intel/fs: Properly handle 64-bit types in LOAD_PAYLOAD

By just assigning dst.type to src[i].type, we ensure that the offset at
the end of the loop actually offsets it by the right number of
registers.  Otherwise, we'll get into a case where we copy with a Q type
and then offset with a D type and things get out of sync.

Reviewed-by: Kenneth Graunke <[email protected]>

2 files changed, 7 insertions, 2 deletions

diff --git a/src/intel/compiler/brw_fs.cpp b/src/intel/compiler/brw_fs.cpp
index 8dd3b94fbd5..303b1c1b272 100644
--- a/src/intel/compiler/brw_fs.cpp
+++ b/src/intel/compiler/brw_fs.cpp
@@ -3799,8 +3799,12 @@ fs_visitor::lower_load_payload()
       }
 
       for (uint8_t i = inst->header_size; i < inst->sources; i++) {
-         if (inst->src[i].file != BAD_FILE)
-            ibld.MOV(retype(dst, inst->src[i].type), inst->src[i]);
+         if (inst->src[i].file != BAD_FILE) {
+            dst.type = inst->src[i].type;
+            ibld.MOV(dst, inst->src[i]);
+         } else {
+            dst.type = BRW_REGISTER_TYPE_UD;
+         }
          dst = offset(dst, ibld, 1);
       }
 
diff --git a/src/intel/compiler/brw_fs_cse.cpp b/src/intel/compiler/brw_fs_cse.cpp
index bd917baaa6d..6efa111b1a4 100644
--- a/src/intel/compiler/brw_fs_cse.cpp
+++ b/src/intel/compiler/brw_fs_cse.cpp
@@ -216,6 +216,7 @@ create_copy_instr(const fs_builder &bld, fs_inst *inst, fs_reg src, bool negate)
          src.offset += REG_SIZE;
       }
       for (int i = inst->header_size; i < inst->sources; i++) {
+         src.type = inst->src[i].type;
          payload[i] = src;
          src = offset(src, bld, 1);
       }