[llvm-branch-commits] [llvm] release/18.x: [GlobalISel] Don't form anyextending atomic loads. (PR #90435)
https://github.com/llvmbot created
https://github.com/llvm/llvm-project/pull/90435
Backport 60fc4ac67a613e4e36cef019fb2d13d70a06cfe8
Requested by: @nikic
>From 00370f3fc3238b7516512c00956221ab9225a2bb Mon Sep 17 00:00:00 2001
From: Amara Emerson
Date: Fri, 5 Apr 2024 10:49:19 -0700
Subject: [PATCH] [GlobalISel] Don't form anyextending atomic loads.
Until we can reliably check the legality and improve our selection of these,
don't form them at all.
(cherry picked from commit 60fc4ac67a613e4e36cef019fb2d13d70a06cfe8)
---
.../lib/CodeGen/GlobalISel/CombinerHelper.cpp | 4 +-
.../Atomics/aarch64-atomic-load-rcpc_immo.ll | 55 +-
.../AArch64/GlobalISel/arm64-atomic.ll| 56 +--
.../AArch64/GlobalISel/arm64-pcsections.ll| 28 +-
.../atomic-anyextending-load-crash.ll | 47
5 files changed, 131 insertions(+), 59 deletions(-)
create mode 100644
llvm/test/CodeGen/AArch64/GlobalISel/atomic-anyextending-load-crash.ll
diff --git a/llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp
b/llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp
index 772229215e798d..61ddc858ba44c7 100644
--- a/llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp
+++ b/llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp
@@ -591,8 +591,8 @@ bool
CombinerHelper::matchCombineExtendingLoads(MachineInstr &MI,
UseMI.getOpcode() == TargetOpcode::G_ZEXT ||
(UseMI.getOpcode() == TargetOpcode::G_ANYEXT)) {
const auto &MMO = LoadMI->getMMO();
- // For atomics, only form anyextending loads.
- if (MMO.isAtomic() && UseMI.getOpcode() != TargetOpcode::G_ANYEXT)
+ // Don't do anything for atomics.
+ if (MMO.isAtomic())
continue;
// Check for legality.
if (!isPreLegalize()) {
diff --git a/llvm/test/CodeGen/AArch64/Atomics/aarch64-atomic-load-rcpc_immo.ll
b/llvm/test/CodeGen/AArch64/Atomics/aarch64-atomic-load-rcpc_immo.ll
index b0507e9d075fab..9687ba683fb7e6 100644
--- a/llvm/test/CodeGen/AArch64/Atomics/aarch64-atomic-load-rcpc_immo.ll
+++ b/llvm/test/CodeGen/AArch64/Atomics/aarch64-atomic-load-rcpc_immo.ll
@@ -35,16 +35,24 @@ define i8 @load_atomic_i8_aligned_monotonic_const(ptr
readonly %ptr) {
}
define i8 @load_atomic_i8_aligned_acquire(ptr %ptr) {
-; CHECK-LABEL: load_atomic_i8_aligned_acquire:
-; CHECK:ldapurb w0, [x0, #4]
+; GISEL-LABEL: load_atomic_i8_aligned_acquire:
+; GISEL:add x8, x0, #4
+; GISEL:ldaprb w0, [x8]
+;
+; SDAG-LABEL: load_atomic_i8_aligned_acquire:
+; SDAG:ldapurb w0, [x0, #4]
%gep = getelementptr inbounds i8, ptr %ptr, i32 4
%r = load atomic i8, ptr %gep acquire, align 1
ret i8 %r
}
define i8 @load_atomic_i8_aligned_acquire_const(ptr readonly %ptr) {
-; CHECK-LABEL: load_atomic_i8_aligned_acquire_const:
-; CHECK:ldapurb w0, [x0, #4]
+; GISEL-LABEL: load_atomic_i8_aligned_acquire_const:
+; GISEL:add x8, x0, #4
+; GISEL:ldaprb w0, [x8]
+;
+; SDAG-LABEL: load_atomic_i8_aligned_acquire_const:
+; SDAG:ldapurb w0, [x0, #4]
%gep = getelementptr inbounds i8, ptr %ptr, i32 4
%r = load atomic i8, ptr %gep acquire, align 1
ret i8 %r
@@ -101,16 +109,24 @@ define i16 @load_atomic_i16_aligned_monotonic_const(ptr
readonly %ptr) {
}
define i16 @load_atomic_i16_aligned_acquire(ptr %ptr) {
-; CHECK-LABEL: load_atomic_i16_aligned_acquire:
-; CHECK:ldapurh w0, [x0, #8]
+; GISEL-LABEL: load_atomic_i16_aligned_acquire:
+; GISEL:add x8, x0, #8
+; GISEL:ldaprh w0, [x8]
+;
+; SDAG-LABEL: load_atomic_i16_aligned_acquire:
+; SDAG:ldapurh w0, [x0, #8]
%gep = getelementptr inbounds i16, ptr %ptr, i32 4
%r = load atomic i16, ptr %gep acquire, align 2
ret i16 %r
}
define i16 @load_atomic_i16_aligned_acquire_const(ptr readonly %ptr) {
-; CHECK-LABEL: load_atomic_i16_aligned_acquire_const:
-; CHECK:ldapurh w0, [x0, #8]
+; GISEL-LABEL: load_atomic_i16_aligned_acquire_const:
+; GISEL:add x8, x0, #8
+; GISEL:ldaprh w0, [x8]
+;
+; SDAG-LABEL: load_atomic_i16_aligned_acquire_const:
+; SDAG:ldapurh w0, [x0, #8]
%gep = getelementptr inbounds i16, ptr %ptr, i32 4
%r = load atomic i16, ptr %gep acquire, align 2
ret i16 %r
@@ -367,16 +383,24 @@ define i8 @load_atomic_i8_unaligned_monotonic_const(ptr
readonly %ptr) {
}
define i8 @load_atomic_i8_unaligned_acquire(ptr %ptr) {
-; CHECK-LABEL: load_atomic_i8_unaligned_acquire:
-; CHECK:ldapurb w0, [x0, #4]
+; GISEL-LABEL: load_atomic_i8_unaligned_acquire:
+; GISEL:add x8, x0, #4
+; GISEL:ldaprb w0, [x8]
+;
+; SDAG-LABEL: load_atomic_i8_unaligned_acquire:
+; SDAG:ldapurb w0, [x0, #4]
%gep = getelementptr inbounds i8, ptr %ptr, i32 4
%r = load atomic i8, ptr %gep acquire, align 1
ret i8 %r
}
define i8 @load_atomic_i8_unaligned_acquire_const(ptr readonly %ptr) {
-; CHECK-LABEL: load_atomic_i8_unaligned_acquire_const:
-; CHECK:ldapurb w0, [x0, #4]
+; GISEL-LABEL: load_atomic_i8_unaligned_acquire_const:
+; GISEL:add
[llvm-branch-commits] [llvm] release/18.x: [GlobalISel] Don't form anyextending atomic loads. (PR #90435)
https://github.com/llvmbot milestoned https://github.com/llvm/llvm-project/pull/90435 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] release/18.x: [GlobalISel] Don't form anyextending atomic loads. (PR #90435)
llvmbot wrote:
@llvm/pr-subscribers-backend-aarch64
Author: None (llvmbot)
Changes
Backport 60fc4ac67a613e4e36cef019fb2d13d70a06cfe8
Requested by: @nikic
---
Full diff: https://github.com/llvm/llvm-project/pull/90435.diff
5 Files Affected:
- (modified) llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp (+2-2)
- (modified) llvm/test/CodeGen/AArch64/Atomics/aarch64-atomic-load-rcpc_immo.ll
(+40-15)
- (modified) llvm/test/CodeGen/AArch64/GlobalISel/arm64-atomic.ll (+28-28)
- (modified) llvm/test/CodeGen/AArch64/GlobalISel/arm64-pcsections.ll (+14-14)
- (added)
llvm/test/CodeGen/AArch64/GlobalISel/atomic-anyextending-load-crash.ll (+47)
``diff
diff --git a/llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp
b/llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp
index 772229215e798d..61ddc858ba44c7 100644
--- a/llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp
+++ b/llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp
@@ -591,8 +591,8 @@ bool
CombinerHelper::matchCombineExtendingLoads(MachineInstr &MI,
UseMI.getOpcode() == TargetOpcode::G_ZEXT ||
(UseMI.getOpcode() == TargetOpcode::G_ANYEXT)) {
const auto &MMO = LoadMI->getMMO();
- // For atomics, only form anyextending loads.
- if (MMO.isAtomic() && UseMI.getOpcode() != TargetOpcode::G_ANYEXT)
+ // Don't do anything for atomics.
+ if (MMO.isAtomic())
continue;
// Check for legality.
if (!isPreLegalize()) {
diff --git a/llvm/test/CodeGen/AArch64/Atomics/aarch64-atomic-load-rcpc_immo.ll
b/llvm/test/CodeGen/AArch64/Atomics/aarch64-atomic-load-rcpc_immo.ll
index b0507e9d075fab..9687ba683fb7e6 100644
--- a/llvm/test/CodeGen/AArch64/Atomics/aarch64-atomic-load-rcpc_immo.ll
+++ b/llvm/test/CodeGen/AArch64/Atomics/aarch64-atomic-load-rcpc_immo.ll
@@ -35,16 +35,24 @@ define i8 @load_atomic_i8_aligned_monotonic_const(ptr
readonly %ptr) {
}
define i8 @load_atomic_i8_aligned_acquire(ptr %ptr) {
-; CHECK-LABEL: load_atomic_i8_aligned_acquire:
-; CHECK:ldapurb w0, [x0, #4]
+; GISEL-LABEL: load_atomic_i8_aligned_acquire:
+; GISEL:add x8, x0, #4
+; GISEL:ldaprb w0, [x8]
+;
+; SDAG-LABEL: load_atomic_i8_aligned_acquire:
+; SDAG:ldapurb w0, [x0, #4]
%gep = getelementptr inbounds i8, ptr %ptr, i32 4
%r = load atomic i8, ptr %gep acquire, align 1
ret i8 %r
}
define i8 @load_atomic_i8_aligned_acquire_const(ptr readonly %ptr) {
-; CHECK-LABEL: load_atomic_i8_aligned_acquire_const:
-; CHECK:ldapurb w0, [x0, #4]
+; GISEL-LABEL: load_atomic_i8_aligned_acquire_const:
+; GISEL:add x8, x0, #4
+; GISEL:ldaprb w0, [x8]
+;
+; SDAG-LABEL: load_atomic_i8_aligned_acquire_const:
+; SDAG:ldapurb w0, [x0, #4]
%gep = getelementptr inbounds i8, ptr %ptr, i32 4
%r = load atomic i8, ptr %gep acquire, align 1
ret i8 %r
@@ -101,16 +109,24 @@ define i16 @load_atomic_i16_aligned_monotonic_const(ptr
readonly %ptr) {
}
define i16 @load_atomic_i16_aligned_acquire(ptr %ptr) {
-; CHECK-LABEL: load_atomic_i16_aligned_acquire:
-; CHECK:ldapurh w0, [x0, #8]
+; GISEL-LABEL: load_atomic_i16_aligned_acquire:
+; GISEL:add x8, x0, #8
+; GISEL:ldaprh w0, [x8]
+;
+; SDAG-LABEL: load_atomic_i16_aligned_acquire:
+; SDAG:ldapurh w0, [x0, #8]
%gep = getelementptr inbounds i16, ptr %ptr, i32 4
%r = load atomic i16, ptr %gep acquire, align 2
ret i16 %r
}
define i16 @load_atomic_i16_aligned_acquire_const(ptr readonly %ptr) {
-; CHECK-LABEL: load_atomic_i16_aligned_acquire_const:
-; CHECK:ldapurh w0, [x0, #8]
+; GISEL-LABEL: load_atomic_i16_aligned_acquire_const:
+; GISEL:add x8, x0, #8
+; GISEL:ldaprh w0, [x8]
+;
+; SDAG-LABEL: load_atomic_i16_aligned_acquire_const:
+; SDAG:ldapurh w0, [x0, #8]
%gep = getelementptr inbounds i16, ptr %ptr, i32 4
%r = load atomic i16, ptr %gep acquire, align 2
ret i16 %r
@@ -367,16 +383,24 @@ define i8 @load_atomic_i8_unaligned_monotonic_const(ptr
readonly %ptr) {
}
define i8 @load_atomic_i8_unaligned_acquire(ptr %ptr) {
-; CHECK-LABEL: load_atomic_i8_unaligned_acquire:
-; CHECK:ldapurb w0, [x0, #4]
+; GISEL-LABEL: load_atomic_i8_unaligned_acquire:
+; GISEL:add x8, x0, #4
+; GISEL:ldaprb w0, [x8]
+;
+; SDAG-LABEL: load_atomic_i8_unaligned_acquire:
+; SDAG:ldapurb w0, [x0, #4]
%gep = getelementptr inbounds i8, ptr %ptr, i32 4
%r = load atomic i8, ptr %gep acquire, align 1
ret i8 %r
}
define i8 @load_atomic_i8_unaligned_acquire_const(ptr readonly %ptr) {
-; CHECK-LABEL: load_atomic_i8_unaligned_acquire_const:
-; CHECK:ldapurb w0, [x0, #4]
+; GISEL-LABEL: load_atomic_i8_unaligned_acquire_const:
+; GISEL:add x8, x0, #4
+; GISEL:ldaprb w0, [x8]
+;
+; SDAG-LABEL: load_atomic_i8_unaligned_acquire_const:
+; SDAG:ldapurb w0, [x0, #4]
%gep = getelementptr inbounds i8, ptr %ptr, i32 4
%r = load atomic i8, ptr %gep acquire, align 1
ret i8 %r
@@ -819,7 +843,8 @@ define i128 @load_atomic_i128_unaligned_seq_cst_const(ptr
r
[llvm-branch-commits] [llvm] release/18.x: [CGP] Drop poison-generating flags after hoisting (#90382) (PR #90437)
https://github.com/llvmbot milestoned https://github.com/llvm/llvm-project/pull/90437 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] release/18.x: [CGP] Drop poison-generating flags after hoisting (#90382) (PR #90437)
https://github.com/llvmbot created
https://github.com/llvm/llvm-project/pull/90437
Backport ab12bba0aad800c1805eca2ea937da958c1854c8
Requested by: @dtcxzyw
>From 237863e781eb5d8f1562faea28db90fbf050e43f Mon Sep 17 00:00:00 2001
From: Yingwei Zheng
Date: Mon, 29 Apr 2024 15:51:49 +0800
Subject: [PATCH] [CGP] Drop poison-generating flags after hoisting (#90382)
See the following case:
```
define i8 @src1(i8 %x) {
entry:
%cmp = icmp eq i8 %x, -1
br i1 %cmp, label %exit, label %if.then
if.then:
%inc = add nuw nsw i8 %x, 1
br label %exit
exit:
%retval = phi i8 [ %inc, %if.then ], [ -1, %entry ]
ret i8 %retval
}
define i8 @tgt1(i8 %x) {
entry:
%inc = add nuw nsw i8 %x, 1
%0 = icmp eq i8 %inc, 0
br i1 %0, label %exit, label %if.then
if.then: ; preds = %entry
br label %exit
exit: ; preds = %if.then, %entry
%retval = phi i8 [ %inc, %if.then ], [ -1, %entry ]
ret i8 %retval
}
```
`optimizeBranch` converts `icmp eq X, -1` into cmp to zero on RISC-V and
hoists the add into the entry block. Poison-generating flags should be
dropped as they don't still hold.
Proof: https://alive2.llvm.org/ce/z/sP7mvK
Fixes https://github.com/llvm/llvm-project/issues/90380
(cherry picked from commit ab12bba0aad800c1805eca2ea937da958c1854c8)
---
llvm/lib/CodeGen/CodeGenPrepare.cpp | 2 +
.../CodeGenPrepare/ARM/branch-on-zero.ll | 23 ++
.../CodeGenPrepare/RISCV/convert-to-eqz.ll| 80 +++
3 files changed, 105 insertions(+)
create mode 100644 llvm/test/Transforms/CodeGenPrepare/RISCV/convert-to-eqz.ll
diff --git a/llvm/lib/CodeGen/CodeGenPrepare.cpp
b/llvm/lib/CodeGen/CodeGenPrepare.cpp
index 8ee1f19e083e4e..1cca56fc19cfd8 100644
--- a/llvm/lib/CodeGen/CodeGenPrepare.cpp
+++ b/llvm/lib/CodeGen/CodeGenPrepare.cpp
@@ -8154,6 +8154,7 @@ static bool optimizeBranch(BranchInst *Branch, const
TargetLowering &TLI,
IRBuilder<> Builder(Branch);
if (UI->getParent() != Branch->getParent())
UI->moveBefore(Branch);
+ UI->dropPoisonGeneratingFlags();
Value *NewCmp = Builder.CreateCmp(ICmpInst::ICMP_EQ, UI,
ConstantInt::get(UI->getType(), 0));
LLVM_DEBUG(dbgs() << "Converting " << *Cmp << "\n");
@@ -8167,6 +8168,7 @@ static bool optimizeBranch(BranchInst *Branch, const
TargetLowering &TLI,
IRBuilder<> Builder(Branch);
if (UI->getParent() != Branch->getParent())
UI->moveBefore(Branch);
+ UI->dropPoisonGeneratingFlags();
Value *NewCmp = Builder.CreateCmp(Cmp->getPredicate(), UI,
ConstantInt::get(UI->getType(), 0));
LLVM_DEBUG(dbgs() << "Converting " << *Cmp << "\n");
diff --git a/llvm/test/Transforms/CodeGenPrepare/ARM/branch-on-zero.ll
b/llvm/test/Transforms/CodeGenPrepare/ARM/branch-on-zero.ll
index ff5cef7e781fe6..25dfb3c53a077b 100644
--- a/llvm/test/Transforms/CodeGenPrepare/ARM/branch-on-zero.ll
+++ b/llvm/test/Transforms/CodeGenPrepare/ARM/branch-on-zero.ll
@@ -211,6 +211,29 @@ else:
ret i32 %l
}
+define i32 @sub10_else_drop_nuw(i32 %a) {
+; CHECK-LABEL: @sub10_else_drop_nuw(
+; CHECK-NEXT: entry:
+; CHECK-NEXT:[[L:%.*]] = sub i32 [[A:%.*]], 10
+; CHECK-NEXT:[[TMP0:%.*]] = icmp eq i32 [[L]], 0
+; CHECK-NEXT:br i1 [[TMP0]], label [[THEN:%.*]], label [[ELSE:%.*]]
+; CHECK: then:
+; CHECK-NEXT:ret i32 0
+; CHECK: else:
+; CHECK-NEXT:ret i32 [[L]]
+;
+entry:
+ %c = icmp eq i32 %a, 10
+ br i1 %c, label %then, label %else
+
+then:
+ ret i32 0
+
+else:
+ %l = sub nuw i32 %a, 10
+ ret i32 %l
+}
+
define i32 @subm10_then(i32 %a) {
; CHECK-LABEL: @subm10_then(
; CHECK-NEXT: entry:
diff --git a/llvm/test/Transforms/CodeGenPrepare/RISCV/convert-to-eqz.ll
b/llvm/test/Transforms/CodeGenPrepare/RISCV/convert-to-eqz.ll
new file mode 100644
index 00..a6909d14913494
--- /dev/null
+++ b/llvm/test/Transforms/CodeGenPrepare/RISCV/convert-to-eqz.ll
@@ -0,0 +1,80 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
UTC_ARGS: --version 4
+; RUN: opt -codegenprepare -S -mtriple=riscv64 < %s | FileCheck %s
+
+define i8 @hoist_add(i8 %x) {
+; CHECK-LABEL: define i8 @hoist_add(
+; CHECK-SAME: i8 [[X:%.*]]) {
+; CHECK-NEXT: entry:
+; CHECK-NEXT:[[INC:%.*]] = add i8 [[X]], 1
+; CHECK-NEXT:[[TMP0:%.*]] = icmp eq i8 [[INC]], 0
+; CHECK-NEXT:br i1 [[TMP0]], label [[EXIT:%.*]], label [[IF_THEN:%.*]]
+; CHECK: if.then:
+; CHECK-NEXT:br label [[EXIT]]
+; CHECK: exit:
+; CHECK-NEXT:[[RETVAL:%.*]] = phi i8 [ [[INC]], [[IF_THEN]] ], [ -1,
[[ENTRY:%.*]] ]
+; CHECK-NEXT:ret i8 [[RETVAL]]
+;
+entry:
+ %cmp = icmp eq i8 %x, -1
+ br i1 %cmp, label %exit, label %if.then
+
+if.then:
+ %inc = add nuw nsw i8 %x, 1
+ br label %exit
+
+exit:
+ %retval = phi i8 [ %inc, %if.then ], [ -1, %entry ]
+ ret i8 %retval
+}
+
+define i8 @hoist_lshr(i8
[llvm-branch-commits] [llvm] release/18.x: [CGP] Drop poison-generating flags after hoisting (#90382) (PR #90437)
llvmbot wrote: @nikic What do you think about merging this PR to the release branch? https://github.com/llvm/llvm-project/pull/90437 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] release/18.x: [CGP] Drop poison-generating flags after hoisting (#90382) (PR #90437)
llvmbot wrote:
@llvm/pr-subscribers-llvm-transforms
Author: None (llvmbot)
Changes
Backport ab12bba0aad800c1805eca2ea937da958c1854c8
Requested by: @dtcxzyw
---
Full diff: https://github.com/llvm/llvm-project/pull/90437.diff
3 Files Affected:
- (modified) llvm/lib/CodeGen/CodeGenPrepare.cpp (+2)
- (modified) llvm/test/Transforms/CodeGenPrepare/ARM/branch-on-zero.ll (+23)
- (added) llvm/test/Transforms/CodeGenPrepare/RISCV/convert-to-eqz.ll (+80)
``diff
diff --git a/llvm/lib/CodeGen/CodeGenPrepare.cpp
b/llvm/lib/CodeGen/CodeGenPrepare.cpp
index 8ee1f19e083e4e..1cca56fc19cfd8 100644
--- a/llvm/lib/CodeGen/CodeGenPrepare.cpp
+++ b/llvm/lib/CodeGen/CodeGenPrepare.cpp
@@ -8154,6 +8154,7 @@ static bool optimizeBranch(BranchInst *Branch, const
TargetLowering &TLI,
IRBuilder<> Builder(Branch);
if (UI->getParent() != Branch->getParent())
UI->moveBefore(Branch);
+ UI->dropPoisonGeneratingFlags();
Value *NewCmp = Builder.CreateCmp(ICmpInst::ICMP_EQ, UI,
ConstantInt::get(UI->getType(), 0));
LLVM_DEBUG(dbgs() << "Converting " << *Cmp << "\n");
@@ -8167,6 +8168,7 @@ static bool optimizeBranch(BranchInst *Branch, const
TargetLowering &TLI,
IRBuilder<> Builder(Branch);
if (UI->getParent() != Branch->getParent())
UI->moveBefore(Branch);
+ UI->dropPoisonGeneratingFlags();
Value *NewCmp = Builder.CreateCmp(Cmp->getPredicate(), UI,
ConstantInt::get(UI->getType(), 0));
LLVM_DEBUG(dbgs() << "Converting " << *Cmp << "\n");
diff --git a/llvm/test/Transforms/CodeGenPrepare/ARM/branch-on-zero.ll
b/llvm/test/Transforms/CodeGenPrepare/ARM/branch-on-zero.ll
index ff5cef7e781fe6..25dfb3c53a077b 100644
--- a/llvm/test/Transforms/CodeGenPrepare/ARM/branch-on-zero.ll
+++ b/llvm/test/Transforms/CodeGenPrepare/ARM/branch-on-zero.ll
@@ -211,6 +211,29 @@ else:
ret i32 %l
}
+define i32 @sub10_else_drop_nuw(i32 %a) {
+; CHECK-LABEL: @sub10_else_drop_nuw(
+; CHECK-NEXT: entry:
+; CHECK-NEXT:[[L:%.*]] = sub i32 [[A:%.*]], 10
+; CHECK-NEXT:[[TMP0:%.*]] = icmp eq i32 [[L]], 0
+; CHECK-NEXT:br i1 [[TMP0]], label [[THEN:%.*]], label [[ELSE:%.*]]
+; CHECK: then:
+; CHECK-NEXT:ret i32 0
+; CHECK: else:
+; CHECK-NEXT:ret i32 [[L]]
+;
+entry:
+ %c = icmp eq i32 %a, 10
+ br i1 %c, label %then, label %else
+
+then:
+ ret i32 0
+
+else:
+ %l = sub nuw i32 %a, 10
+ ret i32 %l
+}
+
define i32 @subm10_then(i32 %a) {
; CHECK-LABEL: @subm10_then(
; CHECK-NEXT: entry:
diff --git a/llvm/test/Transforms/CodeGenPrepare/RISCV/convert-to-eqz.ll
b/llvm/test/Transforms/CodeGenPrepare/RISCV/convert-to-eqz.ll
new file mode 100644
index 00..a6909d14913494
--- /dev/null
+++ b/llvm/test/Transforms/CodeGenPrepare/RISCV/convert-to-eqz.ll
@@ -0,0 +1,80 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
UTC_ARGS: --version 4
+; RUN: opt -codegenprepare -S -mtriple=riscv64 < %s | FileCheck %s
+
+define i8 @hoist_add(i8 %x) {
+; CHECK-LABEL: define i8 @hoist_add(
+; CHECK-SAME: i8 [[X:%.*]]) {
+; CHECK-NEXT: entry:
+; CHECK-NEXT:[[INC:%.*]] = add i8 [[X]], 1
+; CHECK-NEXT:[[TMP0:%.*]] = icmp eq i8 [[INC]], 0
+; CHECK-NEXT:br i1 [[TMP0]], label [[EXIT:%.*]], label [[IF_THEN:%.*]]
+; CHECK: if.then:
+; CHECK-NEXT:br label [[EXIT]]
+; CHECK: exit:
+; CHECK-NEXT:[[RETVAL:%.*]] = phi i8 [ [[INC]], [[IF_THEN]] ], [ -1,
[[ENTRY:%.*]] ]
+; CHECK-NEXT:ret i8 [[RETVAL]]
+;
+entry:
+ %cmp = icmp eq i8 %x, -1
+ br i1 %cmp, label %exit, label %if.then
+
+if.then:
+ %inc = add nuw nsw i8 %x, 1
+ br label %exit
+
+exit:
+ %retval = phi i8 [ %inc, %if.then ], [ -1, %entry ]
+ ret i8 %retval
+}
+
+define i8 @hoist_lshr(i8 %x) {
+; CHECK-LABEL: define i8 @hoist_lshr(
+; CHECK-SAME: i8 [[X:%.*]]) {
+; CHECK-NEXT: entry:
+; CHECK-NEXT:[[INC:%.*]] = lshr i8 [[X]], 3
+; CHECK-NEXT:[[TMP0:%.*]] = icmp eq i8 [[INC]], 0
+; CHECK-NEXT:br i1 [[TMP0]], label [[EXIT:%.*]], label [[IF_THEN:%.*]]
+; CHECK: if.then:
+; CHECK-NEXT:br label [[EXIT]]
+; CHECK: exit:
+; CHECK-NEXT:[[RETVAL:%.*]] = phi i8 [ [[INC]], [[IF_THEN]] ], [ -1,
[[ENTRY:%.*]] ]
+; CHECK-NEXT:ret i8 [[RETVAL]]
+;
+entry:
+ %cmp = icmp ult i8 %x, 8
+ br i1 %cmp, label %exit, label %if.then
+
+if.then:
+ %inc = lshr exact i8 %x, 3
+ br label %exit
+
+exit:
+ %retval = phi i8 [ %inc, %if.then ], [ -1, %entry ]
+ ret i8 %retval
+}
+
+define i8 @nomove_add(i8 %x) {
+; CHECK-LABEL: define i8 @nomove_add(
+; CHECK-SAME: i8 [[X:%.*]]) {
+; CHECK-NEXT: entry:
+; CHECK-NEXT:[[INC:%.*]] = add i8 [[X]], 1
+; CHECK-NEXT:[[TMP0:%.*]] = icmp eq i8 [[INC]], 0
+; CHECK-NEXT:br i1 [[TMP0]], label [[EXIT:%.*]], label [[IF_THEN:%.*]]
+; CHECK: if.then:
+; CHECK-NEXT:br label [[EXIT]]
+; CHECK: exit:
+; CHECK-NEXT:[[RETVAL:%.*]] = phi i8 [ [[INC]], [[IF_THEN]] ], [
[llvm-branch-commits] [llvm] release/18.x: [X86][EVEX512] Check hasEVEX512 for canExtendTo512DQ (#90390) (PR #90422)
https://github.com/RKSimon approved this pull request. LGTM https://github.com/llvm/llvm-project/pull/90422 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] release/18.x: [CGP] Drop poison-generating flags after hoisting (#90382) (PR #90437)
https://github.com/nikic approved this pull request. LGTM https://github.com/llvm/llvm-project/pull/90437 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [LoongArch][Codegen] Add support for TLSDESC (PR #90159)
https://github.com/SixWeining edited https://github.com/llvm/llvm-project/pull/90159 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [LoongArch][Codegen] Add support for TLSDESC (PR #90159)
https://github.com/SixWeining approved this pull request. LGTM except a nit. https://github.com/llvm/llvm-project/pull/90159 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [LoongArch][Codegen] Add support for TLSDESC (PR #90159)
@@ -737,7 +737,7 @@ bool tools::isTLSDESCEnabled(const ToolChain &TC,
StringRef V = A->getValue();
bool SupportedArgument = false, EnableTLSDESC = false;
bool Unsupported = !Triple.isOSBinFormatELF();
- if (Triple.isRISCV()) {
+ if (Triple.isRISCV() || Triple.isLoongArch()) {
SixWeining wrote:
Is it better to sort alphabetically?
