[llvm] r278594 - [X86] Remove patterns for (vzmovl (insert_subvector undef, (scalar_to_vector))) as the (vzmovl VR256) pattern has higher priority. NFC

Craig Topper via llvm-commits llvm-commits at lists.llvm.org
Fri Aug 12 23:02:19 PDT 2016


Author: ctopper
Date: Sat Aug 13 01:02:19 2016
New Revision: 278594

URL: http://llvm.org/viewvc/llvm-project?rev=278594&view=rev
Log:
[X86] Remove patterns for (vzmovl (insert_subvector undef, (scalar_to_vector))) as the (vzmovl VR256) pattern has higher priority. NFC

Modified:
    llvm/trunk/lib/Target/X86/X86InstrSSE.td

Modified: llvm/trunk/lib/Target/X86/X86InstrSSE.td
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/X86/X86InstrSSE.td?rev=278594&r1=278593&r2=278594&view=diff
==============================================================================
--- llvm/trunk/lib/Target/X86/X86InstrSSE.td (original)
+++ llvm/trunk/lib/Target/X86/X86InstrSSE.td Sat Aug 13 01:02:19 2016
@@ -7170,17 +7170,6 @@ let Predicates = [UseAVX] in {
             (VBLENDPDYrri (v4f64 (AVX_SET0)), VR256:$src, (i8 1))>;
   }
 
-  def : Pat<(v8f32 (X86vzmovl (insert_subvector undef,
-                   (v4f32 (scalar_to_vector FR32:$src)), (iPTR 0)))),
-            (SUBREG_TO_REG (i32 0),
-                           (v4f32 (VMOVSSrr (v4f32 (V_SET0)), FR32:$src)),
-                           sub_xmm)>;
-  def : Pat<(v4f64 (X86vzmovl (insert_subvector undef,
-                   (v2f64 (scalar_to_vector FR64:$src)), (iPTR 0)))),
-            (SUBREG_TO_REG (i64 0),
-                           (v2f64 (VMOVSDrr (v2f64 (V_SET0)), FR64:$src)),
-                           sub_xmm)>;
-
   // These will incur an FP/int domain crossing penalty, but it may be the only
   // way without AVX2. Do not add any complexity because we may be able to match
   // more optimal patterns defined earlier in this file.




More information about the llvm-commits mailing list