https://github.com/philnik777 closed
https://github.com/llvm/llvm-project/pull/67805
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/philnik777 updated
https://github.com/llvm/llvm-project/pull/67805
>From 71742243aeab0e51238eaa6df9487a0c37e97087 Mon Sep 17 00:00:00 2001
From: Nikolas Klauser
Date: Fri, 29 Sep 2023 15:45:56 +0200
Subject: [PATCH] [Clang] Add __datasizeof
---
clang/docs/LanguageExtensions
https://github.com/AaronBallman approved this pull request.
LGTM with the new codegen test, thank you!
https://github.com/llvm/llvm-project/pull/67805
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listin
https://github.com/philnik777 updated
https://github.com/llvm/llvm-project/pull/67805
>From 8010f6178df82305b5790610f0114ddfb3c0a5f8 Mon Sep 17 00:00:00 2001
From: Nikolas Klauser
Date: Fri, 29 Sep 2023 15:45:56 +0200
Subject: [PATCH] [Clang] Add __datasizeof
---
clang/docs/LanguageExtensions
@@ -0,0 +1,19 @@
+// RUN: %clang_cc1 -triple x86_64 -emit-llvm %s -o - | FileCheck %s
+
+// CHECK: define dso_local noundef i32 @_Z4testi(i32 noundef %i) #0 {
+// CHECK-NEXT: entry:
+// CHECK-NEXT: %i.addr = alloca i32, align 4
+// CHECK-NEXT: store i32 %i, ptr %i.addr, a
philnik777 wrote:
@AaronBallman This did indeed catch a bug. Also, I think there was something
about how codegen tests should be written, but I don't remember exactly
anymore. Could you take a look at the test?
https://github.com/llvm/llvm-project/pull/67805
___
https://github.com/philnik777 updated
https://github.com/llvm/llvm-project/pull/67805
>From f384bc5feab5b2a0ae71d4c96d123c3fe0d13a73 Mon Sep 17 00:00:00 2001
From: Nikolas Klauser
Date: Fri, 29 Sep 2023 15:45:56 +0200
Subject: [PATCH] [Clang] Add __datasizeof
---
clang/docs/LanguageExtensions
https://github.com/AaronBallman approved this pull request.
LGTM assuming the requested codegen tests don't bring up any surprises for you.
https://github.com/llvm/llvm-project/pull/67805
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://
@@ -0,0 +1,34 @@
+// RUN: %clang_cc1 -fsyntax-only -triple x86_64-linux-gnu -verify %s
AaronBallman wrote:
Yeah, a codegen test would make sense (just to show that we actually do emit
code for that monstrosity).
https://github.com/llvm/llvm-project/pull/67805
_
https://github.com/AaronBallman edited
https://github.com/llvm/llvm-project/pull/67805
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -0,0 +1,34 @@
+// RUN: %clang_cc1 -fsyntax-only -triple x86_64-linux-gnu -verify %s
philnik777 wrote:
It looks like the VLA test can't be constant evaluated. Should I add a CodeGen
test for that?
https://github.com/llvm/llvm-project/pull/67805
__
github-actions[bot] wrote:
:warning: C/C++ code formatter, clang-format found issues in your code.
:warning:
You can test this locally with the following command:
``bash
git-clang-format --diff 5e1c2bf3e6fca35ee0445b2a81d47e8576024186
e6770ca7735285e1e669032a5a99b07709957803 --
https://github.com/philnik777 updated
https://github.com/llvm/llvm-project/pull/67805
>From e6770ca7735285e1e669032a5a99b07709957803 Mon Sep 17 00:00:00 2001
From: Nikolas Klauser
Date: Fri, 29 Sep 2023 15:45:56 +0200
Subject: [PATCH] [Clang] Add __datasizeof
---
clang/docs/LanguageExtensions
@@ -236,6 +236,7 @@ FEATURE(shadow_call_stack,
FEATURE(tls, PP.getTargetInfo().isTLSSupported())
FEATURE(underlying_type, LangOpts.CPlusPlus)
FEATURE(experimental_library, LangOpts.ExperimentalLibrary)
+FEATURE(datasizeof, LangOpts.CPlusPlus)
philnik777 wrote:
@@ -236,6 +236,7 @@ FEATURE(shadow_call_stack,
FEATURE(tls, PP.getTargetInfo().isTLSSupported())
FEATURE(underlying_type, LangOpts.CPlusPlus)
FEATURE(experimental_library, LangOpts.ExperimentalLibrary)
+FEATURE(datasizeof, LangOpts.CPlusPlus)
AaronBallman wrote
@@ -1016,4 +1016,7 @@ def warn_unpacked_field
def warn_unaligned_access : Warning<
"field %1 within %0 is less aligned than %2 and is usually due to %0 being "
"packed, which can lead to unaligned accesses">, InGroup,
DefaultIgnore;
+
+def err_cannot_mangle_expression : Er
@@ -1016,4 +1016,7 @@ def warn_unpacked_field
def warn_unaligned_access : Warning<
