mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2025-01-31 20:51:52 +01:00
e59cd1a981
This patch implements a correct, but not terribly useful, transform. In particular, if we have a dynamic alloca in a loop which is guaranteed to execute, and provably not captured, we hoist the alloca out of the loop. The capture tracking is needed so that we can prove that each previous stack region dies before the next one is allocated. The transform decreases the amount of stack allocation needed by a linear factor (e.g. the iteration count of the loop). Now, I really hope no one is actually using dynamic allocas. As such, why this patch? Well, the actual problem I'm hoping to make progress on is allocation hoisting. There's a large draft patch out for review (https://reviews.llvm.org/D60056), and this patch was the smallest chunk of testable functionality I could come up with which takes a step vaguely in that direction. Once this is in, it makes motivating the changes to capture tracking mentioned in TODOs testable. After that, I hope to extend this to trivial malloc free regions (i.e. free dominating all loop exits) and allocation functions for GCed languages. Differential Revision: https://reviews.llvm.org/D69227
169 lines
5.3 KiB
LLVM
169 lines
5.3 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
|
|
; RUN: opt -S -licm < %s | FileCheck %s
|
|
|
|
@G = external global i64
|
|
|
|
define void @test(i64 %n) {
|
|
; CHECK-LABEL: @test(
|
|
; CHECK-NEXT: entry:
|
|
; CHECK-NEXT: [[A:%.*]] = alloca i64
|
|
; CHECK-NEXT: [[VAL:%.*]] = load i64, i64* [[A]]
|
|
; CHECK-NEXT: store i64 [[VAL]], i64* @G
|
|
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
|
|
; CHECK: for.body:
|
|
; CHECK-NEXT: [[IV:%.*]] = phi i64 [ [[IV_NEXT:%.*]], [[FOR_BODY]] ], [ 0, [[ENTRY:%.*]] ]
|
|
; CHECK-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
|
|
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp ult i64 [[IV]], [[N:%.*]]
|
|
; CHECK-NEXT: br i1 [[EXITCOND]], label [[FOR_BODY]], label [[EXIT:%.*]]
|
|
; CHECK: exit:
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
entry:
|
|
br label %for.body
|
|
|
|
for.body:
|
|
%iv = phi i64 [ %iv.next, %for.body ], [ 0, %entry ]
|
|
%a = alloca i64
|
|
%val = load i64, i64* %a
|
|
store i64 %val, i64* @G
|
|
%iv.next = add nuw nsw i64 %iv, 1
|
|
%exitcond = icmp ult i64 %iv, %n
|
|
br i1 %exitcond, label %for.body, label %exit
|
|
exit:
|
|
ret void
|
|
}
|
|
|
|
define void @test2(i64 %n) {
|
|
; CHECK-LABEL: @test2(
|
|
; CHECK-NEXT: entry:
|
|
; CHECK-NEXT: [[A:%.*]] = alloca i64
|
|
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
|
|
; CHECK: for.body:
|
|
; CHECK-NEXT: [[IV:%.*]] = phi i64 [ [[IV_NEXT:%.*]], [[FOR_BODY]] ], [ 0, [[ENTRY:%.*]] ]
|
|
; CHECK-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
|
|
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp ult i64 [[IV]], [[N:%.*]]
|
|
; CHECK-NEXT: br i1 [[EXITCOND]], label [[FOR_BODY]], label [[EXIT:%.*]]
|
|
; CHECK: exit:
|
|
; CHECK-NEXT: [[IV_LCSSA:%.*]] = phi i64 [ [[IV]], [[FOR_BODY]] ]
|
|
; CHECK-NEXT: store i64 [[IV_LCSSA]], i64* [[A]], align 4
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
entry:
|
|
br label %for.body
|
|
|
|
for.body:
|
|
%iv = phi i64 [ %iv.next, %for.body ], [ 0, %entry ]
|
|
%a = alloca i64
|
|
store i64 %iv, i64* %a
|
|
%iv.next = add nuw nsw i64 %iv, 1
|
|
%exitcond = icmp ult i64 %iv, %n
|
|
br i1 %exitcond, label %for.body, label %exit
|
|
exit:
|
|
ret void
|
|
}
|
|
|
|
|
|
define void @test3(i64 %n) {
|
|
; CHECK-LABEL: @test3(
|
|
; CHECK-NEXT: entry:
|
|
; CHECK-NEXT: [[A:%.*]] = alloca i64
|
|
; CHECK-NEXT: [[A_I8:%.*]] = bitcast i64* [[A]] to i8*
|
|
; CHECK-NEXT: [[A_OFFSET:%.*]] = getelementptr i8, i8* [[A_I8]], i64 4
|
|
; CHECK-NEXT: store i8 0, i8* [[A_OFFSET]]
|
|
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
|
|
; CHECK: for.body:
|
|
; CHECK-NEXT: [[IV:%.*]] = phi i64 [ [[IV_NEXT:%.*]], [[FOR_BODY]] ], [ 0, [[ENTRY:%.*]] ]
|
|
; CHECK-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
|
|
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp ult i64 [[IV]], [[N:%.*]]
|
|
; CHECK-NEXT: br i1 [[EXITCOND]], label [[FOR_BODY]], label [[EXIT:%.*]]
|
|
; CHECK: exit:
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
entry:
|
|
br label %for.body
|
|
|
|
for.body:
|
|
%iv = phi i64 [ %iv.next, %for.body ], [ 0, %entry ]
|
|
%a = alloca i64
|
|
%a.i8 = bitcast i64* %a to i8*
|
|
%a.offset = getelementptr i8, i8* %a.i8, i64 4
|
|
store i8 0, i8* %a.offset
|
|
%iv.next = add nuw nsw i64 %iv, 1
|
|
%exitcond = icmp ult i64 %iv, %n
|
|
br i1 %exitcond, label %for.body, label %exit
|
|
exit:
|
|
ret void
|
|
}
|
|
|
|
; This example is subtle. Because the dynamic alloca isn't reclaimed until
|
|
; end of function scope, the captured value can legally point to a dynamic
|
|
; alloca stack region from a previous iteration.
