X-Git-Url: http://drtracing.org/?a=blobdiff_plain;f=drivers%2Fgpu%2Fdrm%2Fi915%2Fi915_gpu_error.c;h=bc4a3ebc2662652b056ba24ae4e43d23b53d6dc8;hb=6361f4ba468264e8203111c5d138c7123e72a5c2;hp=831895b8cb75d61ec00857e9f66b0d529386a071;hpb=b340941af9d85ca74d3efacb7210e95296c484aa;p=deliverable%2Flinux.git diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index 831895b8cb75..bc4a3ebc2662 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -30,9 +30,9 @@ #include #include "i915_drv.h" -static const char *ring_str(int ring) +static const char *engine_str(int engine) { - switch (ring) { + switch (engine) { case RCS: return "render"; case VCS: return "bsd"; case BCS: return "blt"; @@ -198,7 +198,7 @@ static void print_error_buffers(struct drm_i915_error_state_buf *m, err->size, err->read_domains, err->write_domain); - for (i = 0; i < I915_NUM_RINGS; i++) + for (i = 0; i < I915_NUM_ENGINES; i++) err_printf(m, "%02x ", err->rseqno[i]); err_printf(m, "] %02x", err->wseqno); @@ -207,8 +207,8 @@ static void print_error_buffers(struct drm_i915_error_state_buf *m, err_puts(m, dirty_flag(err->dirty)); err_puts(m, purgeable_flag(err->purgeable)); err_puts(m, err->userptr ? " userptr" : ""); - err_puts(m, err->ring != -1 ? " " : ""); - err_puts(m, ring_str(err->ring)); + err_puts(m, err->engine != -1 ? " " : ""); + err_puts(m, engine_str(err->engine)); err_puts(m, i915_cache_level_str(m->i915, err->cache_level)); if (err->name) @@ -230,8 +230,6 @@ static const char *hangcheck_action_to_str(enum intel_ring_hangcheck_action a) return "wait"; case HANGCHECK_ACTIVE: return "active"; - case HANGCHECK_ACTIVE_LOOP: - return "active (loop)"; case HANGCHECK_KICK: return "kick"; case HANGCHECK_HUNG: @@ -241,69 +239,65 @@ static const char *hangcheck_action_to_str(enum intel_ring_hangcheck_action a) return "unknown"; } -static void i915_ring_error_state(struct drm_i915_error_state_buf *m, - struct drm_device *dev, - struct drm_i915_error_state *error, - int ring_idx) +static void error_print_engine(struct drm_i915_error_state_buf *m, + struct drm_i915_error_engine *ee) { - struct drm_i915_error_ring *ring = &error->ring[ring_idx]; - - if (!ring->valid) - return; - - err_printf(m, "%s command stream:\n", ring_str(ring_idx)); - err_printf(m, " START: 0x%08x\n", ring->start); - err_printf(m, " HEAD: 0x%08x\n", ring->head); - err_printf(m, " TAIL: 0x%08x\n", ring->tail); - err_printf(m, " CTL: 0x%08x\n", ring->ctl); - err_printf(m, " HWS: 0x%08x\n", ring->hws); - err_printf(m, " ACTHD: 0x%08x %08x\n", (u32)(ring->acthd>>32), (u32)ring->acthd); - err_printf(m, " IPEIR: 0x%08x\n", ring->ipeir); - err_printf(m, " IPEHR: 0x%08x\n", ring->ipehr); - err_printf(m, " INSTDONE: 0x%08x\n", ring->instdone); - if (INTEL_INFO(dev)->gen >= 4) { - err_printf(m, " BBADDR: 0x%08x %08x\n", (u32)(ring->bbaddr>>32), (u32)ring->bbaddr); - err_printf(m, " BB_STATE: 0x%08x\n", ring->bbstate); - err_printf(m, " INSTPS: 0x%08x\n", ring->instps); + err_printf(m, "%s command stream:\n", engine_str(ee->engine_id)); + err_printf(m, " START: 0x%08x\n", ee->start); + err_printf(m, " HEAD: 0x%08x\n", ee->head); + err_printf(m, " TAIL: 0x%08x\n", ee->tail); + err_printf(m, " CTL: 0x%08x\n", ee->ctl); + err_printf(m, " HWS: 0x%08x\n", ee->hws); + err_printf(m, " ACTHD: 0x%08x %08x\n", + (u32)(ee->acthd>>32), (u32)ee->acthd); + err_printf(m, " IPEIR: 0x%08x\n", ee->ipeir); + err_printf(m, " IPEHR: 0x%08x\n", ee->ipehr); + err_printf(m, " INSTDONE: 0x%08x\n", ee->instdone); + if (INTEL_GEN(m->i915) >= 4) { + err_printf(m, " BBADDR: 0x%08x %08x\n", + (u32)(ee->bbaddr>>32), (u32)ee->bbaddr); + err_printf(m, " BB_STATE: 0x%08x\n", ee->bbstate); + err_printf(m, " INSTPS: 0x%08x\n", ee->instps); } - err_printf(m, " INSTPM: 0x%08x\n", ring->instpm); - err_printf(m, " FADDR: 0x%08x %08x\n", upper_32_bits(ring->faddr), - lower_32_bits(ring->faddr)); - if (INTEL_INFO(dev)->gen >= 6) { - err_printf(m, " RC PSMI: 0x%08x\n", ring->rc_psmi); - err_printf(m, " FAULT_REG: 0x%08x\n", ring->fault_reg); + err_printf(m, " INSTPM: 0x%08x\n", ee->instpm); + err_printf(m, " FADDR: 0x%08x %08x\n", upper_32_bits(ee->faddr), + lower_32_bits(ee->faddr)); + if (INTEL_GEN(m->i915) >= 6) { + err_printf(m, " RC PSMI: 0x%08x\n", ee->rc_psmi); + err_printf(m, " FAULT_REG: 0x%08x\n", ee->fault_reg); err_printf(m, " SYNC_0: 0x%08x [last synced 0x%08x]\n", - ring->semaphore_mboxes[0], - ring->semaphore_seqno[0]); + ee->semaphore_mboxes[0], + ee->semaphore_seqno[0]); err_printf(m, " SYNC_1: 0x%08x [last synced 0x%08x]\n", - ring->semaphore_mboxes[1], - ring->semaphore_seqno[1]); - if (HAS_VEBOX(dev)) { + ee->semaphore_mboxes[1], + ee->semaphore_seqno[1]); + if (HAS_VEBOX(m->i915)) { err_printf(m, " SYNC_2: 0x%08x [last synced 0x%08x]\n", - ring->semaphore_mboxes[2], - ring->semaphore_seqno[2]); + ee->semaphore_mboxes[2], + ee->semaphore_seqno[2]); } } - if (USES_PPGTT(dev)) { - err_printf(m, " GFX_MODE: 0x%08x\n", ring->vm_info.gfx_mode); + if (USES_PPGTT(m->i915)) { + err_printf(m, " GFX_MODE: 0x%08x\n", ee->vm_info.gfx_mode); - if (INTEL_INFO(dev)->gen >= 8) { + if (INTEL_GEN(m->i915) >= 8) { int i; for (i = 0; i < 4; i++) err_printf(m, " PDP%d: 0x%016llx\n", - i, ring->vm_info.pdp[i]); + i, ee->vm_info.pdp[i]); } else { err_printf(m, " PP_DIR_BASE: 0x%08x\n", - ring->vm_info.pp_dir_base); + ee->vm_info.pp_dir_base); } } - err_printf(m, " seqno: 0x%08x\n", ring->seqno); - err_printf(m, " waiting: %s\n", yesno(ring->waiting)); - err_printf(m, " ring->head: 0x%08x\n", ring->cpu_ring_head); - err_printf(m, " ring->tail: 0x%08x\n", ring->cpu_ring_tail); + err_printf(m, " seqno: 0x%08x\n", ee->seqno); + err_printf(m, " last_seqno: 0x%08x\n", ee->last_seqno); + err_printf(m, " waiting: %s\n", yesno(ee->waiting)); + err_printf(m, " ring->head: 0x%08x\n", ee->cpu_ring_head); + err_printf(m, " ring->tail: 0x%08x\n", ee->cpu_ring_tail); err_printf(m, " hangcheck: %s [%d]\n", - hangcheck_action_to_str(ring->hangcheck_action), - ring->hangcheck_score); + hangcheck_action_to_str(ee->hangcheck_action), + ee->hangcheck_score); } void i915_error_printf(struct drm_i915_error_state_buf *e, const char *f, ...) @@ -333,7 +327,7 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, const struct i915_error_state_file_priv *error_priv) { struct drm_device *dev = error_priv->dev; - struct drm_i915_private *dev_priv = dev->dev_private; + struct drm_i915_private *dev_priv = to_i915(dev); struct drm_i915_error_state *error = error_priv->error; struct drm_i915_error_object *obj; int i, j, offset, elt; @@ -349,17 +343,17 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, error->time.tv_usec); err_printf(m, "Kernel: " UTS_RELEASE "\n"); max_hangcheck_score = 0; - for (i = 0; i < ARRAY_SIZE(error->ring); i++) { - if (error->ring[i].hangcheck_score > max_hangcheck_score) - max_hangcheck_score = error->ring[i].hangcheck_score; + for (i = 0; i < ARRAY_SIZE(error->engine); i++) { + if (error->engine[i].hangcheck_score > max_hangcheck_score) + max_hangcheck_score = error->engine[i].hangcheck_score; } - for (i = 0; i < ARRAY_SIZE(error->ring); i++) { - if (error->ring[i].hangcheck_score == max_hangcheck_score && - error->ring[i].pid != -1) { + for (i = 0; i < ARRAY_SIZE(error->engine); i++) { + if (error->engine[i].hangcheck_score == max_hangcheck_score && + error->engine[i].pid != -1) { err_printf(m, "Active process (on ring %s): %s [%d]\n", - ring_str(i), - error->ring[i].comm, - error->ring[i].pid); + engine_str(i), + error->engine[i].comm, + error->engine[i].pid); } } err_printf(m, "Reset count: %u\n", error->reset_count); @@ -412,11 +406,13 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, err_printf(m, "DONE_REG: 0x%08x\n", error->done_reg); } - if (INTEL_INFO(dev)->gen == 7) + if (IS_GEN7(dev)) err_printf(m, "ERR_INT: 0x%08x\n", error->err_int); - for (i = 0; i < ARRAY_SIZE(error->ring); i++) - i915_ring_error_state(m, dev, error, i); + for (i = 0; i < ARRAY_SIZE(error->engine); i++) { + if (error->engine[i].engine_id != -1) + error_print_engine(m, &error->engine[i]); + } for (i = 0; i < error->vm_count; i++) { err_printf(m, "vm[%d]\n", i); @@ -430,48 +426,62 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, error->pinned_bo_count[i]); } - for (i = 0; i < ARRAY_SIZE(error->ring); i++) { - obj = error->ring[i].batchbuffer; + for (i = 0; i < ARRAY_SIZE(error->engine); i++) { + struct drm_i915_error_engine *ee = &error->engine[i]; + + obj = ee->batchbuffer; if (obj) { - err_puts(m, dev_priv->ring[i].name); - if (error->ring[i].pid != -1) + err_puts(m, dev_priv->engine[i].name); + if (ee->pid != -1) err_printf(m, " (submitted by %s [%d])", - error->ring[i].comm, - error->ring[i].pid); + ee->comm, + ee->pid); err_printf(m, " --- gtt_offset = 0x%08x %08x\n", upper_32_bits(obj->gtt_offset), lower_32_bits(obj->gtt_offset)); print_error_obj(m, obj); } - obj = error->ring[i].wa_batchbuffer; + obj = ee->wa_batchbuffer; if (obj) { err_printf(m, "%s (w/a) --- gtt_offset = 0x%08x\n", - dev_priv->ring[i].name, + dev_priv->engine[i].name, lower_32_bits(obj->gtt_offset)); print_error_obj(m, obj); } - if (error->ring[i].num_requests) { + if (ee->num_requests) { err_printf(m, "%s --- %d requests\n", - dev_priv->ring[i].name, - error->ring[i].num_requests); - for (j = 0; j < error->ring[i].num_requests; j++) { + dev_priv->engine[i].name, + ee->num_requests); + for (j = 0; j < ee->num_requests; j++) { err_printf(m, " seqno 0x%08x, emitted %ld, tail 0x%08x\n", - error->ring[i].requests[j].seqno, - error->ring[i].requests[j].jiffies, - error->ring[i].requests[j].tail); + ee->requests[j].seqno, + ee->requests[j].jiffies, + ee->requests[j].tail); } } - if ((obj = error->ring[i].ringbuffer)) { + if (ee->num_waiters) { + err_printf(m, "%s --- %d waiters\n", + dev_priv->engine[i].name, + ee->num_waiters); + for (j = 0; j < ee->num_waiters; j++) { + err_printf(m, " seqno 0x%08x for %s [%d]\n", + ee->waiters[j].seqno, + ee->waiters[j].comm, + ee->waiters[j].pid); + } + } + + if ((obj = ee->ringbuffer)) { err_printf(m, "%s --- ringbuffer = 0x%08x\n", - dev_priv->ring[i].name, + dev_priv->engine[i].name, lower_32_bits(obj->gtt_offset)); print_error_obj(m, obj); } - if ((obj = error->ring[i].hws_page)) { + if ((obj = ee->hws_page)) { u64 hws_offset = obj->gtt_offset; u32 *hws_page = &obj->pages[0][0]; @@ -480,7 +490,7 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, hws_page = &obj->pages[LRC_PPHWSP_PN][0]; } err_printf(m, "%s --- HW Status = 0x%08llx\n", - dev_priv->ring[i].name, hws_offset); + dev_priv->engine[i].name, hws_offset); offset = 0; for (elt = 0; elt < PAGE_SIZE/16; elt += 4) { err_printf(m, "[%04x] %08x %08x %08x %08x\n", @@ -489,13 +499,35 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, hws_page[elt+1], hws_page[elt+2], hws_page[elt+3]); - offset += 16; + offset += 16; + } + } + + obj = ee->wa_ctx; + if (obj) { + u64 wa_ctx_offset = obj->gtt_offset; + u32 *wa_ctx_page = &obj->pages[0][0]; + struct intel_engine_cs *engine = &dev_priv->engine[RCS]; + u32 wa_ctx_size = (engine->wa_ctx.indirect_ctx.size + + engine->wa_ctx.per_ctx.size); + + err_printf(m, "%s --- WA ctx batch buffer = 0x%08llx\n", + dev_priv->engine[i].name, wa_ctx_offset); + offset = 0; + for (elt = 0; elt < wa_ctx_size; elt += 4) { + err_printf(m, "[%04x] %08x %08x %08x %08x\n", + offset, + wa_ctx_page[elt + 0], + wa_ctx_page[elt + 1], + wa_ctx_page[elt + 2], + wa_ctx_page[elt + 3]); + offset += 16; } } - if ((obj = error->ring[i].ctx)) { + if ((obj = ee->ctx)) { err_printf(m, "%s --- HW Context = 0x%08x\n", - dev_priv->ring[i].name, + dev_priv->engine[i].name, lower_32_bits(obj->gtt_offset)); print_error_obj(m, obj); } @@ -578,13 +610,18 @@ static void i915_error_state_free(struct kref *error_ref) typeof(*error), ref); int i; - for (i = 0; i < ARRAY_SIZE(error->ring); i++) { - i915_error_object_free(error->ring[i].batchbuffer); - i915_error_object_free(error->ring[i].wa_batchbuffer); - i915_error_object_free(error->ring[i].ringbuffer); - i915_error_object_free(error->ring[i].hws_page); - i915_error_object_free(error->ring[i].ctx); - kfree(error->ring[i].requests); + for (i = 0; i < ARRAY_SIZE(error->engine); i++) { + struct drm_i915_error_engine *ee = &error->engine[i]; + + i915_error_object_free(ee->batchbuffer); + i915_error_object_free(ee->wa_batchbuffer); + i915_error_object_free(ee->ringbuffer); + i915_error_object_free(ee->hws_page); + i915_error_object_free(ee->ctx); + i915_error_object_free(ee->wa_ctx); + + kfree(ee->requests); + kfree(ee->waiters); } i915_error_object_free(error->semaphore_obj); @@ -606,6 +643,7 @@ i915_error_object_create(struct drm_i915_private *dev_priv, struct drm_i915_gem_object *src, struct i915_address_space *vm) { + struct i915_ggtt *ggtt = &dev_priv->ggtt; struct drm_i915_error_object *dst; struct i915_vma *vma = NULL; int num_pages; @@ -632,7 +670,7 @@ i915_error_object_create(struct drm_i915_private *dev_priv, vma = i915_gem_obj_to_ggtt(src); use_ggtt = (src->cache_level == I915_CACHE_NONE && vma && (vma->bound & GLOBAL_BIND) && - reloc_offset + num_pages * PAGE_SIZE <= dev_priv->gtt.mappable_end); + reloc_offset + num_pages * PAGE_SIZE <= ggtt->mappable_end); /* Cannot access stolen address directly, try to use the aperture */ if (src->stolen) { @@ -642,12 +680,13 @@ i915_error_object_create(struct drm_i915_private *dev_priv, goto unwind; reloc_offset = i915_gem_obj_ggtt_offset(src); - if (reloc_offset + num_pages * PAGE_SIZE > dev_priv->gtt.mappable_end) + if (reloc_offset + num_pages * PAGE_SIZE > ggtt->mappable_end) goto unwind; } /* Cannot access snooped pages through the aperture */ - if (use_ggtt && src->cache_level != I915_CACHE_NONE && !HAS_LLC(dev_priv->dev)) + if (use_ggtt && src->cache_level != I915_CACHE_NONE && + !HAS_LLC(dev_priv)) goto unwind; dst->page_count = num_pages; @@ -668,7 +707,7 @@ i915_error_object_create(struct drm_i915_private *dev_priv, * captures what the GPU read. */ - s = io_mapping_map_atomic_wc(dev_priv->gtt.mappable, + s = io_mapping_map_atomic_wc(ggtt->mappable, reloc_offset); memcpy_fromio(d, s, PAGE_SIZE); io_mapping_unmap_atomic(s); @@ -701,7 +740,7 @@ unwind: return NULL; } #define i915_error_ggtt_object_create(dev_priv, src) \ - i915_error_object_create((dev_priv), (src), &(dev_priv)->gtt.base) + i915_error_object_create((dev_priv), (src), &(dev_priv)->ggtt.base) static void capture_bo(struct drm_i915_error_buffer *err, struct i915_vma *vma) @@ -711,7 +750,7 @@ static void capture_bo(struct drm_i915_error_buffer *err, err->size = obj->base.size; err->name = obj->base.name; - for (i = 0; i < I915_NUM_RINGS; i++) + for (i = 0; i < I915_NUM_ENGINES; i++) err->rseqno[i] = i915_gem_request_get_seqno(obj->last_read_req[i]); err->wseqno = i915_gem_request_get_seqno(obj->last_write_req); err->gtt_offset = vma->node.start; @@ -725,8 +764,8 @@ static void capture_bo(struct drm_i915_error_buffer *err, err->dirty = obj->dirty; err->purgeable = obj->madv != I915_MADV_WILLNEED; err->userptr = obj->userptr.mm != NULL; - err->ring = obj->last_write_req ? - i915_gem_request_get_ring(obj->last_write_req)->id : -1; + err->engine = obj->last_write_req ? + i915_gem_request_get_engine(obj->last_write_req)->id : -1; err->cache_level = obj->cache_level; } @@ -778,7 +817,7 @@ static u32 capture_pinned_bo(struct drm_i915_error_buffer *err, */ static uint32_t i915_error_generate_code(struct drm_i915_private *dev_priv, struct drm_i915_error_state *error, - int *ring_id) + int *engine_id) { uint32_t error_code = 0; int i; @@ -788,136 +827,173 @@ static uint32_t i915_error_generate_code(struct drm_i915_private *dev_priv, * synchronization commands which almost always appear in the case * strictly a client bug. Use instdone to differentiate those some. */ - for (i = 0; i < I915_NUM_RINGS; i++) { - if (error->ring[i].hangcheck_action == HANGCHECK_HUNG) { - if (ring_id) - *ring_id = i; + for (i = 0; i < I915_NUM_ENGINES; i++) { + if (error->engine[i].hangcheck_action == HANGCHECK_HUNG) { + if (engine_id) + *engine_id = i; - return error->ring[i].ipehr ^ error->ring[i].instdone; + return error->engine[i].ipehr ^ error->engine[i].instdone; } } return error_code; } -static void i915_gem_record_fences(struct drm_device *dev, +static void i915_gem_record_fences(struct drm_i915_private *dev_priv, struct drm_i915_error_state *error) { - struct drm_i915_private *dev_priv = dev->dev_private; int i; - if (IS_GEN3(dev) || IS_GEN2(dev)) { + if (IS_GEN3(dev_priv) || IS_GEN2(dev_priv)) { for (i = 0; i < dev_priv->num_fence_regs; i++) error->fence[i] = I915_READ(FENCE_REG(i)); - } else if (IS_GEN5(dev) || IS_GEN4(dev)) { + } else if (IS_GEN5(dev_priv) || IS_GEN4(dev_priv)) { for (i = 0; i < dev_priv->num_fence_regs; i++) error->fence[i] = I915_READ64(FENCE_REG_965_LO(i)); - } else if (INTEL_INFO(dev)->gen >= 6) { + } else if (INTEL_GEN(dev_priv) >= 6) { for (i = 0; i < dev_priv->num_fence_regs; i++) error->fence[i] = I915_READ64(FENCE_REG_GEN6_LO(i)); } } -static void gen8_record_semaphore_state(struct drm_i915_private *dev_priv, - struct drm_i915_error_state *error, - struct intel_engine_cs *ring, - struct drm_i915_error_ring *ering) +static void gen8_record_semaphore_state(struct drm_i915_error_state *error, + struct intel_engine_cs *engine, + struct drm_i915_error_engine *ee) { + struct drm_i915_private *dev_priv = engine->i915; struct intel_engine_cs *to; - int i; - - if (!i915_semaphore_is_enabled(dev_priv->dev)) - return; + enum intel_engine_id id; if (!error->semaphore_obj) - error->semaphore_obj = - i915_error_ggtt_object_create(dev_priv, - dev_priv->semaphore_obj); + return; - for_each_ring(to, dev_priv, i) { + for_each_engine_id(to, dev_priv, id) { int idx; u16 signal_offset; u32 *tmp; - if (ring == to) + if (engine == to) continue; - signal_offset = (GEN8_SIGNAL_OFFSET(ring, i) & (PAGE_SIZE - 1)) - / 4; + signal_offset = + (GEN8_SIGNAL_OFFSET(engine, id) & (PAGE_SIZE - 1)) / 4; tmp = error->semaphore_obj->pages[0]; - idx = intel_ring_sync_index(ring, to); + idx = intel_ring_sync_index(engine, to); - ering->semaphore_mboxes[idx] = tmp[signal_offset]; - ering->semaphore_seqno[idx] = ring->semaphore.sync_seqno[idx]; + ee->semaphore_mboxes[idx] = tmp[signal_offset]; + ee->semaphore_seqno[idx] = engine->semaphore.sync_seqno[idx]; } } -static void gen6_record_semaphore_state(struct drm_i915_private *dev_priv, - struct intel_engine_cs *ring, - struct drm_i915_error_ring *ering) +static void gen6_record_semaphore_state(struct intel_engine_cs *engine, + struct drm_i915_error_engine *ee) { - ering->semaphore_mboxes[0] = I915_READ(RING_SYNC_0(ring->mmio_base)); - ering->semaphore_mboxes[1] = I915_READ(RING_SYNC_1(ring->mmio_base)); - ering->semaphore_seqno[0] = ring->semaphore.sync_seqno[0]; - ering->semaphore_seqno[1] = ring->semaphore.sync_seqno[1]; - - if (HAS_VEBOX(dev_priv->dev)) { - ering->semaphore_mboxes[2] = - I915_READ(RING_SYNC_2(ring->mmio_base)); - ering->semaphore_seqno[2] = ring->semaphore.sync_seqno[2]; + struct drm_i915_private *dev_priv = engine->i915; + + ee->semaphore_mboxes[0] = I915_READ(RING_SYNC_0(engine->mmio_base)); + ee->semaphore_mboxes[1] = I915_READ(RING_SYNC_1(engine->mmio_base)); + ee->semaphore_seqno[0] = engine->semaphore.sync_seqno[0]; + ee->semaphore_seqno[1] = engine->semaphore.sync_seqno[1]; + + if (HAS_VEBOX(dev_priv)) { + ee->semaphore_mboxes[2] = + I915_READ(RING_SYNC_2(engine->mmio_base)); + ee->semaphore_seqno[2] = engine->semaphore.sync_seqno[2]; } } -static void i915_record_ring_state(struct drm_device *dev, - struct drm_i915_error_state *error, - struct intel_engine_cs *ring, - struct drm_i915_error_ring *ering) +static void error_record_engine_waiters(struct intel_engine_cs *engine, + struct drm_i915_error_engine *ee) { - struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_breadcrumbs *b = &engine->breadcrumbs; + struct drm_i915_error_waiter *waiter; + struct rb_node *rb; + int count; + + ee->num_waiters = 0; + ee->waiters = NULL; + + spin_lock(&b->lock); + count = 0; + for (rb = rb_first(&b->waiters); rb != NULL; rb = rb_next(rb)) + count++; + spin_unlock(&b->lock); + + waiter = NULL; + if (count) + waiter = kmalloc_array(count, + sizeof(struct drm_i915_error_waiter), + GFP_ATOMIC); + if (!waiter) + return; - if (INTEL_INFO(dev)->gen >= 6) { - ering->rc_psmi = I915_READ(RING_PSMI_CTL(ring->mmio_base)); - ering->fault_reg = I915_READ(RING_FAULT_REG(ring)); - if (INTEL_INFO(dev)->gen >= 8) - gen8_record_semaphore_state(dev_priv, error, ring, ering); + ee->waiters = waiter; + + spin_lock(&b->lock); + for (rb = rb_first(&b->waiters); rb; rb = rb_next(rb)) { + struct intel_wait *w = container_of(rb, typeof(*w), node); + + strcpy(waiter->comm, w->tsk->comm); + waiter->pid = w->tsk->pid; + waiter->seqno = w->seqno; + waiter++; + + if (++ee->num_waiters == count) + break; + } + spin_unlock(&b->lock); +} + +static void error_record_engine_registers(struct drm_i915_error_state *error, + struct intel_engine_cs *engine, + struct drm_i915_error_engine *ee) +{ + struct drm_i915_private *dev_priv = engine->i915; + + if (INTEL_GEN(dev_priv) >= 6) { + ee->rc_psmi = I915_READ(RING_PSMI_CTL(engine->mmio_base)); + ee->fault_reg = I915_READ(RING_FAULT_REG(engine)); + if (INTEL_GEN(dev_priv) >= 8) + gen8_record_semaphore_state(error, engine, ee); else - gen6_record_semaphore_state(dev_priv, ring, ering); + gen6_record_semaphore_state(engine, ee); } - if (INTEL_INFO(dev)->gen >= 4) { - ering->faddr = I915_READ(RING_DMA_FADD(ring->mmio_base)); - ering->ipeir = I915_READ(RING_IPEIR(ring->mmio_base)); - ering->ipehr = I915_READ(RING_IPEHR(ring->mmio_base)); - ering->instdone = I915_READ(RING_INSTDONE(ring->mmio_base)); - ering->instps = I915_READ(RING_INSTPS(ring->mmio_base)); - ering->bbaddr = I915_READ(RING_BBADDR(ring->mmio_base)); - if (INTEL_INFO(dev)->gen >= 8) { - ering->faddr |= (u64) I915_READ(RING_DMA_FADD_UDW(ring->mmio_base)) << 32; - ering->bbaddr |= (u64) I915_READ(RING_BBADDR_UDW(ring->mmio_base)) << 32; + if (INTEL_GEN(dev_priv) >= 4) { + ee->faddr = I915_READ(RING_DMA_FADD(engine->mmio_base)); + ee->ipeir = I915_READ(RING_IPEIR(engine->mmio_base)); + ee->ipehr = I915_READ(RING_IPEHR(engine->mmio_base)); + ee->instdone = I915_READ(RING_INSTDONE(engine->mmio_base)); + ee->instps = I915_READ(RING_INSTPS(engine->mmio_base)); + ee->bbaddr = I915_READ(RING_BBADDR(engine->mmio_base)); + if (INTEL_GEN(dev_priv) >= 8) { + ee->faddr |= (u64) I915_READ(RING_DMA_FADD_UDW(engine->mmio_base)) << 32; + ee->bbaddr |= (u64) I915_READ(RING_BBADDR_UDW(engine->mmio_base)) << 32; } - ering->bbstate = I915_READ(RING_BBSTATE(ring->mmio_base)); + ee->bbstate = I915_READ(RING_BBSTATE(engine->mmio_base)); } else { - ering->faddr = I915_READ(DMA_FADD_I8XX); - ering->ipeir = I915_READ(IPEIR); - ering->ipehr = I915_READ(IPEHR); - ering->instdone = I915_READ(GEN2_INSTDONE); + ee->faddr = I915_READ(DMA_FADD_I8XX); + ee->ipeir = I915_READ(IPEIR); + ee->ipehr = I915_READ(IPEHR); + ee->instdone = I915_READ(GEN2_INSTDONE); } - ering->waiting = waitqueue_active(&ring->irq_queue); - ering->instpm = I915_READ(RING_INSTPM(ring->mmio_base)); - ering->seqno = ring->get_seqno(ring, false); - ering->acthd = intel_ring_get_active_head(ring); - ering->start = I915_READ_START(ring); - ering->head = I915_READ_HEAD(ring); - ering->tail = I915_READ_TAIL(ring); - ering->ctl = I915_READ_CTL(ring); - - if (I915_NEED_GFX_HWS(dev)) { + ee->waiting = intel_engine_has_waiter(engine); + ee->instpm = I915_READ(RING_INSTPM(engine->mmio_base)); + ee->acthd = intel_ring_get_active_head(engine); + ee->seqno = intel_engine_get_seqno(engine); + ee->last_seqno = engine->last_submitted_seqno; + ee->start = I915_READ_START(engine); + ee->head = I915_READ_HEAD(engine); + ee->tail = I915_READ_TAIL(engine); + ee->ctl = I915_READ_CTL(engine); + + if (I915_NEED_GFX_HWS(dev_priv)) { i915_reg_t mmio; - if (IS_GEN7(dev)) { - switch (ring->id) { + if (IS_GEN7(dev_priv)) { + switch (engine->id) { default: case RCS: mmio = RENDER_HWS_PGA_GEN7; @@ -932,51 +1008,51 @@ static void i915_record_ring_state(struct drm_device *dev, mmio = VEBOX_HWS_PGA_GEN7; break; } - } else if (IS_GEN6(ring->dev)) { - mmio = RING_HWS_PGA_GEN6(ring->mmio_base); + } else if (IS_GEN6(engine->i915)) { + mmio = RING_HWS_PGA_GEN6(engine->mmio_base); } else { /* XXX: gen8 returns to sanity */ - mmio = RING_HWS_PGA(ring->mmio_base); + mmio = RING_HWS_PGA(engine->mmio_base); } - ering->hws = I915_READ(mmio); + ee->hws = I915_READ(mmio); } - ering->hangcheck_score = ring->hangcheck.score; - ering->hangcheck_action = ring->hangcheck.action; + ee->hangcheck_score = engine->hangcheck.score; + ee->hangcheck_action = engine->hangcheck.action; - if (USES_PPGTT(dev)) { + if (USES_PPGTT(dev_priv)) { int i; - ering->vm_info.gfx_mode = I915_READ(RING_MODE_GEN7(ring)); + ee->vm_info.gfx_mode = I915_READ(RING_MODE_GEN7(engine)); - if (IS_GEN6(dev)) - ering->vm_info.pp_dir_base = - I915_READ(RING_PP_DIR_BASE_READ(ring)); - else if (IS_GEN7(dev)) - ering->vm_info.pp_dir_base = - I915_READ(RING_PP_DIR_BASE(ring)); - else if (INTEL_INFO(dev)->gen >= 8) + if (IS_GEN6(dev_priv)) + ee->vm_info.pp_dir_base = + I915_READ(RING_PP_DIR_BASE_READ(engine)); + else if (IS_GEN7(dev_priv)) + ee->vm_info.pp_dir_base = + I915_READ(RING_PP_DIR_BASE(engine)); + else if (INTEL_GEN(dev_priv) >= 8) for (i = 0; i < 4; i++) { - ering->vm_info.pdp[i] = - I915_READ(GEN8_RING_PDP_UDW(ring, i)); - ering->vm_info.pdp[i] <<= 32; - ering->vm_info.pdp[i] |= - I915_READ(GEN8_RING_PDP_LDW(ring, i)); + ee->vm_info.pdp[i] = + I915_READ(GEN8_RING_PDP_UDW(engine, i)); + ee->vm_info.pdp[i] <<= 32; + ee->vm_info.pdp[i] |= + I915_READ(GEN8_RING_PDP_LDW(engine, i)); } } } -static void i915_gem_record_active_context(struct intel_engine_cs *ring, +static void i915_gem_record_active_context(struct intel_engine_cs *engine, struct drm_i915_error_state *error, - struct drm_i915_error_ring *ering) + struct drm_i915_error_engine *ee) { - struct drm_i915_private *dev_priv = ring->dev->dev_private; + struct drm_i915_private *dev_priv = engine->i915; struct drm_i915_gem_object *obj; /* Currently render ring is the only HW context user */ - if (ring->id != RCS || !error->ccid) + if (engine->id != RCS || !error->ccid) return; list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) { @@ -984,53 +1060,61 @@ static void i915_gem_record_active_context(struct intel_engine_cs *ring, continue; if ((error->ccid & PAGE_MASK) == i915_gem_obj_ggtt_offset(obj)) { - ering->ctx = i915_error_ggtt_object_create(dev_priv, obj); + ee->ctx = i915_error_ggtt_object_create(dev_priv, obj); break; } } } -static void i915_gem_record_rings(struct drm_device *dev, +static void i915_gem_record_rings(struct drm_i915_private *dev_priv, struct drm_i915_error_state *error) { - struct drm_i915_private *dev_priv = dev->dev_private; + struct i915_ggtt *ggtt = &dev_priv->ggtt; struct drm_i915_gem_request *request; int i, count; - for (i = 0; i < I915_NUM_RINGS; i++) { - struct intel_engine_cs *ring = &dev_priv->ring[i]; - struct intel_ringbuffer *rbuf; + if (dev_priv->semaphore_obj) { + error->semaphore_obj = + i915_error_ggtt_object_create(dev_priv, + dev_priv->semaphore_obj); + } - error->ring[i].pid = -1; + for (i = 0; i < I915_NUM_ENGINES; i++) { + struct intel_engine_cs *engine = &dev_priv->engine[i]; + struct drm_i915_error_engine *ee = &error->engine[i]; - if (ring->dev == NULL) + ee->pid = -1; + ee->engine_id = -1; + + if (!intel_engine_initialized(engine)) continue; - error->ring[i].valid = true; + ee->engine_id = i; - i915_record_ring_state(dev, error, ring, &error->ring[i]); + error_record_engine_registers(error, engine, ee); + error_record_engine_waiters(engine, ee); - request = i915_gem_find_active_request(ring); + request = i915_gem_find_active_request(engine); if (request) { struct i915_address_space *vm; + struct intel_ringbuffer *rb; - vm = request->ctx && request->ctx->ppgtt ? - &request->ctx->ppgtt->base : - &dev_priv->gtt.base; + vm = request->ctx->ppgtt ? + &request->ctx->ppgtt->base : &ggtt->base; /* We