[Mesa-dev] [PATCH 2/2] blorp: Embed a wm_prog_data in blorp_prog_data
Pohjolainen, Topi
topi.pohjolainen at gmail.com
Wed Feb 1 08:01:31 UTC 2017
On Tue, Jan 31, 2017 at 11:05:28AM -0800, Jason Ekstrand wrote:
> While we're at it, we rename it to remove the brw_ prefix
Nice! Both patches:
Reviewed-by: Topi Pohjolainen <topi.pohjolainen at intel.com>
>
> Signed-off-by: Jason Ekstrand <jason at jlekstrand.net>
> ---
> src/intel/blorp/blorp.c | 26 +++++---------
> src/intel/blorp/blorp_blit.c | 2 +-
> src/intel/blorp/blorp_clear.c | 2 +-
> src/intel/blorp/blorp_genX_exec.h | 72 +++++++++++++++++++--------------------
> src/intel/blorp/blorp_priv.h | 30 ++++------------
> 5 files changed, 52 insertions(+), 80 deletions(-)
>
> diff --git a/src/intel/blorp/blorp.c b/src/intel/blorp/blorp.c
> index 2f27274..8a90c3b 100644
> --- a/src/intel/blorp/blorp.c
> +++ b/src/intel/blorp/blorp.c
> @@ -145,7 +145,7 @@ const unsigned *
> brw_blorp_compile_nir_shader(struct blorp_context *blorp, struct nir_shader *nir,
> const struct brw_wm_prog_key *wm_key,
> bool use_repclear,
> - struct brw_blorp_prog_data *prog_data,
> + struct blorp_prog_data *prog_data,
> unsigned *program_size)
> {
> const struct brw_compiler *compiler = blorp->compiler;
> @@ -160,15 +160,14 @@ brw_blorp_compile_nir_shader(struct blorp_context *blorp, struct nir_shader *nir
> nir->options =
> compiler->glsl_compiler_options[MESA_SHADER_FRAGMENT].NirOptions;
>
> - struct brw_wm_prog_data wm_prog_data;
> - memset(&wm_prog_data, 0, sizeof(wm_prog_data));
> + memset(prog_data, 0, sizeof(*prog_data));
>
> - wm_prog_data.base.nr_params = 0;
> - wm_prog_data.base.param = NULL;
> + prog_data->wm.base.nr_params = 0;
> + prog_data->wm.base.param = NULL;
>
> /* BLORP always just uses the first two binding table entries */
> - wm_prog_data.binding_table.render_target_start = BLORP_RENDERBUFFER_BT_INDEX;
> - wm_prog_data.base.binding_table.texture_start = BLORP_TEXTURE_BT_INDEX;
> + prog_data->wm.binding_table.render_target_start = BLORP_RENDERBUFFER_BT_INDEX;
> + prog_data->wm.base.binding_table.texture_start = BLORP_TEXTURE_BT_INDEX;
>
> nir = brw_preprocess_nir(compiler, nir);
> nir_remove_dead_variables(nir, nir_var_shader_in);
> @@ -176,21 +175,12 @@ brw_blorp_compile_nir_shader(struct blorp_context *blorp, struct nir_shader *nir
>
> const unsigned *program =
> brw_compile_fs(compiler, blorp->driver_ctx, mem_ctx,
> - wm_key, &wm_prog_data, nir,
> + wm_key, &prog_data->wm, nir,
> NULL, -1, -1, false, use_repclear, program_size, NULL);
>
> - /* Copy the relavent bits of wm_prog_data over into the blorp prog data */
> - prog_data->dispatch_8 = wm_prog_data.dispatch_8;
> - prog_data->dispatch_16 = wm_prog_data.dispatch_16;
> - prog_data->first_curbe_grf_0 = wm_prog_data.base.dispatch_grf_start_reg;
> - prog_data->first_curbe_grf_2 = wm_prog_data.dispatch_grf_start_reg_2;
> - prog_data->ksp_offset_2 = wm_prog_data.prog_offset_2;
> - prog_data->persample_msaa_dispatch = wm_prog_data.persample_dispatch;
> - prog_data->flat_inputs = wm_prog_data.flat_inputs;
> - prog_data->num_varying_inputs = wm_prog_data.num_varying_inputs;
> prog_data->inputs_read = nir->info.inputs_read;
>
> - assert(wm_prog_data.base.nr_params == 0);
> + assert(prog_data->wm.base.nr_params == 0);
>
> return program;
> }
> diff --git a/src/intel/blorp/blorp_blit.c b/src/intel/blorp/blorp_blit.c
> index af46389..fc9e737 100644
> --- a/src/intel/blorp/blorp_blit.c
> +++ b/src/intel/blorp/blorp_blit.c
> @@ -1235,7 +1235,7 @@ brw_blorp_get_blit_kernel(struct blorp_context *blorp,
>
> const unsigned *program;
> unsigned program_size;
> - struct brw_blorp_prog_data prog_data;
> + struct blorp_prog_data prog_data;
>
> /* Try and compile with NIR first. If that fails, fall back to the old
> * method of building shaders manually.
