diff options
Diffstat (limited to 'drivers/gpu/drm/msm/adreno')
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a2xx_gpu.c | 3 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a3xx_gpu.c | 3 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a4xx_gpu.c | 3 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a5xx_debugfs.c | 4 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a5xx_gpu.c | 55 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 25 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 15 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a6xx_gpu.h | 10 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c | 104 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a6xx_hfi.c | 10 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a6xx_hfi.h | 11 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/adreno_device.c | 38 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/adreno_gpu.c | 4 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/adreno_gpu.h | 7 |
14 files changed, 223 insertions, 69 deletions
diff --git a/drivers/gpu/drm/msm/adreno/a2xx_gpu.c b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c index bdc989183c64..22e8295a5e2b 100644 --- a/drivers/gpu/drm/msm/adreno/a2xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c @@ -12,7 +12,6 @@ static bool a2xx_idle(struct msm_gpu *gpu); static void a2xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) { - struct msm_drm_private *priv = gpu->dev->dev_private; struct msm_ringbuffer *ring = submit->ring; unsigned int i; @@ -23,7 +22,7 @@ static void a2xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) break; case MSM_SUBMIT_CMD_CTX_RESTORE_BUF: /* ignore if there has not been a ctx switch: */ - if (priv->lastctx == submit->queue->ctx) + if (gpu->cur_ctx_seqno == submit->queue->ctx->seqno) break; fallthrough; case MSM_SUBMIT_CMD_BUF: diff --git a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c index 8fb847c174ff..2e481e2692ba 100644 --- a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c @@ -30,7 +30,6 @@ static bool a3xx_idle(struct msm_gpu *gpu); static void a3xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) { - struct msm_drm_private *priv = gpu->dev->dev_private; struct msm_ringbuffer *ring = submit->ring; unsigned int i; @@ -41,7 +40,7 @@ static void a3xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) break; case MSM_SUBMIT_CMD_CTX_RESTORE_BUF: /* ignore if there has not been a ctx switch: */ - if (priv->lastctx == submit->queue->ctx) + if (gpu->cur_ctx_seqno == submit->queue->ctx->seqno) break; fallthrough; case MSM_SUBMIT_CMD_BUF: diff --git a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c index a96ee79cc5e0..c5524d6e8705 100644 --- a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c @@ -24,7 +24,6 @@ static bool a4xx_idle(struct msm_gpu *gpu); static void a4xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) { - struct msm_drm_private *priv = gpu->dev->dev_private; struct msm_ringbuffer *ring = submit->ring; unsigned int i; @@ -35,7 +34,7 @@ static void a4xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) break; case MSM_SUBMIT_CMD_CTX_RESTORE_BUF: /* ignore if there has not been a ctx switch: */ - if (priv->lastctx == submit->queue->ctx) + if (gpu->cur_ctx_seqno == submit->queue->ctx->seqno) break; fallthrough; case MSM_SUBMIT_CMD_BUF: diff --git