https://github.com/llvm/llvm-project/pull/90159
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [flang] acc59b6 - Revert "[flang] Improve debug info for functions. (#90083)"
Author: David Spickett
Date: 2024-04-29T11:01:28+01:00
New Revision: acc59b62bc10fe034faf876c9a23f92dd4146505
URL:
https://github.com/llvm/llvm-project/commit/acc59b62bc10fe034faf876c9a23f92dd4146505
DIFF:
https://github.com/llvm/llvm-project/commit/acc59b62bc10fe034faf876c9a23f92dd4146505.diff
LOG: Revert "[flang] Improve debug info for functions. (#90083)"
This reverts commit f029da5cfce66e2d460d5f54b17582810a9111ad.
Added:
Modified:
flang/lib/Optimizer/Transforms/AddDebugInfo.cpp
flang/lib/Optimizer/Transforms/CMakeLists.txt
flang/test/Transforms/debug-line-table-inc-file.fir
Removed:
flang/lib/Optimizer/Transforms/DebugTypeGenerator.cpp
flang/lib/Optimizer/Transforms/DebugTypeGenerator.h
flang/test/Transforms/debug-fn-info.f90
diff --git a/flang/lib/Optimizer/Transforms/AddDebugInfo.cpp
b/flang/lib/Optimizer/Transforms/AddDebugInfo.cpp
index 5108469cb6c84a..18d98a11ef3c40 100644
--- a/flang/lib/Optimizer/Transforms/AddDebugInfo.cpp
+++ b/flang/lib/Optimizer/Transforms/AddDebugInfo.cpp
@@ -11,7 +11,6 @@
/// This pass populates some debug information for the module and functions.
//===--===//
-#include "DebugTypeGenerator.h"
#include "flang/Common/Version.h"
#include "flang/Optimizer/Builder/FIRBuilder.h"
#include "flang/Optimizer/Builder/Todo.h"
@@ -107,27 +106,14 @@ void AddDebugInfoPass::runOnOperation() {
filePath =
llvm::sys::path::parent_path(funcLoc.getFilename().getValue());
}
-mlir::StringAttr fullName =
-mlir::StringAttr::get(context, funcOp.getName());
-auto result = fir::NameUniquer::deconstruct(funcOp.getName());
mlir::StringAttr funcName =
-mlir::StringAttr::get(context, result.second.name);
-
-llvm::SmallVector types;
-fir::DebugTypeGenerator typeGen(module);
-for (auto resTy : funcOp.getResultTypes()) {
- auto tyAttr =
- typeGen.convertType(resTy, fileAttr, cuAttr, funcOp.getLoc());
- types.push_back(tyAttr);
-}
-for (auto inTy : funcOp.getArgumentTypes()) {
- auto tyAttr = typeGen.convertType(fir::unwrapRefType(inTy), fileAttr,
-cuAttr, funcOp.getLoc());
- types.push_back(tyAttr);
-}
-
+mlir::StringAttr::get(context, funcOp.getName());
+mlir::LLVM::DIBasicTypeAttr bT = mlir::LLVM::DIBasicTypeAttr::get(
+context, llvm::dwarf::DW_TAG_base_type, "void", /*sizeInBits=*/0,
+/*encoding=*/1);
+// FIXME: Provide proper type for subroutine
mlir::LLVM::DISubroutineTypeAttr subTypeAttr =
-mlir::LLVM::DISubroutineTypeAttr::get(context, CC, types);
+mlir::LLVM::DISubroutineTypeAttr::get(context, CC, {bT, bT});
mlir::LLVM::DIFileAttr funcFileAttr =
mlir::LLVM::DIFileAttr::get(context, fileName, filePath);
@@ -144,13 +130,11 @@ void AddDebugInfoPass::runOnOperation() {
subprogramFlags =
subprogramFlags | mlir::LLVM::DISubprogramFlags::Definition;
}
-unsigned line = 1;
-if (auto funcLoc = l.dyn_cast())
- line = funcLoc.getLine();
-
+// FIXME: Provide proper line and scopeline.
auto spAttr = mlir::LLVM::DISubprogramAttr::get(
-context, id, compilationUnit, fileAttr, funcName, fullName,
-funcFileAttr, line, line, subprogramFlags, subTypeAttr);
+context, id, compilationUnit, fileAttr, funcName, funcName,
+funcFileAttr, /*line=*/1, /*scopeline=*/1, subprogramFlags,
+subTypeAttr);
funcOp->setLoc(builder.getFusedLoc({funcOp->getLoc()}, spAttr));
});
}
diff --git a/flang/lib/Optimizer/Transforms/CMakeLists.txt
b/flang/lib/Optimizer/Transforms/CMakeLists.txt
index 5a542f237f8f98..fc08d67540ceb0 100644
--- a/flang/lib/Optimizer/Transforms/CMakeLists.txt
+++ b/flang/lib/Optimizer/Transforms/CMakeLists.txt
@@ -22,7 +22,6 @@ add_flang_library(FIRTransforms
OMPMarkDeclareTarget.cpp
VScaleAttr.cpp
FunctionAttr.cpp
- DebugTypeGenerator.cpp
DEPENDS
FIRDialect
diff --git a/flang/lib/Optimizer/Transforms/DebugTypeGenerator.cpp
b/flang/lib/Optimizer/Transforms/DebugTypeGenerator.cpp
deleted file mode 100644
index 0057e08f8a763b..00
--- a/flang/lib/Optimizer/Transforms/DebugTypeGenerator.cpp
+++ /dev/null
@@ -1,63 +0,0 @@
-//===-- DebugTypeGenerator.cpp -- type conversion ---*- C++
-*-===//
-//
-// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
-// See https://llvm.org/LICENSE.txt for license information.
-// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
-//
-//===--===//
-//
-// Coding style: https://mlir.llvm.org/getting_started/DeveloperGuide/
-//
-//===--===//
-
-#define DEBUG_TYPE "flang-debug-type-gene
[llvm-branch-commits] [clang] [CIR] Add options to emit ClangIR and enable the ClangIR pipeline (PR #89030)
https://github.com/erichkeane approved this pull request. https://github.com/llvm/llvm-project/pull/89030 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] release/18.x: [clang][CoverageMapping] do not emit a gap region when either end doesn't have valid source locations (#89564) (PR #90369)
https://github.com/ZequanWu approved this pull request. https://github.com/llvm/llvm-project/pull/90369 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [GSYM][DWARF] Include end_sequence debug_line rows when looking up addr ranges (PR #89703)
https://github.com/avillega updated https://github.com/llvm/llvm-project/pull/89703 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [GSYM][DWARF] Include end_sequence debug_line rows when looking up addr ranges (PR #89703)
https://github.com/avillega updated https://github.com/llvm/llvm-project/pull/89703 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [Serialization] Code cleanups and polish 83233 (PR #83237)
alexfh wrote: Apologies for the delay with the reproducer. @ilya-biryukov is on vacation, but maybe @sam-mccall can help with an isolated test case. Either way it seems to be challenging. https://github.com/llvm/llvm-project/pull/83237 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [release/18.x][COFF][Aarch64] Add _InterlockedAdd64 intrinsic (#81849) (PR #89951)
https://github.com/dpaoliello updated
https://github.com/llvm/llvm-project/pull/89951
>From d1a4febc0974cb76bce1692be9d62aaa9a5dcefb Mon Sep 17 00:00:00 2001
From: Pierrick Bouvier
Date: Fri, 16 Feb 2024 15:20:08 +0400
Subject: [PATCH] [COFF][Aarch64] Add _InterlockedAdd64 intrinsic (#81849)
Found when compiling openssl master branch using clang-cl.
This commit introduces usage of InterlockedAdd64:
https://github.com/openssl/openssl/commit/d0e1a0ae701cfaca7f3dd3bf28a3f934a6408813
https://learn.microsoft.com/en-us/cpp/intrinsics/interlockedadd-intrinsic-functions
---
clang/include/clang/Basic/BuiltinsAArch64.def | 1 +
clang/lib/CodeGen/CGBuiltin.cpp | 3 ++-
clang/lib/Headers/intrin.h | 1 +
clang/test/CodeGen/arm64-microsoft-intrinsics.c | 14 ++
clang/test/CodeGen/ms-intrinsics-other.c| 9 +
clang/test/CodeGen/ms-intrinsics-underaligned.c | 6 ++
6 files changed, 33 insertions(+), 1 deletion(-)
diff --git a/clang/include/clang/Basic/BuiltinsAArch64.def
b/clang/include/clang/Basic/BuiltinsAArch64.def
index 31ec84143f65c1..b5cbe90c8fd6a3 100644
--- a/clang/include/clang/Basic/BuiltinsAArch64.def
+++ b/clang/include/clang/Basic/BuiltinsAArch64.def
@@ -139,6 +139,7 @@ TARGET_HEADER_BUILTIN(_BitScanForward64, "UcUNi*ULLi",
"nh", INTRIN_H, ALL_MS_LA
TARGET_HEADER_BUILTIN(_BitScanReverse64, "UcUNi*ULLi", "nh", INTRIN_H,
ALL_MS_LANGUAGES, "")
TARGET_HEADER_BUILTIN(_InterlockedAdd, "NiNiD*Ni","nh",
INTRIN_H, ALL_MS_LANGUAGES, "")
+TARGET_HEADER_BUILTIN(_InterlockedAdd64, "LLiLLiD*LLi", "nh",
INTRIN_H, ALL_MS_LANGUAGES, "")
TARGET_HEADER_BUILTIN(_InterlockedAnd64, "LLiLLiD*LLi", "nh",
INTRIN_H, ALL_MS_LANGUAGES, "")
TARGET_HEADER_BUILTIN(_InterlockedDecrement64, "LLiLLiD*","nh",
INTRIN_H, ALL_MS_LANGUAGES, "")
TARGET_HEADER_BUILTIN(_InterlockedExchange64,"LLiLLiD*LLi", "nh",
INTRIN_H, ALL_MS_LANGUAGES, "")
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index a4f26a6f0eb19b..05a898a24b5890 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -12044,7 +12044,8 @@ Value *CodeGenFunction::EmitAArch64BuiltinExpr(unsigned
BuiltinID,
"vgetq_lane");
}
- case clang::AArch64::BI_InterlockedAdd: {
+ case clang::AArch64::BI_InterlockedAdd:
+ case clang::AArch64::BI_InterlockedAdd64: {
Address DestAddr = CheckAtomicAlignment(*this, E);
Value *Val = EmitScalarExpr(E->getArg(1));
AtomicRMWInst *RMWI =
diff --git a/clang/lib/Headers/intrin.h b/clang/lib/Headers/intrin.h
index 9ebaea9fee9421..a6395143db54c2 100644
--- a/clang/lib/Headers/intrin.h
+++ b/clang/lib/Headers/intrin.h
@@ -551,6 +551,7 @@ static __inline__ void __DEFAULT_FN_ATTRS __nop(void) {
#if defined(__aarch64__)
unsigned __int64 __getReg(int);
long _InterlockedAdd(long volatile *Addend, long Value);
+__int64 _InterlockedAdd64(__int64 volatile *Addend, __int64 Value);
__int64 _ReadStatusReg(int);
void _WriteStatusReg(int, __int64);
diff --git a/clang/test/CodeGen/arm64-microsoft-intrinsics.c
b/clang/test/CodeGen/arm64-microsoft-intrinsics.c
index 44b2ee28fe5681..a354ed948ca5f1 100644
--- a/clang/test/CodeGen/arm64-microsoft-intrinsics.c
+++ b/clang/test/CodeGen/arm64-microsoft-intrinsics.c
@@ -21,6 +21,20 @@ long test_InterlockedAdd_constant(long volatile *Addend) {
// CHECK-MSVC: ret i32 %[[NEWVAL:[0-9]+]]
// CHECK-LINUX: error: call to undeclared function '_InterlockedAdd'
+__int64 test_InterlockedAdd64(__int64 volatile *Addend, __int64 Value) {
+ return _InterlockedAdd64(Addend, Value);
+}
+
+__int64 test_InterlockedAdd64_constant(__int64 volatile *Addend) {
+ return _InterlockedAdd64(Addend, -1);
+}
+
+// CHECK-LABEL: define {{.*}} i64 @test_InterlockedAdd64(ptr %Addend, i64
%Value) {{.*}} {
+// CHECK-MSVC: %[[OLDVAL:[0-9]+]] = atomicrmw add ptr %1, i64 %2 seq_cst,
align 8
+// CHECK-MSVC: %[[NEWVAL:[0-9]+]] = add i64 %[[OLDVAL:[0-9]+]], %2
+// CHECK-MSVC: ret i64 %[[NEWVAL:[0-9]+]]
+// CHECK-LINUX: error: call to undeclared function '_InterlockedAdd64'
+
void check__dmb(void) {
__dmb(0);
}
diff --git a/clang/test/CodeGen/ms-intrinsics-other.c
b/clang/test/CodeGen/ms-intrinsics-other.c
index 76f54add749669..36c40dddcbb4f5 100644
--- a/clang/test/CodeGen/ms-intrinsics-other.c
+++ b/clang/test/CodeGen/ms-intrinsics-other.c
@@ -240,6 +240,15 @@ LONG test_InterlockedAdd(LONG volatile *Addend, LONG
Value) {
// CHECK-ARM-ARM64: %[[OLDVAL:[0-9]+]] = atomicrmw add ptr %Addend, i32 %Value
seq_cst, align 4
// CHECK-ARM-ARM64: %[[NEWVAL:[0-9]+]] = add i32 %[[OLDVAL:[0-9]+]], %Value
// CHECK-ARM-ARM64: ret i32 %[[NEWVAL:[0-9]+]]
+
+__int64 test_InterlockedAdd64(__int64 volatile *Addend, __int64 Value) {
+ return _InterlockedAdd64(Addend, Value);
+}
+
+// CHECK-ARM-ARM64: define{{.*}}i64 @test_InterlockedAdd64(ptr{{[a-z_
]*}}%Addend, i64 noundef %Value) {{.*}} {
+// CHECK-ARM-ARM6
[llvm-branch-commits] [clang] [release/18.x][COFF][Aarch64] Add _InterlockedAdd64 intrinsic (#81849) (PR #89951)
dpaoliello wrote: > I think BuiltinsAArch64.def is part of clang's ABI, so changing it violates > the backport rules. > > Otherwise, I'd be inclined to accept; it's kind of late to request, but it's > low risk. @tstellar can you please advise if this change is ok to backport? https://github.com/llvm/llvm-project/pull/89951 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm][NFC] Document cl::opt variable and fix typo (PR #90145)
https://github.com/ilovepi updated https://github.com/llvm/llvm-project/pull/90145 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm][NFC] Document cl::opt variable and fix typo (PR #90145)
https://github.com/ilovepi updated https://github.com/llvm/llvm-project/pull/90145 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm][IR] Extend BranchWeightMetadata to track provenance of weights (PR #86609)
https://github.com/ilovepi updated https://github.com/llvm/llvm-project/pull/86609 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm][IR] Extend BranchWeightMetadata to track provenance of weights (PR #86609)
https://github.com/ilovepi updated https://github.com/llvm/llvm-project/pull/86609 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm][misexpect] Update MisExpect to use provenance tracking metadata (PR #86610)
https://github.com/ilovepi updated https://github.com/llvm/llvm-project/pull/86610 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm][misexpect] Update MisExpect to use provenance tracking metadata (PR #86610)
https://github.com/ilovepi updated https://github.com/llvm/llvm-project/pull/86610 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm][ProfDataUtils] provide getNumBranchWeights API (PR #90146)
https://github.com/ilovepi updated https://github.com/llvm/llvm-project/pull/90146 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm][ProfDataUtils] provide getNumBranchWeights API (PR #90146)
https://github.com/ilovepi updated https://github.com/llvm/llvm-project/pull/90146 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][CallGraphSection] Add type id metadata to indirect call and targets (PR #87573)
https://github.com/Prabhuk updated
https://github.com/llvm/llvm-project/pull/87573
>From a8a5848885e12c771f12cfa33b4dbc6a0272e925 Mon Sep 17 00:00:00 2001
From: Prabhuk
Date: Mon, 22 Apr 2024 11:34:04 -0700
Subject: [PATCH 1/3] Update clang/lib/CodeGen/CodeGenModule.cpp
Cleaner if checks.
Co-authored-by: Matt Arsenault
---
clang/lib/CodeGen/CodeGenModule.cpp | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/clang/lib/CodeGen/CodeGenModule.cpp
b/clang/lib/CodeGen/CodeGenModule.cpp
index e19bbee996f582..ff1586d2fa8abe 100644
--- a/clang/lib/CodeGen/CodeGenModule.cpp
+++ b/clang/lib/CodeGen/CodeGenModule.cpp
@@ -2711,7 +2711,7 @@ void
CodeGenModule::CreateFunctionTypeMetadataForIcall(const FunctionDecl *FD,
void CodeGenModule::CreateFunctionTypeMetadataForIcall(const QualType &QT,
llvm::CallBase *CB) {
// Only if needed for call graph section and only for indirect calls.
- if (!(CodeGenOpts.CallGraphSection && CB && CB->isIndirectCall()))
+ if (!CodeGenOpts.CallGraphSection || !CB || !CB->isIndirectCall())
return;
auto *MD = CreateMetadataIdentifierGeneralized(QT);
>From 019b2ca5e1c263183ed114e0b967b4e77b4a17a8 Mon Sep 17 00:00:00 2001
From: Prabhuk
Date: Mon, 22 Apr 2024 11:34:31 -0700
Subject: [PATCH 2/3] Update clang/lib/CodeGen/CodeGenModule.cpp
Update the comments as suggested.
Co-authored-by: Matt Arsenault
---
clang/lib/CodeGen/CodeGenModule.cpp | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/clang/lib/CodeGen/CodeGenModule.cpp
b/clang/lib/CodeGen/CodeGenModule.cpp
index ff1586d2fa8abe..5635a87d2358a7 100644
--- a/clang/lib/CodeGen/CodeGenModule.cpp
+++ b/clang/lib/CodeGen/CodeGenModule.cpp
@@ -2680,9 +2680,9 @@ void
CodeGenModule::CreateFunctionTypeMetadataForIcall(const FunctionDecl *FD,
bool EmittedMDIdGeneralized = false;
if (CodeGenOpts.CallGraphSection &&
(!F->hasLocalLinkage() ||
- F->getFunction().hasAddressTaken(nullptr, /* IgnoreCallbackUses */ true,
-/* IgnoreAssumeLikeCalls */ true,
-/* IgnoreLLVMUsed */ false))) {
+ F->getFunction().hasAddressTaken(nullptr, /*IgnoreCallbackUses=*/ true,
+/*IgnoreAssumeLikeCalls=*/ true,
+/*IgnoreLLVMUsed=*/ false))) {
F->addTypeMetadata(0, CreateMetadataIdentifierGeneralized(FD->getType()));
EmittedMDIdGeneralized = true;
}
>From 99242900c51778abd4b7e7f4361b09202b7abcda Mon Sep 17 00:00:00 2001
From: Prabhuk
Date: Mon, 29 Apr 2024 11:53:40 -0700
Subject: [PATCH 3/3] dyn_cast to isa
Created using spr 1.3.6-beta.1
---
clang/lib/CodeGen/CGCall.cpp | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/clang/lib/CodeGen/CGCall.cpp b/clang/lib/CodeGen/CGCall.cpp
index 526a63b24ff834..45033ced1d8344 100644
--- a/clang/lib/CodeGen/CGCall.cpp
+++ b/clang/lib/CodeGen/CGCall.cpp
@@ -5713,8 +5713,8 @@ RValue CodeGenFunction::EmitCall(const CGFunctionInfo
&CallInfo,
if (callOrInvoke && *callOrInvoke && (*callOrInvoke)->isIndirectCall()) {
if (const FunctionDecl *FD = dyn_cast_or_null(TargetDecl))
{
// Type id metadata is set only for C/C++ contexts.
