[clang] Introduce paged vector (PR #66430)

[Giulio Eulisse via cfe-commits]

@@ -0,0 +1,322 @@
+//===- llvm/ADT/PagedVector.h - 'Lazyly allocated' vectors *- 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
+//
+//===--===//
+//
+// This file defines the PagedVector class.
+//
+//===--===//
+#ifndef LLVM_ADT_PAGEDVECTOR_H
+#define LLVM_ADT_PAGEDVECTOR_H
+
+#include "llvm/ADT/PointerIntPair.h"
+#include "llvm/Support/Allocator.h"
+#include 
+#include 
+#include 
+
+namespace llvm {
+// A vector that allocates memory in pages.
+// Order is kept, but memory is allocated only when one element of the page is
+// accessed. This introduces a level of indirection, but it is useful when you
+// have a sparsely initialised vector where the full size is allocated upfront
+// with the default constructor and elements are initialised later, on first
+// access.
+//
+// Notice that this does not have iterators, because if you
+// have iterators it probably means you are going to touch
+// all the memory in any case, so better use a std::vector in
+// the first place.
+//
+// Pages are allocated in SLAB_SIZE chunks, using the BumpPtrAllocator.
+template 
+class PagedVector {
+  static_assert(PAGE_SIZE > 0, "PAGE_SIZE must be greater than 0. Most likely "
+   "you want it to be greater than 16.");
+  // The actual number of element in the vector which can be accessed.
+  std::size_t Size = 0;
+
+  // The position of the initial element of the page in the Data vector.
+  // Pages are allocated contiguously in the Data vector.
+  mutable std::vector PageToDataIdx;

ktf wrote:

No, indeed, I have not checked. Sorry for not replying earlier. I will try to 
have some benchmark as soon as I finish some other unrelated unrelated tests.

https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [InstCombine] Canonicalize `and(zext(A), B)` into `select A, B & 1, 0` (PR #66740)

[via cfe-commits]

DianQK wrote:

We should use rebase instead of merge.
I recommend force push to change the code before the review starts.

https://github.com/llvm/llvm-project/pull/66740
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][dataflow] Fix two null pointer dereferences in `getMemberForAccessor()`. (PR #66742)

[via cfe-commits]

https://github.com/martinboehme resolved 
https://github.com/llvm/llvm-project/pull/66742
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][dataflow] Fix two null pointer dereferences in `getMemberForAccessor()`. (PR #66742)

[via cfe-commits]

@@ -289,11 +289,14 @@ static void insertIfFunction(const Decl &D,
 }
 
 static MemberExpr *getMemberForAccessor(const CXXMemberCallExpr &C) {
+  if (!C.getMethodDecl())
+return nullptr;
   auto *Body = dyn_cast_or_null(C.getMethodDecl()->getBody());
   if (!Body || Body->size() != 1)
 return nullptr;
   if (auto *RS = dyn_cast(*Body->body_begin()))
-return dyn_cast(RS->getRetValue()->IgnoreParenImpCasts());
+if (RS->getRetValue() != nullptr)

martinboehme wrote:

Done, thanks!

https://github.com/llvm/llvm-project/pull/66742
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][dataflow] Fix two null pointer dereferences in `getMemberForAccessor()`. (PR #66742)

[via cfe-commits]

https://github.com/martinboehme updated 
https://github.com/llvm/llvm-project/pull/66742

>From 5cdc525940af9b3b3e7432212b9e6852c6c7a602 Mon Sep 17 00:00:00 2001
From: Martin Braenne 
Date: Tue, 19 Sep 2023 06:49:27 +
Subject: [PATCH] [clang][dataflow] Fix two null pointer dereferences in
 `getMemberForAccessor()`.

The additions to the test trigger crashes without the fixes.
---
 clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp | 5 -
 clang/unittests/Analysis/FlowSensitive/TransferTest.cpp  | 9 +
 2 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp 
b/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp
index 26e097349057238..9dc528567038ac4 100644
--- a/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp
+++ b/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp
@@ -289,11 +289,14 @@ static void insertIfFunction(const Decl &D,
 }
 
 static MemberExpr *getMemberForAccessor(const CXXMemberCallExpr &C) {
+  if (!C.getMethodDecl())
+return nullptr;
   auto *Body = dyn_cast_or_null(C.getMethodDecl()->getBody());
   if (!Body || Body->size() != 1)
 return nullptr;
   if (auto *RS = dyn_cast(*Body->body_begin()))
-return dyn_cast(RS->getRetValue()->IgnoreParenImpCasts());
+if (auto *Return = RS->getRetValue())
+  return dyn_cast(Return->IgnoreParenImpCasts());
   return nullptr;
 }
 
diff --git a/clang/unittests/Analysis/FlowSensitive/TransferTest.cpp 
b/clang/unittests/Analysis/FlowSensitive/TransferTest.cpp
index 14188f5acd5b36e..e8cbca756460369 100644
--- a/clang/unittests/Analysis/FlowSensitive/TransferTest.cpp
+++ b/clang/unittests/Analysis/FlowSensitive/TransferTest.cpp
@@ -1463,6 +1463,7 @@ TEST(TransferTest, StructModeledFieldsWithAccessor) {
   int getIntNotAccessed() const { return IntNotAccessed; }
   int getIntNoDefinition() const;
   int &getIntRef() { return IntRef; }
+  void returnVoid() const { return; }
 };
 
 void target() {
@@ -1473,6 +1474,14 @@ TEST(TransferTest, StructModeledFieldsWithAccessor) {
   int i2 = s.getWithInc(1);
   int i3 = s.getIntNoDefinition();
   int &iref = s.getIntRef();
+
+  // Regression test: Don't crash on an indirect call (which doesn't have
+  // an associated `CXXMethodDecl`).
+  auto ptr_to_member_fn = &S::getPtr;
+  p1 = (s.*ptr_to_member_fn)();
+
+  // Regression test: Don't crash on a return statement without a value.
+  s.returnVoid();
   // [[p]]
 }
   )";

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][Driver] Obey -fuse-ld=... for -print-prog-name=ld output (PR #66698)

[Fangrui Song via cfe-commits]

MaskRay wrote:

> > Can you comment on what projects need this? Can't we fix the projects 
> > instead? And what do they do with the linker path? Why don't they invoke 
> > the clang driver to run a link action?
> 
> They are legion. It comes from autoconf checks for libtool. It doesn't really 
> matter why they're doing it. They are doing it and it will be years before 
> they can all be changed. If you want to use Clang as the compiler for random 
> open source code, it needs to be compatible with GCC in this regard.

Do your have some examples of such projects? As mentioned, I know multiple lld 
based distributions and they don't find `clang -fuse-ld=ld 
--print-prog-name=ld` not behave like GCC a problem. If you think there are 
many and existing distributions haven't detected them, I'd like to know the 
project.

BTW: if you make `ld` a symlink to `lld` or `ld.lld`, you'll pretty much not 
notice anything.


https://github.com/llvm/llvm-project/pull/66698
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] In ExprRequirement building, treat OverloadExpr as dependent (PR #66683)

[via cfe-commits]

https://github.com/cor3ntin approved this pull request.

LGTM (but now you have conflicts in the release notes!)

https://github.com/llvm/llvm-project/pull/66683
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Introduce paged vector (PR #66430)

[Giulio Eulisse via cfe-commits]

https://github.com/ktf resolved https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][dataflow] Fix two null pointer dereferences in `getMemberForAccessor()`. (PR #66742)

[Haojian Wu via cfe-commits]

@@ -289,11 +289,14 @@ static void insertIfFunction(const Decl &D,
 }
 
 static MemberExpr *getMemberForAccessor(const CXXMemberCallExpr &C) {
+  if (!C.getMethodDecl())
+return nullptr;
   auto *Body = dyn_cast_or_null(C.getMethodDecl()->getBody());
   if (!Body || Body->size() != 1)
 return nullptr;
   if (auto *RS = dyn_cast(*Body->body_begin()))
-return dyn_cast(RS->getRetValue()->IgnoreParenImpCasts());
+if (RS->getRetValue() != nullptr)

hokein wrote:

nit: `if (auto* ReturnValue = RS->getRetValue())` 

https://github.com/llvm/llvm-project/pull/66742
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][dataflow] Fix two null pointer dereferences in `getMemberForAccessor()`. (PR #66742)

[Haojian Wu via cfe-commits]

https://github.com/hokein approved this pull request.


https://github.com/llvm/llvm-project/pull/66742
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][dataflow] Fix two null pointer dereferences in `getMemberForAccessor()`. (PR #66742)

[Haojian Wu via cfe-commits]

https://github.com/hokein edited https://github.com/llvm/llvm-project/pull/66742
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][dataflow] Fix two null pointer dereferences in `getMemberForAccessor()`. (PR #66742)

[Fangrui Song via cfe-commits]

https://github.com/MaskRay approved this pull request.


https://github.com/llvm/llvm-project/pull/66742
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][dataflow] Fix two null pointer dereferences in `getMemberForAccessor()`. (PR #66742)

[Sam McCall via cfe-commits]

https://github.com/sam-mccall approved this pull request.


https://github.com/llvm/llvm-project/pull/66742
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][dataflow] Fix two null pointer dereferences in `getMemberForAccessor()`. (PR #66742)

[via cfe-commits]

llvmbot wrote:




@llvm/pr-subscribers-clang-analysis


Changes

The additions to the test trigger crashes without the fixes.


---
Full diff: https://github.com/llvm/llvm-project/pull/66742.diff


2 Files Affected:

- (modified) clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp (+4-1) 
- (modified) clang/unittests/Analysis/FlowSensitive/TransferTest.cpp (+9) 


``diff
diff --git a/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp 
b/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp
index 26e097349057238..98128693e145da4 100644
--- a/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp
+++ b/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp
@@ -289,11 +289,14 @@ static void insertIfFunction(const Decl &D,
 }
 
 static MemberExpr *getMemberForAccessor(const CXXMemberCallExpr &C) {
+  if (!C.getMethodDecl())
+return nullptr;
   auto *Body = dyn_cast_or_null(C.getMethodDecl()->getBody());
   if (!Body || Body->size() != 1)
 return nullptr;
   if (auto *RS = dyn_cast(*Body->body_begin()))
-return dyn_cast(RS->getRetValue()->IgnoreParenImpCasts());
+if (RS->getRetValue() != nullptr)
+  return dyn_cast(RS->getRetValue()->IgnoreParenImpCasts());
   return nullptr;
 }
 
diff --git a/clang/unittests/Analysis/FlowSensitive/TransferTest.cpp 
b/clang/unittests/Analysis/FlowSensitive/TransferTest.cpp
index 14188f5acd5b36e..e8cbca756460369 100644
--- a/clang/unittests/Analysis/FlowSensitive/TransferTest.cpp
+++ b/clang/unittests/Analysis/FlowSensitive/TransferTest.cpp
@@ -1463,6 +1463,7 @@ TEST(TransferTest, StructModeledFieldsWithAccessor) {
   int getIntNotAccessed() const { return IntNotAccessed; }
   int getIntNoDefinition() const;
   int &getIntRef() { return IntRef; }
+  void returnVoid() const { return; }
 };
 
 void target() {
@@ -1473,6 +1474,14 @@ TEST(TransferTest, StructModeledFieldsWithAccessor) {
   int i2 = s.getWithInc(1);
   int i3 = s.getIntNoDefinition();
   int &iref = s.getIntRef();
+
+  // Regression test: Don't crash on an indirect call (which doesn't have
+  // an associated `CXXMethodDecl`).
+  auto ptr_to_member_fn = &S::getPtr;
+  p1 = (s.*ptr_to_member_fn)();
+
+  // Regression test: Don't crash on a return statement without a value.
+  s.returnVoid();
   // [[p]]
 }
   )";

``




https://github.com/llvm/llvm-project/pull/66742
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][dataflow] Fix two null pointer dereferences in `getMemberForAccessor()`. (PR #66742)

[via cfe-commits]

https://github.com/martinboehme created 
https://github.com/llvm/llvm-project/pull/66742

The additions to the test trigger crashes without the fixes.


>From 110ac1d623c25a2cbde494dd708c408d6c638cc2 Mon Sep 17 00:00:00 2001
From: Martin Braenne 
Date: Tue, 19 Sep 2023 06:39:57 +
Subject: [PATCH] [clang][dataflow] Fix two null pointer dereferences in
 `getMemberForAccessor()`.

The additions to the test trigger crashes without the fixes.
---
 clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp | 5 -
 clang/unittests/Analysis/FlowSensitive/TransferTest.cpp  | 9 +
 2 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp 
b/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp
index 26e097349057238..98128693e145da4 100644
--- a/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp
+++ b/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp
@@ -289,11 +289,14 @@ static void insertIfFunction(const Decl &D,
 }
 
 static MemberExpr *getMemberForAccessor(const CXXMemberCallExpr &C) {
+  if (!C.getMethodDecl())
+return nullptr;
   auto *Body = dyn_cast_or_null(C.getMethodDecl()->getBody());
   if (!Body || Body->size() != 1)
 return nullptr;
   if (auto *RS = dyn_cast(*Body->body_begin()))
-return dyn_cast(RS->getRetValue()->IgnoreParenImpCasts());
+if (RS->getRetValue() != nullptr)
+  return dyn_cast(RS->getRetValue()->IgnoreParenImpCasts());
   return nullptr;
 }
 
diff --git a/clang/unittests/Analysis/FlowSensitive/TransferTest.cpp 
b/clang/unittests/Analysis/FlowSensitive/TransferTest.cpp
index 14188f5acd5b36e..e8cbca756460369 100644
--- a/clang/unittests/Analysis/FlowSensitive/TransferTest.cpp
+++ b/clang/unittests/Analysis/FlowSensitive/TransferTest.cpp
@@ -1463,6 +1463,7 @@ TEST(TransferTest, StructModeledFieldsWithAccessor) {
   int getIntNotAccessed() const { return IntNotAccessed; }
   int getIntNoDefinition() const;
   int &getIntRef() { return IntRef; }
+  void returnVoid() const { return; }
 };
 
 void target() {
@@ -1473,6 +1474,14 @@ TEST(TransferTest, StructModeledFieldsWithAccessor) {
   int i2 = s.getWithInc(1);
   int i3 = s.getIntNoDefinition();
   int &iref = s.getIntRef();
+
+  // Regression test: Don't crash on an indirect call (which doesn't have
+  // an associated `CXXMethodDecl`).
+  auto ptr_to_member_fn = &S::getPtr;
+  p1 = (s.*ptr_to_member_fn)();
+
+  // Regression test: Don't crash on a return statement without a value.
+  s.returnVoid();
   // [[p]]
 }
   )";

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D147844: [clang][Sema]Print diagnostic warning about precedence when integer expression is used without parentheses in an conditional operator expression

[Fangrui Song via Phabricator via cfe-commits]

MaskRay added a comment.

In D147844#4646633 , @MaskRay wrote:

> FWIW: adding parentheses for expressions, such as `overflow = (4608 * 1024 * 
> 1024) ?  4608 * 1024 * 1024 : 0;`, and `results.reason = (actions & 
> _UA_SEARCH_PHASE) ? ... : ...`, looks unnatural and is too noisy to me.

From quuxplusone:

The first expression

  overflow = (4608 * 1024 * 1024) ?  4608 * 1024 * 1024 : 0;

is a tautology; it is not possible for (4608 * 1024 * 1024) to be zero. (If 
it's signed integer overflow, it's UB, not zero; and the compiler knows this.)
But if you make it defined by adding `u`, then IMO it's very little hardship to 
explicitly write the comparison to zero:

  overflow = (4608u * 1024 * 1024) != 0 ?  4608u * 1024 * 1024 : 0;  // OK, no 
UB, no warning

The second expression

  results.reason = (actions & _UA_SEARCH_PHASE) ? ... : ...

should IMHO also be written as

  results.reason = (actions & _UA_SEARCH_PHASE) != 0 ? ... : ...

but I think you would get good support if you proposed that the compiler should 
treat `&` as a "logical operator" in this specific case. This just means 
changing line 9312 to

  return OP->isComparisonOp() || OP->isLogicalOp() || (OP->getOpcode() == 
BO_And);


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D147844/new/

https://reviews.llvm.org/D147844

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [ValueTracking] Simplify uaddo pattern (PR #65910)

[Yingwei Zheng via cfe-commits]

dtcxzyw wrote:

Ping.


https://github.com/llvm/llvm-project/pull/65910
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang-tools-extra] [ValueTracking] Simplify uaddo pattern (PR #65910)

[Yingwei Zheng via cfe-commits]

dtcxzyw wrote:

Ping.


https://github.com/llvm/llvm-project/pull/65910
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D159436: [clang-tidy] Add support for optional parameters in config.

[Piotr Zegar via Phabricator via cfe-commits]

PiotrZSL accepted this revision.
PiotrZSL added a comment.
This revision is now accepted and ready to land.

Except few nits in documentation looks fine. Probably one day in future we can 
add same for bools.
Make sure that documentation of impacted checks is correct, so it says that 
argument is positive integer, or none to force default value.




Comment at: clang-tools-extra/clang-tidy/readability/FunctionSizeCheck.h:20-34
+///   * `LineThreshold` - flag functions exceeding this number of lines. This
+/// parameter is disabled by default.
 ///   * `StatementThreshold` - flag functions exceeding this number of
 /// statements. This may differ significantly from the number of lines for
 /// macro-heavy code. The default is `800`.
 ///   * `BranchThreshold` - flag functions exceeding this number of control
+/// statements. This parameter is disabled by default.

Those options documentation could be removed from this file. Only keep check 
description.



Comment at: clang-tools-extra/docs/ReleaseNotes.rst:116-118
+- Added support for optional parameters. Parameters that previously used -1 to 
disable
+  their effect can now be set to `none`, `null`, or left empty to get the same
+  behaviour.

I'm not sure if this release notes entry is needed. Maybe this change should be 
documented per changed check basic.



Comment at: 
clang-tools-extra/docs/clang-tidy/checks/readability/function-size.rst:15
 
-   Flag functions exceeding this number of lines. The default is `-1` (ignore
-   the number of lines).
+   Flag functions exceeding this number of lines. This parameter is disabled
+   by default.

Leave previous but change -1 into none.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D159436/new/

https://reviews.llvm.org/D159436

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D154576: [RISCV] RISCV vector calling convention (1/2)

[Wang Pengcheng via Phabricator via cfe-commits]

wangpc added inline comments.



Comment at: llvm/lib/Target/RISCV/RISCVCallingConv.td:52
 // Same as CSR_Interrupt, but including all 64-bit FP registers.
 def CSR_XLEN_F64_Interrupt: CalleeSavedRegs<(add CSR_Interrupt,
  (sequence "F%u_D", 0, 31))>;

Should we add CSRs for interrupt functions? And Should we save `vtype`, 
`vstart`, `vxrm`, `vxsat`, etc. registers?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D154576/new/

https://reviews.llvm.org/D154576

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [RISCV] Install sifive_vector.h to riscv-resource-headers (PR #66330)

[Brandon Wu via cfe-commits]

https://github.com/4vtomat resolved 
https://github.com/llvm/llvm-project/pull/66330
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [RISCV] Install sifive_vector.h to riscv-resource-headers (PR #66330)

[Brandon Wu via cfe-commits]

https://github.com/4vtomat updated 
https://github.com/llvm/llvm-project/pull/66330

>From 8eb805beb037ba824fdf01dd8528b797a80de7ad Mon Sep 17 00:00:00 2001
From: 4vtomat 
Date: Wed, 13 Sep 2023 22:59:30 -0700
Subject: [PATCH 1/3] [RISCV] Install sifive_vector.h to riscv-resource-headers

---
 clang/lib/Headers/CMakeLists.txt | 5 +
 1 file changed, 5 insertions(+)

diff --git a/clang/lib/Headers/CMakeLists.txt b/clang/lib/Headers/CMakeLists.txt
index 334bc13799409bb..1868357b2eb4b7a 100644
--- a/clang/lib/Headers/CMakeLists.txt
+++ b/clang/lib/Headers/CMakeLists.txt
@@ -623,6 +623,11 @@ install(
   EXCLUDE_FROM_ALL
   COMPONENT riscv-resource-headers)
 
+install(
+  FILES ${sifive_files}
+  DESTINATION ${header_install_dir}
+  COMPONENT riscv-resource-headers)
+
 install(
   FILES ${systemz_files}
   DESTINATION ${header_install_dir}

>From 99857a112e1af6997cc93e7e692a3089f65156f3 Mon Sep 17 00:00:00 2001
From: 4vtomat 
Date: Thu, 14 Sep 2023 09:17:00 -0700
Subject: [PATCH 2/3] fixup! [RISCV] Install sifive_vector.h to
 riscv-resource-headers

---
 clang/test/Headers/riscv-sifive-header.c | 6 ++
 1 file changed, 6 insertions(+)
 create mode 100644 clang/test/Headers/riscv-sifive-header.c

diff --git a/clang/test/Headers/riscv-sifive-header.c 
b/clang/test/Headers/riscv-sifive-header.c
new file mode 100644
index 000..d90f4990c49f7a9
--- /dev/null
+++ b/clang/test/Headers/riscv-sifive-header.c
@@ -0,0 +1,6 @@
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v %s
+// REQUIRES: riscv-registered-target
+
+// expected-no-diagnostics
+
+#include 

>From a8af2ed9859eb8fe875f2efc4c7e56bb5a1e19cc Mon Sep 17 00:00:00 2001
From: 4vtomat 
Date: Mon, 18 Sep 2023 22:12:44 -0700
Subject: [PATCH 3/3] fixup! [RISCV] Install sifive_vector.h to
 riscv-resource-headers

---
 clang/lib/Headers/CMakeLists.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/clang/lib/Headers/CMakeLists.txt b/clang/lib/Headers/CMakeLists.txt
index 1868357b2eb4b7a..8e3bbdeb1d16b60 100644
--- a/clang/lib/Headers/CMakeLists.txt
+++ b/clang/lib/Headers/CMakeLists.txt
@@ -477,7 +477,7 @@ add_header_target("loongarch-resource-headers" 
"${loongarch_files}")
 add_header_target("mips-resource-headers" "${mips_msa_files}")
 add_header_target("ppc-resource-headers" "${ppc_files};${ppc_wrapper_files}")
 add_header_target("ppc-htm-resource-headers" "${ppc_htm_files}")
-add_header_target("riscv-resource-headers" 
"${riscv_files};${riscv_generated_files}")
+add_header_target("riscv-resource-headers" 
"${riscv_files};${riscv_generated_files};${sifive_files}")
 add_header_target("systemz-resource-headers" "${systemz_files}")
 add_header_target("ve-resource-headers" "${ve_files}")
 add_header_target("webassembly-resource-headers" "${webassembly_files}")

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [RISCV] Install sifive_vector.h to riscv-resource-headers (PR #66330)

[Brandon Wu via cfe-commits]

https://github.com/4vtomat edited 
https://github.com/llvm/llvm-project/pull/66330
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [RISCV] Install sifive_vector.h to riscv-resource-headers (PR #66330)

[Brandon Wu via cfe-commits]

@@ -623,6 +623,11 @@ install(
   EXCLUDE_FROM_ALL
   COMPONENT riscv-resource-headers)
 
+install(

4vtomat wrote:

Yes, we are missing from add_header_target("riscv-resource-headers" 
"${riscv_files};${riscv_generated_files}").
I think we can just put sifive_vector.h in riscv_files, I'll have another patch 
for it, thanks!

https://github.com/llvm/llvm-project/pull/66330
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D159483: [Modules] Add a flag to control builtin headers being in system modules

[Ian Anderson via Phabricator via cfe-commits]

iana marked an inline comment as done.
iana added inline comments.



