https://github.com/erichkeane edited
https://github.com/llvm/llvm-project/pull/152870
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/jhuber6 approved this pull request.
Seems fine, though I do wonder what's the point of SPIR-V being generic if we
hard-code behavior for different 'flavors' anyway.
https://github.com/llvm/llvm-project/pull/135251
___
cfe-commits ma
@@ -62,8 +76,23 @@ class IncrementalExecutor {
llvm::orc::LLJIT &GetExecutionEngine() { return *Jit; }
+#ifndef _WIN32
+ pid_t getOutOfProcessChildPid() const { return OutOfProcessChildPid; }
+#endif
vgvassilev wrote:
```suggestion
pid_t getOutOfProcess
https://github.com/AlexVlx updated
https://github.com/llvm/llvm-project/pull/154867
>From f65fa1deedae45ad8f3d007e39c914c091387be2 Mon Sep 17 00:00:00 2001
From: Alex Voicu
Date: Fri, 22 Aug 2025 01:02:42 +0100
Subject: [PATCH 1/3] Use "device" instead of "agent" and "subgroup" instead of
"wav
@@ -1444,15 +1444,16 @@ void clang::emitBackendOutput(CompilerInstance &CI,
CodeGenOptions &CGOpts,
// Verify clang's TargetInfo DataLayout against the LLVM TargetMachine's
// DataLayout.
- if (AsmHelper.TM) {
-std::string DLDesc = M->getDataLayout().getStringReprese
@@ -915,7 +948,14 @@ __m128i test_mm_mask_abs_epi8(__m128i __W, __mmask16 __U,
__m128i __A) {
// CHECK: select <16 x i1> %{{.*}}, <16 x i8> [[ABS]], <16 x i8> %{{.*}}
return _mm_mask_abs_epi8(__W,__U,__A);
}
-TEST_CONSTEXPR(match_v16qi(_mm_mask_abs_epi8((__m128i)(__v16qi)
https://github.com/RKSimon requested changes to this pull request.
https://github.com/llvm/llvm-project/pull/156234
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/lknknm updated
https://github.com/llvm/llvm-project/pull/156620
>From 10b4bcf3653c484652e9a832e613eb820530c34f Mon Sep 17 00:00:00 2001
From: Lucas <108239558+lkn...@users.noreply.github.com>
Date: Mon, 1 Sep 2025 22:54:46 +0200
Subject: [PATCH 1/8] Fix: Change Reference to 'B
github-actions[bot] wrote:
:warning: C/C++ code formatter, clang-format found issues in your code.
:warning:
You can test this locally with the following command:
``bash
git-clang-format --diff origin/main HEAD --extensions cpp,h,c --
clang/lib/CodeGen/CGDecl.cpp clang/lib/CodeG
@@ -28,7 +28,7 @@ Clang - Features and Goals
End-User Features:
-Fast compiles and low memory use
+Fast Compile-times and low memory use
lknknm wrote:
Sure! Just changed :)
https://github.com/llvm/llvm-project/pull/156620
___
jplehr wrote:
Attached is the IR from pre codegen. Apologies that this may again not be what
is needed. I always find the clang-linker-wrapper thing a bit confusing.
So, I hope this helps and thanks for the time.
[libc-gpu-precodegen.txt](https://github.com/user-attachments/files/22116762/libc-g
@@ -183,7 +183,7 @@ Library Based
Architecture
levels.
-As an example of the power of this library based design If you wanted
to
+As some examples for the power of this library based design: If you wanted
to
lknknm wrote:
Yes, mostly I just wante
https://github.com/SahilPatidar created
https://github.com/llvm/llvm-project/pull/156649
Reimplement value printing on top of ORC MemoryAccess. The previous
implementation only supported in-process evaluation; with this new
design it now works in both in-process and out-of-process modes.