-if (dyn_cast(FD) || dyn_cast(FD) ||
-dyn_cast(FD)) {
+if (isa(FD) || isa(FD) ||
+isa(FD)) {
CGM.CreateFunctionTypeMetadataForIcall(FD->getType(), *callOrInvoke);
}
}
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [CallSiteInfo][CallGraphSection] Extend CallSiteInfo with TypeId (PR #87574)
https://github.com/Prabhuk updated https://github.com/llvm/llvm-project/pull/87574 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [CallSiteInfo][CallGraphSection] Extend CallSiteInfo with TypeId (PR #87574)
https://github.com/Prabhuk updated https://github.com/llvm/llvm-project/pull/87574 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [CallSiteInfo][CallGraphSection] Extract and propagate indirect call type ids (PR #87575)
https://github.com/Prabhuk updated https://github.com/llvm/llvm-project/pull/87575 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [CallSiteInfo][CallGraphSection] Extract and propagate indirect call type ids (PR #87575)
https://github.com/Prabhuk updated https://github.com/llvm/llvm-project/pull/87575 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [AsmPrinter][CallGraphSection] Emit call graph section (PR #87576)
https://github.com/Prabhuk updated https://github.com/llvm/llvm-project/pull/87576 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [AsmPrinter][CallGraphSection] Emit call graph section (PR #87576)
https://github.com/Prabhuk updated https://github.com/llvm/llvm-project/pull/87576 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][CallGraphSection] Add type id metadata to indirect call and targets (PR #87573)
@@ -5687,6 +5688,39 @@ RValue CodeGenFunction::EmitCall(const CGFunctionInfo
&CallInfo,
AllocAlignAttrEmitter AllocAlignAttrEmitter(*this, TargetDecl, CallArgs);
Attrs = AllocAlignAttrEmitter.TryEmitAsCallSiteAttribute(Attrs);
+ if (CGM.getCodeGenOpts().CallGraphSection) {
+// FIXME: create operand bundle only for indirect calls, not for all
+
+assert((TargetDecl && TargetDecl->getFunctionType() ||
+Callee.getAbstractInfo().getCalleeFunctionProtoType()) &&
+ "cannot find callsite type");
+
+QualType CST;
+if (TargetDecl && TargetDecl->getFunctionType())
+ CST = QualType(TargetDecl->getFunctionType(), 0);
+else if (const auto *FPT =
+ Callee.getAbstractInfo().getCalleeFunctionProtoType())
+ CST = QualType(FPT, 0);
+
+if (!CST.isNull()) {
+ auto *TypeIdMD = CGM.CreateMetadataIdentifierGeneralized(CST);
+ auto *TypeIdMDVal =
+ llvm::MetadataAsValue::get(getLLVMContext(), TypeIdMD);
+ BundleList.emplace_back("type", TypeIdMDVal);
+}
+
+// Set type identifier metadata of indirect calls for call graph section.
+if (callOrInvoke && *callOrInvoke && (*callOrInvoke)->isIndirectCall()) {
+ if (const FunctionDecl *FD = dyn_cast_or_null(TargetDecl))
{
+// Type id metadata is set only for C/C++ contexts.
+if (dyn_cast(FD) || dyn_cast(FD) ||
+dyn_cast(FD)) {
Prabhuk wrote:
Thank you. Done.
https://github.com/llvm/llvm-project/pull/87573
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [BOLT] Use heuristic for matching split local functions (PR #90424)
https://github.com/aaupov updated
https://github.com/llvm/llvm-project/pull/90424
>From b9a8ff53a491f126fb0dcf8cc371ed7a053e859b Mon Sep 17 00:00:00 2001
From: Amir Ayupov
Date: Mon, 29 Apr 2024 12:48:24 -0700
Subject: [PATCH 1/3] Revert cdsplit-symbol-names.s
---
bolt/test/X86/cdsplit-symbol-names.s | 24
1 file changed, 4 insertions(+), 20 deletions(-)
diff --git a/bolt/test/X86/cdsplit-symbol-names.s
b/bolt/test/X86/cdsplit-symbol-names.s
index 0d9149f49db908..e53863e22246d6 100644
--- a/bolt/test/X86/cdsplit-symbol-names.s
+++ b/bolt/test/X86/cdsplit-symbol-names.s
@@ -2,14 +2,12 @@
# Warm section should have name .text.warm and warm function fragments should
# have symbol names ending in warm.
-# RUN: split-file %s %t
-# RUN: llvm-mc --filetype=obj --triple x86_64-unknown-unknown %t/main.s -o %t.o
-# RUN: llvm-mc --filetype=obj --triple x86_64-unknown-unknown %t/chain.s -o
%t.chain.o
+# RUN: llvm-mc --filetype=obj --triple x86_64-unknown-unknown %s -o %t.o
# RUN: link_fdata %s %t.o %t.fdata
# RUN: llvm-strip --strip-unneeded %t.o
-# RUN: %clang %cflags %t.o %t.chain.o -o %t.exe -Wl,-q
+# RUN: %clang %cflags %t.o -o %t.exe -Wl,-q
# RUN: llvm-bolt %t.exe -o %t.bolt --split-functions --split-strategy=cdsplit \
-# RUN: --call-scale=2 --data=%t.fdata --reorder-blocks=ext-tsp --enable-bat
+# RUN: --call-scale=2 --data=%t.fdata --reorder-blocks=ext-tsp
# RUN: llvm-objdump --syms %t.bolt | FileCheck %s
--check-prefix=CHECK-SYMS-WARM
# CHECK-SYMS-WARM: l df *ABS* bolt-pseudo.o
@@ -18,22 +16,8 @@
# CHECK-SYMS-WARM: .text.cold
# CHECK-SYMS-WARM-SAME: dummy.cold
-# RUN: link_fdata %s %t.bolt %t.preagg PREAGG
-# PREAGG: B X:0 #chain.warm# 1 0
-# RUN: perf2bolt %t.bolt -p %t.preagg --pa -o %t.bat.fdata -w %t.bat.yaml -v=1
\
-# RUN: | FileCheck %s --check-prefix=CHECK-REGISTER
-
-# CHECK-REGISTER: BOLT-INFO: marking chain.warm/1(*2) as a fragment of
chain/2(*2)
-
-#--- chain.s
-.text
-.type chain, @function
-chain:
-ret
-.size chain, .-chain
-
-#--- main.s
.text
+.globl chain
.type chain, @function
chain:
pushq %rbp
>From 7c92aca6a0cba71bfde9eb2127575d93e210d346 Mon Sep 17 00:00:00 2001
From: Amir Ayupov
Date: Mon, 29 Apr 2024 12:48:39 -0700
Subject: [PATCH 2/3] Add register-fragments-bolt-symbols.s
---
.../X86/register-fragments-bolt-symbols.s | 32 +++
1 file changed, 32 insertions(+)
create mode 100644 bolt/test/X86/register-fragments-bolt-symbols.s
diff --git a/bolt/test/X86/register-fragments-bolt-symbols.s
b/bolt/test/X86/register-fragments-bolt-symbols.s
new file mode 100644
index 00..fa9b70e0b2d891
--- /dev/null
+++ b/bolt/test/X86/register-fragments-bolt-symbols.s
@@ -0,0 +1,32 @@
+# Test the heuristics for matching BOLT-added split functions.
+
+# RUN: llvm-mc --filetype=obj --triple x86_64-unknown-unknown
%S/cdsplit-symbol-names.s -o %t.main.o
+# RUN: llvm-mc --filetype=obj --triple x86_64-unknown-unknown %s -o %t.chain.o
+# RUN: link_fdata %S/cdsplit-symbol-names.s %t.main.o %t.fdata
+# RUN: sed -i 's|chain|chain/2|g' %t.fdata
+# RUN: llvm-strip --strip-unneeded %t.main.o
+# RUN: llvm-objcopy --localize-symbol=chain %t.main.o
+# RUN: %clang %cflags %t.chain.o %t.main.o -o %t.exe -Wl,-q
+# RUN: llvm-bolt %t.exe -o %t.bolt --split-functions --split-strategy=randomN \
+# RUN: --reorder-blocks=ext-tsp --enable-bat --bolt-seed=7 --data=%t.fdata
+# RUN: llvm-objdump --syms %t.bolt | FileCheck %s --check-prefix=CHECK-SYMS
+
+# RUN: link_fdata %s %t.bolt %t.preagg PREAGG
+# PREAGG: B X:0 #chain.cold.0# 1 0
+# RUN: perf2bolt %t.bolt -p %t.preagg --pa -o %t.bat.fdata -w %t.bat.yaml -v=1
\
+# RUN: | FileCheck %s --check-prefix=CHECK-REGISTER
+
+# CHECK-SYMS: l df *ABS* [[#]] chain.s
+# CHECK-SYMS: l F .bolt.org.text [[#]] chain
+# CHECK-SYMS: l F .text.cold [[#]] chain.cold.0
+# CHECK-SYMS: l F .text [[#]] chain
+# CHECK-SYMS: l df *ABS* [[#]] bolt-pseudo.o
+
+# CHECK-REGISTER: BOLT-INFO: marking chain.cold.0/1(*2) as a fragment of
chain/2(*2)
+
+.file "chain.s"
+.text
+.type chain, @function
+chain:
+ret
+.size chain, .-chain
>From a94e14566de73e4a54a28086c95d742a1f09e7b4 Mon Sep 17 00:00:00 2001
From: Amir Ayupov
Date: Mon, 29 Apr 2024 12:50:07 -0700
Subject: [PATCH 3/3] Fix a mistake in NameOrError
---
bolt/lib/Rewrite/RewriteInstance.cpp | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/bolt/lib/Rewrite/RewriteInstance.cpp
b/bolt/lib/Rewrite/RewriteInstance.cpp
index 8eb2e5a9d9120a..5e7bcd06df82c4 100644
--- a/bolt/lib/Rewrite/RewriteInstance.cpp
+++ b/bolt/lib/Rewrite/RewriteInstance.cpp
@@ -1515,7 +1515,7 @@ void RewriteInstance::registerFragments() {
// symbol.
for (ELFSymbolRef NextSymbol = Symbol; NextSymbol < StopSymbol;
NextSymbol.moveNext()) {
- Expected NameOrError = Sy
[llvm-branch-commits] [llvm] [BOLT] Use heuristic for matching split local functions (PR #90424)
https://github.com/aaupov edited https://github.com/llvm/llvm-project/pull/90424 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [BOLT] Use heuristic for matching split local functions (PR #90424)
https://github.com/maksfb approved this pull request. https://github.com/llvm/llvm-project/pull/90424 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [CallGraphSection] Add call graph section options and documentation (PR #87572)
ilovepi wrote: A few comments. 1. I think its better to split the LLVM and clang changes into separate patches. 2. Please update the commit messages w/ the correct [clang] or [llvm] tag. 3. These should probably be some of the last patches to land. Enabling driver support in clang + documentation and release notes are normally the last step in landing a complex feature. The LLVM flag support is less important about where it is in the stack, since it's not exposed externally. https://github.com/llvm/llvm-project/pull/87572 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [MTE] add stack frame history buffer (PR #86356)
https://github.com/fmayer updated
https://github.com/llvm/llvm-project/pull/86356
>From a64c5d63a4df7f59845291ca0d634466713b1ff8 Mon Sep 17 00:00:00 2001
From: Florian Mayer
Date: Fri, 29 Mar 2024 16:53:52 -0700
Subject: [PATCH 1/2] update
Created using spr 1.3.4
---
llvm/lib/Target/AArch64/AArch64StackTagging.cpp | 1 -
1 file changed, 1 deletion(-)
diff --git a/llvm/lib/Target/AArch64/AArch64StackTagging.cpp
b/llvm/lib/Target/AArch64/AArch64StackTagging.cpp
index a6e236386d5baa..6538abea832907 100644
--- a/llvm/lib/Target/AArch64/AArch64StackTagging.cpp
+++ b/llvm/lib/Target/AArch64/AArch64StackTagging.cpp
@@ -489,7 +489,6 @@ Instruction *AArch64StackTagging::insertBaseTaggedPointer(
auto *IntptrTy = IRB.getIntPtrTy(M.getDataLayout());
Value *SlotPtr = memtag::getAndroidSlotPtr(IRB, StackMteSlot);
-SlotPtr->setName("TLS_SLOT_STACK_MTE");
auto *ThreadLong = IRB.CreateLoad(IntptrTy, SlotPtr);
Value *TaggedFP = IRB.CreateOr(
memtag::getFP(IRB),
>From 8591fb38c7e065862a0814792a368e2983b8b10c Mon Sep 17 00:00:00 2001
From: Florian Mayer
Date: Mon, 29 Apr 2024 14:45:16 -0700
Subject: [PATCH 2/2] api lvl
Created using spr 1.3.4
---
llvm/lib/Target/AArch64/AArch64StackTagging.cpp | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/llvm/lib/Target/AArch64/AArch64StackTagging.cpp
b/llvm/lib/Target/AArch64/AArch64StackTagging.cpp
index 840635315eee04..e38cce94a5cc0b 100644
--- a/llvm/lib/Target/AArch64/AArch64StackTagging.cpp
+++ b/llvm/lib/Target/AArch64/AArch64StackTagging.cpp
@@ -484,7 +484,7 @@ Instruction *AArch64StackTagging::insertBaseTaggedPointer(
Base->setName("basetag");
auto TargetTriple = Triple(M.getTargetTriple());
if (ClRecordStackHistory == instr && TargetTriple.isAndroid() &&
- TargetTriple.isAArch64() && !TargetTriple.isAndroidVersionLT(35)) {
+ TargetTriple.isAArch64() && !TargetTriple.isAndroidVersionLT(36)) {
constexpr int StackMteSlot = -3;
constexpr uint64_t TagMask = 0xFULL << 56;
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [flang] [flang][cuda] Lower device/managed/unified allocation to cuda ops (PR #90526)
https://github.com/clementval created
https://github.com/llvm/llvm-project/pull/90526
Lower locals allocation of cuda device, managed and unified variables to
fir.cuda_alloc. Add fir.cuda_free in the function context finalization.
>From 02d1ef45cae1ba973a51e5898f092403395c Mon Sep 17 00:00:00 2001
From: Valentin Clement
Date: Mon, 29 Apr 2024 14:30:46 -0700
Subject: [PATCH] [flang][cuda] Lower device/managed/unified allocation to cuda
ops
---
.../flang/Optimizer/Builder/FIRBuilder.h | 7 +
flang/include/flang/Semantics/tools.h | 17 +++
flang/lib/Lower/ConvertVariable.cpp | 29 +++
flang/lib/Optimizer/Builder/FIRBuilder.cpp| 25 +---
flang/lib/Optimizer/Dialect/FIROps.cpp| 15 ++
flang/test/Lower/CUDA/cuda-data-attribute.cuf | 25
6 files changed, 107 insertions(+), 11 deletions(-)
diff --git a/flang/include/flang/Optimizer/Builder/FIRBuilder.h
b/flang/include/flang/Optimizer/Builder/FIRBuilder.h
index e4c954159f71be..0d650f830b64e0 100644
--- a/flang/include/flang/Optimizer/Builder/FIRBuilder.h
+++ b/flang/include/flang/Optimizer/Builder/FIRBuilder.h
@@ -708,6 +708,13 @@ mlir::Value createNullBoxProc(fir::FirOpBuilder &builder,
mlir::Location loc,
/// Set internal linkage attribute on a function.
void setInternalLinkage(mlir::func::FuncOp);
+
+llvm::SmallVector
+elideExtentsAlreadyInType(mlir::Type type, mlir::ValueRange shape);
+
+llvm::SmallVector
+elideLengthsAlreadyInType(mlir::Type type, mlir::ValueRange lenParams);
+
} // namespace fir::factory
#endif // FORTRAN_OPTIMIZER_BUILDER_FIRBUILDER_H
diff --git a/flang/include/flang/Semantics/tools.h
b/flang/include/flang/Semantics/tools.h
index da10969ebc7021..c9eb5bc857ac01 100644
--- a/flang/include/flang/Semantics/tools.h
+++ b/flang/include/flang/Semantics/tools.h
@@ -222,6 +222,23 @@ inline bool HasCUDAAttr(const Symbol &sym) {
return false;
}
+inline bool NeedCUDAAlloc(const Symbol &sym) {
+ bool inDeviceSubprogram{IsCUDADeviceContext(&sym.owner())};
+ if (const auto *details{
+ sym.GetUltimate().detailsIf()}) {
+if (details->cudaDataAttr() &&
+(*details->cudaDataAttr() == common::CUDADataAttr::Device ||
+*details->cudaDataAttr() == common::CUDADataAttr::Managed ||
+*details->cudaDataAttr() == common::CUDADataAttr::Unified)) {
+ // Descriptor is allocated on host when in host context.
+ if (Fortran::semantics::IsAllocatable(sym))
+return inDeviceSubprogram;
+ return true;
+}
+ }
+ return false;
+}
+
const Scope *FindCUDADeviceContext(const Scope *);
std::optional GetCUDADataAttr(const Symbol *);
diff --git a/flang/lib/Lower/ConvertVariable.cpp
b/flang/lib/Lower/ConvertVariable.cpp
index 21db0cac11bf6a..9a17acf5b15c36 100644
--- a/flang/lib/Lower/ConvertVariable.cpp
+++ b/flang/lib/Lower/ConvertVariable.cpp
@@ -693,6 +693,22 @@ static mlir::Value
createNewLocal(Fortran::lower::AbstractConverter &converter,
if (ultimateSymbol.test(Fortran::semantics::Symbol::Flag::CrayPointee))
return builder.create(loc, fir::ReferenceType::get(ty));
+ if (Fortran::semantics::NeedCUDAAlloc(ultimateSymbol)) {
+fir::CUDADataAttributeAttr cudaAttr =
+Fortran::lower::translateSymbolCUDADataAttribute(builder.getContext(),
+ ultimateSymbol);
+llvm::SmallVector indices;
+llvm::SmallVector elidedShape =
+fir::factory::elideExtentsAlreadyInType(ty, shape);
+llvm::SmallVector elidedLenParams =
+fir::factory::elideLengthsAlreadyInType(ty, lenParams);
+auto idxTy = builder.getIndexType();
+for (mlir::Value sh : elidedShape)
+ indices.push_back(builder.createConvert(loc, idxTy, sh));
+return builder.create(loc, ty, nm, symNm, cudaAttr,
+lenParams, indices);
+ }
+
// Let the builder do all the heavy lifting.
if (!Fortran::semantics::IsProcedurePointer(ultimateSymbol))
return builder.allocateLocal(loc, ty, nm, symNm, shape, lenParams, isTarg);
@@ -927,6 +943,19 @@ static void
instantiateLocal(Fortran::lower::AbstractConverter &converter,
});
}
}
+ if (Fortran::semantics::NeedCUDAAlloc(var.getSymbol())) {
+auto *builder = &converter.getFirOpBuilder();
+mlir::Location loc = converter.getCurrentLocation();
+fir::ExtendedValue exv =
+converter.getSymbolExtendedValue(var.getSymbol(), &symMap);
+auto *sym = &var.getSymbol();
+converter.getFctCtx().attachCleanup([builder, loc, exv, sym]() {
+ fir::CUDADataAttributeAttr cudaAttr =
+ Fortran::lower::translateSymbolCUDADataAttribute(
+ builder->getContext(), *sym);
+ builder->create(loc, fir::getBase(exv), cudaAttr);
+});
+ }
}
//======//
diff --git a/flang/lib/Optimizer/Builder/FIRBuilder.cpp
b/flang/lib/Op
[llvm-branch-commits] [flang] [flang][cuda] Lower device/managed/unified allocation to cuda ops (PR #90526)
llvmbot wrote:
@llvm/pr-subscribers-flang-fir-hlfir
Author: Valentin Clement (バレンタイン クレメン) (clementval)
Changes
Lower locals allocation of cuda device, managed and unified variables to
fir.cuda_alloc. Add fir.cuda_free in the function context finalization.
---
Full diff: https://github.com/llvm/llvm-project/pull/90526.diff
6 Files Affected:
- (modified) flang/include/flang/Optimizer/Builder/FIRBuilder.h (+7)
- (modified) flang/include/flang/Semantics/tools.h (+17)
- (modified) flang/lib/Lower/ConvertVariable.cpp (+29)
- (modified) flang/lib/Optimizer/Builder/FIRBuilder.cpp (+14-11)
- (modified) flang/lib/Optimizer/Dialect/FIROps.cpp (+15)
- (modified) flang/test/Lower/CUDA/cuda-data-attribute.cuf (+25)
``diff
diff --git a/flang/include/flang/Optimizer/Builder/FIRBuilder.h
b/flang/include/flang/Optimizer/Builder/FIRBuilder.h
index e4c954159f71be..0d650f830b64e0 100644
--- a/flang/include/flang/Optimizer/Builder/FIRBuilder.h
+++ b/flang/include/flang/Optimizer/Builder/FIRBuilder.h
@@ -708,6 +708,13 @@ mlir::Value createNullBoxProc(fir::FirOpBuilder &builder,
mlir::Location loc,
/// Set internal linkage attribute on a function.
void setInternalLinkage(mlir::func::FuncOp);
+
+llvm::SmallVector
+elideExtentsAlreadyInType(mlir::Type type, mlir::ValueRange shape);
+
+llvm::SmallVector
+elideLengthsAlreadyInType(mlir::Type type, mlir::ValueRange lenParams);
+
} // namespace fir::factory
#endif // FORTRAN_OPTIMIZER_BUILDER_FIRBUILDER_H
diff --git a/flang/include/flang/Semantics/tools.h
b/flang/include/flang/Semantics/tools.h
index da10969ebc7021..c9eb5bc857ac01 100644
--- a/flang/include/flang/Semantics/tools.h
+++ b/flang/include/flang/Semantics/tools.h
@@ -222,6 +222,23 @@ inline bool HasCUDAAttr(const Symbol &sym) {
return false;
}
+inline bool NeedCUDAAlloc(const Symbol &sym) {
+ bool inDeviceSubprogram{IsCUDADeviceContext(&sym.owner())};
+ if (const auto *details{
+ sym.GetUltimate().detailsIf()}) {
+if (details->cudaDataAttr() &&
+(*details->cudaDataAttr() == common::CUDADataAttr::Device ||
+*details->cudaDataAttr() == common::CUDADataAttr::Managed ||
+*details->cudaDataAttr() == common::CUDADataAttr::Unified)) {
+ // Descriptor is allocated on host when in host context.
+ if (Fortran::semantics::IsAllocatable(sym))
+return inDeviceSubprogram;
+ return true;
+}
+ }
+ return false;
+}
+
const Scope *FindCUDADeviceContext(const Scope *);
std::optional GetCUDADataAttr(const Symbol *);
diff --git a/flang/lib/Lower/ConvertVariable.cpp
b/flang/lib/Lower/ConvertVariable.cpp
index 21db0cac11bf6a..9a17acf5b15c36 100644
--- a/flang/lib/Lower/ConvertVariable.cpp
+++ b/flang/lib/Lower/ConvertVariable.cpp
@@ -693,6 +693,22 @@ static mlir::Value
createNewLocal(Fortran::lower::AbstractConverter &converter,
if (ultimateSymbol.test(Fortran::semantics::Symbol::Flag::CrayPointee))
return builder.create(loc, fir::ReferenceType::get(ty));
+ if (Fortran::semantics::NeedCUDAAlloc(ultimateSymbol)) {
+fir::CUDADataAttributeAttr cudaAttr =
+Fortran::lower::translateSymbolCUDADataAttribute(builder.getContext(),
+ ultimateSymbol);
+llvm::SmallVector indices;
+llvm::SmallVector elidedShape =
+fir::factory::elideExtentsAlreadyInType(ty, shape);
+llvm::SmallVector elidedLenParams =
+fir::factory::elideLengthsAlreadyInType(ty, lenParams);
+auto idxTy = builder.getIndexType();
+for (mlir::Value sh : elidedShape)
+ indices.push_back(builder.createConvert(loc, idxTy, sh));
+return builder.create(loc, ty, nm, symNm, cudaAttr,
+lenParams, indices);
+ }
+
// Let the builder do all the heavy lifting.
if (!Fortran::semantics::IsProcedurePointer(ultimateSymbol))
return builder.allocateLocal(loc, ty, nm, symNm, shape, lenParams, isTarg);
@@ -927,6 +943,19 @@ static void
instantiateLocal(Fortran::lower::AbstractConverter &converter,
});
}
}
+ if (Fortran::semantics::NeedCUDAAlloc(var.getSymbol())) {
+auto *builder = &converter.getFirOpBuilder();
+mlir::Location loc = converter.getCurrentLocation();
+fir::ExtendedValue exv =
+converter.getSymbolExtendedValue(var.getSymbol(), &symMap);
+auto *sym = &var.getSymbol();
+converter.getFctCtx().attachCleanup([builder, loc, exv, sym]() {
+ fir::CUDADataAttributeAttr cudaAttr =
+ Fortran::lower::translateSymbolCUDADataAttribute(
+ builder->getContext(), *sym);
+ builder->create(loc, fir::getBase(exv), cudaAttr);
+});
+ }
}
//======//
diff --git a/flang/lib/Optimizer/Builder/FIRBuilder.cpp
b/flang/lib/Optimizer/Builder/FIRBuilder.cpp
index a0fbae5b614cc7..a813b646087d73 100644
--- a/flang/lib/Optimizer/Builder/FIRBuilder.cpp
+++ b/flang/lib/Optimizer/Builder/F
[llvm-branch-commits] [flang] [flang][cuda] Lower device/managed/unified allocation to cuda ops (PR #90526)
llvmbot wrote:
@llvm/pr-subscribers-flang-semantics
Author: Valentin Clement (バレンタイン クレメン) (clementval)
Changes
Lower locals allocation of cuda device, managed and unified variables to
fir.cuda_alloc. Add fir.cuda_free in the function context finalization.
