From bcf1d9fa5d03c1e86733411b5dc3a76ba7d75e6e Mon Sep 17 00:00:00 2001 From: Jordan Crouse Date: Tue, 24 Jul 2018 10:33:28 -0600 Subject: drm/msm/adreno: Convert the show/crash file format Convert the format of the 'show' debugfs file and the crash dump to a format resembling YAML. This should be easier to parse and be more flexible for future changes and expansions. v2: Use a standard .rst for the msm crashdump documentation Signed-off-by: Jordan Crouse Signed-off-by: Rob Clark --- Documentation/gpu/msm-crash-dump.rst | 71 ++++++++++++++++++++++++++++++++++++ 1 file changed, 71 insertions(+) create mode 100644 Documentation/gpu/msm-crash-dump.rst (limited to 'Documentation/gpu/msm-crash-dump.rst') diff --git a/Documentation/gpu/msm-crash-dump.rst b/Documentation/gpu/msm-crash-dump.rst new file mode 100644 index 000000000000..75ab1d541c03 --- /dev/null +++ b/Documentation/gpu/msm-crash-dump.rst @@ -0,0 +1,71 @@ +===================== +MSM Crash Dump Format +===================== + +Following a GPU hang the MSM driver outputs debugging information via +/sys/kernel/dri/X/show or via devcoredump (/sys/class/devcoredump/dcdX/data). +This document describes how the output is formatted. + +Each entry is in the form key: value. Sections headers will not have a value +and all the contents of a section will be indented two spaces from the header. +Each section might have multiple array entries the start of which is designated +by a (-). + +Mappings +-------- + +kernel + The kernel version that generated the dump (UTS_RELEASE). + +module + The module that generated the crashdump. + +time + The kernel time at crash formated as seconds.microseconds. + +comm + Comm string for the binary that generated the fault. + +cmdline + Command line for the binary that generated the fault. + +revision + ID of the GPU that generated the crash formatted as + core.major.minor.patchlevel separated by dots. + +rbbm-status + The current value of RBBM_STATUS which shows what top level GPU + components are in use at the time of crash. + +ringbuffer + Section containing the contents of each ringbuffer. Each ringbuffer is + identified with an id number. + + id + Ringbuffer ID (0 based index). Each ringbuffer in the section + will have its own unique id. + iova + GPU address of the ringbuffer. + + last-fence + The last fence that was issued on the ringbuffer + + retired-fence + The last fence retired on the ringbuffer. + + rptr + The current read pointer (rptr) for the ringbuffer. + + wptr + The current write pointer (wptr) for the ringbuffer. + +registers + Set of registers values. Each entry is on its own line enclosed + by brackets { }. + + offset + Byte offset of the register from the start of the + GPU memory region. + + value + Hexadecimal value of the register. -- cgit v1.2.3 From 43a56687d15db09f3cf7b9d53b182bdef86c17c0 Mon Sep 17 00:00:00 2001 From: Jordan Crouse Date: Tue, 24 Jul 2018 10:33:29 -0600 Subject: drm/msm/adreno: Add ringbuffer data to the GPU state Add the contents of each ringbuffer to the GPU state and dump the data in the crash file encoded with ascii85. To save space only the used portions of the ringbuffer are dumped. Signed-off-by: Jordan Crouse Signed-off-by: Rob Clark --- Documentation/gpu/msm-crash-dump.rst | 7 ++++++ drivers/gpu/drm/msm/adreno/adreno_gpu.c | 39 +++++++++++++++++++++++++++++++++ drivers/gpu/drm/msm/msm_gpu.h | 2 ++ 3 files changed, 48 insertions(+) (limited to 'Documentation/gpu/msm-crash-dump.rst') diff --git a/Documentation/gpu/msm-crash-dump.rst b/Documentation/gpu/msm-crash-dump.rst index 75ab1d541c03..35e87004e006 100644 --- a/Documentation/gpu/msm-crash-dump.rst +++ b/Documentation/gpu/msm-crash-dump.rst @@ -59,6 +59,13 @@ ringbuffer wptr The current write pointer (wptr) for the ringbuffer. + size + Maximum size of the ringbuffer programmed in the hardware. + + data + The contents of the ring encoded as ascii85. Only the used + portions of the ring will be printed. + registers Set of registers values. Each entry is on its own line enclosed by brackets { }. diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index c7a998d9dc85..808d4fc9c4a1 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -17,6 +17,7 @@ * this program. If not, see . */ +#include #include #include "adreno_gpu.h" #include "msm_gem.h" @@ -383,11 +384,29 @@ struct msm_gpu_state *adreno_gpu_state_get(struct msm_gpu *gpu) do_gettimeofday(&state->time); for (i = 0; i < gpu->nr_rings; i++) { + int size = 0, j; + state->ring[i].fence = gpu->rb[i]->memptrs->fence; state->ring[i].iova = gpu->rb[i]->iova; state->ring[i].seqno = gpu->rb[i]->seqno; state->ring[i].rptr = get_rptr(adreno_gpu, gpu->rb[i]); state->ring[i].wptr = get_wptr(gpu->rb[i]); + + /* Copy at least 'wptr' dwords of the data */ + size = state->ring[i].wptr; + + /* After wptr find the last non zero dword to save space */ + for (j = state->ring[i].wptr; j < MSM_GPU_RINGBUFFER_SZ >> 2; j++) + if (gpu->rb[i]->start[j]) + size = j + 1; + + if (size) { + state->ring[i].data = kmalloc(size << 2, GFP_KERNEL); + if (state->ring[i].data) { + memcpy(state->ring[i].data, gpu->rb[i]->start, size << 2); + state->ring[i].data_size = size << 2; + } + } } /* Count the number of registers */ @@ -418,9 +437,13 @@ struct msm_gpu_state *adreno_gpu_state_get(struct msm_gpu *gpu) static void adreno_gpu_state_destroy(struct kref *kref) { + int i; struct msm_gpu_state *state = container_of(kref, struct msm_gpu_state, ref); + for (i = 0; i < ARRAY_SIZE(state->ring); i++) + kfree(state->ring[i].data); + kfree(state->comm); kfree(state->cmd); kfree(state->registers); @@ -461,6 +484,22 @@ void adreno_show(struct msm_gpu *gpu, struct msm_gpu_state *state, drm_printf(p, " retired-fence: %d\n", state->ring[i].fence); drm_printf(p, " rptr: %d\n", state->ring[i].rptr); drm_printf(p, " wptr: %d\n", state->ring[i].wptr); + drm_printf(p, " size: %d\n", MSM_GPU_RINGBUFFER_SZ); + + if (state->ring[i].data && state->ring[i].data_size) { + u32 *ptr = (u32 *) state->ring[i].data; + char out[ASCII85_BUFSZ]; + long len = ascii85_encode_len(state->ring[i].data_size); + int j; + + drm_printf(p, " data: !!ascii85 |\n"); + drm_printf(p, " "); + + for (j = 0; j < len; j++) + drm_printf(p, ascii85_encode(ptr[j], out)); + + drm_printf(p, "\n"); + } } drm_puts(p, "registers:\n"); diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h index b1cb44922441..878090b57e90 100644 --- a/drivers/gpu/drm/msm/msm_gpu.h +++ b/drivers/gpu/drm/msm/msm_gpu.h @@ -191,6 +191,8 @@ struct msm_gpu_state { u32 seqno; u32 rptr; u32 wptr; + void *data; + int data_size; } ring[MSM_GPU_MAX_RINGS]; int nr_registers; -- cgit v1.2.3 From 50f8d21863b9b774b198e631d2b14878f6a54b5b Mon Sep 17 00:00:00 2001 From: Jordan Crouse Date: Tue, 24 Jul 2018 10:33:30 -0600 Subject: drm/msm/adreno: Add a5xx specific registers for the GPU state HLSQ, SP and TP registers are only accessible from a special aperture and to make matters worse the aperture is blocked from the CPU on targets that can support secure rendering. Luckily the GPU hardware has its own purpose built register dumper that can access the registers from the aperture. Add a5xx specific code to program the crashdumper and retrieve the wayward registers and dump them for the crash state. Also, remove a block of registers the regular CPU accessible list that aren't useful for debug which helps reduce the size of the crash state file by a goodly amount. Signed-off-by: Jordan Crouse Signed-off-by: Rob Clark --- Documentation/gpu/msm-crash-dump.rst | 4 + drivers/gpu/drm/msm/adreno/a3xx_gpu.c | 8 +- drivers/gpu/drm/msm/adreno/a4xx_gpu.c | 8 +- drivers/gpu/drm/msm/adreno/a5xx_gpu.c | 236 ++++++++++++++++++++++++++++++-- drivers/gpu/drm/msm/adreno/adreno_gpu.c | 23 ++-- drivers/gpu/drm/msm/adreno/adreno_gpu.h | 4 +- 6 files changed, 252 insertions(+), 31 deletions(-) (limited to 'Documentation/gpu/msm-crash-dump.rst') diff --git a/Documentation/gpu/msm-crash-dump.rst b/Documentation/gpu/msm-crash-dump.rst index 35e87004e006..7943f43f70d6 100644 --- a/Documentation/gpu/msm-crash-dump.rst +++ b/Documentation/gpu/msm-crash-dump.rst @@ -76,3 +76,7 @@ registers value Hexadecimal value of the register. + +registers-hlsq + (5xx only) Register values from the HLSQ aperture. + Same format as the register section. diff --git a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c index fc502e412132..669c2d4b070d 100644 --- a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c @@ -421,10 +421,12 @@ static void a3xx_dump(struct msm_gpu *gpu) static struct msm_gpu_state *a3xx_gpu_state_get(struct msm_gpu *gpu) { - struct msm_gpu_state *state = adreno_gpu_state_get(gpu); + struct msm_gpu_state *state = kzalloc(sizeof(*state), GFP_KERNEL); - if (IS_ERR(state)) - return state; + if (!state) + return ERR_PTR(-ENOMEM); + + adreno_gpu_state_get(gpu, state); state->rbbm_status = gpu_read(gpu, REG_A3XX_RBBM_STATUS); diff --git a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c index 8129cf037db1..7c4e6dc1ed59 100644 --- a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c @@ -457,10 +457,12 @@ static const unsigned int a4xx_registers[] = { static struct msm_gpu_state *a4xx_gpu_state_get(struct msm_gpu *gpu) { - struct msm_gpu_state *state = adreno_gpu_state_get(gpu); + struct msm_gpu_state *state = kzalloc(sizeof(*state), GFP_KERNEL); - if (IS_ERR(state)) - return state; + if (!state) + return ERR_PTR(-ENOMEM); + + adreno_gpu_state_get(gpu, state); state->rbbm_status = gpu_read(gpu, REG_A4XX_RBBM_STATUS); diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c index 16074fa6bf1e..bd84f71d27d8 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c @@ -19,6 +19,7 @@ #include #include #include +#include #include "msm_gem.h" #include "msm_mmu.h" #include "a5xx_gpu.h" @@ -1123,8 +1124,9 @@ static const u32 a5xx_registers[] = { 0xE800, 0xE806, 0xE810, 0xE89A, 0xE8A0, 0xE8A4, 0xE8AA, 0xE8EB, 0xE900, 0xE905, 0xEB80, 0xEB8F, 0xEBB0, 0xEBB0, 0xEC00, 0xEC05, 0xEC08, 0xECE9, 0xECF0, 0xECF0, 0xEA80, 0xEA80, 0xEA82, 0xEAA3, - 0xEAA5, 0xEAC2, 0xA800, 0xA8FF, 0xAC60, 0xAC60, 0xB000, 0xB97F, - 0xB9A0, 0xB9BF, ~0 + 0xEAA5, 0xEAC2, 0xA800, 0xA800, 0xA820, 0xA828, 0xA840, 0xA87D, + 0XA880, 0xA88D, 0xA890, 0xA8A3, 0xA8D0, 0xA8D8, 0xA8E0, 0xA8F5, + 0xAC60, 0xAC60, ~0, }; static void a5xx_dump(struct msm_gpu *gpu) @@ -1195,25 +1197,233 @@ static int a5xx_get_timestamp(struct msm_gpu *gpu, uint64_t *value) return 0; } +struct a5xx_crashdumper { + void *ptr; + struct drm_gem_object *bo; + u64 iova; +}; + +struct a5xx_gpu_state { + struct msm_gpu_state base; + u32 *hlsqregs; +}; + +#define gpu_poll_timeout(gpu, addr, val, cond, interval, timeout) \ + readl_poll_timeout((gpu)->mmio + ((addr) << 2), val, cond, \ + interval, timeout) + +static int a5xx_crashdumper_init(struct msm_gpu *gpu, + struct a5xx_crashdumper *dumper) +{ + dumper->ptr = msm_gem_kernel_new_locked(gpu->dev, + SZ_1M, MSM_BO_UNCACHED, gpu->aspace, + &dumper->bo, &dumper->iova); + + if (IS_ERR(dumper->ptr)) + return PTR_ERR(dumper->ptr); + + return 0; +} + +static void a5xx_crashdumper_free(struct msm_gpu *gpu, + struct a5xx_crashdumper *dumper) +{ + msm_gem_put_iova(dumper->bo, gpu->aspace); + msm_gem_put_vaddr(dumper->bo); + + drm_gem_object_unreference(dumper->bo); +} + +static int a5xx_crashdumper_run(struct msm_gpu *gpu, + struct a5xx_crashdumper *dumper) +{ + u32 val; + + if (IS_ERR_OR_NULL(dumper->ptr)) + return -EINVAL; + + gpu_write64(gpu, REG_A5XX_CP_CRASH_SCRIPT_BASE_LO, + REG_A5XX_CP_CRASH_SCRIPT_BASE_HI, dumper->iova); + + gpu_write(gpu, REG_A5XX_CP_CRASH_DUMP_CNTL, 1); + + return gpu_poll_timeout(gpu, REG_A5XX_CP_CRASH_DUMP_CNTL, val, + val & 0x04, 100, 10000); +} + +/* + * These are a list of the registers that need to be read through the HLSQ + * aperture through the crashdumper. These are not nominally accessible from + * the CPU on a secure platform. + */ +static const struct { + u32 type; + u32 regoffset; + u32 count; +} a5xx_hlsq_aperture_regs[] = { + { 0x35, 0xe00, 0x32 }, /* HSLQ non-context */ + { 0x31, 0x2080, 0x1 }, /* HLSQ 2D context 0 */ + { 0x33, 0x2480, 0x1 }, /* HLSQ 2D context 1 */ + { 0x32, 0xe780, 0x62 }, /* HLSQ 3D context 0 */ + { 0x34, 0xef80, 0x62 }, /* HLSQ 3D context 1 */ + { 0x3f, 0x0ec0, 0x40 }, /* SP non-context */ + { 0x3d, 0x2040, 0x1 }, /* SP 2D context 0 */ + { 0x3b, 0x2440, 0x1 }, /* SP 2D context 1 */ + { 0x3e, 0xe580, 0x170 }, /* SP 3D context 0 */ + { 0x3c, 0xed80, 0x170 }, /* SP 3D context 1 */ + { 0x3a, 0x0f00, 0x1c }, /* TP non-context */ + { 0x38, 0x2000, 0xa }, /* TP 2D context 0 */ + { 0x36, 0x2400, 0xa }, /* TP 2D context 1 */ + { 0x39, 0xe700, 0x80 }, /* TP 3D context 0 */ + { 0x37, 0xef00, 0x80 }, /* TP 3D context 1 */ +}; + +static void a5xx_gpu_state_get_hlsq_regs(struct msm_gpu *gpu, + struct a5xx_gpu_state *a5xx_state) +{ + struct a5xx_crashdumper dumper = { 0 }; + u32 offset, count = 0; + u64 *ptr; + int i; + + if (a5xx_crashdumper_init(gpu, &dumper)) + return; + + /* The