The im
SahilPatidar wrote:
@vgvassilev This is still a draft — some parts are not implemented yet (e.g.
record and function types), but I shared it so you can see the idea and check
if the approach looks feasible.
https://github.com/llvm/llvm-project/pull/156649
SahilPatidar wrote:
And also I need a way to detect whether the interpreter is running in-process
or out-of-process.
https://github.com/llvm/llvm-project/pull/156649
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bi
https://github.com/lknknm updated
https://github.com/llvm/llvm-project/pull/156620
>From 10b4bcf3653c484652e9a832e613eb820530c34f Mon Sep 17 00:00:00 2001
From: Lucas <108239558+lkn...@users.noreply.github.com>
Date: Mon, 1 Sep 2025 22:54:46 +0200
Subject: [PATCH 01/10] Fix: Change Reference to
@@ -443,7 +443,8 @@ implementation.
+-+---+---+--+
| Traits for default device envirable
llvm-ci wrote:
LLVM Buildbot has detected a new failure on builder `ppc64le-flang-rhel-clang`
running on `ppc64le-flang-rhel-test` while building `clang` at step 5
"build-unified-tree".
Full details are available at:
https://lab.llvm.org/buildbot/#/builders/157/builds/38662
Here is the rele
@@ -541,7 +588,33 @@ struct scan_results
};
} // unnamed namespace
+} // extern "C"
+
+namespace {
+// The logical model for casting authenticated function pointers makes
+// it impossible to directly cast them without breaking the authentication,
+// as a result we need this
@@ -557,7 +596,19 @@ set_registers(_Unwind_Exception* unwind_exception,
_Unwind_Context* context,
reinterpret_cast(unwind_exception));
_Unwind_SetGR(context, __builtin_eh_return_data_regno(1),
static_cast(results.ttypeIndex));
+#if defined(__A
kovdan01 wrote:
Cannot add inline comment, so commenting on the whole file: `stepWithDwarf`
function contains piece of logic for handling signed RA. This works properly
for pac-ret, but it interferes with logic for ptrauth-returns which is a part
of pauthtest
@@ -377,13 +395,32 @@ const char *CFI_Parser::parseCIE(A &addressSpace,
pint_t cie,
case 'z':
cieInfo->fdesHaveAugmentationData = true;
break;
- case 'P':
+ case 'P': {
cieInfo->personalityEncoding = addressSpace.get8(p);
++p;
@@ -103,6 +103,52 @@
#define _LIBCXXABI_DTOR_FUNC
#endif
+#if __has_include()
+# include
+#endif
kovdan01 wrote:
@ojhunt Can we get rid of `__has_include()` check here and switch to
`__has_feature(ptrauth_calls) || defined(__PTRAUTH__)` just as a couple of
@@ -16,6 +16,12 @@
#include
#endif
+#if __has_feature(ptrauth_calls) || defined(__PTRAUTH__)
kovdan01 wrote:
It's probably worth submitting changes related to `__PTRAUTH__` introduced in
#153912 as a separate PR while keeping this PR only focused on unwind
@@ -557,7 +630,21 @@ set_registers(_Unwind_Exception* unwind_exception,
_Unwind_Context* context,
reinterpret_cast(unwind_exception));
_Unwind_SetGR(context, __builtin_eh_return_data_regno(1),
static_cast(results.ttypeIndex));
+#if __has_featu
@@ -30,6 +30,45 @@ EXCEPTION_DISPOSITION
_GCC_specific_handler(PEXCEPTION_RECORD, void *, PCONTEXT,
_Unwind_Personality_Fn);
#endif
+#if __has_feature(ptrauth_qualifier)
+#include
+#if __has_feature(ptrauth_restricted_intptr_qualif
@@ -238,7 +282,20 @@ COMPILER_RT_ABI _Unwind_Reason_Code __gcc_personality_v0(
_Unwind_SetGR(context, __builtin_eh_return_data_regno(0),
(uintptr_t)exceptionObject);
_Unwind_SetGR(context, __builtin_eh_return_data_regno(1), 0);
- _Unwind_Set
https://github.com/kovdan01 requested changes to this pull request.
https://github.com/llvm/llvm-project/pull/143230
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -377,13 +391,31 @@ const char *CFI_Parser::parseCIE(A &addressSpace,
pint_t cie,
case 'z':
cieInfo->fdesHaveAugmentationData = true;
break;
- case 'P':
+ case 'P': {
cieInfo->personalityEncoding = addressSpace.get8(p);
++p;
kovdan01 wrote:
Thanks @ojhunt for updates! These resolve some of the previously found issues,
but many problems are still present (both previously reported and newly found).