---
Full diff: https://github.com/llvm/llvm-project/pull/90526.diff
6 Files Affected:
- (modified) flang/include/flang/Optimizer/Builder/FIRBuilder.h (+7)
- (modified) flang/include/flang/Semantics/tools.h (+17)
- (modified) flang/lib/Lower/ConvertVariable.cpp (+29)
- (modified) flang/lib/Optimizer/Builder/FIRBuilder.cpp (+14-11)
- (modified) flang/lib/Optimizer/Dialect/FIROps.cpp (+15)
- (modified) flang/test/Lower/CUDA/cuda-data-attribute.cuf (+25)
``diff
diff --git a/flang/include/flang/Optimizer/Builder/FIRBuilder.h
b/flang/include/flang/Optimizer/Builder/FIRBuilder.h
index e4c954159f71be..0d650f830b64e0 100644
--- a/flang/include/flang/Optimizer/Builder/FIRBuilder.h
+++ b/flang/include/flang/Optimizer/Builder/FIRBuilder.h
@@ -708,6 +708,13 @@ mlir::Value createNullBoxProc(fir::FirOpBuilder &builder,
mlir::Location loc,
/// Set internal linkage attribute on a function.
void setInternalLinkage(mlir::func::FuncOp);
+
+llvm::SmallVector
+elideExtentsAlreadyInType(mlir::Type type, mlir::ValueRange shape);
+
+llvm::SmallVector
+elideLengthsAlreadyInType(mlir::Type type, mlir::ValueRange lenParams);
+
} // namespace fir::factory
#endif // FORTRAN_OPTIMIZER_BUILDER_FIRBUILDER_H
diff --git a/flang/include/flang/Semantics/tools.h
b/flang/include/flang/Semantics/tools.h
index da10969ebc7021..c9eb5bc857ac01 100644
--- a/flang/include/flang/Semantics/tools.h
+++ b/flang/include/flang/Semantics/tools.h
@@ -222,6 +222,23 @@ inline bool HasCUDAAttr(const Symbol &sym) {
return false;
}
+inline bool NeedCUDAAlloc(const Symbol &sym) {
+ bool inDeviceSubprogram{IsCUDADeviceContext(&sym.owner())};
+ if (const auto *details{
+ sym.GetUltimate().detailsIf()}) {
+if (details->cudaDataAttr() &&
+(*details->cudaDataAttr() == common::CUDADataAttr::Device ||
+*details->cudaDataAttr() == common::CUDADataAttr::Managed ||
+*details->cudaDataAttr() == common::CUDADataAttr::Unified)) {
+ // Descriptor is allocated on host when in host context.
+ if (Fortran::semantics::IsAllocatable(sym))
+return inDeviceSubprogram;
+ return true;
+}
+ }
+ return false;
+}
+
const Scope *FindCUDADeviceContext(const Scope *);
std::optional GetCUDADataAttr(const Symbol *);
diff --git a/flang/lib/Lower/ConvertVariable.cpp
b/flang/lib/Lower/ConvertVariable.cpp
index 21db0cac11bf6a..9a17acf5b15c36 100644
--- a/flang/lib/Lower/ConvertVariable.cpp
+++ b/flang/lib/Lower/ConvertVariable.cpp
@@ -693,6 +693,22 @@ static mlir::Value
createNewLocal(Fortran::lower::AbstractConverter &converter,
if (ultimateSymbol.test(Fortran::semantics::Symbol::Flag::CrayPointee))
return builder.create(loc, fir::ReferenceType::get(ty));
+ if (Fortran::semantics::NeedCUDAAlloc(ultimateSymbol)) {
+fir::CUDADataAttributeAttr cudaAttr =
+Fortran::lower::translateSymbolCUDADataAttribute(builder.getContext(),
+ ultimateSymbol);
+llvm::SmallVector indices;
+llvm::SmallVector elidedShape =
+fir::factory::elideExtentsAlreadyInType(ty, shape);
+llvm::SmallVector elidedLenParams =
+fir::factory::elideLengthsAlreadyInType(ty, lenParams);
+auto idxTy = builder.getIndexType();
+for (mlir::Value sh : elidedShape)
+ indices.push_back(builder.createConvert(loc, idxTy, sh));
+return builder.create(loc, ty, nm, symNm, cudaAttr,
+lenParams, indices);
+ }
+
// Let the builder do all the heavy lifting.
if (!Fortran::semantics::IsProcedurePointer(ultimateSymbol))
return builder.allocateLocal(loc, ty, nm, symNm, shape, lenParams, isTarg);
@@ -927,6 +943,19 @@ static void
instantiateLocal(Fortran::lower::AbstractConverter &converter,
});
}
}
+ if (Fortran::semantics::NeedCUDAAlloc(var.getSymbol())) {
+auto *builder = &converter.getFirOpBuilder();
+mlir::Location loc = converter.getCurrentLocation();
+fir::ExtendedValue exv =
+converter.getSymbolExtendedValue(var.getSymbol(), &symMap);
+auto *sym = &var.getSymbol();
+converter.getFctCtx().attachCleanup([builder, loc, exv, sym]() {
+ fir::CUDADataAttributeAttr cudaAttr =
+ Fortran::lower::translateSymbolCUDADataAttribute(
+ builder->getContext(), *sym);
+ builder->create(loc, fir::getBase(exv), cudaAttr);
+});
+ }
}
//======//
diff --git a/flang/lib/Optimizer/Builder/FIRBuilder.cpp
b/flang/lib/Optimizer/Builder/FIRBuilder.cpp
index a0fbae5b614cc7..a813b646087d73 100644
--- a/flang/lib/Optimizer/Builder/FIRBuilder.cpp
+++ b/flang/lib/Optimizer/Builder/F
[llvm-branch-commits] [libcxx] [flang] [lldb] [lld] [clang] [libc] [clang-tools-extra] [compiler-rt] [llvm] [msan] Unwind stack before fatal reports (PR #77168)
https://github.com/vitalybuka edited https://github.com/llvm/llvm-project/pull/77168 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [compiler-rt] [test][hwasan] Test function name in summaries (PR #77391)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/77391
>From 23bf2a98a428fa6a866f64e2499793734b1a97a9 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Mon, 8 Jan 2024 15:27:19 -0800
Subject: [PATCH] undo some failing cases
Created using spr 1.3.4
---
compiler-rt/test/hwasan/TestCases/halt-on-error.cpp | 6 +++---
compiler-rt/test/hwasan/TestCases/report-unmapped.cpp | 2 +-
compiler-rt/test/hwasan/TestCases/use-after-free.c| 2 +-
3 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/compiler-rt/test/hwasan/TestCases/halt-on-error.cpp
b/compiler-rt/test/hwasan/TestCases/halt-on-error.cpp
index f587ee0fcabd30..7666baf9de79de 100644
--- a/compiler-rt/test/hwasan/TestCases/halt-on-error.cpp
+++ b/compiler-rt/test/hwasan/TestCases/halt-on-error.cpp
@@ -26,15 +26,15 @@ int main() {
// COMMON: READ of size 4 at
// When instrumenting with callbacks, main is actually #1, and #0 is
__hwasan_load4.
// COMMON: #{{.*}} in main {{.*}}halt-on-error.cpp:[[@LINE-3]]
- // COMMON: SUMMARY: HWAddressSanitizer: tag-mismatch {{.*}} in SigTrap
+ // COMMON: SUMMARY: HWAddressSanitizer: tag-mismatch
// RECOVER: READ of size 1 at
// RECOVER: #{{.*}} in main {{.*}}halt-on-error.cpp:[[@LINE-7]]
- // RECOVER: SUMMARY: HWAddressSanitizer: tag-mismatch {{.*}} in SigTrap
+ // RECOVER: SUMMARY: HWAddressSanitizer: tag-mismatch
// RECOVER: READ of size 1 at
// RECOVER: #{{.*}} in main {{.*}}halt-on-error.cpp:[[@LINE-11]]
- // RECOVER: SUMMARY: HWAddressSanitizer: tag-mismatch {{.*}} in SigTrap
+ // RECOVER: SUMMARY: HWAddressSanitizer: tag-mismatch
// COMMON-NOT: tag-mismatch
}
diff --git a/compiler-rt/test/hwasan/TestCases/report-unmapped.cpp
b/compiler-rt/test/hwasan/TestCases/report-unmapped.cpp
index 1802e920f390a2..935cd71c8c9c96 100644
--- a/compiler-rt/test/hwasan/TestCases/report-unmapped.cpp
+++ b/compiler-rt/test/hwasan/TestCases/report-unmapped.cpp
@@ -36,4 +36,4 @@ int main(int argc, char **argv) {
// CHECK: Tags for short granules around
// Check that report is complete.
-// CHECK: SUMMARY: HWAddressSanitizer: tag-mismatch {{.*}} in SigTrap
+// CHECK: SUMMARY: HWAddressSanitizer: tag-mismatch
diff --git a/compiler-rt/test/hwasan/TestCases/use-after-free.c
b/compiler-rt/test/hwasan/TestCases/use-after-free.c
index e0afb767b05937..b3eed88600726c 100644
--- a/compiler-rt/test/hwasan/TestCases/use-after-free.c
+++ b/compiler-rt/test/hwasan/TestCases/use-after-free.c
@@ -38,6 +38,6 @@ int main() {
// CHECK: #1 {{.*}} in main {{.*}}use-after-free.c:[[@LINE-24]]
// CHECK: Memory tags around the buggy address (one tag corresponds to 16
bytes):
// CHECK: =>{{.*}}[[MEM_TAG]]
- // CHECK: SUMMARY: HWAddressSanitizer: tag-mismatch {{.*}} in SigTrap
+ // CHECK: SUMMARY: HWAddressSanitizer: tag-mismatch
return r;
}
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [compiler-rt] [msan] Unwind stack before fatal reports (PR #77168)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/77168
>From a127373cf1ac1676ce17ce8dca909d0c3bce9d18 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Mon, 8 Jan 2024 11:45:37 -0800
Subject: [PATCH 1/2] [NFC][msan] Switch allocator interface to use
BufferedStackTrace
We will need it to unwind for fatal errors.
Pull Request: https://github.com/llvm/llvm-project/pull/77363
---
compiler-rt/lib/msan/msan.h | 23 ++--
compiler-rt/lib/msan/msan_allocator.cpp | 29 +
2 files changed, 27 insertions(+), 25 deletions(-)
diff --git a/compiler-rt/lib/msan/msan.h b/compiler-rt/lib/msan/msan.h
index 25fa2212bdadd3..753e6b260734f9 100644
--- a/compiler-rt/lib/msan/msan.h
+++ b/compiler-rt/lib/msan/msan.h
@@ -255,18 +255,19 @@ char *GetProcSelfMaps();
void InitializeInterceptors();
void MsanAllocatorInit();
-void MsanDeallocate(StackTrace *stack, void *ptr);
-
-void *msan_malloc(uptr size, StackTrace *stack);
-void *msan_calloc(uptr nmemb, uptr size, StackTrace *stack);
-void *msan_realloc(void *ptr, uptr size, StackTrace *stack);
-void *msan_reallocarray(void *ptr, uptr nmemb, uptr size, StackTrace *stack);
-void *msan_valloc(uptr size, StackTrace *stack);
-void *msan_pvalloc(uptr size, StackTrace *stack);
-void *msan_aligned_alloc(uptr alignment, uptr size, StackTrace *stack);
-void *msan_memalign(uptr alignment, uptr size, StackTrace *stack);
+void MsanDeallocate(BufferedStackTrace *stack, void *ptr);
+
+void *msan_malloc(uptr size, BufferedStackTrace *stack);
+void *msan_calloc(uptr nmemb, uptr size, BufferedStackTrace *stack);
+void *msan_realloc(void *ptr, uptr size, BufferedStackTrace *stack);
+void *msan_reallocarray(void *ptr, uptr nmemb, uptr size,
+BufferedStackTrace *stack);
+void *msan_valloc(uptr size, BufferedStackTrace *stack);
+void *msan_pvalloc(uptr size, BufferedStackTrace *stack);
+void *msan_aligned_alloc(uptr alignment, uptr size, BufferedStackTrace *stack);
+void *msan_memalign(uptr alignment, uptr size, BufferedStackTrace *stack);
int msan_posix_memalign(void **memptr, uptr alignment, uptr size,
-StackTrace *stack);
+BufferedStackTrace *stack);
void InstallTrapHandler();
void InstallAtExitHandler();
diff --git a/compiler-rt/lib/msan/msan_allocator.cpp
b/compiler-rt/lib/msan/msan_allocator.cpp
index 72a7f980d39fb0..987c894c79d45e 100644
--- a/compiler-rt/lib/msan/msan_allocator.cpp
+++ b/compiler-rt/lib/msan/msan_allocator.cpp
@@ -178,7 +178,7 @@ void MsanThreadLocalMallocStorage::CommitBack() {
allocator.DestroyCache(GetAllocatorCache(this));
}
-static void *MsanAllocate(StackTrace *stack, uptr size, uptr alignment,
+static void *MsanAllocate(BufferedStackTrace *stack, uptr size, uptr alignment,
bool zeroise) {
if (size > max_malloc_size) {
if (AllocatorMayReturnNull()) {
@@ -229,7 +229,7 @@ static void *MsanAllocate(StackTrace *stack, uptr size,
uptr alignment,
return allocated;
}
-void MsanDeallocate(StackTrace *stack, void *p) {
+void MsanDeallocate(BufferedStackTrace *stack, void *p) {
CHECK(p);
UnpoisonParam(1);
RunFreeHooks(p);
@@ -259,8 +259,8 @@ void MsanDeallocate(StackTrace *stack, void *p) {
}
}
-static void *MsanReallocate(StackTrace *stack, void *old_p, uptr new_size,
-uptr alignment) {
+static void *MsanReallocate(BufferedStackTrace *stack, void *old_p,
+uptr new_size, uptr alignment) {
Metadata *meta = reinterpret_cast(allocator.GetMetaData(old_p));
uptr old_size = meta->requested_size;
uptr actually_allocated_size = allocator.GetActuallyAllocatedSize(old_p);
@@ -284,7 +284,7 @@ static void *MsanReallocate(StackTrace *stack, void *old_p,
uptr new_size,
return new_p;
}
-static void *MsanCalloc(StackTrace *stack, uptr nmemb, uptr size) {
+static void *MsanCalloc(BufferedStackTrace *stack, uptr nmemb, uptr size) {
if (UNLIKELY(CheckForCallocOverflow(size, nmemb))) {
if (AllocatorMayReturnNull())
return nullptr;
@@ -320,15 +320,15 @@ static uptr AllocationSizeFast(const void *p) {
return reinterpret_cast(allocator.GetMetaData(p))->requested_size;
}
-void *msan_malloc(uptr size, StackTrace *stack) {
+void *msan_malloc(uptr size, BufferedStackTrace *stack) {
return SetErrnoOnNull(MsanAllocate(stack, size, sizeof(u64), false));
}
-void *msan_calloc(uptr nmemb, uptr size, StackTrace *stack) {
+void *msan_calloc(uptr nmemb, uptr size, BufferedStackTrace *stack) {
return SetErrnoOnNull(MsanCalloc(stack, nmemb, size));
}
-void *msan_realloc(void *ptr, uptr size, StackTrace *stack) {
+void *msan_realloc(void *ptr, uptr size, BufferedStackTrace *stack) {
if (!ptr)
return SetErrnoOnNull(MsanAllocate(stack, size, sizeof(u64), false));
if (size == 0) {
@@ -338,7 +338,8 @@ void *msan_realloc(void *ptr, uptr size, StackTrace *stack)
{
return SetErrno
[llvm-branch-commits] [compiler-rt] [test][hwasan] Test function name in summaries (PR #77391)
https://github.com/vitalybuka closed https://github.com/llvm/llvm-project/pull/77391 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [llvm] [InstCombiner] Remove unused `llvm.experimental.hot()` (PR #84851)
https://github.com/vitalybuka ready_for_review https://github.com/llvm/llvm-project/pull/84851 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [llvm] [InstCombiner] Remove unused `llvm.experimental.hot()` (PR #84851)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/84851
>From c15aa098b7a8db614f574d47c220852674d4681d Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Mon, 11 Mar 2024 17:13:07 -0700
Subject: [PATCH] rebase
Created using spr 1.3.4
---
clang/lib/CodeGen/BackendUtil.cpp | 6 +-
clang/lib/CodeGen/CGExpr.cpp | 15 +---
clang/test/CodeGen/remote-traps.c | 23 +
.../Instrumentation/RemoveTrapsPass.cpp | 26 +++---
.../Transforms/RemoveTraps/remove-traps.ll| 89 ++-
5 files changed, 43 insertions(+), 116 deletions(-)
diff --git a/clang/lib/CodeGen/BackendUtil.cpp
b/clang/lib/CodeGen/BackendUtil.cpp
index 7e53469a48d42c..82b30b8d815629 100644
--- a/clang/lib/CodeGen/BackendUtil.cpp
+++ b/clang/lib/CodeGen/BackendUtil.cpp
@@ -100,9 +100,9 @@ using namespace llvm;
namespace llvm {
extern cl::opt PrintPipelinePasses;
-static cl::opt ClRemoveTraps("clang-remove-traps", cl::Optional,
- cl::desc("Insert remove-traps pass."),
- cl::init(false));
+cl::opt ClRemoveTraps("clang-remove-traps", cl::Optional,
+cl::desc("Insert remove-traps pass."),
+cl::init(false));
// Experiment to move sanitizers earlier.
static cl::opt ClSanitizeOnOptimizerEarlyEP(
diff --git a/clang/lib/CodeGen/CGExpr.cpp b/clang/lib/CodeGen/CGExpr.cpp
index 3a27622f165995..59a7fe8925001c 100644
--- a/clang/lib/CodeGen/CGExpr.cpp
+++ b/clang/lib/CodeGen/CGExpr.cpp
@@ -56,14 +56,7 @@ using namespace CodeGen;
// Experiment to make sanitizers easier to debug
static llvm::cl::opt ClSanitizeDebugDeoptimization(
"ubsan-unique-traps", llvm::cl::Optional,
-llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check."),
-llvm::cl::init(false));
-
-// TODO: Introduce frontend options to enabled per sanitizers, similar to
-// `fsanitize-trap`.
-static llvm::cl::opt ClSanitizeExpHot(
-"ubsan-exp-hot", llvm::cl::Optional,
-llvm::cl::desc("Pass UBSAN checks if `llvm.experimental.hot()` is true."),
+llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check"),
llvm::cl::init(false));
//======//
@@ -3812,12 +3805,6 @@ void CodeGenFunction::EmitTrapCheck(llvm::Value *Checked,
SanitizerHandler CheckHandlerID) {
llvm::BasicBlock *Cont = createBasicBlock("cont");
- if (ClSanitizeExpHot) {
-Checked =
-Builder.CreateOr(Checked, Builder.CreateCall(CGM.getIntrinsic(
- llvm::Intrinsic::experimental_hot)));
- }
-
// If we're optimizing, collapse all calls to trap down to just one per
// check-type per function to save on code size.
if ((int)TrapBBs.size() <= CheckHandlerID)
diff --git a/clang/test/CodeGen/remote-traps.c
b/clang/test/CodeGen/remote-traps.c
index 16e4ebecb5c326..6751afb96d25f2 100644
--- a/clang/test/CodeGen/remote-traps.c
+++ b/clang/test/CodeGen/remote-traps.c
@@ -1,7 +1,5 @@
-// RUN: %clang_cc1 -O1 %s -o - -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -ubsan-exp-hot | FileCheck %s
-// RUN: %clang_cc1 -O1 %s -o - -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -ubsan-exp-hot -mllvm
-clang-remove-traps -mllvm -remove-traps-random-rate=1 %s -o - | FileCheck %s
--check-prefixes=REMOVE
-
-#include
+// RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow %s -o - | FileCheck %s
+// RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -clang-remove-traps -mllvm
-remove-traps-random-rate=1 %s -o - | FileCheck %s --implicit-check-not="call
void @llvm.ubsantrap" --check-prefixes=REMOVE
int test(int x) {
return x + 123;
@@ -14,19 +12,4 @@ int test(int x) {
// CHECK-NEXT: unreachable
// REMOVE-LABEL: define {{.*}}i32 @test(
-// REMOVE: add i32 %x, 123
-// REMOVE-NEXT: ret i32
-
-
-bool experimental_hot() __asm("llvm.experimental.hot");
-
-bool test_asm() {
- return experimental_hot();
-}
-
-// CHECK-LABEL: define {{.*}}i1 @test_asm(
-// CHECK: [[R:%.*]] = tail call zeroext i1 @llvm.experimental.hot()
-// CHECK: ret i1 [[R]]
-
-// REMOVE-LABEL: define {{.*}}i1 @test_asm(
-// REMOVE: ret i1 true
+// REMOVE: call { i32, i1 } @llvm.sadd.with.overflow.i32(
diff --git a/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
b/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
index fa4716f2e7a403..d87f7482a21d25 100644
--- a/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
+++ b/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
@@ -11,7 +11,6 @@
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/Analysis/ProfileSummaryInfo.h"
-#include "llvm/
[llvm-branch-commits] [clang] [llvm] [InstCombiner] Remove unused `llvm.experimental.hot()` (PR #84851)
@@ -56,7 +56,14 @@ using namespace CodeGen;
// Experiment to make sanitizers easier to debug
static llvm::cl::opt ClSanitizeDebugDeoptimization(
"ubsan-unique-traps", llvm::cl::Optional,
-llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check"),
+llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check."),
+llvm::cl::init(false));
+
+// TODO: Introduce frontend options to enabled per sanitizers, similar to
+// `fsanitize-trap`.
+static llvm::cl::opt ClSanitizeExpHot(
+"ubsan-exp-hot", llvm::cl::Optional,
+llvm::cl::desc("Pass UBSAN checks if `llvm.experimental.hot()` is true."),
vitalybuka wrote:
I guess #84858
https://github.com/llvm/llvm-project/pull/84851
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [llvm] [InstCombiner] Remove unused `llvm.experimental.hot()` (PR #84851)
https://github.com/kstoimenov approved this pull request. https://github.com/llvm/llvm-project/pull/84851 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [IR] Introduce `llvm.experimental.hot()` (PR #84850)
https://github.com/vitalybuka updated https://github.com/llvm/llvm-project/pull/84850 >From 679055efe79a97c84c8ef675d7eb450c5df58a51 Mon Sep 17 00:00:00 2001 From: Vitaly Buka Date: Tue, 12 Mar 2024 12:01:01 -0700 Subject: [PATCH] Update llvm/docs/LangRef.rst Co-authored-by: Nikita Popov --- llvm/docs/LangRef.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/llvm/docs/LangRef.rst b/llvm/docs/LangRef.rst index 36f4c964ee296c..a707ab721a2b2a 100644 --- a/llvm/docs/LangRef.rst +++ b/llvm/docs/LangRef.rst @@ -27656,7 +27656,7 @@ This intrinsic returns true iff it's known that containing basic block is hot in profile. When used with profile based optimization allows to change program behaviour -deppending on the code hotness. +depending on the code hotness. Arguments: "" ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [IR] Introduce `llvm.experimental.hot()` (PR #84850)
@@ -27639,6 +27639,54 @@ constant `true`. However it is always correct to replace it with any other `i1` value. Any pass can freely do it if it can benefit from non-default lowering. +'``llvm.experimental.hot``' Intrinsic +^ + +Syntax: +""" + +:: + + declare i1 @llvm.experimental.hot() + +Overview: +" + +This intrinsic returns true iff it's known that containing basic block is hot in +profile. + +When used with profile based optimization allows to change program behaviour +deppending on the code hotness. + +Arguments: +"" + +None. + +Semantics: +"" + +The intrinsic ``@llvm.experimental.hot()`` returns either `true` or `false`, +deppending on profile used. Expresion is evaluated as `true` iff profile and +summary are availible and profile counter for the block reach hotness threshold. +For each evaluation of a call to this intrinsic, the program must be valid and +correct both if it returns `true` and if it returns `false`. + +When used in a branch condition, it allows us to choose between +two alternative correct solutions for the same problem, like +in example below: + +.. code-block:: text + +%cond = call i1 @llvm.experimental.hot() +br i1 %cond, label %fast_path, label %slow_path + + label %fast_path: vitalybuka wrote: copypasted mistake from @llvm.experimental.widenable.condition :) https://github.com/llvm/llvm-project/pull/84850 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [llvm] [InstCombiner] Remove unused `llvm.experimental.hot()` (PR #84851)
@@ -2827,6 +2827,14 @@ Instruction *InstCombinerImpl::visitCallInst(CallInst
&CI) {
}))
return nullptr;
break;
+ case Intrinsic::experimental_hot: {
+// The intrinsic declaration includes sideeffects to avoid it moved. This
+// prevents removing even if the intrinsic is unused. We should remove
+// unused ones to enabled other optimizations.
