_ctx_lock); + + return r; +} + static void amdgpu_ctx_fini(struct kref *ref) { struct amdgpu_ctx *ctx = container_of(ref, struct amdgpu_ctx, refcount); @@ -270,7 +351,7 @@ static void amdgpu_ctx_fini(struct kref *ref) ctx->entities[i][j] = NULL; } } - + amdgpu_ctx_set_stable_pstate(ctx, AMDGPU_CTX_STABLE_PSTATE_NONE); mutex_destroy(&ctx->lock); kfree(ctx); } @@ -467,11 +548,41 @@ static int amdgpu_ctx_query2(struct amdgpu_device *adev, return 0; } + + +static int amdgpu_ctx_stable_pstate(struct amdgpu_device *adev, + struct amdgpu_fpriv *fpriv, uint32_t id, + bool set, u32 *stable_pstate) +{ + struct amdgpu_ctx *ctx; + struct amdgpu_ctx_mgr *mgr; + int r; + + if (!fpriv) + return -EINVAL; + + mgr = &fpriv->ctx_mgr; + mutex_lock(&mgr->lock); + ctx = idr_find(&mgr->ctx_handles, id); + if (!ctx) { + mutex_unlock(&mgr->lock); + return -EINVAL; + } + + if (set) + r = amdgpu_ctx_set_stable_pstate(ctx, *stable_pstate); + else + r = amdgpu_ctx_get_stable_pstate(ctx, stable_pstate); + + mutex_unlock(&mgr->lock); + return r; +} + int amdgpu_ctx_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) { int r; - uint32_t id; + uint32_t id, stable_pstate; int32_t priority; union drm_amdgpu_ctx *args = data; @@ -500,6 +611,20 @@ int amdgpu_ctx_ioctl(struct drm_device *dev, void *data, case AMDGPU_CTX_OP_QUERY_STATE2: r = amdgpu_ctx_query2(adev, fpriv, id, &args->out); break; + case AMDGPU_CTX_OP_GET_STABLE_PSTATE: + if (args->in.flags) + return -EINVAL; + r = amdgpu_ctx_stable_pstate(adev, fpriv, id, false, &stable_pstate); + args->out.pstate.flags = stable_pstate; + break; + case AMDGPU_CTX_OP_SET_STABLE_PSTATE: + if (args->in.flags & ~AMDGPU_CTX_STABLE_PSTATE_FLAGS_MASK) + return -EINVAL; + stable_pstate = args->in.flags & AMDGPU_CTX_STABLE_PSTATE_FLAGS_MASK; + if (stable_pstate > AMDGPU_CTX_STABLE_PSTATE_PEAK) + return -EINVAL; + r = amdgpu_ctx_stable_pstate(adev, fpriv, id, true, &stable_pstate); + break; default: return -EINVAL; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h index a44b8b8ed39c..142f2f87d44c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h @@ -53,6 +53,7 @@ struct amdgpu_ctx { atomic_t guilty; unsigned long ras_counter_ce; unsigned long ras_counter_ue; + uint32_t stable_pstate; }; struct amdgpu_ctx_mgr { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 6ba57ad88640..660fb4085c97 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -3510,6 +3510,7 @@ int amdgpu_device_init(struct amdgpu_device *adev, init_rwsem(&adev->reset_sem); mutex_init(&adev->psp.mutex); mutex_init(&adev->notifier_lock); + mutex_init(&adev->pm.stable_pstate_ctx_lock); amdgpu_device_init_apu_flags(adev); diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c index 837a31a46596..d68e7132da2c 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c @@ -336,11 +336,16 @@ static ssize_t amdgpu_set_power_dpm_force_performance_level(struct device *dev, return ret; } + mutex_lock(&adev->pm.stable_pstate_ctx_lock); if (amdgpu_dpm_force_performance_level(adev, level)) { pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); + mutex_unlock(&adev->pm.stable_pstate_ctx_lock); return -EINVAL; } + /* override whatever a user ctx may have set */ + adev->pm.stable_pstate_ctx = NULL; + mutex_unlock(&adev->pm.stable_pstate_ctx_lock); pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); diff --git a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h index 2766b88ecf96..5cc05110cdae 100644 --- a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h +++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h @@ -338,6 +338,9 @@ struct amdgpu_pm { uint32_t smu_debug_mask; bool pp_force_state_enabled; + + struct mutex stable_pstate_ctx_lock; + struct amdgpu_ctx *stable_pstate_ctx; }; u32 amdgpu_dpm_get_vblank_time(struct amdgpu_device *adev); diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index be4f9111f478..76b580d10a52 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -206,6 +206,8 @@ union drm_amdgpu_bo_list { #define AMDGPU_CTX_OP_FREE_CTX 2 #define AMDGPU_CTX_OP_QUERY_STATE 3 #define AMDGPU_CTX_OP_QUERY_STATE2 4 +#define AMDGPU_CTX_OP_GET_STABLE_PSTATE 5 +#define AMDGPU_CTX_OP_SET_STABLE_PSTATE 6 /* GPU reset status */ #define AMDGPU_CTX_NO_RESET 0 @@ -238,10 +240,18 @@ union drm_amdgpu_bo_list { #define AMDGPU_CTX_PRIORITY_HIGH 512 #define AMDGPU_CTX_PRIORITY_VERY_HIGH 1023 +/* select a stable profiling pstate for perfmon tools */ +#define AMDGPU_CTX_STABLE_PSTATE_FLAGS_MASK 0xf +#define AMDGPU_CTX_STABLE_PSTATE_NONE 0 +#define AMDGPU_CTX_STABLE_PSTATE_STANDARD 1 +#define AMDGPU_CTX_STABLE_PSTATE_MIN_SCLK 2 +#define AMDGPU_CTX_STABLE_PSTATE_MIN_MCLK 3 +#define AMDGPU_CTX_STABLE_PSTATE_PEAK 4 + struct drm_amdgpu_ctx_in { /** AMDGPU_CTX_OP_* */ __u32 op; - /** For future use, no flags defined so far */ + /** Flags */ __u32 flags; __u32 ctx_id; /** AMDGPU_CTX_PRIORITY_* */ @@ -262,6 +272,11 @@ union drm_amdgpu_ctx_out { /** Reset status since the last call of the ioctl. */ __u32 reset_status; } state; + + struct { + __u32 flags; + __u32 _pad; + } pstate; }; union drm_amdgpu_ctx { -- cgit v1.2.3 From a3574119826d9a4ef807fb973cf5150c3b90da43 Mon Sep 17 00:00:00 2001 From: Simon Ser Date: Thu, 3 Feb 2022 13:38:06 +0000 Subject: drm: document struct drm_mode_fb_cmd2 Follow-up for the DRM_IOCTL_MODE_GETFB2 docs. v2: (Daniel Stone) - Replace fourcc.org with drm_fourcc.h because this is the authoritative source and the website may have mismatches. - Drop assumption that offsets will generally be 0. - Mention that unused entries must be zero'ed out. v3: (Pekka) - Mention that a handle can be re-used - Add unit for pitches/offsets Signed-off-by: Simon Ser Reviewed-by: Daniel Vetter Acked-by: Pekka Paalanen Cc: Daniel Stone Reviewed-by: Daniel Stone Link: https://patchwork.freedesktop.org/patch/msgid/20220203133753.261507-1-contact@emersion.fr --- include/uapi/drm/drm_mode.h | 88 ++++++++++++++++++++++++++++++--------------- 1 file changed, 60 insertions(+), 28 deletions(-) (limited to 'include/uapi/drm') diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h index e1e351682872..0a0d56a6158e 100644 --- a/include/uapi/drm/drm_mode.h +++ b/include/uapi/drm/drm_mode.h @@ -663,41 +663,73 @@ struct drm_mode_fb_cmd { #define DRM_MODE_FB_INTERLACED (1<<0) /* for interlaced framebuffers */ #define DRM_MODE_FB_MODIFIERS (1<<1) /* enables ->modifer[] */ +/** + * struct drm_mode_fb_cmd2 - Frame-buffer metadata. + * + * This struct holds frame-buffer metadata. There are two ways to use it: + * + * - User-space can fill this struct and perform a &DRM_IOCTL_MODE_ADDFB2 + * ioctl to register a new frame-buffer. The new frame-buffer object ID will + * be set by the kernel in @fb_id. + * - User-space can set @fb_id and perform a &DRM_IOCTL_MODE_GETFB2 ioctl to + * fetch metadata about an existing frame-buffer. + * + * In case of planar formats, this struct allows up to 4 buffer objects with + * offsets and pitches per plane. The pitch and offset order is dictated by the + * format FourCC as defined by ``drm_fourcc.h``, e.g. NV12 is described as: + * + * YUV 4:2:0 image with a plane of 8 bit Y samples followed by an + * interleaved U/V plane containing 8 bit 2x2 subsampled colour difference + * samples. + * + * So it would consist of a Y plane at ``offsets[0]`` and a UV plane at + * ``offsets[1]``. + * + * To accommodate tiled, compressed, etc formats, a modifier can be specified. + * For more information see the "Format Modifiers" section. Note that even + * though it looks like we have a modifier per-plane, we in fact do not. The + * modifier for each plane must be identical. Thus all combinations of + * different data layouts for multi-plane formats must be enumerated as + * separate modifiers. + * + * All of the entries in @handles, @pitches, @offsets and @modifier must be + * zero when unused. Warning, for @offsets and @modifier zero can't be used to + * figure out whether the entry is used or not since it's a valid value (a zero + * offset is common, and a zero modifier is &DRM_FORMAT_MOD_LINEAR). + */ struct drm_mode_fb_cmd2 { + /** @fb_id: Object ID of the frame-buffer. */ __u32 fb_id; + /** @width: Width of the frame-buffer. */ __u32 width; + /** @height: Height of the frame-buffer. */ __u32 height; - __u32 pixel_format; /* fourcc code from drm_fourcc.h */ - __u32 flags; /* see above flags */ + /** + * @pixel_format: FourCC format code, see ``DRM_FORMAT_*`` constants in + * ``drm_fourcc.h``. + */ + __u32 pixel_format; + /** + * @flags: Frame-buffer flags (see &DRM_MODE_FB_INTERLACED and + * &DRM_MODE_FB_MODIFIERS). + */ + __u32 flags; - /* - * In case of planar formats, this ioctl allows up to 4 - * buffer objects with offsets and pitches per plane. - * The pitch and offset order is dictated by the fourcc, - * e.g. NV12 (https://fourcc.org/yuv.php#NV12) is described as: - * - * YUV 4:2:0 image with a plane of 8 bit Y samples - * followed by an interleaved U/V plane containing - * 8 bit 2x2 subsampled colour difference samples. - * - * So it would consist of Y as offsets[0] and UV as - * offsets[1]. Note that offsets[0] will generally - * be 0 (but this is not required). - * - * To accommodate tiled, compressed, etc formats, a - * modifier can be specified. The default value of zero - * indicates "native" format as specified by the fourcc. - * Vendor specific modifier token. Note that even though - * it looks like we have a modifier per-plane, we in fact - * do not. The modifier for each plane must be identical. - * Thus all combinations of different data layouts for - * multi plane formats must be enumerated as separate - * modifiers. + /** + * @handles: GEM buffer handle, one per plane. Set to 0 if the plane is + * unused. The same handle can be used for multiple planes. */ __u32 handles[4]; - __u32 pitches[4]; /* pitch for each plane */ - __u32 offsets[4]; /* offset of each plane */ - __u64 modifier[4]; /* ie, tiling, compress */ + /** @pitches: Pitch (aka. stride) in bytes, one per plane. */ + __u32 pitches[4]; + /** @offsets: Offset into the buffer in bytes, one per plane. */ + __u32 offsets[4]; + /** + * @modifier: Format modifier, one per plane. See ``DRM_FORMAT_MOD_*`` + * constants in ``drm_fourcc.h``. All planes must use the same + * modifier. Ignored unless &DRM_MODE_FB_MODIFIERS is set in @flags. + */ + __u64 modifier[4]; }; #define DRM_MODE_FB_DIRTY_ANNOTATE_COPY 0x01 -- cgit v1.2.3 From a65dbf7cded724a5ed4a5e1a718616b048ca0c34 Mon Sep 17 00:00:00 2001 From: Prike Liang Date: Thu, 23 Dec 2021 09:35:37 +0800 Subject: drm/amdgpu/gfx10: Add GC 10.3.7 Support Needed to properly initialize GC 10.3.7. Signed-off-by: Prike Liang Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 5 +++ drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 53 ++++++++++++++++++++++++++- include/uapi/drm/amdgpu_drm.h | 1 + 3 files changed, 58 insertions(+), 1 deletion(-) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index 0274de05e333..7c7e28fd912e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -1424,6 +1424,7 @@ static int amdgpu_discovery_set_gc_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 7): amdgpu_device_ip_block_add(adev, &gfx_v10_0_ip_block); break; default: @@ -1766,6 +1767,9 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(10, 3, 3): adev->family = AMDGPU_FAMILY_YC; break; + case IP_VERSION(10, 3, 7): + adev->family = AMDGPU_FAMILY_GC_10_3_7; + break; default: return -EINVAL; } @@ -1778,6 +1782,7 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(10, 1, 4): case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 7): adev->flags |= AMD_IS_APU; break; default: diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index dfbe65c1ae0b..95b8f76d0e7a 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -258,6 +258,13 @@ MODULE_FIRMWARE("amdgpu/cyan_skillfish2_mec.bin"); MODULE_FIRMWARE("amdgpu/cyan_skillfish2_mec2.bin"); MODULE_FIRMWARE("amdgpu/cyan_skillfish2_rlc.bin"); +MODULE_FIRMWARE("amdgpu/gc_10_3_7_ce.bin"); +MODULE_FIRMWARE("amdgpu/gc_10_3_7_pfp.bin"); +MODULE_FIRMWARE("amdgpu/gc_10_3_7_me.bin"); +MODULE_FIRMWARE("amdgpu/gc_10_3_7_mec.bin"); +MODULE_FIRMWARE("amdgpu/gc_10_3_7_mec2.bin"); +MODULE_FIRMWARE("amdgpu/gc_10_3_7_rlc.bin"); + static const struct soc15_reg_golden golden_settings_gc_10_1[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_4, 0xffffffff, 0x00400014), @@ -3408,6 +3415,31 @@ static const struct soc15_reg_golden golden_settings_gc_10_0_cyan_skillfish[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0x00800000, 0x00800000) }; +static const struct soc15_reg_golden golden_settings_gc_10_3_7[] = { + SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CS_CLK_CTRL, 0xff7f0fff, 0x78000100), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmCH_PIPE_STEER, 0x000000ff, 0x000000e4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmCPF_GCR_CNTL, 0x0007ffff, 0x0000c200), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG3, 0xffffffff, 0x00000280), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG4, 0xffffffff, 0x00800000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_ADDR_CONFIG, 0x0c1807ff, 0x00000041), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCR_GENERAL_CNTL, 0x1ff1ffff, 0x00000500), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL1_PIPE_STEER, 0x000000ff, 0x000000e4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2_PIPE_STEER_0, 0x77777777, 0x32103210), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2_PIPE_STEER_1, 0x77777777, 0x32103210), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2A_ADDR_MATCH_MASK, 0xffffffff, 0xffffffff), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_ADDR_MATCH_MASK, 0xffffffff, 0xffffffff), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CM_CTRL1, 0xff8fff0f, 0x580f1008), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CTRL3, 0xf7ffffff, 0x00f80988), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmLDS_CONFIG, 0x000001ff, 0x00000020), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_CL_ENHANCE, 0xf000003f, 0x01200007), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_BINNER_TIMEOUT_COUNTER, 0xffffffff, 0x00000800), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE_2, 0xffffffbf, 0x00000820), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQG_CONFIG, 0x000017ff, 0x00001000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSX_DEBUG_1, 0xffffff7f, 0x00010020), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX, 0xfff7ffff, 0x01030000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0xffffffff, 0x00100000) +}; + #define DEFAULT_SH_MEM_CONFIG \ ((SH_MEM_ADDRESS_MODE_64 << SH_MEM_CONFIG__ADDRESS_MODE__SHIFT) | \ (SH_MEM_ALIGNMENT_MODE_UNALIGNED << SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT) | \ @@ -3646,6 +3678,11 @@ static void gfx_v10_0_init_golden_registers(struct amdgpu_device *adev) golden_settings_gc_10_0_cyan_skillfish, (const u32)ARRAY_SIZE(golden_settings_gc_10_0_cyan_skillfish)); break; + case IP_VERSION(10, 3, 7): + soc15_program_register_sequence(adev, + golden_settings_gc_10_3_7, + (const u32)ARRAY_SIZE(golden_settings_gc_10_3_7)); + break; default: break; } @@ -3835,6 +3872,7 @@ static void gfx_v10_0_check_fw_write_wait(struct amdgpu_device *adev) case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 7): adev->gfx.cp_fw_write_wait = true; break; default: @@ -3962,6 +4000,9 @@ static int gfx_v10_0_init_microcode(struct amdgpu_device *adev) else chip_name = "cyan_skillfish"; break; + case IP_VERSION(10, 3, 7): + chip_name = "gc_10_3_7"; + break; default: BUG(); } @@ -4558,6 +4599,7 @@ static void gfx_v10_0_gpu_early_init(struct amdgpu_device *adev) case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 7): adev->gfx.config.max_hw_contexts = 8; adev->gfx.config.sc_prim_fifo_size_frontend = 0x20; adev->gfx.config.sc_prim_fifo_size_backend = 0x100; @@ -4695,6 +4737,7 @@ static int gfx_v10_0_sw_init(void *handle) case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 7): adev->gfx.me.num_me = 1; adev->gfx.me.num_pipe_per_me = 1; adev->gfx.me.num_queue_per_pipe = 1; @@ -6207,6 +6250,7 @@ static void gfx_v10_0_cp_gfx_set_doorbell(struct amdgpu_device *adev, case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 7): tmp = REG_SET_FIELD(0, CP_RB_DOORBELL_RANGE_LOWER, DOORBELL_RANGE_LOWER_Sienna_Cichlid, ring->doorbell_index); WREG32_SOC15(GC, 0, mmCP_RB_DOORBELL_RANGE_LOWER, tmp); @@ -6344,6 +6388,7 @@ static void gfx_v10_0_cp_compute_enable(struct amdgpu_device *adev, bool enable) case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 7): WREG32_SOC15(GC, 0, mmCP_MEC_CNTL_Sienna_Cichlid, 0); break; default: @@ -6358,6 +6403,7 @@ static void gfx_v10_0_cp_compute_enable(struct amdgpu_device *adev, bool enable) case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 7): WREG32_SOC15(GC, 0, mmCP_MEC_CNTL_Sienna_Cichlid, (CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK)); @@ -7185,6 +7231,7 @@ static bool gfx_v10_0_check_grbm_cam_remapping(struct amdgpu_device *adev) break; case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 7): return true; default: data = RREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE); @@ -7220,6 +7267,7 @@ static void gfx_v10_0_setup_grbm_cam_remapping(struct amdgpu_device *adev) case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 7): /* mmVGT_TF_RING_SIZE_UMD -> mmVGT_TF_RING_SIZE */ data = (SOC15_REG_OFFSET(GC, 0, mmVGT_TF_RING_SIZE_UMD) << GRBM_CAM_DATA__CAM_ADDR__SHIFT) | @@ -7672,6 +7720,7 @@ static int gfx_v10_0_early_init(void *handle) case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 7): adev->gfx.num_gfx_rings = GFX10_NUM_GFX_RINGS_Sienna_Cichlid; break; default: @@ -9430,6 +9479,7 @@ static void gfx_v10_0_set_rlc_funcs(struct amdgpu_device *adev) case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 7): adev->gfx.rlc.funcs = &gfx_v10_0_rlc_funcs; break; case IP_VERSION(10, 1, 2): @@ -9522,7 +9572,8 @@ static int gfx_v10_0_get_cu_info(struct amdgpu_device *adev, for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { bitmap = i * adev->gfx.config.max_sh_per_se + j; if (((adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 0)) || - (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 3))) && + (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 3)) || + (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 7))) && ((gfx_v10_3_get_disabled_sa(adev) >> bitmap) & 1)) continue; mask = 1; diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index 76b580d10a52..55fa660e35f4 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -1151,6 +1151,7 @@ struct drm_amdgpu_info_video_caps { #define AMDGPU_FAMILY_NV 143 /* Navi10 */ #define AMDGPU_FAMILY_VGH 144 /* Van Gogh */ #define AMDGPU_FAMILY_YC 146 /* Yellow Carp */ +#define AMDGPU_FAMILY_GC_10_3_7 151 /* GC 10.3.7 */ #if defined(__cplusplus) } -- cgit v1.2.3 From 874bfdfa4735cbb1b0d6e0c6157c712a312371a1 Mon Sep 17 00:00:00 2001 From: Yifan Zhang Date: Thu, 10 Feb 2022 14:42:09 -0500 Subject: drm/amdgpu: add gc 10.3.6 support this patch adds gc 10.3.6 support. Signed-off-by: Yifan Zhang Reviewed-by: Alex Deucher Reviewed-by: Huang Rui Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 6 ++ drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 87 ++++++++++++++++++++++++++- include/uapi/drm/amdgpu_drm.h | 1 + 3 files changed, 91 insertions(+), 3 deletions(-) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index 1ba0b055d44b..3501ade93f44 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -1421,6 +1421,7 @@ static int amdgpu_discovery_set_gc_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): case IP_VERSION(10, 3, 7): amdgpu_device_ip_block_add(adev, &gfx_v10_0_ip_block); @@ -1559,6 +1560,7 @@ static int amdgpu_discovery_set_mes_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(10, 3, 3): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): amdgpu_device_ip_block_add(adev, &mes_v10_1_ip_block); break; default: @@ -1765,6 +1767,9 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(10, 3, 3): adev->family = AMDGPU_FAMILY_YC; break; + case IP_VERSION(10, 3, 6): + adev->family = AMDGPU_FAMILY_GC_10_3_6; + break; case IP_VERSION(10, 3, 7): adev->family = AMDGPU_FAMILY_GC_10_3_7; break; @@ -1780,6 +1785,7 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(10, 1, 4): case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 7): adev->flags |= AMD_IS_APU; break; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index 95b8f76d0e7a..90158289cd30 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -106,6 +106,12 @@ #define mmGOLDEN_TSC_COUNT_UPPER_Vangogh_BASE_IDX 1 #define mmGOLDEN_TSC_COUNT_LOWER_Vangogh 0x0026 #define mmGOLDEN_TSC_COUNT_LOWER_Vangogh_BASE_IDX 1 + +#define mmGOLDEN_TSC_COUNT_UPPER_GC_10_3_6 0x002d +#define mmGOLDEN_TSC_COUNT_UPPER_GC_10_3_6_BASE_IDX 1 +#define mmGOLDEN_TSC_COUNT_LOWER_GC_10_3_6 0x002e +#define mmGOLDEN_TSC_COUNT_LOWER_GC_10_3_6_BASE_IDX 1 + #define mmSPI_CONFIG_CNTL_1_Vangogh 0x2441 #define mmSPI_CONFIG_CNTL_1_Vangogh_BASE_IDX 1 #define mmVGT_TF_MEMORY_BASE_HI_Vangogh 0x2261 @@ -258,6 +264,13 @@ MODULE_FIRMWARE("amdgpu/cyan_skillfish2_mec.bin"); MODULE_FIRMWARE("amdgpu/cyan_skillfish2_mec2.bin"); MODULE_FIRMWARE("amdgpu/cyan_skillfish2_rlc.bin"); +MODULE_FIRMWARE("amdgpu/gc_10_3_6_ce.bin"); +MODULE_FIRMWARE("amdgpu/gc_10_3_6_pfp.bin"); +MODULE_FIRMWARE("amdgpu/gc_10_3_6_me.bin"); +MODULE_FIRMWARE("amdgpu/gc_10_3_6_mec.bin"); +MODULE_FIRMWARE("amdgpu/gc_10_3_6_mec2.bin"); +MODULE_FIRMWARE("amdgpu/gc_10_3_6_rlc.bin"); + MODULE_FIRMWARE("amdgpu/gc_10_3_7_ce.bin"); MODULE_FIRMWARE("amdgpu/gc_10_3_7_pfp.bin"); MODULE_FIRMWARE("amdgpu/gc_10_3_7_me.bin"); @@ -3415,6 +3428,32 @@ static const struct soc15_reg_golden golden_settings_gc_10_0_cyan_skillfish[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0x00800000, 0x00800000) }; +static const struct soc15_reg_golden golden_settings_gc_10_3_6[] = +{ + SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CS_CLK_CTRL, 0xff7f0fff, 0x78000100), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmCH_PIPE_STEER, 0x000000ff, 0x00000044), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmCPF_GCR_CNTL, 0x0007ffff, 0x0000c200), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG3, 0xffffffff, 0x00000280), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG4, 0xffffffff, 0x00800000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_ADDR_CONFIG, 0x0c1807ff, 0x00000042), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCR_GENERAL_CNTL, 0x1ff1ffff, 0x00000500), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL1_PIPE_STEER, 0x000000ff, 0x00000044), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2_PIPE_STEER_0, 0x77777777, 0x32103210), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2_PIPE_STEER_1, 0x77777777, 0x32103210), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2A_ADDR_MATCH_MASK, 0xffffffff, 0xfffffff3), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_ADDR_MATCH_MASK, 0xffffffff, 0xfffffff3), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CM_CTRL1, 0xff8fff0f, 0x580f1008), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CTRL3, 0xf7ffffff, 0x00f80988), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmLDS_CONFIG, 0x000001ff, 0x00000020), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_CL_ENHANCE, 0xf17fffff, 0x01200007), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_BINNER_TIMEOUT_COUNTER, 0xffffffff, 0x00000800), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE_2, 0xffffffbf, 0x00000820), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQG_CONFIG, 0x000017ff, 0x00001000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSX_DEBUG_1, 0xffffff7f, 0x00010020), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX, 0xfff7ffff, 0x01030000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0xffffffff, 0x00100000) +}; + static const struct soc15_reg_golden golden_settings_gc_10_3_7[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CS_CLK_CTRL, 0xff7f0fff, 0x78000100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCH_PIPE_STEER, 0x000000ff, 0x000000e4), @@ -3678,6 +3717,11 @@ static void gfx_v10_0_init_golden_registers(struct amdgpu_device *adev) golden_settings_gc_10_0_cyan_skillfish, (const u32)ARRAY_SIZE(golden_settings_gc_10_0_cyan_skillfish)); break; + case IP_VERSION(10, 3, 6): + soc15_program_register_sequence(adev, + golden_settings_gc_10_3_6, + (const u32)ARRAY_SIZE(golden_settings_gc_10_3_6)); + break; case IP_VERSION(10, 3, 7): soc15_program_register_sequence(adev, golden_settings_gc_10_3_7, @@ -3871,6 +3915,7 @@ static void gfx_v10_0_check_fw_write_wait(struct amdgpu_device *adev) case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): case IP_VERSION(10, 3, 7): adev->gfx.cp_fw_write_wait = true; @@ -3993,6 +4038,9 @@ static int gfx_v10_0_init_microcode(struct amdgpu_device *adev) case IP_VERSION(10, 3, 3): chip_name = "yellow_carp"; break; + case IP_VERSION(10, 3, 6): + chip_name = "gc_10_3_6"; + break; case IP_VERSION(10, 1, 3): case IP_VERSION(10, 1, 4): if (adev->apu_flags & AMD_APU_IS_CYAN_SKILLFISH2) @@ -4598,6 +4646,7 @@ static void gfx_v10_0_gpu_early_init(struct amdgpu_device *adev) case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): case IP_VERSION(10, 3, 7): adev->gfx.config.max_hw_contexts = 8; @@ -4736,6 +4785,7 @@ static int gfx_v10_0_sw_init(void *handle) case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): case IP_VERSION(10, 3, 7): adev->gfx.me.num_me = 1; @@ -4975,7 +5025,8 @@ static void gfx_v10_0_setup_rb(struct amdgpu_device *adev) for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { bitmap = i * adev->gfx.config.max_sh_per_se + j; if (((adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 0)) || - (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 3))) && + (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 3)) || + (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 6))) && ((gfx_v10_3_get_disabled_sa(adev) >> bitmap) & 1)) continue; gfx_v10_0_select_se_sh(adev, i, j, 0xffffffff); @@ -6249,6 +6300,7 @@ static void gfx_v10_0_cp_gfx_set_doorbell(struct amdgpu_device *adev, case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): case IP_VERSION(10, 3, 7): tmp = REG_SET_FIELD(0, CP_RB_DOORBELL_RANGE_LOWER, @@ -6387,6 +6439,7 @@ static void gfx_v10_0_cp_compute_enable(struct amdgpu_device *adev, bool enable) case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): case IP_VERSION(10, 3, 7): WREG32_SOC15(GC, 0, mmCP_MEC_CNTL_Sienna_Cichlid, 0); @@ -6402,6 +6455,7 @@ static void gfx_v10_0_cp_compute_enable(struct amdgpu_device *adev, bool enable) case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): case IP_VERSION(10, 3, 7): WREG32_SOC15(GC, 0, mmCP_MEC_CNTL_Sienna_Cichlid, @@ -6501,6 +6555,7 @@ static void gfx_v10_0_kiq_setting(struct amdgpu_ring *ring) case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): tmp = RREG32_SOC15(GC, 0, mmRLC_CP_SCHEDULERS_Sienna_Cichlid); tmp &= 0xffffff00; @@ -7231,6 +7286,7 @@ static bool gfx_v10_0_check_grbm_cam_remapping(struct amdgpu_device *adev) break; case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 7): return true; default: @@ -7266,6 +7322,7 @@ static void gfx_v10_0_setup_grbm_cam_remapping(struct amdgpu_device *adev) case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): case IP_VERSION(10, 3, 7): /* mmVGT_TF_RING_SIZE_UMD -> mmVGT_TF_RING_SIZE */ @@ -7587,6 +7644,7 @@ static int gfx_v10_0_soft_reset(void *handle) case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): if (REG_GET_FIELD(tmp, GRBM_STATUS2, RLC_BUSY_Sienna_Cichlid)) grbm_soft_reset = REG_SET_FIELD(grbm_soft_reset, @@ -7654,6 +7712,21 @@ static uint64_t gfx_v10_0_get_gpu_clock_counter(struct amdgpu_device *adev) preempt_enable(); clock = clock_lo | (clock_hi << 32ULL); break; + case IP_VERSION(10, 3, 6): + preempt_disable(); + clock_hi = RREG32_SOC15_NO_KIQ(SMUIO, 0, mmGOLDEN_TSC_COUNT_UPPER_GC_10_3_6); + clock_lo = RREG32_SOC15_NO_KIQ(SMUIO, 0, mmGOLDEN_TSC_COUNT_LOWER_GC_10_3_6); + hi_check = RREG32_SOC15_NO_KIQ(SMUIO, 0, mmGOLDEN_TSC_COUNT_UPPER_GC_10_3_6); + /* The SMUIO TSC clock frequency is 100MHz, which sets 32-bit carry over + * roughly every 42 seconds. + */ + if (hi_check != clock_hi) { + clock_lo = RREG32_SOC15_NO_KIQ(SMUIO, 0, mmGOLDEN_TSC_COUNT_LOWER_GC_10_3_6); + clock_hi = hi_check; + } + preempt_enable(); + clock = clock_lo | (clock_hi << 32ULL); + break; default: preempt_disable(); clock_hi = RREG32_SOC15_NO_KIQ(SMUIO, 0, mmGOLDEN_TSC_COUNT_UPPER); @@ -7719,6 +7792,7 @@ static int gfx_v10_0_early_init(void *handle) case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): case IP_VERSION(10, 3, 7): adev->gfx.num_gfx_rings = GFX10_NUM_GFX_RINGS_Sienna_Cichlid; @@ -7781,6 +7855,7 @@ static void gfx_v10_0_set_safe_mode(struct amdgpu_device *adev) case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): WREG32_SOC15(GC, 0, mmRLC_SAFE_MODE_Sienna_Cichlid, data); @@ -7817,6 +7892,7 @@ static void gfx_v10_0_unset_safe_mode(struct amdgpu_device *adev) case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): WREG32_SOC15(GC, 0, mmRLC_SAFE_MODE_Sienna_Cichlid, data); break; @@ -8271,6 +8347,7 @@ static void gfx_v10_cntl_power_gating(struct amdgpu_device *adev, bool enable) switch (adev->ip_versions[GC_HWIP][0]) { case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 6): data = 0x4E20 & RLC_PG_DELAY_3__CGCG_ACTIVE_BEFORE_CGPG_MASK_Vangogh; WREG32_SOC15(GC, 0, mmRLC_PG_DELAY_3, data); break; @@ -8339,6 +8416,7 @@ static int gfx_v10_0_set_powergating_state(void *handle, break; case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 3): + case IP_VERSION(10, 3, 6): gfx_v10_cntl_pg(adev, enable); amdgpu_gfx_off_ctrl(adev, enable); break; @@ -8365,6 +8443,7 @@ static int gfx_v10_0_set_clockgating_state(void *handle, case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): gfx_v10_0_update_gfx_clock_gating(adev, state == AMD_CG_STATE_GATE); @@ -9478,6 +9557,7 @@ static void gfx_v10_0_set_rlc_funcs(struct amdgpu_device *adev) case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 4): case IP_VERSION(10, 3, 5): + case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): case IP_VERSION(10, 3, 7): adev->gfx.rlc.funcs = &gfx_v10_0_rlc_funcs; @@ -9572,8 +9652,9 @@ static int gfx_v10_0_get_cu_info(struct amdgpu_device *adev, for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { bitmap = i * adev->gfx.config.max_sh_per_se + j; if (((adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 0)) || - (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 3)) || - (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 7))) && + (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 3)) || + (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 6)) || + (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 7))) && ((gfx_v10_3_get_disabled_sa(adev) >> bitmap) & 1)) continue; mask = 1; diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index 55fa660e35f4..1d65c1fbc4ec 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -1151,6 +1151,7 @@ struct drm_amdgpu_info_video_caps { #define AMDGPU_FAMILY_NV 143 /* Navi10 */ #define AMDGPU_FAMILY_VGH 144 /* Van Gogh */ #define AMDGPU_FAMILY_YC 146 /* Yellow Carp */ +#define AMDGPU_FAMILY_GC_10_3_6 149 /* GC 10.3.6 */ #define AMDGPU_FAMILY_GC_10_3_7 151 /* GC 10.3.7 */ #if defined(__cplusplus) -- cgit v1.2.3 From caa574ffc4aaf4f29b890223878c63e2e7772f62 Mon Sep 17 00:00:00 2001 From: Matthew Auld Date: Sat, 19 Feb 2022 00:17:49 +0530 Subject: drm/i915/uapi: document behaviour for DG2 64K support MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit On discrete platforms like DG2, we need to support a minimum page size of 64K when dealing with device local-memory. This is quite tricky for various reasons, so try to document the new implicit