I tried to resolve all the threads which are now no longer relevant, so please
treat all the opened threads as issues w
@@ -681,7 +681,18 @@
DEFINE_LIBUNWIND_FUNCTION(__libunwind_Registers_arm64_jumpto)
// context struct, because it is allocated on the stack, and an exception
// could clobber the de-allocated portion of the stack after sp has been
// restored.
- ldrx16, [x0, #0x0
@@ -1047,18 +1051,26 @@ class UnwindCursor : public AbstractUnwindCursor{
bool getInfoFromFdeCie(const typename CFI_Parser::FDE_Info &fdeInfo,
const typename CFI_Parser::CIE_Info &cieInfo,
pint_t pc, uintptr_t dso_base);
- bo
@@ -126,6 +130,36 @@ _LIBUNWIND_HIDDEN int __unw_set_reg(unw_cursor_t *cursor,
unw_regnum_t regNum,
// First, get the FDE for the old location and then update it.
co->getInfo(&info);
co->setInfoBasedOnIPRegister(false);
+
+#if __has_feature(ptrauth_calls)
+
https://github.com/DavidSpickett approved this pull request.
LGTM.
You need to adjust your GitHub email settings before I can merge this. See
https://llvm.org/docs/DeveloperPolicy.html#email-addresses.
https://github.com/llvm/llvm-project/pull/156620
___
@@ -1479,7 +1492,85 @@ class SemaOpenMP : public SemaBase {
bool checkTransformableLoopNest(
OpenMPDirectiveKind Kind, Stmt *AStmt, int NumLoops,
SmallVectorImpl &LoopHelpers,
- Stmt *&Body, SmallVectorImpl> &OriginalInits);
+ Stmt *&Body, SmallVectorImp
https://github.com/alejandro-alvarez-sonarsource updated
https://github.com/llvm/llvm-project/pull/156507
From 736d0dcf2e31f402607656bf14f8c8dd6539 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Alejandro=20=C3=81lvarez=20Ayll=C3=B3n?=
Date: Fri, 29 Aug 2025 09:51:04 +0200
Subject: [PATCH 1/3] [
@@ -6776,6 +6776,28 @@ void Verifier::visitIntrinsicCall(Intrinsic::ID ID,
CallBase &Call) {
"invalid vector type for format", &Call, Src1,
Call.getArgOperand(2));
break;
}
+ case Intrinsic::amdgcn_cooperative_atomic_load_32x4B:
+ case Intrinsic::amdgcn_coop
https://github.com/Endilll closed
https://github.com/llvm/llvm-project/pull/142120
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/Naghasan commented:
sorry for the delay @sarnex
I have 2 questions:
- It seems strange to have different address spaces for 32/64. I understand you
probably don't care about 32, but it is tested.
- why mapping to generic ? This would conflict with
https://github.com/intel/
wenju-he wrote:
Applying weak attribute to a small set of libclc functions might be acceptable
because libclc will eventually set linkage to linkonce_odr for all functions.
The small set of libclc functions, that were not optimized by
PostOrderFunctionAttrsPass during libclc build, will be opt
@@ -1778,10 +1778,20 @@ def ext_bit_int : Extension<
} // end of Parse Issue category.
let CategoryName = "Modules Issue" in {
-def err_unexpected_module_decl : Error<
- "module declaration can only appear at the top level">;
+def err_invalid_module_or_import_directive : Erro
bolshakov-a wrote:
I always forget about release notes, thanks! However, there isn't "Potential
AST breaking changes" section but there is "AST Dumping Potentially Breaking
Changes" one. But this PR doesn't change AST dumps. Maybe, "C++ Specific
Potentially Breaking Changes"?
https://github.c
erichkeane wrote:
> > IS a pattern folks actually use intentionally. So making this an error
> > isn't something we can really do.
