From: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Date: Wed, 14 Sep 2016 20:15:14 +0000 (+0200)
Subject: nv50/ir: optimize SHLADD(a, b, c) to MOV((a << b) + c)
X-Git-Tag: upstream/17.1.0~6105
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=115c79be10bf3712a1e1bc25a563c90388c1bcaa;p=platform%2Fupstream%2Fmesa.git

nv50/ir: optimize SHLADD(a, b, c) to MOV((a << b) + c)

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
---

diff --git a/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp b/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp
index cbbe34d..9875738 100644
--- a/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp
+++ b/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp
@@ -778,6 +778,9 @@ ConstantFolding::expr(Instruction *i,
       }
       break;
    }
+   case OP_SHLADD:
+      res.data.u32 = (a->data.u32 << b->data.u32) + c->data.u32;
+      break;
    default:
       return;
    }