[Mesa-dev] [PATCH v2 03/13] intel/decoder: tools: Use engine for decoding batch instructions
Lionel Landwerlin
lionel.g.landwerlin at intel.com
Tue Nov 6 12:11:16 UTC 2018
On 31/10/2018 13:12, Toni Lönnberg wrote:
> The engine to which the batch was sent to is now set to the decoder context when
> decoding the batch. This is needed so that we can distinguish between
> instructions as the render and video pipe share some of the instruction opcodes.
>
> v2: The engine is now in the decoder context and the batch decoder uses a local
> function for finding the instruction for an engine.
> ---
> src/intel/common/gen_batch_decoder.c | 25 +++++++++++++++---------
> src/intel/common/gen_decoder.c | 7 +++++--
> src/intel/common/gen_decoder.h | 6 +++++-
> src/intel/tools/aubinator.c | 3 ++-
> src/intel/tools/aubinator_error_decode.c | 16 +++++++++++++++
> 5 files changed, 44 insertions(+), 13 deletions(-)
>
> diff --git a/src/intel/common/gen_batch_decoder.c b/src/intel/common/gen_batch_decoder.c
> index 63f04627572..d5482a4d455 100644
> --- a/src/intel/common/gen_batch_decoder.c
> +++ b/src/intel/common/gen_batch_decoder.c
> @@ -45,6 +45,7 @@ gen_batch_decode_ctx_init(struct gen_batch_decode_ctx *ctx,
> ctx->fp = fp;
> ctx->flags = flags;
> ctx->max_vbo_decoded_lines = -1; /* No limit! */
> + ctx->engine = I915_ENGINE_CLASS_RENDER;
>
> if (xml_path == NULL)
> ctx->spec = gen_spec_load(devinfo);
> @@ -192,10 +193,16 @@ ctx_print_buffer(struct gen_batch_decode_ctx *ctx,
> fprintf(ctx->fp, "\n");
> }
>
> +static struct gen_group *
> +gen_ctx_find_instruction(struct gen_batch_decode_ctx *ctx, const uint32_t *p)
> +{
> + return gen_spec_find_instruction(ctx->spec, ctx->engine, p);
> +}
> +
> static void
> handle_state_base_address(struct gen_batch_decode_ctx *ctx, const uint32_t *p)
> {
> - struct gen_group *inst = gen_spec_find_instruction(ctx->spec, p);
> + struct gen_group *inst = gen_ctx_find_instruction(ctx, p);
>
> struct gen_field_iterator iter;
> gen_field_iterator_init(&iter, inst, p, 0, false);
> @@ -309,7 +316,7 @@ static void
> handle_media_interface_descriptor_load(struct gen_batch_decode_ctx *ctx,
> const uint32_t *p)
> {
> - struct gen_group *inst = gen_spec_find_instruction(ctx->spec, p);
> + struct gen_group *inst = gen_ctx_find_instruction(ctx, p);
> struct gen_group *desc =
> gen_spec_find_struct(ctx->spec, "INTERFACE_DESCRIPTOR_DATA");
>
> @@ -373,7 +380,7 @@ static void
> handle_3dstate_vertex_buffers(struct gen_batch_decode_ctx *ctx,
> const uint32_t *p)
> {
> - struct gen_group *inst = gen_spec_find_instruction(ctx->spec, p);
> + struct gen_group *inst = gen_ctx_find_instruction(ctx, p);
> struct gen_group *vbs = gen_spec_find_struct(ctx->spec, "VERTEX_BUFFER_STATE");
>
> struct gen_batch_decode_bo vb = {};
> @@ -436,7 +443,7 @@ static void
> handle_3dstate_index_buffer(struct gen_batch_decode_ctx *ctx,
> const uint32_t *p)
> {
> - struct gen_group *inst = gen_spec_find_instruction(ctx->spec, p);
> + struct gen_group *inst = gen_ctx_find_instruction(ctx, p);
>
> struct gen_batch_decode_bo ib = {};
> uint32_t ib_size = 0;
> @@ -486,7 +493,7 @@ handle_3dstate_index_buffer(struct gen_batch_decode_ctx *ctx,
