[PATCH] D40254: [MI scheduler] Fix VADD in cortex-a57 model
Eugene Leviant via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Mon Nov 20 07:42:23 PST 2017
evgeny777 created this revision.
The "VADD" regex incorrectly captures VADDfq and other vector floating-point adds, which have 5 cycles latency instead of 3.
Repository:
rL LLVM
https://reviews.llvm.org/D40254
Files:
lib/Target/ARM/ARMScheduleA57.td
test/CodeGen/ARM/cortex-a57-misched-vadd.ll
Index: test/CodeGen/ARM/cortex-a57-misched-vadd.ll
===================================================================
--- test/CodeGen/ARM/cortex-a57-misched-vadd.ll
+++ test/CodeGen/ARM/cortex-a57-misched-vadd.ll
@@ -0,0 +1,24 @@
+; REQUIRES: asserts
+; RUN: llc < %s -mtriple=armv8r-eabi -mcpu=cortex-a57 -misched-postra -enable-misched -verify-misched -debug-only=machine-scheduler -o - 2>&1 > /dev/null | FileCheck %s
+
+; CHECK: SU(8): {{.*}} VADDv4i32
+; CHECK-NEXT: # preds left
+; CHECK-NEXT: # succs left
+; CHECK-NEXT: # rdefs left
+; CHECK-NEXT: Latency : 3
+
+; CHECK: SU(8): {{.*}} VADDfq
+; CHECK-NEXT: # preds left
+; CHECK-NEXT: # succs left
+; CHECK-NEXT: # rdefs left
+; CHECK-NEXT: Latency : 5
+
+define <4 x i32> @addv_i32(<4 x i32>, <4 x i32>) {
+ %3 = add <4 x i32> %1, %0
+ ret <4 x i32> %3
+}
+
+define <4 x float> @addv_f32(<4 x float>, <4 x float>) {
+ %3 = fadd <4 x float> %0, %1
+ ret <4 x float> %3
+}
Index: lib/Target/ARM/ARMScheduleA57.td
===================================================================
--- lib/Target/ARM/ARMScheduleA57.td
+++ lib/Target/ARM/ARMScheduleA57.td
@@ -971,7 +971,7 @@
def : InstRW<[A57Write_3cyc_1V], (instregex "VABDL(s|u)")>;
// ASIMD arith, basic
-def : InstRW<[A57Write_3cyc_1V], (instregex "VADD", "VADDL", "VADDW",
+def : InstRW<[A57Write_3cyc_1V], (instregex "VADDv", "VADDL", "VADDW",
"VNEG(s8d|s16d|s32d|s8q|s16q|s32q|d|q)",
"VPADDi", "VPADDL", "VSUB", "VSUBL", "VSUBW")>;
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D40254.123593.patch
Type: text/x-patch
Size: 1495 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20171120/14a4c36c/attachment.bin>
More information about the llvm-commits
mailing list