a/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c b/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c index dd593ec2bc56..6bd397a85834 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c @@ -107,7 +107,7 @@ reset_set(void *data, u64 val) * try to reset an active GPU. */ - mutex_lock(&dev->struct_mutex); + mutex_lock(&gpu->lock); release_firmware(adreno_gpu->fw[ADRENO_FW_PM4]); adreno_gpu->fw[ADRENO_FW_PM4] = NULL; @@ -133,7 +133,7 @@ reset_set(void *data, u64 val) gpu->funcs->recover(gpu); pm_runtime_put_sync(&gpu->pdev->dev); - mutex_unlock(&dev->struct_mutex); + mutex_unlock(&gpu->lock); return 0; } diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c index 5e2750eb3810..3d28fcf841a6 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c @@ -65,7 +65,6 @@ void a5xx_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring, static void a5xx_submit_in_rb(struct msm_gpu *gpu, struct msm_gem_submit *submit) { - struct msm_drm_private *priv = gpu->dev->dev_private; struct msm_ringbuffer *ring = submit->ring; struct msm_gem_object *obj; uint32_t *ptr, dwords; @@ -76,7 +75,7 @@ static void a5xx_submit_in_rb(struct msm_gpu *gpu, struct msm_gem_submit *submit case MSM_SUBMIT_CMD_IB_TARGET_BUF: break; case MSM_SUBMIT_CMD_CTX_RESTORE_BUF: - if (priv->lastctx == submit->queue->ctx) + if (gpu->cur_ctx_seqno == submit->queue->ctx->seqno) break; fallthrough; case MSM_SUBMIT_CMD_BUF: @@ -126,12 +125,11 @@ static void a5xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); - struct msm_drm_private *priv = gpu->dev->dev_private; struct msm_ringbuffer *ring = submit->ring; unsigned int i, ibs = 0; if (IS_ENABLED(CONFIG_DRM_MSM_GPU_SUDO) && submit->in_rb) { - priv->lastctx = NULL; + gpu->cur_ctx_seqno = 0; a5xx_submit_in_rb(gpu, submit); return; } @@ -166,7 +164,7 @@ static void a5xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) case MSM_SUBMIT_CMD_IB_TARGET_BUF: break; case MSM_SUBMIT_CMD_CTX_RESTORE_BUF: - if (priv->lastctx == submit->queue->ctx) + if (gpu->cur_ctx_seqno == submit->queue->ctx->seqno) break; fallthrough; case MSM_SUBMIT_CMD_BUF: @@ -441,7 +439,7 @@ void a5xx_set_hwcg(struct msm_gpu *gpu, bool state) const struct adreno_five_hwcg_regs *regs; unsigned int i, sz; - if (adreno_is_a508(adreno_gpu)) { + if (adreno_is_a506(adreno_gpu) || adreno_is_a508(adreno_gpu)) { regs = a50x_hwcg; sz = ARRAY_SIZE(a50x_hwcg); } else if (adreno_is_a509(adreno_gpu) || adreno_is_a512(adreno_gpu)) { @@ -485,7 +483,7 @@ static int a5xx_me_init(struct msm_gpu *gpu) OUT_RING(ring, 0x00000000); /* Specify workarounds for various microcode issues */ - if (adreno_is_a530(adreno_gpu)) { + if (adreno_is_a506(adreno_gpu) || adreno_is_a530(adreno_gpu)) { /* Workaround for token end syncs * Force a WFI after every direct-render 3D mode draw and every * 2D mode 3 draw @@ -620,8 +618,16 @@ static int a5xx_ucode_init(struct msm_gpu *gpu) static int a5xx_zap_shader_resume(struct msm_gpu *gpu) { + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); int ret; + /* + * Adreno 506 have CPZ Retention feature and doesn't require + * to resume zap shader + */ + if (adreno_is_a506(adreno_gpu)) + return 0; + ret = qcom_scm_set_remote_state(SCM_GPU_ZAP_SHADER_RESUME, GPU_PAS_ID); if (ret) DRM_ERROR("%s: zap-shader resume failed: %d\n", @@ -733,9 +739,10 @@ static int a5xx_hw_init(struct msm_gpu *gpu) 0x00100000 + adreno_gpu->gmem - 