> static void
> decode_single_ksp(struct gen_batch_decode_ctx *ctx, const uint32_t *p)
> {
> - struct gen_group *inst = gen_spec_find_instruction(ctx->spec, p);
> + struct gen_group *inst = gen_ctx_find_instruction(ctx, p);
>
> uint64_t ksp = 0;
> bool is_simd8 = false; /* vertex shaders on Gen8+ only */
> @@ -528,7 +535,7 @@ decode_single_ksp(struct gen_batch_decode_ctx *ctx, const uint32_t *p)
> static void
> decode_ps_kernels(struct gen_batch_decode_ctx *ctx, const uint32_t *p)
> {
> - struct gen_group *inst = gen_spec_find_instruction(ctx->spec, p);
> + struct gen_group *inst = gen_ctx_find_instruction(ctx, p);
>
> uint64_t ksp[3] = {0, 0, 0};
> bool enabled[3] = {false, false, false};
> @@ -576,7 +583,7 @@ decode_ps_kernels(struct gen_batch_decode_ctx *ctx, const uint32_t *p)
> static void
> decode_3dstate_constant(struct gen_batch_decode_ctx *ctx, const uint32_t *p)
> {
> - struct gen_group *inst = gen_spec_find_instruction(ctx->spec, p);
> + struct gen_group *inst = gen_ctx_find_instruction(ctx, p);
> struct gen_group *body =
> gen_spec_find_struct(ctx->spec, "3DSTATE_CONSTANT_BODY");
>
> @@ -658,7 +665,7 @@ decode_dynamic_state_pointers(struct gen_batch_decode_ctx *ctx,
> const char *struct_type, const uint32_t *p,
> int count)
> {
> - struct gen_group *inst = gen_spec_find_instruction(ctx->spec, p);
> + struct gen_group *inst = gen_ctx_find_instruction(ctx, p);
>
> uint32_t state_offset = 0;
>
> @@ -802,7 +809,7 @@ gen_print_batch(struct gen_batch_decode_ctx *ctx,
> struct gen_group *inst;
>
> for (p = batch; p < end; p += length) {
> - inst = gen_spec_find_instruction(ctx->spec, p);
> + inst = gen_ctx_find_instruction(ctx, p);
> length = gen_group_get_length(inst, p);
> assert(inst == NULL || length > 0);
> length = MAX2(1, length);
> diff --git a/src/intel/common/gen_decoder.c b/src/intel/common/gen_decoder.c
> index 43e5b6e3561..bb1e0d3980f 100644
> --- a/src/intel/common/gen_decoder.c
> +++ b/src/intel/common/gen_decoder.c
> @@ -719,12 +719,15 @@ void gen_spec_destroy(struct gen_spec *spec)
> }
>
> struct gen_group *
> -gen_spec_find_instruction(struct gen_spec *spec, const uint32_t *p)
> +gen_spec_find_instruction(struct gen_spec *spec,
> + enum drm_i915_gem_engine_class engine,
> + const uint32_t *p)
> {
> hash_table_foreach(spec->commands, entry) {
> struct gen_group *command = entry->data;
> uint32_t opcode = *p & command->opcode_mask;
> - if (opcode == command->opcode)
> + if ((command->engine & I915_ENGINE_CLASS_TO_MASK(engine)) &&
> + opcode == command->opcode)
> return command;
> }
>
> diff --git a/src/intel/common/gen_decoder.h b/src/intel/common/gen_decoder.h
> index e9efa6d1451..dfb07b24a9f 100644
> --- a/src/intel/common/gen_decoder.h
> +++ b/src/intel/common/gen_decoder.h
> @@ -55,7 +55,9 @@ struct gen_spec *gen_spec_load_from_path(const struct gen_device_info *devinfo,
> const char *path);
> void gen_spec_destroy(struct gen_spec *spec);
> uint32_t gen_spec_get_gen(struct gen_spec *spec);
> -struct gen_group *gen_spec_find_instruction(struct gen_spec *spec, const uint32_t *p);
> +struct gen_group *gen_spec_find_instruction(struct gen_spec *spec,
