radv_buffer_get_va(cmd_buffer->upload.upload_bo);
cmd_buffer->gfx9_fence_va += fence_offset;
- /* Allocate a buffer for the EOP bug on GFX9. */
- radv_cmd_buffer_upload_alloc(cmd_buffer, 16 * num_db, 8,
- &eop_bug_offset, &fence_ptr);
- cmd_buffer->gfx9_eop_bug_va =
- radv_buffer_get_va(cmd_buffer->upload.upload_bo);
- cmd_buffer->gfx9_eop_bug_va += eop_bug_offset;
+ if (cmd_buffer->device->physical_device->rad_info.chip_class == GFX9) {
+ /* Allocate a buffer for the EOP bug on GFX9. */
+ radv_cmd_buffer_upload_alloc(cmd_buffer, 16 * num_db, 8,
+ &eop_bug_offset, &fence_ptr);
+ cmd_buffer->gfx9_eop_bug_va =
+ radv_buffer_get_va(cmd_buffer->upload.upload_bo);
+ cmd_buffer->gfx9_eop_bug_va += eop_bug_offset;
+ }
}
cmd_buffer->status = RADV_CMD_BUFFER_STATUS_INITIAL;
cmd_buffer->state.context_roll_without_scissor_emitted = true;
}
+static void
+radv_update_binning_state(struct radv_cmd_buffer *cmd_buffer,
+ struct radv_pipeline *pipeline)
+{
+ const struct radv_pipeline *old_pipeline = cmd_buffer->state.emitted_pipeline;
+
+
+ if (pipeline->device->physical_device->rad_info.chip_class < GFX9)
+ return;
+
+ if (old_pipeline &&
+ old_pipeline->graphics.binning.pa_sc_binner_cntl_0 == pipeline->graphics.binning.pa_sc_binner_cntl_0 &&
+ old_pipeline->graphics.binning.db_dfsm_control == pipeline->graphics.binning.db_dfsm_control)
+ return;
+
+ bool binning_flush = false;
+ if (cmd_buffer->device->physical_device->rad_info.family == CHIP_VEGA12 ||
+ cmd_buffer->device->physical_device->rad_info.family == CHIP_VEGA20 ||
+ cmd_buffer->device->physical_device->rad_info.family == CHIP_RAVEN2 ||
+ cmd_buffer->device->physical_device->rad_info.chip_class >= GFX10) {
+ binning_flush = !old_pipeline ||
+ G_028C44_BINNING_MODE(old_pipeline->graphics.binning.pa_sc_binner_cntl_0) !=
+ G_028C44_BINNING_MODE(pipeline->graphics.binning.pa_sc_binner_cntl_0);
+ }
+
+ radeon_set_context_reg(cmd_buffer->cs, R_028C44_PA_SC_BINNER_CNTL_0,
+ pipeline->graphics.binning.pa_sc_binner_cntl_0 |
+ S_028C44_FLUSH_ON_BINNING_TRANSITION(!!binning_flush));
+
+ if (cmd_buffer->device->physical_device->rad_info.chip_class >= GFX10) {
+ radeon_set_context_reg(cmd_buffer->cs, R_028038_DB_DFSM_CONTROL,
+ pipeline->graphics.binning.db_dfsm_control);
+ } else {
+ radeon_set_context_reg(cmd_buffer->cs, R_028060_DB_DFSM_CONTROL,
+ pipeline->graphics.binning.db_dfsm_control);
+ }
+
+ cmd_buffer->state.context_roll_without_scissor_emitted = true;
+}
+
+
static void
radv_emit_shader_prefetch(struct radv_cmd_buffer *cmd_buffer,
struct radv_shader_variant *shader)
if (mask & RADV_PREFETCH_GS) {
radv_emit_shader_prefetch(cmd_buffer,
pipeline->shaders[MESA_SHADER_GEOMETRY]);
- if (pipeline->gs_copy_shader)
+ if (radv_pipeline_has_gs_copy_shader(pipeline))
radv_emit_shader_prefetch(cmd_buffer, pipeline->gs_copy_shader);
}
return;
radv_update_multisample_state(cmd_buffer, pipeline);
+ radv_update_binning_state(cmd_buffer, pipeline);
cmd_buffer->scratch_size_needed =
MAX2(cmd_buffer->scratch_size_needed,
pipeline->shaders[i]->bo);
}
- if (radv_pipeline_has_gs(pipeline) && pipeline->gs_copy_shader)
+ if (radv_pipeline_has_gs_copy_shader(pipeline))