uapi for this. v4: Kdoc modification. v3: fix typos and less emphasis v2: Fixed suggestions on formatting [Daniel] Signed-off-by: Matthew Auld Signed-off-by: Ramalingam C Signed-off-by: Robert Beckett Acked-by: Jordan Justen Reviewed-by: Ramalingam C Reviewed-by: Thomas Hellström cc: Simon Ser cc: Pekka Paalanen Cc: Jordan Justen Cc: Kenneth Graunke Cc: mesa-dev@lists.freedesktop.org Cc: Tony Ye Cc: Slawomir Milczarek Signed-off-by: Lucas De Marchi Link: https://patchwork.freedesktop.org/patch/msgid/20220218184752.7524-13-ramalingam.c@intel.com --- include/uapi/drm/i915_drm.h | 45 ++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 40 insertions(+), 5 deletions(-) (limited to 'include/uapi/drm') diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index 914ebd9290e5..05c3642aaece 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -1118,10 +1118,16 @@ struct drm_i915_gem_exec_object2 { /** * When the EXEC_OBJECT_PINNED flag is specified this is populated by * the user with the GTT offset at which this object will be pinned. + * * When the I915_EXEC_NO_RELOC flag is specified this must contain the * presumed_offset of the object. + * * During execbuffer2 the kernel populates it with the value of the * current GTT offset of the object, for future presumed_offset writes. + * + * See struct drm_i915_gem_create_ext for the rules when dealing with + * alignment restrictions with I915_MEMORY_CLASS_DEVICE, on devices with + * minimum page sizes, like DG2. */ __u64 offset; @@ -3144,11 +3150,40 @@ struct drm_i915_gem_create_ext { * * The (page-aligned) allocated size for the object will be returned. * - * Note that for some devices we have might have further minimum - * page-size restrictions(larger than 4K), like for device local-memory. - * However in general the final size here should always reflect any - * rounding up, if for example using the I915_GEM_CREATE_EXT_MEMORY_REGIONS - * extension to place the object in device local-memory. + * + * DG2 64K min page size implications: + * + * On discrete platforms, starting from DG2, we have to contend with GTT + * page size restrictions when dealing with I915_MEMORY_CLASS_DEVICE + * objects. Specifically the hardware only supports 64K or larger GTT + * page sizes for such memory. The kernel will already ensure that all + * I915_MEMORY_CLASS_DEVICE memory is allocated using 64K or larger page + * sizes underneath. + * + * Note that the returned size here will always reflect any required + * rounding up done by the kernel, i.e 4K will now become 64K on devices + * such as DG2. + * + * Special DG2 GTT address alignment requirement: + * + * The GTT alignment will also need to be at least 2M for such objects. + * + * Note that due to how the hardware implements 64K GTT page support, we + * have some further complications: + * + * 1) The entire PDE (which covers a 2MB virtual address range), must + * contain only 64K PTEs, i.e mixing 4K and 64K PTEs in the same + * PDE is forbidden by the hardware. + * + * 2) We still need to support 4K PTEs for I915_MEMORY_CLASS_SYSTEM + * objects. + * + * To keep things simple for userland, we mandate that any GTT mappings + * must be aligned to and rounded up to 2MB. The kernel will internally + * pad them out to the next 2MB boundary. As this only wastes virtual + * address space and avoids userland having to copy any needlessly + * complicated PDE sharing scheme (coloring) and only affects DG2, this + * is deemed to be a good compromise. */ __u64 size; /** -- cgit v1.2.3 From c6e7deb0f092616bd8cb19e8c436b212c64daaab Mon Sep 17 00:00:00 2001 From: Stanislav Lisovskiy Date: Tue, 18 Jan 2022 13:55:43 +0200 Subject: drm/i915: Introduce new Tile 4 format This tiling layout uses 4KB tiles in a row-major layout. It has the same shape as Tile Y at two granularities: 4KB (128B x 32) and 64B (16B x 4). It only differs from Tile Y at the 256B granularity in between. At this granularity, Tile Y has a shape of 16B x 32 rows, but this tiling has a shape of 64B x 8 rows. Reviewed-by: Imre Deak Acked-by: Nanley Chery Signed-off-by: Stanislav Lisovskiy Link: https://patchwork.freedesktop.org/patch/msgid/20220118115544.15116-2-stanislav.lisovskiy@intel.com --- include/uapi/drm/drm_fourcc.h | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'include/uapi/drm') diff --git a/include/uapi/drm/drm_fourcc.h b/include/uapi/drm/drm_fourcc.h index fc0c1454d275..b73fe6797fc3 100644 --- a/include/uapi/drm/drm_fourcc.h +++ b/include/uapi/drm/drm_fourcc.h @@ -572,6 +572,17 @@ extern "C" { */ #define I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS_CC fourcc_mod_code(INTEL, 8) +/* + * Intel Tile 4 layout + * + * This is a tiled layout using 4KB tiles in a row-major layout. It has the same + * shape as Tile Y at two granularities: 4KB (128B x 32) and 64B (16B x 4). It + * only differs from Tile Y at the 256B granularity in between. At this + * granularity, Tile Y has a shape of 16B x 32 rows, but this tiling has a shape + * of 64B x 8 rows. + */ +#define I915_FORMAT_MOD_4_TILED fourcc_mod_code(INTEL, 9) + /* * Tiled, NV12MT, grouped in 64 (pixels) x 32 (lines) -sized macroblocks * -- cgit v1.2.3 From f7ddbf5581b474fe4a0a29244acaa1bf72234675 Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Thu, 3 Mar 2022 16:52:15 -0800 Subject: drm/msm: Add SET_PARAM ioctl It was always expected to have a use for this some day, so we left a placeholder. Now we do. (And I expect another use in the not too distant future when we start allowing userspace to allocate GPU iova.) Signed-off-by: Rob Clark Link: https://lore.kernel.org/r/20220304005317.776110-3-robdclark@gmail.com --- drivers/gpu/drm/msm/adreno/a2xx_gpu.c | 1 + drivers/gpu/drm/msm/adreno/a3xx_gpu.c | 1 + drivers/gpu/drm/msm/adreno/a4xx_gpu.c | 1 + drivers/gpu/drm/msm/adreno/a5xx_gpu.c | 1 + drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 1 + drivers/gpu/drm/msm/adreno/adreno_gpu.c | 10 ++++++++++ drivers/gpu/drm/msm/adreno/adreno_gpu.h | 2 ++ drivers/gpu/drm/msm/msm_drv.c | 20 ++++++++++++++++++++ drivers/gpu/drm/msm/msm_gpu.h | 2 ++ include/uapi/drm/msm_drm.h | 27 +++++++++++++++------------ 10 files changed, 54 insertions(+), 12 deletions(-) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/msm/adreno/a2xx_gpu.c b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c index 22e8295a5e2b..6c9a747eb4ad 100644 --- a/drivers/gpu/drm/msm/adreno/a2xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c @@ -471,6 +471,7 @@ static u32 a2xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) static const struct adreno_gpu_funcs funcs = { .base = { .get_param = adreno_get_param, + .set_param = adreno_set_param, .hw_init = a2xx_hw_init, .pm_suspend = msm_gpu_pm_suspend, .pm_resume = msm_gpu_pm_resume, diff --git a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c index 2e481e2692ba..0ab0e1dd8bbb 100644 --- a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c @@ -486,6 +486,7 @@ static u32 a3xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) static const struct adreno_gpu_funcs funcs = { .base = { .get_param = adreno_get_param, + .set_param = adreno_set_param, .hw_init = a3xx_hw_init, .pm_suspend = msm_gpu_pm_suspend, .pm_resume = msm_gpu_pm_resume, diff --git a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c index c5524d6e8705..0c6b2a6d0b4c 100644 --- a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c @@ -621,6 +621,7 @@ static u32 a4xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) static const struct adreno_gpu_funcs funcs = { .base = { .get_param = adreno_get_param, + .set_param = adreno_set_param, .hw_init = a4xx_hw_init, .pm_suspend = a4xx_pm_suspend, .pm_resume = a4xx_pm_resume, diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c index 3d28fcf841a6..407f50a15faa 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c @@ -1700,6 +1700,7 @@ static uint32_t a5xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) static const struct adreno_gpu_funcs funcs = { .base = { .get_param = adreno_get_param, + .set_param = adreno_set_param, .hw_init = a5xx_hw_init, .pm_suspend = a5xx_pm_suspend, .pm_resume = a5xx_pm_resume, diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c index 7d23c741db4a..237c2e7a7baa 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c @@ -1800,6 +1800,7 @@ done: static const struct adreno_gpu_funcs funcs = { .base = { .get_param = adreno_get_param, + .set_param = adreno_set_param, .hw_init = a6xx_hw_init, .pm_suspend = a6xx_pm_suspend, .pm_resume = a6xx_pm_resume, diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index 15c8997b7251..6a37d409653b 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -283,6 +283,16 @@ int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx, } } +int adreno_set_param(struct msm_gpu *gpu, struct msm_file_private *ctx, + uint32_t param, uint64_t value) +{ + switch (param) { + default: + DBG("%s: invalid param: %u", gpu->name, param); + return -EINVAL; + } +} + const struct firmware * adreno_request_fw(struct adreno_gpu *adreno_gpu, const char *fwname) { diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h index b1ee453d627d..0490c5fbb780 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h @@ -282,6 +282,8 @@ static inline int adreno_is_a650_family(struct adreno_gpu *gpu) int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx, uint32_t param, uint64_t *value); +int adreno_set_param(struct msm_gpu *gpu, struct msm_file_private *ctx, + uint32_t param, uint64_t value); const struct firmware *adreno_request_fw(struct adreno_gpu *adreno_gpu, const char *fwname); struct drm_gem_object *adreno_fw_create_bo(struct msm_gpu *gpu, diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 30fd18ca88c4..c4d90a9b7010 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -613,6 +613,25 @@ static int msm_ioctl_get_param(struct drm_device *dev, void *data, args->param, &args->value); } +static int msm_ioctl_set_param(struct drm_device *dev, void *data, + struct drm_file *file) +{ + struct msm_drm_private *priv = dev->dev_private; + struct drm_msm_param *args = data; + struct msm_gpu *gpu; + + if (args->pipe != MSM_PIPE_3D0) + return -EINVAL; + + gpu = priv->gpu; + + if (!gpu) + return -ENXIO; + + return gpu->funcs->set_param(gpu, file->driver_priv, + args->param, args->value); +} + static int msm_ioctl_gem_new(struct drm_device *dev, void *data, struct drm_file *file) { @@ -898,6 +917,7 @@ static int msm_ioctl_submitqueue_close(struct drm_device *dev, void *data, static const struct drm_ioctl_desc msm_ioctls[] = { DRM_IOCTL_DEF_DRV(MSM_GET_PARAM, msm_ioctl_get_param, DRM_RENDER_ALLOW), + DRM_IOCTL_DEF_DRV(MSM_SET_PARAM, msm_ioctl_set_param, DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(MSM_GEM_NEW, msm_ioctl_gem_new, DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(MSM_GEM_INFO, msm_ioctl_gem_info, DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(MSM_GEM_CPU_PREP, msm_ioctl_gem_cpu_prep, DRM_RENDER_ALLOW), diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h index c99627fc99dd..07ee6573a301 100644 --- a/drivers/gpu/drm/msm/msm_gpu.h +++ b/drivers/gpu/drm/msm/msm_gpu.h @@ -44,6 +44,8 @@ struct msm_gpu_config { struct msm_gpu_funcs { int (*get_param)(struct msm_gpu *gpu, struct msm_file_private *ctx, uint32_t param, uint64_t *value); + int (*set_param)(struct msm_gpu *gpu, struct msm_file_private *ctx, + uint32_t param, uint64_t value); int (*hw_init)(struct msm_gpu *gpu); int (*pm_suspend)(struct msm_gpu *gpu); int (*pm_resume)(struct msm_gpu *gpu); diff --git a/include/uapi/drm/msm_drm.h b/include/uapi/drm/msm_drm.h index 6b8fffc28a50..cf5de53836e7 100644 --- a/include/uapi/drm/msm_drm.h +++ b/include/uapi/drm/msm_drm.h @@ -67,16 +67,20 @@ struct drm_msm_timespec { __s64 tv_nsec; /* nanoseconds */ }; -#define MSM_PARAM_GPU_ID 0x01 -#define MSM_PARAM_GMEM_SIZE 0x02 -#define MSM_PARAM_CHIP_ID 0x03 -#define MSM_PARAM_MAX_FREQ 0x04 -#define MSM_PARAM_TIMESTAMP 0x05 -#define MSM_PARAM_GMEM_BASE 0x06 -#define MSM_PARAM_PRIORITIES 0x07 /* The # of priority levels */ -#define MSM_PARAM_PP_PGTABLE 0x08 /* => 1 for per-process pagetables, else 0 */ -#define MSM_PARAM_FAULTS 0x09 -#define MSM_PARAM_SUSPENDS 0x0a +/* Below "RO" indicates a read-only param, "WO" indicates write-only, and + * "RW" indicates a param that can be both read (GET_PARAM) and written + * (SET_PARAM) + */ +#define MSM_PARAM_GPU_ID 0x01 /* RO */ +#define MSM_PARAM_GMEM_SIZE 0x02 /* RO */ +#define MSM_PARAM_CHIP_ID 0x03 /* RO */ +#define MSM_PARAM_MAX_FREQ 0x04 /* RO */ +#define MSM_PARAM_TIMESTAMP 0x05 /* RO */ +#define MSM_PARAM_GMEM_BASE 0x06 /* RO */ +#define MSM_PARAM_PRIORITIES 0x07 /* RO: The # of priority levels */ +#define MSM_PARAM_PP_PGTABLE 0x08 /* RO: Deprecated, always returns zero */ +#define MSM_PARAM_FAULTS 0x09 /* RO */ +#define MSM_PARAM_SUSPENDS 0x0a /* RO */ /* For backwards compat. The original support for preemption was based on * a single ring per priority level so # of priority levels equals the # @@ -333,9 +337,7 @@ struct drm_msm_submitqueue_query { }; #define DRM_MSM_GET_PARAM 0x00 -/* placeholder: #define DRM_MSM_SET_PARAM 0x01 - */ #define DRM_MSM_GEM_NEW 0x02 #define DRM_MSM_GEM_INFO 0x03 #define DRM_MSM_GEM_CPU_PREP 0x04 @@ -351,6 +353,7 @@ struct drm_msm_submitqueue_query { #define DRM_MSM_SUBMITQUEUE_QUERY 0x0C #define DRM_IOCTL_MSM_GET_PARAM DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_GET_PARAM, struct drm_msm_param) +#define DRM_IOCTL_MSM_SET_PARAM DRM_IOW (DRM_COMMAND_BASE + DRM_MSM_SET_PARAM, struct drm_msm_param) #define DRM_IOCTL_MSM_GEM_NEW DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_GEM_NEW, struct drm_msm_gem_new) #define DRM_IOCTL_MSM_GEM_INFO DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_GEM_INFO, struct drm_msm_gem_info) #define DRM_IOCTL_MSM_GEM_CPU_PREP DRM_IOW (DRM_COMMAND_BASE + DRM_MSM_GEM_CPU_PREP, struct drm_msm_gem_cpu_prep) -- cgit v1.2.3 From 90f45c42d7d7b0ec0fd797485c07fc421c474e12 Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Thu, 3 Mar 2022 16:52:16 -0800 Subject: drm/msm: Add SYSPROF param (v2) Add a SYSPROF param for system profiling tools like Mesa's pps-producer (perfetto) to control behavior related to system-wide performance counter collection. In particular, for profiling, one wants to ensure that GPU context switches do not effect perfcounter state, and might want to suppress suspend (which would cause counters to lose state). v2: Swap the order in msm_file_private_set_sysprof() [sboyd] and initialize the sysprof_active refcount to one (because the under/ overflow checking in refcount_t doesn't expect a 0->1 transition) meaning that values greater than 1 means sysprof is active. Signed-off-by: Rob Clark Link: https://lore.kernel.org/r/20220304005317.776110-4-robdclark@gmail.com --- drivers/gpu/drm/msm/adreno/adreno_gpu.c | 4 ++++ drivers/gpu/drm/msm/msm_drv.c | 8 +++++++ drivers/gpu/drm/msm/msm_gpu.c | 2 ++ drivers/gpu/drm/msm/msm_gpu.h | 27 +++++++++++++++++++++++ drivers/gpu/drm/msm/msm_submitqueue.c | 39 +++++++++++++++++++++++++++++++++ include/uapi/drm/msm_drm.h | 1 + 6 files changed, 81 insertions(+) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index 6a37d409653b..c91ea363c373 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -287,6 +287,10 @@ int adreno_set_param(struct msm_gpu *gpu, struct msm_file_private *ctx, uint32_t param, uint64_t value) { switch (param) { + case MSM_PARAM_SYSPROF: + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + return msm_file_private_set_sysprof(ctx, gpu, value); default: DBG("%s: invalid param: %u", gpu->name, param); return -EINVAL; diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index c4d90a9b7010..a2834663dec4 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -558,8 +558,16 @@ static void context_close(struct msm_file_private *ctx) static void msm_postclose(struct drm_device *dev, struct drm_file *file) { + struct msm_drm_private *priv = dev->dev_private; struct msm_file_private *ctx = file->driver_priv; + /* + * It is not possible to set sysprof param to non-zero if gpu + * is not initialized: + */ + if (priv->gpu) + msm_file_private_set_sysprof(ctx, priv->gpu, 0); + context_close(ctx); } diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index bacdabbaad96..faf0c242874e 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -959,6 +959,8 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev, gpu->nr_rings = nr_rings; + refcount_set(&gpu->sysprof_active, 1); + return 0; fail: diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h index 07ee6573a301..02419f2ca2bc 100644 --- a/drivers/gpu/drm/msm/msm_gpu.h +++ b/drivers/gpu/drm/msm/msm_gpu.h @@ -159,6 +159,13 @@ struct msm_gpu { struct msm_ringbuffer *rb[MSM_GPU_MAX_RINGS]; int nr_rings; + /** + * sysprof_active: + * + * The count of contexts that have enabled system profiling. + */ + refcount_t sysprof_active; + /** * cur_ctx_seqno: * @@ -329,6 +336,24 @@ struct msm_file_private { struct kref ref; int seqno; + /** + * sysprof: + * + * The value of MSM_PARAM_SYSPROF set by userspace. This is + * intended to be used by system profiling tools like Mesa's + * pps-producer (perfetto), and restricted to CAP_SYS_ADMIN. + * + * Setting a value of 1 will preserve performance counters across + * context switches. Setting a value of 2 will in addition + * suppress suspend. (Performance counters lose state across + * power collapse, which is undesirable for profiling in some + * cases.) + * + * The value automatically reverts to zero when the drm device + * file is closed. + */ + int sysprof; + /** * entities: * @@ -525,6 +550,8 @@ void msm_submitqueue_close(struct msm_file_private *ctx); void msm_submitqueue_destroy(struct kref *kref); +int msm_file_private_set_sysprof(struct msm_file_private *ctx, + struct msm_gpu *gpu, int sysprof); void __msm_file_private_destroy(struct kref *kref); static inline void msm_file_private_put(struct msm_file_private *ctx) diff --git a/drivers/gpu/drm/msm/msm_submitqueue.c b/drivers/gpu/drm/msm/msm_submitqueue.c index 7cb158bcbcf6..79b6ccd6ce64 100644 --- a/drivers/gpu/drm/msm/msm_submitqueue.c +++ b/drivers/gpu/drm/msm/msm_submitqueue.c @@ -7,6 +7,45 @@ #include "msm_gpu.h" +int msm_file_private_set_sysprof(struct msm_file_private *ctx, + struct msm_gpu *gpu, int sysprof) +{ + /* + * Since pm_runtime and sysprof_active are both refcounts, we + * call apply the new value first, and then unwind the previous + * value + */ + + switch (sysprof) { + default: + return -EINVAL; + case 2: + pm_runtime_get_sync(&gpu->pdev->dev); + fallthrough; + case 1: + refcount_inc(&gpu->sysprof_active); + fallthrough; + case 0: + break; + } + + /* unwind old value: */ + switch (ctx->sysprof) { + case 2: + pm_runtime_put_autosuspend(&gpu->pdev->dev); + fallthrough; + case 1: + refcount_dec(&gpu->sysprof_active); + fallthrough; + case 0: + break; + } + + ctx->sysprof = sysprof; + + return 0; +} + void __msm_file_private_destroy(struct kref *kref) { struct msm_file_private *ctx = container_of(kref, diff --git a/include/uapi/drm/msm_drm.h b/include/uapi/drm/msm_drm.h index cf5de53836e7..2ee03ba08681 100644 --- a/include/uapi/drm/msm_drm.h +++ b/include/uapi/drm/msm_drm.h @@ -81,6 +81,7 @@ struct drm_msm_timespec { #define MSM_PARAM_PP_PGTABLE 0x08 /* RO: Deprecated, always returns zero */ #define MSM_PARAM_FAULTS 0x09 /* RO */ #define MSM_PARAM_SUSPENDS 0x0a /* RO */ +#define MSM_PARAM_SYSPROF 0x0b /* WO: 1 preserves perfcntrs, 2 also disables suspend */ /* For backwards compat. The original support for preemption was based on * a single ring per priority level so # of priority levels equals the # -- cgit v1.2.3 From 17154addc5c1a175bcf3441ff0d9598efa1f05cd Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Thu, 24 Feb 2022 14:23:21 -0800 Subject: drm/msm: Add MSM_SUBMIT_FENCE_SN_IN Add a way for userspace to specify the sequence number fence used to track completion of the submit. As the seqno fence is simply an incrementing counter which is local to the submitqueue, it is easy for userspace to know the next value. This is useful for native userspace drivers in a vm guest, as the guest to host roundtrip can have high latency. Assigning the fence seqno in the guest userspace allows the guest to continue without waiting for response from the host. Signed-off-by: Rob Clark Link: https://lore.kernel.org/r/20220224222321.60653-1-robdclark@gmail.com --- drivers/gpu/drm/msm/msm_drv.c | 3 ++- drivers/gpu/drm/msm/msm_gem_submit.c | 42 ++++++++++++++++++++++++++++++------ include/uapi/drm/msm_drm.h | 4 +++- 3 files changed, 41 insertions(+), 8 deletions(-) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index a2834663dec4..e88c4b46a56f 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -41,9 +41,10 @@ * - 1.6.0 - Syncobj support * - 1.7.0 - Add MSM_PARAM_SUSPENDS to access suspend count * - 1.8.0 - Add MSM_BO_CACHED_COHERENT for supported GPUs (a6xx) + * - 1.9.0 - Add MSM_SUBMIT_FENCE_SN_IN */ #define MSM_VERSION_MAJOR 1 -#define MSM_VERSION_MINOR 8 +#define MSM_VERSION_MINOR 9 #define MSM_VERSION_PATCHLEVEL 0 static const struct drm_mode_config_funcs mode_config_funcs = { diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c index 6cfa984dee6a..c6d60c8d286d 100644 --- a/drivers/gpu/drm/msm/msm_gem_submit.c +++ b/drivers/gpu/drm/msm/msm_gem_submit.c @@ -872,16 +872,46 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, submit->nr_cmds = i; + /* + * If using userspace provided seqno fence, validate that the id + * is available before arming sched job. Since access to fence_idr + * is serialized on the queue lock, the slot should be still avail + * after the job is armed + */ + if ((args->flags & MSM_SUBMIT_FENCE_SN_IN) && + idr_find(&queue->fence_idr, args->fence)) { + ret = -EINVAL; + goto out; + } + drm_sched_job_arm(&submit->base); submit->user_fence = dma_fence_get(&submit->base.s_fence->finished); - /* - * Allocate an id which can be used by WAIT_FENCE ioctl to map back - * to the underlying fence. - */ - submit->fence_id = idr_alloc_cyclic(&queue->fence_idr, - submit->user_fence, 1, INT_MAX, GFP_KERNEL); + if (args->flags & MSM_SUBMIT_FENCE_SN_IN) { + /* + * Userspace has assigned the seqno fence that it wants + * us to use. It is an error to pick a fence sequence + * number that is not available. + */ + submit->fence_id = args->fence; + ret = idr_alloc_u32(&queue->fence_idr, submit->user_fence, + &submit->fence_id, submit->fence_id, + GFP_KERNEL); + /* + * We've already validated that the fence_id slot is valid, + * so if idr_alloc_u32 failed, it is a kernel bug + */ + WARN_ON(ret); + } else { + /* + * Allocate an id which can be used by WAIT_FENCE ioctl to map + * back to the underlying fence. + */ + submit->fence_id = idr_alloc_cyclic(&queue->fence_idr, + submit->user_fence, 1, + INT_MAX, GFP_KERNEL); + } if (submit->fence_id < 0) { ret = submit->fence_id = 0; submit->fence_id = 0; diff --git a/include/uapi/drm/msm_drm.h b/include/uapi/drm/msm_drm.h index 2ee03ba08681..07efc8033492 100644 --- a/include/uapi/drm/msm_drm.h +++ b/include/uapi/drm/msm_drm.h @@ -232,6 +232,7 @@ struct drm_msm_gem_submit_bo { #define MSM_SUBMIT_SUDO 0x10000000 /* run submitted cmds from RB */ #define MSM_SUBMIT_SYNCOBJ_IN 0x08000000 /* enable input syncobj */ #define MSM_SUBMIT_SYNCOBJ_OUT 0x04000000 /* enable output syncobj */ +#define MSM_SUBMIT_FENCE_SN_IN 0x02000000 /* userspace passes in seqno fence */ #define MSM_SUBMIT_FLAGS ( \ MSM_SUBMIT_NO_IMPLICIT | \ MSM_SUBMIT_FENCE_FD_IN | \ @@ -239,6 +240,7 @@ struct drm_msm_gem_submit_bo { MSM_SUBMIT_SUDO | \ MSM_SUBMIT_SYNCOBJ_IN | \ MSM_SUBMIT_SYNCOBJ_OUT | \ + MSM_SUBMIT_FENCE_SN_IN | \ 0) #define MSM_SUBMIT_SYNCOBJ_RESET 0x00000001 /* Reset syncobj after wait. */ @@ -258,7 +260,7 @@ struct drm_msm_gem_submit_syncobj { */ struct drm_msm_gem_submit { __u32 flags; /* MSM_PIPE_x | MSM_SUBMIT_x */ - __u32 fence; /* out */ + __u32 fence; /* out (or in with MSM_SUBMIT_FENCE_SN_IN flag) */ __u32 nr_bos; /* in, number of submit_bo's */ __u32 nr_cmds; /* in, number of submit_cmd's */ __u64 bos; /* in, ptr to array of submit_bo's */ -- cgit v1.2.3 From 23b0e695d4ca92e9ebd97689e2e51e497957cb48 Mon Sep 17 00:00:00 2001 From: Zack Rusin Date: Wed, 2 Mar 2022 10:24:23 -0500 Subject: drm/vmwgfx: Allow querying of the SVGA PCI id from the userspace Mesa3D loaders require knowledge of the devices PCI id. SVGAv2 and v3 have different PCI id's, but the same driver is used to handle them both. To allow Mesa3D svga driver to be loaded automatically for both SVGAv2 and SVGAv3 make the kernel return the PCI id of the currently running device. Signed-off-by: Zack Rusin Reviewed-by: Martin Krastev Reviewed-by: Maaz Mombasawala Link: https://patchwork.freedesktop.org/patch/msgid/20220302152426.885214-6-zack@kde.org --- drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c | 27 +++++++++++++++------------ include/uapi/drm/vmwgfx_drm.h | 9 ++++++++- 2 files changed, 23 insertions(+), 13 deletions(-) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c index 471da2b4c177..a1da5678c731 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 OR MIT /************************************************************************** * - * Copyright 2009-2015 VMware, Inc., Palo Alto, CA., USA + * Copyright 2009-2022 VMware, Inc., Palo Alto, CA., USA * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the @@ -27,9 +27,11 @@ #include "vmwgfx_drv.h" #include "vmwgfx_devcaps.h" -#include #include "vmwgfx_kms.h" +#include +#include + int vmw_getparam_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv) { @@ -62,17 +64,15 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data, break; case DRM_VMW_PARAM_FIFO_HW_VERSION: { - if ((dev_priv->capabilities & SVGA_CAP_GBOBJECTS)) { + if ((dev_priv->capabilities & SVGA_CAP_GBOBJECTS)) param->value = SVGA3D_HWVERSION_WS8_B1; - break; - } - - param->value = - vmw_fifo_mem_read(dev_priv, - ((vmw_fifo_caps(dev_priv) & - SVGA_FIFO_CAP_3D_HWVERSION_REVISED) ? - SVGA_FIFO_3D_HWVERSION_REVISED : - SVGA_FIFO_3D_HWVERSION)); + else + param->value = vmw_fifo_mem_read( + dev_priv, + ((vmw_fifo_caps(dev_priv) & + SVGA_FIFO_CAP_3D_HWVERSION_REVISED) ? + SVGA_FIFO_3D_HWVERSION_REVISED : + SVGA_FIFO_3D_HWVERSION)); break; } case DRM_VMW_PARAM_MAX_SURF_MEMORY: @@ -108,6 +108,9 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data, case DRM_VMW_PARAM_GL43: param->value = has_gl43_context(dev_priv); break; + case DRM_VMW_PARAM_DEVICE_ID: + param->value = to_pci_dev(dev_priv->drm.dev)->device; + break; default: return -EINVAL; } diff --git a/include/uapi/drm/vmwgfx_drm.h b/include/uapi/drm/vmwgfx_drm.h index 8277644c1144..26549c86a91f 100644 --- a/include/uapi/drm/vmwgfx_drm.h +++ b/include/uapi/drm/vmwgfx_drm.h @@ -1,6 +1,6 @@ /************************************************************************** * - * Copyright © 2009-2015 VMware, Inc., Palo Alto, CA., USA + * Copyright © 2009-2022 VMware, Inc., Palo Alto, CA., USA * All Rights Reserved. * * Permission is hereby granted, free of charge, to any person obtaining a @@ -92,6 +92,12 @@ extern "C" { * * DRM_VMW_PARAM_SM5 * SM5 support is enabled. + * + * DRM_VMW_PARAM_GL43 + * SM5.1+GL4.3 support is enabled. + * + * DRM_VMW_PARAM_DEVICE_ID + * PCI ID of the underlying SVGA device. */ #define DRM_VMW_PARAM_NUM_STREAMS 0 @@ -111,6 +117,7 @@ extern "C" { #define DRM_VMW_PARAM_SM4_1 14 #define DRM_VMW_PARAM_SM5 15 #define DRM_VMW_PARAM_GL43 16 +#define DRM_VMW_PARAM_DEVICE_ID 17 /** * enum drm_vmw_handle_type - handle type for ref ioctls -- cgit v1.2.3 From 78e1fb3112c0ac440bf9dce080a4aefbaa547d52 Mon Sep 17 00:00:00 2001 From: Rodrigo Vivi Date: Sun, 6 Mar 2022 15:21:57 -0800 Subject: drm/i915/uapi: Add query for hwconfig blob In this interface i915 is returning a blob of data which it receives from the guc software. This blob provides some useful data about the hardware for drivers. The format of this blob will be documented in the Programmer Reference Manuals when released. Cc: Tvrtko Ursulin Cc: Kenneth Graunke Cc: Michal Wajdeczko Cc: Slawomir Milczarek Cc: Joonas Lahtinen Signed-off-by: Rodrigo Vivi Signed-off-by: John Harrison Reviewed-by: Matthew Brost Acked-by: Jordan Justen Tested-by: Jordan Justen Acked-by: Jon Bloomfield Signed-off-by: John Harrison Link: https://patchwork.freedesktop.org/patch/msgid/20220306232157.1174335-3-jordan.l.justen@intel.com --- drivers/gpu/drm/i915/i915_query.c | 23 +++++++++++++++++++++++ include/uapi/drm/i915_drm.h | 1 + 2 files changed, 24 insertions(+) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/i915/i915_query.c b/drivers/gpu/drm/i915/i915_query.c index 2dfbc22857a3..b5ca00cb6cf6 100644 --- a/drivers/gpu/drm/i915/i915_query.c +++ b/drivers/gpu/drm/i915/i915_query.c @@ -479,12 +479,35 @@ static int query_memregion_info(struct drm_i915_private *i915, return total_length; } +static int query_hwconfig_blob(struct drm_i915_private *i915, + struct drm_i915_query_item *query_item) +{ + struct intel_gt *gt = to_gt(i915); + struct intel_hwconfig *hwconfig = >->info.hwconfig; + + if (!hwconfig->size || !hwconfig->ptr) + return -ENODEV; + + if (query_item->length == 0) + return hwconfig->size; + + if (query_item->length < hwconfig->size) + return -EINVAL; + + if (copy_to_user(u64_to_user_ptr(query_item->data_ptr), + hwconfig->ptr, hwconfig->size)) + return -EFAULT; + + return hwconfig->size; +} + static int (* const i915_query_funcs[])(struct drm_i915_private *dev_priv, struct drm_i915_query_item *query_item) = { query_topology_info, query_engine_info, query_perf_config, query_memregion_info, + query_hwconfig_blob, }; int i915_query_ioctl(struct drm_device *dev, void *data, struct drm_file *file) diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index 05c3642aaece..071ffd9d51f1 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -2691,6 +2691,7 @@ struct drm_i915_query_item { #define DRM_I915_QUERY_ENGINE_INFO 2 #define DRM_I915_QUERY_PERF_CONFIG 3 #define DRM_I915_QUERY_MEMORY_REGIONS 4 +#define DRM_I915_QUERY_HWCONFIG_BLOB 5 /* Must be kept compact -- no holes and well documented */ /** -- cgit v1.2.3 From 034d47b25b2ce627f1e1b04242d72db4180f3525 Mon Sep 17 00:00:00 2001 From: Tvrtko Ursulin Date: Fri, 25 Mar 2022 09:49:16 +0000 Subject: drm/i915/uapi: Document DRM_I915_QUERY_HWCONFIG_BLOB UAPI with absolutely no documentation should not have been added - clarify blob format and content will be described externally. Fixes: 78e1fb3112c0 ("drm/i915/uapi: Add query for hwconfig blob") Signed-off-by: Tvrtko Ursulin Co-developed-by: Jordan Justen Cc: Jon Bloomfield Cc: Daniel Vetter Cc: John Harrison Cc: Jon Ewins Cc: Joonas Lahtinen Cc: Rodrigo Vivi Reviewed-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/20220325094916.2186367-1-tvrtko.ursulin@linux.intel.com [tursulin: Fixed spelling s/meading/meaning/.] --- include/uapi/drm/i915_drm.h | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'include/uapi/drm') diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index 071ffd9d51f1..9ab021c4d632 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -2683,6 +2683,9 @@ struct drm_i915_perf_oa_config { * * The behaviour is determined by the @query_id. Note that exactly what * @data_ptr is also depends on the specific @query_id. + * + * For specific queries see: + * * `GuC HWCONFIG blob uAPI`_ */ struct drm_i915_query_item { /** @query_id: The id for this query */ @@ -3135,6 +3138,16 @@ struct drm_i915_query_memory_regions { struct drm_i915_memory_region_info regions[]; }; +/** + * DOC: GuC HWCONFIG blob uAPI + * + * The GuC produces a blob with information about the current device. + * i915 reads this blob from GuC and makes it available via this uAPI. + * + * The format and meaning of the blob content are documented in the + * Programmer's Reference Manual. + */ + /** * struct drm_i915_gem_create_ext - Existing gem_create