[Mesa-dev] [PATCH 05/11] i965/fs: Add support for "if" statements in 16-wide mode.

[Eric Anholt]

It turns out there's nothing in the hardware preventing this, it was
just the lame backend we used to have.

Improves glbenchmark Egypt framerate 4.4% +/- 0.3% (n=3), and Pro by
2.6% +/- 0.6% (n=3).
---
 src/mesa/drivers/dri/i965/brw_eu_emit.c |    6 +++++-
 src/mesa/drivers/dri/i965/brw_fs.cpp    |    9 ++++-----
 2 files changed, 9 insertions(+), 6 deletions(-)

diff --git a/src/mesa/drivers/dri/i965/brw_eu_emit.c b/src/mesa/drivers/dri/i965/brw_eu_emit.c
index 859068e..afa1982 100644
--- a/src/mesa/drivers/dri/i965/brw_eu_emit.c
+++ b/src/mesa/drivers/dri/i965/brw_eu_emit.c
@@ -913,7 +913,11 @@ gen6_IF(struct brw_compile *p, uint32_t conditional,
    insn = next_insn(p, BRW_OPCODE_IF);
 
    brw_set_dest(p, insn, brw_imm_w(0));
-   insn->header.execution_size = BRW_EXECUTE_8;
+   if (p->compressed) {
+      insn->header.execution_size = BRW_EXECUTE_16;
+   } else {
+      insn->header.execution_size = BRW_EXECUTE_8;
+   }
    insn->bits1.branch_gen6.jump_count = 0;
    brw_set_src0(insn, src0);
    brw_set_src1(insn, src1);
diff --git a/src/mesa/drivers/dri/i965/brw_fs.cpp b/src/mesa/drivers/dri/i965/brw_fs.cpp
index 7db1314..ef51369 100644
--- a/src/mesa/drivers/dri/i965/brw_fs.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs.cpp
@@ -1797,10 +1797,6 @@ fs_visitor::visit(ir_if *ir)
 {
    fs_inst *inst;
 
-   if (c->dispatch_width == 16) {
-      fail("Can't support (non-uniform) control flow on 16-wide\n");
-   }
-
    /* Don't point the annotation at the if statement, because then it plus
     * the then and else blocks get printed.
     */
@@ -3803,8 +3799,11 @@ fs_visitor::generate_code()
 	    assert(intel->gen >= 6);
 	    if_stack[if_stack_depth] = gen6_IF(p, inst->conditional_mod, src[0], src[1]);
 	 } else {
-	    if_stack[if_stack_depth] = brw_IF(p, BRW_EXECUTE_8);
+	    if_stack[if_stack_depth] = brw_IF(p, c->dispatch_width == 16 ?
+					      BRW_EXECUTE_16 :
+					      BRW_EXECUTE_8);
 	 }
+
 	 if_depth_in_loop[loop_stack_depth]++;
 	 if_stack_depth++;
 	 if (if_stack_array_size <= if_stack_depth) {
-- 
1.7.5.1