Comment at: clang/include/clang/Basic/LangOptions.def:176
 COMPATIBLE_LANGOPT(CPlusPlusModules, 1, 0, "C++ modules syntax")
+LANGOPT(BuiltinHeadersInSystemModules, 1, 0, "builtin headers belong to system 
modules, and _Builtin_ modules are ignored for cstdlib headers")
 BENIGN_ENUM_LANGOPT(CompilingModule, CompilingModuleKind, 3, CMK_None,

vsapsai wrote:
> Why not to make the default value `true` to preserve the old behavior and 
> switch on the new behavior in the driver? It's not a veiled way to force you 
> to change it, I'm genuinely curious and want to consider pro/cons of various 
> alternatives.
I did it this way because the current behavior is nonintuitive and just kind of 
bizarre. I had no idea that `Darwin.C.stdint` also included the compiler's 
stdint.h, and that the OS's stdint.h was treated textually. I thought it would 
be better for the default behavior to be for modules to behave consistently and 
obviously, not have secret special cases, and be usable with C++.

The con of course is that if anyone besides Apple depends on the current 
behavior they're going to either have to fix their OS modules or add this flag. 
I'm not sure if other platforms have a convenient driver class where they can 
do that.



Comment at: clang/lib/Driver/ToolChains/Darwin.cpp:2860-2861
+return SDKVersion >= VersionTuple(99U);
+  default:
+return true;
+  }

vsapsai wrote:
> Another question regarding defaults. Doesn't look consistent with your 
> position
> > [...] I thought it safer to default to the current behavior which is false.
> 
> Personally I gravitate towards `false` for extra safety but it's not a strong 
> opinion.
default only covers DriverKit, which can be true. Any future platforms should 
also default to true.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D159483/new/

https://reviews.llvm.org/D159483

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Workaround for MSVC ARM64 build performance regression (PR #65215)

[Tobias Hieta via cfe-commits]

https://github.com/tru approved this pull request.

LGTM

https://github.com/llvm/llvm-project/pull/65215
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D159064: [Modules] Make clang modules for the C standard library headers

[Ian Anderson via Phabricator via cfe-commits]

iana added a comment.

In D159064#4647709 , @vsapsai wrote:

> Still going through the patch and through the discussion. But wanted to ask 
> if there are any complications in reverting this change? For libc++ we've 
> discussed that we don't really know the perf impact of multiple small 
> modules. So I want to make sure we aren't making decisions that make it 
> impossible to revert the change later.

I don't think we can really revert the C stdlib headers each being in their own 
module; there isn't any other way around the module layering maze. libc++ is a 
bit different, it's not just the C stdlib headers that are each in their own 
module, it's every one of the ~1000 headers right now. I have misgivings about 
that, I made D155141  to reduce that closer 
to 150 modules instead of 1000 but it was poorly received. I might try to 
address some of the concerns after I get this one worked through.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D159064/new/

https://reviews.llvm.org/D159064

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D154576: [RISCV] RISCV vector calling convention (1/2)

[Brandon Wu via Phabricator via cfe-commits]

4vtomat updated this revision to Diff 556995.
4vtomat added a comment.

Add VRM2, VRM4 and VRM8 to CalleeSaved register list.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D154576/new/

https://reviews.llvm.org/D154576

Files:
  clang/include/clang/Basic/Attr.td
  clang/include/clang/Basic/AttrDocs.td
  clang/include/clang/Basic/Specifiers.h
  clang/lib/AST/ItaniumMangle.cpp
  clang/lib/AST/Type.cpp
  clang/lib/AST/TypePrinter.cpp
  clang/lib/Basic/Targets/RISCV.cpp
  clang/lib/Basic/Targets/RISCV.h
  clang/lib/CodeGen/CGCall.cpp
  clang/lib/Sema/SemaDeclAttr.cpp
  clang/lib/Sema/SemaType.cpp
  clang/test/CodeGen/RISCV/riscv-vector-callingconv-llvm-ir.c
  clang/test/CodeGen/RISCV/riscv-vector-callingconv.cpp
  llvm/include/llvm/AsmParser/LLToken.h
  llvm/include/llvm/IR/CallingConv.h
  llvm/lib/AsmParser/LLLexer.cpp
  llvm/lib/AsmParser/LLParser.cpp
  llvm/lib/IR/AsmWriter.cpp
  llvm/lib/Target/RISCV/RISCVCallingConv.td
  llvm/lib/Target/RISCV/RISCVFrameLowering.cpp
  llvm/lib/Target/RISCV/RISCVISelLowering.cpp
  llvm/lib/Target/RISCV/RISCVRegisterInfo.cpp
  llvm/test/CodeGen/RISCV/rvv/callee-saved-regs.ll

Index: llvm/test/CodeGen/RISCV/rvv/callee-saved-regs.ll
===
--- /dev/null
+++ llvm/test/CodeGen/RISCV/rvv/callee-saved-regs.ll
@@ -0,0 +1,95 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv32 -mattr=+m -mattr=+v -O2 < %s \
+; RUN:| FileCheck --check-prefix=SPILL-O2 %s
+
+define  @test_vector_std( %va) nounwind {
+; SPILL-O2-LABEL: test_vector_std:
+; SPILL-O2:   # %bb.0: # %entry
+; SPILL-O2-NEXT:addi sp, sp, -16
+; SPILL-O2-NEXT:csrr a0, vlenb
+; SPILL-O2-NEXT:slli a0, a0, 1
+; SPILL-O2-NEXT:sub sp, sp, a0
+; SPILL-O2-NEXT:addi a0, sp, 16
+; SPILL-O2-NEXT:vs1r.v v8, (a0) # Unknown-size Folded Spill
+; SPILL-O2-NEXT:#APP
+; SPILL-O2-NEXT:#NO_APP
+; SPILL-O2-NEXT:vl1r.v v8, (a0) # Unknown-size Folded Reload
+; SPILL-O2-NEXT:csrr a0, vlenb
+; SPILL-O2-NEXT:slli a0, a0, 1
+; SPILL-O2-NEXT:add sp, sp, a0
+; SPILL-O2-NEXT:addi sp, sp, 16
+; SPILL-O2-NEXT:ret
+entry:
+  call void asm sideeffect "",
+  "~{v0},~{v1},~{v2},~{v3},~{v4},~{v5},~{v6},~{v7},~{v8},~{v9},~{v10},~{v11},~{v12},~{v13},~{v14},~{v15},~{v16},~{v17},~{v18},~{v19},~{v20},~{v21},~{v22},~{v23},~{v24},~{v25},~{v26},~{v27},~{v28},~{v29},~{v30},~{v31}"()
+
+  ret  %va
+}
+
+define riscv_vector_cc  @test_vector_callee( %va) nounwind {
+; SPILL-O2-LABEL: test_vector_callee:
+; SPILL-O2:   # %bb.0: # %entry
+; SPILL-O2-NEXT:addi sp, sp, -16
+; SPILL-O2-NEXT:csrr a0, vlenb
+; SPILL-O2-NEXT:slli a0, a0, 4
+; SPILL-O2-NEXT:sub sp, sp, a0
+; SPILL-O2-NEXT:csrr a0, vlenb
+; SPILL-O2-NEXT:slli a1, a0, 4
+; SPILL-O2-NEXT:sub a0, a1, a0
+; SPILL-O2-NEXT:add a0, sp, a0
+; SPILL-O2-NEXT:addi a0, a0, 16
+; SPILL-O2-NEXT:vs1r.v v1, (a0) # Unknown-size Folded Spill
+; SPILL-O2-NEXT:csrr a0, vlenb
+; SPILL-O2-NEXT:li a1, 13
+; SPILL-O2-NEXT:mul a0, a0, a1
+; SPILL-O2-NEXT:add a0, sp, a0
+; SPILL-O2-NEXT:addi a0, a0, 16
+; SPILL-O2-NEXT:vs2r.v v2, (a0) # Unknown-size Folded Spill
+; SPILL-O2-NEXT:csrr a0, vlenb
+; SPILL-O2-NEXT:slli a1, a0, 3
+; SPILL-O2-NEXT:add a0, a1, a0
+; SPILL-O2-NEXT:add a0, sp, a0
+; SPILL-O2-NEXT:addi a0, a0, 16
+; SPILL-O2-NEXT:vs4r.v v4, (a0) # Unknown-size Folded Spill
+; SPILL-O2-NEXT:csrr a0, vlenb
+; SPILL-O2-NEXT:add a0, sp, a0
+; SPILL-O2-NEXT:addi a0, a0, 16
+; SPILL-O2-NEXT:vs8r.v v24, (a0) # Unknown-size Folded Spill
+; SPILL-O2-NEXT:addi a0, sp, 16
+; SPILL-O2-NEXT:vs1r.v v8, (a0) # Unknown-size Folded Spill
+; SPILL-O2-NEXT:#APP
+; SPILL-O2-NEXT:#NO_APP
+; SPILL-O2-NEXT:vl1r.v v8, (a0) # Unknown-size Folded Reload
+; SPILL-O2-NEXT:csrr a0, vlenb
+; SPILL-O2-NEXT:slli a1, a0, 4
+; SPILL-O2-NEXT:sub a0, a1, a0
+; SPILL-O2-NEXT:add a0, sp, a0
+; SPILL-O2-NEXT:addi a0, a0, 16
+; SPILL-O2-NEXT:vl1r.v v1, (a0) # Unknown-size Folded Reload
+; SPILL-O2-NEXT:csrr a0, vlenb
+; SPILL-O2-NEXT:li a1, 13
+; SPILL-O2-NEXT:mul a0, a0, a1
+; SPILL-O2-NEXT:add a0, sp, a0
+; SPILL-O2-NEXT:addi a0, a0, 16
+; SPILL-O2-NEXT:vl2r.v v2, (a0) # Unknown-size Folded Reload
+; SPILL-O2-NEXT:csrr a0, vlenb
+; SPILL-O2-NEXT:slli a1, a0, 3
+; SPILL-O2-NEXT:add a0, a1, a0
+; SPILL-O2-NEXT:add a0, sp, a0
+; SPILL-O2-NEXT:addi a0, a0, 16
+; SPILL-O2-NEXT:vl4r.v v4, (a0) # Unknown-size Folded Reload
+; SPILL-O2-NEXT:csrr a0, vlenb
+; SPILL-O2-NEXT:add a0, sp, a0
+; SPILL-O2-NEXT:addi a0, a0, 16
+; SPILL-O2-NEXT:vl8r.v v24, (a0) # Unknown-size Folded Reload
+; SPILL-O2-NEXT:csrr a0, vlenb
+; SPILL-O2-NEXT:slli a0, a0, 4
+; SPILL-O2-NEXT:add sp, sp, a0
+; SPILL-O2-NEXT:addi sp, sp, 16
+; SPILL-O2-NEXT:   

[clang] Introduce paged vector (PR #66430)

[Jakub Kuderski via cfe-commits]

@@ -0,0 +1,322 @@
+//===- llvm/ADT/PagedVector.h - 'Lazyly allocated' vectors *- 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
+//
+//===--===//
+//
+// This file defines the PagedVector class.
+//
+//===--===//
+#ifndef LLVM_ADT_PAGEDVECTOR_H
+#define LLVM_ADT_PAGEDVECTOR_H
+
+#include "llvm/ADT/PointerIntPair.h"
+#include "llvm/Support/Allocator.h"
+#include 
+#include 
+#include 
+
+namespace llvm {
+// A vector that allocates memory in pages.
+// Order is kept, but memory is allocated only when one element of the page is
+// accessed. This introduces a level of indirection, but it is useful when you
+// have a sparsely initialised vector where the full size is allocated upfront
+// with the default constructor and elements are initialised later, on first
+// access.
+//
+// Notice that this does not have iterators, because if you
+// have iterators it probably means you are going to touch
+// all the memory in any case, so better use a std::vector in
+// the first place.
+//
+// Pages are allocated in SLAB_SIZE chunks, using the BumpPtrAllocator.
+template 
+class PagedVector {
+  static_assert(PAGE_SIZE > 0, "PAGE_SIZE must be greater than 0. Most likely "
+   "you want it to be greater than 16.");
+  // The actual number of element in the vector which can be accessed.
+  std::size_t Size = 0;
+
+  // The position of the initial element of the page in the Data vector.
+  // Pages are allocated contiguously in the Data vector.
+  mutable std::vector PageToDataIdx;
+  // Actual page data. All the page elements are added to this vector on the
+  // first access of any of the elements of the page. Elements default
+  // constructed and elements of the page are stored contiguously. The order of
+  // the elements however depends on the order of access of the pages.
+  PointerIntPair Allocator;
+
+  constexpr static uintptr_t InvalidPage = SIZE_MAX;
+
+public:
+  using value_type = T;
+
+  // Default constructor. We build our own allocator.
+  PagedVector() : Allocator(new BumpPtrAllocator, true) {}
+  PagedVector(BumpPtrAllocator *A) : Allocator(A, false) {}
+
+  ~PagedVector() {
+// If we own the allocator, delete it.
+if (Allocator.getInt() == true)
+  delete Allocator.getPointer();
+  }
+
+  // Lookup an element at position i.
+  // If the associated page is not filled, it will be filled with default
+  // constructed elements. If the associated page is filled, return the 
element.
+  T &operator[](std::size_t Index) const {
+assert(Index < Size);
+assert(Index / PAGE_SIZE < PageToDataIdx.size());
+uintptr_t &PagePtr = PageToDataIdx[Index / PAGE_SIZE];
+// If the page was not yet allocated, allocate it.
+if (PagePtr == InvalidPage) {
+  T *NewPagePtr = Allocator.getPointer()->template Allocate(PAGE_SIZE);
+  // We need to invoke the default constructor on all the elements of the
+  // page.
+  for (std::size_t I = 0; I < PAGE_SIZE; ++I)
+new (NewPagePtr + I) T();
+
+  PagePtr = reinterpret_cast(NewPagePtr);
+}
+// Dereference the element in the page.
+return *((Index % PAGE_SIZE) + reinterpret_cast(PagePtr));
+  }
+
+  // Return the capacity of the vector. I.e. the maximum size it can be 
expanded
+  // to with the resize method without allocating more pages.
+  [[nodiscard]] std::size_t capacity() const {
+return PageToDataIdx.size() * PAGE_SIZE;
+  }
+
+  // Return the size of the vector. I.e. the maximum index that can be
+  // accessed, i.e. the maximum value which was used as argument of the
+  // resize method.
+  [[nodiscard]] std::size_t size() const { return Size; }
+
+  // Expands the vector to the given NewSize number of elements.
+  // If the vector was smaller, allocates new pages as needed.
+  // It should be called only with NewSize >= Size.
+  void resize(std::size_t NewSize) {
+// Handle shrink case: delete the pages and update the size.
+if (NewSize < Size) {
+  std::size_t NewLastPage = (NewSize - 1) / PAGE_SIZE;
+  for (std::size_t I = NewLastPage + 1; I < PageToDataIdx.size(); ++I) {
+uintptr_t PagePtr = PageToDataIdx[I];
+if (PagePtr == InvalidPage)
+  continue;
+T *Page = reinterpret_cast(PagePtr);
+// We need to invoke the destructor on all the elements of the page.
+for (std::size_t J = 0; J < PAGE_SIZE; ++J)
+  Page[J].~T();
+Allocator.getPointer()->Deallocate(Page);
+  }
+  // Delete the extra ones in the new last page.
+  uintptr_t PagePtr = PageToDataIdx[NewLastPage];
+  if (PagePtr != InvalidPage) {
+T *Page = reinterpret_cast(PagePtr)

[clang] Introduce paged vector (PR #66430)

[Jakub Kuderski via cfe-commits]

@@ -0,0 +1,322 @@
+//===- llvm/ADT/PagedVector.h - 'Lazyly allocated' vectors *- 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
+//
+//===--===//
+//
+// This file defines the PagedVector class.
+//
+//===--===//
+#ifndef LLVM_ADT_PAGEDVECTOR_H
+#define LLVM_ADT_PAGEDVECTOR_H
+
+#include "llvm/ADT/PointerIntPair.h"
+#include "llvm/Support/Allocator.h"
+#include 
+#include 
+#include 
+
+namespace llvm {
+// A vector that allocates memory in pages.
+// Order is kept, but memory is allocated only when one element of the page is
+// accessed. This introduces a level of indirection, but it is useful when you
+// have a sparsely initialised vector where the full size is allocated upfront
+// with the default constructor and elements are initialised later, on first
+// access.
+//
+// Notice that this does not have iterators, because if you
+// have iterators it probably means you are going to touch
+// all the memory in any case, so better use a std::vector in
+// the first place.
+//
+// Pages are allocated in SLAB_SIZE chunks, using the BumpPtrAllocator.
+template 
+class PagedVector {
+  static_assert(PAGE_SIZE > 0, "PAGE_SIZE must be greater than 0. Most likely "
+   "you want it to be greater than 16.");
+  // The actual number of element in the vector which can be accessed.
+  std::size_t Size = 0;
+
+  // The position of the initial element of the page in the Data vector.
+  // Pages are allocated contiguously in the Data vector.
+  mutable std::vector PageToDataIdx;
+  // Actual page data. All the page elements are added to this vector on the
+  // first access of any of the elements of the page. Elements default
+  // constructed and elements of the page are stored contiguously. The order of
+  // the elements however depends on the order of access of the pages.
+  PointerIntPair Allocator;
+
+  constexpr static uintptr_t InvalidPage = SIZE_MAX;
+
+public:
+  using value_type = T;
+
+  // Default constructor. We build our own allocator.
+  PagedVector() : Allocator(new BumpPtrAllocator, true) {}
+  PagedVector(BumpPtrAllocator *A) : Allocator(A, false) {}
+
+  ~PagedVector() {
+// If we own the allocator, delete it.
+if (Allocator.getInt() == true)
+  delete Allocator.getPointer();
+  }
+
+  // Lookup an element at position i.

kuhar wrote:

```suggestion
  // Look up an element at position `Index`.
```

https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Introduce paged vector (PR #66430)

[Jakub Kuderski via cfe-commits]

@@ -0,0 +1,322 @@
+//===- llvm/ADT/PagedVector.h - 'Lazyly allocated' vectors *- 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
+//
+//===--===//
+//
+// This file defines the PagedVector class.
+//
+//===--===//
+#ifndef LLVM_ADT_PAGEDVECTOR_H
+#define LLVM_ADT_PAGEDVECTOR_H
+
+#include "llvm/ADT/PointerIntPair.h"
+#include "llvm/Support/Allocator.h"
+#include 
+#include 
+#include 
+
+namespace llvm {
+// A vector that allocates memory in pages.
+// Order is kept, but memory is allocated only when one element of the page is
+// accessed. This introduces a level of indirection, but it is useful when you
+// have a sparsely initialised vector where the full size is allocated upfront
+// with the default constructor and elements are initialised later, on first
+// access.
+//
+// Notice that this does not have iterators, because if you
+// have iterators it probably means you are going to touch
+// all the memory in any case, so better use a std::vector in
+// the first place.
+//
+// Pages are allocated in SLAB_SIZE chunks, using the BumpPtrAllocator.
+template 
+class PagedVector {
+  static_assert(PAGE_SIZE > 0, "PAGE_SIZE must be greater than 0. Most likely "
+   "you want it to be greater than 16.");
+  // The actual number of element in the vector which can be accessed.
+  std::size_t Size = 0;
+
+  // The position of the initial element of the page in the Data vector.
+  // Pages are allocated contiguously in the Data vector.
+  mutable std::vector PageToDataIdx;
+  // Actual page data. All the page elements are added to this vector on the
+  // first access of any of the elements of the page. Elements default
+  // constructed and elements of the page are stored contiguously. The order of
+  // the elements however depends on the order of access of the pages.
+  PointerIntPair Allocator;
+
+  constexpr static uintptr_t InvalidPage = SIZE_MAX;
+
+public:
+  using value_type = T;
+
+  // Default constructor. We build our own allocator.
+  PagedVector() : Allocator(new BumpPtrAllocator, true) {}
+  PagedVector(BumpPtrAllocator *A) : Allocator(A, false) {}
+
+  ~PagedVector() {
+// If we own the allocator, delete it.
+if (Allocator.getInt() == true)
+  delete Allocator.getPointer();
+  }
+
+  // Lookup an element at position i.
+  // If the associated page is not filled, it will be filled with default
+  // constructed elements. If the associated page is filled, return the 
element.
+  T &operator[](std::size_t Index) const {
+assert(Index < Size);
+assert(Index / PAGE_SIZE < PageToDataIdx.size());
+uintptr_t &PagePtr = PageToDataIdx[Index / PAGE_SIZE];
+// If the page was not yet allocated, allocate it.
+if (PagePtr == InvalidPage) {
+  T *NewPagePtr = Allocator.getPointer()->template Allocate(PAGE_SIZE);
+  // We need to invoke the default constructor on all the elements of the
+  // page.
+  for (std::size_t I = 0; I < PAGE_SIZE; ++I)
+new (NewPagePtr + I) T();
+
+  PagePtr = reinterpret_cast(NewPagePtr);
+}
+// Dereference the element in the page.
+return *((Index % PAGE_SIZE) + reinterpret_cast(PagePtr));
+  }
+
+  // Return the capacity of the vector. I.e. the maximum size it can be 
expanded
+  // to with the resize method without allocating more pages.
+  [[nodiscard]] std::size_t capacity() const {
+return PageToDataIdx.size() * PAGE_SIZE;
+  }
+
+  // Return the size of the vector. I.e. the maximum index that can be
+  // accessed, i.e. the maximum value which was used as argument of the
+  // resize method.
+  [[nodiscard]] std::size_t size() const { return Size; }
+
+  // Expands the vector to the given NewSize number of elements.
+  // If the vector was smaller, allocates new pages as needed.
+  // It should be called only with NewSize >= Size.
+  void resize(std::size_t NewSize) {
+// Handle shrink case: delete the pages and update the size.
+if (NewSize < Size) {
+  std::size_t NewLastPage = (NewSize - 1) / PAGE_SIZE;
+  for (std::size_t I = NewLastPage + 1; I < PageToDataIdx.size(); ++I) {
+uintptr_t PagePtr = PageToDataIdx[I];
+if (PagePtr == InvalidPage)
+  continue;
+T *Page = reinterpret_cast(PagePtr);
+// We need to invoke the destructor on all the elements of the page.
+for (std::size_t J = 0; J < PAGE_SIZE; ++J)
+  Page[J].~T();
+Allocator.getPointer()->Deallocate(Page);
+  }
+  // Delete the extra ones in the new last page.
+  uintptr_t PagePtr = PageToDataIdx[NewLastPage];
+  if (PagePtr != InvalidPage) {
+T *Page = reinterpret_cast(PagePtr)