radv_cs_add_buffer(cmd_buffer->device->ws, cmd_buffer->cs,
pipeline->gs_copy_shader->bo);
cb->cb_color_attrib2);
radeon_set_context_reg(cmd_buffer->cs, R_028EE0_CB_COLOR0_ATTRIB3 + index * 4,
cb->cb_color_attrib3);
- } else if (cmd_buffer->device->physical_device->rad_info.chip_class >= GFX9) {
+ } else if (cmd_buffer->device->physical_device->rad_info.chip_class == GFX9) {
radeon_set_context_reg_seq(cmd_buffer->cs, R_028C60_CB_COLOR0_BASE + index * 0x3c, 11);
radeon_emit(cmd_buffer->cs, cb->cb_color_base);
radeon_emit(cmd_buffer->cs, S_028C64_BASE_256B(cb->cb_color_base >> 32));
uint32_t db_z_info = ds->db_z_info;
uint32_t db_z_info_reg;
- if (!radv_image_is_tc_compat_htile(image))
+ if (!cmd_buffer->device->physical_device->has_tc_compat_zrange_bug ||
+ !radv_image_is_tc_compat_htile(image))
return;
if (!radv_layout_has_htile(image, layout,
radeon_emit(cmd_buffer->cs, ds->db_z_read_base >> 32);
radeon_emit(cmd_buffer->cs, ds->db_stencil_read_base >> 32);
radeon_emit(cmd_buffer->cs, ds->db_htile_data_base >> 32);
- } else if (cmd_buffer->device->physical_device->rad_info.chip_class >= GFX9) {
+ } else if (cmd_buffer->device->physical_device->rad_info.chip_class == GFX9) {
radeon_set_context_reg_seq(cmd_buffer->cs, R_028014_DB_HTILE_DATA_BASE, 3);
radeon_emit(cmd_buffer->cs, ds->db_htile_data_base);
radeon_emit(cmd_buffer->cs, S_028018_BASE_HI(ds->db_htile_data_base >> 32));
{
struct radeon_cmdbuf *cs = cmd_buffer->cs;
uint64_t va = radv_buffer_get_va(image->bo);
+
+ if (!cmd_buffer->device->physical_device->has_tc_compat_zrange_bug)
+ return;
+
va += image->offset + image->tc_compat_zrange_offset;
radeon_emit(cs, PKT3(PKT3_WRITE_DATA, 3, cmd_buffer->state.predicating));
struct radv_attachment_info *att = &framebuffer->attachments[idx];
struct radv_image *image = att->attachment->image;
radv_cs_add_buffer(cmd_buffer->device->ws, cmd_buffer->cs, att->attachment->bo);
- MAYBE_UNUSED uint32_t queue_mask = radv_image_queue_family_mask(image,
+ ASSERTED uint32_t queue_mask = radv_image_queue_family_mask(image,
cmd_buffer->queue_family_index,
cmd_buffer->queue_family_index);
/* We currently don't support writing decompressed HTILE */
S_028424_DISABLE_CONSTANT_ENCODE_REG(disable_constant_encode));
}
- if (cmd_buffer->device->dfsm_allowed) {
+ if (cmd_buffer->device->pbb_allowed) {
radeon_emit(cmd_buffer->cs, PKT3(PKT3_EVENT_WRITE, 0, 0));
radeon_emit(cmd_buffer->cs, EVENT_TYPE(V_028A90_BREAK_BATCH) | EVENT_INDEX(0));
}
if (flush_indirect_descriptors)
radv_flush_indirect_descriptor_sets(cmd_buffer, bind_point);
- MAYBE_UNUSED unsigned cdw_max = radeon_check_space(cmd_buffer->device->ws,
+ ASSERTED unsigned cdw_max = radeon_check_space(cmd_buffer->device->ws,
cmd_buffer->cs,
MAX_SETS * MESA_SHADER_STAGES * 4);
va = radv_buffer_get_va(cmd_buffer->upload.upload_bo);
va += offset;
- MAYBE_UNUSED unsigned cdw_max =
+ ASSERTED unsigned cdw_max =
radeon_check_space(cmd_buffer->device->ws,
cmd_buffer->cs, MESA_SHADER_STAGES * 4);
base_reg + loc->sgpr_idx * 4, va, false);
}
- if (pipeline->gs_copy_shader) {
+ if (radv_pipeline_has_gs_copy_shader(pipeline)) {
loc = &pipeline->gs_copy_shader->info.user_sgprs_locs.shader_data[AC_UD_STREAMOUT_BUFFERS];
if (loc->sgpr_idx != -1) {
base_reg = R_00B130_SPI_SHADER_USER_DATA_VS_0;
desc[3] = S_008F0C_DST_SEL_X(V_008F0C_SQ_SEL_X) |
S_008F0C_DST_SEL_Y(V_008F0C_SQ_SEL_Y) |
S_008F0C_DST_SEL_Z(V_008F0C_SQ_SEL_Z) |
- S_008F0C_DST_SEL_W(V_008F0C_SQ_SEL_W) |
- S_008F0C_DATA_FORMAT(V_008F0C_BUF_DATA_FORMAT_32);
+ S_008F0C_DST_SEL_W(V_008F0C_SQ_SEL_W);
+
+ if (cmd_buffer->device->physical_device->rad_info.chip_class >= GFX10) {
+ desc[3] |= S_008F0C_FORMAT(V_008F0C_IMG_FORMAT_32_FLOAT) |
+ S_008F0C_OOB_SELECT(3) |
+ S_008F0C_RESOURCE_LEVEL(1);
+ } else {
+ desc[3] |= S_008F0C_DATA_FORMAT(V_008F0C_BUF_DATA_FORMAT_32);
+ }
}
va = radv_buffer_get_va(cmd_buffer->upload.upload_bo);
uint64_t strmout_buffer_offset;
};
+static uint32_t
+radv_get_primitive_reset_index(struct radv_cmd_buffer *cmd_buffer)
+{
+ switch (cmd_buffer->state.index_type) {
+ case V_028A7C_VGT_INDEX_8:
+ return 0xffu;
+ case V_028A7C_VGT_INDEX_16:
+ return 0xffffu;
+ case V_028A7C_VGT_INDEX_32:
+ return 0xffffffffu;
+ default:
+ unreachable("invalid index type");
+ }
+}
+
static void
si_emit_ia_multi_vgt_param(struct radv_cmd_buffer *cmd_buffer,
bool instanced_draw, bool indirect_draw,
draw_vertex_count);
if (state->last_ia_multi_vgt_param != ia_multi_vgt_param) {
- if (info->chip_class >= GFX9) {
+ if (info->chip_class == GFX9) {
radeon_set_uconfig_reg_idx(cmd_buffer->device->physical_device,
cs,
R_030960_IA_MULTI_VGT_PARAM,
if (primitive_reset_en) {
uint32_t primitive_reset_index =
- state->index_type ? 0xffffffffu : 0xffffu;
+ radv_get_primitive_reset_index(cmd_buffer);
if (primitive_reset_index != state->last_primitive_reset_index) {
radeon_set_context_reg(cs,
cmd_buffer->state.dirty |= RADV_CMD_DIRTY_VERTEX_BUFFER;
}
+static uint32_t
+vk_to_index_type(VkIndexType type)
+{
+ switch (type) {
+ case VK_INDEX_TYPE_UINT8_EXT:
+ return V_028A7C_VGT_INDEX_8;
+ case VK_INDEX_TYPE_UINT16:
+ return V_028A7C_VGT_INDEX_16;
+ case VK_INDEX_TYPE_UINT32:
+ return V_028A7C_VGT_INDEX_32;
+ default:
+ unreachable("invalid index type");
+ }
+}
+
+static uint32_t
+radv_get_vgt_index_size(uint32_t type)
+{
+ switch (type) {
+ case V_028A7C_VGT_INDEX_8:
+ return 1;
+ case V_028A7C_VGT_INDEX_16:
+ return 2;
+ case V_028A7C_VGT_INDEX_32:
+ return 4;
+ default:
+ unreachable("invalid index type");
+ }
+}
+
void radv_CmdBindIndexBuffer(
VkCommandBuffer commandBuffer,
VkBuffer buffer,
cmd_buffer->state.index_buffer = index_buffer;
cmd_buffer->state.index_offset = offset;
- cmd_buffer->state.index_type = indexType; /* vk matches hw */
+ cmd_buffer->state.index_type = vk_to_index_type(indexType);
cmd_buffer->state.index_va = radv_buffer_get_va(index_buffer->bo);
cmd_buffer->state.index_va += index_buffer->offset + offset;
- int index_size_shift = cmd_buffer->state.index_type ? 2 : 1;
- cmd_buffer->state.max_index_count = (index_buffer->size - offset) >> index_size_shift;
+ int index_size = radv_get_vgt_index_size(indexType);
+ cmd_buffer->state.max_index_count = (index_buffer->size - offset) / index_size;
cmd_buffer->state.dirty |= RADV_CMD_DIRTY_INDEX_BUFFER;
radv_cs_add_buffer(cmd_buffer->device->ws, cmd_buffer->cs, index_buffer->bo);
}
* because it is invalid, according to Vulkan spec.
*/
for (int i = 0; i < descriptorWriteCount; i++) {
- MAYBE_UNUSED const VkWriteDescriptorSet *writeset = &pDescriptorWrites[i];
+ ASSERTED const VkWriteDescriptorSet *writeset = &pDescriptorWrites[i];
assert(writeset->descriptorType != VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT);
}
{
RADV_FROM_HANDLE(radv_cmd_buffer, cmd_buffer, commandBuffer);
struct radv_cmd_state *state = &cmd_buffer->state;
- MAYBE_UNUSED const uint32_t total_count = firstViewport + viewportCount;
+ ASSERTED const uint32_t total_count = firstViewport + viewportCount;
assert(firstViewport < MAX_VIEWPORTS);
assert(total_count >= 1 && total_count <= MAX_VIEWPORTS);
{
RADV_FROM_HANDLE(radv_cmd_buffer, cmd_buffer, commandBuffer);
struct radv_cmd_state *state = &cmd_buffer->state;
- MAYBE_UNUSED const uint32_t total_count = firstScissor + scissorCount;
+ ASSERTED const uint32_t total_count = firstScissor + scissorCount;
assert(firstScissor < MAX_SCISSORS);
assert(total_count >= 1 && total_count <= MAX_SCISSORS);
{
RADV_FROM_HANDLE(radv_cmd_buffer, cmd_buffer, commandBuffer);
struct radv_cmd_state *state = &cmd_buffer->state;
- MAYBE_UNUSED const uint32_t total_count = firstDiscardRectangle + discardRectangleCount;
+ ASSERTED const uint32_t total_count = firstDiscardRectangle + discardRectangleCount;
assert(firstDiscardRectangle < MAX_DISCARD_RECTANGLES);
assert(total_count >= 1 && total_count <= MAX_DISCARD_RECTANGLES);
struct radv_cmd_state *state = &cmd_buffer->state;
struct radv_subpass *subpass = &state->pass->subpasses[subpass_id];
- MAYBE_UNUSED unsigned cdw_max = radeon_check_space(cmd_buffer->device->ws,
+ ASSERTED unsigned cdw_max = radeon_check_space(cmd_buffer->device->ws,
cmd_buffer->cs, 4096);
radv_subpass_barrier(cmd_buffer, &subpass->start_barrier);
radeon_set_sh_reg(cmd_buffer->cs, base_reg + loc->sgpr_idx * 4, index);
}
- if (pipeline->gs_copy_shader) {
+ if (radv_pipeline_has_gs_copy_shader(pipeline)) {
struct radv_userdata_info *loc = &pipeline->gs_copy_shader->info.user_sgprs_locs.shader_data[AC_UD_VIEW_INDEX];
if (loc->sgpr_idx != -1) {
uint32_t base_reg = R_00B130_SPI_SHADER_USER_DATA_VS_0;
}
if (info->indexed) {
- int index_size = state->index_type ? 4 : 2;
+ int index_size = radv_get_vgt_index_size(state->index_type);
uint64_t index_va;
index_va = state->index_va;
if (cmd_buffer->state.dirty & used_states)
return true;
+ uint32_t primitive_reset_index =
+ radv_get_primitive_reset_index(cmd_buffer);
+
if (info->indexed && state->pipeline->graphics.prim_restart_enable &&
- (state->index_type ? 0xffffffffu : 0xffffu) != state->last_primitive_reset_index)
+ primitive_reset_index != state->last_primitive_reset_index)
return true;
return false;
(cmd_buffer->state.dirty & RADV_CMD_DIRTY_PIPELINE) &&
cmd_buffer->state.pipeline != cmd_buffer->state.emitted_pipeline;
- MAYBE_UNUSED unsigned cdw_max =
+ ASSERTED unsigned cdw_max =
radeon_check_space(cmd_buffer->device->ws,
cmd_buffer->cs, 4096);
loc = radv_lookup_user_sgpr(pipeline, MESA_SHADER_COMPUTE,
AC_UD_CS_GRID_SIZE);
- MAYBE_UNUSED unsigned cdw_max = radeon_check_space(ws, cs, 25);
+ ASSERTED unsigned cdw_max = radeon_check_space(ws, cs, 25);
if (info->indirect) {
uint64_t va = radv_buffer_get_va(info->indirect->bo);
radv_cs_add_buffer(cmd_buffer->device->ws, cs, event->bo);
- MAYBE_UNUSED unsigned cdw_max = radeon_check_space(cmd_buffer->device->ws, cs, 7);
+ ASSERTED unsigned cdw_max = radeon_check_space(cmd_buffer->device->ws, cs, 7);
radv_cp_wait_mem(cs, WAIT_REG_MEM_EQUAL, va, 1, 0xffffffff);
assert(cmd_buffer->cs->cdw <= cdw_max);
radv_cs_add_buffer(cmd_buffer->device->ws, cs, event->bo);
- MAYBE_UNUSED unsigned cdw_max = radeon_check_space(cmd_buffer->device->ws, cs, 21);
+ ASSERTED unsigned cdw_max = radeon_check_space(cmd_buffer->device->ws, cs, 21);
/* Flags that only require a top-of-pipe event. */
VkPipelineStageFlags top_of_pipe_flags =
cmd_buffer->device->physical_device->rad_info.chip_class,
radv_cmd_buffer_uses_mec(cmd_buffer),
V_028A90_BOTTOM_OF_PIPE_TS, 0,
+ EOP_DST_SEL_MEM,
EOP_DATA_SEL_VALUE_32BIT, va, value,
cmd_buffer->gfx9_eop_bug_va);
}
radeon_emit(cs, 4); /* poll interval */
}
-void radv_CmdBeginTransformFeedbackEXT(
- VkCommandBuffer commandBuffer,
- uint32_t firstCounterBuffer,
- uint32_t counterBufferCount,
- const VkBuffer* pCounterBuffers,
- const VkDeviceSize* pCounterBufferOffsets)
+static void
+radv_emit_streamout_begin(struct radv_cmd_buffer *cmd_buffer,
+ uint32_t firstCounterBuffer,
+ uint32_t counterBufferCount,
+ const VkBuffer *pCounterBuffers,
+ const VkDeviceSize *pCounterBufferOffsets)
+
{
- RADV_FROM_HANDLE(radv_cmd_buffer, cmd_buffer, commandBuffer);
struct radv_streamout_binding *sb = cmd_buffer->streamout_bindings;
struct radv_streamout_state *so = &cmd_buffer->state.streamout;
struct radeon_cmdbuf *cs = cmd_buffer->cs;
radv_set_streamout_enable(cmd_buffer, true);
}
-void radv_CmdEndTransformFeedbackEXT(
+void radv_CmdBeginTransformFeedbackEXT(
VkCommandBuffer commandBuffer,
uint32_t firstCounterBuffer,
uint32_t counterBufferCount,
const VkDeviceSize* pCounterBufferOffsets)
{
RADV_FROM_HANDLE(radv_cmd_buffer, cmd_buffer, commandBuffer);
+
+ radv_emit_streamout_begin(cmd_buffer,
+ firstCounterBuffer, counterBufferCount,
+ pCounterBuffers, pCounterBufferOffsets);
+}
+
+static void
+radv_emit_streamout_end(struct radv_cmd_buffer *cmd_buffer,
+ uint32_t firstCounterBuffer,
+ uint32_t counterBufferCount,
+ const VkBuffer *pCounterBuffers,
+ const VkDeviceSize *pCounterBufferOffsets)
+{
struct radv_streamout_state *so = &cmd_buffer->state.streamout;
struct radeon_cmdbuf *cs = cmd_buffer->cs;
uint32_t i;
radv_set_streamout_enable(cmd_buffer, false);
}
+void radv_CmdEndTransformFeedbackEXT(
+ VkCommandBuffer commandBuffer,
+ uint32_t firstCounterBuffer,
+ uint32_t counterBufferCount,
+ const VkBuffer* pCounterBuffers,
+ const VkDeviceSize* pCounterBufferOffsets)
+{
+ RADV_FROM_HANDLE(radv_cmd_buffer, cmd_buffer, commandBuffer);
+
+ radv_emit_streamout_end(cmd_buffer,
+ firstCounterBuffer, counterBufferCount,
+ pCounterBuffers, pCounterBufferOffsets);
+}
+
void radv_CmdDrawIndirectByteCountEXT(
VkCommandBuffer commandBuffer,
uint32_t instanceCount,
cmd_buffer->device->physical_device->rad_info.chip_class,
radv_cmd_buffer_uses_mec(cmd_buffer),
V_028A90_BOTTOM_OF_PIPE_TS, 0,
+ EOP_DST_SEL_MEM,
EOP_DATA_SEL_VALUE_32BIT,
va, marker,
cmd_buffer->gfx9_eop_bug_va);