behaviour, with added * extension support using struct i915_user_extension. -- cgit v1.2.3 From 764b2668cffafdfde47f51f22b0949315bd0b96a Mon Sep 17 00:00:00 2001 From: Matt Roper Date: Mon, 11 Apr 2022 17:34:02 +0300 Subject: drm/fourcc: Introduce format modifiers for DG2 render and media compression The render/media engines on DG2 unify render compression and media compression into a single format for the first time, using the Tile 4 layout for main surfaces. The compression algorithm is different from any previous platform and the display engine must still be configured to decompress either a render or media compressed surface; as such, we need new RC and MC framebuffer modifiers to represent buffers in this format. v2: Clarify modifier layout description. Cc: dri-devel@lists.freedesktop.org Signed-off-by: Matt Roper Signed-off-by: Imre Deak Acked-by: Nanley Chery Reviewed-by: Juha-Pekka Heikkila Acked-by: Maarten Lankhorst Link: https://patchwork.freedesktop.org/patch/msgid/20220411143405.1073845-2-imre.deak@intel.com --- include/uapi/drm/drm_fourcc.h | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) (limited to 'include/uapi/drm') diff --git a/include/uapi/drm/drm_fourcc.h b/include/uapi/drm/drm_fourcc.h index b73fe6797fc3..4a5117715db3 100644 --- a/include/uapi/drm/drm_fourcc.h +++ b/include/uapi/drm/drm_fourcc.h @@ -583,6 +583,28 @@ extern "C" { */ #define I915_FORMAT_MOD_4_TILED fourcc_mod_code(INTEL, 9) +/* + * Intel color control surfaces (CCS) for DG2 render compression. + * + * The main surface is Tile 4 and at plane index 0. The CCS data is stored + * outside of the GEM object in a reserved memory area dedicated for the + * storage of the CCS data for all RC/RC_CC/MC compressible GEM objects. The + * main surface pitch is required to be a multiple of four Tile 4 widths. + */ +#define I915_FORMAT_MOD_4_TILED_DG2_RC_CCS fourcc_mod_code(INTEL, 10) + +/* + * Intel color control surfaces (CCS) for DG2 media compression. + * + * The main surface is Tile 4 and at plane index 0. For semi-planar formats + * like NV12, the Y and UV planes are Tile 4 and are located at plane indices + * 0 and 1, respectively. The CCS for all planes are stored outside of the + * GEM object in a reserved memory area dedicated for the storage of the + * CCS data for all RC/RC_CC/MC compressible GEM objects. The main surface + * pitch is required to be a multiple of four Tile 4 widths. + */ +#define I915_FORMAT_MOD_4_TILED_DG2_MC_CCS fourcc_mod_code(INTEL, 11) + /* * Tiled, NV12MT, grouped in 64 (pixels) x 32 (lines) -sized macroblocks * -- cgit v1.2.3 From 9035039e1ed691cd893777a42e048003a2f349d6 Mon Sep 17 00:00:00 2001 From: Mika Kahola Date: Mon, 11 Apr 2022 17:34:04 +0300 Subject: drm/fourcc: Introduce format modifier for DG2 clear color MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit DG2 clear color render compression uses Tile4 layout. Therefore, we need to define a new format modifier for uAPI to support clear color rendering. v2: Display version is fixed. [Imre] KDoc is enhanced for cc modifier. [Nanley & Lionel] v3: Split out the modifier addition to a separate patch. Clarify the modifier layout description. Cc: dri-devel@lists.freedesktop.org Signed-off-by: Mika Kahola cc: Anshuman Gupta Signed-off-by: Juha-Pekka Heikkilä Signed-off-by: Ramalingam C Signed-off-by: Imre Deak Acked-by: Nanley Chery Reviewed-by: Juha-Pekka Heikkila Acked-by: Maarten Lankhorst Link: https://patchwork.freedesktop.org/patch/msgid/20220411143405.1073845-4-imre.deak@intel.com --- include/uapi/drm/drm_fourcc.h | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'include/uapi/drm') diff --git a/include/uapi/drm/drm_fourcc.h b/include/uapi/drm/drm_fourcc.h index 4a5117715db3..e5074162bcdd 100644 --- a/include/uapi/drm/drm_fourcc.h +++ b/include/uapi/drm/drm_fourcc.h @@ -605,6 +605,20 @@ extern "C" { */ #define I915_FORMAT_MOD_4_TILED_DG2_MC_CCS fourcc_mod_code(INTEL, 11) +/* + * Intel Color Control Surface with Clear Color (CCS) for DG2 render compression. + * + * The main surface is Tile 4 and at plane index 0. The CCS data is stored + * outside of the GEM object in a reserved memory area dedicated for the + * storage of the CCS data for all RC/RC_CC/MC compressible GEM objects. The + * main surface pitch is required to be a multiple of four Tile 4 widths. The + * clear color is stored at plane index 1 and the pitch should be ignored. The + * format of the 256 bits of clear color data matches the one used for the + * I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS_CC modifier, see its description + * for details. + */ +#define I915_FORMAT_MOD_4_TILED_DG2_RC_CCS_CC fourcc_mod_code(INTEL, 12) + /* * Tiled, NV12MT, grouped in 64 (pixels) x 32 (lines) -sized macroblocks * -- cgit v1.2.3 From 462ac1cdf4d7acf100f9b21ad8bf72c88ceb902f Mon Sep 17 00:00:00 2001 From: Matt Roper Date: Thu, 14 Apr 2022 12:22:27 -0700 Subject: drm/i915/doc: Convert drm_i915_query_topology_info comment to kerneldoc This structure has a great comment describing the fields, but it's not currently in kerneldoc form and does not show up in the generated documentation. Let's fix that and also clarify the description of what "subslice" refers to on gen12 platforms and beyond and that "slice" is no longer meaningful on Xe_HP and beyond. Signed-off-by: Matt Roper Reviewed-by: Francisco Jerez Link: https://patchwork.freedesktop.org/patch/msgid/20220414192230.749771-1-matthew.d.roper@intel.com --- include/uapi/drm/i915_drm.h | 110 +++++++++++++++++++++++++++++++------------- 1 file changed, 78 insertions(+), 32 deletions(-) (limited to 'include/uapi/drm') diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index 9ab021c4d632..73e1c6180ddb 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -2775,66 +2775,112 @@ struct drm_i915_query { __u64 items_ptr; }; -/* - * Data written by the kernel with query DRM_I915_QUERY_TOPOLOGY_INFO : - * - * data: contains the 3 pieces of information : - * - * - the slice mask with one bit per slice telling whether a slice is - * available. The availability of slice X can be queried with the following - * formula : - * - * (data[X / 8] >> (X % 8)) & 1 - * - * - the subslice mask for each slice with one bit per subslice telling - * whether a subslice is available. Gen12 has dual-subslices, which are - * similar to two gen11 subslices. For gen12, this array represents dual- - * subslices. The availability of subslice Y in slice X can be queried - * with the following formula : - * - * (data[subslice_offset + - * X * subslice_stride + - * Y / 8] >> (Y % 8)) & 1 - * - * - the EU mask for each subslice in each slice with one bit per EU telling - * whether an EU is available. The availability of EU Z in subslice Y in - * slice X can be queried with the following formula : +/** + * struct drm_i915_query_topology_info * - * (data[eu_offset + - * (X * max_subslices + Y) * eu_stride + - * Z / 8] >> (Z % 8)) & 1 + * Describes slice/subslice/EU information queried by + * %DRM_I915_QUERY_TOPOLOGY_INFO */ struct drm_i915_query_topology_info { - /* + /** + * @flags: + * * Unused for now. Must be cleared to zero. */ __u16 flags; + /** + * @max_slices: + * + * The number of bits used to express the slice mask. + */ __u16 max_slices; + + /** + * @max_subslices: + * + * The number of bits used to express the subslice mask. + */ __u16 max_subslices; + + /** + * @max_eus_per_subslice: + * + * The number of bits in the EU mask that correspond to a single + * subslice's EUs. + */ __u16 max_eus_per_subslice; - /* + /** + * @subslice_offset: + * * Offset in data[] at which the subslice masks are stored. */ __u16 subslice_offset; - /* + /** + * @subslice_stride: + * * Stride at which each of the subslice masks for each slice are * stored. */ __u16 subslice_stride; - /* + /** + * @eu_offset: + * * Offset in data[] at which the EU masks are stored. */ __u16 eu_offset; - /* + /** + * @eu_stride: + * * Stride at which each of the EU masks for each subslice are stored. */ __u16 eu_stride; + /** + * @data: + * + * Contains 3 pieces of information : + * + * - The slice mask with one bit per slice telling whether a slice is + * available. The availability of slice X can be queried with the + * following formula : + * + * .. code:: c + * + * (data[X / 8] >> (X % 8)) & 1 + * + * Starting with Xe_HP platforms, Intel hardware no longer has + * traditional slices so i915 will always report a single slice + * (hardcoded slicemask = 0x1) which contains all of the platform's + * subslices. I.e., the mask here does not reflect any of the newer + * hardware concepts such as "gslices" or "cslices" since userspace + * is capable of inferring those from the subslice mask. + * + * - The subslice mask for each slice with one bit per subslice telling + * whether a subslice is available. Starting with Gen12 we use the + * term "subslice" to refer to what the hardware documentation + * describes as a "dual-subslices." The availability of subslice Y + * in slice X can be queried with the following formula : + * + * .. code:: c + * + * (data[subslice_offset + X * subslice_stride + Y / 8] >> (Y % 8)) & 1 + * + * - The EU mask for each subslice in each slice, with one bit per EU + * telling whether an EU is available. The availability of EU Z in + * subslice Y in slice X can be queried with the following formula : + * + * .. code:: c + * + * (data[eu_offset + + * (X * max_subslices + Y) * eu_stride + + * Z / 8 + * ] >> (Z % 8)) & 1 + */ __u8 data[]; }; -- cgit v1.2.3 From a2e5402691e2326991704b1fb73646fa2efde1c3 Mon Sep 17 00:00:00 2001 From: Matt Roper Date: Thu, 14 Apr 2022 12:22:28 -0700 Subject: drm/i915/doc: Convert perf UAPI comments to kerneldoc Convert the comments for drm_i915_query_perf_config and drm_i915_perf_oa_config to kerneldoc so that they will show up in the generated documentation. Also correct a couple places that referred to query_id when they actually meant to refer to query_item.flags. Signed-off-by: Matt Roper Reviewed-by: Francisco Jerez Link: https://patchwork.freedesktop.org/patch/msgid/20220414192230.749771-2-matthew.d.roper@intel.com --- include/uapi/drm/i915_drm.h | 116 ++++++++++++++++++++++++++++++++------------ 1 file changed, 86 insertions(+), 30 deletions(-) (limited to 'include/uapi/drm') diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index 73e1c6180ddb..787dab98f7b0 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -2657,24 +2657,65 @@ enum drm_i915_perf_record_type { DRM_I915_PERF_RECORD_MAX /* non-ABI */ }; -/* +/** + * struct drm_i915_perf_oa_config + * * Structure to upload perf dynamic configuration into the kernel. */ struct drm_i915_perf_oa_config { - /** String formatted like "%08x-%04x-%04x-%04x-%012x" */ + /** + * @uuid: + * + * String formatted like "%\08x-%\04x-%\04x-%\04x-%\012x" + */ char uuid[36]; + /** + * @n_mux_regs: + * + * Number of mux regs in &mux_regs_ptr. + */ __u32 n_mux_regs; + + /** + * @n_boolean_regs: + * + * Number of boolean regs in &boolean_regs_ptr. + */ __u32 n_boolean_regs; + + /** + * @n_flex_regs: + * + * Number of flex regs in &flex_regs_ptr. + */ __u32 n_flex_regs; - /* - * These fields are pointers to tuples of u32 values (register address, - * value). For example the expected length of the buffer pointed by - * mux_regs_ptr is (2 * sizeof(u32) * n_mux_regs). + /** + * @mux_regs_ptr: + * + * Pointer to tuples of u32 values (register address, value) for mux + * registers. Expected length of buffer is (2 * sizeof(u32) * + * &n_mux_regs). */ __u64 mux_regs_ptr; + + /** + * @boolean_regs_ptr: + * + * Pointer to tuples of u32 values (register address, value) for mux + * registers. Expected length of buffer is (2 * sizeof(u32) * + * &n_boolean_regs). + */ __u64 boolean_regs_ptr; + + /** + * @flex_regs_ptr: + * + * Pointer to tuples of u32 values (register address, value) for mux + * registers. Expected length of buffer is (2 * sizeof(u32) * + * &n_flex_regs). + */ __u64 flex_regs_ptr; }; @@ -3001,52 +3042,67 @@ struct drm_i915_query_engine_info { struct drm_i915_engine_info engines[]; }; -/* - * Data written by the kernel with query DRM_I915_QUERY_PERF_CONFIG. +/** + * struct drm_i915_query_perf_config + * + * Data written by the kernel with query %DRM_I915_QUERY_PERF_CONFIG. */ struct drm_i915_query_perf_config { union { - /* - * When query_item.flags == DRM_I915_QUERY_PERF_CONFIG_LIST, i915 sets - * this fields to the number of configurations available. + /** + * @n_configs: + * + * When &drm_i915_query_item.flags == + * %DRM_I915_QUERY_PERF_CONFIG_LIST, i915 sets this fields to + * the number of configurations available. */ __u64 n_configs; - /* - * When query_id == DRM_I915_QUERY_PERF_CONFIG_DATA_FOR_ID, - * i915 will use the value in this field as configuration - * identifier to decide what data to write into config_ptr. + /** + * @config: + * + * When &drm_i915_query_item.flags == + * %DRM_I915_QUERY_PERF_CONFIG_DATA_FOR_ID, i915 will use the + * value in this field as configuration identifier to decide + * what data to write into config_ptr. */ __u64 config; - /* - * When query_id == DRM_I915_QUERY_PERF_CONFIG_DATA_FOR_UUID, - * i915 will use the value in this field as configuration - * identifier to decide what data to write into config_ptr. + /** + * @uuid: + * + * When &drm_i915_query_item.flags == + * %DRM_I915_QUERY_PERF_CONFIG_DATA_FOR_UUID, i915 will use the + * value in this field as configuration identifier to decide + * what data to write into config_ptr. * * String formatted like "%08x-%04x-%04x-%04x-%012x" */ char uuid[36]; }; - /* + /** + * @flags: + * * Unused for now. Must be cleared to zero. */ __u32 flags; - /* - * When query_item.flags == DRM_I915_QUERY_PERF_CONFIG_LIST, i915 will - * write an array of __u64 of configuration identifiers. + /** + * @data: + * + * When &drm_i915_query_item.flags == %DRM_I915_QUERY_PERF_CONFIG_LIST, + * i915 will write an array of __u64 of configuration identifiers. * - * When query_item.flags == DRM_I915_QUERY_PERF_CONFIG_DATA, i915 will - * write a struct drm_i915_perf_oa_config. If the following fields of - * drm_i915_perf_oa_config are set not set to 0, i915 will write into - * the associated pointers the values of submitted when the + * When &drm_i915_query_item.flags == %DRM_I915_QUERY_PERF_CONFIG_DATA, + * i915 will write a struct drm_i915_perf_oa_config. If the following + * fields of struct drm_i915_perf_oa_config are not set to 0, i915 will + * write into the associated pointers the values of submitted when the * configuration was created : * - * - n_mux_regs - * - n_boolean_regs - * - n_flex_regs + * - &drm_i915_perf_oa_config.n_mux_regs + * - &drm_i915_perf_oa_config.n_boolean_regs + * - &drm_i915_perf_oa_config.n_flex_regs */ __u8 data[]; }; -- cgit v1.2.3 From 1c671ad753dbbf5f5da7b18175d05acc3cccf1a8 Mon Sep 17 00:00:00 2001 From: Matt Roper Date: Thu, 14 Apr 2022 12:22:29 -0700 Subject: drm/i915/doc: Link query items to their uapi structs Document the possible options for drm_i915_query_item.query_id with links to the corresponding uapi structures. Signed-off-by: Matt Roper Reviewed-by: Francisco Jerez Link: https://patchwork.freedesktop.org/patch/msgid/20220414192230.749771-3-matthew.d.roper@intel.com --- include/uapi/drm/i915_drm.h | 34 ++++++++++++++++++++-------------- 1 file changed, 20 insertions(+), 14 deletions(-) (limited to 'include/uapi/drm') diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index 787dab98f7b0..097a7935a510 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -2724,18 +2724,24 @@ struct drm_i915_perf_oa_config { * * The behaviour is determined by the @query_id. Note that exactly what * @data_ptr is also depends on the specific @query_id. - * - * For specific queries see: - * * `GuC HWCONFIG blob uAPI`_ */ struct drm_i915_query_item { - /** @query_id: The id for this query */ + /** + * @query_id: + * + * The id for this query. Currently accepted query IDs are: + * - %DRM_I915_QUERY_TOPOLOGY_INFO (see struct drm_i915_query_topology_info) + * - %DRM_I915_QUERY_ENGINE_INFO (see struct drm_i915_engine_info) + * - %DRM_I915_QUERY_PERF_CONFIG (see struct drm_i915_query_perf_config) + * - %DRM_I915_QUERY_MEMORY_REGIONS (see struct drm_i915_query_memory_regions) + * - %DRM_I915_QUERY_HWCONFIG_BLOB (see `GuC HWCONFIG blob uAPI`) + */ __u64 query_id; -#define DRM_I915_QUERY_TOPOLOGY_INFO 1 -#define DRM_I915_QUERY_ENGINE_INFO 2 -#define DRM_I915_QUERY_PERF_CONFIG 3 -#define DRM_I915_QUERY_MEMORY_REGIONS 4 -#define DRM_I915_QUERY_HWCONFIG_BLOB 5 +#define DRM_I915_QUERY_TOPOLOGY_INFO 1 +#define DRM_I915_QUERY_ENGINE_INFO 2 +#define DRM_I915_QUERY_PERF_CONFIG 3 +#define DRM_I915_QUERY_MEMORY_REGIONS 4 +#define DRM_I915_QUERY_HWCONFIG_BLOB 5 /* Must be kept compact -- no holes and well documented */ /** @@ -2751,14 +2757,14 @@ struct drm_i915_query_item { /** * @flags: * - * When query_id == DRM_I915_QUERY_TOPOLOGY_INFO, must be 0. + * When &query_id == %DRM_I915_QUERY_TOPOLOGY_INFO, must be 0. * - * When query_id == DRM_I915_QUERY_PERF_CONFIG, must be one of the + * When &query_id == %DRM_I915_QUERY_PERF_CONFIG, must be one of the * following: * - * - DRM_I915_QUERY_PERF_CONFIG_LIST - * - DRM_I915_QUERY_PERF_CONFIG_DATA_FOR_UUID - * - DRM_I915_QUERY_PERF_CONFIG_FOR_UUID + * - %DRM_I915_QUERY_PERF_CONFIG_LIST + * - %DRM_I915_QUERY_PERF_CONFIG_DATA_FOR_UUID + * - %DRM_I915_QUERY_PERF_CONFIG_FOR_UUID */ __u32 flags; #define DRM_I915_QUERY_PERF_CONFIG_LIST 1 -- cgit v1.2.3 From c94fde8f516610b0961b65a59bc8885e6df23cb9 Mon Sep 17 00:00:00 2001 From: Matt Atwood Date: Thu, 14 Apr 2022 12:22:30 -0700 Subject: drm/i915/uapi: Add DRM_I915_QUERY_GEOMETRY_SUBSLICES Newer platforms have DSS that aren't necessarily available for both geometry and compute, two queries will need to exist. This introduces the first, when passing a valid engine class and engine instance in the flags returns a topology describing geometry. Based on past discussion, we currently only support this new query item on Xe_HP and beyond; earlier platforms do not need to worry about geometry and compute pipelines having access to different topology and should continue to use the existing topology query. v2: fix white space errors v3: change flags from hosting 2 8 bit numbers to holding a i915_engine_class_instance struct v4: add error if non rcs engine passed. v5 (by MattR): - Improve kerneldoc and cross references to related structs/enums. (Daniel) - Clarify that geometry query is only supported on render engines (Francisco) - Clarify that the new query is only supported on Xe_HP+. - Fix checkpatch warnings. Cc: Ashutosh Dixit Cc: Matt Roper Cc: Joonas Lahtinen Cc: Francisco Jerez UMD (mesa): https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14143 Testcase: igt@i915_query@test-query-geometry-subslices Signed-off-by: Matt Atwood Signed-off-by: Matt Roper Reviewed-by: Francisco Jerez Link: https://patchwork.freedesktop.org/patch/msgid/20220414192230.749771-4-matthew.d.roper@intel.com --- drivers/gpu/drm/i915/i915_query.c | 71 ++++++++++++++++++++++++++++----------- include/uapi/drm/i915_drm.h | 27 ++++++++++++--- 2 files changed, 75 insertions(+), 23 deletions(-) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/i915/i915_query.c b/drivers/gpu/drm/i915/i915_query.c index b5ca00cb6cf6..7584cec53d5d 100644 --- a/drivers/gpu/drm/i915/i915_query.c +++ b/drivers/gpu/drm/i915/i915_query.c @@ -9,6 +9,7 @@ #include "i915_drv.h" #include "i915_perf.h" #include "i915_query.h" +#include "gt/intel_engine_user.h" #include static int copy_query_item(void *query_hdr, size_t query_sz, @@ -28,36 +29,30 @@ static int copy_query_item(void *query_hdr, size_t query_sz, return 0; } -static int query_topology_info(struct drm_i915_private *dev_priv, - struct drm_i915_query_item *query_item) +static int fill_topology_info(const struct sseu_dev_info *sseu, + struct drm_i915_query_item *query_item, + const u8 *subslice_mask) { - const struct sseu_dev_info *sseu = &to_gt(dev_priv)->info.sseu; struct drm_i915_query_topology_info topo; u32 slice_length, subslice_length, eu_length, total_length; int ret; - if (query_item->flags != 0) - return -EINVAL; + BUILD_BUG_ON(sizeof(u8) != sizeof(sseu->slice_mask)); if (sseu->max_slices == 0) return -ENODEV; - BUILD_BUG_ON(sizeof(u8) != sizeof(sseu->slice_mask)); - slice_length = sizeof(sseu->slice_mask); subslice_length = sseu->max_slices * sseu->ss_stride; eu_length = sseu->max_slices * sseu->max_subslices * sseu->eu_stride; total_length = sizeof(topo) + slice_length + subslice_length + eu_length; - ret = copy_query_item(&topo, sizeof(topo), total_length, - query_item); + ret = copy_query_item(&topo, sizeof(topo), total_length, query_item); + if (ret != 0) return ret; - if (topo.flags != 0) - return -EINVAL; - memset(&topo, 0, sizeof(topo)); topo.max_slices = sseu->max_slices; topo.max_subslices = sseu->max_subslices; @@ -69,27 +64,64 @@ static int query_topology_info(struct drm_i915_private *dev_priv, topo.eu_stride = sseu->eu_stride; if (copy_to_user(u64_to_user_ptr(query_item->data_ptr), - &topo, sizeof(topo))) + &topo, sizeof(topo))) return -EFAULT; if (copy_to_user(u64_to_user_ptr(query_item->data_ptr + sizeof(topo)), - &sseu->slice_mask, slice_length)) + &sseu->slice_mask, slice_length)) return -EFAULT; if (copy_to_user(u64_to_user_ptr(query_item->data_ptr + - sizeof(topo) + slice_length), - sseu->subslice_mask, subslice_length)) + sizeof(topo) + slice_length), + subslice_mask, subslice_length)) return -EFAULT; if (copy_to_user(u64_to_user_ptr(query_item->data_ptr + - sizeof(topo) + - slice_length + subslice_length), - sseu->eu_mask, eu_length)) + sizeof(topo) + + slice_length + subslice_length), + sseu->eu_mask, eu_length)) return -EFAULT; return total_length; } +static int query_topology_info(struct drm_i915_private *dev_priv, + struct drm_i915_query_item *query_item) +{ + const struct sseu_dev_info *sseu = &to_gt(dev_priv)->info.sseu; + + if (query_item->flags != 0) + return -EINVAL; + + return fill_topology_info(sseu, query_item, sseu->subslice_mask); +} + +static int query_geometry_subslices(struct drm_i915_private *i915, + struct drm_i915_query_item *query_item) +{ + const struct sseu_dev_info *sseu; + struct intel_engine_cs *engine; + struct i915_engine_class_instance classinstance; + + if (GRAPHICS_VER_FULL(i915) < IP_VER(12, 50)) + return -ENODEV; + + classinstance = *((struct i915_engine_class_instance *)&query_item->flags); + + engine = intel_engine_lookup_user(i915, (u8)classinstance.engine_class, + (u8)classinstance.engine_instance); + + if (!engine) + return -EINVAL; + + if (engine->class != RENDER_CLASS) + return -EINVAL; + + sseu = &engine->gt->info.sseu; + + return fill_topology_info(sseu, query_item, sseu->geometry_subslice_mask); +} + static int query_engine_info(struct drm_i915_private *i915, struct drm_i915_query_item *query_item) @@ -508,6 +540,7 @@ static int (* const i915_query_funcs[])(struct drm_i915_private *dev_priv, query_perf_config, query_memregion_info, query_hwconfig_blob, + query_geometry_subslices, }; int i915_query_ioctl(struct drm_device *dev, void *data, struct drm_file *file) diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index 097a7935a510..35ca528803fd 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -172,7 +172,9 @@ enum drm_i915_gem_engine_class { I915_ENGINE_CLASS_INVALID = -1 }; -/* +/** + * struct i915_engine_class_instance - Engine class/instance identifier + * * There may be more than one engine fulfilling any role within the system. * Each engine of a class is given a unique instance number and therefore * any engine can be specified by its class:instance tuplet. APIs that allow @@ -180,10 +182,21 @@ enum drm_i915_gem_engine_class { * for this identification. */ struct i915_engine_class_instance { - __u16 engine_class; /* see enum drm_i915_gem_engine_class */ - __u16 engine_instance; + /** + * @engine_class: + * + * Engine class from enum drm_i915_gem_engine_class + */ + __u16 engine_class; #define I915_ENGINE_CLASS_INVALID_NONE -1 #define I915_ENGINE_CLASS_INVALID_VIRTUAL -2 + + /** + * @engine_instance: + * + * Engine instance. + */ + __u16 engine_instance; }; /** @@ -2735,6 +2748,7 @@ struct drm_i915_query_item { * - %DRM_I915_QUERY_PERF_CONFIG (see struct drm_i915_query_perf_config) * - %DRM_I915_QUERY_MEMORY_REGIONS (see struct drm_i915_query_memory_regions) * - %DRM_I915_QUERY_HWCONFIG_BLOB (see `GuC HWCONFIG blob uAPI`) + * - %DRM_I915_QUERY_GEOMETRY_SUBSLICES (see struct drm_i915_query_topology_info) */ __u64 query_id; #define DRM_I915_QUERY_TOPOLOGY_INFO 1 @@ -2742,6 +2756,7 @@ struct drm_i915_query_item { #define DRM_I915_QUERY_PERF_CONFIG 3 #define DRM_I915_QUERY_MEMORY_REGIONS 4 #define DRM_I915_QUERY_HWCONFIG_BLOB 5 +#define DRM_I915_QUERY_GEOMETRY_SUBSLICES 6 /* Must be kept compact -- no holes and well documented */ /** @@ -2765,6 +2780,9 @@ struct drm_i915_query_item { * - %DRM_I915_QUERY_PERF_CONFIG_LIST * - %DRM_I915_QUERY_PERF_CONFIG_DATA_FOR_UUID * - %DRM_I915_QUERY_PERF_CONFIG_FOR_UUID + * + * When &query_id == %DRM_I915_QUERY_GEOMETRY_SUBSLICES must contain + * a struct i915_engine_class_instance that references a render engine. */ __u32 flags; #define DRM_I915_QUERY_PERF_CONFIG_LIST 1 @@ -3051,7 +3069,8 @@ struct drm_i915_query_engine_info { /** * struct drm_i915_query_perf_config * - * Data written by the kernel with query %DRM_I915_QUERY_PERF_CONFIG. + * Data written by the kernel with query %DRM_I915_QUERY_PERF_CONFIG and + * %DRM_I915_QUERY_GEOMETRY_SUBSLICES. */ struct drm_i915_query_perf_config { union { -- cgit v1.2.3 From 651877176c0846616add00f50b719ea6d59dff8d Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Sat, 4 Sep 2021 10:06:03 -0700 Subject: drm/fourcc: Add QCOM tiled modifiers These are mainly used internally in mesa, although I believe the display should be able to scan out the TILED3 format. Currently we define this modifier internally in mesa for use with modifier based allocation. But we can get rid of that hack if we define the modfiers properly. Signed-off-by: Rob Clark Link: https://lore.kernel.org/r/20210904170603.1739137-1-robdclark@gmail.com Signed-off-by: Rob Clark --- include/uapi/drm/drm_fourcc.h | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) (limited to 'include/uapi/drm') diff --git a/include/uapi/drm/drm_fourcc.h b/include/uapi/drm/drm_fourcc.h index fc0c1454d275..4bdf8dded9e7 100644 --- a/include/uapi/drm/drm_fourcc.h +++ b/include/uapi/drm/drm_fourcc.h @@ -609,6 +609,28 @@ extern "C" { */ #define DRM_FORMAT_MOD_QCOM_COMPRESSED fourcc_mod_code(QCOM, 1) +/* + * Qualcomm Tiled Format + * + * Similar to DRM_FORMAT_MOD_QCOM_COMPRESSED but not compressed. + * Implementation may be platform and base-format specific. + * + * Each macrotile consists of m x n (mostly 4 x 4) tiles. + * Pixel data pitch/stride is aligned with macrotile width. + * Pixel data height is aligned with macrotile height. + * Entire pixel data buffer is aligned with 4k(bytes). + */ +#define DRM_FORMAT_MOD_QCOM_TILED3 fourcc_mod_code(QCOM, 3) + +/* + * Qualcomm Alternate Tiled Format + * + * Alternate tiled format typically only used within GMEM. + * Implementation may be platform and base-format specific. + */ +#define DRM_FORMAT_MOD_QCOM_TILED2 fourcc_mod_code(QCOM, 2) + + /* Vivante framebuffer modifiers */ /* -- cgit v1.2.3 From 4bfba71640f8578b3daadb551d6bce12a2d436d8 Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Thu, 17 Mar 2022 09:51:38 -0700 Subject: drm/msm: Add support for pointer params The 64b value field is already suffient to hold a pointer instead of immediate, but we also need a length field. Signed-off-by: Rob Clark Link: https://lore.kernel.org/r/20220317165144.222101-2-robdclark@gmail.com Signed-off-by: Rob Clark --- drivers/gpu/drm/msm/adreno/adreno_gpu.c | 12 ++++++++++-- drivers/gpu/drm/msm/adreno/adreno_gpu.h | 4 ++-- drivers/gpu/drm/msm/msm_drv.c | 8 ++++---- drivers/gpu/drm/msm/msm_gpu.h | 4 ++-- drivers/gpu/drm/msm/msm_rd.c | 5 +++-- include/uapi/drm/msm_drm.h | 2 ++ 6 files changed, 23 insertions(+), 12 deletions(-) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index 9efc84929be0..3d307b34854d 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -229,10 +229,14 @@ adreno_iommu_create_address_space(struct msm_gpu *gpu, } int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx, - uint32_t param, uint64_t *value) + uint32_t param, uint64_t *value, uint32_t *len) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + /* No pointer params yet */ + if (*len != 0) + return -EINVAL; + switch (param) { case MSM_PARAM_GPU_ID: *value = adreno_gpu->info->revn; @@ -284,8 +288,12 @@ int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx, } int adreno_set_param(struct msm_gpu *gpu, struct msm_file_private *ctx, - uint32_t param, uint64_t value) + uint32_t param, uint64_t value, uint32_t len) { + /* No pointer params yet */ + if (len != 0) + return -EINVAL; + switch (param) { case MSM_PARAM_SYSPROF: if (!capable(CAP_SYS_ADMIN)) diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h index 0490c5fbb780..ab3b5ef80332 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h @@ -281,9 +281,9 @@ static inline int adreno_is_a650_family(struct adreno_gpu *gpu) } int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx, - uint32_t param, uint64_t *value); + uint32_t param, uint64_t *value, uint32_t *len); int adreno_set_param(struct msm_gpu *gpu, struct msm_file_private *ctx, - uint32_t param, uint64_t value); + uint32_t param, uint64_t value, uint32_t len); const struct firmware *adreno_request_fw(struct adreno_gpu *adreno_gpu, const char *fwname); struct drm_gem_object *adreno_fw_create_bo(struct msm_gpu *gpu, diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index affa95eb05fc..81b40c67e399 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -613,7 +613,7 @@ static int msm_ioctl_get_param(struct drm_device *dev, void *data, /* for now, we just have 3d pipe.. eventually this would need to * be more clever to dispatch to appropriate gpu module: */ - if (args->pipe != MSM_PIPE_3D0) + if ((args->pipe != MSM_PIPE_3D0) || (args->pad != 0)) return -EINVAL; gpu = priv->gpu; @@ -622,7 +622,7 @@ static int msm_ioctl_get_param(struct drm_device *dev, void *data, return -ENXIO; return gpu->funcs->get_param(gpu, file->driver_priv, - args->param, &args->value); + args->param, &args->value, &args->len); } static int msm_ioctl_set_param(struct drm_device *dev, void *data, @@ -632,7 +632,7 @@ static int msm_ioctl_set_param(struct drm_device *dev, void *data, struct drm_msm_param *args = data; struct msm_gpu *gpu; - if (args->pipe != MSM_PIPE_3D0) + if ((args->pipe != MSM_PIPE_3D0) || (args->pad != 0)) return -EINVAL; gpu = priv->gpu; @@ -641,7 +641,7 @@ static int msm_ioctl_set_param(struct drm_device *dev, void *data, return -ENXIO; return gpu->funcs->set_param(gpu, file->driver_priv, - args->param, args->value); + args->param, args->value, args->len); } static int msm_ioctl_gem_new(struct drm_device *dev, void *data, diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h index cb5340c8af8c..ab5d31f5df81 100644 --- a/drivers/gpu/drm/msm/msm_gpu.h +++ b/drivers/gpu/drm/msm/msm_gpu.h @@ -44,9 +44,9 @@ struct msm_gpu_config { */ struct msm_gpu_funcs { int (*get_param)(struct msm_gpu *gpu, struct msm_file_private *ctx, - uint32_t param, uint64_t *value); + uint32_t param, uint64_t *value, uint32_t *len); int (*set_param)(struct msm_gpu *gpu, struct msm_file_private *ctx, - uint32_t param, uint64_t value); + uint32_t param, uint64_t value, uint32_t len); int (*hw_init)(struct msm_gpu *gpu); int (*pm_suspend)(struct msm_gpu *gpu); int (*pm_resume)(struct msm_gpu *gpu); diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c index 9d835331f214..a92ffde53f0b 100644 --- a/drivers/gpu/drm/msm/msm_rd.c +++ b/drivers/gpu/drm/msm/msm_rd.c @@ -180,6 +180,7 @@ static