script will be written at offset 0 */ + ptr = dumper.ptr; + + /* Start writing the data at offset 256k */ + offset = dumper.iova + (256 * SZ_1K); + + /* Count how many additional registers to get from the HLSQ aperture */ + for (i = 0; i < ARRAY_SIZE(a5xx_hlsq_aperture_regs); i++) + count += a5xx_hlsq_aperture_regs[i].count; + + a5xx_state->hlsqregs = kcalloc(count, sizeof(u32), GFP_KERNEL); + if (!a5xx_state->hlsqregs) + return; + + /* Build the crashdump script */ + for (i = 0; i < ARRAY_SIZE(a5xx_hlsq_aperture_regs); i++) { + u32 type = a5xx_hlsq_aperture_regs[i].type; + u32 c = a5xx_hlsq_aperture_regs[i].count; + + /* Write the register to select the desired bank */ + *ptr++ = ((u64) type << 8); + *ptr++ = (((u64) REG_A5XX_HLSQ_DBG_READ_SEL) << 44) | + (1 << 21) | 1; + + *ptr++ = offset; + *ptr++ = (((u64) REG_A5XX_HLSQ_DBG_AHB_READ_APERTURE) << 44) + | c; + + offset += c * sizeof(u32); + } + + /* Write two zeros to close off the script */ + *ptr++ = 0; + *ptr++ = 0; + + if (a5xx_crashdumper_run(gpu, &dumper)) { + kfree(a5xx_state->hlsqregs); + a5xx_crashdumper_free(gpu, &dumper); + return; + } + + /* Copy the data from the crashdumper to the state */ + memcpy(a5xx_state->hlsqregs, dumper.ptr + (256 * SZ_1K), + count * sizeof(u32)); + + a5xx_crashdumper_free(gpu, &dumper); +} + static struct msm_gpu_state *a5xx_gpu_state_get(struct msm_gpu *gpu) { - struct msm_gpu_state *state; + struct a5xx_gpu_state *a5xx_state = kzalloc(sizeof(*a5xx_state), + GFP_KERNEL); - /* - * Temporarily disable hardware clock gating before going into - * adreno_show to avoid issues while reading the registers - */ + if (!a5xx_state) + return ERR_PTR(-ENOMEM); + + /* Temporarily disable hardware clock gating before reading the hw */ a5xx_set_hwcg(gpu, false); - state = adreno_gpu_state_get(gpu); + /* First get the generic state from the adreno core */ + adreno_gpu_state_get(gpu, &(a5xx_state->base)); + + a5xx_state->base.rbbm_status = gpu_read(gpu, REG_A5XX_RBBM_STATUS); - if (!IS_ERR(state)) - state->rbbm_status = gpu_read(gpu, REG_A5XX_RBBM_STATUS); + /* Get the HLSQ regs with the help of the crashdumper */ + a5xx_gpu_state_get_hlsq_regs(gpu, a5xx_state); a5xx_set_hwcg(gpu, true); - return state; + return &a5xx_state->base; +} + +static void a5xx_gpu_state_destroy(struct kref *kref) +{ + struct msm_gpu_state *state = container_of(kref, + struct msm_gpu_state, ref); + struct a5xx_gpu_state *a5xx_state = container_of(state, + struct a5xx_gpu_state, base); + + kfree(a5xx_state->hlsqregs); + + adreno_gpu_state_destroy(state); + kfree(a5xx_state); +} + +int a5xx_gpu_state_put(struct msm_gpu_state *state) +{ + if (IS_ERR_OR_NULL(state)) + return 1; + + return kref_put(&state->ref, a5xx_gpu_state_destroy); +} + + +#if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP) +void a5xx_show(struct msm_gpu *gpu, struct msm_gpu_state *state, + struct drm_printer *p) +{ + int i, j; + u32 pos = 0; + struct a5xx_gpu_state *a5xx_state = container_of(state, + struct a5xx_gpu_state, base); + + if (IS_ERR_OR_NULL(state)) + return; + + adreno_show(gpu, state, p); + + /* Dump the additional a5xx HLSQ registers */ + if (!a5xx_state->hlsqregs) + return; + + drm_printf(p, "registers-hlsq:\n"); + + for (i = 0; i < ARRAY_SIZE(a5xx_hlsq_aperture_regs); i++) { + u32 o = a5xx_hlsq_aperture_regs[i].regoffset; + u32 c = a5xx_hlsq_aperture_regs[i].count; + + for (j = 0; j < c; j++, pos++, o++) { + /* + * To keep the crashdump simple we pull the entire range + * for each register type but not all of the registers + * in the range are valid. Fortunately invalid registers + * stick out like a sore thumb with a value of + * 0xdeadbeef + */ + if (a5xx_state->hlsqregs[pos] == 0xdeadbeef) + continue; + + drm_printf(p, " - { offset: 0x%04x, value: 0x%08x }\n", + o << 2, a5xx_state->hlsqregs[pos]); + } + } } +#endif static struct msm_ringbuffer *a5xx_active_ring(struct msm_gpu *gpu) { @@ -1244,14 +1454,14 @@ static const struct adreno_gpu_funcs funcs = { .irq = a5xx_irq, .destroy = a5xx_destroy, #if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP) - .show = adreno_show, + .show = a5xx_show, #endif #if defined(CONFIG_DEBUG_FS) .debugfs_init = a5xx_debugfs_init, #endif .gpu_busy = a5xx_gpu_busy, .gpu_state_get = a5xx_gpu_state_get, - .gpu_state_put = adreno_gpu_state_put, + .gpu_state_put = a5xx_gpu_state_put, }, .get_timestamp = a5xx_get_timestamp, }; diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index 808d4fc9c4a1..cd418dab7a62 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -369,16 +369,11 @@ bool adreno_idle(struct msm_gpu *gpu, struct msm_ringbuffer *ring) return false; } -struct msm_gpu_state *adreno_gpu_state_get(struct msm_gpu *gpu) +int adreno_gpu_state_get(struct msm_gpu *gpu, struct msm_gpu_state *state) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); - struct msm_gpu_state *state; int i, count = 0; - state = kzalloc(sizeof(*state), GFP_KERNEL); - if (!state) - return ERR_PTR(-ENOMEM); - kref_init(&state->ref); do_gettimeofday(&state->time); @@ -432,14 +427,12 @@ struct msm_gpu_state *adreno_gpu_state_get(struct msm_gpu *gpu) state->nr_registers = count; } - return state; + return 0; } -static void adreno_gpu_state_destroy(struct kref *kref) +void adreno_gpu_state_destroy(struct msm_gpu_state *state) { int i; - struct msm_gpu_state *state = container_of(kref, - struct msm_gpu_state, ref); for (i = 0; i < ARRAY_SIZE(state->ring); i++) kfree(state->ring[i].data); @@ -447,6 +440,14 @@ static void adreno_gpu_state_destroy(struct kref *kref) kfree(state->comm); kfree(state->cmd); kfree(state->registers); +} + +static void adreno_gpu_state_kref_destroy(struct kref *kref) +{ + struct msm_gpu_state *state = container_of(kref, + struct msm_gpu_state, ref); + + adreno_gpu_state_destroy(state); kfree(state); } @@ -455,7 +456,7 @@ int adreno_gpu_state_put(struct msm_gpu_state *state) if (IS_ERR_OR_NULL(state)) return 1; - return kref_put(&state->ref, adreno_gpu_state_destroy); + return kref_put(&state->ref, adreno_gpu_state_kref_destroy); } #if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP) diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h index 4a868aaf1a70..4406776597fd 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h @@ -230,7 +230,9 @@ int adreno_gpu_init(struct drm_device *drm, struct platform_device *pdev, void adreno_gpu_cleanup(struct adreno_gpu *gpu); -struct msm_gpu_state *adreno_gpu_state_get(struct msm_gpu *gpu); +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); /* ringbuffer helpers (the parts that are adreno specific) */ -- cgit v1.2.3 From cdb95931dea32981545e34a3b1dfc9e172425d95 Mon Sep 17 00:00:00 2001 From: Jordan Crouse Date: Tue, 24 Jul 2018 10:33:31 -0600 Subject: drm/msm/gpu: Add the buffer objects from the submit to the crash dump For hangs, dump copy out the contents of the buffer objects attached to the guilty submission and print them in the crash dump report. Signed-off-by: Jordan Crouse Signed-off-by: Rob Clark --- Documentation/gpu/msm-crash-dump.rst | 14 ++++++++ drivers/gpu/drm/msm/adreno/adreno_gpu.c | 58 +++++++++++++++++++++++++++------ drivers/gpu/drm/msm/msm_gpu.c | 48 +++++++++++++++++++++++++-- drivers/gpu/drm/msm/msm_gpu.h | 9 +++++ 4 files changed, 116 insertions(+), 13 deletions(-) (limited to 'Documentation/gpu/msm-crash-dump.rst') diff --git a/Documentation/gpu/msm-crash-dump.rst b/Documentation/gpu/msm-crash-dump.rst index 7943f43f70d6..757cd257e0d8 100644 --- a/Documentation/gpu/msm-crash-dump.rst +++ b/Documentation/gpu/msm-crash-dump.rst @@ -66,6 +66,20 @@ ringbuffer The contents of the ring encoded as ascii85. Only the used portions of the ring will be printed. +bo + List of buffers from the hanging submission if available. + Each buffer object will have a uinque iova. + + iova + GPU address of the buffer object. + + size + Allocated size of the buffer object. + + data + The contents of the buffer object encoded with ascii85. Only + Trailing zeros at the end of the buffer will be skipped. + registers Set of registers values. Each entry is on its own line enclosed by brackets { }. diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index cd418dab7a62..08d3c618b7de 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -437,6 +437,10 @@ void adreno_gpu_state_destroy(struct msm_gpu_state *state) for (i = 0; i < ARRAY_SIZE(state->ring); i++) kfree(state->ring[i].data); + for (i = 0; state->bos && i < state->nr_bos; i++) + kvfree(state->bos[i].data); + + kfree(state->bos); kfree(state->comm); kfree(state->cmd); kfree(state->registers); @@ -460,6 +464,39 @@ int adreno_gpu_state_put(struct msm_gpu_state *state) } #if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP) + +static void adreno_show_object(struct drm_printer *p, u32 *ptr, int len) +{ + char out[ASCII85_BUFSZ]; + long l, datalen, i; + + if (!ptr || !len) + return; + + /* + * Only dump the non-zero part of the buffer - rarely will any data + * completely fill the entire allocated size of the buffer + */ + for (datalen = 0, i = 0; i < len >> 2; i++) { + if (ptr[i]) + datalen = (i << 2) + 1; + } + + /* Skip printing the object if it is empty */ + if (datalen == 0) + return; + + l = ascii85_encode_len(datalen); + + drm_puts(p, " data: !!ascii85 |\n"); + drm_puts(p, " "); + + for (i = 0; i < l; i++) + drm_puts(p, ascii85_encode(ptr[i], out)); + + drm_puts(p, "\n"); +} + void adreno_show(struct msm_gpu *gpu, struct msm_gpu_state *state, struct drm_printer *p) { @@ -487,19 +524,20 @@ void adreno_show(struct msm_gpu *gpu, struct msm_gpu_state *state, drm_printf(p, " wptr: %d\n", state->ring[i].wptr); drm_printf(p, " size: %d\n", MSM_GPU_RINGBUFFER_SZ); - if (state->ring[i].data && state->ring[i].data_size) { - u32 *ptr = (u32 *) state->ring[i].data; - char out[ASCII85_BUFSZ]; - long len = ascii85_encode_len(state->ring[i].data_size); - int j; + adreno_show_object(p, state->ring[i].data, + state->ring[i].data_size); + } - drm_printf(p, " data: !!ascii85 |\n"); - drm_printf(p, " "); + if (state->bos) { + drm_puts(p, "bos:\n"); - for (j = 0; j < len; j++) - drm_printf(p, ascii85_encode(ptr[j], out)); + for (i = 0; i < state->nr_bos; i++) { + drm_printf(p, " - iova: 0x%016llx\n", + state->bos[i].iova); + drm_printf(p, " size: %zd\n", state->bos[i].size); - drm_printf(p, "\n"); + adreno_show_object(p, state->bos[i].data, + state->bos[i].size); } } diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index 5f39549d9a8b..3cf8e8d29812 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -318,8 +318,39 @@ static void msm_gpu_devcoredump_free(void *data) msm_gpu_crashstate_put(gpu); } -static void msm_gpu_crashstate_capture(struct msm_gpu *gpu, char *comm, - char *cmd) +static void msm_gpu_crashstate_get_bo(struct msm_gpu_state *state, + struct msm_gem_object *obj, u64 iova, u32 flags) +{ + struct msm_gpu_state_bo *state_bo = &state->bos[state->nr_bos]; + + /* Don't record write only objects */ + + state_bo->size = obj->base.size; + state_bo->iova = iova; + + /* Only store the data for buffer objects marked for read */ + if ((flags & MSM_SUBMIT_BO_READ)) { + void *ptr; + + state_bo->data = kvmalloc(obj->base.size, GFP_KERNEL); + if (!state_bo->data) + return; + + ptr = msm_gem_get_vaddr_active(&obj->base); + if (IS_ERR(ptr)) { + kvfree(state_bo->data); + return; + } + + memcpy(state_bo->data, ptr, obj->base.size); + msm_gem_put_vaddr(&obj->base); + } + + state->nr_bos++; +} + +static void msm_gpu_crashstate_capture(struct msm_gpu *gpu, + struct msm_gem_submit *submit, char *comm, char *cmd) { struct msm_gpu_state *state; @@ -335,6 +366,17 @@ static void msm_gpu_crashstate_capture(struct msm_gpu *gpu, char *comm, state->comm = kstrdup(comm, GFP_KERNEL); state->cmd = kstrdup(cmd, GFP_KERNEL); + if (submit) { + int i; + + state->bos = kcalloc(submit->nr_bos, + sizeof(struct msm_gpu_state_bo), GFP_KERNEL); + + for (i = 0; state->bos && i < submit->nr_bos; i++) + msm_gpu_crashstate_get_bo(state, submit->bos[i].obj, + submit->bos[i].iova, submit->bos[i].flags); + } + /* Set the active crash state to be dumped on failure */ gpu->crashstate = state; @@ -434,7 +476,7 @@ static void recover_worker(struct work_struct *work) /* Record the crash state */ pm_runtime_get_sync(&gpu->pdev->dev); - msm_gpu_crashstate_capture(gpu, comm, cmd); + msm_gpu_crashstate_capture(gpu, submit, comm, cmd); pm_runtime_put_sync(&gpu->pdev->dev); kfree(cmd); diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h index 878090b57e90..57380ef8d1f7 100644 --- a/drivers/gpu/drm/msm/msm_gpu.h +++ b/drivers/gpu/drm/msm/msm_gpu.h @@ -181,6 +181,12 @@ struct msm_gpu_submitqueue { struct kref ref; }; +struct msm_gpu_state_bo { + u64 iova; + size_t size; + void *data; +}; + struct msm_gpu_state { struct kref ref; struct timeval time; @@ -202,6 +208,9 @@ struct msm_gpu_state { char *comm; char *cmd; + + int nr_bos; + struct msm_gpu_state_bo *bos; }; static inline void gpu_write(struct msm_gpu *gpu, u32 reg, u32 data) -- cgit v1.2.3