> diff --git a/src/intel/blorp/blorp_clear.c b/src/intel/blorp/blorp_clear.c
> index ce9b292..01889b8 100644
> --- a/src/intel/blorp/blorp_clear.c
> +++ b/src/intel/blorp/blorp_clear.c
> @@ -73,7 +73,7 @@ blorp_params_get_clear_kernel(struct blorp_context *blorp,
> struct brw_wm_prog_key wm_key;
> brw_blorp_init_wm_prog_key(&wm_key);
>
> - struct brw_blorp_prog_data prog_data;
> + struct blorp_prog_data prog_data;
> unsigned program_size;
> const unsigned *program =
> brw_blorp_compile_nir_shader(blorp, b.shader, &wm_key, use_replicated_data,
> diff --git a/src/intel/blorp/blorp_genX_exec.h b/src/intel/blorp/blorp_genX_exec.h
> index 71beadc..bb665e2 100644
> --- a/src/intel/blorp/blorp_genX_exec.h
> +++ b/src/intel/blorp/blorp_genX_exec.h
> @@ -156,7 +156,7 @@ emit_urb_config(struct blorp_batch *batch,
> * where 'n' stands for number of varying inputs expressed as vec4s.
> */
> const unsigned num_varyings =
> - params->wm_prog_data ? params->wm_prog_data->num_varying_inputs : 0;
> + params->wm_prog_data ? params->wm_prog_data->wm.num_varying_inputs : 0;
> const unsigned total_needed = 16 + 16 + num_varyings * 16;
>
> /* The URB size is expressed in units of 64 bytes (512 bits) */
> @@ -191,7 +191,7 @@ blorp_emit_input_varying_data(struct blorp_batch *batch,
> const unsigned vec4_size_in_bytes = 4 * sizeof(float);
> const unsigned max_num_varyings =
> DIV_ROUND_UP(sizeof(params->wm_inputs), vec4_size_in_bytes);
> - const unsigned num_varyings = params->wm_prog_data->num_varying_inputs;
> + const unsigned num_varyings = params->wm_prog_data->wm.num_varying_inputs;
>
> *size = num_varyings * vec4_size_in_bytes;
>
> @@ -239,7 +239,7 @@ blorp_emit_vertex_buffers(struct blorp_batch *batch,
> vb[0].EndAddress.offset += size - 1;
> #endif
>
> - if (params->wm_prog_data && params->wm_prog_data->num_varying_inputs) {
> + if (params->wm_prog_data && params->wm_prog_data->wm.num_varying_inputs) {
> blorp_emit_input_varying_data(batch, params,
> &vb[1].BufferStartingAddress, &size);
> vb[1].VertexBufferIndex = 1;
> @@ -273,7 +273,7 @@ blorp_emit_vertex_elements(struct blorp_batch *batch,
> const struct blorp_params *params)
> {
> const unsigned num_varyings =
> - params->wm_prog_data ? params->wm_prog_data->num_varying_inputs : 0;
> + params->wm_prog_data ? params->wm_prog_data->wm.num_varying_inputs : 0;
> const unsigned num_elements = 2 + num_varyings;
>
> struct GENX(VERTEX_ELEMENT_STATE) ve[num_elements];
> @@ -381,7 +381,7 @@ static void
> blorp_emit_sf_config(struct blorp_batch *batch,
> const struct blorp_params *params)
> {
> - const struct brw_blorp_prog_data *prog_data = params->wm_prog_data;
