diff options
| -rw-r--r-- | drivers/gpu/drm/xe/abi/guc_actions_abi.h | 4 | ||||
| -rw-r--r-- | drivers/gpu/drm/xe/abi/guc_klvs_abi.h | 15 | ||||
| -rw-r--r-- | drivers/gpu/drm/xe/xe_guc.c | 56 | ||||
| -rw-r--r-- | drivers/gpu/drm/xe/xe_guc.h | 1 | ||||
| -rw-r--r-- | drivers/gpu/drm/xe/xe_guc_submit.c | 21 | ||||
| -rw-r--r-- | drivers/gpu/drm/xe/xe_uc.c | 4 |
6 files changed, 98 insertions, 3 deletions
diff --git a/drivers/gpu/drm/xe/abi/guc_actions_abi.h b/drivers/gpu/drm/xe/abi/guc_actions_abi.h index ff4f412c28d8..81eb046aeebf 100644 --- a/drivers/gpu/drm/xe/abi/guc_actions_abi.h +++ b/drivers/gpu/drm/xe/abi/guc_actions_abi.h @@ -142,6 +142,7 @@ enum xe_guc_action { XE_GUC_ACTION_SET_ENG_UTIL_BUFF = 0x550A, XE_GUC_ACTION_SET_DEVICE_ENGINE_ACTIVITY_BUFFER = 0x550C, XE_GUC_ACTION_SET_FUNCTION_ENGINE_ACTIVITY_BUFFER = 0x550D, + XE_GUC_ACTION_OPT_IN_FEATURE_KLV = 0x550E, XE_GUC_ACTION_NOTIFY_MEMORY_CAT_ERROR = 0x6000, XE_GUC_ACTION_REPORT_PAGE_FAULT_REQ_DESC = 0x6002, XE_GUC_ACTION_PAGE_FAULT_RES_DESC = 0x6003, @@ -271,4 +272,7 @@ enum xe_guc_g2g_type { #define XE_G2G_DEREGISTER_TILE REG_GENMASK(15, 12) #define XE_G2G_DEREGISTER_TYPE REG_GENMASK(11, 8) +/* invalid type for XE_GUC_ACTION_NOTIFY_MEMORY_CAT_ERROR */ +#define XE_GUC_CAT_ERR_TYPE_INVALID 0xdeadbeef + #endif diff --git a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h index 7de8f827281f..5b2502bec2dc 100644 --- a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h +++ b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h @@ -16,6 +16,7 @@ * +===+=======+==============================================================+ * | 0 | 31:16 | **KEY** - KLV key identifier | * | | | - `GuC Self Config KLVs`_ | + * | | | - `GuC Opt In Feature KLVs`_ | * | | | - `GuC VGT Policy KLVs`_ | * | | | - `GuC VF Configuration KLVs`_ | * | | | | @@ -125,6 +126,20 @@ enum { }; /** + * DOC: GuC Opt In Feature KLVs + * + * `GuC KLV`_ keys available for use with OPT_IN_FEATURE_KLV + * + * _`GUC_KLV_OPT_IN_FEATURE_EXT_CAT_ERR_TYPE` : 0x4001 + * Adds an extra dword to the XE_GUC_ACTION_NOTIFY_MEMORY_CAT_ERROR G2H + * containing the type of the CAT error. On HW that does not support + * reporting the CAT error type, the extra dword is set to 0xdeadbeef. + */ + +#define GUC_KLV_OPT_IN_FEATURE_EXT_CAT_ERR_TYPE_KEY 0x4001 +#define GUC_KLV_OPT_IN_FEATURE_EXT_CAT_ERR_TYPE_LEN 0u + +/** * DOC: GuC VGT Policy KLVs * * `GuC KLV`_ keys available for use with PF2GUC_UPDATE_VGT_POLICY. diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c index 7a4207a2c1e0..28e24fe923a4 100644 --- a/drivers/gpu/drm/xe/xe_guc.c +++ b/drivers/gpu/drm/xe/xe_guc.c @@ -29,6 +29,7 @@ #include "xe_guc_db_mgr.h" #include "xe_guc_engine_activity.h" #include "xe_guc_hwconfig.h" +#include "xe_guc_klv_helpers.h" #include "xe_guc_log.h" #include "xe_guc_pc.h" #include "xe_guc_relay.h" @@ -570,6 +571,57 @@ err_deregister: return err; } +static int __guc_opt_in_features_enable(struct xe_guc *guc, u64 addr, u32 num_dwords) +{ + u32 action[] = { + XE_GUC_ACTION_OPT_IN_FEATURE_KLV, + lower_32_bits(addr), + upper_32_bits(addr), + num_dwords + }; + + return xe_guc_ct_send_block(&guc->ct, action, ARRAY_SIZE(action)); +} + +#define OPT_IN_MAX_DWORDS 16 +int xe_guc_opt_in_features_enable(struct xe_guc *guc) +{ + struct xe_device *xe = guc_to_xe(guc); + CLASS(xe_guc_buf, buf)(&guc->buf, OPT_IN_MAX_DWORDS); + u32 count = 0; + u32 *klvs; + int ret; + + if (!xe_guc_buf_is_valid(buf)) + return -ENOBUFS; + + klvs = xe_guc_buf_cpu_ptr(buf); + + /* + * The extra CAT error type opt-in was added in GuC v70.17.0, which maps + * to compatibility version v1.7.0. + * Note that the GuC allows enabling this KLV even on platforms that do + * not support the extra type; in such case the returned type variable + * will be set to a known invalid value which we can check against. + */ + if (GUC_SUBMIT_VER(guc) >= MAKE_GUC_VER(1, 7, 0)) + klvs[count++] = PREP_GUC_KLV_TAG(OPT_IN_FEATURE_EXT_CAT_ERR_TYPE); + + if (count) { + xe_assert(xe, count <= OPT_IN_MAX_DWORDS); + + ret = __guc_opt_in_features_enable(guc, xe_guc_buf_flush(buf), count); + if (ret < 0) { + xe_gt_err(guc_to_gt(guc), + "failed to enable GuC opt-in features: %pe\n", + ERR_PTR(ret)); + return ret; + } + } + + return 0; +} + static void guc_fini_hw(void *arg) { struct xe_guc *guc = arg; @@ -789,6 +841,10 @@ int xe_guc_post_load_init(struct xe_guc *guc) xe_guc_ads_populate_post_load(&guc->ads); + ret = xe_guc_opt_in_features_enable(guc); + if (ret) + return ret; + if (xe_guc_g2g_wanted(guc_to_xe(guc))) { ret = guc_g2g_start(guc); if (ret) diff --git a/drivers/gpu/drm/xe/xe_guc.h b/drivers/gpu/drm/xe/xe_guc.h index 965bf7291200..22cf019a11bf 100644 --- a/drivers/gpu/drm/xe/xe_guc.h +++ b/drivers/gpu/drm/xe/xe_guc.h @@ -34,6 +34,7 @@ int xe_guc_reset(struct xe_guc *guc); int xe_guc_upload(struct xe_guc *guc); int xe_guc_min_load_for_hwconfig(struct xe_guc *guc); int xe_guc_enable_communication(struct xe_guc *guc); +int xe_guc_opt_in_features_enable(struct xe_guc *guc); int xe_guc_suspend(struct xe_guc *guc); void xe_guc_notify(struct xe_guc *guc); int xe_guc_auth_huc(struct xe_guc *guc, u32 rsa_addr); diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c index 72477ccc5c5e..5f91b9a6ab7d 100644 --- a/drivers/gpu/drm/xe/xe_guc_submit.c +++ b/drivers/gpu/drm/xe/xe_guc_submit.c @@ -2092,12 +2092,16 @@ int xe_guc_exec_queue_memory_cat_error_handler(struct xe_guc *guc, u32 *msg, struct xe_gt *gt = guc_to_gt(guc); struct xe_exec_queue *q; u32 guc_id; + u32 type = XE_GUC_CAT_ERR_TYPE_INVALID; - if (unlikely(len < 1)) + if (unlikely(!len || len > 2)) return -EPROTO; guc_id = msg[0]; + if (len == 2) + type = msg[1]; + if (guc_id == GUC_ID_UNKNOWN) { /* * GuC uses GUC_ID_UNKNOWN if it can not map the CAT fault to any PF/VF @@ -2111,8 +2115,19 @@ int xe_guc_exec_queue_memory_cat_error_handler(struct xe_guc *guc, u32 *msg, if (unlikely(!q)) return -EPROTO; - xe_gt_dbg(gt, "Engine memory cat error: engine_class=%s, logical_mask: 0x%x, guc_id=%d", - xe_hw_engine_class_to_str(q->class), q->logical_mask, guc_id); + /* + * The type is HW-defined and changes based on platform, so we don't + * decode it in the kernel and only check if it is valid. + * See bspec 54047 and 72187 for details. + */ + if (type != XE_GUC_CAT_ERR_TYPE_INVALID) + xe_gt_dbg(gt, + "Engine memory CAT error [%u]: class=%s, logical_mask: 0x%x, guc_id=%d", + type, xe_hw_engine_class_to_str(q->class), q->logical_mask, guc_id); + else + xe_gt_dbg(gt, + "Engine memory CAT error: class=%s, logical_mask: 0x%x, guc_id=%d", + xe_hw_engine_class_to_str(q->class), q->logical_mask, guc_id); trace_xe_exec_queue_memory_cat_error(q); diff --git a/drivers/gpu/drm/xe/xe_uc.c b/drivers/gpu/drm/xe/xe_uc.c index ccfd059382fe..6431ba3a2c53 100644 --- a/drivers/gpu/drm/xe/xe_uc.c +++ b/drivers/gpu/drm/xe/xe_uc.c @@ -162,6 +162,10 @@ static int vf_uc_load_hw(struct xe_uc *uc) uc->guc.submission_state.enabled = true; + err = xe_guc_opt_in_features_enable(&uc->guc); + if (err) + return err; + err = xe_gt_record_default_lrcs(uc_to_gt(uc)); if (err) return err; |