[clang] Introduce paged vector (PR #66430)

[Jakub Kuderski via cfe-commits]

@@ -0,0 +1,322 @@
+//===- llvm/ADT/PagedVector.h - 'Lazyly allocated' vectors *- 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
+//
+//===--===//
+//
+// This file defines the PagedVector class.
+//
+//===--===//
+#ifndef LLVM_ADT_PAGEDVECTOR_H
+#define LLVM_ADT_PAGEDVECTOR_H
+
+#include "llvm/ADT/PointerIntPair.h"
+#include "llvm/Support/Allocator.h"
+#include 
+#include 
+#include 
+
+namespace llvm {
+// A vector that allocates memory in pages.
+// Order is kept, but memory is allocated only when one element of the page is
+// accessed. This introduces a level of indirection, but it is useful when you
+// have a sparsely initialised vector where the full size is allocated upfront
+// with the default constructor and elements are initialised later, on first
+// access.
+//
+// Notice that this does not have iterators, because if you
+// have iterators it probably means you are going to touch
+// all the memory in any case, so better use a std::vector in
+// the first place.

kuhar wrote:

nit: could you reflow this comment? The lines wrap well before the 80 character 
line limit

https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Introduce paged vector (PR #66430)

[Jakub Kuderski via cfe-commits]

@@ -0,0 +1,322 @@
+//===- llvm/ADT/PagedVector.h - 'Lazyly allocated' vectors *- 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
+//
+//===--===//
+//
+// This file defines the PagedVector class.
+//
+//===--===//
+#ifndef LLVM_ADT_PAGEDVECTOR_H
+#define LLVM_ADT_PAGEDVECTOR_H
+
+#include "llvm/ADT/PointerIntPair.h"
+#include "llvm/Support/Allocator.h"
+#include 
+#include 
+#include 
+
+namespace llvm {
+// A vector that allocates memory in pages.
+// Order is kept, but memory is allocated only when one element of the page is
+// accessed. This introduces a level of indirection, but it is useful when you
+// have a sparsely initialised vector where the full size is allocated upfront
+// with the default constructor and elements are initialised later, on first
+// access.
+//
+// Notice that this does not have iterators, because if you
+// have iterators it probably means you are going to touch
+// all the memory in any case, so better use a std::vector in
+// the first place.
+//
+// Pages are allocated in SLAB_SIZE chunks, using the BumpPtrAllocator.
+template 
+class PagedVector {
+  static_assert(PAGE_SIZE > 0, "PAGE_SIZE must be greater than 0. Most likely "

kuhar wrote:

If we assert that page size is > 2, we can also use `IntPointerPair` for 
invalid pages, no? Maybe just enforce the minimum size of 16? This would allow 
us to get rid of those reinterpret casts.

https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Introduce paged vector (PR #66430)

[Jakub Kuderski via cfe-commits]

https://github.com/kuhar commented:

Thanks for all the fixes, this is looking very good. Did another pass and left 
some local suggestions.

https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Introduce paged vector (PR #66430)

[Jakub Kuderski via cfe-commits]

@@ -0,0 +1,322 @@
+//===- llvm/ADT/PagedVector.h - 'Lazyly allocated' vectors *- 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
+//
+//===--===//
+//
+// This file defines the PagedVector class.
+//
+//===--===//
+#ifndef LLVM_ADT_PAGEDVECTOR_H
+#define LLVM_ADT_PAGEDVECTOR_H
+
+#include "llvm/ADT/PointerIntPair.h"
+#include "llvm/Support/Allocator.h"
+#include 
+#include 
+#include 
+
+namespace llvm {
+// A vector that allocates memory in pages.
+// Order is kept, but memory is allocated only when one element of the page is
+// accessed. This introduces a level of indirection, but it is useful when you
+// have a sparsely initialised vector where the full size is allocated upfront
+// with the default constructor and elements are initialised later, on first
+// access.
+//
+// Notice that this does not have iterators, because if you
+// have iterators it probably means you are going to touch
+// all the memory in any case, so better use a std::vector in
+// the first place.
+//
+// Pages are allocated in SLAB_SIZE chunks, using the BumpPtrAllocator.
+template 
+class PagedVector {
+  static_assert(PAGE_SIZE > 0, "PAGE_SIZE must be greater than 0. Most likely "
+   "you want it to be greater than 16.");
+  // The actual number of element in the vector which can be accessed.
+  std::size_t Size = 0;
+
+  // The position of the initial element of the page in the Data vector.
+  // Pages are allocated contiguously in the Data vector.
+  mutable std::vector PageToDataIdx;
+  // Actual page data. All the page elements are added to this vector on the
+  // first access of any of the elements of the page. Elements default
+  // constructed and elements of the page are stored contiguously. The order of
+  // the elements however depends on the order of access of the pages.
+  PointerIntPair Allocator;
+
+  constexpr static uintptr_t InvalidPage = SIZE_MAX;
+
+public:
+  using value_type = T;
+
+  // Default constructor. We build our own allocator.
+  PagedVector() : Allocator(new BumpPtrAllocator, true) {}
+  PagedVector(BumpPtrAllocator *A) : Allocator(A, false) {}
+
+  ~PagedVector() {
+// If we own the allocator, delete it.
+if (Allocator.getInt() == true)
+  delete Allocator.getPointer();
+  }
+
+  // Lookup an element at position i.
+  // If the associated page is not filled, it will be filled with default
+  // constructed elements. If the associated page is filled, return the 
element.
+  T &operator[](std::size_t Index) const {
+assert(Index < Size);
+assert(Index / PAGE_SIZE < PageToDataIdx.size());
+uintptr_t &PagePtr = PageToDataIdx[Index / PAGE_SIZE];
+// If the page was not yet allocated, allocate it.
+if (PagePtr == InvalidPage) {
+  T *NewPagePtr = Allocator.getPointer()->template Allocate(PAGE_SIZE);
+  // We need to invoke the default constructor on all the elements of the
+  // page.
+  for (std::size_t I = 0; I < PAGE_SIZE; ++I)
+new (NewPagePtr + I) T();
+
+  PagePtr = reinterpret_cast(NewPagePtr);
+}
+// Dereference the element in the page.
+return *((Index % PAGE_SIZE) + reinterpret_cast(PagePtr));
+  }
+
+  // Return the capacity of the vector. I.e. the maximum size it can be 
expanded
+  // to with the resize method without allocating more pages.
+  [[nodiscard]] std::size_t capacity() const {
+return PageToDataIdx.size() * PAGE_SIZE;
+  }
+
+  // Return the size of the vector. I.e. the maximum index that can be
+  // accessed, i.e. the maximum value which was used as argument of the
+  // resize method.
+  [[nodiscard]] std::size_t size() const { return Size; }
+
+  // Expands the vector to the given NewSize number of elements.
+  // If the vector was smaller, allocates new pages as needed.
+  // It should be called only with NewSize >= Size.
+  void resize(std::size_t NewSize) {
+// Handle shrink case: delete the pages and update the size.
+if (NewSize < Size) {
+  std::size_t NewLastPage = (NewSize - 1) / PAGE_SIZE;
+  for (std::size_t I = NewLastPage + 1; I < PageToDataIdx.size(); ++I) {
+uintptr_t PagePtr = PageToDataIdx[I];
+if (PagePtr == InvalidPage)
+  continue;
+T *Page = reinterpret_cast(PagePtr);
+// We need to invoke the destructor on all the elements of the page.
+for (std::size_t J = 0; J < PAGE_SIZE; ++J)
+  Page[J].~T();
+Allocator.getPointer()->Deallocate(Page);
+  }
+  // Delete the extra ones in the new last page.
+  uintptr_t PagePtr = PageToDataIdx[NewLastPage];
+  if (PagePtr != InvalidPage) {
+T *Page = reinterpret_cast(PagePtr)

[clang] Introduce paged vector (PR #66430)

[Jakub Kuderski via cfe-commits]

@@ -0,0 +1,322 @@
+//===- llvm/ADT/PagedVector.h - 'Lazyly allocated' vectors *- 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
+//
+//===--===//
+//
+// This file defines the PagedVector class.
+//
+//===--===//
+#ifndef LLVM_ADT_PAGEDVECTOR_H
+#define LLVM_ADT_PAGEDVECTOR_H
+
+#include "llvm/ADT/PointerIntPair.h"
+#include "llvm/Support/Allocator.h"
+#include 
+#include 
+#include 
+
+namespace llvm {
+// A vector that allocates memory in pages.
+// Order is kept, but memory is allocated only when one element of the page is
+// accessed. This introduces a level of indirection, but it is useful when you
+// have a sparsely initialised vector where the full size is allocated upfront
+// with the default constructor and elements are initialised later, on first
+// access.
+//
+// Notice that this does not have iterators, because if you
+// have iterators it probably means you are going to touch
+// all the memory in any case, so better use a std::vector in
+// the first place.
+//
+// Pages are allocated in SLAB_SIZE chunks, using the BumpPtrAllocator.
+template 
+class PagedVector {
+  static_assert(PAGE_SIZE > 0, "PAGE_SIZE must be greater than 0. Most likely "
+   "you want it to be greater than 16.");
+  // The actual number of element in the vector which can be accessed.
+  std::size_t Size = 0;
+
+  // The position of the initial element of the page in the Data vector.
+  // Pages are allocated contiguously in the Data vector.
+  mutable std::vector PageToDataIdx;
+  // Actual page data. All the page elements are added to this vector on the
+  // first access of any of the elements of the page. Elements default
+  // constructed and elements of the page are stored contiguously. The order of
+  // the elements however depends on the order of access of the pages.
+  PointerIntPair Allocator;
+
+  constexpr static uintptr_t InvalidPage = SIZE_MAX;
+
+public:
+  using value_type = T;
+
+  // Default constructor. We build our own allocator.

kuhar wrote:

Maybe explain the value of the `bool` in `Allocator`:
```suggestion
  // Default constructor. We build our own allocator and mark it as such with 
`true` in the second pair element.
```

https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Introduce paged vector (PR #66430)

[Jakub Kuderski via cfe-commits]

@@ -0,0 +1,322 @@
+//===- llvm/ADT/PagedVector.h - 'Lazyly allocated' vectors *- 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
+//
+//===--===//
+//
+// This file defines the PagedVector class.
+//
+//===--===//
+#ifndef LLVM_ADT_PAGEDVECTOR_H
+#define LLVM_ADT_PAGEDVECTOR_H
+
+#include "llvm/ADT/PointerIntPair.h"
+#include "llvm/Support/Allocator.h"
+#include 
+#include 
+#include 
+
+namespace llvm {
+// A vector that allocates memory in pages.
+// Order is kept, but memory is allocated only when one element of the page is
+// accessed. This introduces a level of indirection, but it is useful when you
+// have a sparsely initialised vector where the full size is allocated upfront
+// with the default constructor and elements are initialised later, on first
+// access.
+//
+// Notice that this does not have iterators, because if you
+// have iterators it probably means you are going to touch
+// all the memory in any case, so better use a std::vector in
+// the first place.
+//
+// Pages are allocated in SLAB_SIZE chunks, using the BumpPtrAllocator.
+template 

kuhar wrote:

nit: I looked at other code under ADT and believe that we use ThisCase for 
template arguments. Could you update the second argument to `PageSize`?

https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Introduce paged vector (PR #66430)

[Jakub Kuderski via cfe-commits]

@@ -1625,6 +1625,38 @@ SmallVector has grown a few other minor advantages over 
std::vector, causing
and is no longer "private to the implementation". A name like
``SmallVectorHeader`` might be more appropriate.
 
+.. _dss_pagedvector:
+
+llvm/ADT/PagedVector.h
+^^
+
+``PagedVector`` is a random access container that allocates
+(PageSize) elements of type Type when the first element of a page is accessed
+via the ``operator[]``.  This is useful for the case in which the number of
+elements is known in advance and their actual initialization is expensive and
+sparse so that it's only done lazily when the element is accessed. When the
+number of used pages is small significant memory savings can be achieved.
+
+The main advantage is that a ``PagedVector`` allows to delay the actual 
allocation
+of the page until it's needed, at the extra cost of one integer per page and 
one
+extra indirection when accessing elements with their positional index. 
+
+In order to maximise the memory footprint of this container, it's important to
+balance the PageSize so that it's not too small (otherwise the overhead of the
+integer per page might become too high) and not too big (otherwise the memory 
is
+wasted if the page is not fully used).
+
+Moreover, while retaining the oder of the elements based on their insertion

kuhar wrote:

```suggestion
Moreover, while retaining the order of the elements based on their insertion
```

https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [Clang][Coroutines] Improve GRO handling to better fit scopes and cleanups (PR #66706)

[Chuanqi Xu via cfe-commits]

ChuanqiXu9 wrote:

I remember that there is a defect that we may place the GRO on the coroutine 
frame. And my instinct reaction is that would this patch be covered by forcing 
GRO to not live on the coroutine frame? 

https://github.com/llvm/llvm-project/pull/66706
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang-tools-extra] [InstCombine] Simplify the pattern `a ne/eq (zext/sext (a ne/eq c))` (PR #65852)

[Yingwei Zheng via cfe-commits]

https://github.com/dtcxzyw resolved 
https://github.com/llvm/llvm-project/pull/65852
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [InstCombine] Simplify the pattern `a ne/eq (zext/sext (a ne/eq c))` (PR #65852)

[Yingwei Zheng via cfe-commits]

https://github.com/dtcxzyw resolved 
https://github.com/llvm/llvm-project/pull/65852
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [InstCombine] Simplify the pattern `a ne/eq (zext/sext (a ne/eq c))` (PR #65852)

[Yingwei Zheng via cfe-commits]

https://github.com/dtcxzyw resolved 
https://github.com/llvm/llvm-project/pull/65852
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang-tools-extra] [InstCombine] Simplify the pattern `a ne/eq (zext/sext (a ne/eq c))` (PR #65852)

[Yingwei Zheng via cfe-commits]

https://github.com/dtcxzyw resolved 
https://github.com/llvm/llvm-project/pull/65852
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang-tools-extra] [InstCombine] Simplify the pattern `a ne/eq (zext/sext (a ne/eq c))` (PR #65852)

[Yingwei Zheng via cfe-commits]

https://github.com/dtcxzyw updated 
https://github.com/llvm/llvm-project/pull/65852

>From d9d8bcbb98e8f5aecb9733329389d61a489bd731 Mon Sep 17 00:00:00 2001
From: Yingwei Zheng 
Date: Sat, 9 Sep 2023 23:07:29 +0800
Subject: [PATCH 1/6] [InstCombine] Simplify the pattern `a ne/eq (zext (a
 ne/eq c))`

---
 .../InstCombine/InstCombineCompares.cpp   |  62 ++
 .../test/Transforms/InstCombine/icmp-range.ll | 181 ++
 2 files changed, 243 insertions(+)

diff --git a/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp 
b/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
index 9fdc46fec631679..837b8e6d2619989 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
@@ -6309,7 +6309,69 @@ Instruction 
*InstCombinerImpl::foldICmpUsingBoolRange(ICmpInst &I) {
   Y->getType()->isIntOrIntVectorTy(1) && Pred == ICmpInst::ICMP_ULE)
 return BinaryOperator::CreateOr(Builder.CreateIsNull(X), Y);
 
+  ICmpInst::Predicate Pred1, Pred2;
   const APInt *C;
+  // icmp eq/ne X, (zext (icmp eq/ne X, C))
+  if (match(&I, m_c_ICmp(Pred1, m_Value(X),
+ m_ZExt(m_ICmp(Pred2, m_Deferred(X), m_APInt(C) &&
+  ICmpInst::isEquality(Pred1) && ICmpInst::isEquality(Pred2)) {
+if (C->isZero()) {
+  if (Pred2 == ICmpInst::ICMP_EQ) {
+// icmp eq X, (zext (icmp eq X, 0)) --> false
+// icmp ne X, (zext (icmp eq X, 0)) --> true
+return replaceInstUsesWith(
+I,
+Constant::getIntegerValue(
+I.getType(),
+APInt(1U, static_cast(Pred1 == ICmpInst::ICMP_NE;
+  } else {
+// icmp eq X, (zext (icmp ne X, 0)) --> icmp ult X, 2
+// icmp ne X, (zext (icmp ne X, 0)) --> icmp ugt X, 1
+return ICmpInst::Create(
+Instruction::ICmp,
+Pred1 == ICmpInst::ICMP_NE ? ICmpInst::ICMP_UGT
+   : ICmpInst::ICMP_ULT,
+X,
+Constant::getIntegerValue(
+X->getType(), APInt(X->getType()->getScalarSizeInBits(),
+Pred1 == ICmpInst::ICMP_NE ? 1 : 2)));
+  }
+} else if (C->isOne()) {
+  if (Pred2 == ICmpInst::ICMP_NE) {
+// icmp eq X, (zext (icmp ne X, 1)) --> false
+// icmp ne X, (zext (icmp ne X, 1)) --> true
+return replaceInstUsesWith(
+I,
+Constant::getIntegerValue(
+I.getType(),
+APInt(1U, static_cast(Pred1 == ICmpInst::ICMP_NE;
+  } else {
+// icmp eq X, (zext (icmp eq X, 1)) --> icmp ult X, 2
+// icmp ne X, (zext (icmp eq X, 1)) --> icmp ugt X, 1
+return ICmpInst::Create(
+Instruction::ICmp,
+Pred1 == ICmpInst::ICMP_NE ? ICmpInst::ICMP_UGT
+   : ICmpInst::ICMP_ULT,
+X,
+Constant::getIntegerValue(
+X->getType(), APInt(X->getType()->getScalarSizeInBits(),
+Pred1 == ICmpInst::ICMP_NE ? 1 : 2)));
+  }
+} else {
+  // C != 0 && C != 1
+  // icmp eq X, (zext (icmp eq X, C)) --> icmp eq X, 0
+  // icmp eq X, (zext (icmp ne X, C)) --> icmp eq X, 1
+  // icmp ne X, (zext (icmp eq X, C)) --> icmp ne X, 0
+  // icmp ne X, (zext (icmp ne X, C)) --> icmp ne X, 1
+  return ICmpInst::Create(
+  Instruction::ICmp, Pred1, X,
+  Constant::getIntegerValue(
+  X->getType(),
+  APInt(X->getType()->getScalarSizeInBits(),
+static_cast(Pred2 == ICmpInst::ICMP_NE;
+}
+  }
+
   if (match(I.getOperand(0), m_c_Add(m_ZExt(m_Value(X)), m_SExt(m_Value(Y 
&&
   match(I.getOperand(1), m_APInt(C)) &&
   X->getType()->isIntOrIntVectorTy(1) &&
diff --git a/llvm/test/Transforms/InstCombine/icmp-range.ll 
b/llvm/test/Transforms/InstCombine/icmp-range.ll
index 4281e09cb0309c8..15424fce33fdeea 100644
--- a/llvm/test/Transforms/InstCombine/icmp-range.ll
+++ b/llvm/test/Transforms/InstCombine/icmp-range.ll
@@ -1034,6 +1034,187 @@ define i1 @icmp_ne_bool_1(ptr %ptr) {
   ret i1 %cmp
 }
 
+define i1 @icmp_ne_zext_eq_zero(i32 %a) {
+; CHECK-LABEL: @icmp_ne_zext_eq_zero(
+; CHECK-NEXT:ret i1 true
+;
+  %cmp = icmp eq i32 %a, 0
+  %conv = zext i1 %cmp to i32
+  %cmp1 = icmp ne i32 %conv, %a
+  ret i1 %cmp1
+}
+
+define i1 @icmp_ne_zext_ne_zero(i32 %a) {
+; CHECK-LABEL: @icmp_ne_zext_ne_zero(
+; CHECK-NEXT:[[CMP1:%.*]] = icmp ugt i32 [[A:%.*]], 1
+; CHECK-NEXT:ret i1 [[CMP1]]
+;
+  %cmp = icmp ne i32 %a, 0
+  %conv = zext i1 %cmp to i32
+  %cmp1 = icmp ne i32 %conv, %a
+  ret i1 %cmp1
+}
+
+define i1 @icmp_eq_zext_eq_zero(i32 %a) {
+; CHECK-LABEL: @icmp_eq_zext_eq_zero(
+; CHECK-NEXT:ret i1 false
+;
+  %cmp = icmp eq i32 %a, 0
+  %conv = zext i1 %cmp to i32
+  %cmp1 = icmp eq i32 %conv, %a
+  ret i1 %cmp1
+}
+
+define i1 @icmp_eq_zext_ne_zero(i32 %

[clang] [InstCombine] Simplify the pattern `a ne/eq (zext/sext (a ne/eq c))` (PR #65852)

[Yingwei Zheng via cfe-commits]

https://github.com/dtcxzyw updated 
https://github.com/llvm/llvm-project/pull/65852

>From d9d8bcbb98e8f5aecb9733329389d61a489bd731 Mon Sep 17 00:00:00 2001
From: Yingwei Zheng 
Date: Sat, 9 Sep 2023 23:07:29 +0800
Subject: [PATCH 1/6] [InstCombine] Simplify the pattern `a ne/eq (zext (a
 ne/eq c))`

