diff options
Diffstat (limited to 'drivers/gpu/drm/msm/adreno')
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a2xx_gpu.c | 5 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a3xx_gpu.c | 10 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a4xx_gpu.c | 10 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a5xx_gpu.c | 47 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 38 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 14 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c | 2 | ||||
| -rw-r--r-- | drivers/gpu/drm/msm/adreno/adreno_gpu.c | 29 |
8 files changed, 109 insertions, 46 deletions
diff --git a/drivers/gpu/drm/msm/adreno/a2xx_gpu.c b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c index 1f83bc18d500..80f3b1da9fc2 100644 --- a/drivers/gpu/drm/msm/adreno/a2xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c @@ -164,6 +164,11 @@ static int a2xx_hw_init(struct msm_gpu *gpu) if (ret) return ret; + gpu_write(gpu, REG_AXXX_CP_RB_CNTL, + MSM_GPU_RB_CNTL_DEFAULT | AXXX_CP_RB_CNTL_NO_UPDATE); + + gpu_write(gpu, REG_AXXX_CP_RB_BASE, lower_32_bits(gpu->rb[0]->iova)); + /* NOTE: PM4/micro-engine firmware registers look to be the same * for a2xx and a3xx.. we could possibly push that part down to * adreno_gpu base class. Or push both PM4 and PFP but diff --git a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c index 5f7e98028eaf..eeba2deeca1e 100644 --- a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c @@ -215,6 +215,16 @@ static int a3xx_hw_init(struct msm_gpu *gpu) if (ret) return ret; + /* + * Use the default ringbuffer size and block size but disable the RPTR + * shadow + */ + gpu_write(gpu, REG_AXXX_CP_RB_CNTL, + MSM_GPU_RB_CNTL_DEFAULT | AXXX_CP_RB_CNTL_NO_UPDATE); + + /* Set the ringbuffer address */ + gpu_write(gpu, REG_AXXX_CP_RB_BASE, lower_32_bits(gpu->rb[0]->iova)); + /* setup access protection: */ gpu_write(gpu, REG_A3XX_CP_PROTECT_CTRL, 0x00000007); diff --git a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c index ab2b752566d8..05cfa81d4c54 100644 --- a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c @@ -265,6 +265,16 @@ static int a4xx_hw_init(struct msm_gpu *gpu) if (ret) return ret; + /* + * Use the default ringbuffer size and block size but disable the RPTR + * shadow + */ + gpu_write(gpu, REG_A4XX_CP_RB_CNTL, + MSM_GPU_RB_CNTL_DEFAULT | AXXX_CP_RB_CNTL_NO_UPDATE); + + /* Set the ringbuffer address */ + gpu_write(gpu, REG_A4XX_CP_RB_BASE, lower_32_bits(gpu->rb[0]->iova)); + /* Load PM4: */ ptr = (uint32_t *)(adreno_gpu->fw[ADRENO_FW_PM4]->data); len = adreno_gpu->fw[ADRENO_FW_PM4]->size / 4; diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c index 99cd6e62a971..c8fb21cc0d6f 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c @@ -677,14 +677,21 @@ static int a5xx_hw_init(struct msm_gpu *gpu) if (ret) return ret; - a5xx_preempt_hw_init(gpu); - a5xx_gpmu_ucode_init(gpu); ret = a5xx_ucode_init(gpu); if (ret) return ret; + /* Set the ringbuffer address */ + gpu_write64(gpu, REG_A5XX_CP_RB_BASE, REG_A5XX_CP_RB_BASE_HI, + gpu->rb[0]->iova); + + gpu_write(gpu, REG_A5XX_CP_RB_CNTL, + MSM_GPU_RB_CNTL_DEFAULT | AXXX_CP_RB_CNTL_NO_UPDATE); + + a5xx_preempt_hw_init(gpu); + /* Disable the interrupts through the initial bringup stage */ gpu_write(gpu, REG_A5XX_RBBM_INT_0_MASK, A5XX_INT_MASK); @@ -1359,6 +1366,10 @@ static unsigned long a5xx_gpu_busy(struct msm_gpu *gpu) { u64 busy_cycles, busy_time; + /* Only read the gpu busy if the hardware is already active */ + if (pm_runtime_get_if_in_use(&gpu->pdev->dev) == 0) + return 0; + busy_cycles = gpu_read64(gpu, REG_A5XX_RBBM_PERFCTR_RBBM_0_LO, REG_A5XX_RBBM_PERFCTR_RBBM_0_HI); @@ -1367,6 +1378,8 @@ static