int rd_open(struct inode *inode, struct file *file) struct msm_gpu *gpu = priv->gpu; uint64_t val; uint32_t gpu_id; + uint32_t zero = 0; int ret = 0; if (!gpu) @@ -200,12 +201,12 @@ static int rd_open(struct inode *inode, struct file *file) * * Note: These particular params do not require a context */ - gpu->funcs->get_param(gpu, NULL, MSM_PARAM_GPU_ID, &val); + gpu->funcs->get_param(gpu, NULL, MSM_PARAM_GPU_ID, &val, &zero); gpu_id = val; rd_write_section(rd, RD_GPU_ID, &gpu_id, sizeof(gpu_id)); - gpu->funcs->get_param(gpu, NULL, MSM_PARAM_CHIP_ID, &val); + gpu->funcs->get_param(gpu, NULL, MSM_PARAM_CHIP_ID, &val, &zero); rd_write_section(rd, RD_CHIP_ID, &val, sizeof(val)); out: diff --git a/include/uapi/drm/msm_drm.h b/include/uapi/drm/msm_drm.h index 07efc8033492..0aa1a8cb4e0d 100644 --- a/include/uapi/drm/msm_drm.h +++ b/include/uapi/drm/msm_drm.h @@ -95,6 +95,8 @@ struct drm_msm_param { __u32 pipe; /* in, MSM_PIPE_x */ __u32 param; /* in, MSM_PARAM_x */ __u64 value; /* out (get_param) or in (set_param) */ + __u32 len; /* zero for non-pointer params */ + __u32 pad; /* must be zero */ }; /* -- cgit v1.2.3 From d4726d7700688835f4784d3b94de6fff2cbe16c2 Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Thu, 17 Mar 2022 09:51:40 -0700 Subject: drm/msm: Add a way to override processes comm/cmdline In the cause of using the GPU via virtgpu, the host side process is really a sort of proxy, and not terribly interesting from the PoV of crash/fault logging. Add a way to override these per process so that we can see the guest process's name. v2: Handle kmalloc failure, add comment to explain kstrdup returns NULL if passed NULL [Dan Carpenter] Signed-off-by: Rob Clark Link: https://lore.kernel.org/r/20220317165144.222101-4-robdclark@gmail.com Signed-off-by: Rob Clark --- drivers/gpu/drm/msm/adreno/adreno_gpu.c | 43 ++++++++++++++++++++++++++++++--- drivers/gpu/drm/msm/msm_gpu.c | 12 +++++++-- drivers/gpu/drm/msm/msm_gpu.h | 6 +++++ drivers/gpu/drm/msm/msm_submitqueue.c | 2 ++ include/uapi/drm/msm_drm.h | 2 ++ 5 files changed, 60 insertions(+), 5 deletions(-) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index 3d307b34854d..45f2c6084aa7 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -290,11 +290,48 @@ int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx, int adreno_set_param(struct msm_gpu *gpu, struct msm_file_private *ctx, uint32_t param, uint64_t value, uint32_t len) { - /* No pointer params yet */ - if (len != 0) - return -EINVAL; + switch (param) { + case MSM_PARAM_COMM: + case MSM_PARAM_CMDLINE: + /* kstrdup_quotable_cmdline() limits to PAGE_SIZE, so + * that should be a reasonable upper bound + */ + if (len > PAGE_SIZE) + return -EINVAL; + break; + default: + if (len != 0) + return -EINVAL; + } switch (param) { + case MSM_PARAM_COMM: + case MSM_PARAM_CMDLINE: { + char *str, **paramp; + + str = kmalloc(len + 1, GFP_KERNEL); + if (!str) + return -ENOMEM; + + if (copy_from_user(str, u64_to_user_ptr(value), len)) { + kfree(str); + return -EFAULT; + } + + /* Ensure string is null terminated: */ + str[len] = '\0'; + + if (param == MSM_PARAM_COMM) { + paramp = &ctx->comm; + } else { + paramp = &ctx->cmdline; + } + + kfree(*paramp); + *paramp = str; + + return 0; + } case MSM_PARAM_SYSPROF: if (!capable(CAP_SYS_ADMIN)) return -EPERM; diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index 7a88dd6568c0..f292b70c1e6d 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -353,14 +353,22 @@ static void retire_submits(struct msm_gpu *gpu); static void get_comm_cmdline(struct msm_gem_submit *submit, char **comm, char **cmd) { + struct msm_file_private *ctx = submit->queue->ctx; struct task_struct *task; + /* Note that kstrdup will return NULL if argument is NULL: */ + *comm = kstrdup(ctx->comm, GFP_KERNEL); + *cmd = kstrdup(ctx->cmdline, GFP_KERNEL); + task = get_pid_task(submit->pid, PIDTYPE_PID); if (!task) return; - *comm = kstrdup(task->comm, GFP_KERNEL); - *cmd = kstrdup_quotable_cmdline(task, GFP_KERNEL); + if (!*comm) + *comm = kstrdup(task->comm, GFP_KERNEL); + + if (!*cmd) + *cmd = kstrdup_quotable_cmdline(task, GFP_KERNEL); put_task_struct(task); } diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h index ab5d31f5df81..321e41de6d42 100644 --- a/drivers/gpu/drm/msm/msm_gpu.h +++ b/drivers/gpu/drm/msm/msm_gpu.h @@ -355,6 +355,12 @@ struct msm_file_private { */ int sysprof; + /** comm: Overridden task comm, see MSM_PARAM_COMM */ + char *comm; + + /** cmdline: Overridden task cmdline, see MSM_PARAM_CMDLINE */ + char *cmdline; + /** * entities: * diff --git a/drivers/gpu/drm/msm/msm_submitqueue.c b/drivers/gpu/drm/msm/msm_submitqueue.c index 79b6ccd6ce64..f486a3cd4e55 100644 --- a/drivers/gpu/drm/msm/msm_submitqueue.c +++ b/drivers/gpu/drm/msm/msm_submitqueue.c @@ -61,6 +61,8 @@ void __msm_file_private_destroy(struct kref *kref) } msm_gem_address_space_put(ctx->aspace); + kfree(ctx->comm); + kfree(ctx->cmdline); kfree(ctx); } diff --git a/include/uapi/drm/msm_drm.h b/include/uapi/drm/msm_drm.h index 0aa1a8cb4e0d..794ad1948497 100644 --- a/include/uapi/drm/msm_drm.h +++ b/include/uapi/drm/msm_drm.h @@ -82,6 +82,8 @@ struct drm_msm_timespec { #define MSM_PARAM_FAULTS 0x09 /* RO */ #define MSM_PARAM_SUSPENDS 0x0a /* RO */ #define MSM_PARAM_SYSPROF 0x0b /* WO: 1 preserves perfcntrs, 2 also disables suspend */ +#define MSM_PARAM_COMM 0x0c /* WO: override for task->comm */ +#define MSM_PARAM_CMDLINE 0x0d /* WO: override for task cmdline */ /* For backwards compat. The original support for preemption was based on * a single ring per priority level so # of priority levels equals the # -- cgit v1.2.3 From a636a0ff11483a4c158acc31ee7ca083d98a0d13 Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Mon, 11 Apr 2022 14:58:39 -0700 Subject: drm/msm: Add a way for userspace to allocate GPU iova The motivation at this point is mainly native userspace mesa driver in a VM guest. The one remaining synchronous "hotpath" is buffer allocation, because guest needs to wait to know the bo's iova before it can start emitting cmdstream/state that references the new bo. By allocating the iova in the guest userspace, we no longer need to wait for a response from the host, but can just rely on the allocation request being processed before the cmdstream submission. Allocation failures (OoM, etc) would just be treated as context-lost (ie. GL_GUILTY_CONTEXT_RESET) or subsequent allocations (or readpix, etc) can raise GL_OUT_OF_MEMORY. v2: Fix inuse check v3: Change mismatched iova case to -EBUSY Signed-off-by: Rob Clark Reviewed-by: Dmitry Baryshkov Reviewed-by: Dmitry Osipenko Link: https://lore.kernel.org/r/20220411215849.297838-11-robdclark@gmail.com Signed-off-by: Rob Clark --- drivers/gpu/drm/msm/adreno/adreno_gpu.c | 10 +++++++ drivers/gpu/drm/msm/msm_drv.c | 21 +++++++++++++++ drivers/gpu/drm/msm/msm_gem.c | 48 +++++++++++++++++++++++++++++++++ drivers/gpu/drm/msm/msm_gem.h | 8 ++++++ drivers/gpu/drm/msm/msm_gem_vma.c | 2 ++ include/uapi/drm/msm_drm.h | 3 +++ 6 files changed, 92 insertions(+) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index 6385ab06632f..4caae0229518 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -281,6 +281,16 @@ int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx, case MSM_PARAM_SUSPENDS: *value = gpu->suspend_count; return 0; + case MSM_PARAM_VA_START: + if (ctx->aspace == gpu->aspace) + return -EINVAL; + *value = ctx->aspace->va_start; + return 0; + case MSM_PARAM_VA_SIZE: + if (ctx->aspace == gpu->aspace) + return -EINVAL; + *value = ctx->aspace->va_size; + return 0; default: DBG("%s: invalid param: %u", gpu->name, param); return -EINVAL; diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 81b40c67e399..30e238682598 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -722,6 +722,23 @@ static int msm_ioctl_gem_info_iova(struct drm_device *dev, return msm_gem_get_iova(obj, ctx->aspace, iova); } +static int msm_ioctl_gem_info_set_iova(struct drm_device *dev, + struct drm_file *file, struct drm_gem_object *obj, + uint64_t iova) +{ + struct msm_drm_private *priv = dev->dev_private; + struct msm_file_private *ctx = file->driver_priv; + + if (!priv->gpu) + return -EINVAL; + + /* Only supported if per-process address space is supported: */ + if (priv->gpu->aspace == ctx->aspace) + return -EOPNOTSUPP; + + return msm_gem_set_iova(obj, ctx->aspace, iova); +} + static int msm_ioctl_gem_info(struct drm_device *dev, void *data, struct drm_file *file) { @@ -736,6 +753,7 @@ static int msm_ioctl_gem_info(struct drm_device *dev, void *data, switch (args->info) { case MSM_INFO_GET_OFFSET: case MSM_INFO_GET_IOVA: + case MSM_INFO_SET_IOVA: /* value returned as immediate, not pointer, so len==0: */ if (args->len) return -EINVAL; @@ -760,6 +778,9 @@ static int msm_ioctl_gem_info(struct drm_device *dev, void *data, case MSM_INFO_GET_IOVA: ret = msm_ioctl_gem_info_iova(dev, file, obj, &args->value); break; + case MSM_INFO_SET_IOVA: + ret = msm_ioctl_gem_info_set_iova(dev, file, obj, args->value); + break; case MSM_INFO_SET_NAME: /* length check should leave room for terminating null: */ if (args->len >= sizeof(msm_obj->name)) { diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c index 21986c321458..52fe6428a341 100644 --- a/drivers/gpu/drm/msm/msm_gem.c +++ b/drivers/gpu/drm/msm/msm_gem.c @@ -525,6 +525,54 @@ int msm_gem_get_iova(struct drm_gem_object *obj, return ret; } +static int clear_iova(struct drm_gem_object *obj, + struct msm_gem_address_space *aspace) +{ + struct msm_gem_vma *vma = lookup_vma(obj, aspace); + + if (!vma) + return 0; + + if (msm_gem_vma_inuse(vma)) + return -EBUSY; + + msm_gem_purge_vma(vma->aspace, vma); + msm_gem_close_vma(vma->aspace, vma); + del_vma(vma); + + return 0; +} + +/* + * Get the requested iova but don't pin it. Fails if the requested iova is + * not available. Doesn't need a put because iovas are currently valid for + * the life of the object. + * + * Setting an iova of zero will clear the vma. + */ +int msm_gem_set_iova(struct drm_gem_object *obj, + struct msm_gem_address_space *aspace, uint64_t iova) +{ + int ret = 0; + + msm_gem_lock(obj); + if (!iova) { + ret = clear_iova(obj, aspace); + } else { + struct msm_gem_vma *vma; + vma = get_vma_locked(obj, aspace, iova, iova + obj->size); + if (IS_ERR(vma)) { + ret = PTR_ERR(vma); + } else if (GEM_WARN_ON(vma->iova != iova)) { + clear_iova(obj, aspace); + ret = -EBUSY; + } + } + msm_gem_unlock(obj); + + return ret; +} + /* * Unpin a iova by updating the reference counts. The memory isn't actually * purged until something else (shrinker, mm_notifier, destroy, etc) decides diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h index 580b6eb95edd..c75d3b879a53 100644 --- a/drivers/gpu/drm/msm/msm_gem.h +++ b/drivers/gpu/drm/msm/msm_gem.h @@ -38,6 +38,12 @@ struct msm_gem_address_space { /* @faults: the number of GPU hangs associated with this address space */ int faults; + + /** @va_start: lowest possible address to allocate */ + uint64_t va_start; + + /** @va_size: the size of the address space (in bytes) */ + uint64_t va_size; }; struct msm_gem_address_space * @@ -144,6 +150,8 @@ struct msm_gem_vma *msm_gem_get_vma_locked(struct drm_gem_object *obj, struct msm_gem_address_space *aspace); int msm_gem_get_iova(struct drm_gem_object *obj, struct msm_gem_address_space *aspace, uint64_t *iova); +int msm_gem_set_iova(struct drm_gem_object *obj, + struct msm_gem_address_space *aspace, uint64_t iova); int msm_gem_get_and_pin_iova_range(struct drm_gem_object *obj, struct msm_gem_address_space *aspace, uint64_t *iova, u64 range_start, u64 range_end); diff --git a/drivers/gpu/drm/msm/msm_gem_vma.c b/drivers/gpu/drm/msm/msm_gem_vma.c index 0cd6770faf41..3c1dc9241831 100644 --- a/drivers/gpu/drm/msm/msm_gem_vma.c +++ b/drivers/gpu/drm/msm/msm_gem_vma.c @@ -184,6 +184,8 @@ msm_gem_address_space_create(struct msm_mmu *mmu, const char *name, spin_lock_init(&aspace->lock); aspace->name = name; aspace->mmu = mmu; + aspace->va_start = va_start; + aspace->va_size = size; drm_mm_init(&aspace->mm, va_start, size); diff --git a/include/uapi/drm/msm_drm.h b/include/uapi/drm/msm_drm.h index 794ad1948497..3c7b097c4e3d 100644 --- a/include/uapi/drm/msm_drm.h +++ b/include/uapi/drm/msm_drm.h @@ -84,6 +84,8 @@ struct drm_msm_timespec { #define MSM_PARAM_SYSPROF 0x0b /* WO: 1 preserves perfcntrs, 2 also disables suspend */ #define MSM_PARAM_COMM 0x0c /* WO: override for task->comm */ #define MSM_PARAM_CMDLINE 0x0d /* WO: override for task cmdline */ +#define MSM_PARAM_VA_START 0x0e /* RO: start of valid GPU iova range */ +#define MSM_PARAM_VA_SIZE 0x0f /* RO: size of valid GPU iova range (bytes) */ /* For backwards compat. The original support for preemption was based on * a single ring per priority level so # of priority levels equals the # @@ -135,6 +137,7 @@ struct drm_msm_gem_new { #define MSM_INFO_GET_IOVA 0x01 /* get iova, returned by value */ #define MSM_INFO_SET_NAME 0x02 /* set the debug name (by pointer) */ #define MSM_INFO_GET_NAME 0x03 /* get debug name, returned by pointer */ +#define MSM_INFO_SET_IOVA 0x04 /* set the iova, passed by value */ struct drm_msm_gem_info { __u32 handle; /* in */ -- cgit v1.2.3 From 991b4de3275728fd746ce15e00087fdbd51c29a1 Mon Sep 17 00:00:00 2001 From: Matt Roper Date: Wed, 27 Apr 2022 21:19:23 -0700 Subject: drm/i915/uapi: Add kerneldoc for engine class enum We'll be adding a new type of engine soon. Let's document the existing engine classes first to help make it clear what each type of engine is used for. Cc: Andi Shyti Signed-off-by: Matt Roper Reviewed-by: Andi Shyti Reviewed-by: Jordan Justen Link: https://patchwork.freedesktop.org/patch/msgid/20220428041926.1483683-2-matthew.d.roper@intel.com --- include/uapi/drm/i915_drm.h | 53 ++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 47 insertions(+), 6 deletions(-) (limited to 'include/uapi/drm') diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index 35ca528803fd..ec000fc6c879 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -154,21 +154,62 @@ enum i915_mocs_table_index { I915_MOCS_CACHED, }; -/* +/** + * enum drm_i915_gem_engine_class - uapi engine type enumeration + * * Different engines serve different roles, and there may be more than one - * engine serving each role. enum drm_i915_gem_engine_class provides a - * classification of the role of the engine, which may be used when requesting - * operations to be performed on a certain subset of engines, or for providing - * information about that group. + * engine serving each role. This enum provides a classification of the role + * of the engine, which may be used when requesting operations to be performed + * on a certain subset of engines, or for providing information about that + * group. */ enum drm_i915_gem_engine_class { + /** + * @I915_ENGINE_CLASS_RENDER: + * + * Render engines support instructions used for 3D, Compute (GPGPU), + * and programmable media workloads. These instructions fetch data and + * dispatch individual work items to threads that operate in parallel. + * The threads run small programs (called "kernels" or "shaders") on + * the GPU's execution units (EUs). + */ I915_ENGINE_CLASS_RENDER = 0, + + /** + * @I915_ENGINE_CLASS_COPY: + * + * Copy engines (also referred to as "blitters") support instructions + * that move blocks of data from one location in memory to another, + * or that fill a specified location of memory with fixed data. + * Copy engines can perform pre-defined logical or bitwise operations + * on the source, destination, or pattern data. + */ I915_ENGINE_CLASS_COPY = 1, + + /** + * @I915_ENGINE_CLASS_VIDEO: + * + * Video engines (also referred to as "bit stream decode" (BSD) or + * "vdbox") support instructions that perform fixed-function media + * decode and encode. + */ I915_ENGINE_CLASS_VIDEO = 2, + + /** + * @I915_ENGINE_CLASS_VIDEO_ENHANCE: + * + * Video enhancement engines (also referred to