---
 .../InstCombine/InstCombineCompares.cpp   |  62 ++
 .../test/Transforms/InstCombine/icmp-range.ll | 181 ++
 2 files changed, 243 insertions(+)

diff --git a/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp 
b/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
index 9fdc46fec631679..837b8e6d2619989 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
@@ -6309,7 +6309,69 @@ Instruction 
*InstCombinerImpl::foldICmpUsingBoolRange(ICmpInst &I) {
   Y->getType()->isIntOrIntVectorTy(1) && Pred == ICmpInst::ICMP_ULE)
 return BinaryOperator::CreateOr(Builder.CreateIsNull(X), Y);
 
+  ICmpInst::Predicate Pred1, Pred2;
   const APInt *C;
+  // icmp eq/ne X, (zext (icmp eq/ne X, C))
+  if (match(&I, m_c_ICmp(Pred1, m_Value(X),
+ m_ZExt(m_ICmp(Pred2, m_Deferred(X), m_APInt(C) &&
+  ICmpInst::isEquality(Pred1) && ICmpInst::isEquality(Pred2)) {
+if (C->isZero()) {
+  if (Pred2 == ICmpInst::ICMP_EQ) {
+// icmp eq X, (zext (icmp eq X, 0)) --> false
+// icmp ne X, (zext (icmp eq X, 0)) --> true
+return replaceInstUsesWith(
+I,
+Constant::getIntegerValue(
+I.getType(),
+APInt(1U, static_cast(Pred1 == ICmpInst::ICMP_NE;
+  } else {
+// icmp eq X, (zext (icmp ne X, 0)) --> icmp ult X, 2
+// icmp ne X, (zext (icmp ne X, 0)) --> icmp ugt X, 1
+return ICmpInst::Create(
+Instruction::ICmp,
+Pred1 == ICmpInst::ICMP_NE ? ICmpInst::ICMP_UGT
+   : ICmpInst::ICMP_ULT,
+X,
+Constant::getIntegerValue(
+X->getType(), APInt(X->getType()->getScalarSizeInBits(),
+Pred1 == ICmpInst::ICMP_NE ? 1 : 2)));
+  }
+} else if (C->isOne()) {
+  if (Pred2 == ICmpInst::ICMP_NE) {
+// icmp eq X, (zext (icmp ne X, 1)) --> false
+// icmp ne X, (zext (icmp ne X, 1)) --> true
+return replaceInstUsesWith(
+I,
+Constant::getIntegerValue(
+I.getType(),
+APInt(1U, static_cast(Pred1 == ICmpInst::ICMP_NE;
+  } else {
+// icmp eq X, (zext (icmp eq X, 1)) --> icmp ult X, 2
+// icmp ne X, (zext (icmp eq X, 1)) --> icmp ugt X, 1
+return ICmpInst::Create(
+Instruction::ICmp,
+Pred1 == ICmpInst::ICMP_NE ? ICmpInst::ICMP_UGT
+   : ICmpInst::ICMP_ULT,
+X,
+Constant::getIntegerValue(
+X->getType(), APInt(X->getType()->getScalarSizeInBits(),
+Pred1 == ICmpInst::ICMP_NE ? 1 : 2)));
+  }
+} else {
+  // C != 0 && C != 1
+  // icmp eq X, (zext (icmp eq X, C)) --> icmp eq X, 0
+  // icmp eq X, (zext (icmp ne X, C)) --> icmp eq X, 1
+  // icmp ne X, (zext (icmp eq X, C)) --> icmp ne X, 0
+  // icmp ne X, (zext (icmp ne X, C)) --> icmp ne X, 1
+  return ICmpInst::Create(
+  Instruction::ICmp, Pred1, X,
+  Constant::getIntegerValue(
+  X->getType(),
+  APInt(X->getType()->getScalarSizeInBits(),
+static_cast(Pred2 == ICmpInst::ICMP_NE;
+}
+  }
+
   if (match(I.getOperand(0), m_c_Add(m_ZExt(m_Value(X)), m_SExt(m_Value(Y 
&&
   match(I.getOperand(1), m_APInt(C)) &&
   X->getType()->isIntOrIntVectorTy(1) &&
diff --git a/llvm/test/Transforms/InstCombine/icmp-range.ll 
b/llvm/test/Transforms/InstCombine/icmp-range.ll
index 4281e09cb0309c8..15424fce33fdeea 100644
--- a/llvm/test/Transforms/InstCombine/icmp-range.ll
+++ b/llvm/test/Transforms/InstCombine/icmp-range.ll
@@ -1034,6 +1034,187 @@ define i1 @icmp_ne_bool_1(ptr %ptr) {
   ret i1 %cmp
 }
 
+define i1 @icmp_ne_zext_eq_zero(i32 %a) {
+; CHECK-LABEL: @icmp_ne_zext_eq_zero(
+; CHECK-NEXT:ret i1 true
+;
+  %cmp = icmp eq i32 %a, 0
+  %conv = zext i1 %cmp to i32
+  %cmp1 = icmp ne i32 %conv, %a
+  ret i1 %cmp1
+}
+
+define i1 @icmp_ne_zext_ne_zero(i32 %a) {
+; CHECK-LABEL: @icmp_ne_zext_ne_zero(
+; CHECK-NEXT:[[CMP1:%.*]] = icmp ugt i32 [[A:%.*]], 1
+; CHECK-NEXT:ret i1 [[CMP1]]
+;
+  %cmp = icmp ne i32 %a, 0
+  %conv = zext i1 %cmp to i32
+  %cmp1 = icmp ne i32 %conv, %a
+  ret i1 %cmp1
+}
+
+define i1 @icmp_eq_zext_eq_zero(i32 %a) {
+; CHECK-LABEL: @icmp_eq_zext_eq_zero(
+; CHECK-NEXT:ret i1 false
+;
+  %cmp = icmp eq i32 %a, 0
+  %conv = zext i1 %cmp to i32
+  %cmp1 = icmp eq i32 %conv, %a
+  ret i1 %cmp1
+}
+
+define i1 @icmp_eq_zext_ne_zero(i32 %

[clang] [clang] Fix null dereference on return in lambda attribute statement expr (PR #66643)

[Piotr Fusik via cfe-commits]

https://github.com/pfusik closed https://github.com/llvm/llvm-project/pull/66643
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] c724ac9 - [clang] Fix null dereference on return in lambda attribute statement expr (#66643)

[via cfe-commits]

Author: Piotr Fusik
Date: 2023-09-19T04:02:04+02:00
New Revision: c724ac9330cb77d9557178fbe3893485264376d9

URL: 
https://github.com/llvm/llvm-project/commit/c724ac9330cb77d9557178fbe3893485264376d9
DIFF: 
https://github.com/llvm/llvm-project/commit/c724ac9330cb77d9557178fbe3893485264376d9.diff

LOG: [clang] Fix null dereference on return in lambda attribute statement expr 
(#66643)

clang was crashing on a lambda attribute with a statement expression
that contained a `return`.
It attempted to access the lambda type which was unknown at that point.

Fixes https://github.com/llvm/llvm-project/issues/48527

Added: 
clang/test/Parser/gh48527.cpp

Modified: 
clang/docs/ReleaseNotes.rst
clang/lib/Sema/SemaStmt.cpp
clang/test/SemaCXX/lambda-expressions.cpp

Removed: 




diff  --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index 9b1578762e301f6..172818114c3b92b 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -301,6 +301,10 @@ Bug Fixes to C++ Support
   makes an invalid call to an immediate function.
   (`#66324 `_)
 
+- Fix crash for a lambda attribute with a statement expression
+  that contains a `return`.
+  (`#48527 `_)
+
 Bug Fixes to AST Handling
 ^
 - Fixed an import failure of recursive friend class template.

diff  --git a/clang/lib/Sema/SemaStmt.cpp b/clang/lib/Sema/SemaStmt.cpp
index 7cc509542d5381d..10adfbc406dfbb5 100644
--- a/clang/lib/Sema/SemaStmt.cpp
+++ b/clang/lib/Sema/SemaStmt.cpp
@@ -3577,6 +3577,8 @@ StmtResult Sema::ActOnCapScopeReturnStmt(SourceLocation 
ReturnLoc,
   CapturingScopeInfo *CurCap = cast(getCurFunction());
   QualType FnRetType = CurCap->ReturnType;
   LambdaScopeInfo *CurLambda = dyn_cast(CurCap);
+  if (CurLambda && CurLambda->CallOperator->getType().isNull())
+return StmtError();
   bool HasDeducedReturnType =
   CurLambda && hasDeducedReturnType(CurLambda->CallOperator);
 

diff  --git a/clang/test/Parser/gh48527.cpp b/clang/test/Parser/gh48527.cpp
new file mode 100644
index 000..420c35be37f5191
--- /dev/null
+++ b/clang/test/Parser/gh48527.cpp
@@ -0,0 +1,10 @@
+// RUN: %clang_cc1 -fsyntax-only -verify %s
+
+int main() { // expected-note {{to match this '{'}}
+auto a = [](void)__attribute__((b(({ // expected-note {{to match this '('}}
+return 0;
+} // expected-error 3 {{expected ')'}} \
+  // expected-error {{expected ';' at end of declaration}}
+// expected-error@+2 {{expected ')'}}
+// expected-error@+1 {{expected body of lambda expression}}
+// expected-error {{expected '}'}}

diff  --git a/clang/test/SemaCXX/lambda-expressions.cpp 
b/clang/test/SemaCXX/lambda-expressions.cpp
index 23745dc14154747..0c9e8584e653473 100644
--- a/clang/test/SemaCXX/lambda-expressions.cpp
+++ b/clang/test/SemaCXX/lambda-expressions.cpp
@@ -714,3 +714,7 @@ void foo() {
   // CHECK-NEXT: ConstantExpr
   // CHECK-NEXT: value: Int 2
 }
+
+void GH48527() {
+  auto a = []()__attribute__((b(({ return 0; }{}; // expected-warning 
{{unknown attribute 'b' ignored}}
+}



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Support target names with dots in more utilities (PR #65812)

[via cfe-commits]

https://github.com/dankm resolved 
https://github.com/llvm/llvm-project/pull/65812
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [MSVC, ARM64] Add _Copy* and _Count* intrinsics (PR #66554)

[Eli Friedman via cfe-commits]

@@ -10752,6 +10752,68 @@ Value 
*CodeGenFunction::EmitAArch64BuiltinExpr(unsigned BuiltinID,
 return Load;
   }
 
+  if (BuiltinID == AArch64::BI_CopyDoubleFromInt64 ||
+  BuiltinID == AArch64::BI_CopyFloatFromInt32 ||
+  BuiltinID == AArch64::BI_CopyInt32FromFloat ||
+  BuiltinID == AArch64::BI_CopyInt64FromDouble) {
+return EmitScalarExpr(E->getArg(0));

efriedma-quic wrote:

I meant a bitcast, not an sitofp.

https://github.com/llvm/llvm-project/pull/66554
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D159064: [Modules] Make clang modules for the C standard library headers

[Volodymyr Sapsai via Phabricator via cfe-commits]

vsapsai added a comment.

Still going through the patch and through the discussion. But wanted to ask if 
there are any complications in reverting this change? For libc++ we've 
discussed that we don't really know the perf impact of multiple small modules. 
So I want to make sure we aren't making decisions that make it impossible to 
revert the change later.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D159064/new/

https://reviews.llvm.org/D159064

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] remove ClassScopeFunctionSpecializationDecl (PR #66636)

[Krystian Stasiowski via cfe-commits]

@@ -650,17 +642,16 @@ void ASTDeclWriter::VisitFunctionDecl(FunctionDecl *D) {
 DependentFunctionTemplateSpecializationInfo *
   DFTSInfo = D->getDependentSpecializationInfo();
 
-// Templates.
-Record.push_back(DFTSInfo->getNumTemplates());
-for (int i=0, e = DFTSInfo->getNumTemplates(); i != e; ++i)
-  Record.AddDeclRef(DFTSInfo->getTemplate(i));
+// Candidates.
+Record.push_back(DFTSInfo->getCandidates().size());

sdkrystian wrote:

`DependentFunctionTemplateSpecializationInfo` stores the number of candidates 
in an `unsigned` member, so it should be fine. 

https://github.com/llvm/llvm-project/pull/66636
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D159483: [Modules] Add a flag to control builtin headers being in system modules

[Volodymyr Sapsai via Phabricator via cfe-commits]

vsapsai added a comment.

In D159483#4641191 , @iana wrote:

> A big assumption this patch makes is that `ModuleMap::isBuiltinHeader` is 
> primarily to support Apple's unfortunate module needs. Thus this patch turns 
> that behavior off by default, which makes things work the way one would 
> expect. That is, when usr/include/module.modulemap references stdint.h, that 
> just means usr/include/stdint.h and it doesn't also pull in the clang builtin 
> stdint.h, it doesn't transform usr/include/stdint.h into a textual header, 
> etc. I'm hoping that's acceptable behavior on non-Apple platforms, but if 
> someone knows otherwise please let me know and we can rethink how the option 
> should be defined and set.

Good way to test the assumptions can be checking how Swift works with this 
change on Linux. There are not as many module maps there as in Apple SDKs but 
there are some.




Comment at: clang/include/clang/Basic/LangOptions.def:176
 COMPATIBLE_LANGOPT(CPlusPlusModules, 1, 0, "C++ modules syntax")
+LANGOPT(BuiltinHeadersInSystemModules, 1, 0, "builtin headers belong to system 
modules, and _Builtin_ modules are ignored for cstdlib headers")
 BENIGN_ENUM_LANGOPT(CompilingModule, CompilingModuleKind, 3, CMK_None,

Why not to make the default value `true` to preserve the old behavior and 
switch on the new behavior in the driver? It's not a veiled way to force you to 
change it, I'm genuinely curious and want to consider pro/cons of various 
alternatives.



Comment at: clang/lib/Driver/ToolChains/Darwin.cpp:2860-2861
+return SDKVersion >= VersionTuple(99U);
+  default:
+return true;
+  }

Another question regarding defaults. Doesn't look consistent with your position
> [...] I thought it safer to default to the current behavior which is false.

Personally I gravitate towards `false` for extra safety but it's not a strong 
opinion.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D159483/new/

https://reviews.llvm.org/D159483

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang-tools-extra] [clang] remove ClassScopeFunctionSpecializationDecl (PR #66636)

[Shafik Yaghmour via cfe-commits]

@@ -4086,7 +4073,7 @@ FunctionDecl 
*Sema::SubstSpaceshipAsEqualEqual(CXXRecordDecl *RD,
   Decl *R;
   if (auto *MD = dyn_cast(Spaceship)) {
 R = Instantiator.VisitCXXMethodDecl(
-MD, nullptr, std::nullopt,
+MD, nullptr,

shafik wrote:

To be consistent with 
[bugprone-argument-comment](https://clang.llvm.org/extra/clang-tidy/checks/bugprone/argument-comment.html)

https://github.com/llvm/llvm-project/pull/66636
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] remove ClassScopeFunctionSpecializationDecl (PR #66636)

[Shafik Yaghmour via cfe-commits]

@@ -4086,7 +4073,7 @@ FunctionDecl 
*Sema::SubstSpaceshipAsEqualEqual(CXXRecordDecl *RD,
   Decl *R;
   if (auto *MD = dyn_cast(Spaceship)) {
 R = Instantiator.VisitCXXMethodDecl(
-MD, nullptr, std::nullopt,
+MD, nullptr,

shafik wrote:

```suggestion
MD, /*TemplateParams=*/nullptr,
```

https://github.com/llvm/llvm-project/pull/66636
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] remove ClassScopeFunctionSpecializationDecl (PR #66636)

[Shafik Yaghmour via cfe-commits]

@@ -650,17 +642,16 @@ void ASTDeclWriter::VisitFunctionDecl(FunctionDecl *D) {
 DependentFunctionTemplateSpecializationInfo *
   DFTSInfo = D->getDependentSpecializationInfo();
 
-// Templates.
-Record.push_back(DFTSInfo->getNumTemplates());
-for (int i=0, e = DFTSInfo->getNumTemplates(); i != e; ++i)
-  Record.AddDeclRef(DFTSInfo->getTemplate(i));
+// Candidates.
+Record.push_back(DFTSInfo->getCandidates().size());

shafik wrote:

`getNumTemplates()` returns `unsigned` while `size()` AFAICT returns `size_t` 
which is not the same type. Is this ok?

https://github.com/llvm/llvm-project/pull/66636
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang-tools-extra] [clang] remove ClassScopeFunctionSpecializationDecl (PR #66636)

[Shafik Yaghmour via cfe-commits]

@@ -1016,21 +1003,20 @@ void ASTDeclReader::VisitFunctionDecl(FunctionDecl *FD) 
{
   }
   case FunctionDecl::TK_DependentFunctionTemplateSpecialization: {
 // Templates.
-UnresolvedSet<8> TemplDecls;
-unsigned NumTemplates = Record.readInt();
-while (NumTemplates--)
-  TemplDecls.addDecl(readDeclAs());
+UnresolvedSet<8> Candidates;
+unsigned NumCandidates = Record.readInt();

shafik wrote:

It looks like this should be `uint64_t`, although it does not seem to do what 
the "label says" it does.

https://github.com/llvm/llvm-project/pull/66636
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang-tools-extra] [clang] remove ClassScopeFunctionSpecializationDecl (PR #66636)

[Shafik Yaghmour via cfe-commits]

https://github.com/shafik commented:

Mostly minor fixes and or questions.

https://github.com/llvm/llvm-project/pull/66636
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] remove ClassScopeFunctionSpecializationDecl (PR #66636)

[Shafik Yaghmour via cfe-commits]

https://github.com/shafik edited https://github.com/llvm/llvm-project/pull/66636
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [X86] Align 128/256 variants to use void * as 512 variants. (PR #66310)

[Freddy Ye via cfe-commits]

https://github.com/FreddyLeaf resolved 
https://github.com/llvm/llvm-project/pull/66310
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [X86] Align 128/256 variants to use void * as 512 variants. (PR #66310)

[Freddy Ye via cfe-commits]

@@ -358,6 +358,12 @@ __m128i test_mm_stream_load_si128(__m128i const *a) {
   return _mm_stream_load_si128(a);
 }
 
+__m128i test_mm_stream_load_si128_void(void const *a) {

FreddyLeaf wrote:

3e463d3c41bb8e8391c62dd95910d009b4b39b7d

https://github.com/llvm/llvm-project/pull/66310
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [X86] Align 128/256 variants to use void * as 512 variants. (PR #66310)

[Freddy Ye via cfe-commits]

https://github.com/FreddyLeaf resolved 
https://github.com/llvm/llvm-project/pull/66310
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [X86] Align 128/256 variants to use void * as 512 variants. (PR #66310)

[Freddy Ye via cfe-commits]

@@ -645,7 +645,7 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS 
_mm_mul_epi32(__m128i __V1,
 /// \returns A 128-bit integer vector containing the data stored at the
 ///specified memory location.
 static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_stream_load_si128(__m128i const *__V) {
+_mm_stream_load_si128(void const *__V) {

FreddyLeaf wrote:

3e463d3c41bb8e8391c62dd95910d009b4b39b7d

https://github.com/llvm/llvm-project/pull/66310
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [X86] Align 128/256 variants to use void * as 512 variants. (PR #66310)

[Freddy Ye via cfe-commits]

https://github.com/FreddyLeaf resolved 
https://github.com/llvm/llvm-project/pull/66310
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [X86] Align 128/256 variants to use void * as 512 variants. (PR #66310)

[Freddy Ye via cfe-commits]

@@ -2979,7 +2979,7 @@ _mm256_xor_si256(__m256i __a, __m256i __b)
 ///A pointer to the 32-byte aligned memory containing the vector to load.
 /// \returns A 256-bit integer vector loaded from memory.
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_stream_load_si256(__m256i const *__V)
+_mm256_stream_load_si256(void const *__V)

FreddyLeaf wrote:

3e463d3c41bb8e8391c62dd95910d009b4b39b7d

https://github.com/llvm/llvm-project/pull/66310
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [X86] Align 128/256 variants to use void * as 512 variants. (PR #66310)

[Freddy Ye via cfe-commits]

https://github.com/FreddyLeaf resolved 
https://github.com/llvm/llvm-project/pull/66310
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [X86] Align 128/256 variants to use void * as 512 variants. (PR #66310)

[Freddy Ye via cfe-commits]

https://github.com/FreddyLeaf edited 
https://github.com/llvm/llvm-project/pull/66310
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [X86] Align 128/256 variants to use void * as 512 variants. (PR #66310)

[Freddy Ye via cfe-commits]

@@ -1223,6 +1223,12 @@ __m256i test_mm256_stream_load_si256(__m256i const *a) {
   return _mm256_stream_load_si256(a);
 }
 
+__m256i test_mm256_stream_load_si256_const(void const *a) {

FreddyLeaf wrote:

3e463d3

https://github.com/llvm/llvm-project/pull/66310
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [X86] Align 128/256 variants to use void * as 512 variants. (PR #66310)

[Freddy Ye via cfe-commits]

https://github.com/FreddyLeaf updated 
https://github.com/llvm/llvm-project/pull/66310

>From 21157a0e3b4c4e4e2430752ef806148685a942a2 Mon Sep 17 00:00:00 2001
From: Freddy Ye 
Date: Thu, 14 Sep 2023 09:17:39 +0800
Subject: [PATCH 1/2] [X86] Align 128/256 variants to use void * as 512
 variants.