unsigned long a5xx_gpu_busy(struct msm_gpu *gpu) gpu->devfreq.busy_cycles = busy_cycles; + pm_runtime_put(&gpu->pdev->dev); + if (WARN_ON(busy_time > ~0LU)) return ~0LU; @@ -1401,18 +1414,31 @@ static const struct adreno_gpu_funcs funcs = { static void check_speed_bin(struct device *dev) { struct nvmem_cell *cell; - u32 bin, val; + u32 val; + + /* + * If the OPP table specifies a opp-supported-hw property then we have + * to set something with dev_pm_opp_set_supported_hw() or the table + * doesn't get populated so pick an arbitrary value that should + * ensure the default frequencies are selected but not conflict with any + * actual bins + */ + val = 0x80; cell = nvmem_cell_get(dev, "speed_bin"); - /* If a nvmem cell isn't defined, nothing to do */ - if (IS_ERR(cell)) - return; + if (!IS_ERR(cell)) { + void *buf = nvmem_cell_read(cell, NULL); + + if (!IS_ERR(buf)) { + u8 bin = *((u8 *) buf); - bin = *((u32 *) nvmem_cell_read(cell, NULL)); - nvmem_cell_put(cell); + val = (1 << bin); + kfree(buf); + } - val = (1 << bin); + nvmem_cell_put(cell); + } dev_pm_opp_set_supported_hw(dev, &val, 1); } @@ -1445,7 +1471,8 @@ struct msm_gpu *a5xx_gpu_init(struct drm_device *dev) check_speed_bin(&pdev->dev); - ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 4); + /* Restricting nr_rings to 1 to temporarily disable preemption */ + ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 1); if (ret) { a5xx_destroy(&(a5xx_gpu->base.base)); return ERR_PTR(ret); diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c index 85f14feafdec..40431a09dc97 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c @@ -107,6 +107,13 @@ static void __a6xx_gmu_set_freq(struct a6xx_gmu *gmu, int index) struct msm_gpu *gpu = &adreno_gpu->base; int ret; + /* + * This can get called from devfreq while the hardware is idle. Don't + * bring up the power if it isn't already active + */ + if (pm_runtime_get_if_in_use(gmu->dev) == 0) + return; + gmu_write(gmu, REG_A6XX_GMU_DCVS_ACK_OPTION, 0); gmu_write(gmu, REG_A6XX_GMU_DCVS_PERF_SETTING, @@ -133,6 +140,7 @@ static void __a6xx_gmu_set_freq(struct a6xx_gmu *gmu, int index) * for now leave it at max so that the performance is nominal. */ icc_set_bw(gpu->icc_path, 0, MBps_to_icc(7216)); + pm_runtime_put(gmu->dev); } void a6xx_gmu_set_freq(struct msm_gpu *gpu, unsigned long freq) @@ -191,12 +199,22 @@ static int a6xx_gmu_start(struct a6xx_gmu *gmu) { int ret; u32 val; + u32 mask, reset_val; + + val = gmu_read(gmu, REG_A6XX_GMU_CM3_DTCM_START + 0xff8); + if (val <= 0x20010004) { + mask = 0xffffffff; + reset_val = 0xbabeface; + } else { + mask = 0x1ff; + reset_val = 0x100; + } gmu_write(gmu, REG_A6XX_GMU_CM3_SYSRESET, 1); gmu_write(gmu, REG_A6XX_GMU_CM3_SYSRESET, 0); ret = gmu_poll_timeout(gmu, REG_A6XX_GMU_CM3_FW_INIT_RESULT, val, - val == 0xbabeface, 100, 10000); + (val & mask) == reset_val, 100, 10000); if (ret) DRM_DEV_ERROR(gmu->dev, "GMU firmware initialization timed out\n"); @@ -705,10 +723,19 @@ int a6xx_gmu_resume(struct a6xx_gpu *a6xx_gpu) /* Turn on the resources */ pm_runtime_get_sync(gmu->dev); + /* + * "enable" the GX power domain which won't actually do anything but it + * will make sure that the refcounting is correct in case we need to + * bring down the GX after a GMU failure + */ + if (!IS_ERR_OR_NULL(gmu->gxpd)) + pm_runtime_get_sync(gmu->gxpd); + /* Use a known rate to bring up the GMU */ clk_set_rate(gmu->core_clk, 200000000); ret = clk_bulk_prepare_enable(gmu->nr_clocks, gmu->clocks); if (ret) { + pm_runtime_put(gmu->gxpd); pm_runtime_put(gmu->dev); return ret; } @@ -744,19 +771,12 @@ int a6xx_gmu_resume(struct a6xx_gpu *a6xx_gpu) /* Set the GPU to the highest power frequency */ __a6xx_gmu_set_freq(gmu, gmu->nr_gpu_freqs - 1); - /* - * "enable" the GX power domain which won't actually do anything but it - * will make sure that the refcounting is correct in case we need to - * bring down the GX after a GMU failure - */ - if (!IS_ERR_OR_NULL(gmu->gxpd)) - pm_runtime_get(gmu->gxpd); - out: /* On failure, shut down the GMU to leave it in a good state */ if (ret) { disable_irq(gmu->gmu_irq); a6xx_rpmh_stop(gmu); + pm_runtime_put(gmu->gxpd); pm_runtime_put(gmu->dev); } diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c index 686c34d706b0..ab75f0309d4b 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c @@ -512,6 +512,13 @@ static int a6xx_hw_init(struct msm_gpu *gpu) if (ret) goto out; + /* Set the ringbuffer address */ + gpu_write64(gpu, REG_A6XX_CP_RB_BASE, REG_A6XX_CP_RB_BASE_HI, + gpu->rb[0]->iova); + + gpu_write(gpu, REG_A6XX_CP_RB_CNTL, + MSM_GPU_RB_CNTL_DEFAULT | AXXX_CP_RB_CNTL_NO_UPDATE); + /* Always come up on rb 0 */ a6xx_gpu->cur_ring = gpu->rb[0]; @@ -803,6 +810,11 @@ static unsigned long a6xx_gpu_busy(struct msm_gpu *gpu) struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); u64 busy_cycles, busy_time; + + /* Only read the gpu busy if the hardware is already active */ + if (pm_runtime_get_if_in_use(a6xx_gpu->gmu.dev) == 0) + return 0; + busy_cycles = gmu_read64(&a6xx_gpu->gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_XOCLK_0_L, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_XOCLK_0_H); @@ -812,6 +824,8 @@ static unsigned long a6xx_gpu_busy(struct msm_gpu *gpu) gpu->devfreq.busy_cycles = busy_cycles; + pm_runtime_put(a6xx_gpu->gmu.dev); + if (WARN_ON(busy_time > ~0LU)) return ~0LU; diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c index 691c1a277d91..dfcbb2b7cdda 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c @@ -834,7 +834,7 @@ static void a6xx_get_indexed_registers(struct msm_gpu *gpu, int i; a6xx_state->indexed_regs = state_kcalloc(a6xx_state, count, - sizeof(a6xx_state->indexed_regs)); + sizeof(*a6xx_state->indexed_regs)); if (!a6xx_state->indexed_regs) return; diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index 048c8be426f3..3802ad38c519 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -350,30 +350,10 @@ int adreno_hw_init(struct msm_gpu *gpu) ring->next = ring->start; /* reset completed fence seqno: */ - ring->memptrs->fence = ring->seqno; + ring->memptrs->fence = ring->fctx->completed_fence; ring->memptrs->rptr = 0; } - /* - * Setup REG_CP_RB_CNTL. The same value is used across targets (with - * the excpetion of A430 that disables the RPTR shadow) - the cacluation - * for the ringbuffer size and block size is moved to msm_gpu.h for the - * pre-processor to deal with and the A430 variant is ORed in here - */ - adreno_gpu_write(adreno_gpu, REG_ADRENO_CP_RB_CNTL, - MSM_GPU_RB_CNTL_DEFAULT | - (adreno_is_a430(adreno_gpu) ? AXXX_CP_RB_CNTL_NO_UPDATE : 0)); - - /* Setup ringbuffer address - use ringbuffer[0] for GPU init */ - adreno_gpu_write64(adreno_gpu, REG_ADRENO_CP_RB_BASE, - REG_ADRENO_CP_RB_BASE_HI, gpu->rb[0]->iova); - - if (!adreno_is_a430(adreno_gpu)) { - adreno_gpu_write64(adreno_gpu, REG_ADRENO_CP_RB_RPTR_ADDR, - REG_ADRENO_CP_RB_RPTR_ADDR_HI, - rbmemptr(gpu->rb[0], rptr)); - } - return 0; } @@ -381,11 +361,8 @@ int adreno_hw_init(struct msm_gpu *gpu) static uint32_t get_rptr(struct adreno_gpu *adreno_gpu, struct msm_ringbuffer *ring) { - if (adreno_is_a430(adreno_gpu)) - return ring->memptrs->rptr = adreno_gpu_read( - adreno_gpu, REG_ADRENO_CP_RB_RPTR); - else - return ring->memptrs->rptr; + return ring->memptrs->rptr = adreno_gpu_read( + adreno_gpu, REG_ADRENO_CP_RB_RPTR); } struct msm_ringbuffer *adreno_active_ring(struct msm_gpu *gpu) |