> + enum drm_i915_gem_engine_class engine,
> + const uint32_t *p);
> struct gen_group *gen_spec_find_register(struct gen_spec *spec, uint32_t offset);
> struct gen_group *gen_spec_find_register_by_name(struct gen_spec *spec, const char *name);
> struct gen_enum *gen_spec_find_enum(struct gen_spec *spec, const char *name);
> @@ -232,6 +234,8 @@ struct gen_batch_decode_ctx {
> uint64_t instruction_base;
>
> int max_vbo_decoded_lines;
> +
> + enum drm_i915_gem_engine_class engine;
> };
>
> void gen_batch_decode_ctx_init(struct gen_batch_decode_ctx *ctx,
> diff --git a/src/intel/tools/aubinator.c b/src/intel/tools/aubinator.c
> index ed758a62503..f3bd006a8ab 100644
> --- a/src/intel/tools/aubinator.c
> +++ b/src/intel/tools/aubinator.c
> @@ -157,7 +157,7 @@ handle_execlist_write(void *user_data, enum drm_i915_gem_engine_class engine, ui
> batch_ctx.get_bo = aub_mem_get_ggtt_bo;
> }
>
> - (void)engine; /* TODO */
> + batch_ctx.engine = engine;
> gen_print_batch(&batch_ctx, commands, ring_buffer_tail - ring_buffer_head,
> 0);
> aub_mem_clear_bo_maps(&mem);
> @@ -170,6 +170,7 @@ handle_ring_write(void *user_data, enum drm_i915_gem_engine_class engine,
> batch_ctx.user_data = &mem;
> batch_ctx.get_bo = aub_mem_get_ggtt_bo;
>
> + batch_ctx.engine = engine;
> gen_print_batch(&batch_ctx, data, data_len, 0);
>
> aub_mem_clear_bo_maps(&mem);
> diff --git a/src/intel/tools/aubinator_error_decode.c b/src/intel/tools/aubinator_error_decode.c
> index c581414eb2c..e7f900e4f00 100644
> --- a/src/intel/tools/aubinator_error_decode.c
> +++ b/src/intel/tools/aubinator_error_decode.c
> @@ -111,6 +111,18 @@ static const struct ring_register_mapping fault_registers[] = {
> { VECS, 0, "VECS_FAULT_REG" },
> };
>
> +static enum drm_i915_gem_engine_class class_to_engine(unsigned int class)
> +{
> + static const enum drm_i915_gem_engine_class map[] = {
> + [RCS] = I915_ENGINE_CLASS_RENDER,
> + [BCS] = I915_ENGINE_CLASS_COPY,
> + [VCS] = I915_ENGINE_CLASS_VIDEO,
> + [VECS] = I915_ENGINE_CLASS_VIDEO_ENHANCE,
> + };
> +
> + return map[class];
> +}
> +
Rather than adding a new function, I would modify the existing one.
They already deal with class/instance tuples so the change should be
pretty mechanical.
Otherwise the rest of the change looks good to me.
Thanks for doing this.
-
Lionel
> static int ring_name_to_class(const char *ring_name,
> unsigned int *class)
> {
> @@ -601,6 +613,9 @@ read_data_file(FILE *file)
>
>
> for (int s = 0; s < num_sections; s++) {
> + unsigned int class;
> + ring_name_to_class(sections[s].ring_name, &class);
> +
> printf("--- %s (%s) at 0x%08x %08x\n",
> sections[s].buffer_name, sections[s].ring_name,
> (unsigned) (sections[s].gtt_offset >> 32),
> @@ -610,6 +625,7 @@ read_data_file(FILE *file)
> strcmp(sections[s].buffer_name, "batch buffer") == 0 ||
> strcmp(sections[s].buffer_name, "ring buffer") == 0 ||
> strcmp(sections[s].buffer_name, "HW Context") == 0) {
> + batch_ctx.engine = class_to_engine(class);
> gen_print_batch(&batch_ctx, sections[s].data,
> sections[s].dword_count * 4,
> sections[s].gtt_offset);
More information about the mesa-dev
mailing list