For *_stream_* series intrinsics.
---
 clang/lib/Headers/avx2intrin.h  |  2 +-
 clang/lib/Headers/avxintrin.h   |  6 +++---
 clang/lib/Headers/emmintrin.h   |  8 
 clang/lib/Headers/smmintrin.h   |  2 +-
 clang/lib/Headers/xmmintrin.h   |  2 +-
 clang/test/CodeGen/X86/avx-builtins.c   | 18 ++
 clang/test/CodeGen/X86/avx2-builtins.c  |  6 ++
 clang/test/CodeGen/X86/sse-builtins.c   |  6 ++
 clang/test/CodeGen/X86/sse2-builtins.c  | 24 
 clang/test/CodeGen/X86/sse41-builtins.c |  6 ++
 10 files changed, 70 insertions(+), 10 deletions(-)

diff --git a/clang/lib/Headers/avx2intrin.h b/clang/lib/Headers/avx2intrin.h
index c45006193eddcc9..675a93bba1c8a4f 100644
--- a/clang/lib/Headers/avx2intrin.h
+++ b/clang/lib/Headers/avx2intrin.h
@@ -2979,7 +2979,7 @@ _mm256_xor_si256(__m256i __a, __m256i __b)
 ///A pointer to the 32-byte aligned memory containing the vector to load.
 /// \returns A 256-bit integer vector loaded from memory.
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_stream_load_si256(__m256i const *__V)
+_mm256_stream_load_si256(void const *__V)
 {
   typedef __v4di __v4di_aligned __attribute__((aligned(32)));
   return (__m256i)__builtin_nontemporal_load((const __v4di_aligned *)__V);
diff --git a/clang/lib/Headers/avxintrin.h b/clang/lib/Headers/avxintrin.h
index 94fac5e6c9da471..b796bb773ec11f0 100644
--- a/clang/lib/Headers/avxintrin.h
+++ b/clang/lib/Headers/avxintrin.h
@@ -3563,7 +3563,7 @@ _mm_maskstore_ps(float *__p, __m128i __m, __m128 __a)
 /// \param __b
 ///A 256-bit integer vector containing the values to be moved.
 static __inline void __DEFAULT_FN_ATTRS
-_mm256_stream_si256(__m256i *__a, __m256i __b)
+_mm256_stream_si256(void *__a, __m256i __b)
 {
   typedef __v4di __v4di_aligned __attribute__((aligned(32)));
   __builtin_nontemporal_store((__v4di_aligned)__b, (__v4di_aligned*)__a);
@@ -3583,7 +3583,7 @@ _mm256_stream_si256(__m256i *__a, __m256i __b)
 /// \param __b
 ///A 256-bit vector of [4 x double] containing the values to be moved.
 static __inline void __DEFAULT_FN_ATTRS
-_mm256_stream_pd(double *__a, __m256d __b)
+_mm256_stream_pd(void *__a, __m256d __b)
 {
   typedef __v4df __v4df_aligned __attribute__((aligned(32)));
   __builtin_nontemporal_store((__v4df_aligned)__b, (__v4df_aligned*)__a);
@@ -3604,7 +3604,7 @@ _mm256_stream_pd(double *__a, __m256d __b)
 /// \param __a
 ///A 256-bit vector of [8 x float] containing the values to be moved.
 static __inline void __DEFAULT_FN_ATTRS
-_mm256_stream_ps(float *__p, __m256 __a)
+_mm256_stream_ps(void *__p, __m256 __a)
 {
   typedef __v8sf __v8sf_aligned __attribute__((aligned(32)));
   __builtin_nontemporal_store((__v8sf_aligned)__a, (__v8sf_aligned*)__p);
diff --git a/clang/lib/Headers/emmintrin.h b/clang/lib/Headers/emmintrin.h
index 064d974936598f8..eacb0182614304d 100644
--- a/clang/lib/Headers/emmintrin.h
+++ b/clang/lib/Headers/emmintrin.h
@@ -3945,7 +3945,7 @@ static __inline__ void __DEFAULT_FN_ATTRS 
_mm_storel_epi64(__m128i_u *__p,
 ///A pointer to the 128-bit aligned memory location used to store the 
value.
 /// \param __a
 ///A vector of [2 x double] containing the 64-bit values to be stored.
-static __inline__ void __DEFAULT_FN_ATTRS _mm_stream_pd(double *__p,
+static __inline__ void __DEFAULT_FN_ATTRS _mm_stream_pd(void *__p,
 __m128d __a) {
   __builtin_nontemporal_store((__v2df)__a, (__v2df *)__p);
 }
@@ -3963,7 +3963,7 @@ static __inline__ void __DEFAULT_FN_ATTRS 
_mm_stream_pd(double *__p,
 ///A pointer to the 128-bit aligned memory location used to store the 
value.
 /// \param __a
 ///A 128-bit integer vector containing the values to be stored.
-static __inline__ void __DEFAULT_FN_ATTRS _mm_stream_si128(__m128i *__p,
+static __inline__ void __DEFAULT_FN_ATTRS _mm_stream_si128(void *__p,
__m128i __a) {
   __builtin_nontemporal_store((__v2di)__a, (__v2di *)__p);
 }
@@ -3983,7 +3983,7 @@ static __inline__ void __DEFAULT_FN_ATTRS 
_mm_stream_si128(__m128i *__p,
 ///A 32-bit integer containing the value to be stored.
 static __inline__ void
 __attribute__((__always_inline__, __nodebug__, __target__("sse2")))
-_mm_stream_si32(int *__p, int __a) {
+_mm_stream_si32(void *__p, int __a) {
   __builtin_ia32_movnti(__p, __a);
 }
 
@@ -4003,7 +4003,7 @@ static __inline__ void
 ///A 64-bit integer containing the value to be stored.
 static __inline__ void
 __attribute__((__always_inline__, __nodebug__, __target__("sse2")))
-_mm_stream_si64(long long *__p, 

[clang] Support target names with dots in more utilities (PR #65812)

[via cfe-commits]

https://github.com/dankm updated https://github.com/llvm/llvm-project/pull/65812

>From 5f5e57a5c01088e7f0b9276ed02dd436a3f9cc0a Mon Sep 17 00:00:00 2001
From: Dan McGregor 
Date: Mon, 14 Aug 2023 18:44:08 -0600
Subject: [PATCH 1/2] Support: hoist lld's executable name code to Path

Instead of using custom code to find the program name throughout
the codebase, write one function as a path helper to consistently
determine the program name. This globally correctly finds target
names with dots in them (ie freebsd13.2).
---
 lld/COFF/Driver.cpp|  2 +-
 lld/Common/Args.cpp|  6 --
 lld/ELF/Driver.cpp |  2 +-
 lld/MachO/Driver.cpp   |  2 +-
 lld/include/lld/Common/Args.h  |  2 --
 lld/wasm/Driver.cpp|  2 +-
 lldb/tools/driver/Driver.cpp   |  2 +-
 llvm/include/llvm/Support/Path.h   | 15 +++
 llvm/lib/Support/Path.cpp  |  7 +++
 llvm/tools/llvm-ar/llvm-ar.cpp |  3 ++-
 llvm/tools/llvm-cov/llvm-cov.cpp   |  2 +-
 llvm/tools/llvm-driver/llvm-driver.cpp |  2 +-
 llvm/tools/llvm-objcopy/llvm-objcopy.cpp   |  2 +-
 llvm/tools/llvm-objdump/llvm-objdump.cpp   |  2 +-
 llvm/tools/llvm-rc/llvm-rc.cpp |  2 +-
 llvm/tools/llvm-readobj/llvm-readobj.cpp   |  2 +-
 llvm/tools/llvm-symbolizer/llvm-symbolizer.cpp |  2 +-
 17 files changed, 36 insertions(+), 21 deletions(-)

diff --git a/lld/COFF/Driver.cpp b/lld/COFF/Driver.cpp
index d7476e91e03e384..03a63bf3a306b4d 100644
--- a/lld/COFF/Driver.cpp
+++ b/lld/COFF/Driver.cpp
@@ -68,7 +68,7 @@ bool link(ArrayRef args, llvm::raw_ostream 
&stdoutOS,
   auto *ctx = new COFFLinkerContext;
 
   ctx->e.initialize(stdoutOS, stderrOS, exitEarly, disableOutput);
-  ctx->e.logName = args::getFilenameWithoutExe(args[0]);
+  ctx->e.logName = sys::path::program_name(args[0]);
   ctx->e.errorLimitExceededMsg = "too many errors emitted, stopping now"
  " (use /errorlimit:0 to see all errors)";
 
diff --git a/lld/Common/Args.cpp b/lld/Common/Args.cpp
index 48c934df3a2c931..b02d023fe19aeb5 100644
--- a/lld/Common/Args.cpp
+++ b/lld/Common/Args.cpp
@@ -85,9 +85,3 @@ std::vector lld::args::getLines(MemoryBufferRef 
mb) {
   }
   return ret;
 }
-
-StringRef lld::args::getFilenameWithoutExe(StringRef path) {
-  if (path.ends_with_insensitive(".exe"))
-return sys::path::stem(path);
-  return sys::path::filename(path);
-}
diff --git a/lld/ELF/Driver.cpp b/lld/ELF/Driver.cpp
index 9d167293574fa64..957d0da1589caf8 100644
--- a/lld/ELF/Driver.cpp
+++ b/lld/ELF/Driver.cpp
@@ -143,7 +143,7 @@ bool link(ArrayRef args, llvm::raw_ostream 
&stdoutOS,
 
 SharedFile::vernauxNum = 0;
   };
-  ctx->e.logName = args::getFilenameWithoutExe(args[0]);
+  ctx->e.logName = sys::path::program_name(args[0]);
   ctx->e.errorLimitExceededMsg = "too many errors emitted, stopping now (use "
  "--error-limit=0 to see all errors)";
 
diff --git a/lld/MachO/Driver.cpp b/lld/MachO/Driver.cpp
index 00ff3439a043be9..ed0b66d4b2ec6ba 100644
--- a/lld/MachO/Driver.cpp
+++ b/lld/MachO/Driver.cpp
@@ -1428,7 +1428,7 @@ bool link(ArrayRef argsArr, 
llvm::raw_ostream &stdoutOS,
 InputFile::resetIdCount();
   };
 
-  ctx->e.logName = args::getFilenameWithoutExe(argsArr[0]);
+  ctx->e.logName = sys::path::program_name(argsArr[0]);
 
   MachOOptTable parser;
   InputArgList args = parser.parse(argsArr.slice(1));
diff --git a/lld/include/lld/Common/Args.h b/lld/include/lld/Common/Args.h
index 60f83fbbbf1a3c9..8822707a6a1f212 100644
--- a/lld/include/lld/Common/Args.h
+++ b/lld/include/lld/Common/Args.h
@@ -38,8 +38,6 @@ uint64_t getZOptionValue(llvm::opt::InputArgList &args, int 
id, StringRef key,
 
 std::vector getLines(MemoryBufferRef mb);
 
-StringRef getFilenameWithoutExe(StringRef path);
-
 } // namespace args
 } // namespace lld
 
diff --git a/lld/wasm/Driver.cpp b/lld/wasm/Driver.cpp
index c2f5f0185781f36..f13aa520f35b100 100644
--- a/lld/wasm/Driver.cpp
+++ b/lld/wasm/Driver.cpp
@@ -89,7 +89,7 @@ bool link(ArrayRef args, llvm::raw_ostream 
&stdoutOS,
   auto *ctx = new CommonLinkerContext;
 
   ctx->e.initialize(stdoutOS, stderrOS, exitEarly, disableOutput);
-  ctx->e.logName = args::getFilenameWithoutExe(args[0]);
+  ctx->e.logName = sys::path::program_name(args[0]);
   ctx->e.errorLimitExceededMsg = "too many errors emitted, stopping now (use "
  "-error-limit=0 to see all errors)";
 
diff --git a/lldb/tools/driver/Driver.cpp b/lldb/tools/driver/Driver.cpp
index f8058f868d53ffe..2fb760ca97a001d 100644
--- a/lldb/tools/driver/Driver.cpp
+++ b/lldb/tools/driver/Driver.cpp
@@ -754,7 +754,7 @@ int main(int argc, char const *argv[]) {
   ArrayRef arg_arr = ArrayRef(argv + 1, argc - 1);
   opt::InputArgList input_args =
   T.ParseArgs(arg_arr, MissingA

[clang] Support target names with dots in more utilities (PR #65812)

[via cfe-commits]

https://github.com/dankm resolved 
https://github.com/llvm/llvm-project/pull/65812
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] Fix null dereference on return in lambda attribute statement expr (PR #66643)

[Piotr Fusik via cfe-commits]

https://github.com/pfusik updated 
https://github.com/llvm/llvm-project/pull/66643

>From 4545c9c691a7affcab6cfe4e1b9b2a7715ab3f8b Mon Sep 17 00:00:00 2001
From: Piotr Fusik 
Date: Mon, 18 Sep 2023 19:56:10 +0200
Subject: [PATCH] [clang] Fix null dereference on return in lambda attribute
 statement expr

clang was crashing on a lambda attribute with a statement expression
that contained a `return`.
It attempted to access the lambda type which was unknown at that point.

Fixes https://github.com/llvm/llvm-project/issues/48527
---
 clang/docs/ReleaseNotes.rst   |  4 
 clang/lib/Sema/SemaStmt.cpp   |  2 ++
 clang/test/Parser/gh48527.cpp | 10 ++
 clang/test/SemaCXX/lambda-expressions.cpp |  4 
 4 files changed, 20 insertions(+)
 create mode 100644 clang/test/Parser/gh48527.cpp

diff --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index 9b1578762e301f6..172818114c3b92b 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -301,6 +301,10 @@ Bug Fixes to C++ Support
   makes an invalid call to an immediate function.
   (`#66324 `_)
 
+- Fix crash for a lambda attribute with a statement expression
+  that contains a `return`.
+  (`#48527 `_)
+
 Bug Fixes to AST Handling
 ^
 - Fixed an import failure of recursive friend class template.
diff --git a/clang/lib/Sema/SemaStmt.cpp b/clang/lib/Sema/SemaStmt.cpp
index 7cc509542d5381d..10adfbc406dfbb5 100644
--- a/clang/lib/Sema/SemaStmt.cpp
+++ b/clang/lib/Sema/SemaStmt.cpp
@@ -3577,6 +3577,8 @@ StmtResult Sema::ActOnCapScopeReturnStmt(SourceLocation 
ReturnLoc,
   CapturingScopeInfo *CurCap = cast(getCurFunction());
   QualType FnRetType = CurCap->ReturnType;
   LambdaScopeInfo *CurLambda = dyn_cast(CurCap);
+  if (CurLambda && CurLambda->CallOperator->getType().isNull())
+return StmtError();
   bool HasDeducedReturnType =
   CurLambda && hasDeducedReturnType(CurLambda->CallOperator);
 
diff --git a/clang/test/Parser/gh48527.cpp b/clang/test/Parser/gh48527.cpp
new file mode 100644
index 000..420c35be37f5191
--- /dev/null
+++ b/clang/test/Parser/gh48527.cpp
@@ -0,0 +1,10 @@
+// RUN: %clang_cc1 -fsyntax-only -verify %s
+
+int main() { // expected-note {{to match this '{'}}
+auto a = [](void)__attribute__((b(({ // expected-note {{to match this '('}}
+return 0;
+} // expected-error 3 {{expected ')'}} \
+  // expected-error {{expected ';' at end of declaration}}
+// expected-error@+2 {{expected ')'}}
+// expected-error@+1 {{expected body of lambda expression}}
+// expected-error {{expected '}'}}
diff --git a/clang/test/SemaCXX/lambda-expressions.cpp 
b/clang/test/SemaCXX/lambda-expressions.cpp
index 23745dc14154747..0c9e8584e653473 100644
--- a/clang/test/SemaCXX/lambda-expressions.cpp
+++ b/clang/test/SemaCXX/lambda-expressions.cpp
@@ -714,3 +714,7 @@ void foo() {
   // CHECK-NEXT: ConstantExpr
   // CHECK-NEXT: value: Int 2
 }
+
+void GH48527() {
+  auto a = []()__attribute__((b(({ return 0; }{}; // expected-warning 
{{unknown attribute 'b' ignored}}
+}

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] Enable descriptions for --print-supported-extensions (PR #66715)

[Balint Cristian via cfe-commits]

https://github.com/cbalint13 updated 
https://github.com/llvm/llvm-project/pull/66715

>From 231c93881e7719b447db3f15a37ad6cecede3be3 Mon Sep 17 00:00:00 2001
From: Balint Cristian 
Date: Tue, 19 Sep 2023 02:59:40 +0300
Subject: [PATCH] [clang] Enable descriptions for --print-supported-extensions

---
 clang/tools/driver/cc1_main.cpp   | 15 ++---
 llvm/include/llvm/MC/MCSubtargetInfo.h|  7 +++
 llvm/include/llvm/Support/RISCVISAInfo.h  |  2 +-
 .../llvm/TargetParser/AArch64TargetParser.h   |  3 ++-
 .../llvm/TargetParser/ARMTargetParser.h   |  3 ++-
 llvm/lib/Support/RISCVISAInfo.cpp | 21 +--
 llvm/lib/TargetParser/AArch64TargetParser.cpp | 16 ++
 llvm/lib/TargetParser/ARMTargetParser.cpp | 15 +
 llvm/unittests/Support/RISCVISAInfoTest.cpp   |  4 +++-
 .../TargetParser/TargetParserTest.cpp |  7 +--
 10 files changed, 70 insertions(+), 23 deletions(-)

diff --git a/clang/tools/driver/cc1_main.cpp b/clang/tools/driver/cc1_main.cpp
index f0d7b5c3889dc1f..a963324705db92d 100644
--- a/clang/tools/driver/cc1_main.cpp
+++ b/clang/tools/driver/cc1_main.cpp
@@ -28,6 +28,7 @@
 #include "llvm/ADT/Statistic.h"
 #include "llvm/Config/llvm-config.h"
 #include "llvm/LinkAllPasses.h"
+#include "llvm/MC/MCSubtargetInfo.h"
 #include "llvm/MC/TargetRegistry.h"
 #include "llvm/Option/Arg.h"
 #include "llvm/Option/ArgList.h"
@@ -198,13 +199,21 @@ static int PrintSupportedExtensions(std::string 
TargetStr) {
   std::unique_ptr TheTargetMachine(
   TheTarget->createTargetMachine(TargetStr, "", "", Options, 
std::nullopt));
   const llvm::Triple &MachineTriple = TheTargetMachine->getTargetTriple();
+  const llvm::MCSubtargetInfo *MCInfo = TheTargetMachine->getMCSubtargetInfo();
+  const llvm::ArrayRef Features =
+MCInfo->getAllProcessorFeatures();
+
+  std::map llvmDescMap;
+  for (const llvm::SubtargetFeatureKV &feature : Features) {
+llvmDescMap.insert(std::make_pair(feature.Key, feature.Desc));
+  }
 
   if (MachineTriple.isRISCV())
-llvm::riscvExtensionsHelp();
+llvm::riscvExtensionsHelp(llvmDescMap);
   else if (MachineTriple.isAArch64())
-llvm::AArch64::PrintSupportedExtensions();
+llvm::AArch64::PrintSupportedExtensions(llvmDescMap);
   else if (MachineTriple.isARM())
-llvm::ARM::PrintSupportedExtensions();
+llvm::ARM::PrintSupportedExtensions(llvmDescMap);
   else {
 // The option was already checked in Driver::HandleImmediateArgs,
 // so we do not expect to get here if we are not a supported architecture.
diff --git a/llvm/include/llvm/MC/MCSubtargetInfo.h 
b/llvm/include/llvm/MC/MCSubtargetInfo.h
index c1533ac8d0059f5..45c58988985e638 100644
--- a/llvm/include/llvm/MC/MCSubtargetInfo.h
+++ b/llvm/include/llvm/MC/MCSubtargetInfo.h
@@ -230,10 +230,17 @@ class MCSubtargetInfo {
 return Found != ProcDesc.end() && StringRef(Found->Key) == CPU;
   }
 
+  /// Return processor descriptions.
   ArrayRef getAllProcessorDescriptions() const {
 return ProcDesc;
   }
 
+  /// Return processor features.
+  ArrayRef getAllProcessorFeatures() const {
+return ProcFeatures;
+  }
+
+
   virtual unsigned getHwMode() const { return 0; }
 
   /// Return the cache size in bytes for the given level of cache.
diff --git a/llvm/include/llvm/Support/RISCVISAInfo.h 
b/llvm/include/llvm/Support/RISCVISAInfo.h
index 9092fe5c272a994..1555f2f9909983e 100644
--- a/llvm/include/llvm/Support/RISCVISAInfo.h
+++ b/llvm/include/llvm/Support/RISCVISAInfo.h
@@ -22,7 +22,7 @@ struct RISCVExtensionInfo {
   unsigned MinorVersion;
 };
 
-void riscvExtensionsHelp();
+void riscvExtensionsHelp(std::map llvmDescMap);
 
 class RISCVISAInfo {
 public:
diff --git a/llvm/include/llvm/TargetParser/AArch64TargetParser.h 
b/llvm/include/llvm/TargetParser/AArch64TargetParser.h
index 190f482044083c0..0d4b5e5a18458c7 100644
--- a/llvm/include/llvm/TargetParser/AArch64TargetParser.h
+++ b/llvm/include/llvm/TargetParser/AArch64TargetParser.h
@@ -19,6 +19,7 @@
 #include "llvm/ADT/StringRef.h"
 #include "llvm/Support/VersionTuple.h"
 #include 
+#include 
 #include 
 
 namespace llvm {
@@ -663,7 +664,7 @@ bool isX18ReservedByDefault(const Triple &TT);
 // themselves, they are sequential (0, 1, 2, 3, ...).
 uint64_t getCpuSupportsMask(ArrayRef FeatureStrs);
 
-void PrintSupportedExtensions();
+void PrintSupportedExtensions(std::map llvmDescMap);
 
 } // namespace AArch64
 } // namespace llvm
diff --git a/llvm/include/llvm/TargetParser/ARMTargetParser.h 
b/llvm/include/llvm/TargetParser/ARMTargetParser.h
index 37a358d1fa415c9..f3472b1e68233ee 100644
--- a/llvm/include/llvm/TargetParser/ARMTargetParser.h
+++ b/llvm/include/llvm/TargetParser/ARMTargetParser.h
@@ -17,6 +17,7 @@
 #include "llvm/ADT/StringRef.h"
 #include "llvm/Support/ARMBuildAttributes.h"
 #include "llvm/TargetParser/ARMTargetParserCommon.h"
+#include 
 #include 
 
 namespace llvm {
@@ -259,7 +260,7 @@ StringRef computeDefaultTargetABI(const Triple &TT, 
St

[clang] Introduce paged vector (PR #66430)

[Giulio Eulisse via cfe-commits]

https://github.com/ktf updated https://github.com/llvm/llvm-project/pull/66430

>From 5d2553755add3e7d2adb0b2119d6a0c355df1551 Mon Sep 17 00:00:00 2001
From: Giulio Eulisse <10544+...@users.noreply.github.com>
Date: Thu, 14 Sep 2023 21:58:21 +0200
Subject: [PATCH] Introduce PagedVector class

The goal of the class is to be an (almost) drop in replacement for
SmallVector and std::vector when those are presized and filled later,
as it happens in SourceManager and ASTReader.