"field %1 within %0 is less aligned than %2 and is usually due to %0 being "
"packed, which can lead to unaligned accesses">, InGroup,
DefaultIgnore;
+
+def err_cannot_mangle_expression : Er
@@ -236,6 +236,7 @@ FEATURE(shadow_call_stack,
FEATURE(tls, PP.getTargetInfo().isTLSSupported())
FEATURE(underlying_type, LangOpts.CPlusPlus)
FEATURE(experimental_library, LangOpts.ExperimentalLibrary)
+FEATURE(datasizeof, LangOpts.CPlusPlus)
philnik777 wrote:
@@ -236,6 +236,7 @@ FEATURE(shadow_call_stack,
FEATURE(tls, PP.getTargetInfo().isTLSSupported())
FEATURE(underlying_type, LangOpts.CPlusPlus)
FEATURE(experimental_library, LangOpts.ExperimentalLibrary)
+FEATURE(datasizeof, LangOpts.CPlusPlus)
AaronBallman wrote
@@ -1016,4 +1016,7 @@ def warn_unpacked_field
def warn_unaligned_access : Warning<
"field %1 within %0 is less aligned than %2 and is usually due to %0 being "
"packed, which can lead to unaligned accesses">, InGroup,
DefaultIgnore;
+
+def err_cannot_mangle_expression : Er
@@ -236,6 +236,7 @@ FEATURE(shadow_call_stack,
FEATURE(tls, PP.getTargetInfo().isTLSSupported())
FEATURE(underlying_type, LangOpts.CPlusPlus)
FEATURE(experimental_library, LangOpts.ExperimentalLibrary)
+FEATURE(datasizeof, LangOpts.CPlusPlus)
philnik777 wrote:
@@ -1016,4 +1016,7 @@ def warn_unpacked_field
def warn_unaligned_access : Warning<
"field %1 within %0 is less aligned than %2 and is usually due to %0 being "
"packed, which can lead to unaligned accesses">, InGroup,
DefaultIgnore;
+
+def err_cannot_mangle_expression : Er
@@ -0,0 +1,34 @@
+// RUN: %clang_cc1 -fsyntax-only -triple x86_64-linux-gnu -verify %s
AaronBallman wrote:
We should also show it's strange behavior with VLAs in C++:
```
int n = 12;
__datasizeof(int[n++]); // Surprise, n is incremented, you're welcome
```
and wi
https://github.com/AaronBallman commented:
I think this is a reasonable minor extension, but it should come with
documentation in `clang/docs/LanguageExtensions.rst` and release note in
`clang/docs/ReleaseNotes.rst` before we're done.
https://github.com/llvm/llvm-project/pull/67805
___
@@ -5038,19 +5039,19 @@ void CXXNameMangler::mangleExpression(const Expr *E,
unsigned Arity,
Out << 'a';
MangleAlignofSizeofArg();
break;
+case UETT_DataSizeOf: {
+ Context.getDiags().Report(diag::err_cannot_mangle_expression)
+ << "__datasi
https://github.com/AaronBallman edited
https://github.com/llvm/llvm-project/pull/67805
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -236,6 +236,7 @@ FEATURE(shadow_call_stack,
FEATURE(tls, PP.getTargetInfo().isTLSSupported())
FEATURE(underlying_type, LangOpts.CPlusPlus)
FEATURE(experimental_library, LangOpts.ExperimentalLibrary)
+FEATURE(datasizeof, LangOpts.CPlusPlus)
AaronBallman wrote
@@ -0,0 +1,34 @@
+// RUN: %clang_cc1 -fsyntax-only -triple x86_64-linux-gnu -verify %s
shafik wrote:
Maybe we should add tests to show parallels with `sizeof` e.g` *void* types and
incomplete types.
Also w/o parens etc
https://github.com/llvm/llvm-project/pu
https://github.com/shafik commented:
Looks good but Aaron should look at it as well.
https://github.com/llvm/llvm-project/pull/67805
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -5038,19 +5039,19 @@ void CXXNameMangler::mangleExpression(const Expr *E,
unsigned Arity,
Out << 'a';
MangleAlignofSizeofArg();
break;
+case UETT_DataSizeOf: {
+ Context.getDiags().Report(diag::err_cannot_mangle_expression)
+ << "__datasi
https://github.com/shafik edited https://github.com/llvm/llvm-project/pull/67805
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
llvmbot wrote:
@llvm/pr-subscribers-clang
Changes
The data size is required for implementing the `memmove` optimization for
`std::copy`, `std::move` etc. correctly as well as replacing
`__compressed_pair` with `[[no_unique_address]]` in libc++. Since the compiler
already knows the data s
https://github.com/philnik777 created
https://github.com/llvm/llvm-project/pull/67805
The data size is required for implementing the `memmove` optimization for
`std::copy`, `std::move` etc. correctly as well as replacing
`__compressed_pair` with `[[no_unique_address]]` in libc++. Since the com
33 matches
Mail list logo