>
> Right, I think I wasn’t quite awake yet earlier because for some reason I was
> only thinking about the very specific use case where the `if` statement
> chec
https://github.com/hekota edited
https://github.com/llvm/llvm-project/pull/156075
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -1482,3 +1482,10 @@ bool CapturedStmt::capturesVariable(const VarDecl *Var)
const {
return false;
}
+
+Stmt *LoopControlStmt::getLabelTarget() const {
+ Stmt *Target = TargetLabel->getStmt();
+ while (isa_and_present(Target))
+Target = cast(Target)->getSubStmt();
+
nikic wrote:
GVN doesn't seem to do anything particularly interesting there. I tried running
this through llc to see how the codegen changes, but that gives:
> LLVM ERROR: Cannot select: intrinsic %llvm.amdgcn.if
Does this need extra llc flags beyond the triple?
https://github.com/llvm/llvm-p
jongmyeong-choi wrote:
I've implemented the change using a deferred cleanup approach.
Initially, I considered pushing cleanups to the EHStack while manipulating
PrologueCleanupDepth, but I felt this approach was too hacky.
If there's a better approach for this fix, please let me know.
https://g
https://github.com/tbaederr created
https://github.com/llvm/llvm-project/pull/156645
In the attached test case, the global variable later only points to gargbage,
because the MaterializeTemporaryExpr used to initialize it is a local variable,
which is gone by the time we try to evaluate the st
https://github.com/carlosgalvezp commented:
LGTM, I wonder why the `bB` was there in the first place. Binary numbers?
https://github.com/llvm/llvm-project/pull/156584
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bi
@@ -723,6 +811,52 @@ void CIRGenFunction::emitCXXDestructorCall(const
CXXDestructorDecl *dd,
delegating, thisAddr, thisTy);
}
+mlir::Value CIRGenFunction::getVTTParameter(GlobalDecl gd, bool forVirtualBase,
+
@@ -549,3 +652,32 @@ bool
CIRGenItaniumCXXABI::isVirtualOffsetNeededForVTableField(
return false;
return needsVTTParameter(cgf.curGD);
}
+
+mlir::Value CIRGenItaniumCXXABI::getVirtualBaseClassOffset(
+mlir::Location loc, CIRGenFunction &cgf, Address thisAddr,
+co
@@ -379,6 +395,23 @@ class CIRGenBuilderTy : public cir::CIRBaseBuilderTy {
/*mem_order=*/cir::MemOrderAttr{});
}
+ cir::LoadOp createAlignedLoad(mlir::Location loc, mlir::Type ty,
+mlir::Value ptr, llvm::MaybeA
@@ -329,10 +330,35 @@ static Address applyNonVirtualAndVirtualOffset(
loc, addr, baseValueTy, nonVirtualOffset.getQuantity(),
assumeNotNull);
}
+ } else {
+baseOffset = virtualOffset;
+ }
+
+ // Apply the base offset. cir.ptr_stride adjusts by a
https://github.com/xlauko edited
https://github.com/llvm/llvm-project/pull/156521
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -723,6 +811,52 @@ void CIRGenFunction::emitCXXDestructorCall(const
CXXDestructorDecl *dd,
delegating, thisAddr, thisTy);
}
+mlir::Value CIRGenFunction::getVTTParameter(GlobalDecl gd, bool forVirtualBase,
+
@@ -329,10 +330,35 @@ static Address applyNonVirtualAndVirtualOffset(
loc, addr, baseValueTy, nonVirtualOffset.getQuantity(),
assumeNotNull);
}
+ } else {
+baseOffset = virtualOffset;
+ }
+
+ // Apply the base offset. cir.ptr_stride adjusts by a
@@ -287,6 +287,10 @@ class CIRGenBuilderTy : public cir::CIRBaseBuilderTy {
cir::ConstantOp getUInt32(uint32_t c, mlir::Location loc) {
return getConstantInt(loc, getUInt32Ty(), c);
}
+ cir::ConstantOp getSInt64(uint64_t c, mlir::Location loc) {
+cir::IntType sInt6
https://github.com/xlauko approved this pull request.
lgtm, with minor nits
https://github.com/llvm/llvm-project/pull/156521
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
ckoparkar wrote:
@RKSimon Not much is left after #155405. I need to clean up the type checking
assertions in the new interpreter, probably just remove them. I'll finish it
soon.