|
|
define void @test4(i64 %n) {
|
|
; CHECK-LABEL: @test4(
|
|
; CHECK-NEXT: entry:
|
|
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
|
|
; CHECK: for.body:
|
|
; CHECK-NEXT: [[IV:%.*]] = phi i64 [ [[IV_NEXT:%.*]], [[FOR_BODY]] ], [ 0, [[ENTRY:%.*]] ]
|
|
; CHECK-NEXT: [[A:%.*]] = alloca i64
|
|
; CHECK-NEXT: store i64 [[IV]], i64* [[A]]
|
|
; CHECK-NEXT: call void @capture(i64* [[A]])
|
|
; CHECK-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
|
|
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp ult i64 [[IV]], [[N:%.*]]
|
|
; CHECK-NEXT: br i1 [[EXITCOND]], label [[FOR_BODY]], label [[EXIT:%.*]]
|
|
; CHECK: exit:
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
entry:
|
|
br label %for.body
|
|
|
|
for.body:
|
|
%iv = phi i64 [ %iv.next, %for.body ], [ 0, %entry ]
|
|
%a = alloca i64
|
|
store i64 %iv, i64* %a
|
|
%a.i8 = bitcast i64* %a to i8*
|
|
call void @capture(i64* %a)
|
|
%iv.next = add nuw nsw i64 %iv, 1
|
|
%exitcond = icmp ult i64 %iv, %n
|
|
br i1 %exitcond, label %for.body, label %exit
|
|
exit:
|
|
ret void
|
|
}
|
|
declare void @capture(i64* %a)
|
|
|
|
|
|
; TODO: not yet handled
|
|
define void @test5(i64 %n) {
|
|
; CHECK-LABEL: @test5(
|
|
; CHECK-NEXT: entry:
|
|
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
|
|
; CHECK: for.body:
|
|
; CHECK-NEXT: [[IV:%.*]] = phi i64 [ [[IV_NEXT:%.*]], [[FOR_BODY]] ], [ 0, [[ENTRY:%.*]] ]
|
|
; CHECK-NEXT: [[A:%.*]] = alloca i64
|
|
; CHECK-NEXT: store i64 [[IV]], i64* [[A]]
|
|
; CHECK-NEXT: [[A_I8:%.*]] = bitcast i64* [[A]] to i8*
|
|
; CHECK-NEXT: [[TMP0:%.*]] = call {}* @llvm.invariant.start.p0i8(i64 8, i8* [[A_I8]])
|
|
; CHECK-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
|
|
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp ult i64 [[IV]], [[N:%.*]]
|
|
; CHECK-NEXT: br i1 [[EXITCOND]], label [[FOR_BODY]], label [[EXIT:%.*]]
|
|
; CHECK: exit:
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
entry:
|
|
br label %for.body
|
|
|
|
for.body:
|
|
%iv = phi i64 [ %iv.next, %for.body ], [ 0, %entry ]
|
|
%a = alloca i64
|
|
store i64 %iv, i64* %a
|
|
%a.i8 = bitcast i64* %a to i8*
|
|
call {}* @llvm.invariant.start.p0i8(i64 8, i8* %a.i8)
|
|
%iv.next = add nuw nsw i64 %iv, 1
|
|
%exitcond = icmp ult i64 %iv, %n
|
|
br i1 %exitcond, label %for.body, label %exit
|
|
exit:
|
|
ret void
|
|
}
|
|
|
|
declare {}* @llvm.invariant.start.p0i8(i64, i8* nocapture) nounwind readonly
|
|
|