+if (CI.use_empty())
+ return eraseInstFromFunction(CI);
nikic wrote:
This should be handled inside wouldInstructionBeTriviallyDead.
https://github.com/llvm/llvm-project/pull/84851
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [IR] Introduce `llvm.experimental.hot()` (PR #84850)
https://github.com/vitalybuka edited https://github.com/llvm/llvm-project/pull/84850 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [IR] Introduce `llvm.experimental.hot()` (PR #84850)
vitalybuka wrote: > Please submit an RFC on discourse for this change. https://discourse.llvm.org/t/rfc-add-llvm-experimental-hot-intrinsic-or-llvm-hot/77641 https://github.com/llvm/llvm-project/pull/84850 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [IR] Introduce `llvm.experimental.hot()` (PR #84850)
@@ -1722,6 +1722,11 @@ def int_debugtrap : Intrinsic<[]>, def int_ubsantrap : Intrinsic<[], [llvm_i8_ty], [IntrNoReturn, IntrCold, ImmArg>]>; +// Return true if profile counter for containing block is hot. +def int_experimental_hot : Intrinsic<[llvm_i1_ty], [], + [IntrInaccessibleMemOnly, IntrWriteMem, vitalybuka wrote: No, we have no more precise attribute. E.g. int_sideeffect is declared this way as well. There are a few comments for similar cases around, I'll add one here to. We need to lock intrinsic in place, without it, it will be moved to the entry block and value shared between users. So it will work as function level, vs block level. With ThinLTO it should not be a problem, in postlink we rerun simplification and optimization with intrinsic lowered to true or false. I tried to remove this attribute and make it function level, but can't measure perf difference. Alternative is to have intrinsic without sideeffects but with dependency like this: ``` %condition = call i1 @llvm.experimenta.hot(i1 %ubsan_condition) ``` This way we can lock it in place. But this will not help us with assert() with sideeffects as we discussed before. So I tried both with UBSAN, ThinLTO and plain, and ``` call i1 @llvm.experimenta.hot() ``` is consistently, but close to noise, faster then: ``` call i1 @llvm.experimenta.hot(i1 %ubsan_condition) ``` I believe because `or i1 %ubsan, @llvm.experimenta.hot()` is more predictable for early transformation passes than `@llvm.experimenta.hot(i1 %ubsan_condition)`. Both intrinsic approaches faster then removing traps. https://github.com/llvm/llvm-project/pull/84850 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [IR] Introduce `llvm.experimental.hot()` (PR #84850)
https://github.com/vitalybuka updated https://github.com/llvm/llvm-project/pull/84850 >From 679055efe79a97c84c8ef675d7eb450c5df58a51 Mon Sep 17 00:00:00 2001 From: Vitaly Buka Date: Tue, 12 Mar 2024 12:01:01 -0700 Subject: [PATCH] Update llvm/docs/LangRef.rst Co-authored-by: Nikita Popov --- llvm/docs/LangRef.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/llvm/docs/LangRef.rst b/llvm/docs/LangRef.rst index 36f4c964ee296c..a707ab721a2b2a 100644 --- a/llvm/docs/LangRef.rst +++ b/llvm/docs/LangRef.rst @@ -27656,7 +27656,7 @@ This intrinsic returns true iff it's known that containing basic block is hot in profile. When used with profile based optimization allows to change program behaviour -deppending on the code hotness. +depending on the code hotness. Arguments: "" ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [IR] Introduce `llvm.experimental.hot()` (PR #84850)
https://github.com/vitalybuka updated https://github.com/llvm/llvm-project/pull/84850 >From 679055efe79a97c84c8ef675d7eb450c5df58a51 Mon Sep 17 00:00:00 2001 From: Vitaly Buka Date: Tue, 12 Mar 2024 12:01:01 -0700 Subject: [PATCH] Update llvm/docs/LangRef.rst Co-authored-by: Nikita Popov --- llvm/docs/LangRef.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/llvm/docs/LangRef.rst b/llvm/docs/LangRef.rst index 36f4c964ee296c..a707ab721a2b2a 100644 --- a/llvm/docs/LangRef.rst +++ b/llvm/docs/LangRef.rst @@ -27656,7 +27656,7 @@ This intrinsic returns true iff it's known that containing basic block is hot in profile. When used with profile based optimization allows to change program behaviour -deppending on the code hotness. +depending on the code hotness. Arguments: "" ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [llvm] [InstCombiner] Remove unused `llvm.experimental.hot()` (PR #84851)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/84851
>From c15aa098b7a8db614f574d47c220852674d4681d Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Mon, 11 Mar 2024 17:13:07 -0700
Subject: [PATCH] rebase
Created using spr 1.3.4
---
clang/lib/CodeGen/BackendUtil.cpp | 6 +-
clang/lib/CodeGen/CGExpr.cpp | 15 +---
clang/test/CodeGen/remote-traps.c | 23 +
.../Instrumentation/RemoveTrapsPass.cpp | 26 +++---
.../Transforms/RemoveTraps/remove-traps.ll| 89 ++-
5 files changed, 43 insertions(+), 116 deletions(-)
diff --git a/clang/lib/CodeGen/BackendUtil.cpp
b/clang/lib/CodeGen/BackendUtil.cpp
index 7e53469a48d42c..82b30b8d815629 100644
--- a/clang/lib/CodeGen/BackendUtil.cpp
+++ b/clang/lib/CodeGen/BackendUtil.cpp
@@ -100,9 +100,9 @@ using namespace llvm;
namespace llvm {
extern cl::opt PrintPipelinePasses;
-static cl::opt ClRemoveTraps("clang-remove-traps", cl::Optional,
- cl::desc("Insert remove-traps pass."),
- cl::init(false));
+cl::opt ClRemoveTraps("clang-remove-traps", cl::Optional,
+cl::desc("Insert remove-traps pass."),
+cl::init(false));
// Experiment to move sanitizers earlier.
static cl::opt ClSanitizeOnOptimizerEarlyEP(
diff --git a/clang/lib/CodeGen/CGExpr.cpp b/clang/lib/CodeGen/CGExpr.cpp
index 3a27622f165995..59a7fe8925001c 100644
--- a/clang/lib/CodeGen/CGExpr.cpp
+++ b/clang/lib/CodeGen/CGExpr.cpp
@@ -56,14 +56,7 @@ using namespace CodeGen;
// Experiment to make sanitizers easier to debug
static llvm::cl::opt ClSanitizeDebugDeoptimization(
"ubsan-unique-traps", llvm::cl::Optional,
-llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check."),
-llvm::cl::init(false));
-
-// TODO: Introduce frontend options to enabled per sanitizers, similar to
-// `fsanitize-trap`.
-static llvm::cl::opt ClSanitizeExpHot(
-"ubsan-exp-hot", llvm::cl::Optional,
-llvm::cl::desc("Pass UBSAN checks if `llvm.experimental.hot()` is true."),
+llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check"),
llvm::cl::init(false));
//======//
@@ -3812,12 +3805,6 @@ void CodeGenFunction::EmitTrapCheck(llvm::Value *Checked,
SanitizerHandler CheckHandlerID) {
llvm::BasicBlock *Cont = createBasicBlock("cont");
- if (ClSanitizeExpHot) {
-Checked =
-Builder.CreateOr(Checked, Builder.CreateCall(CGM.getIntrinsic(
- llvm::Intrinsic::experimental_hot)));
- }
-
// If we're optimizing, collapse all calls to trap down to just one per
// check-type per function to save on code size.
if ((int)TrapBBs.size() <= CheckHandlerID)
diff --git a/clang/test/CodeGen/remote-traps.c
b/clang/test/CodeGen/remote-traps.c
index 16e4ebecb5c326..6751afb96d25f2 100644
--- a/clang/test/CodeGen/remote-traps.c
+++ b/clang/test/CodeGen/remote-traps.c
@@ -1,7 +1,5 @@
-// RUN: %clang_cc1 -O1 %s -o - -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -ubsan-exp-hot | FileCheck %s
-// RUN: %clang_cc1 -O1 %s -o - -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -ubsan-exp-hot -mllvm
-clang-remove-traps -mllvm -remove-traps-random-rate=1 %s -o - | FileCheck %s
--check-prefixes=REMOVE
-
-#include
+// RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow %s -o - | FileCheck %s
+// RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -clang-remove-traps -mllvm
-remove-traps-random-rate=1 %s -o - | FileCheck %s --implicit-check-not="call
void @llvm.ubsantrap" --check-prefixes=REMOVE
int test(int x) {
return x + 123;
@@ -14,19 +12,4 @@ int test(int x) {
// CHECK-NEXT: unreachable
// REMOVE-LABEL: define {{.*}}i32 @test(
-// REMOVE: add i32 %x, 123
-// REMOVE-NEXT: ret i32
-
-
-bool experimental_hot() __asm("llvm.experimental.hot");
-
-bool test_asm() {
- return experimental_hot();
-}
-
-// CHECK-LABEL: define {{.*}}i1 @test_asm(
-// CHECK: [[R:%.*]] = tail call zeroext i1 @llvm.experimental.hot()
-// CHECK: ret i1 [[R]]
-
-// REMOVE-LABEL: define {{.*}}i1 @test_asm(
-// REMOVE: ret i1 true
+// REMOVE: call { i32, i1 } @llvm.sadd.with.overflow.i32(
diff --git a/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
b/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
index fa4716f2e7a403..d87f7482a21d25 100644
--- a/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
+++ b/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
@@ -11,7 +11,6 @@
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/Analysis/ProfileSummaryInfo.h"
-#include "llvm/
[llvm-branch-commits] [llvm] a6319c3 - Revert "Make two texts static in `ReplayInlineAdvisor` (#79489)"
Author: Vitaly Buka
Date: 2024-02-21T16:05:56-08:00
New Revision: a6319c3d1c4a7083b5f148b3ca33cb2757ed469c
URL:
https://github.com/llvm/llvm-project/commit/a6319c3d1c4a7083b5f148b3ca33cb2757ed469c
DIFF:
https://github.com/llvm/llvm-project/commit/a6319c3d1c4a7083b5f148b3ca33cb2757ed469c.diff
LOG: Revert "Make two texts static in `ReplayInlineAdvisor` (#79489)"
This reverts commit 67f9c35ffee1aea3fa8a18b57a7b27400a03435a.
Added:
Modified:
llvm/lib/Analysis/ReplayInlineAdvisor.cpp
Removed:
diff --git a/llvm/lib/Analysis/ReplayInlineAdvisor.cpp
b/llvm/lib/Analysis/ReplayInlineAdvisor.cpp
index 0814483db343ce..2ca02eb1741712 100644
--- a/llvm/lib/Analysis/ReplayInlineAdvisor.cpp
+++ b/llvm/lib/Analysis/ReplayInlineAdvisor.cpp
@@ -43,8 +43,8 @@ ReplayInlineAdvisor::ReplayInlineAdvisor(
// main:3:1.1;
// We use the callsite string after `at callsite` to replay inlining.
line_iterator LineIt(*BufferOrErr.get(), /*SkipBlanks=*/true);
- static const std::string PositiveRemark = "' inlined into '";
- static const std::string NegativeRemark = "' will not be inlined into '";
+ const std::string PositiveRemark = "' inlined into '";
+ const std::string NegativeRemark = "' will not be inlined into '";
for (; !LineIt.is_at_eof(); ++LineIt) {
StringRef Line = *LineIt;
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [llvm] [InstCombiner] Remove trivially dead `llvm.allow.{runtime, ubsan}.check()` (PR #84851)
https://github.com/vitalybuka edited https://github.com/llvm/llvm-project/pull/84851 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [llvm] [InstCombiner] Remove unused `llvm.experimental.hot()` (PR #84851)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/84851
>From c15aa098b7a8db614f574d47c220852674d4681d Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Mon, 11 Mar 2024 17:13:07 -0700
Subject: [PATCH] rebase
Created using spr 1.3.4
---
clang/lib/CodeGen/BackendUtil.cpp | 6 +-
clang/lib/CodeGen/CGExpr.cpp | 15 +---
clang/test/CodeGen/remote-traps.c | 23 +
.../Instrumentation/RemoveTrapsPass.cpp | 26 +++---
.../Transforms/RemoveTraps/remove-traps.ll| 89 ++-
5 files changed, 43 insertions(+), 116 deletions(-)
diff --git a/clang/lib/CodeGen/BackendUtil.cpp
b/clang/lib/CodeGen/BackendUtil.cpp
index 7e53469a48d42c..82b30b8d815629 100644
--- a/clang/lib/CodeGen/BackendUtil.cpp
+++ b/clang/lib/CodeGen/BackendUtil.cpp
@@ -100,9 +100,9 @@ using namespace llvm;
namespace llvm {
extern cl::opt PrintPipelinePasses;
-static cl::opt ClRemoveTraps("clang-remove-traps", cl::Optional,
- cl::desc("Insert remove-traps pass."),
- cl::init(false));
+cl::opt ClRemoveTraps("clang-remove-traps", cl::Optional,
+cl::desc("Insert remove-traps pass."),
+cl::init(false));
// Experiment to move sanitizers earlier.
static cl::opt ClSanitizeOnOptimizerEarlyEP(
diff --git a/clang/lib/CodeGen/CGExpr.cpp b/clang/lib/CodeGen/CGExpr.cpp
index 3a27622f165995..59a7fe8925001c 100644
--- a/clang/lib/CodeGen/CGExpr.cpp
+++ b/clang/lib/CodeGen/CGExpr.cpp
@@ -56,14 +56,7 @@ using namespace CodeGen;
// Experiment to make sanitizers easier to debug
static llvm::cl::opt ClSanitizeDebugDeoptimization(
"ubsan-unique-traps", llvm::cl::Optional,
-llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check."),
-llvm::cl::init(false));
-
-// TODO: Introduce frontend options to enabled per sanitizers, similar to
-// `fsanitize-trap`.
-static llvm::cl::opt ClSanitizeExpHot(
-"ubsan-exp-hot", llvm::cl::Optional,
-llvm::cl::desc("Pass UBSAN checks if `llvm.experimental.hot()` is true."),
+llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check"),
llvm::cl::init(false));
//======//
@@ -3812,12 +3805,6 @@ void CodeGenFunction::EmitTrapCheck(llvm::Value *Checked,
SanitizerHandler CheckHandlerID) {
llvm::BasicBlock *Cont = createBasicBlock("cont");
- if (ClSanitizeExpHot) {
-Checked =
-Builder.CreateOr(Checked, Builder.CreateCall(CGM.getIntrinsic(
- llvm::Intrinsic::experimental_hot)));
- }
-
// If we're optimizing, collapse all calls to trap down to just one per
// check-type per function to save on code size.
if ((int)TrapBBs.size() <= CheckHandlerID)
diff --git a/clang/test/CodeGen/remote-traps.c
b/clang/test/CodeGen/remote-traps.c
index 16e4ebecb5c326..6751afb96d25f2 100644
--- a/clang/test/CodeGen/remote-traps.c
+++ b/clang/test/CodeGen/remote-traps.c
@@ -1,7 +1,5 @@
-// RUN: %clang_cc1 -O1 %s -o - -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -ubsan-exp-hot | FileCheck %s
-// RUN: %clang_cc1 -O1 %s -o - -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -ubsan-exp-hot -mllvm
-clang-remove-traps -mllvm -remove-traps-random-rate=1 %s -o - | FileCheck %s
--check-prefixes=REMOVE
-
-#include
+// RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow %s -o - | FileCheck %s
+// RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -clang-remove-traps -mllvm
-remove-traps-random-rate=1 %s -o - | FileCheck %s --implicit-check-not="call
void @llvm.ubsantrap" --check-prefixes=REMOVE
int test(int x) {
return x + 123;
@@ -14,19 +12,4 @@ int test(int x) {
// CHECK-NEXT: unreachable
// REMOVE-LABEL: define {{.*}}i32 @test(
-// REMOVE: add i32 %x, 123
-// REMOVE-NEXT: ret i32
-
-
-bool experimental_hot() __asm("llvm.experimental.hot");
-
-bool test_asm() {
- return experimental_hot();
-}
-
-// CHECK-LABEL: define {{.*}}i1 @test_asm(
-// CHECK: [[R:%.*]] = tail call zeroext i1 @llvm.experimental.hot()
-// CHECK: ret i1 [[R]]
-
-// REMOVE-LABEL: define {{.*}}i1 @test_asm(
-// REMOVE: ret i1 true
+// REMOVE: call { i32, i1 } @llvm.sadd.with.overflow.i32(
diff --git a/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
b/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
index fa4716f2e7a403..d87f7482a21d25 100644
--- a/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
+++ b/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
@@ -11,7 +11,6 @@
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/Analysis/ProfileSummaryInfo.h"
-#include "llvm/
[llvm-branch-commits] [clang] [llvm] [InstCombiner] Remove trivially dead `llvm.allow.{runtime, ubsan}.check()` (PR #84851)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/84851
>From c15aa098b7a8db614f574d47c220852674d4681d Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Mon, 11 Mar 2024 17:13:07 -0700
Subject: [PATCH] rebase
Created using spr 1.3.4
---
clang/lib/CodeGen/BackendUtil.cpp | 6 +-
clang/lib/CodeGen/CGExpr.cpp | 15 +---
clang/test/CodeGen/remote-traps.c | 23 +
.../Instrumentation/RemoveTrapsPass.cpp | 26 +++---
.../Transforms/RemoveTraps/remove-traps.ll| 89 ++-
5 files changed, 43 insertions(+), 116 deletions(-)
diff --git a/clang/lib/CodeGen/BackendUtil.cpp
b/clang/lib/CodeGen/BackendUtil.cpp
index 7e53469a48d42c..82b30b8d815629 100644
--- a/clang/lib/CodeGen/BackendUtil.cpp
+++ b/clang/lib/CodeGen/BackendUtil.cpp
@@ -100,9 +100,9 @@ using namespace llvm;
namespace llvm {
extern cl::opt PrintPipelinePasses;
-static cl::opt ClRemoveTraps("clang-remove-traps", cl::Optional,
- cl::desc("Insert remove-traps pass."),
- cl::init(false));
+cl::opt ClRemoveTraps("clang-remove-traps", cl::Optional,
+cl::desc("Insert remove-traps pass."),
+cl::init(false));
// Experiment to move sanitizers earlier.
static cl::opt ClSanitizeOnOptimizerEarlyEP(
diff --git a/clang/lib/CodeGen/CGExpr.cpp b/clang/lib/CodeGen/CGExpr.cpp
index 3a27622f165995..59a7fe8925001c 100644
--- a/clang/lib/CodeGen/CGExpr.cpp
+++ b/clang/lib/CodeGen/CGExpr.cpp
@@ -56,14 +56,7 @@ using namespace CodeGen;
// Experiment to make sanitizers easier to debug
static llvm::cl::opt ClSanitizeDebugDeoptimization(
"ubsan-unique-traps", llvm::cl::Optional,
-llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check."),
-llvm::cl::init(false));
-
-// TODO: Introduce frontend options to enabled per sanitizers, similar to
-// `fsanitize-trap`.
-static llvm::cl::opt ClSanitizeExpHot(
-"ubsan-exp-hot", llvm::cl::Optional,
-llvm::cl::desc("Pass UBSAN checks if `llvm.experimental.hot()` is true."),
+llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check"),
llvm::cl::init(false));
//======//
@@ -3812,12 +3805,6 @@ void CodeGenFunction::EmitTrapCheck(llvm::Value *Checked,
SanitizerHandler CheckHandlerID) {
llvm::BasicBlock *Cont = createBasicBlock("cont");
- if (ClSanitizeExpHot) {
-Checked =
-Builder.CreateOr(Checked, Builder.CreateCall(CGM.getIntrinsic(
- llvm::Intrinsic::experimental_hot)));
- }
-
// If we're optimizing, collapse all calls to trap down to just one per
// check-type per function to save on code size.
if ((int)TrapBBs.size() <= CheckHandlerID)
diff --git a/clang/test/CodeGen/remote-traps.c
b/clang/test/CodeGen/remote-traps.c
index 16e4ebecb5c326..6751afb96d25f2 100644
--- a/clang/test/CodeGen/remote-traps.c
+++ b/clang/test/CodeGen/remote-traps.c
@@ -1,7 +1,5 @@
-// RUN: %clang_cc1 -O1 %s -o - -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -ubsan-exp-hot | FileCheck %s
-// RUN: %clang_cc1 -O1 %s -o - -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -ubsan-exp-hot -mllvm
-clang-remove-traps -mllvm -remove-traps-random-rate=1 %s -o - | FileCheck %s
--check-prefixes=REMOVE
-
-#include
+// RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow %s -o - | FileCheck %s
+// RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -clang-remove-traps -mllvm
-remove-traps-random-rate=1 %s -o - | FileCheck %s --implicit-check-not="call
void @llvm.ubsantrap" --check-prefixes=REMOVE
int test(int x) {
return x + 123;
@@ -14,19 +12,4 @@ int test(int x) {
// CHECK-NEXT: unreachable
// REMOVE-LABEL: define {{.*}}i32 @test(
-// REMOVE: add i32 %x, 123
-// REMOVE-NEXT: ret i32
-
-
-bool experimental_hot() __asm("llvm.experimental.hot");
-
-bool test_asm() {
- return experimental_hot();
-}
-
-// CHECK-LABEL: define {{.*}}i1 @test_asm(
-// CHECK: [[R:%.*]] = tail call zeroext i1 @llvm.experimental.hot()
-// CHECK: ret i1 [[R]]
-
-// REMOVE-LABEL: define {{.*}}i1 @test_asm(
-// REMOVE: ret i1 true
+// REMOVE: call { i32, i1 } @llvm.sadd.with.overflow.i32(
diff --git a/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
b/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
index fa4716f2e7a403..d87f7482a21d25 100644
--- a/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
+++ b/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
@@ -11,7 +11,6 @@
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/Analysis/ProfileSummaryInfo.h"
-#include "llvm/
[llvm-branch-commits] [clang] [llvm] [InstCombiner] Remove trivially dead `llvm.allow.{runtime, ubsan}.check()` (PR #84851)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/84851
>From c15aa098b7a8db614f574d47c220852674d4681d Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Mon, 11 Mar 2024 17:13:07 -0700
Subject: [PATCH 1/2] rebase
Created using spr 1.3.4
---
clang/lib/CodeGen/BackendUtil.cpp | 6 +-
clang/lib/CodeGen/CGExpr.cpp | 15 +---
clang/test/CodeGen/remote-traps.c | 23 +
.../Instrumentation/RemoveTrapsPass.cpp | 26 +++---
.../Transforms/RemoveTraps/remove-traps.ll| 89 ++-
5 files changed, 43 insertions(+), 116 deletions(-)
diff --git a/clang/lib/CodeGen/BackendUtil.cpp
b/clang/lib/CodeGen/BackendUtil.cpp
index 7e53469a48d42c..82b30b8d815629 100644
--- a/clang/lib/CodeGen/BackendUtil.cpp
+++ b/clang/lib/CodeGen/BackendUtil.cpp
@@ -100,9 +100,9 @@ using namespace llvm;
namespace llvm {
extern cl::opt PrintPipelinePasses;
-static cl::opt ClRemoveTraps("clang-remove-traps", cl::Optional,
- cl::desc("Insert remove-traps pass."),
- cl::init(false));
+cl::opt ClRemoveTraps("clang-remove-traps", cl::Optional,
+cl::desc("Insert remove-traps pass."),
+cl::init(false));
// Experiment to move sanitizers earlier.
static cl::opt ClSanitizeOnOptimizerEarlyEP(
diff --git a/clang/lib/CodeGen/CGExpr.cpp b/clang/lib/CodeGen/CGExpr.cpp
index 3a27622f165995..59a7fe8925001c 100644
--- a/clang/lib/CodeGen/CGExpr.cpp
+++ b/clang/lib/CodeGen/CGExpr.cpp
@@ -56,14 +56,7 @@ using namespace CodeGen;
// Experiment to make sanitizers easier to debug
static llvm::cl::opt ClSanitizeDebugDeoptimization(
"ubsan-unique-traps", llvm::cl::Optional,
-llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check."),
-llvm::cl::init(false));
-
-// TODO: Introduce frontend options to enabled per sanitizers, similar to
-// `fsanitize-trap`.
