From: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Date: Sun, 14 Jun 2020 21:24:29 +0000 (-0400)
Subject: zink: add lengthy comment and remove assert from discard_if ntv pass
X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=359c9384837f332edb3140f43675c761341b8bcd;p=mesa.git

zink: add lengthy comment and remove assert from discard_if ntv pass

as in the comment, while we may want to try verifying that discard will be
the last instruction in a block, it's a bit problematic given that other nir
passes we're doing may insert instructions after a discard as part of e.g.,
nir_opt_dead_cf in the process of removing another block

fixes shaders@glsl-fs-discard-04

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5852>
---

diff --git a/src/gallium/drivers/zink/zink_compiler.c b/src/gallium/drivers/zink/zink_compiler.c
index 65d4a2e3390..33e7631130b 100644
--- a/src/gallium/drivers/zink/zink_compiler.c
+++ b/src/gallium/drivers/zink/zink_compiler.c
@@ -52,8 +52,49 @@ lower_discard_if_instr(nir_intrinsic_instr *instr, nir_builder *b)
       nir_instr_remove(&instr->instr);
       return true;
    }
-   assert(instr->intrinsic != nir_intrinsic_discard ||
-          nir_block_last_instr(instr->instr.block) == &instr->instr);
+   /* a shader like this (shaders@glsl-fs-discard-04):
+
+      uniform int j, k;
+
+      void main()
+      {
+       for (int i = 0; i < j; i++) {
+        if (i > k)
+         continue;
+        discard;
+       }
+       gl_FragColor = vec4(0.0, 1.0, 0.0, 0.0);
+      }
+
+
+
+      will generate nir like:
+
+      loop   {
+         //snip
+         if   ssa_11   {
+            block   block_5:
+            /   preds:   block_4   /
+            vec1   32   ssa_17   =   iadd   ssa_50,   ssa_31
+            /   succs:   block_7   /
+         }   else   {
+            block   block_6:
+            /   preds:   block_4   /
+            intrinsic   discard   ()   () <-- not last instruction
+            vec1   32   ssa_23   =   iadd   ssa_50,   ssa_31 <-- dead code loop itr increment
+            /   succs:   block_7   /
+         }
+         //snip
+      }
+
+      which means that we can't assert like this:
+
+      assert(instr->intrinsic != nir_intrinsic_discard ||
+             nir_block_last_instr(instr->instr.block) == &instr->instr);
+
+
+      and it's unnecessary anyway since post-vtn optimizing will dce the instructions following the discard
+    */
 
    return false;
 }