From ad31d10fabf065f3a64afdff8bc9203167df984b Mon Sep 17 00:00:00 2001 From: Tom Stellard Date: Tue, 15 Dec 2015 17:11:17 +0000 Subject: [PATCH] AMDGPU/SI: Fix bitcast between v2f32 and f64 MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit The radeonsi fp64 support can hit these now that some redundant bitcasts are folded. Patch by: Michel Dänzer Signed-off-by: Michel Dänzer git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@255657 91177308-0d34-0410-b5e6-96231b3b80d8 --- lib/Target/AMDGPU/SIInstructions.td | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/lib/Target/AMDGPU/SIInstructions.td b/lib/Target/AMDGPU/SIInstructions.td index 08149d4eab2..8163419cd48 100644 --- a/lib/Target/AMDGPU/SIInstructions.td +++ b/lib/Target/AMDGPU/SIInstructions.td @@ -2561,7 +2561,9 @@ def : BitConvert ; def : BitConvert ; def : BitConvert ; def : BitConvert ; +def : BitConvert ; def : BitConvert ; +def : BitConvert ; def : BitConvert ; def : BitConvert ; def : BitConvert ; @@ -2570,7 +2572,9 @@ def : BitConvert ; def : BitConvert ; def : BitConvert ; +def : BitConvert ; def : BitConvert ; +def : BitConvert ; def : BitConvert ; -- 2.34.1