-static llvm::cl::opt ClSanitizeExpHot(
-"ubsan-exp-hot", llvm::cl::Optional,
-llvm::cl::desc("Pass UBSAN checks if `llvm.experimental.hot()` is true."),
+llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check"),
llvm::cl::init(false));
//======//
@@ -3812,12 +3805,6 @@ void CodeGenFunction::EmitTrapCheck(llvm::Value *Checked,
SanitizerHandler CheckHandlerID) {
llvm::BasicBlock *Cont = createBasicBlock("cont");
- if (ClSanitizeExpHot) {
-Checked =
-Builder.CreateOr(Checked, Builder.CreateCall(CGM.getIntrinsic(
- llvm::Intrinsic::experimental_hot)));
- }
-
// If we're optimizing, collapse all calls to trap down to just one per
// check-type per function to save on code size.
if ((int)TrapBBs.size() <= CheckHandlerID)
diff --git a/clang/test/CodeGen/remote-traps.c
b/clang/test/CodeGen/remote-traps.c
index 16e4ebecb5c326..6751afb96d25f2 100644
--- a/clang/test/CodeGen/remote-traps.c
+++ b/clang/test/CodeGen/remote-traps.c
@@ -1,7 +1,5 @@
-// RUN: %clang_cc1 -O1 %s -o - -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -ubsan-exp-hot | FileCheck %s
-// RUN: %clang_cc1 -O1 %s -o - -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -ubsan-exp-hot -mllvm
-clang-remove-traps -mllvm -remove-traps-random-rate=1 %s -o - | FileCheck %s
--check-prefixes=REMOVE
-
-#include
+// RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow %s -o - | FileCheck %s
+// RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -clang-remove-traps -mllvm
-remove-traps-random-rate=1 %s -o - | FileCheck %s --implicit-check-not="call
void @llvm.ubsantrap" --check-prefixes=REMOVE
int test(int x) {
return x + 123;
@@ -14,19 +12,4 @@ int test(int x) {
// CHECK-NEXT: unreachable
// REMOVE-LABEL: define {{.*}}i32 @test(
-// REMOVE: add i32 %x, 123
-// REMOVE-NEXT: ret i32
-
-
-bool experimental_hot() __asm("llvm.experimental.hot");
-
-bool test_asm() {
- return experimental_hot();
-}
-
-// CHECK-LABEL: define {{.*}}i1 @test_asm(
-// CHECK: [[R:%.*]] = tail call zeroext i1 @llvm.experimental.hot()
-// CHECK: ret i1 [[R]]
-
-// REMOVE-LABEL: define {{.*}}i1 @test_asm(
-// REMOVE: ret i1 true
+// REMOVE: call { i32, i1 } @llvm.sadd.with.overflow.i32(
diff --git a/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
b/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
index fa4716f2e7a403..d87f7482a21d25 100644
--- a/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
+++ b/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
@@ -11,7 +11,6 @@
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/Analysis/ProfileSummaryInfo.h"
-#include "l
[llvm-branch-commits] [llvm] [CodeGen] Add default lowering for llvm.allow.{runtime, ubsan}.check() (PR #86049)
https://github.com/vitalybuka updated https://github.com/llvm/llvm-project/pull/86049 >From d44531592108a483e537e23ee650713e12e1f919 Mon Sep 17 00:00:00 2001 From: Vitaly Buka Date: Thu, 21 Mar 2024 11:09:50 -0700 Subject: [PATCH] unsaved aarch64 Created using spr 1.3.4 --- llvm/test/CodeGen/AArch64/allow-check.ll | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/llvm/test/CodeGen/AArch64/allow-check.ll b/llvm/test/CodeGen/AArch64/allow-check.ll index c315b216e222c0..9e4a47357906c6 100644 --- a/llvm/test/CodeGen/AArch64/allow-check.ll +++ b/llvm/test/CodeGen/AArch64/allow-check.ll @@ -1,7 +1,7 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4 -; RUN: llc < %s -mtriple=aarch64 | FileCheck %s -; RUN: llc < %s -mtriple=aarch64 -global-isel | FileCheck %s -; RUN: llc < %s -mtriple=aarch64 -fast-isel | FileCheck %s +; RUN: llc < %s -mtriple=aarch64 -global-isel=0 -fast-isel=0 | FileCheck %s +; RUN: llc < %s -mtriple=aarch64 -global-isel=1 -fast-isel=0 | FileCheck %s +; RUN: llc < %s -mtriple=aarch64 -global-isel=0 -fast-isel=1 | FileCheck %s target triple = "aarch64-linux" ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [CodeGen] Add default lowering for llvm.allow.{runtime, ubsan}.check() (PR #86049)
https://github.com/arsenm approved this pull request. https://github.com/llvm/llvm-project/pull/86049 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [llvm] [InstCombiner] Remove trivially dead `llvm.allow.{runtime, ubsan}.check()` (PR #84851)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/84851
>From c15aa098b7a8db614f574d47c220852674d4681d Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Mon, 11 Mar 2024 17:13:07 -0700
Subject: [PATCH] rebase
Created using spr 1.3.4
---
clang/lib/CodeGen/BackendUtil.cpp | 6 +-
clang/lib/CodeGen/CGExpr.cpp | 15 +---
clang/test/CodeGen/remote-traps.c | 23 +
.../Instrumentation/RemoveTrapsPass.cpp | 26 +++---
.../Transforms/RemoveTraps/remove-traps.ll| 89 ++-
5 files changed, 43 insertions(+), 116 deletions(-)
diff --git a/clang/lib/CodeGen/BackendUtil.cpp
b/clang/lib/CodeGen/BackendUtil.cpp
index 7e53469a48d42c..82b30b8d815629 100644
--- a/clang/lib/CodeGen/BackendUtil.cpp
+++ b/clang/lib/CodeGen/BackendUtil.cpp
@@ -100,9 +100,9 @@ using namespace llvm;
namespace llvm {
extern cl::opt PrintPipelinePasses;
-static cl::opt ClRemoveTraps("clang-remove-traps", cl::Optional,
- cl::desc("Insert remove-traps pass."),
- cl::init(false));
+cl::opt ClRemoveTraps("clang-remove-traps", cl::Optional,
+cl::desc("Insert remove-traps pass."),
+cl::init(false));
// Experiment to move sanitizers earlier.
static cl::opt ClSanitizeOnOptimizerEarlyEP(
diff --git a/clang/lib/CodeGen/CGExpr.cpp b/clang/lib/CodeGen/CGExpr.cpp
index 3a27622f165995..59a7fe8925001c 100644
--- a/clang/lib/CodeGen/CGExpr.cpp
+++ b/clang/lib/CodeGen/CGExpr.cpp
@@ -56,14 +56,7 @@ using namespace CodeGen;
// Experiment to make sanitizers easier to debug
static llvm::cl::opt ClSanitizeDebugDeoptimization(
"ubsan-unique-traps", llvm::cl::Optional,
-llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check."),
-llvm::cl::init(false));
-
-// TODO: Introduce frontend options to enabled per sanitizers, similar to
-// `fsanitize-trap`.
-static llvm::cl::opt ClSanitizeExpHot(
-"ubsan-exp-hot", llvm::cl::Optional,
-llvm::cl::desc("Pass UBSAN checks if `llvm.experimental.hot()` is true."),
+llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check"),
llvm::cl::init(false));
//======//
@@ -3812,12 +3805,6 @@ void CodeGenFunction::EmitTrapCheck(llvm::Value *Checked,
SanitizerHandler CheckHandlerID) {
llvm::BasicBlock *Cont = createBasicBlock("cont");
- if (ClSanitizeExpHot) {
-Checked =
-Builder.CreateOr(Checked, Builder.CreateCall(CGM.getIntrinsic(
- llvm::Intrinsic::experimental_hot)));
- }
-
// If we're optimizing, collapse all calls to trap down to just one per
// check-type per function to save on code size.
if ((int)TrapBBs.size() <= CheckHandlerID)
diff --git a/clang/test/CodeGen/remote-traps.c
b/clang/test/CodeGen/remote-traps.c
index 16e4ebecb5c326..6751afb96d25f2 100644
--- a/clang/test/CodeGen/remote-traps.c
+++ b/clang/test/CodeGen/remote-traps.c
@@ -1,7 +1,5 @@
-// RUN: %clang_cc1 -O1 %s -o - -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -ubsan-exp-hot | FileCheck %s
-// RUN: %clang_cc1 -O1 %s -o - -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -ubsan-exp-hot -mllvm
-clang-remove-traps -mllvm -remove-traps-random-rate=1 %s -o - | FileCheck %s
--check-prefixes=REMOVE
-
-#include
+// RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow %s -o - | FileCheck %s
+// RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -clang-remove-traps -mllvm
-remove-traps-random-rate=1 %s -o - | FileCheck %s --implicit-check-not="call
void @llvm.ubsantrap" --check-prefixes=REMOVE
int test(int x) {
return x + 123;
@@ -14,19 +12,4 @@ int test(int x) {
// CHECK-NEXT: unreachable
// REMOVE-LABEL: define {{.*}}i32 @test(
-// REMOVE: add i32 %x, 123
-// REMOVE-NEXT: ret i32
-
-
-bool experimental_hot() __asm("llvm.experimental.hot");
-
-bool test_asm() {
- return experimental_hot();
-}
-
-// CHECK-LABEL: define {{.*}}i1 @test_asm(
-// CHECK: [[R:%.*]] = tail call zeroext i1 @llvm.experimental.hot()
-// CHECK: ret i1 [[R]]
-
-// REMOVE-LABEL: define {{.*}}i1 @test_asm(
-// REMOVE: ret i1 true
+// REMOVE: call { i32, i1 } @llvm.sadd.with.overflow.i32(
diff --git a/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
b/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
index fa4716f2e7a403..d87f7482a21d25 100644
--- a/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
+++ b/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
@@ -11,7 +11,6 @@
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/Analysis/ProfileSummaryInfo.h"
-#include "llvm/
[llvm-branch-commits] [clang] [llvm] [InstCombiner] Remove trivially dead `llvm.allow.{runtime, ubsan}.check()` (PR #84851)
https://github.com/nikic approved this pull request. LGTM https://github.com/llvm/llvm-project/pull/84851 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [compiler-rt] [test][hwasan] Test function name in summaries (PR #77391)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/77391
>From 23bf2a98a428fa6a866f64e2499793734b1a97a9 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Mon, 8 Jan 2024 15:27:19 -0800
Subject: [PATCH 1/2] undo some failing cases
Created using spr 1.3.4
---
compiler-rt/test/hwasan/TestCases/halt-on-error.cpp | 6 +++---
compiler-rt/test/hwasan/TestCases/report-unmapped.cpp | 2 +-
compiler-rt/test/hwasan/TestCases/use-after-free.c| 2 +-
3 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/compiler-rt/test/hwasan/TestCases/halt-on-error.cpp
b/compiler-rt/test/hwasan/TestCases/halt-on-error.cpp
index f587ee0fcabd30..7666baf9de79de 100644
--- a/compiler-rt/test/hwasan/TestCases/halt-on-error.cpp
+++ b/compiler-rt/test/hwasan/TestCases/halt-on-error.cpp
@@ -26,15 +26,15 @@ int main() {
// COMMON: READ of size 4 at
// When instrumenting with callbacks, main is actually #1, and #0 is
__hwasan_load4.
// COMMON: #{{.*}} in main {{.*}}halt-on-error.cpp:[[@LINE-3]]
- // COMMON: SUMMARY: HWAddressSanitizer: tag-mismatch {{.*}} in SigTrap
+ // COMMON: SUMMARY: HWAddressSanitizer: tag-mismatch
// RECOVER: READ of size 1 at
// RECOVER: #{{.*}} in main {{.*}}halt-on-error.cpp:[[@LINE-7]]
- // RECOVER: SUMMARY: HWAddressSanitizer: tag-mismatch {{.*}} in SigTrap
+ // RECOVER: SUMMARY: HWAddressSanitizer: tag-mismatch
// RECOVER: READ of size 1 at
// RECOVER: #{{.*}} in main {{.*}}halt-on-error.cpp:[[@LINE-11]]
- // RECOVER: SUMMARY: HWAddressSanitizer: tag-mismatch {{.*}} in SigTrap
+ // RECOVER: SUMMARY: HWAddressSanitizer: tag-mismatch
// COMMON-NOT: tag-mismatch
}
diff --git a/compiler-rt/test/hwasan/TestCases/report-unmapped.cpp
b/compiler-rt/test/hwasan/TestCases/report-unmapped.cpp
index 1802e920f390a2..935cd71c8c9c96 100644
--- a/compiler-rt/test/hwasan/TestCases/report-unmapped.cpp
+++ b/compiler-rt/test/hwasan/TestCases/report-unmapped.cpp
@@ -36,4 +36,4 @@ int main(int argc, char **argv) {
// CHECK: Tags for short granules around
// Check that report is complete.
-// CHECK: SUMMARY: HWAddressSanitizer: tag-mismatch {{.*}} in SigTrap
+// CHECK: SUMMARY: HWAddressSanitizer: tag-mismatch
diff --git a/compiler-rt/test/hwasan/TestCases/use-after-free.c
b/compiler-rt/test/hwasan/TestCases/use-after-free.c
index e0afb767b05937..b3eed88600726c 100644
--- a/compiler-rt/test/hwasan/TestCases/use-after-free.c
+++ b/compiler-rt/test/hwasan/TestCases/use-after-free.c
@@ -38,6 +38,6 @@ int main() {
// CHECK: #1 {{.*}} in main {{.*}}use-after-free.c:[[@LINE-24]]
// CHECK: Memory tags around the buggy address (one tag corresponds to 16
bytes):
// CHECK: =>{{.*}}[[MEM_TAG]]
- // CHECK: SUMMARY: HWAddressSanitizer: tag-mismatch {{.*}} in SigTrap
+ // CHECK: SUMMARY: HWAddressSanitizer: tag-mismatch
return r;
}
>From 3dc1e55b0a49a902cdb9abfdce3af9bb85489b70 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Mon, 8 Jan 2024 15:30:08 -0800
Subject: [PATCH 2/2] undo more
Created using spr 1.3.4
---
compiler-rt/test/hwasan/TestCases/halt-on-error.cpp | 6 +++---
compiler-rt/test/hwasan/TestCases/report-unmapped.cpp | 2 +-
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/compiler-rt/test/hwasan/TestCases/halt-on-error.cpp
b/compiler-rt/test/hwasan/TestCases/halt-on-error.cpp
index 7666baf9de79de..1a32e4bf4cc4b5 100644
--- a/compiler-rt/test/hwasan/TestCases/halt-on-error.cpp
+++ b/compiler-rt/test/hwasan/TestCases/halt-on-error.cpp
@@ -26,15 +26,15 @@ int main() {
// COMMON: READ of size 4 at
// When instrumenting with callbacks, main is actually #1, and #0 is
__hwasan_load4.
// COMMON: #{{.*}} in main {{.*}}halt-on-error.cpp:[[@LINE-3]]
- // COMMON: SUMMARY: HWAddressSanitizer: tag-mismatch
+ // COMMON: SUMMARY: HWAddressSanitizer: tag-mismatch {{.*}} in
// RECOVER: READ of size 1 at
// RECOVER: #{{.*}} in main {{.*}}halt-on-error.cpp:[[@LINE-7]]
- // RECOVER: SUMMARY: HWAddressSanitizer: tag-mismatch
+ // RECOVER: SUMMARY: HWAddressSanitizer: tag-mismatch {{.*}} in
// RECOVER: READ of size 1 at
// RECOVER: #{{.*}} in main {{.*}}halt-on-error.cpp:[[@LINE-11]]
- // RECOVER: SUMMARY: HWAddressSanitizer: tag-mismatch
+ // RECOVER: SUMMARY: HWAddressSanitizer: tag-mismatch {{.*}} in
// COMMON-NOT: tag-mismatch
}
diff --git a/compiler-rt/test/hwasan/TestCases/report-unmapped.cpp
b/compiler-rt/test/hwasan/TestCases/report-unmapped.cpp
index 935cd71c8c9c96..a58e50a78d8750 100644
--- a/compiler-rt/test/hwasan/TestCases/report-unmapped.cpp
+++ b/compiler-rt/test/hwasan/TestCases/report-unmapped.cpp
@@ -36,4 +36,4 @@ int main(int argc, char **argv) {
// CHECK: Tags for short granules around
// Check that report is complete.
-// CHECK: SUMMARY: HWAddressSanitizer: tag-mismatch
+// CHECK: SUMMARY: HWAddressSanitizer
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listin
[llvm-branch-commits] [clang] [llvm] [InstCombiner] Remove unused `llvm.experimental.hot()` (PR #84851)
@@ -56,7 +56,14 @@ using namespace CodeGen;
// Experiment to make sanitizers easier to debug
static llvm::cl::opt ClSanitizeDebugDeoptimization(
"ubsan-unique-traps", llvm::cl::Optional,
-llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check"),
+llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check."),
+llvm::cl::init(false));
+
+// TODO: Introduce frontend options to enabled per sanitizers, similar to
+// `fsanitize-trap`.
+static llvm::cl::opt ClSanitizeExpHot(
+"ubsan-exp-hot", llvm::cl::Optional,
+llvm::cl::desc("Pass UBSAN checks if `llvm.experimental.hot()` is true."),
jroelofs wrote:
moved here: https://github.com/llvm/llvm-project/pull/84858/files#r1520603139
https://github.com/llvm/llvm-project/pull/84851
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [CodeGen] Add default lowering for llvm.allow.{runtime, ubsan}.check() (PR #86049)
https://github.com/vitalybuka updated https://github.com/llvm/llvm-project/pull/86049 >From d44531592108a483e537e23ee650713e12e1f919 Mon Sep 17 00:00:00 2001 From: Vitaly Buka Date: Thu, 21 Mar 2024 11:09:50 -0700 Subject: [PATCH] unsaved aarch64 Created using spr 1.3.4 --- llvm/test/CodeGen/AArch64/allow-check.ll | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/llvm/test/CodeGen/AArch64/allow-check.ll b/llvm/test/CodeGen/AArch64/allow-check.ll index c315b216e222c0..9e4a47357906c6 100644 --- a/llvm/test/CodeGen/AArch64/allow-check.ll +++ b/llvm/test/CodeGen/AArch64/allow-check.ll @@ -1,7 +1,7 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4 -; RUN: llc < %s -mtriple=aarch64 | FileCheck %s -; RUN: llc < %s -mtriple=aarch64 -global-isel | FileCheck %s -; RUN: llc < %s -mtriple=aarch64 -fast-isel | FileCheck %s +; RUN: llc < %s -mtriple=aarch64 -global-isel=0 -fast-isel=0 | FileCheck %s +; RUN: llc < %s -mtriple=aarch64 -global-isel=1 -fast-isel=0 | FileCheck %s +; RUN: llc < %s -mtriple=aarch64 -global-isel=0 -fast-isel=1 | FileCheck %s target triple = "aarch64-linux" ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [CodeGen] Add default lowering for llvm.allow.{runtime, ubsan}.check() (PR #86049)
https://github.com/vitalybuka updated https://github.com/llvm/llvm-project/pull/86049 >From d44531592108a483e537e23ee650713e12e1f919 Mon Sep 17 00:00:00 2001 From: Vitaly Buka Date: Thu, 21 Mar 2024 11:09:50 -0700 Subject: [PATCH] unsaved aarch64 Created using spr 1.3.4 --- llvm/test/CodeGen/AArch64/allow-check.ll | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/llvm/test/CodeGen/AArch64/allow-check.ll b/llvm/test/CodeGen/AArch64/allow-check.ll index c315b216e222c0..9e4a47357906c6 100644 --- a/llvm/test/CodeGen/AArch64/allow-check.ll +++ b/llvm/test/CodeGen/AArch64/allow-check.ll @@ -1,7 +1,7 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4 -; RUN: llc < %s -mtriple=aarch64 | FileCheck %s -; RUN: llc < %s -mtriple=aarch64 -global-isel | FileCheck %s -; RUN: llc < %s -mtriple=aarch64 -fast-isel | FileCheck %s +; RUN: llc < %s -mtriple=aarch64 -global-isel=0 -fast-isel=0 | FileCheck %s +; RUN: llc < %s -mtriple=aarch64 -global-isel=1 -fast-isel=0 | FileCheck %s +; RUN: llc < %s -mtriple=aarch64 -global-isel=0 -fast-isel=1 | FileCheck %s target triple = "aarch64-linux" ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [Analysis] Exclude llvm.allow.{runtime, ubsan}.check() from AliasSetTracker (PR #86065)
https://github.com/vitalybuka updated https://github.com/llvm/llvm-project/pull/86065 >From f9cb87f89f2c2d4b17de4a4609ead16335102cf1 Mon Sep 17 00:00:00 2001 From: Vitaly Buka Date: Sun, 31 Mar 2024 22:45:20 -0700 Subject: [PATCH] rebase Created using spr 1.3.4 --- llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp | 1 + 1 file changed, 1 insertion(+) diff --git a/llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp b/llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp index 47e980e05281fc..0a3801bf965ff9 100644 --- a/llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp +++ b/llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp @@ -14,6 +14,7 @@ #include "llvm/ADT/STLExtras.h" #include "llvm/ADT/ScopeExit.h" #include "llvm/ADT/SmallSet.h" + #include "llvm/ADT/SmallVector.h" #include "llvm/Analysis/AliasAnalysis.h" #include "llvm/Analysis/AssumptionCache.h" ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [Analysis] Exclude llvm.allow.{runtime, ubsan}.check() from AliasSetTracker (PR #86065)
https://github.com/vitalybuka updated https://github.com/llvm/llvm-project/pull/86065 >From f9cb87f89f2c2d4b17de4a4609ead16335102cf1 Mon Sep 17 00:00:00 2001 From: Vitaly Buka Date: Sun, 31 Mar 2024 22:45:20 -0700 Subject: [PATCH 1/2] rebase Created using spr 1.3.4 --- llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp | 1 + 1 file changed, 1 insertion(+) diff --git a/llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp b/llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp index 47e980e05281fc..0a3801bf965ff9 100644 --- a/llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp +++ b/llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp @@ -14,6 +14,7 @@ #include "llvm/ADT/STLExtras.h" #include "llvm/ADT/ScopeExit.h" #include "llvm/ADT/SmallSet.h" + #include "llvm/ADT/SmallVector.h" #include "llvm/Analysis/AliasAnalysis.h" #include "llvm/Analysis/AssumptionCache.h" >From abc4c2f5049ef74720e8fd9f056791d54026466c Mon Sep 17 00:00:00 2001 From: Vitaly Buka Date: Sun, 31 Mar 2024 22:46:55 -0700 Subject: [PATCH 2/2] rebase Created using spr 1.3.4 --- llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp | 1 - 1 file changed, 1 deletion(-) diff --git a/llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp b/llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp index 0a3801bf965ff9..47e980e05281fc 100644 --- a/llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp +++ b/llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp @@ -14,7 +14,6 @@ #include "llvm/ADT/STLExtras.h" #include "llvm/ADT/ScopeExit.h" #include "llvm/ADT/SmallSet.h" - #include "llvm/ADT/SmallVector.h" #include "llvm/Analysis/AliasAnalysis.h" #include "llvm/Analysis/AssumptionCache.h" ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [llvm] [InstCombiner] Remove trivially dead `llvm.allow.{runtime, ubsan}.check()` (PR #84851)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/84851
>From c15aa098b7a8db614f574d47c220852674d4681d Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Mon, 11 Mar 2024 17:13:07 -0700
Subject: [PATCH 1/2] rebase
Created using spr 1.3.4
---
clang/lib/CodeGen/BackendUtil.cpp | 6 +-
clang/lib/CodeGen/CGExpr.cpp | 15 +---
clang/test/CodeGen/remote-traps.c | 23 +
.../Instrumentation/RemoveTrapsPass.cpp | 26 +++---
.../Transforms/RemoveTraps/remove-traps.ll| 89 ++-
5 files changed, 43 insertions(+), 116 deletions(-)
diff --git a/clang/lib/CodeGen/BackendUtil.cpp
b/clang/lib/CodeGen/BackendUtil.cpp
index 7e53469a48d42c..82b30b8d815629 100644
--- a/clang/lib/CodeGen/BackendUtil.cpp
+++ b/clang/lib/CodeGen/BackendUtil.cpp
@@ -100,9 +100,9 @@ using namespace llvm;
namespace llvm {
extern cl::opt PrintPipelinePasses;
-static cl::opt ClRemoveTraps("clang-remove-traps", cl::Optional,
- cl::desc("Insert remove-traps pass."),
- cl::init(false));
+cl::opt ClRemoveTraps("clang-remove-traps", cl::Optional,
+cl::desc("Insert remove-traps pass."),
+cl::init(false));
// Experiment to move sanitizers earlier.
static cl::opt ClSanitizeOnOptimizerEarlyEP(
diff --git a/clang/lib/CodeGen/CGExpr.cpp b/clang/lib/CodeGen/CGExpr.cpp
index 3a27622f165995..59a7fe8925001c 100644
--- a/clang/lib/CodeGen/CGExpr.cpp
+++ b/clang/lib/CodeGen/CGExpr.cpp
@@ -56,14 +56,7 @@ using namespace CodeGen;
// Experiment to make sanitizers easier to debug
static llvm::cl::opt ClSanitizeDebugDeoptimization(
"ubsan-unique-traps", llvm::cl::Optional,
-llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check."),
-llvm::cl::init(false));
-
-// TODO: Introduce frontend options to enabled per sanitizers, similar to
-// `fsanitize-trap`.