https://github.com/llvm/llvm-project/pull/153572
___
cfe-commits mailing
@@ -21,20 +20,526 @@
#include "clang/AST/APValue.h"
#include "clang/AST/ASTContext.h"
#include "clang/AST/Attr.h"
+#include "clang/AST/CharUnits.h"
#include "clang/AST/OperationKinds.h"
#include "clang/AST/RecordLayout.h"
#include "clang/AST/StmtVisitor.h"
#include "clang/B
https://github.com/andykaylor updated
https://github.com/llvm/llvm-project/pull/155697
>From 36b02a82c306c31e8b824e365b0c7a6db34238d2 Mon Sep 17 00:00:00 2001
From: Andy Kaylor
Date: Wed, 27 Aug 2025 12:00:37 -0700
Subject: [PATCH 1/2] [CIR] Emit copy for aggregate initialization
This adds the
juj wrote:
Re-reading back the original description of the PR, I see there's
> This solves many open issues, like:
>
> * https://github.com/llvm/llvm-project/issues/64526
though this PR doesn't quite solve this, since it (silently?) will try to fix
up the call sites with mismatching point
@@ -909,3 +909,46 @@ void foo33(__builtin_va_list a) {
// OGCG: %[[B_IMAG_PTR:.*]] = getelementptr inbounds nuw { float, float },
ptr %[[B_ADDR]], i32 0, i32 1
// OGCG: store float %[[RESULT_REAL]], ptr %[[B_REAL_PTR]], align 4
// OGCG: store float %[[RESULT_IMAG]], ptr %[[
shiltian wrote:
> It isn't clear what these are, why/whether we want them, what their signature
> is, etc.
It is the support for [`__cluster_dims__`
attribute](https://docs.nvidia.com/cuda/cuda-c-programming-guide/#thread-block-clusters).
> Additionally, we do not allow adding undocumented at
erichkeane wrote:
> > It isn't clear what these are, why/whether we want them, what their
> > signature is, etc.
>
> It is the support for [`__cluster_dims__`
> attribute](https://docs.nvidia.com/cuda/cuda-c-programming-guide/#thread-block-clusters).
ALL of this (including a summary that does
https://github.com/erichkeane created
https://github.com/llvm/llvm-project/pull/156716
Patch #156545 is introducing a different syntax for the 'destroy' section of a
recipe, which takes the 'original' value as the first argument, and the
one-to-be-destroyed as the 2nd. This patch corrects the
@@ -828,11 +828,27 @@ llvm::Instruction
*CGHLSLRuntime::getConvergenceToken(BasicBlock &BB) {
class OpaqueValueVisitor : public RecursiveASTVisitor {
public:
- llvm::SmallPtrSet OVEs;
+ llvm::SmallVector OVEs;
+ llvm::SmallPtrSet Visited;
OpaqueValueVisitor() {}
+ bo
llvmbot wrote:
@llvm/pr-subscribers-clang
Author: Erich Keane (erichkeane)
Changes
Patch #156545 is introducing a different syntax for the 'destroy'
section of a recipe, which takes the 'original' value as the first argument,
and the one-to-be-destroyed as the 2nd. This patch corrects t
llvmbot wrote:
@llvm/pr-subscribers-clangir
Author: Erich Keane (erichkeane)
Changes
Patch #156545 is introducing a different syntax for the 'destroy'
section of a recipe, which takes the 'original' value as the first argument,
and the one-to-be-destroyed as the 2nd. This patch corrects
@@ -21,20 +20,526 @@
#include "clang/AST/APValue.h"
#include "clang/AST/ASTContext.h"
#include "clang/AST/Attr.h"
+#include "clang/AST/CharUnits.h"
#include "clang/AST/OperationKinds.h"
#include "clang/AST/RecordLayout.h"
#include "clang/AST/StmtVisitor.h"
#include "clang/B
@@ -30,7 +30,8 @@ llvm::Align CIRDataLayout::getAlignment(mlir::Type ty, bool
useABIAlign) const {
return llvm::Align(1);
// Get the layout annotation... which is lazily created on demand.
-llvm_unreachable("getAlignment()) for record type is not implemented");
@@ -199,6 +199,10 @@ Removed Compiler Flags
Attribute Changes in Clang
--
+- A new attribute ``[[clang::candiscard]]`` can be applied to a function
returning a nodiscard type
Sirraide wrote:
I’m not aware of any discussion but perhaps
@@ -557,7 +630,21 @@ set_registers(_Unwind_Exception* unwind_exception,
_Unwind_Context* context,
reinterpret_cast(unwind_exception));
_Unwind_SetGR(context, __builtin_eh_return_data_regno(1),
static_cast(results.ttypeIndex));
+#if __has_featu
turran wrote:
> Re-reading back the original description of the PR, I see there's
>
> > This solves many open issues, like:
> > ```
> > * https://github.com/llvm/llvm-project/issues/64526
> > ```
>
> though this PR doesn't quite solve this, since it (silently?) will try to fix
> up the call si
@@ -1047,18 +1051,26 @@ class UnwindCursor : public AbstractUnwindCursor{
bool getInfoFromFdeCie(const typename CFI_Parser::FDE_Info &fdeInfo,
const typename CFI_Parser::CIE_Info &cieInfo,
pint_t pc, uintptr_t dso_base);
- bo
@@ -30,6 +30,45 @@ EXCEPTION_DISPOSITION
_GCC_specific_handler(PEXCEPTION_RECORD, void *, PCONTEXT,
_Unwind_Personality_Fn);
#endif
+#if __has_feature(ptrauth_qualifier)
+#include
+#if __has_feature(ptrauth_restricted_intptr_qualif
@@ -541,7 +588,33 @@ struct scan_results
};
} // unnamed namespace
+} // extern "C"
+
+namespace {
+// The logical model for casting authenticated function pointers makes
+// it impossible to directly cast them without breaking the authentication,
+// as a result we need this
@@ -103,6 +103,52 @@
#define _LIBCXXABI_DTOR_FUNC
#endif
+#if __has_include()
+# include
+#endif
ojhunt wrote:
Once I get my local build setup again I will be doing a bunch of the ptrauth.h
cleanup - there are many places that were obviously transitively i
https://github.com/RKSimon updated
https://github.com/llvm/llvm-project/pull/156678
>From 5bdb914dc978150575a82f48e087f732f0e833ec Mon Sep 17 00:00:00 2001
From: Bhasawut Singhaphan
Date: Wed, 3 Sep 2025 20:44:26 +0700
Subject: [PATCH] [Headers][X86] Allow MMX/SSE integer min/max intrinsics to
@@ -98,21 +110,98 @@ void applySPIRVDistance(MachineInstr &MI,
MachineRegisterInfo &MRI,
SPIRVGlobalRegistry *GR =
MI.getMF()->getSubtarget().getSPIRVGlobalRegistry();
- auto RemoveAllUses = [&](Register Reg) {
-SmallVector UsesToErase(
-llvm::make_pointe
@@ -103,10 +104,68 @@
#define _LIBCXXABI_DTOR_FUNC
#endif
-#if __cplusplus < 201103L
-# define _LIBCXXABI_NOEXCEPT throw()
-#else
-# define _LIBCXXABI_NOEXCEPT noexcept
+#if __has_include()
+# include
#endif
+#if __has_extension(ptrauth_qualifier)
+
+// The actual value
https://github.com/RKSimon auto_merge_enabled
https://github.com/llvm/llvm-project/pull/156678
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/RKSimon approved this pull request.