By splitting the actual vector in pages of the same size and allocating
the pages only when they are needed, using this containers reduces the
memory usage by a factor 4 for the cases relevant to the ALICE
experiment ROOT / cling usage.
---
 clang/include/clang/Basic/SourceManager.h |   3 +-
 clang/include/clang/Serialization/ASTReader.h |   5 +-
 clang/lib/Basic/SourceManager.cpp |  10 +-
 clang/lib/Serialization/ASTReader.cpp |   5 +-
 llvm/docs/ProgrammersManual.rst   |  32 ++
 llvm/include/llvm/ADT/PagedVector.h   | 322 ++
 llvm/unittests/ADT/CMakeLists.txt |   1 +
 llvm/unittests/ADT/PagedVectorTest.cpp| 265 ++
 8 files changed, 633 insertions(+), 10 deletions(-)
 create mode 100644 llvm/include/llvm/ADT/PagedVector.h
 create mode 100644 llvm/unittests/ADT/PagedVectorTest.cpp

diff --git a/clang/include/clang/Basic/SourceManager.h 
b/clang/include/clang/Basic/SourceManager.h
index 2f846502d6f3327..e37caa2252532f9 100644
--- a/clang/include/clang/Basic/SourceManager.h
+++ b/clang/include/clang/Basic/SourceManager.h
@@ -43,6 +43,7 @@
 #include "llvm/ADT/DenseMap.h"
 #include "llvm/ADT/DenseSet.h"
 #include "llvm/ADT/IntrusiveRefCntPtr.h"
+#include "llvm/ADT/PagedVector.h"
 #include "llvm/ADT/PointerIntPair.h"
 #include "llvm/ADT/SmallVector.h"
 #include "llvm/ADT/StringRef.h"
@@ -699,7 +700,7 @@ class SourceManager : public RefCountedBase {
   ///
   /// Negative FileIDs are indexes into this table. To get from ID to an index,
   /// use (-ID - 2).
-  SmallVector LoadedSLocEntryTable;
+  llvm::PagedVector LoadedSLocEntryTable;
 
   /// The starting offset of the next local SLocEntry.
   ///
diff --git a/clang/include/clang/Serialization/ASTReader.h 
b/clang/include/clang/Serialization/ASTReader.h
index dc1eb21c27801fe..65e19c6e44cf571 100644
--- a/clang/include/clang/Serialization/ASTReader.h
+++ b/clang/include/clang/Serialization/ASTReader.h
@@ -38,6 +38,7 @@
 #include "llvm/ADT/DenseSet.h"
 #include "llvm/ADT/IntrusiveRefCntPtr.h"
 #include "llvm/ADT/MapVector.h"
+#include "llvm/ADT/PagedVector.h"
 #include "llvm/ADT/STLExtras.h"
 #include "llvm/ADT/SetVector.h"
 #include "llvm/ADT/SmallPtrSet.h"
@@ -487,7 +488,7 @@ class ASTReader
   ///
   /// When the pointer at index I is non-NULL, the type with
   /// ID = (I + 1) << FastQual::Width has already been loaded
-  std::vector TypesLoaded;
+  llvm::PagedVector TypesLoaded;
 
   using GlobalTypeMapType =
   ContinuousRangeMap;
@@ -501,7 +502,7 @@ class ASTReader
   ///
   /// When the pointer at index I is non-NULL, the declaration with ID
   /// = I + 1 has already been loaded.
-  std::vector DeclsLoaded;
+  llvm::PagedVector DeclsLoaded;
 
   using GlobalDeclMapType =
   ContinuousRangeMap;
diff --git a/clang/lib/Basic/SourceManager.cpp 
b/clang/lib/Basic/SourceManager.cpp
index 0521ac7b30339ab..7fa8b8096ac4931 100644
--- a/clang/lib/Basic/SourceManager.cpp
+++ b/clang/lib/Basic/SourceManager.cpp
@@ -2344,11 +2344,11 @@ SourceManager::MemoryBufferSizes 
SourceManager::getMemoryBufferSizes() const {
 }
 
 size_t SourceManager::getDataStructureSizes() const {
-  size_t size = llvm::capacity_in_bytes(MemBufferInfos)
-+ llvm::capacity_in_bytes(LocalSLocEntryTable)
-+ llvm::capacity_in_bytes(LoadedSLocEntryTable)
-+ llvm::capacity_in_bytes(SLocEntryLoaded)
-+ llvm::capacity_in_bytes(FileInfos);
+  size_t size = llvm::capacity_in_bytes(MemBufferInfos) +
+llvm::capacity_in_bytes(LocalSLocEntryTable) +
+llvm::capacity_in_bytes(LoadedSLocEntryTable) +
+llvm::capacity_in_bytes(SLocEntryLoaded) +
+llvm::capacity_in_bytes(FileInfos);
 
   if (OverriddenFilesInfo)
 size += llvm::capacity_in_bytes(OverriddenFilesInfo->OverriddenFiles);
diff --git a/clang/lib/Serialization/ASTReader.cpp 
b/clang/lib/Serialization/ASTReader.cpp
index 0952244d037a77c..badd54987af18dd 100644
--- a/clang/lib/Serialization/ASTReader.cpp
+++ b/clang/lib/Serialization/ASTReader.cpp
@@ -7944,9 +7944,10 @@ void ASTReader::PrintStats() {
   std::fprintf(stderr, "*** AST File Statistics:\n");
 
   unsigned NumTypesLoaded =
-  TypesLoaded.size() - llvm::count(TypesLoaded, QualType());
+  TypesLoaded.size() - llvm::count(TypesLoaded.materialised(), QualType());
   unsigned NumDeclsLoaded =
-  DeclsLoaded.size() - llvm::count(DeclsLoaded, (Decl *)nullptr);
+  DeclsLoaded.size() -
+  llvm::count(DeclsLoaded.materialised(), (De

[PATCH] D158476: [driver] Search for compatible Android runtime directories

[Shoaib Meenai via Phabricator via cfe-commits]

smeenai added a comment.

In D158476#4647652 , @MaskRay wrote:

> nits

Sorry, this came in right as I committed the diff. I pushed 
rG915ebb07dfc53486eccf0dc09b6411929a463e98 
 to 
address it.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D158476/new/

https://reviews.llvm.org/D158476

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] 915ebb0 - [driver] Address missed feedback from https://reviews.llvm.org/D158476

[Shoaib Meenai via cfe-commits]

Author: Shoaib Meenai
Date: 2023-09-18T16:28:04-07:00
New Revision: 915ebb07dfc53486eccf0dc09b6411929a463e98

URL: 
https://github.com/llvm/llvm-project/commit/915ebb07dfc53486eccf0dc09b6411929a463e98
DIFF: 
https://github.com/llvm/llvm-project/commit/915ebb07dfc53486eccf0dc09b6411929a463e98.diff

LOG: [driver] Address missed feedback from https://reviews.llvm.org/D158476

I missed this before I committed.

Added: 


Modified: 
clang/docs/ReleaseNotes.rst
clang/test/Driver/android-unversioned-fallback-warning.cpp

Removed: 




diff  --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index 130dbe870f26d9a..9b1578762e301f6 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -346,7 +346,7 @@ Android Support
 
 - Android target triples are usually suffixed with a version. Clang searches 
for
   target-specific runtime and standard libraries in directories named after the
-  target (e.g. if you're building with ``-target 
aarch64-none-linux-android21``,
+  target (e.g. if you're building with 
``--target=aarch64-none-linux-android21``,
   Clang will look for ``lib/aarch64-none-linux-android21`` under its resource
   directory to find runtime libraries). If an exact match isn't found, Clang
   would previously fall back to a directory without any version (which would be

diff  --git a/clang/test/Driver/android-unversioned-fallback-warning.cpp 
b/clang/test/Driver/android-unversioned-fallback-warning.cpp
index c6c21d63fd41313..62a951d14effafc 100644
--- a/clang/test/Driver/android-unversioned-fallback-warning.cpp
+++ b/clang/test/Driver/android-unversioned-fallback-warning.cpp
@@ -11,12 +11,12 @@
 // RUN: mkdir -p %t/resource/lib/aarch64-none-linux-android23
 
 // Using an unversioned directory for an unversioned triple isn't a warning.
-// RUN: %clang -target aarch64-none-linux-android -ccc-install-dir %t/bin \
+// RUN: %clang --target=aarch64-none-linux-android -ccc-install-dir %t/bin \
 // RUN: -resource-dir %t/resource -### -c %s 2>&1 | \
 // RUN:   FileCheck --check-prefix=NO-WARNING %s
 // NO-WARNING-NOT: Using unversioned Android target directory
 
-// RUN: %clang -target aarch64-none-linux-android21 -ccc-install-dir %t/bin \
+// RUN: %clang --target=aarch64-none-linux-android21 -ccc-install-dir %t/bin \
 // RUN: -resource-dir %t/resource -### -c %s 2>&1 | \
 // RUN:   FileCheck --check-prefix=ANDROID21 -DDIR=%t -DSEP=%{fs-sep} %s
 // ANDROID21-DAG: Using unversioned Android target directory 
[[DIR]]/bin[[SEP]]..[[SEP]]include[[SEP]]aarch64-none-linux-android
@@ -24,10 +24,10 @@
 // ANDROID21-DAG: Using unversioned Android target directory 
[[DIR]]/resource[[SEP]]lib[[SEP]]aarch64-none-linux-android
 
 // 23 or newer should use the versioned directory
-// RUN: %clang -target aarch64-none-linux-android23 -ccc-install-dir %t/bin \
+// RUN: %clang --target=aarch64-none-linux-android23 -ccc-install-dir %t/bin \
 // RUN: -resource-dir %t/resource -### -c %s 2>&1 | \
 // RUN:   FileCheck --check-prefix=NO-WARNING %s
 
-// RUN: %clang -target aarch64-none-linux-android28 -ccc-install-dir %t/bin \
+// RUN: %clang --target=aarch64-none-linux-android28 -ccc-install-dir %t/bin \
 // RUN: -resource-dir %t/resource -### -c %s 2>&1 | \
 // RUN:   FileCheck --check-prefix=NO-WARNING %s



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Introduce paged vector (PR #66430)

[Giulio Eulisse via cfe-commits]

https://github.com/ktf resolved https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [RISCV] Recognize veyron-v1 processor in clang driver. (PR #66703)

[Michael Maitland via cfe-commits]

https://github.com/michaelmaitland approved this pull request.

LGTM.

https://github.com/llvm/llvm-project/pull/66703
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][doc] Add documentation for the ASTs used to represent C++ templates (PR #66436)

[Scott McPeak via cfe-commits]

smcpeak wrote:

And more generally, is there a preferred overall approach to documenting AST 
structures?  My basic idea was to follow the example set by the [The AST 
Library](https://clang.llvm.org/docs/InternalsManual.html#the-ast-library) in 
the Internals Manual, but expanded to cover templates.  I'm open to doing 
something completely different, for example primarily or exclusively adding 
Doxygen comments.  The goal is simply to make it easier for someone new to 
learn the ASTs used to represent templates.

https://github.com/llvm/llvm-project/pull/66436
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D158476: [driver] Search for compatible Android runtime directories

[Shoaib Meenai via Phabricator via cfe-commits]

This revision was automatically updated to reflect the committed changes.
Closed by commit rG58288c6c1214: [driver] Search for compatible Android runtime 
directories (authored by smeenai).

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D158476/new/

https://reviews.llvm.org/D158476

Files:
  clang/docs/ReleaseNotes.rst
  clang/include/clang/Basic/DiagnosticDriverKinds.td
  clang/include/clang/Driver/ToolChain.h
  clang/lib/Driver/ToolChain.cpp
  
clang/test/Driver/Inputs/basic_android_libcxx_tree/usr/lib/aarch64-unknown-linux-android21/libc++.so
  
clang/test/Driver/Inputs/basic_android_libcxx_tree/usr/lib/aarch64-unknown-linux-android23/libc++.so
  
clang/test/Driver/Inputs/basic_android_libcxx_tree/usr/lib/aarch64-unknown-linux-android29/libc++.so
  
clang/test/Driver/Inputs/resource_dir_with_per_target_subdir/lib/aarch64-unknown-linux-android21/libclang_rt.builtins.a
  
clang/test/Driver/Inputs/resource_dir_with_per_target_subdir/lib/aarch64-unknown-linux-android23/libclang_rt.builtins.a
  
clang/test/Driver/Inputs/resource_dir_with_per_target_subdir/lib/aarch64-unknown-linux-android29/libclang_rt.builtins.a
  clang/test/Driver/android-installed-libcxx.cpp
  clang/test/Driver/android-unversioned-fallback-warning.cpp
  clang/test/Driver/linux-per-target-runtime-dir.c

Index: clang/test/Driver/linux-per-target-runtime-dir.c
===
--- clang/test/Driver/linux-per-target-runtime-dir.c
+++ clang/test/Driver/linux-per-target-runtime-dir.c
@@ -17,12 +17,25 @@
 // RUN: %clang --target=aarch64-unknown-linux-android21 -print-file-name=libc++.so 2>&1 \
 // RUN: -ccc-install-dir %S/Inputs/basic_android_libcxx_tree/usr/bin \
 // RUN:   | FileCheck --check-prefix=CHECK-LIBCXX-ANDROID21 %s
-// CHECK-LIBCXX-ANDROID21: ..{{/|\\}}lib{{/|\\}}aarch64-unknown-linux-android21{{/|\\}}libc++.so
+// CHECK-LIBCXX-ANDROID21: ..{{/|\\}}lib{{/|\\}}aarch64-unknown-linux-android{{/|\\}}libc++.so
 
 // RUN: %clang --target=aarch64-unknown-linux-android23 -print-file-name=libc++.so 2>&1 \
 // RUN: -ccc-install-dir %S/Inputs/basic_android_libcxx_tree/usr/bin \
 // RUN:   | FileCheck --check-prefix=CHECK-LIBCXX-ANDROID23 %s
-// CHECK-LIBCXX-ANDROID23: ..{{/|\\}}lib{{/|\\}}aarch64-unknown-linux-android{{/|\\}}libc++.so
+// CHECK-LIBCXX-ANDROID23: ..{{/|\\}}lib{{/|\\}}aarch64-unknown-linux-android23{{/|\\}}libc++.so
+
+// RUN: %clang --target=aarch64-unknown-linux-android26 -print-file-name=libc++.so 2>&1 \
+// RUN: -ccc-install-dir %S/Inputs/basic_android_libcxx_tree/usr/bin \
+// RUN:   | FileCheck --check-prefix=CHECK-LIBCXX-ANDROID23 %s
+
+// RUN: %clang --target=aarch64-unknown-linux-android29 -print-file-name=libc++.so 2>&1 \
+// RUN: -ccc-install-dir %S/Inputs/basic_android_libcxx_tree/usr/bin \
+// RUN:   | FileCheck --check-prefix=CHECK-LIBCXX-ANDROID29 %s
+// CHECK-LIBCXX-ANDROID29: ..{{/|\\}}lib{{/|\\}}aarch64-unknown-linux-android29{{/|\\}}libc++.so
+
+// RUN: %clang --target=aarch64-unknown-linux-android31 -print-file-name=libc++.so 2>&1 \
+// RUN: -ccc-install-dir %S/Inputs/basic_android_libcxx_tree/usr/bin \
+// RUN:   | FileCheck --check-prefix=CHECK-LIBCXX-ANDROID29 %s
 
 // RUN: %clang --target=aarch64-unknown-linux-android -print-file-name=libc++.so 2>&1 \
 // RUN: -ccc-install-dir %S/Inputs/basic_android_libcxx_tree/usr/bin \
@@ -45,13 +58,29 @@
 // RUN: --target=aarch64-unknown-linux-android21 \
 // RUN: -resource-dir=%S/Inputs/resource_dir_with_per_target_subdir \
 // RUN:   | FileCheck --check-prefix=CHECK-FILE-NAME-ANDROID21 %s
-// CHECK-FILE-NAME-ANDROID21: lib{{/|\\}}aarch64-unknown-linux-android21{{/|\\}}libclang_rt.builtins.a
+// CHECK-FILE-NAME-ANDROID21: lib{{/|\\}}aarch64-unknown-linux-android{{/|\\}}libclang_rt.builtins.a
 
 // RUN: %clang -rtlib=compiler-rt -print-file-name=libclang_rt.builtins.a 2>&1 \
 // RUN: --target=aarch64-unknown-linux-android23 \
 // RUN: -resource-dir=%S/Inputs/resource_dir_with_per_target_subdir \
 // RUN:   | FileCheck --check-prefix=CHECK-FILE-NAME-ANDROID23 %s
-// CHECK-FILE-NAME-ANDROID23: lib{{/|\\}}aarch64-unknown-linux-android{{/|\\}}libclang_rt.builtins.a
+// CHECK-FILE-NAME-ANDROID23: lib{{/|\\}}aarch64-unknown-linux-android23{{/|\\}}libclang_rt.builtins.a
+
+// RUN: %clang -rtlib=compiler-rt -print-file-name=libclang_rt.builtins.a 2>&1 \
+// RUN: --target=aarch64-unknown-linux-android26 \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_per_target_subdir \
+// RUN:   | FileCheck --check-prefix=CHECK-FILE-NAME-ANDROID23 %s
+
+// RUN: %clang -rtlib=compiler-rt -print-file-name=libclang_rt.builtins.a 2>&1 \
+// RUN: --target=aarch64-unknown-linux-android29 \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_per_target_subdir \
+// RUN:   | FileCheck --check-prefix=CHECK-FILE-NAME-ANDROID29 %s
+// CHECK-FILE-NAME-ANDROID29: lib{{/|\\}}aarch64-unknown-linux-android29{{/|\\}}libclang_rt.builtins.a
+
+// RUN: %cla

[clang] 58288c6 - [driver] Search for compatible Android runtime directories

[Shoaib Meenai via cfe-commits]

Author: Shoaib Meenai
Date: 2023-09-18T16:14:18-07:00
New Revision: 58288c6c1214f8a3a0e32e003406437652e098bd

URL: 
https://github.com/llvm/llvm-project/commit/58288c6c1214f8a3a0e32e003406437652e098bd
DIFF: 
https://github.com/llvm/llvm-project/commit/58288c6c1214f8a3a0e32e003406437652e098bd.diff

LOG: [driver] Search for compatible Android runtime directories

Android triples include a version number, which makes direct triple
comparisons for per-target runtime directory searching not always work.
Instead, look for the triple with the highest compatible version number
and use that per-target runtime directory instead. This maintains the
existing fallback to a triple without any version number, but I'm hoping
we can remove that in the future. https://discourse.llvm.org/t/62717
discusses this further.

The one remaining triple mismatch after this is that Android armv7
triples usually have an environment of `androideabi`, which Clang
normalizes to `android`. If you use the `androideabi` triple when
building the runtimes with a per-target runtimes dir, the directory will
get created with `androideabi` in its name, but Clang's triple search
uses the normalized triple and will look for an `android` directory
instead. https://reviews.llvm.org/D140925 will fix that by normalizing
triples when creating the per-target runtimes directories as well.

Reviewed By: phosek, pirama

Differential Revision: https://reviews.llvm.org/D158476

Added: 

clang/test/Driver/Inputs/basic_android_libcxx_tree/usr/lib/aarch64-unknown-linux-android23/libc++.so

clang/test/Driver/Inputs/basic_android_libcxx_tree/usr/lib/aarch64-unknown-linux-android29/libc++.so

clang/test/Driver/Inputs/resource_dir_with_per_target_subdir/lib/aarch64-unknown-linux-android23/libclang_rt.builtins.a

clang/test/Driver/Inputs/resource_dir_with_per_target_subdir/lib/aarch64-unknown-linux-android29/libclang_rt.builtins.a
clang/test/Driver/android-unversioned-fallback-warning.cpp

Modified: 
clang/docs/ReleaseNotes.rst
clang/include/clang/Basic/DiagnosticDriverKinds.td
clang/include/clang/Driver/ToolChain.h
clang/lib/Driver/ToolChain.cpp
clang/test/Driver/android-installed-libcxx.cpp
clang/test/Driver/linux-per-target-runtime-dir.c

Removed: 

clang/test/Driver/Inputs/basic_android_libcxx_tree/usr/lib/aarch64-unknown-linux-android21/libc++.so

clang/test/Driver/Inputs/resource_dir_with_per_target_subdir/lib/aarch64-unknown-linux-android21/libclang_rt.builtins.a



diff  --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index d48bcb24e74ddb8..130dbe870f26d9a 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -341,6 +341,23 @@ X86 Support
 Arm and AArch64 Support
 ^^^
 
+Android Support
+^^^
+
+- Android target triples are usually suffixed with a version. Clang searches 
for
+  target-specific runtime and standard libraries in directories named after the
+  target (e.g. if you're building with ``-target 
aarch64-none-linux-android21``,
+  Clang will look for ``lib/aarch64-none-linux-android21`` under its resource
+  directory to find runtime libraries). If an exact match isn't found, Clang
+  would previously fall back to a directory without any version (which would be
+  ``lib/aarch64-none-linux-android`` in our example). Clang will now look for
+  directories for lower versions and use the newest version it finds instead,
+  e.g. if you have ``lib/aarch64-none-linux-android21`` and
+  ``lib/aarch64-none-linux-android29``, ``-target 
aarch64-none-linux-android23``
+  will use the former and ``-target aarch64-none-linux-android30`` will use the
+  latter. Falling back to a versionless directory will now emit a warning, and
+  the fallback will be removed in Clang 19.
+
 Windows Support
 ^^^
 - Fixed an assertion failure that occurred due to a failure to propagate

diff  --git a/clang/include/clang/Basic/DiagnosticDriverKinds.td 
b/clang/include/clang/Basic/DiagnosticDriverKinds.td
index 9349ff85ca8a1d3..61ac792d6fda46a 100644
--- a/clang/include/clang/Basic/DiagnosticDriverKinds.td
+++ b/clang/include/clang/Basic/DiagnosticDriverKinds.td
@@ -749,4 +749,10 @@ def warn_drv_missing_multilib : Warning<
   InGroup>;
 def note_drv_available_multilibs : Note<
   "available multilibs are:%0">;
+
+def warn_android_unversioned_fallback : Warning<
+  "Using unversioned Android target directory %0 for target %1. Unversioned"
+  " directories will not be used in Clang 19. Provide a versioned directory"
+  " for the target version or lower instead.">,
+  InGroup>;
 }

diff  --git a/clang/include/clang/Driver/ToolChain.h 
b/clang/include/clang/Driver/ToolChain.h
index 96d76c2384a9aa6..2d0c1f826c1728a 100644
--- a/clang/include/clang/Driver/ToolChain.h
+++ b/clang/include/clang/Driver/ToolChain.h
@@ -182,6 +182,9 @@ class ToolChain {
 EffectiveTriple = std:

[PATCH] D158476: [driver] Search for compatible Android runtime directories

[Fangrui Song via Phabricator via cfe-commits]

MaskRay accepted this revision.
MaskRay added a comment.

nits




Comment at: clang/docs/ReleaseNotes.rst:295
+  target-specific runtime and standard libraries in directories named after the
+  target (e.g. if you're building with ``-target 
aarch64-none-linux-android21``,
+  Clang will look for ``lib/aarch64-none-linux-android21`` under its resource

`--target=`



Comment at: clang/test/Driver/android-unversioned-fallback-warning.cpp:14
+// Using an unversioned directory for an unversioned triple isn't a warning.
+// RUN: %clang -target aarch64-none-linux-android -ccc-install-dir %t/bin \
+// RUN: -resource-dir %t/resource -### -c %s 2>&1 | \

`--target=`

`-target ` has been deprecated since Clang 3.4


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D158476/new/

https://reviews.llvm.org/D158476

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang-tools-extra] [libc++] Prevent calling the projection more than three times (PR #66315)

[Jocelyn Castellano via cfe-commits]

https://github.com/pandaninjas updated 
https://github.com/llvm/llvm-project/pull/66315

>From ead65bfcb70be46788bc9e88c891e7ae7f91b8d7 Mon Sep 17 00:00:00 2001
From: PandaNinjas 
Date: Wed, 13 Sep 2023 17:38:17 -0700
Subject: [PATCH 01/17] [libc++] Prevent calling the projection more than three
 times

---
 libcxx/include/__algorithm/ranges_clamp.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/libcxx/include/__algorithm/ranges_clamp.h 
b/libcxx/include/__algorithm/ranges_clamp.h
index 9613f7f37720a6c..ca46675eb4b3041 100644
--- a/libcxx/include/__algorithm/ranges_clamp.h
+++ b/libcxx/include/__algorithm/ranges_clamp.h
@@ -37,9 +37,10 @@ struct __fn {
 _LIBCPP_ASSERT_UNCATEGORIZED(!bool(std::invoke(__comp, std::invoke(__proj, 
__high), std::invoke(__proj, __low))),
  "Bad bounds passed to std::ranges::clamp");
 
-if (std::invoke(__comp, std::invoke(__proj, __value), std::invoke(__proj, 
__low)))
+auto &projection = std::invoke(__proj, __value);
+if (std::invoke(__comp, projection, std::invoke(__proj, __low)))
   return __low;
-else if (std::invoke(__comp, std::invoke(__proj, __high), 
std::invoke(__proj, __value)))
+else if (std::invoke(__comp, std::invoke(__proj, __high), projection))
   return __high;
 else
   return __value;

>From c18d60870ac342a95a5528396a8e0c7b91717cbb Mon Sep 17 00:00:00 2001
From: PandaNinjas 
Date: Wed, 13 Sep 2023 18:56:44 -0700
Subject: [PATCH 02/17] [libc++] Run clang-format on file

---
 libcxx/include/__algorithm/ranges_clamp.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/libcxx/include/__algorithm/ranges_clamp.h 
b/libcxx/include/__algorithm/ranges_clamp.h
index ca46675eb4b3041..3469a6419b2270f 100644
--- a/libcxx/include/__algorithm/ranges_clamp.h
+++ b/libcxx/include/__algorithm/ranges_clamp.h
@@ -37,7 +37,7 @@ struct __fn {
 _LIBCPP_ASSERT_UNCATEGORIZED(!bool(std::invoke(__comp, std::invoke(__proj, 
__high), std::invoke(__proj, __low))),
  "Bad bounds passed to std::ranges::clamp");
 
-auto &projection = std::invoke(__proj, __value);
+auto& projection = std::invoke(__proj, __value);
 if (std::invoke(__comp, projection, std::invoke(__proj, __low)))
   return __low;
 else if (std::invoke(__comp, std::invoke(__proj, __high), projection))

>From b40e791f0e9fedbb19936851e1e71decf00331fa Mon Sep 17 00:00:00 2001
From: Jocelyn Castellano 
Date: Wed, 13 Sep 2023 19:11:20 -0700
Subject: [PATCH 03/17] [libcxx] CamelCase projection and make variable name
 more descriptive

---
 libcxx/include/__algorithm/ranges_clamp.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/libcxx/include/__algorithm/ranges_clamp.h 
b/libcxx/include/__algorithm/ranges_clamp.h
index 3469a6419b2270f..3adb5fa828e1ee5 100644
--- a/libcxx/include/__algorithm/ranges_clamp.h
+++ b/libcxx/include/__algorithm/ranges_clamp.h
@@ -37,7 +37,7 @@ struct __fn {
 _LIBCPP_ASSERT_UNCATEGORIZED(!bool(std::invoke(__comp, std::invoke(__proj, 
__high), std::invoke(__proj, __low))),
  "Bad bounds passed to std::ranges::clamp");
 
-auto& projection = std::invoke(__proj, __value);
+auto& ValueProjection = std::invoke(__proj, __value);
 if (std::invoke(__comp, projection, std::invoke(__proj, __low)))
   return __low;
 else if (std::invoke(__comp, std::invoke(__proj, __high), projection))

>From a8907624defa4cc4f47520a2d93a8bd042816aa2 Mon Sep 17 00:00:00 2001
From: Jocelyn Castellano 
Date: Wed, 13 Sep 2023 19:11:47 -0700
Subject: [PATCH 04/17] [libcxx] properly change variable name

---
 libcxx/include/__algorithm/ranges_clamp.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/libcxx/include/__algorithm/ranges_clamp.h 
b/libcxx/include/__algorithm/ranges_clamp.h
index 3adb5fa828e1ee5..3d7a224b3649a3f 100644
--- a/libcxx/include/__algorithm/ranges_clamp.h
+++ b/libcxx/include/__algorithm/ranges_clamp.h
@@ -38,9 +38,9 @@ struct __fn {
  "Bad bounds passed to std::ranges::clamp");
 
 auto& ValueProjection = std::invoke(__proj, __value);
-if (std::invoke(__comp, projection, std::invoke(__proj, __low)))
+if (std::invoke(__comp, ValueProjection, std::invoke(__proj, __low)))
   return __low;
-else if (std::invoke(__comp, std::invoke(__proj, __high), projection))
+else if (std::invoke(__comp, std::invoke(__proj, __high), ValueProjection))
   return __high;
 else
   return __value;

>From 15d3b2b79fbd61f97b0312e0913cede36b5b202d Mon Sep 17 00:00:00 2001
From: Jocelyn Castellano 
Date: Thu, 14 Sep 2023 10:37:34 -0700
Subject: [PATCH 05/17] Apply suggestions from code review

---
 libcxx/include/__algorithm/ranges_clamp.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/libcxx/include/__algorithm/ranges_clamp.h 
b/libcxx/include/__algorithm/ranges_clamp.h
index 3d7a224b3649a3f..

[clang] [clang][Driver] Obey -fuse-ld=... for -print-prog-name=ld output (PR #66698)

[Roland McGrath via cfe-commits]

frobtech wrote:

> Can you comment on what projects need this? Can't we fix the projects 
> instead? And what do they do with the linker path? Why don't they invoke the 
> clang driver to run a link action?

They are legion. It comes from autoconf checks for libtool. It doesn't really 
matter why they're doing it. They are doing it and it will be years before they 
can all be changed. If you want to use Clang as the compiler for random open 
source code, it needs to be compatible with GCC in this regard.

> > ISTM that GetLinkerPath ought to produce the right name regardless and if 
> > it doesn't we can fix it to do so.
> 
> It is, but the current test will fail.

I'll be happy to adjust the test so that it matches all the right names, or so 
that it's restricted to only running on the targets where we know what the 
names might be. Can you suggest how to do that?

https://github.com/llvm/llvm-project/pull/66698
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] Enable descriptions for --print-supported-extensions (PR #66715)

[Balint Cristian via cfe-commits]

https://github.com/cbalint13 updated 
https://github.com/llvm/llvm-project/pull/66715

>From 671cbe3bbb0d201d6d938dbc0e2993b074dd1001 Mon Sep 17 00:00:00 2001
From: Balint Cristian 
Date: Tue, 19 Sep 2023 02:13:19 +0300
Subject: [PATCH] [clang] Enable descriptions for --print-supported-extensions

---
 clang/tools/driver/cc1_main.cpp   | 15 ++---
 llvm/include/llvm/MC/MCSubtargetInfo.h|  7 +++
 llvm/include/llvm/Support/RISCVISAInfo.h  |  2 +-
 .../llvm/TargetParser/AArch64TargetParser.h   |  3 ++-
 .../llvm/TargetParser/ARMTargetParser.h   |  3 ++-
 llvm/lib/Support/RISCVISAInfo.cpp | 21 +--
 llvm/lib/TargetParser/AArch64TargetParser.cpp | 16 ++
 llvm/lib/TargetParser/ARMTargetParser.cpp | 15 +
 llvm/unittests/Support/RISCVISAInfoTest.cpp   |  4 +++-
 .../TargetParser/TargetParserTest.cpp |  7 +--
 10 files changed, 70 insertions(+), 23 deletions(-)

diff --git a/clang/tools/driver/cc1_main.cpp b/clang/tools/driver/cc1_main.cpp
index f0d7b5c3889dc1f..a963324705db92d 100644
--- a/clang/tools/driver/cc1_main.cpp
+++ b/clang/tools/driver/cc1_main.cpp
@@ -28,6 +28,7 @@
 #include "llvm/ADT/Statistic.h"
 #include "llvm/Config/llvm-config.h"
 #include "llvm/LinkAllPasses.h"
+#include "llvm/MC/MCSubtargetInfo.h"
 #include "llvm/MC/TargetRegistry.h"
 #include "llvm/Option/Arg.h"
 #include "llvm/Option/ArgList.h"
@@ -198,13 +199,21 @@ static int PrintSupportedExtensions(std::string 
TargetStr) {
   std::unique_ptr TheTargetMachine(
   TheTarget->createTargetMachine(TargetStr, "", "", Options, 
std::nullopt));
   const llvm::Triple &MachineTriple = TheTargetMachine->getTargetTriple();
+  const llvm::MCSubtargetInfo *MCInfo = TheTargetMachine->getMCSubtargetInfo();
+  const llvm::ArrayRef Features =
+MCInfo->getAllProcessorFeatures();
+
+  std::map llvmDescMap;
+  for (const llvm::SubtargetFeatureKV &feature : Features) {
+llvmDescMap.insert(std::make_pair(feature.Key, feature.Desc));
+  }
 
   if (MachineTriple.isRISCV())
-llvm::riscvExtensionsHelp();
+llvm::riscvExtensionsHelp(llvmDescMap);
   else if (MachineTriple.isAArch64())
-llvm::AArch64::PrintSupportedExtensions();
+llvm::AArch64::PrintSupportedExtensions(llvmDescMap);
   else if (MachineTriple.isARM())
-llvm::ARM::PrintSupportedExtensions();
+llvm::ARM::PrintSupportedExtensions(llvmDescMap);
   else {
 // The option was already checked in Driver::HandleImmediateArgs,
 // so we do not expect to get here if we are not a supported architecture.
diff --git a/llvm/include/llvm/MC/MCSubtargetInfo.h 
b/llvm/include/llvm/MC/MCSubtargetInfo.h
index c1533ac8d0059f5..45c58988985e638 100644
--- a/llvm/include/llvm/MC/MCSubtargetInfo.h
+++ b/llvm/include/llvm/MC/MCSubtargetInfo.h
@@ -230,10 +230,17 @@ class MCSubtargetInfo {
 return Found != ProcDesc.end() && StringRef(Found->Key) == CPU;
   }
 
+  /// Return processor descriptions.
   ArrayRef getAllProcessorDescriptions() const {
 return ProcDesc;
   }
 
+  /// Return processor features.
+  ArrayRef getAllProcessorFeatures() const {
+return ProcFeatures;
+  }
+
+
   virtual unsigned getHwMode() const { return 0; }
 
   /// Return the cache size in bytes for the given level of cache.
diff --git a/llvm/include/llvm/Support/RISCVISAInfo.h 
b/llvm/include/llvm/Support/RISCVISAInfo.h
index 9092fe5c272a994..1555f2f9909983e 100644
--- a/llvm/include/llvm/Support/RISCVISAInfo.h
+++ b/llvm/include/llvm/Support/RISCVISAInfo.h
@@ -22,7 +22,7 @@ struct RISCVExtensionInfo {
   unsigned MinorVersion;
 };
 
-void riscvExtensionsHelp();
+void riscvExtensionsHelp(std::map llvmDescMap);
 
 class RISCVISAInfo {
 public:
diff --git a/llvm/include/llvm/TargetParser/AArch64TargetParser.h 
b/llvm/include/llvm/TargetParser/AArch64TargetParser.h
index 190f482044083c0..e050f840fd0a483 100644
--- a/llvm/include/llvm/TargetParser/AArch64TargetParser.h
+++ b/llvm/include/llvm/TargetParser/AArch64TargetParser.h
@@ -19,6 +19,7 @@
 #include "llvm/ADT/StringRef.h"
 #include "llvm/Support/VersionTuple.h"
 #include 
+#include 
 #include 
 
 namespace llvm {
@@ -663,7 +664,7 @@ bool isX18ReservedByDefault(const Triple &TT);
 // themselves, they are sequential (0, 1, 2, 3, ...).
 uint64_t getCpuSupportsMask(ArrayRef FeatureStrs);
 
-void PrintSupportedExtensions();
+void PrintSupportedExtensions(std::map  llvmDescMap);
 
 } // namespace AArch64
 } // namespace llvm
diff --git a/llvm/include/llvm/TargetParser/ARMTargetParser.h 
b/llvm/include/llvm/TargetParser/ARMTargetParser.h
index 37a358d1fa415c9..dbc10c8e28c076c 100644
--- a/llvm/include/llvm/TargetParser/ARMTargetParser.h
+++ b/llvm/include/llvm/TargetParser/ARMTargetParser.h
@@ -17,6 +17,7 @@
 #include "llvm/ADT/StringRef.h"
 #include "llvm/Support/ARMBuildAttributes.h"
 #include "llvm/TargetParser/ARMTargetParserCommon.h"
+#include 
 #include 
 
 namespace llvm {
@@ -259,7 +260,7 @@ StringRef computeDefaultTargetABI(const Triple &TT, 
S

[PATCH] D155610: [Clang][Sema] Fix display of characters on static assertion failure

[Corentin Jabot via Phabricator via cfe-commits]

cor3ntin added inline comments.



Comment at: clang/test/SemaCXX/static-assert.cpp:287
+  static_assert((char16_t)L'ゆ' == L"C̵̭̯̠̎͌ͅť̺"[1], ""); // expected-error 
{{failed}} \
+  // expected-note {{evaluates 
to ''ゆ' (0x3086) == '̵' (0x335)'}}
+  static_assert(L"＼／"[1] == u'\xFFFD', ""); // expected-error {{failed}} \

hubert.reinterpretcast wrote:
> cor3ntin wrote:
> > hubert.reinterpretcast wrote:
> > > tahonermann wrote:
> > > > cor3ntin wrote:
> > > > > hubert.reinterpretcast wrote:
> > > > > > hubert.reinterpretcast wrote:
> > > > > > > cor3ntin wrote:
> > > > > > > > hubert.reinterpretcast wrote:
> > > > > > > > > The C++23 escaped string formatting facility would not 
> > > > > > > > > generate a trailing combining character like this. I 
> > > > > > > > > recommend following suit.
> > > > > > > > > 
> > > > > > > > > Info on U+0335: 
> > > > > > > > > https://util.unicode.org/UnicodeJsps/character.jsp?a=0335
> > > > > > > > > 
> > > > > > > > This is way outside the scope of the patch. The diagnostic 
> > > > > > > > output facility has no understanding of combining characters or 
> > > > > > > > graphemes and do not attempt to match std::print. It probably 
> > > > > > > > would be an improvement but this patch is not trying to modify 
> > > > > > > > how all diagnostics are printed. (all of that logic is in 
> > > > > > > > Diagnostic.cpp)
> > > > > > > This patch is pushing the envelope of what appears in 
> > > > > > > diagnostics. One can also argue that someone writing
> > > > > > > ```
> > > > > > > static_assert(false, "\u0301");
> > > > > > > ```
> > > > > > > gets what they deserve, but that case does not have a big problem 
> > > > > > > anyway (because the provided message text appears after `: `).
> > > > > > > 
> > > > > > > This patch increases the exposure of the diagnostic output 
> > > > > > > facility to input that it does not handle well. I disagree that 
> > > > > > > it is outside the scope of this patch to insist that it does not 
> > > > > > > generate such inputs to the diagnostic output facility (even if a 
> > > > > > > possible solution is to modify the diagnostic output facility 
> > > > > > > first).
> > > > > > @cor3ntin, do you have status quo examples for how 
> > > > > > grapheme-extending characters that are not already "problematic" in 
> > > > > > their original context are emitted in diagnostics in contexts where 
> > > > > > they are?
> > > > > Are you looking for that sort of examples? 
> > > > > https://godbolt.org/z/c79xWr7Me
> > > > > That shows that clang has no understanding of graphemes
> > > > gcc and MSVC get that case "right" (probably by accident). 
> > > > https://godbolt.org/z/Tjd6xnEon
> > > I was more looking for cases where the output grapheme includes elements 
> > > that were part of the fixed message text (gobbling quotes, etc.). Also, 
> > > this patch is in the wrong shape for handling this concern in the 
> > > diagnostic printer because the delimiting of the replacement text happens 
> > > in this patch.
> > Could you craft a message that becomes a graphene after substitution by the 
> > engine? Maybe? You would have to try very hard and `static_assert` 
> > diagnostics are not of the right shape.
> > 
> > > This patch increases the exposure of the diagnostic output facility to 
> > > input that it does not handle well. I disagree that it is outside the 
> > > scope of this patch to insist that it does not generate such inputs to 
> > > the diagnostic output facility (even if a possible solution is to modify 
> > > the diagnostic output facility first).
> > 
> > I still don't see it. None of the output produce in this patch are even 
> > close to what i could be problematic. ie this patch is only ever producing 
> > ASCII or single codepoints that gets escaped when they are not printable
> > Could you craft a message that becomes a graphene after substitution by the 
> > engine? Maybe? You would have to try very hard and `static_assert` 
> > diagnostics are not of the right shape.
> 
> That is what I meant by this patch introducing new situations.
> 
> > > This patch increases the exposure of the diagnostic output facility to 
> > > input that it does not handle well. I disagree that it is outside the 
> > > scope of this patch to insist that it does not generate such inputs to 
> > > the diagnostic output facility (even if a possible solution is to modify 
> > > the diagnostic output facility first).
> > 
> > I still don't see it. None of the output produce in this patch are even 
> > close to what i could be problematic. ie this patch is only ever producing 
> > ASCII or single codepoints that gets escaped when they are not printable
> 
> This patch produces single codepoints that are not escaped even when they may 
> combine with a `'` delimiter. This patch also (currently) forms the string 
> with the `'` and the potentially-combining character directly

[clang] Introduce paged vector (PR #66430)

[Giulio Eulisse via cfe-commits]

https://github.com/ktf resolved https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Introduce paged vector (PR #66430)

[Giulio Eulisse via cfe-commits]

https://github.com/ktf resolved https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Introduce paged vector (PR #66430)

[Giulio Eulisse via cfe-commits]

https://github.com/ktf resolved https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Introduce paged vector (PR #66430)

[Giulio Eulisse via cfe-commits]

@@ -7944,9 +7944,13 @@ void ASTReader::PrintStats() {
   std::fprintf(stderr, "*** AST File Statistics:\n");
 
   unsigned NumTypesLoaded =
-  TypesLoaded.size() - llvm::count(TypesLoaded, QualType());
+  TypesLoaded.size() - std::count(TypesLoaded.materialisedBegin(),

ktf wrote:

Now it does.

https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [MSVC, ARM64] Add _Copy* and _Count* intrinsics (PR #66554)

[Amy Huang via cfe-commits]

@@ -10752,6 +10752,68 @@ Value 
*CodeGenFunction::EmitAArch64BuiltinExpr(unsigned BuiltinID,
 return Load;
   }
 
+  if (BuiltinID == AArch64::BI_CopyDoubleFromInt64 ||
+  BuiltinID == AArch64::BI_CopyFloatFromInt32 ||
+  BuiltinID == AArch64::BI_CopyInt32FromFloat ||
+  BuiltinID == AArch64::BI_CopyInt64FromDouble) {
+return EmitScalarExpr(E->getArg(0));

amykhuang wrote:

Oh, ok -- added a int to float cast. It seems like MSVC only emits an `fmov` 
for this code, so for an example snippet

```
double copy_double_from_int64(__int64 w) {
  return _CopyDoubleFromInt64((__int64) w);
}
```

MSVC assembly output looks like

```
sub   sp,sp,#0x10
str x0,[sp]
ldr x8,[sp]
fmov d0,x8
add   sp,sp,#0x10
ret
```

whereas clang assembly output (with the cast) looks like 

```
sub sp, sp, #16
str   x0, [sp, #8]
ldr  d0, [sp, #8]
scvtf   d0, d0
addsp, sp, #16
ret
```

https://github.com/llvm/llvm-project/pull/66554
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D74094: Reapply: [IRGen] Emit lifetime intrinsics around temporary aggregate argument allocas

[Nick Desaulniers via Phabricator via cfe-commits]

nickdesaulniers added a comment.

(Sorry for the wall of text)

In D74094#4646297 , @tellenbach wrote:

> No real comment on the issue itself but on the example as a former Eigen 
> maintainer (sorry for the noise if that's all obvious for you):

Woah! Deus Ex Machina?