-static llvm::cl::opt ClSanitizeExpHot(
-"ubsan-exp-hot", llvm::cl::Optional,
-llvm::cl::desc("Pass UBSAN checks if `llvm.experimental.hot()` is true."),
+llvm::cl::desc("Deoptimize traps for UBSAN so there is 1 trap per check"),
llvm::cl::init(false));
//======//
@@ -3812,12 +3805,6 @@ void CodeGenFunction::EmitTrapCheck(llvm::Value *Checked,
SanitizerHandler CheckHandlerID) {
llvm::BasicBlock *Cont = createBasicBlock("cont");
- if (ClSanitizeExpHot) {
-Checked =
-Builder.CreateOr(Checked, Builder.CreateCall(CGM.getIntrinsic(
- llvm::Intrinsic::experimental_hot)));
- }
-
// If we're optimizing, collapse all calls to trap down to just one per
// check-type per function to save on code size.
if ((int)TrapBBs.size() <= CheckHandlerID)
diff --git a/clang/test/CodeGen/remote-traps.c
b/clang/test/CodeGen/remote-traps.c
index 16e4ebecb5c326..6751afb96d25f2 100644
--- a/clang/test/CodeGen/remote-traps.c
+++ b/clang/test/CodeGen/remote-traps.c
@@ -1,7 +1,5 @@
-// RUN: %clang_cc1 -O1 %s -o - -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -ubsan-exp-hot | FileCheck %s
-// RUN: %clang_cc1 -O1 %s -o - -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -ubsan-exp-hot -mllvm
-clang-remove-traps -mllvm -remove-traps-random-rate=1 %s -o - | FileCheck %s
--check-prefixes=REMOVE
-
-#include
+// RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow %s -o - | FileCheck %s
+// RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow
-fsanitize-trap=signed-integer-overflow -mllvm -clang-remove-traps -mllvm
-remove-traps-random-rate=1 %s -o - | FileCheck %s --implicit-check-not="call
void @llvm.ubsantrap" --check-prefixes=REMOVE
int test(int x) {
return x + 123;
@@ -14,19 +12,4 @@ int test(int x) {
// CHECK-NEXT: unreachable
// REMOVE-LABEL: define {{.*}}i32 @test(
-// REMOVE: add i32 %x, 123
-// REMOVE-NEXT: ret i32
-
-
-bool experimental_hot() __asm("llvm.experimental.hot");
-
-bool test_asm() {
- return experimental_hot();
-}
-
-// CHECK-LABEL: define {{.*}}i1 @test_asm(
-// CHECK: [[R:%.*]] = tail call zeroext i1 @llvm.experimental.hot()
-// CHECK: ret i1 [[R]]
-
-// REMOVE-LABEL: define {{.*}}i1 @test_asm(
-// REMOVE: ret i1 true
+// REMOVE: call { i32, i1 } @llvm.sadd.with.overflow.i32(
diff --git a/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
b/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
index fa4716f2e7a403..d87f7482a21d25 100644
--- a/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
+++ b/llvm/lib/Transforms/Instrumentation/RemoveTrapsPass.cpp
@@ -11,7 +11,6 @@
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/Analysis/ProfileSummaryInfo.h"
-#include "l
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
https://github.com/thurstond approved this pull request. https://github.com/llvm/llvm-project/pull/87568 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] 2f46953 - Revert "[clang][UBSan] Add implicit conversion check for bitfields (#75481)"
Author: Vitaly Buka Date: 2024-04-03T10:25:39-07:00 New Revision: 2f4695348d335ea70b167b738c48019d591c86b0 URL: https://github.com/llvm/llvm-project/commit/2f4695348d335ea70b167b738c48019d591c86b0 DIFF: https://github.com/llvm/llvm-project/commit/2f4695348d335ea70b167b738c48019d591c86b0.diff LOG: Revert "[clang][UBSan] Add implicit conversion check for bitfields (#75481)" This reverts commit 1189e87951e59a81ee097eae847c06008276fef1. This reverts commit 450f1952aced87584a53485d1ba1c2f77c3835a1. Added: Modified: clang/docs/ReleaseNotes.rst clang/docs/UndefinedBehaviorSanitizer.rst clang/include/clang/Basic/Sanitizers.def clang/lib/CodeGen/CGExpr.cpp clang/lib/CodeGen/CGExprScalar.cpp clang/lib/CodeGen/CodeGenFunction.h clang/test/Driver/fsanitize.c compiler-rt/lib/ubsan/ubsan_handlers.cpp compiler-rt/lib/ubsan/ubsan_handlers.h Removed: clang/test/CodeGen/ubsan-bitfield-conversion.c clang/test/CodeGenCXX/ubsan-bitfield-conversion.cpp diff --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst index e4c0e49ed6fc1c..8fc925350849cd 100644 --- a/clang/docs/ReleaseNotes.rst +++ b/clang/docs/ReleaseNotes.rst @@ -198,10 +198,6 @@ Non-comprehensive list of changes in this release New Compiler Flags -- -- ``-fsanitize=implicit-bitfield-conversion`` checks implicit truncation and - sign change. -- ``-fsanitize=implicit-integer-conversion`` a group that replaces the previous - group ``-fsanitize=implicit-conversion``. - ``-Wmissing-designated-field-initializers``, grouped under ``-Wmissing-field-initializers``. This diagnostic can be disabled to make ``-Wmissing-field-initializers`` behave @@ -215,9 +211,6 @@ Modified Compiler Flags - Added a new diagnostic flag ``-Wreturn-mismatch`` which is grouped under ``-Wreturn-type``, and moved some of the diagnostics previously controlled by ``-Wreturn-type`` under this new flag. Fixes #GH72116. -- ``-fsanitize=implicit-conversion`` is now a group for both - ``-fsanitize=implicit-integer-conversion`` and - ``-fsanitize=implicit-bitfield-conversion``. - Added ``-Wcast-function-type-mismatch`` under the ``-Wcast-function-type`` warning group. Moved the diagnostic previously controlled by diff --git a/clang/docs/UndefinedBehaviorSanitizer.rst b/clang/docs/UndefinedBehaviorSanitizer.rst index 531d56e313826c..8f58c92bd2a163 100644 --- a/clang/docs/UndefinedBehaviorSanitizer.rst +++ b/clang/docs/UndefinedBehaviorSanitizer.rst @@ -148,11 +148,6 @@ Available checks are: Issues caught by this sanitizer are not undefined behavior, but are often unintentional. - ``-fsanitize=integer-divide-by-zero``: Integer division by zero. - - ``-fsanitize=implicit-bitfield-conversion``: Implicit conversion from - integer of larger bit width to smaller bitfield, if that results in data - loss. This includes unsigned/signed truncations and sign changes, similarly - to how the ``-fsanitize=implicit-integer-conversion`` group works, but - explicitly for bitfields. - ``-fsanitize=nonnull-attribute``: Passing null pointer as a function parameter which is declared to never be null. - ``-fsanitize=null``: Use of a null pointer or creation of a null @@ -198,8 +193,8 @@ Available checks are: signed division overflow (``INT_MIN/-1``). Note that checks are still added even when ``-fwrapv`` is enabled. This sanitizer does not check for lossy implicit conversions performed before the computation (see - ``-fsanitize=implicit-integer-conversion``). Both of these two issues are handled - by ``-fsanitize=implicit-integer-conversion`` group of checks. + ``-fsanitize=implicit-conversion``). Both of these two issues are handled + by ``-fsanitize=implicit-conversion`` group of checks. - ``-fsanitize=unreachable``: If control flow reaches an unreachable program point. - ``-fsanitize=unsigned-integer-overflow``: Unsigned integer overflow, where @@ -207,7 +202,7 @@ Available checks are: type. Unlike signed integer overflow, this is not undefined behavior, but it is often unintentional. This sanitizer does not check for lossy implicit conversions performed before such a computation - (see ``-fsanitize=implicit-integer-conversion``). + (see ``-fsanitize=implicit-conversion``). - ``-fsanitize=vla-bound``: A variable-length array whose bound does not evaluate to a positive value. - ``-fsanitize=vptr``: Use of an object whose vptr indicates that it is of @@ -229,15 +224,11 @@ You can also use the following check groups: - ``-fsanitize=implicit-integer-arithmetic-value-change``: Catches implicit conversions that change the arithmetic value of the integer. Enables ``implicit-signed-integer-truncation`` and ``implicit-integer-sign-change``. - - ``-fsanitize=implicit-integer-conv
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/87568
>From 7ca1d437690a8202cc000aa3a431c26e26b9c358 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:30:10 -0700
Subject: [PATCH 1/2] undo random change
Created using spr 1.3.4
---
clang/lib/CodeGen/CGBuiltin.cpp | 12
1 file changed, 12 insertions(+)
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index f4809f09e025a2..d95a016c25dc92 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -3423,6 +3423,18 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl
GD, unsigned BuiltinID,
Builder.CreateCall(FnAssume, ArgValue);
return RValue::get(nullptr);
}
+ case Builtin::BI__builtin_assume_separate_storage: {
+const Expr *Arg0 = E->getArg(0);
+const Expr *Arg1 = E->getArg(1);
+
+Value *Value0 = EmitScalarExpr(Arg0);
+Value *Value1 = EmitScalarExpr(Arg1);
+
+Value *Values[] = {Value0, Value1};
+OperandBundleDefT OBD("separate_storage", Values);
+Builder.CreateAssumption(ConstantInt::getTrue(getLLVMContext()), {OBD});
+return RValue::get(nullptr);
+ }
case Builtin::BI__builtin_allow_runtime_check: {
StringRef Kind =
cast(E->getArg(0)->IgnoreParenCasts())->getString();
>From 8bb91d88e0d394b61bef885ecf9303018a855bd0 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:43:33 -0700
Subject: [PATCH 2/2] drop one IgnoreParenImpCasts
Created using spr 1.3.4
---
clang/lib/Sema/SemaChecking.cpp | 2 +-
clang/test/Sema/builtin-allow-runtime-check.c | 2 ++
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index 3ac456d5c54885..277836961a2951 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -3237,7 +3237,7 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl,
unsigned BuiltinID,
break;
case Builtin::BI__builtin_allow_runtime_check: {
-Expr *Arg = TheCall->getArg(0)->IgnoreParenImpCasts();
+Expr *Arg = TheCall->getArg(0);
// Check if the argument is a string literal.
if (!isa(Arg->IgnoreParenImpCasts())) {
Diag(TheCall->getBeginLoc(), diag::err_expr_not_string_literal)
diff --git a/clang/test/Sema/builtin-allow-runtime-check.c
b/clang/test/Sema/builtin-allow-runtime-check.c
index f4a94fa66321e5..b656861755 100644
--- a/clang/test/Sema/builtin-allow-runtime-check.c
+++ b/clang/test/Sema/builtin-allow-runtime-check.c
@@ -6,6 +6,8 @@ extern const char *str;
int main(void) {
int r = 0;
+ r |= __builtin_allow_runtime_check(); // expected-error {{too few arguments
to function call}}
+
r |= __builtin_allow_runtime_check(str); // expected-error {{expression is
not a string literal}}
r |= __builtin_allow_runtime_check(5); // expected-error {{incompatible
integer to pointer conversion}} expected-error {{expression is not a string
literal}}
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [llvm] [clang][ubsan] Switch UBSAN optimization to `llvm.allow.{runtime, ubsan}.check()` (PR #84858)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/84858
>From 852eb20b4f091a535ef758407d8555798b0ad809 Mon Sep 17 00:00:00 2001
From: Craig Topper
Date: Thu, 4 Apr 2024 16:17:19 -0700
Subject: [PATCH 1/2] [RISCV][GISel] Make register bank selection for unary and
binary arithmetic ops more generic. (#87593)
This is inspired by AArch64's getSameKindOfOperandsMapping, but based on
what RISC-V currently needs.
This removes the special vector case for G_ADD/SUB and unifies integer
and FP operations into the same handler.
G_SEXTLOAD/ZEXTLOAD have been separated from integer since they should
only be scalar integer and never vector.
---
.../RISCV/GISel/RISCVRegisterBankInfo.cpp | 45 ---
1 file changed, 28 insertions(+), 17 deletions(-)
diff --git a/llvm/lib/Target/RISCV/GISel/RISCVRegisterBankInfo.cpp
b/llvm/lib/Target/RISCV/GISel/RISCVRegisterBankInfo.cpp
index 8534024b6002b6..86e44343b50865 100644
--- a/llvm/lib/Target/RISCV/GISel/RISCVRegisterBankInfo.cpp
+++ b/llvm/lib/Target/RISCV/GISel/RISCVRegisterBankInfo.cpp
@@ -290,16 +290,7 @@ RISCVRegisterBankInfo::getInstrMapping(const MachineInstr
&MI) const {
switch (Opc) {
case TargetOpcode::G_ADD:
- case TargetOpcode::G_SUB: {
-if (MRI.getType(MI.getOperand(0).getReg()).isVector()) {
- LLT Ty = MRI.getType(MI.getOperand(0).getReg());
- return getInstructionMapping(
- DefaultMappingID, /*Cost=*/1,
- getVRBValueMapping(Ty.getSizeInBits().getKnownMinValue()),
- NumOperands);
-}
- }
-LLVM_FALLTHROUGH;
+ case TargetOpcode::G_SUB:
case TargetOpcode::G_SHL:
case TargetOpcode::G_ASHR:
case TargetOpcode::G_LSHR:
@@ -320,10 +311,6 @@ RISCVRegisterBankInfo::getInstrMapping(const MachineInstr
&MI) const {
case TargetOpcode::G_PTR_ADD:
case TargetOpcode::G_PTRTOINT:
case TargetOpcode::G_INTTOPTR:
- case TargetOpcode::G_SEXTLOAD:
- case TargetOpcode::G_ZEXTLOAD:
-return getInstructionMapping(DefaultMappingID, /*Cost=*/1, GPRValueMapping,
- NumOperands);
case TargetOpcode::G_FADD:
case TargetOpcode::G_FSUB:
case TargetOpcode::G_FMUL:
@@ -334,10 +321,34 @@ RISCVRegisterBankInfo::getInstrMapping(const MachineInstr
&MI) const {
case TargetOpcode::G_FMAXNUM:
case TargetOpcode::G_FMINNUM: {
LLT Ty = MRI.getType(MI.getOperand(0).getReg());
-return getInstructionMapping(DefaultMappingID, /*Cost=*/1,
- getFPValueMapping(Ty.getSizeInBits()),
- NumOperands);
+TypeSize Size = Ty.getSizeInBits();
+
+const ValueMapping *Mapping;
+if (Ty.isVector())
+ Mapping = getVRBValueMapping(Size.getKnownMinValue());
+else if (isPreISelGenericFloatingPointOpcode(Opc))
+ Mapping = getFPValueMapping(Size.getFixedValue());
+else
+ Mapping = GPRValueMapping;
+
+#ifndef NDEBUG
+// Make sure all the operands are using similar size and type.
+for (unsigned Idx = 1; Idx != NumOperands; ++Idx) {
+ LLT OpTy = MRI.getType(MI.getOperand(Idx).getReg());
+ assert(Ty.isVector() == OpTy.isVector() &&
+ "Operand has incompatible type");
+ // Don't check size for GPR.
+ if (OpTy.isVector() || isPreISelGenericFloatingPointOpcode(Opc))
+assert(Size == OpTy.getSizeInBits() && "Operand has incompatible
size");
+}
+#endif // End NDEBUG
+
+return getInstructionMapping(DefaultMappingID, 1, Mapping, NumOperands);
}
+ case TargetOpcode::G_SEXTLOAD:
+ case TargetOpcode::G_ZEXTLOAD:
+return getInstructionMapping(DefaultMappingID, /*Cost=*/1, GPRValueMapping,
+ NumOperands);
case TargetOpcode::G_IMPLICIT_DEF: {
Register Dst = MI.getOperand(0).getReg();
LLT DstTy = MRI.getType(Dst);
>From 634a277f9750af0b74b8e40e3fb509ca74b66fe1 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Mon, 11 Mar 2024 16:58:15 -0700
Subject: [PATCH 2/2] [clang][ubsan] Switch UBSAN optimization to
`llvm.allow.{runtime,ubsan}.check()`
Intrinsic introduced with #84850. Intrinsics improves performance
by 3% comparing to removing traps (on
"test-suite/MultiSource/Benchmarks" with PGO+ThinLTO).
The pass will be renamed with #84853.
RFC:
https://discourse.llvm.org/t/rfc-add-llvm-experimental-hot-intrinsic-or-llvm-hot/77641
Reviewers: aeubanks, nikic, kstoimenov, dvyukov
Reviewed By: kstoimenov
Pull Request: https://github.com/llvm/llvm-project/pull/84858
---
clang/test/CodeGen/remote-traps.c | 37 --
.../Instrumentation/RemoveTrapsPass.cpp | 25 ++--
.../Transforms/RemoveTraps/remove-traps.ll| 115 ++
3 files changed, 105 insertions(+), 72 deletions(-)
delete mode 100644 clang/test/CodeGen/remote-traps.c
diff --git a/clang/test/CodeGen/remote-traps.c
b/clang/test/CodeGen/remote-traps.c
deleted file mode 100644
index b12c2c6e23b2d1..00
--- a/clang/test/CodeGen/remote-traps.c
+++ /dev/null
@
[llvm-branch-commits] [clang] [compiler-rt] [flang] [libc] [libcxx] [lldb] [llvm] [mlir] [GlobalISel] Handle more commutable instructions in `commute_constant_to_rhs` (PR #87424)
Jan =?utf-8?q?Kokemüller?= ,Michael Buch ,Spenser Bauman ,Joseph Huber ,Daniil Kovalev ,A. Jiang =?utf-8?q?,?=Ryotaro KASUGA ,Joseph Huber ,Matthias Springer ,Craig Topper ,Jan Svoboda ,Fangrui Song ,Joseph Huber ,Cinhi Young ,Vinayak Dev <104419489+vinayakd...@users.noreply.github.com>,Fangrui Song ,Vinayak Dev <104419489+vinayakd...@users.noreply.github.com>,Chuanqi Xu ,Jonas Devlieghere ,smanna12 ,Slava Zakharin ,Mingming Liu ,hanbeom ,Bevin Hansson <59652494+bevin-hans...@users.noreply.github.com>,Chuanqi Xu ,Chuanqi Xu ,Phoebe Wang ,David Green ,Phoebe Wang ,Elizaveta Noskova <159026035+enoskova...@users.noreply.github.com>,Chuanqi Xu ,Chuanqi Xu ,Florian Hahn ,Chen Zheng ,Jay Foad ,Daniel Grumberg ,Han-Kuan Chen ,Louis Dionne ,Louis Dionne ,Haojian Wu ,Simon Camphausen ,Longsheng Mou ,Gleb Popov <6year...@gmail.com>,Jacek Caban ,Simon Pilgrim ,AinsleySnow <772571...@qq.com>,Louis Dionne ,Louis Dionne ,Florian Hahn ,Simon Pilgrim ,Daniel Chen ,Christian Ulmann ,Axel Lundberg <19574357+zonot...@users.noreply.github.com>,Dominik Adamski ,Simon Camphausen ,Florian Hahn ,Fanbo Meng ,Alexey Bataev ,aniplcc ,Simon Pilgrim ,Adrian Kuegel ,Simon Pilgrim ,Weining Lu ,David Spickett ,Paul Robinson ,Jay Foad , Amaury =?utf-8?q?Séchet?= ,Joe Nash ,Simon Pilgrim ,Mark de Wever ,Hsiangkai Wang ,Owen Pan ,Mark de Wever ,Chris Bieneman ,Nathan Chancellor ,Jonathan Peyton ,Krzysztof Pszeniczny ,Prashant Kumar ,Kazu Hirata ,Johannes Doerfert ,Alexey Bataev ,Slava Zakharin ,Vitaly Buka ,Vitaly Buka ,Aaron Ballman ,Michael Maitland ,maflcko <6399679+mafl...@users.noreply.github.com>,Alexey Bataev ,Rafael Ubal ,Louis Dionne ,Chenguang Wang ,Lei Wang ,Vitaly Buka ,David Green ,Lei Wang ,Alexey Bataev ,Simon Pilgrim ,David Green ,Changpeng Fang ,Joe Nash ,Kevin Frei ,Emma Pilkington ,Alexey Bataev =?utf-8?q?,?=Alexey Bataev ,Christopher Di Bella ,Shourya Goel ,Alexey Bataev ,Simon Pilgrim ,Slava Zakharin ,Damien L-G ,Vitaly Buka ,Michael Maitland ,Michael Maitland ,Michael Maitland ,Michael Maitland ,Michael Maitland ,Gulfem Savrun Yeniceri ,Kojo Acquah ,Kojo Acquah ,Dan Liew ,Chelsea Cassanova ,Yeoul Na ,Jason Molenda ,Han-Chung Wang ,Craig Topper ,Shih-Po Hung ,Vitaly Buka ,darkbuck ,Craig Topper ,Matthias Springer ,Hristo Hristov ,Luke Lau ,Sourabh Singh Tomar ,Gulfem Savrun Yeniceri ,Luke Lau ,Kazu Hirata ,Daniil Kovalev ,Andreas Jonson ,Christian Ulmann ,Jan Patrick Lehr =?utf-8?q?,?=Matthias Springer ,Haojian Wu ,Kadir Cetinkaya ,Matthias Springer ,Guillaume Chatelet ,Vyacheslav Levytskyy ,Guillaume Chatelet ,Fraser Cormack ,Jay Foad ,Stephen Tozer ,Tim Northover ,Daniil Kovalev ,Guillaume Chatelet ,Tom Eccles ,Stephen Tozer ,Ramkumar Ramachandra =?utf-8?q?,?=Tom Eccles ,Simon Pilgrim ,jeanPerier ,"Andrew V. Teylu" ,Simon Pilgrim ,Tom Eccles ,Tom Eccles ,Simon Pilgrim ,Zahira Ammarguellat ,Philip Lassen ,Luke Lau ,Florian Hahn ,Guillaume Chatelet ,Simon Pilgrim ,Simon Pilgrim ,Simon Pilgrim ,Victor Campos ,OverMighty ,Jacek Caban ,Mehdi Amini ,Yitzhak Mandelbaum , Andrzej =?utf-8?q?Warzyński?= ,Jacek Caban ,Christian Ulmann ,Mikael Holmen ,Phoebe Wang ,jeanPerier ,YunQiang Su ,Simon Pilgrim ,Simon Pilgrim ,Simon Pilgrim ,Simon Pilgrim ,Simon Pilgrim ,Simon Pilgrim ,Cyndy Ishida ,Piotr Sobczak ,Craig Topper ,Jay Foad ,Mark de Wever ,Simon Pilgrim ,Andrzej Warzynski ,lntue <35648136+ln...@users.noreply.github.com>,Simon Pilgrim ,Jay Foad ,Fangrui Song ,Fangrui Song ,Gulfem Savrun Yeniceri ,Julian Nagele ,Arthur Eubanks ,Fraser Cormack ,Mark de Wever ,Arthur Eubanks ,Arthur Eubanks ,Arthur Eubanks ,Arthur Eubanks ,Kazu Hirata ,Fangrui Song ,Fabian Mora ,Noah Goldstein ,Noah Goldstein ,Noah Goldstein ,Noah Goldstein ,Cyndy Ishida ,Daniil Kovalev ,Simon Pilgrim ,Eli Friedman ,Alexey Bataev ,Joseph Huber ,Kevin Frei ,Mark de Wever ,Jonathan Peyton ,Alexey Bataev ,Vitaly Buka ,Vitaly Buka ,Florian Mayer ,Keyi Zhang ,Vitaly Buka ,Jeff Niu , Valentin Clement =?utf-8?b?KOODkOODrOODsw=?=,Vitaly Buka ,Florian Mayer ,Ian Anderson ,Victor Campos ,Kazu Hirata ,MaheshRavishankar <1663364+maheshravishan...@users.noreply.github.com>,Nikolas Klauser ,Slava Zakharin ,Vitaly Buka ,Michael Jones ,Chenguang Wang ,Vitaly Buka ,Vitaly Buka ,Koakuma ,Vitaly Buka ,Michael Kruse ,Keith Smiley ,Paul Kirth ,Alexey Bataev ,Shubham Rastogi ,Vitaly Buka ,Craig Topper ,Shilei Tian ,Vitaly Buka ,Qizhi Hu <836744...@qq.com>,Youngsuk Kim ,Vitaly Buka ,Owen Pan ,Keith Smiley ,Farzon Lotfi <1802579+farz...@users.noreply.github.com>,Youngsuk Kim ,Congcong Cai ,Younan Zhang ,Matthias Springer ,Matthias Springer ,Vitaly Buka ,Matthias Springer ,Cyndy Ishida ,Christian Ulmann ,Christian Ulmann ,Qizhi Hu <836744...@qq.com>,mlevesquedion ,David Spickett ,David Spickett ,David Spickett ,Andrzej Warzynski ,AtariDreams ,NagyDonat ,NagyDonat ,Mitch Phillips ,Benjamin Maxwell ,Mehdi Amini ,Mehdi Amini
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
vitalybuka wrote: @MaskRay @nikic @efriedma-quic Any comments? https://github.com/llvm/llvm-project/pull/87568 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
vitalybuka wrote: I assume clang/docs/LanguageExtensions.rst counts as https://github.com/llvm/llvm-project/blob/main/clang/CodeOwners.rst#all-parts-of-clang-not-covered-by-someone-else https://github.com/llvm/llvm-project/pull/87568 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
efriedma-quic wrote: I think it's worth re-posting the builtin as a separate RFC on Discourse, since the original RFC hadn't settled on the exact design for the clang builtin you're using here. Code changes look fine. https://github.com/llvm/llvm-project/pull/87568 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/87568
>From 7ca1d437690a8202cc000aa3a431c26e26b9c358 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:30:10 -0700
Subject: [PATCH 1/2] undo random change
Created using spr 1.3.4
---
clang/lib/CodeGen/CGBuiltin.cpp | 12
1 file changed, 12 insertions(+)
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index f4809f09e025a2..d95a016c25dc92 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -3423,6 +3423,18 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl
GD, unsigned BuiltinID,
Builder.CreateCall(FnAssume, ArgValue);
return RValue::get(nullptr);
}
+ case Builtin::BI__builtin_assume_separate_storage: {
+const Expr *Arg0 = E->getArg(0);
+const Expr *Arg1 = E->getArg(1);
+
+Value *Value0 = EmitScalarExpr(Arg0);
+Value *Value1 = EmitScalarExpr(Arg1);
+
+Value *Values[] = {Value0, Value1};
+OperandBundleDefT OBD("separate_storage", Values);
+Builder.CreateAssumption(ConstantInt::getTrue(getLLVMContext()), {OBD});
+return RValue::get(nullptr);
+ }
case Builtin::BI__builtin_allow_runtime_check: {
StringRef Kind =
cast(E->getArg(0)->IgnoreParenCasts())->getString();
>From 8bb91d88e0d394b61bef885ecf9303018a855bd0 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:43:33 -0700
Subject: [PATCH 2/2] drop one IgnoreParenImpCasts
Created using spr 1.3.4
---
clang/lib/Sema/SemaChecking.cpp | 2 +-
clang/test/Sema/builtin-allow-runtime-check.c | 2 ++
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index 3ac456d5c54885..277836961a2951 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -3237,7 +3237,7 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl,
unsigned BuiltinID,
break;
case Builtin::BI__builtin_allow_runtime_check: {
-Expr *Arg = TheCall->getArg(0)->IgnoreParenImpCasts();
+Expr *Arg = TheCall->getArg(0);
// Check if the argument is a string literal.