as "vebox") support + * instructions related to image enhancement. + */ I915_ENGINE_CLASS_VIDEO_ENHANCE = 3, - /* should be kept compact */ + /* Values in this enum should be kept compact. */ + /** + * @I915_ENGINE_CLASS_INVALID: + * + * Placeholder value to represent an invalid engine class assignment. + */ I915_ENGINE_CLASS_INVALID = -1 }; -- cgit v1.2.3 From ecf8eca51f33dbfddcc23902bdee65c17587247c Mon Sep 17 00:00:00 2001 From: Matt Roper Date: Wed, 27 Apr 2022 21:19:25 -0700 Subject: drm/i915/xehp: Add compute engine ABI We're now ready to start exposing compute engines to userspace. v2: - Move kerneldoc for other engine classes to a separate patch. (Andi) Cc: Daniele Ceraolo Spurio Cc: Tvrtko Ursulin Cc: Vinay Belgaumkar Cc: Jordan Justen Cc: Szymon Morek UMD (mesa): https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14395 Signed-off-by: Matt Roper Acked-by: Tvrtko Ursulin Reviewed-by: Andi Shyti Reviewed-by: Jordan Justen Tested-by: Jordan Justen # mesa anvil & iris Link: https://patchwork.freedesktop.org/patch/msgid/20220428041926.1483683-4-matthew.d.roper@intel.com --- drivers/gpu/drm/i915/gt/intel_engine_user.c | 2 +- drivers/gpu/drm/i915/i915_drm_client.c | 1 + drivers/gpu/drm/i915/i915_drm_client.h | 2 +- include/uapi/drm/i915_drm.h | 9 +++++++++ 4 files changed, 12 insertions(+), 2 deletions(-) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/i915/gt/intel_engine_user.c b/drivers/gpu/drm/i915/gt/intel_engine_user.c index 0f6cd96b459f..46a174f8aa00 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_user.c +++ b/drivers/gpu/drm/i915/gt/intel_engine_user.c @@ -47,7 +47,7 @@ static const u8 uabi_classes[] = { [COPY_ENGINE_CLASS] = I915_ENGINE_CLASS_COPY, [VIDEO_DECODE_CLASS] = I915_ENGINE_CLASS_VIDEO, [VIDEO_ENHANCEMENT_CLASS] = I915_ENGINE_CLASS_VIDEO_ENHANCE, - /* TODO: Add COMPUTE_CLASS mapping once ABI is available */ + [COMPUTE_CLASS] = I915_ENGINE_CLASS_COMPUTE, }; static int engine_cmp(void *priv, const struct list_head *A, diff --git a/drivers/gpu/drm/i915/i915_drm_client.c b/drivers/gpu/drm/i915/i915_drm_client.c index 475a6f824cad..18d38cb59923 100644 --- a/drivers/gpu/drm/i915/i915_drm_client.c +++ b/drivers/gpu/drm/i915/i915_drm_client.c @@ -81,6 +81,7 @@ static const char * const uabi_class_names[] = { [I915_ENGINE_CLASS_COPY] = "copy", [I915_ENGINE_CLASS_VIDEO] = "video", [I915_ENGINE_CLASS_VIDEO_ENHANCE] = "video-enhance", + [I915_ENGINE_CLASS_COMPUTE] = "compute", }; static u64 busy_add(struct i915_gem_context *ctx, unsigned int class) diff --git a/drivers/gpu/drm/i915/i915_drm_client.h b/drivers/gpu/drm/i915/i915_drm_client.h index 5f5b02b01ba0..f796c5e8e060 100644 --- a/drivers/gpu/drm/i915/i915_drm_client.h +++ b/drivers/gpu/drm/i915/i915_drm_client.h @@ -13,7 +13,7 @@ #include "gt/intel_engine_types.h" -#define I915_LAST_UABI_ENGINE_CLASS I915_ENGINE_CLASS_VIDEO_ENHANCE +#define I915_LAST_UABI_ENGINE_CLASS I915_ENGINE_CLASS_COMPUTE struct drm_i915_private; diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index ec000fc6c879..a2def7b27009 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -203,6 +203,15 @@ enum drm_i915_gem_engine_class { */ I915_ENGINE_CLASS_VIDEO_ENHANCE = 3, + /** + * @I915_ENGINE_CLASS_COMPUTE: + * + * Compute engines support a subset of the instructions available + * on render engines: compute engines support Compute (GPGPU) and + * programmable media workloads, but do not support the 3D pipeline. + */ + I915_ENGINE_CLASS_COMPUTE = 4, + /* Values in this enum should be kept compact. */ /** -- cgit v1.2.3 From 5eca8379c73d1bb8cd84e335b9529c263f699468 Mon Sep 17 00:00:00 2001 From: Hawking Zhang Date: Tue, 22 Feb 2022 13:38:51 +0800 Subject: drm/amdgpu: add GC v11_0_0 family id Add GC v11_0_0 family id Signed-off-by: Hawking Zhang Reviewed-by: Likun Gao Signed-off-by: Alex Deucher --- include/uapi/drm/amdgpu_drm.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/uapi/drm') diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index 1d65c1fbc4ec..992851219441 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -1150,6 +1150,7 @@ struct drm_amdgpu_info_video_caps { #define AMDGPU_FAMILY_RV 142 /* Raven */ #define AMDGPU_FAMILY_NV 143 /* Navi10 */ #define AMDGPU_FAMILY_VGH 144 /* Van Gogh */ +#define AMDGPU_FAMILY_GC_11_0_0 145 /* GC 11.0.0 */ #define AMDGPU_FAMILY_YC 146 /* Yellow Carp */ #define AMDGPU_FAMILY_GC_10_3_6 149 /* GC 10.3.6 */ #define AMDGPU_FAMILY_GC_10_3_7 151 /* GC 10.3.7 */ -- cgit v1.2.3 From cbe757ecdb4b0c8ac4c62b03cda3fa58ba6d23fd Mon Sep 17 00:00:00 2001 From: Huang Rui Date: Mon, 2 Aug 2021 09:06:03 +0800 Subject: drm/amdgpu: add GC 11.0.1 family id Add a flag to define GC 11.0.1. Signed-off-by: Huang Rui Reviewed-by: Aaron Liu Signed-off-by: Alex Deucher --- include/uapi/drm/amdgpu_drm.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/uapi/drm') diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index 992851219441..9a1d210d135d 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -1152,6 +1152,7 @@ struct drm_amdgpu_info_video_caps { #define AMDGPU_FAMILY_VGH 144 /* Van Gogh */ #define AMDGPU_FAMILY_GC_11_0_0 145 /* GC 11.0.0 */ #define AMDGPU_FAMILY_YC 146 /* Yellow Carp */ +#define AMDGPU_FAMILY_GC_11_0_1 148 /* GC 11.0.1 */ #define AMDGPU_FAMILY_GC_10_3_6 149 /* GC 10.3.6 */ #define AMDGPU_FAMILY_GC_10_3_7 151 /* GC 10.3.7 */ -- cgit v1.2.3 From d534ca7128d7bf681ed6d462c09b9d6ffb3bed91 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 23 May 2022 11:24:31 -0400 Subject: drm/amdgpu: differentiate between LP and non-LP DDR memory Some applications want to know whether the memory is LP or not. Reviewed-by: Hawking Zhang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c | 8 ++++++-- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 4 +++- include/uapi/drm/amdgpu_drm.h | 2 ++ 3 files changed, 11 insertions(+), 3 deletions(-) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c index 63e0293edc5f..fd8f3731758e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c @@ -188,13 +188,17 @@ static int convert_atom_mem_type_to_vram_type(struct amdgpu_device *adev, vram_type = AMDGPU_VRAM_TYPE_DDR3; break; case Ddr4MemType: - case LpDdr4MemType: vram_type = AMDGPU_VRAM_TYPE_DDR4; break; + case LpDdr4MemType: + vram_type = AMDGPU_VRAM_TYPE_LPDDR4; + break; case Ddr5MemType: - case LpDdr5MemType: vram_type = AMDGPU_VRAM_TYPE_DDR5; break; + case LpDdr5MemType: + vram_type = AMDGPU_VRAM_TYPE_LPDDR5; + break; default: vram_type = AMDGPU_VRAM_TYPE_UNKNOWN; break; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index 5444515c1476..e1e6441c475f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -1018,7 +1018,9 @@ static const char *amdgpu_vram_names[] = { "DDR3", "DDR4", "GDDR6", - "DDR5" + "DDR5", + "LPDDR4", + "LPDDR5" }; /** diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index 9a1d210d135d..d9d475d65c76 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -988,6 +988,8 @@ struct drm_amdgpu_info_vbios { #define AMDGPU_VRAM_TYPE_DDR4 8 #define AMDGPU_VRAM_TYPE_GDDR6 9 #define AMDGPU_VRAM_TYPE_DDR5 10 +#define AMDGPU_VRAM_TYPE_LPDDR4 11 +#define AMDGPU_VRAM_TYPE_LPDDR5 12 struct drm_amdgpu_info_device { /** PCI Device ID */ -- cgit v1.2.3 From fab2cc8335839867a3db38f195441b9c7c6460f6 Mon Sep 17 00:00:00 2001 From: Christian König Date: Fri, 6 May 2022 13:08:34 +0200 Subject: drm/amdgpu: add AMDGPU_GEM_CREATE_DISCARDABLE MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add a AMDGPU_GEM_CREATE_DISCARDABLE flag to note that the content of a BO doesn't needs to be preserved during eviction. KFD was already using a similar functionality for SVM BOs so replace the internal flag with the new UAPI. Signed-off-by: Christian König Reviewed-by: Alex Deucher Reviewed-by: Marek Olšák Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 4 ++-- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 1 + drivers/gpu/drm/amd/amdgpu/amdgpu_object.h | 1 - drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 2 +- drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 2 +- include/uapi/drm/amdgpu_drm.h | 4 ++++ 6 files changed, 9 insertions(+), 5 deletions(-) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 652571267077..56f4c9aa87c6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -296,8 +296,8 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data, AMDGPU_GEM_CREATE_VRAM_CLEARED | AMDGPU_GEM_CREATE_VM_ALWAYS_VALID | AMDGPU_GEM_CREATE_EXPLICIT_SYNC | - AMDGPU_GEM_CREATE_ENCRYPTED)) - + AMDGPU_GEM_CREATE_ENCRYPTED | + AMDGPU_GEM_CREATE_DISCARDABLE)) return -EINVAL; /* reject invalid gem domains */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index e1e6441c475f..2c82b1d5a0d7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -567,6 +567,7 @@ int amdgpu_bo_create(struct amdgpu_device *adev, bp->domain; bo->allowed_domains = bo->preferred_domains; if (bp->type != ttm_bo_type_kernel && + !(bp->flags & AMDGPU_GEM_CREATE_DISCARDABLE) && bo->allowed_domains == AMDGPU_GEM_DOMAIN_VRAM) bo->allowed_domains |= AMDGPU_GEM_DOMAIN_GTT; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h index 4c9cbdc66995..147b79c10cbb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h @@ -41,7 +41,6 @@ /* BO flag to indicate a KFD userptr BO */ #define AMDGPU_AMDKFD_CREATE_USERPTR_BO (1ULL << 63) -#define AMDGPU_AMDKFD_CREATE_SVM_BO (1ULL << 62) #define to_amdgpu_bo_user(abo) container_of((abo), struct amdgpu_bo_user, bo) #define to_amdgpu_bo_vm(abo) container_of((abo), struct amdgpu_bo_vm, bo) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index ec26edd4f4d8..be6f76a30ac6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -117,7 +117,7 @@ static void amdgpu_evict_flags(struct ttm_buffer_object *bo, } abo = ttm_to_amdgpu_bo(bo); - if (abo->flags & AMDGPU_AMDKFD_CREATE_SVM_BO) { + if (abo->flags & AMDGPU_GEM_CREATE_DISCARDABLE) { placement->num_placement = 0; placement->num_busy_placement = 0; return; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c index 29e9ebf6d8d5..2ebf0132c25b 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c @@ -531,7 +531,7 @@ svm_range_vram_node_new(struct amdgpu_device *adev, struct svm_range *prange, bp.domain = AMDGPU_GEM_DOMAIN_VRAM; bp.flags = AMDGPU_GEM_CREATE_NO_CPU_ACCESS; bp.flags |= clear ? AMDGPU_GEM_CREATE_VRAM_CLEARED : 0; - bp.flags |= AMDGPU_AMDKFD_CREATE_SVM_BO; + bp.flags |= AMDGPU_GEM_CREATE_DISCARDABLE; bp.type = ttm_bo_type_device; bp.resv = NULL; diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index d9d475d65c76..a81bef5cfeaa 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -140,6 +140,10 @@ extern "C" { * not require GTT memory accounting */ #define AMDGPU_GEM_CREATE_PREEMPTIBLE (1 << 11) +/* Flag that BO can be discarded under memory pressure without keeping the + * content. + */ +#define AMDGPU_GEM_CREATE_DISCARDABLE (1 << 12) struct drm_amdgpu_gem_create_in { /** the requested memory size */ -- cgit v1.2.3 From b6c65a2c92aa880e8050a91ca83288b85fc32575 Mon Sep 17 00:00:00 2001 From: Christian König Date: Fri, 6 May 2022 13:11:41 +0200 Subject: drm/amdgpu: add AMDGPU_VM_NOALLOC v2 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add the AMDGPU_VM_NOALLOC flag to let userspace control MALL allocation. v2: also add the flag to the allowed flags. Signed-off-by: Christian König Reviewed-by: Alex Deucher (v1) Reviewed-by: Marek Olšák Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 5 ++++- drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c | 3 +++ drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c | 3 +++ include/uapi/drm/amdgpu_drm.h | 2 ++ 4 files changed, 12 insertions(+), 1 deletion(-) (limited to 'include/uapi/drm') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 56f4c9aa87c6..8ef31d687ef3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -645,6 +645,8 @@ uint64_t amdgpu_gem_va_map_flags(struct amdgpu_device *adev, uint32_t flags) pte_flag |= AMDGPU_PTE_WRITEABLE; if (flags & AMDGPU_VM_PAGE_PRT) pte_flag |= AMDGPU_PTE_PRT; + if (flags & AMDGPU_VM_PAGE_NOALLOC) + pte_flag |= AMDGPU_PTE_NOALLOC; if (adev->gmc.gmc_funcs->map_mtype) pte_flag |= amdgpu_gmc_map_mtype(adev, @@ -658,7 +660,8 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, { const uint32_t valid_flags = AMDGPU_VM_DELAY_UPDATE | AMDGPU_VM_PAGE_READABLE | AMDGPU_VM_PAGE_WRITEABLE | - AMDGPU_VM_PAGE_EXECUTABLE | AMDGPU_VM_MTYPE_MASK; + AMDGPU_VM_PAGE_EXECUTABLE | AMDGPU_VM_MTYPE_MASK | + AMDGPU_VM_PAGE_NOALLOC; const uint32_t prt_flags = AMDGPU_VM_DELAY_UPDATE | AMDGPU_VM_PAGE_PRT; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index b8c79789e1e4..9077dfccaf3c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -613,6 +613,9 @@ static void gmc_v10_0_get_vm_pte(struct amdgpu_device *adev, *flags &= ~AMDGPU_PTE_MTYPE_NV10_MASK; *flags |= (mapping->flags & AMDGPU_PTE_MTYPE_NV10_MASK); + *flags &= ~AMDGPU_PTE_NOALLOC; + *flags |= (mapping->flags & AMDGPU_PTE_NOALLOC); + if (mapping->flags & AMDGPU_PTE_PRT) { *flags |= AMDGPU_PTE_PRT; *flags |= AMDGPU_PTE_SNOOPED; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c index 477f67d9b07c..a0c0b7d9f444 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c @@ -500,6 +500,9 @@ static void gmc_v11_0_get_vm_pte(struct amdgpu_device *adev, *flags &= ~AMDGPU_PTE_MTYPE_NV10_MASK; *flags |= (mapping->flags & AMDGPU_PTE_MTYPE_NV10_MASK); + *flags &= ~AMDGPU_PTE_NOALLOC; + *flags |= (mapping->flags & AMDGPU_PTE_NOALLOC); + if (mapping->flags & AMDGPU_PTE_PRT) { *flags |= AMDGPU_PTE_PRT; *flags |= AMDGPU_PTE_SNOOPED; diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index a81bef5cfeaa..18d3246d636e 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -533,6 +533,8 @@ struct drm_amdgpu_gem_op { #define AMDGPU_VM_MTYPE_UC (4 << 5) /* Use Read Write MTYPE instead of default MTYPE */ #define AMDGPU_VM_MTYPE_RW (5 << 5) +/* don't allocate MALL */ +#define AMDGPU_VM_PAGE_NOALLOC (1 << 9) struct drm_amdgpu_gem_va { /** GEM object handle */ -- cgit v1.2.3 From 20b8264394b33adb1640a485a62a84bc1388b6a3 Mon Sep 17 00:00:00 2001 From: Carlos Llamas Date: Tue, 21 Jun 2022 20:39:21 +0000 Subject: drm/fourcc: fix integer type usage in uapi header MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Kernel uapi headers are supposed to use __[us]{8,16,32,64} types defined by as opposed to 'uint32_t' and similar. See [1] for the relevant discussion about this topic. In this particular case, the usage of 'uint64_t' escaped headers_check as these macros are not being called here. However, the following program triggers a compilation error: #include int main() { unsigned long x = AMD_FMT_MOD_CLEAR(RB); return 0; } gcc error: drm.c:5:27: error: ‘uint64_t’ undeclared (first use in this function) 5 | unsigned long x = AMD_FMT_MOD_CLEAR(RB); | ^~~~~~~~~~~~~~~~~ This patch changes AMD_FMT_MOD_{SET,CLEAR} macros to use the correct integer types, which fixes the above issue. [1] https://lkml.org/lkml/2019/6/5/18 Fixes: 8ba16d599374 ("drm/fourcc: Add AMD DRM modifiers.") Signed-off-by: Carlos Llamas Reviewed-by: Simon Ser Signed-off-by: Alex Deucher --- include/uapi/drm/drm_fourcc.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/uapi/drm') diff --git a/include/uapi/drm/drm_fourcc.h b/include/uapi/drm/drm_fourcc.h index f1972154a594..0980678d502d 100644 --- a/include/uapi/drm/drm_fourcc.h +++ b/include/uapi/drm/drm_fourcc.h @@ -1444,11 +1444,11 @@ drm_fourcc_canonicalize_nvidia_format_mod(__u64 modifier) #define AMD_FMT_MOD_PIPE_MASK 0x7 #define AMD_FMT_MOD_SET(field, value) \ - ((uint64_t)(value) << AMD_FMT_MOD_##field##_SHIFT) + ((__u64)(value) << AMD_FMT_MOD_##field##_SHIFT) #define AMD_FMT_MOD_GET(field, value) \ (((value) >> AMD_FMT_MOD_##field##_SHIFT) & AMD_FMT_MOD_##field##_MASK) #define AMD_FMT_MOD_CLEAR(field) \ - (~((uint64_t)AMD_FMT_MOD_##field##_MASK << AMD_FMT_MOD_##field##_SHIFT)) + (~((__u64)AMD_FMT_MOD_##field##_MASK << AMD_FMT_MOD_##field##_SHIFT)) #if defined(__cplusplus) } -- cgit v1.2.3