>   auto round (Tensor m) {
>   return (m + 0.5f).cast().cast();
>   }
>
> does not return a Tensor but an expression encoding the computation which is 
> evaluated during the assignment `const Tensor t3 = round(a.log().exp());` 
> using an overloaded `operator=`. With "evaluation" I mean evaluation the in 
> the sense of performing the intended mathematical computation.
>
> Many things can go wrong when using `auto` in such cases, of which two seem 
> to be relevant here:
>
> 1. Eigen can (this is an implementation detail(!)) decide to evaluate 
> subexpressions into temporaries. The returned expression would then reference 
> the result of such an evaluation beyond its lifetime.
> 2. If 1. does not happen, the unevaluated expression would reference its 
> arguments. The immediate `0.5f` is copied since that's cheap, but the tensor 
> would be referenced. Your example function `round` takes its parameter 
> by-value and the returned expression would reference it. I'm unsure if the 
> lifetime would be extended in this case (again, the exact details of C++ 
> lifetime rules are not my area of expertise). I think if the lifetime would 
> be extended, ASAN complaining after applying this patch is wrong, if the 
> lifetime is not extended ASAN should complain and the example is wrong.
>
> Btw, these issues are so common that Eigen documents the recommendation to 
> avoid using `auto` with Eigen types all together: 
> https://eigen.tuxfamily.org/dox/TopicPitfalls.html#title3

Thanks for the info, https://eigen.tuxfamily.org/dox/TopicPitfalls.html#title3 
is quite helpful.  That helps explain why the return value was deduced to a 
`const Eigen::TensorConversionOp<`. Sounds like this is some kind of lazy 
evaluation?

Of note, I modified the reduced test case to use a static function with `auto` 
return type. But the code as written originally used a lambda!

  auto round = [](Tensor m) {
return (m + 0.5f).cast().cast();
  }

So the original author was well intentioned; generally you want to use `auto` 
rather than the full type of the lambda.  But the author may not have been 
aware of the advice or hazards described by 
https://eigen.tuxfamily.org/dox/TopicPitfalls.html#title3.  Our fix which we've 
committed is to use the `->` for the return type of the lambda.

  -auto round = [](Tensor m) {
  +auto round = [](Tensor m) -> Tensor {
 return (m + 0.5f).cast().cast();
   }

Which I think is inline wrt. the advice in 
https://eigen.tuxfamily.org/dox/TopicPitfalls.html#title3. (i.e. evaluate the 
expression so that we don't have some sort of partially evaluated calculation 
containing references to temporaries).  So I think the code in question (that 
we reverted this over) was just wrong.

One thing that's problematic is that neither clang or gcc do a great job with 
`-Wreturn-stack-address` (clang) or `-Wreturn-local-addr` (GCC): 
https://godbolt.org/z/6oq5nKnY7 Clang only spots obviously incorrect examples; 
GCC depends on optimization level.

I wonder if there's anything more we can do to help spot these at compile-time 
during semantic analysis.  It seems for more complex types, it could even be 
very beneficial.  The examples in eigen come to mind (could we catch those, for 
example?)

---

In D74094#4646954 , @rjmccall wrote:

> Okay, thanks, I can see how that works, and I definitely would never had 
> figured that out from just looking at this code.  The formal lifetime of a 
> parameter of type `Tensor` is definitely in the land of 
> implementation-defined behavior, so this code seems to be at best 
> non-portable.

Right, so that was my first question here; sounds like the optimization is 
correct wrt. the example that broke.  Though I still could not tell you 
precisely why.  Perhaps, "`round` returns an object, but one that contains 
references to temporary variables`."

> Nick, maybe we can take a new look at this patch from that perspective.  
> You've been trying to use very tight lifetime bounds for these objects that 
> only cover the duration of call, which essentially means you're defining the 
> lifetime of parameter objects to just be the call rather than the 
> full-expression (at least, when the parameter type doesn't have a 
> destructor).  In the abstract, that's probably the right rule for Clang to 
> adopt, because I think it matches programmer intuition (it's always wrong to 
> return the address of a local, and that includes by-value parameters), and it 
> means we'd have a consistent rule for all types and targets.  It's also a 
> fairly aggressive rule that's likely to uncover a fair amount of code like 
> this that's 

[clang] [clang][CodeGen] Emit annotations for function declarations. (PR #66716)

[Brendan Dahl via cfe-commits]

brendandahl wrote:

This is relanding the patch from [here](https://reviews.llvm.org/D156172). It 
fixes the [backout 
failure](https://reviews.llvm.org/rG88b7e06dcf9723d0869b0c6bee030b4140e4366d) 
and adds a test for it.

https://github.com/llvm/llvm-project/pull/66716
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Introduce paged vector (PR #66430)

[Giulio Eulisse via cfe-commits]

https://github.com/ktf updated https://github.com/llvm/llvm-project/pull/66430

>From e99bda447b2de2d356539028db7503bdc2e93273 Mon Sep 17 00:00:00 2001
From: Giulio Eulisse <10544+...@users.noreply.github.com>
Date: Thu, 14 Sep 2023 21:58:21 +0200
Subject: [PATCH] Introduce PagedVector class

The goal of the class is to be an (almost) drop in replacement for
SmallVector and std::vector when those are presized and filled later,
as it happens in SourceManager and ASTReader.

By splitting the actual vector in pages of the same size and allocating
the pages only when they are needed, using this containers reduces the
memory usage by a factor 4 for the cases relevant to the ALICE
experiment ROOT / cling usage.
---
 clang/include/clang/Basic/SourceManager.h |   3 +-
 clang/include/clang/Serialization/ASTReader.h |   5 +-
 clang/lib/Basic/SourceManager.cpp |  10 +-
 clang/lib/Serialization/ASTReader.cpp |   5 +-
 llvm/docs/ProgrammersManual.rst   |  32 ++
 llvm/include/llvm/ADT/PagedVector.h   | 322 ++
 llvm/unittests/ADT/CMakeLists.txt |   1 +
 llvm/unittests/ADT/PagedVectorTest.cpp| 265 ++
 8 files changed, 633 insertions(+), 10 deletions(-)
 create mode 100644 llvm/include/llvm/ADT/PagedVector.h
 create mode 100644 llvm/unittests/ADT/PagedVectorTest.cpp

diff --git a/clang/include/clang/Basic/SourceManager.h 
b/clang/include/clang/Basic/SourceManager.h
index 2f846502d6f3327..e37caa2252532f9 100644
--- a/clang/include/clang/Basic/SourceManager.h
+++ b/clang/include/clang/Basic/SourceManager.h
@@ -43,6 +43,7 @@
 #include "llvm/ADT/DenseMap.h"
 #include "llvm/ADT/DenseSet.h"
 #include "llvm/ADT/IntrusiveRefCntPtr.h"
+#include "llvm/ADT/PagedVector.h"
 #include "llvm/ADT/PointerIntPair.h"
 #include "llvm/ADT/SmallVector.h"
 #include "llvm/ADT/StringRef.h"
@@ -699,7 +700,7 @@ class SourceManager : public RefCountedBase {
   ///
   /// Negative FileIDs are indexes into this table. To get from ID to an index,
   /// use (-ID - 2).
-  SmallVector LoadedSLocEntryTable;
+  llvm::PagedVector LoadedSLocEntryTable;
 
   /// The starting offset of the next local SLocEntry.
   ///
diff --git a/clang/include/clang/Serialization/ASTReader.h 
b/clang/include/clang/Serialization/ASTReader.h
index dc1eb21c27801fe..65e19c6e44cf571 100644
--- a/clang/include/clang/Serialization/ASTReader.h
+++ b/clang/include/clang/Serialization/ASTReader.h
@@ -38,6 +38,7 @@
 #include "llvm/ADT/DenseSet.h"
 #include "llvm/ADT/IntrusiveRefCntPtr.h"
 #include "llvm/ADT/MapVector.h"
+#include "llvm/ADT/PagedVector.h"
 #include "llvm/ADT/STLExtras.h"
 #include "llvm/ADT/SetVector.h"
 #include "llvm/ADT/SmallPtrSet.h"
@@ -487,7 +488,7 @@ class ASTReader
   ///
   /// When the pointer at index I is non-NULL, the type with
   /// ID = (I + 1) << FastQual::Width has already been loaded
-  std::vector TypesLoaded;
+  llvm::PagedVector TypesLoaded;
 
   using GlobalTypeMapType =
   ContinuousRangeMap;
@@ -501,7 +502,7 @@ class ASTReader
   ///
   /// When the pointer at index I is non-NULL, the declaration with ID
   /// = I + 1 has already been loaded.
-  std::vector DeclsLoaded;
+  llvm::PagedVector DeclsLoaded;
 
   using GlobalDeclMapType =
   ContinuousRangeMap;
diff --git a/clang/lib/Basic/SourceManager.cpp 
b/clang/lib/Basic/SourceManager.cpp
index 0521ac7b30339ab..7fa8b8096ac4931 100644
--- a/clang/lib/Basic/SourceManager.cpp
+++ b/clang/lib/Basic/SourceManager.cpp
@@ -2344,11 +2344,11 @@ SourceManager::MemoryBufferSizes 
SourceManager::getMemoryBufferSizes() const {
 }
 
 size_t SourceManager::getDataStructureSizes() const {
-  size_t size = llvm::capacity_in_bytes(MemBufferInfos)
-+ llvm::capacity_in_bytes(LocalSLocEntryTable)
-+ llvm::capacity_in_bytes(LoadedSLocEntryTable)
-+ llvm::capacity_in_bytes(SLocEntryLoaded)
-+ llvm::capacity_in_bytes(FileInfos);
+  size_t size = llvm::capacity_in_bytes(MemBufferInfos) +
+llvm::capacity_in_bytes(LocalSLocEntryTable) +
+llvm::capacity_in_bytes(LoadedSLocEntryTable) +
+llvm::capacity_in_bytes(SLocEntryLoaded) +
+llvm::capacity_in_bytes(FileInfos);
 
   if (OverriddenFilesInfo)
 size += llvm::capacity_in_bytes(OverriddenFilesInfo->OverriddenFiles);
diff --git a/clang/lib/Serialization/ASTReader.cpp 
b/clang/lib/Serialization/ASTReader.cpp
index 0952244d037a77c..badd54987af18dd 100644
--- a/clang/lib/Serialization/ASTReader.cpp
+++ b/clang/lib/Serialization/ASTReader.cpp
@@ -7944,9 +7944,10 @@ void ASTReader::PrintStats() {
   std::fprintf(stderr, "*** AST File Statistics:\n");
 
   unsigned NumTypesLoaded =
-  TypesLoaded.size() - llvm::count(TypesLoaded, QualType());
+  TypesLoaded.size() - llvm::count(TypesLoaded.materialised(), QualType());
   unsigned NumDeclsLoaded =
-  DeclsLoaded.size() - llvm::count(DeclsLoaded, (Decl *)nullptr);
+  DeclsLoaded.size() -
+  llvm::count(DeclsLoaded.materialised(), (De

[clang] [clang][CodeGen] Emit annotations for function declarations. (PR #66716)

[via cfe-commits]

llvmbot wrote:




@llvm/pr-subscribers-clang


Changes

Previously, annotations were only emitted for function definitions. With this 
change annotations are also emitted for declarations. Also, emitting function 
annotations is now deferred until the end so that the most up to date 
declaration is used which will have any inherited annotations.

---
Full diff: https://github.com/llvm/llvm-project/pull/66716.diff


9 Files Affected:

- (modified) clang/lib/CodeGen/CodeGenModule.cpp (+21-2) 
- (modified) clang/lib/CodeGen/CodeGenModule.h (+4) 
- (added) clang/test/CodeGen/annotations-decl-use-decl.c (+16) 
- (added) clang/test/CodeGen/annotations-decl-use-define.c (+16) 
- (added) clang/test/CodeGen/annotations-declaration.c (+17) 
- (modified) clang/test/CodeGen/annotations-global.c (+4-4) 
- (added) clang/test/CodeGenCXX/attr-annotate-constructor.cpp (+10) 
- (added) clang/test/CodeGenCXX/attr-annotate-destructor.cpp (+10) 
- (modified) clang/test/CodeGenCXX/attr-annotate.cpp (+3-3) 


``diff
diff --git a/clang/lib/CodeGen/CodeGenModule.cpp 
b/clang/lib/CodeGen/CodeGenModule.cpp
index 8b0c9340775cbe9..5108e6c91bfb30c 100644
--- a/clang/lib/CodeGen/CodeGenModule.cpp
+++ b/clang/lib/CodeGen/CodeGenModule.cpp
@@ -697,6 +697,7 @@ void CodeGenModule::checkAliases() {
 void CodeGenModule::clear() {
   DeferredDeclsToEmit.clear();
   EmittedDeferredDecls.clear();
+  DeferredAnnotations.clear();
   if (OpenMPRuntime)
 OpenMPRuntime->clear();
 }
@@ -3093,6 +3094,13 @@ void CodeGenModule::EmitVTablesOpportunistically() {
 }
 
 void CodeGenModule::EmitGlobalAnnotations() {
+  for (const auto& [MangledName, VD] : DeferredAnnotations) {
+llvm::GlobalValue *GV = GetGlobalValue(MangledName);
+if (GV)
+  AddGlobalAnnotations(VD, GV);
+  }
+  DeferredAnnotations.clear();
+
   if (Annotations.empty())
 return;
 
@@ -3597,6 +3605,14 @@ void CodeGenModule::EmitGlobal(GlobalDecl GD) {
 
   // Ignore declarations, they will be emitted on their first use.
   if (const auto *FD = dyn_cast(Global)) {
+// Update deferred annotations with the latest declaration if the function
+// function was already used or defined.
+if (FD->hasAttr()) {
+  StringRef MangledName = getMangledName(GD);
+  if (GetGlobalValue(MangledName))
+DeferredAnnotations[MangledName] = FD;
+}
+
 // Forward declarations are emitted lazily on first use.
 if (!FD->doesThisDeclarationHaveABody()) {
   if (!FD->doesDeclarationForceExternallyVisibleDefinition())
@@ -4370,6 +4386,11 @@ llvm::Constant *CodeGenModule::GetOrCreateLLVMFunction(
   llvm::Function::Create(FTy, llvm::Function::ExternalLinkage,
  Entry ? StringRef() : MangledName, &getModule());
 
+  // Store the declaration associated with this function so it is potentially
+  // updated by further declarations or definitions and emitted at the end.
+  if (D && D->hasAttr())
+DeferredAnnotations[MangledName] = cast(D);
+
   // If we already created a function with the same mangled name (but different
   // type) before, take its name and add it to the list of functions to be
   // replaced with F at the end of CodeGen.
@@ -5664,8 +5685,6 @@ void 
CodeGenModule::EmitGlobalFunctionDefinition(GlobalDecl GD,
 AddGlobalCtor(Fn, CA->getPriority());
   if (const DestructorAttr *DA = D->getAttr())
 AddGlobalDtor(Fn, DA->getPriority(), true);
-  if (D->hasAttr())
-AddGlobalAnnotations(D, Fn);
   if (getLangOpts().OpenMP && D->hasAttr())
 getOpenMPRuntime().emitDeclareTargetFunction(D, GV);
 }
diff --git a/clang/lib/CodeGen/CodeGenModule.h 
b/clang/lib/CodeGen/CodeGenModule.h
index 073b471c6e3cc11..8b0d68afbd0ecd2 100644
--- a/clang/lib/CodeGen/CodeGenModule.h
+++ b/clang/lib/CodeGen/CodeGenModule.h
@@ -431,6 +431,10 @@ class CodeGenModule : public CodeGenTypeCache {
   /// Global annotations.
   std::vector Annotations;
 
+  // Store deferred function annotations so they can be emitted at the end with
+  // most up to date ValueDecl that will have all the inherited annotations.
+  llvm::DenseMap DeferredAnnotations;
+
   /// Map used to get unique annotation strings.
   llvm::StringMap AnnotationStrings;
 
diff --git a/clang/test/CodeGen/annotations-decl-use-decl.c 
b/clang/test/CodeGen/annotations-decl-use-decl.c
new file mode 100644
index 000..f43ba91a34d876f
--- /dev/null
+++ b/clang/test/CodeGen/annotations-decl-use-decl.c
@@ -0,0 +1,16 @@
+// RUN: %clang_cc1 %s -emit-llvm -o - | FileCheck %s
+
+// Test annotation attributes are still emitted when the function is used 
before
+// it is defined with annotations.
+
+void foo(void);
+void *xxx = (void*)foo;
+void __attribute__((annotate("bar"))) foo();
+
+// CHECK: target triple
+// CHECK-DAG: private unnamed_addr constant [4 x i8] c"bar\00", section 
"llvm.metadata"
+
+// CHECK: @llvm.global.annotations = appending global [1 x { ptr, ptr, ptr, 
i32, ptr }] [{
+// CHECK-SAME: { ptr @foo,
+// CHECK-SAME: }], section "llvm.metadata"

[clang] Introduce paged vector (PR #66430)

[Giulio Eulisse via cfe-commits]

https://github.com/ktf resolved https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Introduce paged vector (PR #66430)

[Giulio Eulisse via cfe-commits]

https://github.com/ktf resolved https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][CodeGen] Emit annotations for function declarations. (PR #66716)

[Brendan Dahl via cfe-commits]

https://github.com/brendandahl created 
https://github.com/llvm/llvm-project/pull/66716

Previously, annotations were only emitted for function definitions. With this 
change annotations are also emitted for declarations. Also, emitting function 
annotations is now deferred until the end so that the most up to date 
declaration is used which will have any inherited annotations.

>From 846deb6e2055a8e458530c9e27bbd512a68deb5c Mon Sep 17 00:00:00 2001
From: Brendan Dahl 
Date: Tue, 12 Sep 2023 12:53:24 -0700
Subject: [PATCH] [clang][CodeGen] Emit annotations for function declarations.

Previously, annotations were only emitted for function definitions. With
this change annotations are also emitted for declarations. Also, emitting
function annotations is now deferred until the end so that the most
up to date declaration is used which will have any inherited annotations.
---
 clang/lib/CodeGen/CodeGenModule.cpp   | 23 +--
 clang/lib/CodeGen/CodeGenModule.h |  4 
 .../test/CodeGen/annotations-decl-use-decl.c  | 16 +
 .../CodeGen/annotations-decl-use-define.c | 16 +
 clang/test/CodeGen/annotations-declaration.c  | 17 ++
 clang/test/CodeGen/annotations-global.c   |  8 +++
 .../CodeGenCXX/attr-annotate-constructor.cpp  | 10 
 .../CodeGenCXX/attr-annotate-destructor.cpp   | 10 
 clang/test/CodeGenCXX/attr-annotate.cpp   |  6 ++---
 9 files changed, 101 insertions(+), 9 deletions(-)
 create mode 100644 clang/test/CodeGen/annotations-decl-use-decl.c
 create mode 100644 clang/test/CodeGen/annotations-decl-use-define.c
 create mode 100644 clang/test/CodeGen/annotations-declaration.c
 create mode 100644 clang/test/CodeGenCXX/attr-annotate-constructor.cpp
 create mode 100644 clang/test/CodeGenCXX/attr-annotate-destructor.cpp

diff --git a/clang/lib/CodeGen/CodeGenModule.cpp 
b/clang/lib/CodeGen/CodeGenModule.cpp
index 8b0c9340775cbe9..5108e6c91bfb30c 100644
--- a/clang/lib/CodeGen/CodeGenModule.cpp
+++ b/clang/lib/CodeGen/CodeGenModule.cpp
@@ -697,6 +697,7 @@ void CodeGenModule::checkAliases() {
 void CodeGenModule::clear() {
   DeferredDeclsToEmit.clear();
   EmittedDeferredDecls.clear();
+  DeferredAnnotations.clear();
   if (OpenMPRuntime)
 OpenMPRuntime->clear();
 }
@@ -3093,6 +3094,13 @@ void CodeGenModule::EmitVTablesOpportunistically() {
 }
 
 void CodeGenModule::EmitGlobalAnnotations() {
+  for (const auto& [MangledName, VD] : DeferredAnnotations) {
+llvm::GlobalValue *GV = GetGlobalValue(MangledName);
+if (GV)
+  AddGlobalAnnotations(VD, GV);
+  }
+  DeferredAnnotations.clear();
+
   if (Annotations.empty())
 return;
 
@@ -3597,6 +3605,14 @@ void CodeGenModule::EmitGlobal(GlobalDecl GD) {
 
   // Ignore declarations, they will be emitted on their first use.
   if (const auto *FD = dyn_cast(Global)) {
+// Update deferred annotations with the latest declaration if the function
+// function was already used or defined.
+if (FD->hasAttr()) {
+  StringRef MangledName = getMangledName(GD);
+  if (GetGlobalValue(MangledName))
+DeferredAnnotations[MangledName] = FD;
+}
+
 // Forward declarations are emitted lazily on first use.
 if (!FD->doesThisDeclarationHaveABody()) {
   if (!FD->doesDeclarationForceExternallyVisibleDefinition())
@@ -4370,6 +4386,11 @@ llvm::Constant *CodeGenModule::GetOrCreateLLVMFunction(
   llvm::Function::Create(FTy, llvm::Function::ExternalLinkage,
  Entry ? StringRef() : MangledName, &getModule());
 
+  // Store the declaration associated with this function so it is potentially
+  // updated by further declarations or definitions and emitted at the end.
+  if (D && D->hasAttr())
+DeferredAnnotations[MangledName] = cast(D);
+
   // If we already created a function with the same mangled name (but different
   // type) before, take its name and add it to the list of functions to be
   // replaced with F at the end of CodeGen.
@@ -5664,8 +5685,6 @@ void 
CodeGenModule::EmitGlobalFunctionDefinition(GlobalDecl GD,
 AddGlobalCtor(Fn, CA->getPriority());
   if (const DestructorAttr *DA = D->getAttr())
 AddGlobalDtor(Fn, DA->getPriority(), true);
-  if (D->hasAttr())
-AddGlobalAnnotations(D, Fn);
   if (getLangOpts().OpenMP && D->hasAttr())
 getOpenMPRuntime().emitDeclareTargetFunction(D, GV);
 }
diff --git a/clang/lib/CodeGen/CodeGenModule.h 
b/clang/lib/CodeGen/CodeGenModule.h
index 073b471c6e3cc11..8b0d68afbd0ecd2 100644
--- a/clang/lib/CodeGen/CodeGenModule.h
+++ b/clang/lib/CodeGen/CodeGenModule.h
@@ -431,6 +431,10 @@ class CodeGenModule : public CodeGenTypeCache {
   /// Global annotations.
   std::vector Annotations;
 
+  // Store deferred function annotations so they can be emitted at the end with
+  // most up to date ValueDecl that will have all the inherited annotations.
+  llvm::DenseMap DeferredAnnotations;
+
   /// Map used to get unique annotation strings.
   llvm::St

[clang] Introduce paged vector (PR #66430)

[Giulio Eulisse via cfe-commits]

https://github.com/ktf resolved https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Introduce paged vector (PR #66430)

[Giulio Eulisse via cfe-commits]

https://github.com/ktf resolved https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Introduce paged vector (PR #66430)

[Giulio Eulisse via cfe-commits]

https://github.com/ktf resolved https://github.com/llvm/llvm-project/pull/66430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits