From 416dd873e8e0cbca1477344a94a7c2e0c9a6c25b Mon Sep 17 00:00:00 2001 From: =?utf8?q?Kristian=20H=C3=B8gsberg?= Date: Mon, 7 Jul 2014 14:43:30 -0700 Subject: [PATCH] i965: Assign PS kernel start pointers when we decide which kernels to use MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit Right now we decide which kernels to use and the GRF start offsets in one place and emit the kernel pointers later. The logic of how to map 8, 16 and 32 kernels to kernel start pointers follows the same logic as which GRF start offsets to use, so lets figure out these two things in one place. Signed-off-by: Kristian Høgsberg Reviewed-by: Kenneth Graunke Reviewed-by: Ben Widawsky --- src/mesa/drivers/dri/i965/gen6_wm_state.c | 19 ++++++++++--------- src/mesa/drivers/dri/i965/gen7_wm_state.c | 18 ++++++++++-------- src/mesa/drivers/dri/i965/gen8_ps_state.c | 14 ++++++++------ 3 files changed, 28 insertions(+), 23 deletions(-) diff --git a/src/mesa/drivers/dri/i965/gen6_wm_state.c b/src/mesa/drivers/dri/i965/gen6_wm_state.c index 047e036f5b6..de95db8b8ab 100644 --- a/src/mesa/drivers/dri/i965/gen6_wm_state.c +++ b/src/mesa/drivers/dri/i965/gen6_wm_state.c @@ -71,7 +71,7 @@ upload_wm_state(struct brw_context *brw) struct gl_context *ctx = &brw->ctx; const struct brw_fragment_program *fp = brw_fragment_program_const(brw->fragment_program); - uint32_t dw2, dw4, dw5, dw6; + uint32_t dw2, dw4, dw5, dw6, ksp0, ksp2; /* _NEW_BUFFERS */ bool multisampled_fbo = ctx->DrawBuffer->Visual.samples > 1; @@ -109,7 +109,7 @@ upload_wm_state(struct brw_context *brw) ADVANCE_BATCH(); } - dw2 = dw4 = dw5 = dw6 = 0; + dw2 = dw4 = dw5 = dw6 = ksp2 = 0; dw4 |= GEN6_WM_STATISTICS_ENABLE; dw5 |= GEN6_WM_LINE_AA_WIDTH_1_0; dw5 |= GEN6_WM_LINE_END_CAP_AA_WIDTH_0_5; @@ -151,14 +151,19 @@ upload_wm_state(struct brw_context *brw) GEN6_WM_DISPATCH_START_GRF_SHIFT_0); dw4 |= (brw->wm.prog_data->dispatch_grf_start_reg_16 << GEN6_WM_DISPATCH_START_GRF_SHIFT_2); - } else + ksp0 = brw->wm.base.prog_offset; + ksp2 = brw->wm.base.prog_offset + brw->wm.prog_data->prog_offset_16; + } else { dw4 |= (brw->wm.prog_data->dispatch_grf_start_reg_16 << GEN6_WM_DISPATCH_START_GRF_SHIFT_0); + ksp0 = brw->wm.base.prog_offset + brw->wm.prog_data->prog_offset_16; + } } else { dw5 |= GEN6_WM_8_DISPATCH_ENABLE; dw4 |= (brw->wm.prog_data->base.dispatch_grf_start_reg << GEN6_WM_DISPATCH_START_GRF_SHIFT_0); + ksp0 = brw->wm.base.prog_offset; } /* CACHE_NEW_WM_PROG | _NEW_COLOR */ @@ -277,10 +282,7 @@ upload_wm_state(struct brw_context *brw) BEGIN_BATCH(9); OUT_BATCH(_3DSTATE_WM << 16 | (9 - 2)); - if (brw->wm.prog_data->prog_offset_16 && min_inv_per_frag > 1) - OUT_BATCH(brw->wm.base.prog_offset + brw->wm.prog_data->prog_offset_16); - else - OUT_BATCH(brw->wm.base.prog_offset); + OUT_BATCH(ksp0); OUT_BATCH(dw2); if (brw->wm.prog_data->total_scratch) { OUT_RELOC(brw->wm.base.scratch_bo, @@ -293,8 +295,7 @@ upload_wm_state(struct brw_context *brw) OUT_BATCH(dw5); OUT_BATCH(dw6); OUT_BATCH(0); /* kernel 1 pointer */ - /* kernel 2 pointer */ - OUT_BATCH(brw->wm.base.prog_offset + brw->wm.prog_data->prog_offset_16); + OUT_BATCH(ksp2); ADVANCE_BATCH(); } diff --git a/src/mesa/drivers/dri/i965/gen7_wm_state.c b/src/mesa/drivers/dri/i965/gen7_wm_state.c index 27a3f44cfc8..c03d19db163 100644 --- a/src/mesa/drivers/dri/i965/gen7_wm_state.c +++ b/src/mesa/drivers/dri/i965/gen7_wm_state.c @@ -139,11 +139,11 @@ static void upload_ps_state(struct brw_context *brw) { struct gl_context *ctx = &brw->ctx; - uint32_t dw2, dw4, dw5; + uint32_t dw2, dw4, dw5, ksp0, ksp2; const int max_threads_shift = brw->is_haswell ? HSW_PS_MAX_THREADS_SHIFT : IVB_PS_MAX_THREADS_SHIFT; - dw2 = dw4 = dw5 = 0; + dw2 = dw4 = dw5 = ksp2 = 0; dw2 |= (ALIGN(brw->wm.base.sampler_count, 4) / 4) << GEN7_PS_SAMPLER_COUNT_SHIFT; @@ -231,22 +231,24 @@ upload_ps_state(struct brw_context *brw) GEN7_PS_DISPATCH_START_GRF_SHIFT_0); dw5 |= (brw->wm.prog_data->dispatch_grf_start_reg_16 << GEN7_PS_DISPATCH_START_GRF_SHIFT_2); - } else + ksp0 = brw->wm.base.prog_offset; + ksp2 = brw->wm.base.prog_offset + brw->wm.prog_data->prog_offset_16; + } else { dw5 |= (brw->wm.prog_data->dispatch_grf_start_reg_16 << GEN7_PS_DISPATCH_START_GRF_SHIFT_0); + ksp0 = brw->wm.base.prog_offset + brw->wm.prog_data->prog_offset_16; + } } else { dw4 |= GEN7_PS_8_DISPATCH_ENABLE; dw5 |= (brw->wm.prog_data->base.dispatch_grf_start_reg << GEN7_PS_DISPATCH_START_GRF_SHIFT_0); + ksp0 = brw->wm.base.prog_offset; } BEGIN_BATCH(8); OUT_BATCH(_3DSTATE_PS << 16 | (8 - 2)); - if (brw->wm.prog_data->prog_offset_16 && min_inv_per_frag > 1) - OUT_BATCH(brw->wm.base.prog_offset + brw->wm.prog_data->prog_offset_16); - else - OUT_BATCH(brw->wm.base.prog_offset); + OUT_BATCH(ksp0); OUT_BATCH(dw2); if (brw->wm.prog_data->total_scratch) { OUT_RELOC(brw->wm.base.scratch_bo, @@ -258,7 +260,7 @@ upload_ps_state(struct brw_context *brw) OUT_BATCH(dw4); OUT_BATCH(dw5); OUT_BATCH(0); /* kernel 1 pointer */ - OUT_BATCH(brw->wm.base.prog_offset + brw->wm.prog_data->prog_offset_16); + OUT_BATCH(ksp2); ADVANCE_BATCH(); } diff --git a/src/mesa/drivers/dri/i965/gen8_ps_state.c b/src/mesa/drivers/dri/i965/gen8_ps_state.c index 3d6d7f0f7a2..f58d49c55cb 100644 --- a/src/mesa/drivers/dri/i965/gen8_ps_state.c +++ b/src/mesa/drivers/dri/i965/gen8_ps_state.c @@ -134,7 +134,7 @@ static void upload_ps_state(struct brw_context *brw) { struct gl_context *ctx = &brw->ctx; - uint32_t dw3 = 0, dw6 = 0, dw7 = 0; + uint32_t dw3 = 0, dw6 = 0, dw7 = 0, ksp0, ksp2 = 0; /* Initialize the execution mask with VMask. Otherwise, derivatives are * incorrect for subspans where some of the pixels are unlit. We believe @@ -203,22 +203,24 @@ upload_ps_state(struct brw_context *brw) GEN7_PS_DISPATCH_START_GRF_SHIFT_0); dw7 |= (brw->wm.prog_data->dispatch_grf_start_reg_16 << GEN7_PS_DISPATCH_START_GRF_SHIFT_2); + ksp0 = brw->wm.base.prog_offset; + ksp2 = brw->wm.base.prog_offset + brw->wm.prog_data->prog_offset_16; } else { dw7 |= (brw->wm.prog_data->dispatch_grf_start_reg_16 << GEN7_PS_DISPATCH_START_GRF_SHIFT_0); + + ksp0 = brw->wm.base.prog_offset + brw->wm.prog_data->prog_offset_16; } } else { dw6 |= GEN7_PS_8_DISPATCH_ENABLE; dw7 |= (brw->wm.prog_data->base.dispatch_grf_start_reg << GEN7_PS_DISPATCH_START_GRF_SHIFT_0); + ksp0 = brw->wm.base.prog_offset; } BEGIN_BATCH(12); OUT_BATCH(_3DSTATE_PS << 16 | (12 - 2)); - if (brw->wm.prog_data->prog_offset_16 && min_invocations_per_fragment > 1) - OUT_BATCH(brw->wm.base.prog_offset + brw->wm.prog_data->prog_offset_16); - else - OUT_BATCH(brw->wm.base.prog_offset); + OUT_BATCH(ksp0); OUT_BATCH(0); OUT_BATCH(dw3); if (brw->wm.prog_data->total_scratch) { @@ -233,7 +235,7 @@ upload_ps_state(struct brw_context *brw) OUT_BATCH(dw7); OUT_BATCH(0); /* kernel 1 pointer */ OUT_BATCH(0); - OUT_BATCH(brw->wm.base.prog_offset + brw->wm.prog_data->prog_offset_16); + OUT_BATCH(ksp2); OUT_BATCH(0); ADVANCE_BATCH(); } -- 2.30.2