[Mesa-dev] [PATCH] i965: Fix last slot calculations

[Kenneth Graunke]

If the VUE map has slots at the end which the shader does not write,
then we'd "flush" (constructing an URB write) on the last output it
actually wrote.  Then, we'd construct another SEND with EOT, but with
no actual payload data.  That's not legal.

For example, SSO programs have clip distance slots allocated no matter
what, but the shader may not write them.  If it doesn't write any user
defined varyings, then the clip distance slots will be the last ones.

Found while debugging
dEQP-VK.tessellation.shader_input_output.gl_position_vs_to_tcs_to_tes

Cc: mesa-stable at lists.freedesktop.org
Signed-off-by: Kenneth Graunke <kenneth at whitecape.org>
---
 src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp b/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp
index 14415bd5c7a..08ac7bc276a 100644
--- a/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp
@@ -672,6 +672,14 @@ fs_visitor::emit_urb_writes(const fs_reg &gs_vertex_count)
    length = 0;
    urb_offset = starting_urb_offset;
    flush = false;
+
+   int last_slot = vue_map->num_slots - 1;
+   while (last_slot > 0 &&
+          (vue_map->slot_to_varying[last_slot] == BRW_VARYING_SLOT_PAD ||
+           outputs[vue_map->slot_to_varying[last_slot]].file == BAD_FILE)) {
+      last_slot--;
+   }
+
    for (slot = 0; slot < vue_map->num_slots; slot++) {
       int varying = vue_map->slot_to_varying[slot];
       switch (varying) {
@@ -757,8 +765,7 @@ fs_visitor::emit_urb_writes(const fs_reg &gs_vertex_count)
        * the last slot or if we need to flush (see BAD_FILE varying case
        * above), emit a URB write send now to flush out the data.
        */
-      int last = slot == vue_map->num_slots - 1;
-      if (length == 8 || last)
+      if (length == 8 || slot == last_slot)
          flush = true;
       if (flush) {
          fs_reg *payload_sources =
@@ -777,7 +784,7 @@ fs_visitor::emit_urb_writes(const fs_reg &gs_vertex_count)
                            header_size);
 
          fs_inst *inst = abld.emit(opcode, reg_undef, payload);
-         inst->eot = last && stage != MESA_SHADER_GEOMETRY;
+         inst->eot = slot == last_slot && stage != MESA_SHADER_GEOMETRY;
          inst->mlen = length + header_size;
          inst->offset = urb_offset;
          urb_offset = starting_urb_offset + slot + 1;
-- 
2.11.0