need to copy these to an anonymous buffer * as the simplest method to avoid being overwritten * by userspace. */ - error->ring[i].batchbuffer = + ee->batchbuffer = i915_error_object_create(dev_priv, request->batch_obj, vm); - if (HAS_BROKEN_CS_TLB(dev_priv->dev)) - error->ring[i].wa_batchbuffer = + if (HAS_BROKEN_CS_TLB(dev_priv)) + ee->wa_batchbuffer = i915_error_ggtt_object_create(dev_priv, - ring->scratch.obj); + engine->scratch.obj); if (request->pid) { struct task_struct *task; @@ -1038,55 +1122,52 @@ static void i915_gem_record_rings(struct drm_device *dev, rcu_read_lock(); task = pid_task(request->pid, PIDTYPE_PID); if (task) { - strcpy(error->ring[i].comm, task->comm); - error->ring[i].pid = task->pid; + strcpy(ee->comm, task->comm); + ee->pid = task->pid; } rcu_read_unlock(); } - } - if (i915.enable_execlists) { - /* TODO: This is only a small fix to keep basic error - * capture working, but we need to add more information - * for it to be useful (e.g. dump the context being - * executed). - */ - if (request) - rbuf = request->ctx->engine[ring->id].ringbuf; - else - rbuf = dev_priv->kernel_context->engine[ring->id].ringbuf; - } else - rbuf = ring->buffer; + error->simulated |= + request->ctx->flags & CONTEXT_NO_ERROR_CAPTURE; - error->ring[i].cpu_ring_head = rbuf->head; - error->ring[i].cpu_ring_tail = rbuf->tail; + rb = request->ringbuf; + ee->cpu_ring_head = rb->head; + ee->cpu_ring_tail = rb->tail; + ee->ringbuffer = + i915_error_ggtt_object_create(dev_priv, + rb->obj); + } - error->ring[i].ringbuffer = - i915_error_ggtt_object_create(dev_priv, rbuf->obj); + ee->hws_page = + i915_error_ggtt_object_create(dev_priv, + engine->status_page.obj); - error->ring[i].hws_page = - i915_error_ggtt_object_create(dev_priv, ring->status_page.obj); + if (engine->wa_ctx.obj) { + ee->wa_ctx = + i915_error_ggtt_object_create(dev_priv, + engine->wa_ctx.obj); + } - i915_gem_record_active_context(ring, error, &error->ring[i]); + i915_gem_record_active_context(engine, error, ee); count = 0; - list_for_each_entry(request, &ring->request_list, list) + list_for_each_entry(request, &engine->request_list, list) count++; - error->ring[i].num_requests = count; - error->ring[i].requests = - kcalloc(count, sizeof(*error->ring[i].requests), - GFP_ATOMIC); - if (error->ring[i].requests == NULL) { - error->ring[i].num_requests = 0; + ee->num_requests = count; + ee->requests = + kcalloc(count, sizeof(*ee->requests), GFP_ATOMIC); + if (!ee->requests) { + ee->num_requests = 0; continue; } count = 0; - list_for_each_entry(request, &ring->request_list, list) { + list_for_each_entry(request, &engine->request_list, list) { struct drm_i915_error_request *erq; - if (count >= error->ring[i].num_requests) { + if (count >= ee->num_requests) { /* * If the ring request list was changed in * between the point where the error request @@ -1105,8 +1186,8 @@ static void i915_gem_record_rings(struct drm_device *dev, break; } - erq = &error->ring[i].requests[count++]; - erq->seqno = request->seqno; + erq = &ee->requests[count++]; + erq->seqno = request->fence.seqno; erq->jiffies = request->emitted_jiffies; erq->tail = request->postfix; } @@ -1200,7 +1281,7 @@ static void i915_gem_capture_buffers(struct drm_i915_private *dev_priv, static void i915_capture_reg_state(struct drm_i915_private *dev_priv, struct drm_i915_error_state *error) { - struct drm_device *dev = dev_priv->dev; + struct drm_device *dev = &dev_priv->drm; int i; /* General organization @@ -1267,35 +1348,34 @@ static void i915_capture_reg_state(struct drm_i915_private *dev_priv, error->eir = I915_READ(EIR); error->pgtbl_er = I915_READ(PGTBL_ER); - i915_get_extra_instdone(dev, error->extra_instdone); + i915_get_extra_instdone(dev_priv, error->extra_instdone); } -static void i915_error_capture_msg(struct drm_device *dev, +static void i915_error_capture_msg(struct drm_i915_private *dev_priv, struct drm_i915_error_state *error, - bool wedged, + u32 engine_mask, const char *error_msg) { - struct drm_i915_private *dev_priv = dev->dev_private; u32 ecode; - int ring_id = -1, len; + int engine_id = -1, len; - ecode = i915_error_generate_code(dev_priv, error, &ring_id); + ecode = i915_error_generate_code(dev_priv, error, &engine_id); len = scnprintf(error->error_msg, sizeof(error->error_msg), "GPU HANG: ecode %d:%d:0x%08x", - INTEL_INFO(dev)->gen, ring_id, ecode); + INTEL_GEN(dev_priv), engine_id, ecode); - if (ring_id != -1 && error->ring[ring_id].pid != -1) + if (engine_id != -1 && error->engine[engine_id].pid != -1) len += scnprintf(error->error_msg + len, sizeof(error->error_msg) - len, ", in %s [%d]", - error->ring[ring_id].comm, - error->ring[ring_id].pid); + error->engine[engine_id].comm, + error->engine[engine_id].pid); scnprintf(error->error_msg + len, sizeof(error->error_msg) - len, ", reason: %s, action: %s", error_msg, - wedged ? "reset" : "continue"); + engine_mask ? "reset" : "continue"); } static void i915_capture_gen_state(struct drm_i915_private *dev_priv, @@ -1318,14 +1398,17 @@ static void i915_capture_gen_state(struct drm_i915_private *dev_priv, * out a structure which becomes available in debugfs for user level tools * to pick up. */ -void i915_capture_error_state(struct drm_device *dev, bool wedged, +void i915_capture_error_state(struct drm_i915_private *dev_priv, + u32 engine_mask, const char *error_msg) { static bool warned; - struct drm_i915_private *dev_priv = dev->dev_private; struct drm_i915_error_state *error; unsigned long flags; + if (READ_ONCE(dev_priv->gpu_error.first_error)) + return; + /* Account for pipe specific data like PIPE*STAT */ error = kzalloc(sizeof(*error), GFP_ATOMIC); if (!error) { @@ -1338,23 +1421,25 @@ void i915_capture_error_state(struct drm_device *dev, bool wedged, i915_capture_gen_state(dev_priv, error); i915_capture_reg_state(dev_priv, error); i915_gem_capture_buffers(dev_priv, error); - i915_gem_record_fences(dev, error); - i915_gem_record_rings(dev, error); + i915_gem_record_fences(dev_priv, error); + i915_gem_record_rings(dev_priv, error); do_gettimeofday(&error->time); - error->overlay = intel_overlay_capture_error_state(dev); - error->display = intel_display_capture_error_state(dev); + error->overlay = intel_overlay_capture_error_state(dev_priv); + error->display = intel_display_capture_error_state(dev_priv); - i915_error_capture_msg(dev, error, wedged, error_msg); + i915_error_capture_msg(dev_priv, error, engine_mask, error_msg); DRM_INFO("%s\n", error->error_msg); - spin_lock_irqsave(&dev_priv->gpu_error.lock, flags); - if (dev_priv->gpu_error.first_error == NULL) { - dev_priv->gpu_error.first_error = error; - error = NULL; + if (!error->simulated) { + spin_lock_irqsave(&dev_priv->gpu_error.lock, flags); + if (!dev_priv->gpu_error.first_error) { + dev_priv->gpu_error.first_error = error; + error = NULL; + } + spin_unlock_irqrestore(&dev_priv->gpu_error.lock, flags); } - spin_unlock_irqrestore(&dev_priv->gpu_error.lock, flags); if (error) { i915_error_state_free(&error->ref); @@ -1366,7 +1451,8 @@ void i915_capture_error_state(struct drm_device *dev, bool wedged, DRM_INFO("Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel\n"); DRM_INFO("drm/i915 developers can then reassign to the right component if it's not a kernel issue.\n"); DRM_INFO("The gpu crash dump is required to analyze gpu hangs, so please always attach it.\n"); - DRM_INFO("GPU crash dump saved to /sys/class/drm/card%d/error\n", dev->primary->index); + DRM_INFO("GPU crash dump saved to /sys/class/drm/card%d/error\n", + dev_priv->drm.primary->index); warned = true; } } @@ -1374,7 +1460,7 @@ void i915_capture_error_state(struct drm_device *dev, bool wedged, void i915_error_state_get(struct drm_device *dev, struct i915_error_state_file_priv *error_priv) { - struct drm_i915_private *dev_priv = dev->dev_private; + struct drm_i915_private *dev_priv = to_i915(dev); spin_lock_irq(&dev_priv->gpu_error.lock); error_priv->error = dev_priv->gpu_error.first_error; @@ -1392,7 +1478,7 @@ void i915_error_state_put(struct i915_error_state_file_priv *error_priv) void i915_destroy_error_state(struct drm_device *dev) { - struct drm_i915_private *dev_priv = dev->dev_private; + struct drm_i915_private *dev_priv = to_i915(dev); struct drm_i915_error_state *error; spin_lock_irq(&dev_priv->gpu_error.lock); @@ -1416,17 +1502,17 @@ const char *i915_cache_level_str(struct drm_i915_private *i915, int type) } /* NB: please notice the memset */ -void i915_get_extra_instdone(struct drm_device *dev, uint32_t *instdone) +void i915_get_extra_instdone(struct drm_i915_private *dev_priv, + uint32_t *instdone) { - struct drm_i915_private *dev_priv = dev->dev_private; memset(instdone, 0, sizeof(*instdone) * I915_NUM_INSTDONE_REG); - if (IS_GEN2(dev) || IS_GEN3(dev)) + if (IS_GEN2(dev_priv) || IS_GEN3(dev_priv)) instdone[0] = I915_READ(GEN2_INSTDONE); - else if (IS_GEN4(dev) || IS_GEN5(dev) || IS_GEN6(dev)) { + else if (IS_GEN4(dev_priv) || IS_GEN5(dev_priv) || IS_GEN6(dev_priv)) { instdone[0] = I915_READ(RING_INSTDONE(RENDER_RING_BASE)); instdone[1] = I915_READ(GEN4_INSTDONE1); - } else if (INTEL_INFO(dev)->gen >= 7) { + } else if (INTEL_GEN(dev_priv) >= 7) { instdone[0] = I915_READ(RING_INSTDONE(RENDER_RING_BASE)); instdone[1] = I915_READ(GEN7_SC_INSTDONE); instdone[2] = I915_READ(GEN7_SAMPLER_INSTDONE);