1); gpu_write(gpu, REG_A5XX_UCHE_GMEM_RANGE_MAX_HI, 0x00000000); - if (adreno_is_a508(adreno_gpu) || adreno_is_a510(adreno_gpu)) { + if (adreno_is_a506(adreno_gpu) || adreno_is_a508(adreno_gpu) || + adreno_is_a510(adreno_gpu)) { gpu_write(gpu, REG_A5XX_CP_MEQ_THRESHOLDS, 0x20); - if (adreno_is_a508(adreno_gpu)) + if (adreno_is_a506(adreno_gpu) || adreno_is_a508(adreno_gpu)) gpu_write(gpu, REG_A5XX_CP_MERCIU_SIZE, 0x400); else gpu_write(gpu, REG_A5XX_CP_MERCIU_SIZE, 0x20); @@ -751,7 +758,7 @@ static int a5xx_hw_init(struct msm_gpu *gpu) gpu_write(gpu, REG_A5XX_CP_ROQ_THRESHOLDS_1, 0x40201B16); } - if (adreno_is_a508(adreno_gpu)) + if (adreno_is_a506(adreno_gpu) || adreno_is_a508(adreno_gpu)) gpu_write(gpu, REG_A5XX_PC_DBG_ECO_CNTL, (0x100 << 11 | 0x100 << 22)); else if (adreno_is_a509(adreno_gpu) || adreno_is_a510(adreno_gpu) || @@ -769,8 +776,8 @@ static int a5xx_hw_init(struct msm_gpu *gpu) * Disable the RB sampler datapath DP2 clock gating optimization * for 1-SP GPUs, as it is enabled by default. */ - if (adreno_is_a508(adreno_gpu) || adreno_is_a509(adreno_gpu) || - adreno_is_a512(adreno_gpu)) + if (adreno_is_a506(adreno_gpu) || adreno_is_a508(adreno_gpu) || + adreno_is_a509(adreno_gpu) || adreno_is_a512(adreno_gpu)) gpu_rmw(gpu, REG_A5XX_RB_DBG_ECO_CNTL, 0, (1 << 9)); /* Disable UCHE global filter as SP can invalidate/flush independently */ @@ -851,10 +858,8 @@ static int a5xx_hw_init(struct msm_gpu *gpu) /* UCHE */ gpu_write(gpu, REG_A5XX_CP_PROTECT(16), ADRENO_PROTECT_RW(0xE80, 16)); - if (adreno_is_a508(adreno_gpu) || adreno_is_a509(adreno_gpu) || - adreno_is_a510(adreno_gpu) || adreno_is_a512(adreno_gpu) || - adreno_is_a530(adreno_gpu)) - gpu_write(gpu, REG_A5XX_CP_PROTECT(17), + /* SMMU */ + gpu_write(gpu, REG_A5XX_CP_PROTECT(17), ADRENO_PROTECT_RW(0x10000, 0x8000)); gpu_write(gpu, REG_A5XX_RBBM_SECVID_TSB_CNTL, 0); @@ -895,8 +900,7 @@ static int a5xx_hw_init(struct msm_gpu *gpu) if (ret) return ret; - if (!(adreno_is_a508(adreno_gpu) || adreno_is_a509(adreno_gpu) || - adreno_is_a510(adreno_gpu) || adreno_is_a512(adreno_gpu))) + if (adreno_is_a530(adreno_gpu) || adreno_is_a540(adreno_gpu)) a5xx_gpmu_ucode_init(gpu); ret = a5xx_ucode_init(gpu); @@ -927,6 +931,8 @@ static int a5xx_hw_init(struct msm_gpu *gpu) if (IS_ERR(a5xx_gpu->shadow)) return PTR_ERR(a5xx_gpu->shadow); + + msm_gem_object_set_name(a5xx_gpu->shadow_bo, "shadow"); } gpu_write64(gpu, REG_A5XX_CP_RB_RPTR_ADDR, @@ -1254,6 +1260,7 @@ static void a5xx_fault_detect_irq(struct msm_gpu *gpu) static irqreturn_t a5xx_irq(struct msm_gpu *gpu) { + struct msm_drm_private *priv = gpu->dev->dev_private; u32 status = gpu_read(gpu, REG_A5XX_RBBM_INT_0_STATUS); /* @@ -1263,6 +1270,11 @@ static irqreturn_t a5xx_irq(struct msm_gpu *gpu) gpu_write(gpu, REG_A5XX_RBBM_INT_CLEAR_CMD, status & ~A5XX_RBBM_INT_0_MASK_RBBM_AHB_ERROR); + if (priv->disable_err_irq) { + status &= A5XX_RBBM_INT_0_MASK_CP_CACHE_FLUSH_TS | + A5XX_RBBM_INT_0_MASK_CP_SW; + } + /* Pass status to a5xx_rbbm_err_irq because we've already cleared it */ if (status & RBBM_ERROR_MASK) a5xx_rbbm_err_irq(gpu, status); @@ -1338,7 +1350,7 @@ static int a5xx_pm_resume(struct msm_gpu *gpu) if (ret) return ret; - /* Adreno 508, 509, 510, 512 needs manual RBBM sus/res control */ + /* Adreno 506, 508, 509, 510, 512 needs manual RBBM sus/res control */ if (!