Mesa (main): radv: reduce number of emitted DWORDS for contiguous context registers

[GitLab Mirror]

Module: Mesa
Branch: main
Commit: f693b94d2fac4590475a4ff95deab3c639cc14b4
URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=f693b94d2fac4590475a4ff95deab3c639cc14b4

Author: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Date:   Wed Jul 28 14:26:32 2021 +0200

radv: reduce number of emitted DWORDS for contiguous context registers

radeon_set_context_reg() emits 3 DWORDS, while
radeon_set_context_reg_seq() + 2 values only needs 4 DWORDS.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas at basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12099>

---

 src/amd/vulkan/radv_cmd_buffer.c |  5 +++--
 src/amd/vulkan/radv_pipeline.c   | 18 +++++++++++-------
 2 files changed, 14 insertions(+), 9 deletions(-)

diff --git a/src/amd/vulkan/radv_cmd_buffer.c b/src/amd/vulkan/radv_cmd_buffer.c
index fa395720d06..3e9f128a7e3 100644
--- a/src/amd/vulkan/radv_cmd_buffer.c
+++ b/src/amd/vulkan/radv_cmd_buffer.c
@@ -1465,8 +1465,9 @@ radv_emit_depth_bounds(struct radv_cmd_buffer *cmd_buffer)
 {
    struct radv_dynamic_state *d = &cmd_buffer->state.dynamic;
 
-   radeon_set_context_reg(cmd_buffer->cs, R_028020_DB_DEPTH_BOUNDS_MIN, fui(d->depth_bounds.min));
-   radeon_set_context_reg(cmd_buffer->cs, R_028024_DB_DEPTH_BOUNDS_MAX, fui(d->depth_bounds.max));
+   radeon_set_context_reg_seq(cmd_buffer->cs, R_028020_DB_DEPTH_BOUNDS_MIN, 2);
+   radeon_emit(cmd_buffer->cs, fui(d->depth_bounds.min));
+   radeon_emit(cmd_buffer->cs, fui(d->depth_bounds.max));
 }
 
 static void
diff --git a/src/amd/vulkan/radv_pipeline.c b/src/amd/vulkan/radv_pipeline.c
index a756d894bc1..0c1ddc1561b 100644
--- a/src/amd/vulkan/radv_pipeline.c
+++ b/src/amd/vulkan/radv_pipeline.c
@@ -4204,8 +4204,10 @@ radv_pipeline_generate_depth_stencil_state(struct radeon_cmdbuf *ctx_cs,
    }
 
    radeon_set_context_reg(ctx_cs, R_028000_DB_RENDER_CONTROL, db_render_control);
-   radeon_set_context_reg(ctx_cs, R_02800C_DB_RENDER_OVERRIDE, db_render_override);
-   radeon_set_context_reg(ctx_cs, R_028010_DB_RENDER_OVERRIDE2, db_render_override2);
+
+   radeon_set_context_reg_seq(ctx_cs, R_02800C_DB_RENDER_OVERRIDE, 2);
+   radeon_emit(ctx_cs, db_render_override);
+   radeon_emit(ctx_cs, db_render_override2);
 }
 
 static void
@@ -4275,10 +4277,12 @@ radv_pipeline_generate_multisample_state(struct radeon_cmdbuf *ctx_cs,
    radeon_emit(ctx_cs, ms->pa_sc_aa_mask[1]);
 
    radeon_set_context_reg(ctx_cs, R_028804_DB_EQAA, ms->db_eqaa);
-   radeon_set_context_reg(ctx_cs, R_028A48_PA_SC_MODE_CNTL_0, ms->pa_sc_mode_cntl_0);
-   radeon_set_context_reg(ctx_cs, R_028A4C_PA_SC_MODE_CNTL_1, ms->pa_sc_mode_cntl_1);
    radeon_set_context_reg(ctx_cs, R_028BE0_PA_SC_AA_CONFIG, ms->pa_sc_aa_config);
 
+   radeon_set_context_reg_seq(ctx_cs, R_028A48_PA_SC_MODE_CNTL_0, 2);
+   radeon_emit(ctx_cs, ms->pa_sc_mode_cntl_0);
+   radeon_emit(ctx_cs, ms->pa_sc_mode_cntl_1);
+
    /* The exclusion bits can be set to improve rasterization efficiency
     * if no sample lies on the pixel boundary (-8 sample offset). It's
     * currently always TRUE because the driver doesn't support 16 samples.
@@ -4974,9 +4978,9 @@ radv_pipeline_generate_fragment_shader(struct radeon_cmdbuf *ctx_cs, struct rade
    radeon_set_context_reg(ctx_cs, R_02880C_DB_SHADER_CONTROL,
                           radv_compute_db_shader_control(pipeline->device, pipeline, ps));
 
-   radeon_set_context_reg(ctx_cs, R_0286CC_SPI_PS_INPUT_ENA, ps->config.spi_ps_input_ena);
-
-   radeon_set_context_reg(ctx_cs, R_0286D0_SPI_PS_INPUT_ADDR, ps->config.spi_ps_input_addr);
+   radeon_set_context_reg_seq(ctx_cs, R_0286CC_SPI_PS_INPUT_ENA, 2);
+   radeon_emit(ctx_cs, ps->config.spi_ps_input_ena);
+   radeon_emit(ctx_cs, ps->config.spi_ps_input_addr);
 
    radeon_set_context_reg(
       ctx_cs, R_0286D8_SPI_PS_IN_CONTROL,