if (!isa(Arg->IgnoreParenImpCasts())) {
Diag(TheCall->getBeginLoc(), diag::err_expr_not_string_literal)
diff --git a/clang/test/Sema/builtin-allow-runtime-check.c
b/clang/test/Sema/builtin-allow-runtime-check.c
index f4a94fa66321e5..b656861755 100644
--- a/clang/test/Sema/builtin-allow-runtime-check.c
+++ b/clang/test/Sema/builtin-allow-runtime-check.c
@@ -6,6 +6,8 @@ extern const char *str;
int main(void) {
int r = 0;
+ r |= __builtin_allow_runtime_check(); // expected-error {{too few arguments
to function call}}
+
r |= __builtin_allow_runtime_check(str); // expected-error {{expression is
not a string literal}}
r |= __builtin_allow_runtime_check(5); // expected-error {{incompatible
integer to pointer conversion}} expected-error {{expression is not a string
literal}}
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/87568
>From 7ca1d437690a8202cc000aa3a431c26e26b9c358 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:30:10 -0700
Subject: [PATCH 1/2] undo random change
Created using spr 1.3.4
---
clang/lib/CodeGen/CGBuiltin.cpp | 12
1 file changed, 12 insertions(+)
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index f4809f09e025a2..d95a016c25dc92 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -3423,6 +3423,18 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl
GD, unsigned BuiltinID,
Builder.CreateCall(FnAssume, ArgValue);
return RValue::get(nullptr);
}
+ case Builtin::BI__builtin_assume_separate_storage: {
+const Expr *Arg0 = E->getArg(0);
+const Expr *Arg1 = E->getArg(1);
+
+Value *Value0 = EmitScalarExpr(Arg0);
+Value *Value1 = EmitScalarExpr(Arg1);
+
+Value *Values[] = {Value0, Value1};
+OperandBundleDefT OBD("separate_storage", Values);
+Builder.CreateAssumption(ConstantInt::getTrue(getLLVMContext()), {OBD});
+return RValue::get(nullptr);
+ }
case Builtin::BI__builtin_allow_runtime_check: {
StringRef Kind =
cast(E->getArg(0)->IgnoreParenCasts())->getString();
>From 8bb91d88e0d394b61bef885ecf9303018a855bd0 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:43:33 -0700
Subject: [PATCH 2/2] drop one IgnoreParenImpCasts
Created using spr 1.3.4
---
clang/lib/Sema/SemaChecking.cpp | 2 +-
clang/test/Sema/builtin-allow-runtime-check.c | 2 ++
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index 3ac456d5c54885..277836961a2951 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -3237,7 +3237,7 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl,
unsigned BuiltinID,
break;
case Builtin::BI__builtin_allow_runtime_check: {
-Expr *Arg = TheCall->getArg(0)->IgnoreParenImpCasts();
+Expr *Arg = TheCall->getArg(0);
// Check if the argument is a string literal.
if (!isa(Arg->IgnoreParenImpCasts())) {
Diag(TheCall->getBeginLoc(), diag::err_expr_not_string_literal)
diff --git a/clang/test/Sema/builtin-allow-runtime-check.c
b/clang/test/Sema/builtin-allow-runtime-check.c
index f4a94fa66321e5..b656861755 100644
--- a/clang/test/Sema/builtin-allow-runtime-check.c
+++ b/clang/test/Sema/builtin-allow-runtime-check.c
@@ -6,6 +6,8 @@ extern const char *str;
int main(void) {
int r = 0;
+ r |= __builtin_allow_runtime_check(); // expected-error {{too few arguments
to function call}}
+
r |= __builtin_allow_runtime_check(str); // expected-error {{expression is
not a string literal}}
r |= __builtin_allow_runtime_check(5); // expected-error {{incompatible
integer to pointer conversion}} expected-error {{expression is not a string
literal}}
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/87568
>From 7ca1d437690a8202cc000aa3a431c26e26b9c358 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:30:10 -0700
Subject: [PATCH 1/2] undo random change
Created using spr 1.3.4
---
clang/lib/CodeGen/CGBuiltin.cpp | 12
1 file changed, 12 insertions(+)
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index f4809f09e025a2..d95a016c25dc92 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -3423,6 +3423,18 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl
GD, unsigned BuiltinID,
Builder.CreateCall(FnAssume, ArgValue);
return RValue::get(nullptr);
}
+ case Builtin::BI__builtin_assume_separate_storage: {
+const Expr *Arg0 = E->getArg(0);
+const Expr *Arg1 = E->getArg(1);
+
+Value *Value0 = EmitScalarExpr(Arg0);
+Value *Value1 = EmitScalarExpr(Arg1);
+
+Value *Values[] = {Value0, Value1};
+OperandBundleDefT OBD("separate_storage", Values);
+Builder.CreateAssumption(ConstantInt::getTrue(getLLVMContext()), {OBD});
+return RValue::get(nullptr);
+ }
case Builtin::BI__builtin_allow_runtime_check: {
StringRef Kind =
cast(E->getArg(0)->IgnoreParenCasts())->getString();
>From 8bb91d88e0d394b61bef885ecf9303018a855bd0 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:43:33 -0700
Subject: [PATCH 2/2] drop one IgnoreParenImpCasts
Created using spr 1.3.4
---
clang/lib/Sema/SemaChecking.cpp | 2 +-
clang/test/Sema/builtin-allow-runtime-check.c | 2 ++
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index 3ac456d5c54885..277836961a2951 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -3237,7 +3237,7 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl,
unsigned BuiltinID,
break;
case Builtin::BI__builtin_allow_runtime_check: {
-Expr *Arg = TheCall->getArg(0)->IgnoreParenImpCasts();
+Expr *Arg = TheCall->getArg(0);
// Check if the argument is a string literal.
if (!isa(Arg->IgnoreParenImpCasts())) {
Diag(TheCall->getBeginLoc(), diag::err_expr_not_string_literal)
diff --git a/clang/test/Sema/builtin-allow-runtime-check.c
b/clang/test/Sema/builtin-allow-runtime-check.c
index f4a94fa66321e5..b656861755 100644
--- a/clang/test/Sema/builtin-allow-runtime-check.c
+++ b/clang/test/Sema/builtin-allow-runtime-check.c
@@ -6,6 +6,8 @@ extern const char *str;
int main(void) {
int r = 0;
+ r |= __builtin_allow_runtime_check(); // expected-error {{too few arguments
to function call}}
+
r |= __builtin_allow_runtime_check(str); // expected-error {{expression is
not a string literal}}
r |= __builtin_allow_runtime_check(5); // expected-error {{incompatible
integer to pointer conversion}} expected-error {{expression is not a string
literal}}
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/87568
>From 7ca1d437690a8202cc000aa3a431c26e26b9c358 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:30:10 -0700
Subject: [PATCH 1/2] undo random change
Created using spr 1.3.4
---
clang/lib/CodeGen/CGBuiltin.cpp | 12
1 file changed, 12 insertions(+)
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index f4809f09e025a2..d95a016c25dc92 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -3423,6 +3423,18 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl
GD, unsigned BuiltinID,
Builder.CreateCall(FnAssume, ArgValue);
return RValue::get(nullptr);
}
+ case Builtin::BI__builtin_assume_separate_storage: {
+const Expr *Arg0 = E->getArg(0);
+const Expr *Arg1 = E->getArg(1);
+
+Value *Value0 = EmitScalarExpr(Arg0);
+Value *Value1 = EmitScalarExpr(Arg1);
+
+Value *Values[] = {Value0, Value1};
+OperandBundleDefT OBD("separate_storage", Values);
+Builder.CreateAssumption(ConstantInt::getTrue(getLLVMContext()), {OBD});
+return RValue::get(nullptr);
+ }
case Builtin::BI__builtin_allow_runtime_check: {
StringRef Kind =
cast(E->getArg(0)->IgnoreParenCasts())->getString();
>From 8bb91d88e0d394b61bef885ecf9303018a855bd0 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:43:33 -0700
Subject: [PATCH 2/2] drop one IgnoreParenImpCasts
Created using spr 1.3.4
---
clang/lib/Sema/SemaChecking.cpp | 2 +-
clang/test/Sema/builtin-allow-runtime-check.c | 2 ++
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index 3ac456d5c54885..277836961a2951 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -3237,7 +3237,7 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl,
unsigned BuiltinID,
break;
case Builtin::BI__builtin_allow_runtime_check: {
-Expr *Arg = TheCall->getArg(0)->IgnoreParenImpCasts();
+Expr *Arg = TheCall->getArg(0);
// Check if the argument is a string literal.
if (!isa(Arg->IgnoreParenImpCasts())) {
Diag(TheCall->getBeginLoc(), diag::err_expr_not_string_literal)
diff --git a/clang/test/Sema/builtin-allow-runtime-check.c
b/clang/test/Sema/builtin-allow-runtime-check.c
index f4a94fa66321e5..b656861755 100644
--- a/clang/test/Sema/builtin-allow-runtime-check.c
+++ b/clang/test/Sema/builtin-allow-runtime-check.c
@@ -6,6 +6,8 @@ extern const char *str;
int main(void) {
int r = 0;
+ r |= __builtin_allow_runtime_check(); // expected-error {{too few arguments
to function call}}
+
r |= __builtin_allow_runtime_check(str); // expected-error {{expression is
not a string literal}}
r |= __builtin_allow_runtime_check(5); // expected-error {{incompatible
integer to pointer conversion}} expected-error {{expression is not a string
literal}}
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
@@ -3464,6 +3464,34 @@ Query for this feature with ``__has_builtin(__builtin_trap)``. ``__builtin_arm_trap`` is lowered to the ``llvm.aarch64.break`` builtin, and then to ``brk #payload``. +``__builtin_allow_runtime_check`` AaronBallman wrote: The docs aren't particularly clear to me -- why would the check at the current program location ever NOT be executed? What strings can be passed in? What compiler options impact it? Will it only accept string literals or are runtime values fine? https://github.com/llvm/llvm-project/pull/87568 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
@@ -3233,6 +3233,17 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl,
unsigned BuiltinID,
if (BuiltinCountZeroBitsGeneric(*this, TheCall))
return ExprError();
break;
+
+ case Builtin::BI__builtin_allow_runtime_check: {
+Expr *Arg = TheCall->getArg(0);
+// Check if the argument is a string literal.
+if (!isa(Arg->IgnoreParenImpCasts())) {
AaronBallman wrote:
This might be a reasonable restriction, but consider other cases like:
`constexpr` variable of string type, `constexpr` function call that returns a
`const char *`, `_Generic` that results in a string constant, etc. Would those
be reasonable to support?
https://github.com/llvm/llvm-project/pull/87568
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
https://github.com/vitalybuka edited https://github.com/llvm/llvm-project/pull/87568 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
@@ -3464,6 +3464,34 @@ Query for this feature with ``__has_builtin(__builtin_trap)``. ``__builtin_arm_trap`` is lowered to the ``llvm.aarch64.break`` builtin, and then to ``brk #payload``. +``__builtin_allow_runtime_check`` vitalybuka wrote: I've started RFC, as @efriedma-quic suggested https://discourse.llvm.org/t/rfc-introduce-new-clang-builtin-builtin-allow-runtime-check/78281 It needs to be compiler time lowered, so I think it should literals. https://github.com/llvm/llvm-project/pull/87568 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/87568
>From 7ca1d437690a8202cc000aa3a431c26e26b9c358 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:30:10 -0700
Subject: [PATCH 1/2] undo random change
Created using spr 1.3.4
---
clang/lib/CodeGen/CGBuiltin.cpp | 12
1 file changed, 12 insertions(+)
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index f4809f09e025a2..d95a016c25dc92 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -3423,6 +3423,18 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl
GD, unsigned BuiltinID,
Builder.CreateCall(FnAssume, ArgValue);
return RValue::get(nullptr);
}
+ case Builtin::BI__builtin_assume_separate_storage: {
+const Expr *Arg0 = E->getArg(0);
+const Expr *Arg1 = E->getArg(1);
+
+Value *Value0 = EmitScalarExpr(Arg0);
+Value *Value1 = EmitScalarExpr(Arg1);
+
+Value *Values[] = {Value0, Value1};
+OperandBundleDefT OBD("separate_storage", Values);
+Builder.CreateAssumption(ConstantInt::getTrue(getLLVMContext()), {OBD});
+return RValue::get(nullptr);
+ }
case Builtin::BI__builtin_allow_runtime_check: {
StringRef Kind =
cast(E->getArg(0)->IgnoreParenCasts())->getString();
>From 8bb91d88e0d394b61bef885ecf9303018a855bd0 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:43:33 -0700
Subject: [PATCH 2/2] drop one IgnoreParenImpCasts
Created using spr 1.3.4
---
clang/lib/Sema/SemaChecking.cpp | 2 +-
clang/test/Sema/builtin-allow-runtime-check.c | 2 ++
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index 3ac456d5c54885..277836961a2951 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -3237,7 +3237,7 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl,
unsigned BuiltinID,
break;
case Builtin::BI__builtin_allow_runtime_check: {
-Expr *Arg = TheCall->getArg(0)->IgnoreParenImpCasts();
+Expr *Arg = TheCall->getArg(0);
// Check if the argument is a string literal.
if (!isa(Arg->IgnoreParenImpCasts())) {
Diag(TheCall->getBeginLoc(), diag::err_expr_not_string_literal)
diff --git a/clang/test/Sema/builtin-allow-runtime-check.c
b/clang/test/Sema/builtin-allow-runtime-check.c
index f4a94fa66321e5..b656861755 100644
--- a/clang/test/Sema/builtin-allow-runtime-check.c
+++ b/clang/test/Sema/builtin-allow-runtime-check.c
@@ -6,6 +6,8 @@ extern const char *str;
int main(void) {
int r = 0;
+ r |= __builtin_allow_runtime_check(); // expected-error {{too few arguments
to function call}}
+
r |= __builtin_allow_runtime_check(str); // expected-error {{expression is
not a string literal}}
r |= __builtin_allow_runtime_check(5); // expected-error {{incompatible
integer to pointer conversion}} expected-error {{expression is not a string
literal}}
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
@@ -3233,6 +3233,17 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl,
unsigned BuiltinID,
if (BuiltinCountZeroBitsGeneric(*this, TheCall))
return ExprError();
break;
+
+ case Builtin::BI__builtin_allow_runtime_check: {
+Expr *Arg = TheCall->getArg(0);
+// Check if the argument is a string literal.
+if (!isa(Arg->IgnoreParenImpCasts())) {
vitalybuka wrote:
My interest is in using this feature to implement `assert` like macros which
will be enabled in cold code, and optimized out in hot. E.g. MY_ASSERT1,
MY_ASSERT2...
Each macro can provide own kind as literal. Number of different kinds is
supposed to be limited. So literal looks enough for me.
We can relax in the future, restricting could be harder.
https://github.com/llvm/llvm-project/pull/87568
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/87568
>From 7ca1d437690a8202cc000aa3a431c26e26b9c358 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:30:10 -0700
Subject: [PATCH 1/2] undo random change
Created using spr 1.3.4
---
clang/lib/CodeGen/CGBuiltin.cpp | 12
1 file changed, 12 insertions(+)
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index f4809f09e025a2..d95a016c25dc92 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -3423,6 +3423,18 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl
GD, unsigned BuiltinID,
Builder.CreateCall(FnAssume, ArgValue);
return RValue::get(nullptr);
}
+ case Builtin::BI__builtin_assume_separate_storage: {
+const Expr *Arg0 = E->getArg(0);
+const Expr *Arg1 = E->getArg(1);
+
+Value *Value0 = EmitScalarExpr(Arg0);
+Value *Value1 = EmitScalarExpr(Arg1);
+
+Value *Values[] = {Value0, Value1};
+OperandBundleDefT OBD("separate_storage", Values);
+Builder.CreateAssumption(ConstantInt::getTrue(getLLVMContext()), {OBD});
+return RValue::get(nullptr);
+ }
case Builtin::BI__builtin_allow_runtime_check: {
StringRef Kind =
cast(E->getArg(0)->IgnoreParenCasts())->getString();
>From 8bb91d88e0d394b61bef885ecf9303018a855bd0 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:43:33 -0700
Subject: [PATCH 2/2] drop one IgnoreParenImpCasts
Created using spr 1.3.4
---
clang/lib/Sema/SemaChecking.cpp | 2 +-
clang/test/Sema/builtin-allow-runtime-check.c | 2 ++
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index 3ac456d5c54885..277836961a2951 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -3237,7 +3237,7 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl,
unsigned BuiltinID,
break;
case Builtin::BI__builtin_allow_runtime_check: {
-Expr *Arg = TheCall->getArg(0)->IgnoreParenImpCasts();
+Expr *Arg = TheCall->getArg(0);
// Check if the argument is a string literal.
if (!isa(Arg->IgnoreParenImpCasts())) {
Diag(TheCall->getBeginLoc(), diag::err_expr_not_string_literal)
diff --git a/clang/test/Sema/builtin-allow-runtime-check.c
b/clang/test/Sema/builtin-allow-runtime-check.c
index f4a94fa66321e5..b656861755 100644
--- a/clang/test/Sema/builtin-allow-runtime-check.c
+++ b/clang/test/Sema/builtin-allow-runtime-check.c
@@ -6,6 +6,8 @@ extern const char *str;
int main(void) {
int r = 0;
+ r |= __builtin_allow_runtime_check(); // expected-error {{too few arguments
to function call}}
+
r |= __builtin_allow_runtime_check(str); // expected-error {{expression is
not a string literal}}
r |= __builtin_allow_runtime_check(5); // expected-error {{incompatible
integer to pointer conversion}} expected-error {{expression is not a string
literal}}
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
https://github.com/nikic commented: This looks very reasonable to me, but I'm not a clang reviewer... https://github.com/llvm/llvm-project/pull/87568 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/87568
>From 7ca1d437690a8202cc000aa3a431c26e26b9c358 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:30:10 -0700
Subject: [PATCH 1/2] undo random change
Created using spr 1.3.4
---
clang/lib/CodeGen/CGBuiltin.cpp | 12
1 file changed, 12 insertions(+)
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index f4809f09e025a2..d95a016c25dc92 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -3423,6 +3423,18 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl
GD, unsigned BuiltinID,
Builder.CreateCall(FnAssume, ArgValue);
return RValue::get(nullptr);
}
+ case Builtin::BI__builtin_assume_separate_storage: {
+const Expr *Arg0 = E->getArg(0);
+const Expr *Arg1 = E->getArg(1);
+
+Value *Value0 = EmitScalarExpr(Arg0);
+Value *Value1 = EmitScalarExpr(Arg1);
+
+Value *Values[] = {Value0, Value1};
+OperandBundleDefT OBD("separate_storage", Values);
+Builder.CreateAssumption(ConstantInt::getTrue(getLLVMContext()), {OBD});
+return RValue::get(nullptr);
+ }
case Builtin::BI__builtin_allow_runtime_check: {
StringRef Kind =
cast(E->getArg(0)->IgnoreParenCasts())->getString();
>From 8bb91d88e0d394b61bef885ecf9303018a855bd0 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:43:33 -0700
Subject: [PATCH 2/2] drop one IgnoreParenImpCasts
Created using spr 1.3.4
---
clang/lib/Sema/SemaChecking.cpp | 2 +-
clang/test/Sema/builtin-allow-runtime-check.c | 2 ++
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index 3ac456d5c54885..277836961a2951 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -3237,7 +3237,7 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl,
unsigned BuiltinID,
break;
case Builtin::BI__builtin_allow_runtime_check: {
-Expr *Arg = TheCall->getArg(0)->IgnoreParenImpCasts();
+Expr *Arg = TheCall->getArg(0);
// Check if the argument is a string literal.
if (!isa(Arg->IgnoreParenImpCasts())) {
Diag(TheCall->getBeginLoc(), diag::err_expr_not_string_literal)
diff --git a/clang/test/Sema/builtin-allow-runtime-check.c
b/clang/test/Sema/builtin-allow-runtime-check.c
index f4a94fa66321e5..b656861755 100644
--- a/clang/test/Sema/builtin-allow-runtime-check.c
+++ b/clang/test/Sema/builtin-allow-runtime-check.c
@@ -6,6 +6,8 @@ extern const char *str;
int main(void) {
int r = 0;
+ r |= __builtin_allow_runtime_check(); // expected-error {{too few arguments
to function call}}
+
r |= __builtin_allow_runtime_check(str); // expected-error {{expression is
not a string literal}}
r |= __builtin_allow_runtime_check(5); // expected-error {{incompatible
integer to pointer conversion}} expected-error {{expression is not a string
literal}}
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
https://github.com/vitalybuka updated
https://github.com/llvm/llvm-project/pull/87568
>From 7ca1d437690a8202cc000aa3a431c26e26b9c358 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:30:10 -0700
Subject: [PATCH 1/2] undo random change
Created using spr 1.3.4
---
clang/lib/CodeGen/CGBuiltin.cpp | 12
1 file changed, 12 insertions(+)
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index f4809f09e025a2..d95a016c25dc92 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -3423,6 +3423,18 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl
GD, unsigned BuiltinID,
Builder.CreateCall(FnAssume, ArgValue);
return RValue::get(nullptr);
}
+ case Builtin::BI__builtin_assume_separate_storage: {
+const Expr *Arg0 = E->getArg(0);
+const Expr *Arg1 = E->getArg(1);
+
+Value *Value0 = EmitScalarExpr(Arg0);
+Value *Value1 = EmitScalarExpr(Arg1);
+
+Value *Values[] = {Value0, Value1};
+OperandBundleDefT OBD("separate_storage", Values);
+Builder.CreateAssumption(ConstantInt::getTrue(getLLVMContext()), {OBD});
+return RValue::get(nullptr);
+ }
case Builtin::BI__builtin_allow_runtime_check: {
StringRef Kind =
cast(E->getArg(0)->IgnoreParenCasts())->getString();
>From 8bb91d88e0d394b61bef885ecf9303018a855bd0 Mon Sep 17 00:00:00 2001
From: Vitaly Buka
Date: Wed, 3 Apr 2024 15:43:33 -0700
Subject: [PATCH 2/2] drop one IgnoreParenImpCasts
Created using spr 1.3.4
---
clang/lib/Sema/SemaChecking.cpp | 2 +-
clang/test/Sema/builtin-allow-runtime-check.c | 2 ++
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index 3ac456d5c54885..277836961a2951 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -3237,7 +3237,7 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl,
unsigned BuiltinID,
break;
case Builtin::BI__builtin_allow_runtime_check: {
-Expr *Arg = TheCall->getArg(0)->IgnoreParenImpCasts();
+Expr *Arg = TheCall->getArg(0);
// Check if the argument is a string literal.
if (!isa(Arg->IgnoreParenImpCasts())) {
Diag(TheCall->getBeginLoc(), diag::err_expr_not_string_literal)
diff --git a/clang/test/Sema/builtin-allow-runtime-check.c
b/clang/test/Sema/builtin-allow-runtime-check.c
index f4a94fa66321e5..b656861755 100644
--- a/clang/test/Sema/builtin-allow-runtime-check.c
+++ b/clang/test/Sema/builtin-allow-runtime-check.c
@@ -6,6 +6,8 @@ extern const char *str;
int main(void) {
int r = 0;
+ r |= __builtin_allow_runtime_check(); // expected-error {{too few arguments
to function call}}
+
r |= __builtin_allow_runtime_check(str); // expected-error {{expression is
not a string literal}}
r |= __builtin_allow_runtime_check(5); // expected-error {{incompatible
integer to pointer conversion}} expected-error {{expression is not a string
literal}}
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang][builtin] Implement __builtin_allow_runtime_check (PR #87568)
@@ -3464,6 +3464,34 @@ Query for this feature with
``__has_builtin(__builtin_trap)``.
``__builtin_arm_trap`` is lowered to the ``llvm.aarch64.break`` builtin, and
then to ``brk #payload``.
+``__builtin_allow_runtime_check``
AaronBallman wrote:
> It needs to be compiler time lowered, so I think it should literals.
In that case, I think we want *constant expression* not just literal, right?
e.g.,
```
constexpr const char *name = "name of check";
if (__builtin_allow_runtime_check(name)) {
}
```
is still able to be lowered at compile time. The reason I ask is because I'm
thinking about template metaprogramming cases where you might want to do
something along the lines of:
```
template
struct S {
void mem_fun() {
if (__builtin_allow_runtime_check(Ty::check_name) && stuff) {
}
}
};
```
but maybe this is not a compelling use case? I don't insist on constant
expression support, more just trying to verify that we support the expected
usage patterns.
https://github.com/llvm/llvm-project/pull/87568
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