(adreno_is_a530(adreno_gpu) || adreno_is_a540(adreno_gpu))) { /* Halt the sp_input_clk at HM level */ gpu_write(gpu, REG_A5XX_RBBM_CLOCK_CNTL, 0x00000055); @@ -1381,8 +1393,9 @@ static int a5xx_pm_suspend(struct msm_gpu *gpu) u32 mask = 0xf; int i, ret; - /* A508, A510 have 3 XIN ports in VBIF */ - if (adreno_is_a508(adreno_gpu) || adreno_is_a510(adreno_gpu)) + /* A506, A508, A510 have 3 XIN ports in VBIF */ + if (adreno_is_a506(adreno_gpu) || adreno_is_a508(adreno_gpu) || + adreno_is_a510(adreno_gpu)) mask = 0x7; /* Clear the VBIF pipe before shutting down */ diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c index 71e52b2b2025..3e325e2a2b1b 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c @@ -1146,7 +1146,7 @@ static void a6xx_gmu_memory_free(struct a6xx_gmu *gmu) } static int a6xx_gmu_memory_alloc(struct a6xx_gmu *gmu, struct a6xx_gmu_bo *bo, - size_t size, u64 iova) + size_t size, u64 iova, const char *name) { struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); struct drm_device *dev = a6xx_gpu->base.base.dev; @@ -1181,6 +1181,8 @@ static int a6xx_gmu_memory_alloc(struct a6xx_gmu *gmu, struct a6xx_gmu_bo *bo, bo->virt = msm_gem_get_vaddr(bo->obj); bo->size = size; + msm_gem_object_set_name(bo->obj, name); + return 0; } @@ -1515,7 +1517,8 @@ int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) */ gmu->dummy.size = SZ_4K; if (adreno_is_a660_family(adreno_gpu)) { - ret = a6xx_gmu_memory_alloc(gmu, &gmu->debug, SZ_4K * 7, 0x60400000); + ret = a6xx_gmu_memory_alloc(gmu, &gmu->debug, SZ_4K * 7, + 0x60400000, "debug"); if (ret) goto err_memory; @@ -1523,42 +1526,46 @@ int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) } /* Allocate memory for the GMU dummy page */ - ret = a6xx_gmu_memory_alloc(gmu, &gmu->dummy, gmu->dummy.size, 0x60000000); + ret = a6xx_gmu_memory_alloc(gmu, &gmu->dummy, gmu->dummy.size, + 0x60000000, "dummy"); if (ret) goto err_memory; + /* Note that a650 family also includes a660 family: */ if (adreno_is_a650_family(adreno_gpu)) { ret = a6xx_gmu_memory_alloc(gmu, &gmu->icache, - SZ_16M - SZ_16K, 0x04000); + SZ_16M - SZ_16K, 0x04000, "icache"); if (ret) goto err_memory; } else if (adreno_is_a640_family(adreno_gpu)) { ret = a6xx_gmu_memory_alloc(gmu, &gmu->icache, - SZ_256K - SZ_16K, 0x04000); + SZ_256K - SZ_16K, 0x04000, "icache"); if (ret) goto err_memory; ret = a6xx_gmu_memory_alloc(gmu, &gmu->dcache, - SZ_256K - SZ_16K, 0x44000); + SZ_256K - SZ_16K, 0x44000, "dcache"); if (ret) goto err_memory; } else { + BUG_ON(adreno_is_a660_family(adreno_gpu)); + /* HFI v1, has sptprac */ gmu->legacy = true; /* Allocate memory for the GMU debug region */ - ret = a6xx_gmu_memory_alloc(gmu, &gmu->debug, SZ_16K, 0); + ret = a6xx_gmu_memory_alloc(gmu, &gmu->debug, SZ_16K, 0, "debug"); if (ret) goto err_memory; } /* Allocate memory for for the HFI queues */ - ret = a6xx_gmu_memory_alloc(gmu, &gmu->hfi, SZ_16K, 0); + ret = a6xx_gmu_memory_alloc(gmu, &gmu->hfi, SZ_16K, 0, "hfi"); if (ret) goto err_memory; /* Allocate memory for the GMU log region */ - ret = a6xx_gmu_memory_alloc(gmu, &gmu->log, SZ_4K, 0); + ret = a6xx_gmu_memory_alloc(gmu, &gmu->log, SZ_4K, 0, "log"); if (ret) goto err_memory; diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c index 78aad5216a61..51b83776951b 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c @@ -106,7 +106,7 @@ static void a6xx_set_pagetable(struct a6xx_gpu *a6xx_gpu, u32 asid; u64 memptr = rbmemptr(ring, ttbr0); - if (ctx->seqno == a6xx_gpu->cur_ctx_seqno) + if (ctx->seqno == a6xx_gpu->base.base.cur_ctx_seqno) return; if (msm_iommu_pagetable_params(ctx->aspace->mmu, &ttbr, &asid)) @@ -138,14 +138,11 @@ static void a6xx_set_pagetable(struct a6xx_gpu *a6xx_gpu, OUT_PKT7(ring, CP_EVENT_WRITE, 1); OUT_RING(ring, 0x31); - - a6xx_gpu->cur_ctx_seqno = ctx->seqno; } static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) { unsigned int index = submit->seqno % MSM_GPU_SUBMIT_STATS_COUNT; - struct msm_drm_private *priv = gpu->dev->dev_private; struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); struct msm_ringbuffer *ring = submit->ring; @@ -177,7 +174,7 @@ static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) case MSM_SUBMIT_CMD_IB_TARGET_BUF: break; case MSM_SUBMIT_CMD_CTX_RESTORE_BUF: - if (priv->lastctx == submit->queue->ctx) + if (gpu->cur_ctx_seqno == submit->queue->ctx->seqno) break; fallthrough; case MSM_SUBMIT_CMD_BUF: @@ -1071,6 +1068,8 @@ static int hw_init(struct msm_gpu *gpu) if (IS_ERR(a6xx_gpu->shadow)) return PTR_ERR(a6xx_gpu->shadow); + + msm_gem_object_set_name(a6xx_gpu->shadow_bo, "shadow"); } gpu_write64(gpu, REG_A6XX_CP_RB_RPTR_ADDR_LO, @@ -1081,7 +1080,7 @@ static int hw_init(struct msm_gpu *gpu) /* Always come up on rb 0 */ a6xx_gpu->cur_ring = gpu->rb[0]; - a6xx_gpu->cur_ctx_seqno = 0; + gpu->cur_ctx_seqno = 0; /* Enable the SQE_to start the CP engine */ gpu_write(gpu, REG_A6XX_CP_SQE_CNTL, 1); @@ -1376,10 +1375,14 @@ static void a6xx_fault_detect_irq(struct msm_gpu *gpu) static irqreturn_t a6xx_irq(struct msm_gpu *gpu) { + struct msm_drm_private *priv = gpu->dev->dev_private; u32 status = gpu_read(gpu, REG_A6XX_RBBM_INT_0_STATUS); gpu_write(gpu, REG_A6XX_RBBM_INT_CLEAR_CMD, status); + if (priv->disable_err_irq) + status &= A6XX_RBBM_INT_0_MASK_CP_CACHE_FLUSH_TS; + if (status & A6XX_RBBM_INT_0_MASK_RBBM_HANG_DETECT) a6xx_fault_detect_irq(gpu); diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h index 8e5527c881b1..86e0a7c3fe6d 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h @@ -20,16 +20,6 @@ struct a6xx_gpu { struct msm_ringbuffer *cur_ring; - /** - * cur_ctx_seqno: - * - * The ctx->seqno value of the context with current pgtables - * installed. Tracked by seqno rather than pointer value to - * avoid dangling pointers, and cases where a ctx can be freed - * and a new one created with the same address. - */ - int cur_ctx_seqno; - struct a6xx_gmu gmu; struct drm_gem_object *shadow_bo; diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c index 6e90209cd543..55f443328d8e 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c @@ -42,7 +42,15 @@ struct a6xx_gpu_state { struct a6xx_gpu_state_obj *cx_debugbus; int nr_cx_debugbus; + struct msm_gpu_state_bo *gmu_log; + struct msm_gpu_state_bo *gmu_hfi; + struct msm_gpu_state_bo *gmu_debug; + + s32 hfi_queue_history[2][HFI_HISTORY_SZ]; + struct list_head objs; + + bool gpu_initialized; }; static inline int CRASHDUMP_WRITE(u64 *in, u32 reg, u32 val) @@ -800,6 +808,45 @@ static void a6xx_get_gmu_registers(struct msm_gpu *gpu, &a6xx_state->gmu_registers[2], false); } +static struct msm_gpu_state_bo *a6xx_snapshot_gmu_bo( + struct a6xx_gpu_state *a6xx_state, struct a6xx_gmu_bo *bo) +{ + struct msm_gpu_state_bo *snapshot; + + snapshot = state_kcalloc(a6xx_state, 1, sizeof(*snapshot)); + if (!snapshot) + return NULL; + + snapshot->iova = bo->iova; + snapshot->size = bo->size; + snapshot->data = kvzalloc(snapshot->size, GFP_KERNEL); + if (!snapshot->data) + return NULL; + + memcpy(snapshot->data, bo->virt, bo->size); + + return snapshot; +} + +static void a6xx_snapshot_gmu_hfi_history(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + unsigned i, j; + + BUILD_BUG_ON(ARRAY_SIZE(gmu->queues) != ARRAY_SIZE(a6xx_state->hfi_queue_history)); + + for (i = 0; i < ARRAY_SIZE(gmu->queues); i++) { + struct a6xx_hfi_queue *queue = &gmu->queues[i]; + for (j = 0; j < HFI_HISTORY_SZ; j++) { + unsigned idx = (j + queue->history_idx) % HFI_HISTORY_SZ; + a6xx_state->hfi_queue_history[i][j] = queue->history[idx]; + } + } +} + #define A6XX_GBIF_REGLIST_SIZE 1 static void a6xx_get_registers(struct msm_gpu *gpu, struct a6xx_gpu_state *a6xx_state, @@ -937,6 +984,12 @@ struct msm_gpu_state *a6xx_gpu_state_get(struct msm_gpu *gpu) a6xx_get_gmu_registers(gpu, a6xx_state); + a6xx_state->gmu_log = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.log); + a6xx_state->gmu_hfi = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.hfi); + a6xx_state->gmu_debug = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.debug); + + a6xx_snapshot_gmu_hfi_history(gpu, a6xx_state); + /* If GX isn't on the rest of the data isn't going to be accessible */ if (!a6xx_gmu_gx_is_on(&a6xx_gpu->gmu)) return &a6xx_state->base; @@ -950,7 +1003,8 @@ struct msm_gpu_state *a6xx_gpu_state_get(struct msm_gpu *gpu) * write out GPU state, so we need to skip this when the SMMU is * stalled in response to an iova fault */ - if (!stalled && !a6xx_crashdumper_init(gpu, &_dumper)) { + if (!stalled && !gpu->needs_hw_init && + !a6xx_crashdumper_init(gpu, &_dumper)) { dumper = &_dumper; } @@ -967,6 +1021,8 @@ struct msm_gpu_state *a6xx_gpu_state_get(struct msm_gpu *gpu) if (snapshot_debugbus) a6xx_get_debugbus(gpu, a6xx_state); + a6xx_state->gpu_initialized = !gpu->needs_hw_init; + return &a6xx_state->base; } @@ -978,6 +1034,12 @@ static void a6xx_gpu_state_destroy(struct kref *kref) struct a6xx_gpu_state *a6xx_state = container_of(state, struct a6xx_gpu_state, base); + if (a6xx_state->gmu_log) + kvfree(a6xx_state->gmu_log->data); + + if (a6xx_state->gmu_hfi) + kvfree(a6xx_state->gmu_hfi->data); + list_for_each_entry_safe(obj, tmp, &a6xx_state->objs, node) kfree(obj); @@ -1189,8 +1251,48 @@ void a6xx_show(struct msm_gpu *gpu, struct msm_gpu_state *state, if (IS_ERR_OR_NULL(state)) return; + drm_printf(p, "gpu-initialized: %d\n", a6xx_state->gpu_initialized); + adreno_show(gpu, state, p); + drm_puts(p, "gmu-log:\n"); + if (a6xx_state->gmu_log) { + struct msm_gpu_state_bo *gmu_log = a6xx_state->gmu_log; + + drm_printf(p, " iova: 0x%016llx\n", gmu_log->iova); + drm_printf(p, " size: %zu\n", gmu_log->size); + adreno_show_object(p, &gmu_log->data, gmu_log->size, + &gmu_log->encoded); + } + + drm_puts(p, "gmu-hfi:\n"); + if (a6xx_state->gmu_hfi) { + struct msm_gpu_state_bo *gmu_hfi = a6xx_state->gmu_hfi; + unsigned i, j; + + drm_printf(p, " iova: 0x%016llx\n", gmu_hfi->iova); + drm_printf(p, " size: %zu\n", gmu_hfi->size); + for (i = 0; i < ARRAY_SIZE(a6xx_state->hfi_queue_history); i++) { + drm_printf(p, " queue-history[%u]:", i); + for (j = 0; j < HFI_HISTORY_SZ; j++) { + drm_printf(p, " %d", a6xx_state->hfi_queue_history[i][j]); + } + drm_printf(p, "\n"); + } + adreno_show_object(p, &gmu_hfi->data, gmu_hfi->size, + &gmu_hfi->encoded); + } + + drm_puts(p, "gmu-debug:\n"); + if (a6xx_state->gmu_debug) { + struct msm_gpu_state_bo *gmu_debug = a6xx_state->gmu_debug; + + drm_printf(p, " iova: 0x%016llx\n", gmu_debug->iova); + drm_printf(p, " size: %zu\n", gmu_debug->size); + adreno_show_object(p, &gmu_debug->data, gmu_debug->size, + &gmu_debug->encoded); + } + drm_puts(p, "registers:\n"); for (i = 0; i < a6xx_state->nr_registers; i++) { struct a6xx_gpu_state_obj *obj = &a6xx_state->registers[i]; diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c index d4c65bf0a1b7..d73fce5fdf1f 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c @@ -36,6 +36,8 @@ static int a6xx_hfi_queue_read(struct a6xx_gmu *gmu, hdr = queue->data[index]; + queue->history[(queue->history_idx++) % HFI_HISTORY_SZ] = index; + /* * If we are to assume that the GMU firmware is in fact a rational actor * and is programmed to not send us a larger response than we expect @@ -75,6 +77,8 @@ static int a6xx_hfi_queue_write(struct a6xx_gmu *gmu, return -ENOSPC; } + queue->history[(queue->history_idx++) % HFI_HISTORY_SZ] = index; + for (i = 0; i < dwords; i++) { queue->data[index] = data[i]; index = (index + 1) % header->size; @@ -600,6 +604,9 @@ void a6xx_hfi_stop(struct a6xx_gmu *gmu) queue->header->read_index = 0; queue->header->write_index = 0; + + memset(&queue->history, 0xff, sizeof(queue->history)); + queue->history_idx = 0; } } @@ -612,6 +619,9 @@ static void a6xx_hfi_queue_init(struct a6xx_hfi_queue *queue, queue->data = virt; atomic_set(&queue->seqnum, 0); + memset(&queue->history, 0xff, sizeof(queue->history)); + queue->history_idx = 0; + /* Set up the shared memory header */ header->iova = iova; header->type = 10 << 8 | id; diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.h b/drivers/gpu/drm/msm/adreno/a6xx_hfi.h index 2bd670ca42d6..528110169398 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_hfi.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.h @@ -33,6 +33,17 @@ struct a6xx_hfi_queue { spinlock_t lock; u32 *data; atomic_t seqnum; + + /* + * Tracking for the start index of the last N messages in the + * queue, for the benefit of devcore dump / crashdec (since + * parsing in the reverse direction to decode the last N + * messages is difficult to do and would rely on heuristics + * which are not guaranteed to be correct) + */ +#define HFI_HISTORY_SZ 8 + s32 history[HFI_HISTORY_SZ]; + u8 history_idx; }; /* This is the outgoing queue to the GMU */ diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c index 2a6ce76656aa..93005839b5da 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_device.c +++ b/drivers/gpu/drm/msm/adreno/adreno_device.c @@ -132,6 +132,24 @@ static const struct adreno_info gpulist[] = { .inactive_period = DRM_MSM_INACTIVE_PERIOD, .init = a4xx_gpu_init, }, { + .rev = ADRENO_REV(5, 0, 6, ANY_ID), + .revn = 506, + .name = "A506", + .fw = { + [ADRENO_FW_PM4] = "a530_pm4.fw", + [ADRENO_FW_PFP] = "a530_pfp.fw", + }, + .gmem = (SZ_128K + SZ_8K), + /* + * Increase inactive period to 250 to avoid bouncing + * the GDSC which appears to make it grumpy + */ + .inactive_period = 250, + .quirks = ADRENO_QUIRK_TWO_PASS_USE_WFI | + ADRENO_QUIRK_LMLOADKILL_DISABLE, + .init = a5xx_gpu_init, + .zapfw = "a506_zap.mdt", + }, { .rev = ADRENO_REV(5, 0, 8, ANY_ID), .revn = 508, .name = "A508", @@ -408,9 +426,9 @@ struct msm_gpu *adreno_load_gpu(struct drm_device *dev) return NULL; } - mutex_lock(&dev->struct_mutex); + mutex_lock(&gpu->lock); ret = msm_gpu_hw_init(gpu); - mutex_unlock(&dev->struct_mutex); + mutex_unlock(&gpu->lock); pm_runtime_put_autosuspend(&pdev->dev); if (ret) { DRM_DEV_ERROR(dev->dev, "gpu hw init failed: %d\n", ret); @@ -427,13 +445,6 @@ struct msm_gpu *adreno_load_gpu(struct drm_device *dev) return gpu; } -static void set_gpu_pdev(struct drm_device *dev, - struct platform_device *pdev) -{ - struct msm_drm_private *priv = dev->dev_private; - priv->gpu_pdev = pdev; -} - static int find_chipid(struct device *dev, struct adreno_rev *rev) { struct device_node *node = dev->of_node; @@ -482,8 +493,8 @@ static int adreno_bind(struct device *dev, struct device *master, void *data) { static struct adreno_platform_config config = {}; const struct adreno_info *info; - struct drm_device *drm = dev_get_drvdata(master); - struct msm_drm_private *priv = drm->dev_private; + struct msm_drm_private *priv = dev_get_drvdata(master); + struct drm_device *drm = priv->dev; struct msm_gpu *gpu; int ret; @@ -492,7 +503,7 @@ static int adreno_bind(struct device *dev, struct device *master, void *data) return ret; dev->platform_data = &config; - set_gpu_pdev(drm, to_platform_device(dev)); + priv->gpu_pdev = to_platform_device(dev); info = adreno_info(config.rev); @@ -521,12 +532,13 @@ static int adreno_bind(struct device *dev, struct device *master, void *data) static void adreno_unbind(struct device *dev, struct device *master, void *data) { + struct msm_drm_private *priv = dev_get_drvdata(master); struct msm_gpu *gpu = dev_to_gpu(dev); pm_runtime_force_suspend(dev); gpu->funcs->destroy(gpu); - set_gpu_pdev(dev_get_drvdata(master), NULL); + priv->gpu_pdev = NULL; } static const struct component_ops a3xx_ops = { diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index 748665232d29..f33cfa4ef1c8 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -504,6 +504,8 @@ int adreno_gpu_state_get(struct msm_gpu *gpu, struct msm_gpu_state *state) struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); int i, count = 0; + WARN_ON(!mutex_is_locked(&gpu->lock)); + kref_init(&state->ref); ktime_get_real_ts64(&state->time); @@ -630,7 +632,7 @@ static char *adreno_gpu_ascii85_encode(u32 *src, size_t len) } /* len is expected to be in bytes */ -static void adreno_show_object(struct drm_printer *p, void **ptr, int len, +void adreno_show_object(struct drm_printer *p, void **ptr, int len, bool *encoded) { if (!*ptr || !len) diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h index 225c277a6223..cffabe7d33c1 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h @@ -201,6 +201,11 @@ static inline int adreno_is_a430(struct adreno_gpu *gpu) return gpu->revn == 430; } +static inline int adreno_is_a506(struct adreno_gpu *gpu) +{ + return gpu->revn == 506; +} + static inline int adreno_is_a508(struct adreno_gpu *gpu) { return gpu->revn == 508; @@ -306,6 +311,8 @@ void adreno_gpu_state_destroy(struct msm_gpu_state *state); int adreno_gpu_state_get(struct msm_gpu *gpu, struct msm_gpu_state *state); int adreno_gpu_state_put(struct msm_gpu_state *state); +void adreno_show_object(struct drm_printer *p, void **ptr, int len, + bool *encoded); /* * Common helper function to initialize the default address space for arm-smmu |