LGTM - cheers
https://github.com/llvm/llvm-project/pull/156678
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/andykaylor updated
https://github.com/llvm/llvm-project/pull/155697
>From 1eab5f0f18485d58934bf87c0f1f3014cb76a995 Mon Sep 17 00:00:00 2001
From: Andy Kaylor
Date: Wed, 27 Aug 2025 12:00:37 -0700
Subject: [PATCH 1/3] [CIR] Emit copy for aggregate initialization
This adds the
https://github.com/changpeng updated
https://github.com/llvm/llvm-project/pull/156595
>From e1dd9629e396f1786903f7cda38072253ffaca00 Mon Sep 17 00:00:00 2001
From: Changpeng Fang
Date: Tue, 2 Sep 2025 23:50:28 -0700
Subject: [PATCH 1/2] [AMDGPU] Support cluster_load_async_to_lds instructions
o
markbhasawut wrote:
Thanks for reviewing! @RKSimon
https://github.com/llvm/llvm-project/pull/156678
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
llvmbot wrote:
@llvm/pr-subscribers-clang
Author: Changpeng Fang (changpeng)
Changes
---
Patch is 56.68 KiB, truncated to 20.00 KiB below, full version:
https://github.com/llvm/llvm-project/pull/156595.diff
16 Files Affected:
- (modified) clang/include/clang/Basic/BuiltinsAMDGPU.def
llvmbot wrote:
@llvm/pr-subscribers-llvm-ir
Author: Changpeng Fang (changpeng)
Changes
---
Patch is 56.68 KiB, truncated to 20.00 KiB below, full version:
https://github.com/llvm/llvm-project/pull/156595.diff
16 Files Affected:
- (modified) clang/include/clang/Basic/BuiltinsAMDGPU.d
llvmbot wrote:
@llvm/pr-subscribers-backend-amdgpu
Author: Changpeng Fang (changpeng)
Changes
---
Patch is 56.68 KiB, truncated to 20.00 KiB below, full version:
https://github.com/llvm/llvm-project/pull/156595.diff
16 Files Affected:
- (modified) clang/include/clang/Basic/BuiltinsA
@@ -16,6 +16,12 @@
#include
#endif
+#if __has_feature(ptrauth_calls) || defined(__PTRAUTH__)
ojhunt wrote:
I found this while updating the __PTRAUTH__ guards -- I would be ok leaving
this as a task for you folk as I think this is a linux specific file?
htt
ojhunt wrote:
hmmm, I thought `ptrauth_calls` was one of the checks guarded by a darwin
target so I might revert the `|| defined(__PTRAUTH__)` changes
https://github.com/llvm/llvm-project/pull/143230
___
cfe-commits mailing list
cfe-commits@lists.llvm
https://github.com/s-perron updated
https://github.com/llvm/llvm-project/pull/156075
>From aaecbc5efd38190e8aedd9b3e3c95b2ede791a09 Mon Sep 17 00:00:00 2001
From: Steven Perron
Date: Fri, 29 Aug 2025 13:40:55 -0400
Subject: [PATCH 1/2] [HLSL] Add copy assignment and construtor to resource
type
llvmbot wrote:
@llvm/pr-subscribers-hlsl
Author: Steven Perron (s-perron)
Changes
The wrapper used to hold the handle for resource type has just the
default copy constructor and assignment operator. This causes clang to
insert memcpys when it does an assignment of a resource type. The
memc
Sirraide wrote:
> I would also like to hear what @Sirraide thinks
I don’t think I have any further comments; I think the implementation lgtm but
I’ll leave it to @erichkeane to approve this.
https://github.com/llvm/llvm-project/pull/154943
___
cfe-co
https://github.com/markbhasawut created
https://github.com/llvm/llvm-project/pull/156722
This PR fixes #155411, add constexpr support for the AVX512F mullox intrinsics:
- _mm512_mullox_epi64
- _mm512_mask_mullox_epi64
Closes #156632.
>From c6d19e89ea5c4878cb04dd30bc410237a53646eb Mon Sep
@@ -1,49 +1,65 @@
-// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.6-compute
-finclude-default-header \
s-perron wrote:
Looking at the change, and it seems like the the CRLF was replaced with LF. My
editors must have done that without me knowing. I can try to r
kevinsala wrote:
My understanding is that the CUDA Driver API requires the arguments in the
buffer to be placed with the proper alignment (i.e., padding between fields).
However, the trailing padding after the last element should not be accounted.
Otherwise, if it's accounted, the `cuLaunchKer
Author: Bhasawut Singhaphan
Date: 2025-09-03T17:46:55Z
New Revision: b176ba78c8ded98cca955d4d32234fac8ada506c
URL:
https://github.com/llvm/llvm-project/commit/b176ba78c8ded98cca955d4d32234fac8ada506c
DIFF:
https://github.com/llvm/llvm-project/commit/b176ba78c8ded98cca955d4d32234fac8ada506c.diff
https://github.com/RKSimon closed
https://github.com/llvm/llvm-project/pull/156678
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/razvanlupusoru approved this pull request.
Thank you!
https://github.com/llvm/llvm-project/pull/156716
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
401 - 500 of 500 matches
Mail list logo