> + const struct blorp_prog_data *prog_data = params->wm_prog_data;
>
> /* 3DSTATE_SF
> *
> @@ -412,11 +412,11 @@ blorp_emit_sf_config(struct blorp_batch *batch,
>
> blorp_emit(batch, GENX(3DSTATE_SBE), sbe) {
> sbe.VertexURBEntryReadOffset = 1;
> - sbe.NumberofSFOutputAttributes = prog_data->num_varying_inputs;
> - sbe.VertexURBEntryReadLength = brw_blorp_get_urb_length(prog_data);
> + sbe.NumberofSFOutputAttributes = prog_data->wm.num_varying_inputs;
> + sbe.VertexURBEntryReadLength = blorp_get_urb_length(prog_data);
> sbe.ForceVertexURBEntryReadLength = true;
> sbe.ForceVertexURBEntryReadOffset = true;
> - sbe.ConstantInterpolationEnable = prog_data->flat_inputs;
> + sbe.ConstantInterpolationEnable = prog_data->wm.flat_inputs;
>
> #if GEN_GEN >= 9
> for (unsigned i = 0; i < 32; i++)
> @@ -441,9 +441,9 @@ blorp_emit_sf_config(struct blorp_batch *batch,
> blorp_emit(batch, GENX(3DSTATE_SBE), sbe) {
> sbe.VertexURBEntryReadOffset = 1;
> if (prog_data) {
> - sbe.NumberofSFOutputAttributes = prog_data->num_varying_inputs;
> - sbe.VertexURBEntryReadLength = brw_blorp_get_urb_length(prog_data);
> - sbe.ConstantInterpolationEnable = prog_data->flat_inputs;
> + sbe.NumberofSFOutputAttributes = prog_data->wm.num_varying_inputs;
> + sbe.VertexURBEntryReadLength = blorp_get_urb_length(prog_data);
> + sbe.ConstantInterpolationEnable = prog_data->wm.flat_inputs;
> } else {
> sbe.NumberofSFOutputAttributes = 0;
> sbe.VertexURBEntryReadLength = 1;
> @@ -461,9 +461,9 @@ blorp_emit_sf_config(struct blorp_batch *batch,
>
> sf.VertexURBEntryReadOffset = 1;
> if (prog_data) {
> - sf.NumberofSFOutputAttributes = prog_data->num_varying_inputs;
> - sf.VertexURBEntryReadLength = brw_blorp_get_urb_length(prog_data);
> - sf.ConstantInterpolationEnable = prog_data->flat_inputs;
> + sf.NumberofSFOutputAttributes = prog_data->wm.num_varying_inputs;
> + sf.VertexURBEntryReadLength = blorp_get_urb_length(prog_data);
> + sf.ConstantInterpolationEnable = prog_data->wm.flat_inputs;
> } else {
> sf.NumberofSFOutputAttributes = 0;
> sf.VertexURBEntryReadLength = 1;
> @@ -477,7 +477,7 @@ static void
> blorp_emit_ps_config(struct blorp_batch *batch,
> const struct blorp_params *params)
> {
> - const struct brw_blorp_prog_data *prog_data = params->wm_prog_data;
> + const struct blorp_prog_data *prog_data = params->wm_prog_data;
>
> /* Even when thread dispatch is disabled, max threads (dw5.25:31) must be
> * nonzero to prevent the GPU from hanging. While the documentation doesn't
> @@ -501,16 +501,16 @@ blorp_emit_ps_config(struct blorp_batch *batch,
> }
>
> ps.DispatchGRFStartRegisterForConstantSetupData0 =
> - prog_data->first_curbe_grf_0;
