replica
/
hanchenye-llvm-project
2
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[PowerPC] exclude more icmps in LSR which is converted in later hardware loop pass 

Differential Revision: https://reviews.llvm.org/D64795

llvm-svn: 366976
This commit is contained in:
Chen Zheng 2019-07-25 01:22:08 +00:00
parent e9be72a3b3
commit a2d74d3d90
3 changed files with 6 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
llvm/lib/Transforms/Scalar/LoopStrengthReduce.cpp
View File

@ -3265,12 +3265,12 @@ void LSRInstance::CollectFixupsAndInitialFormulae() {
// requirements for both N and i at the same time. Limiting this code to
// equality icmps is not a problem because all interesting loops use
// equality icmps, thanks to IndVarSimplify.
if (ICmpInst *CI = dyn_cast<ICmpInst>(UserInst))
if (ICmpInst *CI = dyn_cast<ICmpInst>(UserInst)) {
// If CI can be saved in some target, like replaced inside hardware loop
// in PowerPC, no need to generate initial formulae for it.
if (SaveCmp && CI == dyn_cast<ICmpInst>(ExitBranch->getCondition()))
continue;
if (CI->isEquality()) {
// If CI can be saved in some target, like replaced inside hardware loop
// in PowerPC, no need to generate initial formulae for it.
if (SaveCmp && CI == dyn_cast<ICmpInst>(ExitBranch->getCondition()))
continue;
// Swap the operands if needed to put the OperandValToReplace on the
// left, for consistency.
Value *NV = CI->getOperand(1);
@ -3298,6 +3298,7 @@ void LSRInstance::CollectFixupsAndInitialFormulae() {
Factors.insert(-(uint64_t)Factors[i]);
Factors.insert(-1);
}
}
// Get or create an LSRUse.
std::pair<size_t, int64_t> P = getUse(S, Kind, AccessTy);

3
llvm/test/CodeGen/PowerPC/ctrloop-le.ll
View File

@ -2,9 +2,6 @@
target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v128:128:128-n32:64"
target triple = "powerpc64-unknown-linux-gnu"
; XFAIL: *
; SE needs improvement
; CHECK: test_pos1_ir_sle
; CHECK: bdnz
; a < b

5
llvm/test/CodeGen/PowerPC/ctrloop-lt.ll
View File

@ -2,9 +2,6 @@
target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v128:128:128-n32:64"
target triple = "powerpc64-unknown-linux-gnu"
; XFAIL: *
; SE needs improvement
; CHECK: test_pos1_ir_slt
; CHECK: bdnz
; a < b
@ -35,7 +32,6 @@ for.end: ; preds = %for.body, %entry
; CHECK: test_pos2_ir_slt
; FIXME: Support this loop!
; CHECK: bdnz
; a < b
define void @test_pos2_ir_slt(i8* nocapture %p, i32 %a, i32 %b) nounwind {
@ -65,7 +61,6 @@ for.end: ; preds = %for.body, %entry
; CHECK: test_pos4_ir_slt
; FIXME: Support this loop!
; CHECK: bdnz
; a < b
define void @test_pos4_ir_slt(i8* nocapture %p, i32 %a, i32 %b) nounwind {