[Mesa-dev] [PATCH v3 5/6] i965: Upload binding tables in hw-generated binding table format.

[Abdiel Janulgue]

When hardware-generated binding tables are enabled, use the hw-generated
binding table format when uploading binding table state.

Normally, the CS will will just consume the binding table pointer commands
as pipelined state. When the RS is enabled however, the RS flushes whatever
edited surface state entries of our on-chip binding table to the binding
table pool before passing the command on to the CS.

Note that the the binding table pointer offset is relative to the binding table
pool base address when resource streamer instead of the surface state base address.

v2: Fix possible buffer overflow when allocating a chunk out of the
    hw-binding table pool (Ken).
v3: Remove extra newline and add missing brace around if-statement (Matt).

Cc: kenneth at whitecape.org
Cc: mattst88 at gmail.com
Signed-off-by: Abdiel Janulgue <abdiel.janulgue at linux.intel.com>
---
 src/mesa/drivers/dri/i965/brw_binding_tables.c | 72 ++++++++++++++++++++------
 1 file changed, 56 insertions(+), 16 deletions(-)

diff --git a/src/mesa/drivers/dri/i965/brw_binding_tables.c b/src/mesa/drivers/dri/i965/brw_binding_tables.c
index b3d592b..cc56dbf 100644
--- a/src/mesa/drivers/dri/i965/brw_binding_tables.c
+++ b/src/mesa/drivers/dri/i965/brw_binding_tables.c
@@ -50,6 +50,26 @@ static const GLuint stage_to_bt_edit[MESA_SHADER_FRAGMENT + 1] = {
    _3DSTATE_BINDING_TABLE_EDIT_PS,
 };
 
+static uint32_t
+reserve_hw_bt_space(struct brw_context *brw, unsigned bytes)
+{
+   if (brw->hw_bt_pool.next_offset + bytes >= brw->hw_bt_pool.bo->size - 128) {
+      gen7_reset_hw_bt_pool_offsets(brw);
+   }
+
+   uint32_t offset = brw->hw_bt_pool.next_offset;
+
+   /* From the Haswell PRM, Volume 2b: Command Reference: Instructions,
+    * 3DSTATE_BINDING_TABLE_POINTERS_xS:
+    *
+    * "If HW Binding Table is enabled, the offset is relative to the
+    *  Binding Table Pool Base Address and the alignment is 64 bytes."
+    */
+   brw->hw_bt_pool.next_offset += ALIGN(bytes, 64);
+
+   return offset;
+}
+
 /**
  * Upload a shader stage's binding table as indirect state.
  *
@@ -70,30 +90,50 @@ brw_upload_binding_table(struct brw_context *brw,
 
       stage_state->bind_bo_offset = 0;
    } else {
-      /* Upload a new binding table. */
-      if (INTEL_DEBUG & DEBUG_SHADER_TIME) {
-         brw->vtbl.emit_buffer_surface_state(
-            brw, &stage_state->surf_offset[
-                    prog_data->binding_table.shader_time_start],
-            brw->shader_time.bo, 0, BRW_SURFACEFORMAT_RAW,
-            brw->shader_time.bo->size, 1, true);
+      /* When RS is enabled use hw-binding table uploads, otherwise fallback to
+       * software-uploads.
+       */
+      if (brw->use_resource_streamer) {
+         gen7_update_binding_table_from_array(brw, stage_state->stage,
+                                              stage_state->surf_offset,
+                                              prog_data->binding_table
+                                              .size_bytes / 4);
+      } else {
+         /* Upload a new binding table. */
+         if (INTEL_DEBUG & DEBUG_SHADER_TIME) {
+            brw->vtbl.emit_buffer_surface_state(
+               brw, &stage_state->surf_offset[
+                  prog_data->binding_table.shader_time_start],
+               brw->shader_time.bo, 0, BRW_SURFACEFORMAT_RAW,
+               brw->shader_time.bo->size, 1, true);
+         }
+
+         uint32_t *bind = brw_state_batch(brw, AUB_TRACE_BINDING_TABLE,
+                                          prog_data->binding_table.size_bytes,
+                                          32,
+                                          &stage_state->bind_bo_offset);
+
+         /* BRW_NEW_SURFACES and BRW_NEW_*_CONSTBUF */
+         memcpy(bind, stage_state->surf_offset,
+                prog_data->binding_table.size_bytes);
       }
-
-      uint32_t *bind = brw_state_batch(brw, AUB_TRACE_BINDING_TABLE,
-                                       prog_data->binding_table.size_bytes, 32,
-                                       &stage_state->bind_bo_offset);
-
-      /* BRW_NEW_SURFACES and BRW_NEW_*_CONSTBUF */
-      memcpy(bind, stage_state->surf_offset,
-             prog_data->binding_table.size_bytes);
    }
 
    brw->ctx.NewDriverState |= brw_new_binding_table;
 
    if (brw->gen >= 7) {
+      if (brw->use_resource_streamer) {
+         stage_state->bind_bo_offset =
+            reserve_hw_bt_space(brw, prog_data->binding_table.size_bytes);
+      }
       BEGIN_BATCH(2);
       OUT_BATCH(packet_name << 16 | (2 - 2));
-      OUT_BATCH(stage_state->bind_bo_offset);
+      /* Align SurfaceStateOffset[16:6] format to [15:5] PS Binding Table field
+       * when hw-generated binding table is enabled.
+       */
+      OUT_BATCH(brw->use_resource_streamer ?
+                (stage_state->bind_bo_offset >> 1) :
+                stage_state->bind_bo_offset);
       ADVANCE_BATCH();
    }
 }
-- 
1.9.1