> + prog_data->wm.base.dispatch_grf_start_reg;
> ps.DispatchGRFStartRegisterForConstantSetupData2 =
> - prog_data->first_curbe_grf_2;
> + prog_data->wm.dispatch_grf_start_reg_2;
>
> - ps._8PixelDispatchEnable = prog_data->dispatch_8;
> - ps._16PixelDispatchEnable = prog_data->dispatch_16;
> + ps._8PixelDispatchEnable = prog_data->wm.dispatch_8;
> + ps._16PixelDispatchEnable = prog_data->wm.dispatch_16;
>
> ps.KernelStartPointer0 = params->wm_prog_kernel;
> ps.KernelStartPointer2 =
> - params->wm_prog_kernel + prog_data->ksp_offset_2;
> + params->wm_prog_kernel + prog_data->wm.prog_offset_2;
>
> /* 3DSTATE_PS expects the number of threads per PSD, which is always 64;
> * it implicitly scales for different GT levels (which have some # of
> @@ -552,9 +552,9 @@ blorp_emit_ps_config(struct blorp_batch *batch,
> if (params->src.addr.buffer)
> psx.PixelShaderKillsPixel = true;
>
> - psx.AttributeEnable = prog_data->num_varying_inputs > 0;
> + psx.AttributeEnable = prog_data->wm.num_varying_inputs > 0;
>
> - if (prog_data && prog_data->persample_msaa_dispatch)
> + if (prog_data && prog_data->wm.persample_dispatch)
> psx.PixelShaderIsPerSample = true;
> }
>
> @@ -586,7 +586,7 @@ blorp_emit_ps_config(struct blorp_batch *batch,
> if (params->dst.surf.samples > 1) {
> wm.MultisampleRasterizationMode = MSRASTMODE_ON_PATTERN;
> wm.MultisampleDispatchMode =
> - (prog_data && prog_data->persample_msaa_dispatch) ?
> + (prog_data && prog_data->wm.persample_dispatch) ?
> MSDISPMODE_PERSAMPLE : MSDISPMODE_PERPIXEL;
> } else {
> wm.MultisampleRasterizationMode = MSRASTMODE_OFF_PIXEL;
> @@ -604,18 +604,18 @@ blorp_emit_ps_config(struct blorp_batch *batch,
>
> if (prog_data) {
> ps.DispatchGRFStartRegisterforConstantSetupData0 =
> - prog_data->first_curbe_grf_0;
> + prog_data->wm.base.dispatch_grf_start_reg;
> ps.DispatchGRFStartRegisterforConstantSetupData2 =
> - prog_data->first_curbe_grf_2;
> + prog_data->wm.dispatch_grf_start_reg_2;
>
> ps.KernelStartPointer0 = params->wm_prog_kernel;
> ps.KernelStartPointer2 =
> - params->wm_prog_kernel + prog_data->ksp_offset_2;
> + params->wm_prog_kernel + prog_data->wm.prog_offset_2;
>
> - ps._8PixelDispatchEnable = prog_data->dispatch_8;
> - ps._16PixelDispatchEnable = prog_data->dispatch_16;
> + ps._8PixelDispatchEnable = prog_data->wm.dispatch_8;
> + ps._16PixelDispatchEnable = prog_data->wm.dispatch_16;
>
> - ps.AttributeEnable = prog_data->num_varying_inputs > 0;
> + ps.AttributeEnable = prog_data->wm.num_varying_inputs > 0;
> } else {
> /* Gen7 hardware gets angry if we don't enable at least one dispatch
> * mode, so just enable 16-pixel dispatch if we don't have a program.
> @@ -666,18 +666,18 @@ blorp_emit_ps_config(struct blorp_batch *batch,
> wm.ThreadDispatchEnable = true;
>
> wm.DispatchGRFStartRegisterforConstantSetupData0 =
> - prog_data->first_curbe_grf_0;
> + prog_data->wm.base.dispatch_grf_start_reg;
> wm.DispatchGRFStartRegisterforConstantSetupData2 =
> - prog_data->first_curbe_grf_2;
> + prog_data->wm.dispatch_grf_start_reg_2;
>
> wm.KernelStartPointer0 = params->wm_prog_kernel;
> wm.KernelStartPointer2 =
> - params->wm_prog_kernel + prog_data->ksp_offset_2;
> + params->wm_prog_kernel + prog_data->wm.prog_offset_2;
>
> - wm._8PixelDispatchEnable = prog_data->dispatch_8;
> - wm._16PixelDispatchEnable = prog_data->dispatch_16;
> + wm._8PixelDispatchEnable = prog_data->wm.dispatch_8;
> + wm._16PixelDispatchEnable = prog_data->wm.dispatch_16;
>
> - wm.NumberofSFOutputAttributes = prog_data->num_varying_inputs;
> + wm.NumberofSFOutputAttributes = prog_data->wm.num_varying_inputs;
> }
>
> if (params->src.addr.buffer) {
> @@ -688,7 +688,7 @@ blorp_emit_ps_config(struct blorp_batch *batch,
> if (params->dst.surf.samples > 1) {
> wm.MultisampleRasterizationMode = MSRASTMODE_ON_PATTERN;
> wm.MultisampleDispatchMode =
> - (prog_data && prog_data->persample_msaa_dispatch) ?
> + (prog_data && prog_data->wm.persample_dispatch) ?
> MSDISPMODE_PERSAMPLE : MSDISPMODE_PERPIXEL;
> } else {
> wm.MultisampleRasterizationMode = MSRASTMODE_OFF_PIXEL;
> diff --git a/src/intel/blorp/blorp_priv.h b/src/intel/blorp/blorp_priv.h
> index a88d0f8..70cdfbc 100644
> --- a/src/intel/blorp/blorp_priv.h
> +++ b/src/intel/blorp/blorp_priv.h
> @@ -135,33 +135,15 @@ struct brw_blorp_wm_inputs
> uint32_t pad[1];
> };
>
> -struct brw_blorp_prog_data
> +struct blorp_prog_data
> {
> - bool dispatch_8;
> - bool dispatch_16;
> + struct brw_wm_prog_data wm;
>
> - uint8_t first_curbe_grf_0;
> - uint8_t first_curbe_grf_2;
> -
> - uint32_t ksp_offset_2;
> -
> - /**
> - * True if the WM program should be run in MSDISPMODE_PERSAMPLE with more
> - * than one sample per pixel.
> - */
> - bool persample_msaa_dispatch;
> -
> - /**
> - * Mask of which FS inputs are marked flat by the shader source. This is
> - * needed for setting up 3DSTATE_SF/SBE.
> - */
> - uint32_t flat_inputs;
> - unsigned num_varying_inputs;
> uint64_t inputs_read;
> };
>
> static inline unsigned
> -brw_blorp_get_urb_length(const struct brw_blorp_prog_data *prog_data)
> +blorp_get_urb_length(const struct blorp_prog_data *prog_data)
> {
> if (prog_data == NULL)
> return 1;
> @@ -170,7 +152,7 @@ brw_blorp_get_urb_length(const struct brw_blorp_prog_data *prog_data)
> *
> * read_length = ceiling((max_source_attr+1)/2)
> */
> - return MAX2((prog_data->num_varying_inputs + 1) / 2, 1);
> + return MAX2((prog_data->wm.num_varying_inputs + 1) / 2, 1);
> }
>
> struct blorp_params
> @@ -190,7 +172,7 @@ struct blorp_params
> unsigned num_draw_buffers;
> unsigned num_layers;
> uint32_t wm_prog_kernel;
> - struct brw_blorp_prog_data *wm_prog_data;
> + struct blorp_prog_data *wm_prog_data;
> };
>
> void blorp_params_init(struct blorp_params *params);
> @@ -306,7 +288,7 @@ const unsigned *
> brw_blorp_compile_nir_shader(struct blorp_context *blorp, struct nir_shader *nir,
> const struct brw_wm_prog_key *wm_key,
> bool use_repclear,
> - struct brw_blorp_prog_data *prog_data,
> + struct blorp_prog_data *prog_data,
> unsigned *program_size);
>
> /** \} */
> --
> 2.5.0.400.gff86faf
>
> _______________________________________________
> mesa-dev mailing list
> mesa-dev at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/mesa-dev
More information about the mesa-dev
mailing list