[clang] [llvm] [KCFI][NFC] Rename the !kcfi_type Function metadata to !cfi_type (PR #109080)
https://github.com/MaskRay requested changes to this pull request. . https://github.com/llvm/llvm-project/pull/109080 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [llvm] [KCFI][NFC] Rename the !kcfi_type Function metadata to !cfi_type (PR #109080)
MaskRay wrote: I understand the intention to generalize naming, but I feel that there is significance chance that the over-generalization may not work with the alternative CFI schemes, and this change could turn out to be needed when the other schemes go with different IR constructs. I feel that we should make such renaming at this time. BTW: when kcfi was proposed, I felt that it was not a good name because the strategy could only cover a very limited subset of LTO-based `-fsanitize=cfi`. kcfi is just like a brand name. https://github.com/llvm/llvm-project/pull/109080 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Driver][Sparc] Default to -mcpu=v9 for 32-bit Linux/sparc64 (PR #109278)
MaskRay wrote: > > Just to be certain: the revised version of the patch is ok to commit? > > I think it would be good to get an ACK from at least one of the SPARC Leon > maintainers. > > CC @andreas-gaisler Thanks for informing the maintainer. Note: this is done as a courtesy. When we bump a setting -mcpu= for good reasons, we do not add exception code like `!isVendorA() && !isVendorB()`. clangDriver is moving out of business to hard code such vendor details and requires adoption of Clang configuration files. https://github.com/llvm/llvm-project/pull/109278 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [flang] [clang][flang] Support -time in both clang and flang (PR #109165)
@@ -0,0 +1,24 @@
+// The -time option prints timing information for the various subcommands in a
+// format similar to that used by gcc. When compiling and linking, this will
+// include the time to call clang-${LLVM_VERSION_MAJOR} and the linker. Since
+// the name of the linker could vary across platforms, and name of the compiler
+// could be something different, just check that whatever is printed to stderr
+// looks like timing information.
+
+// RUN: %clang -time -c -O3 -o /dev/null %s 2>&1 \
+// RUN: | FileCheck %s --check-prefix=COMPILE-ONLY
+// RUN: %clang -time -S -emit-llvm -O3 -o /dev/null %s 2>&1 \
+// RUN: | FileCheck %s --check-prefix=COMPILE-ONLY
+// RUN: %clang -time -S -O3 -o /dev/null %s 2>&1 \
+// RUN: | FileCheck %s --check-prefix=COMPILE-ONLY
+// RUN: %clang -time -O3 -o /dev/null %s 2>&1 \
+// RUN: | FileCheck %s --check-prefix=COMPILE-AND-LINK
+
+// COMPILE-ONLY: # {{.+}} {{[0-9]+(.[0-9]+)?}} {{[0-9]+(.[0-9]+)?}}
MaskRay wrote:
add `COMPILE-ONLY-NOT: {{.}}` to ensure that there is one single line
https://github.com/llvm/llvm-project/pull/109165
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [flang] [clang][flang] Support -time in both clang and flang (PR #109165)
@@ -0,0 +1,24 @@
+// The -time option prints timing information for the various subcommands in a
+// format similar to that used by gcc. When compiling and linking, this will
+// include the time to call clang-${LLVM_VERSION_MAJOR} and the linker. Since
+// the name of the linker could vary across platforms, and name of the compiler
+// could be something different, just check that whatever is printed to stderr
+// looks like timing information.
+
+// RUN: %clang -time -c -O3 -o /dev/null %s 2>&1 \
MaskRay wrote:
This requires a hard-coded target triple like `--target=x86_64` and `REQUIRES:
x86-registered-target`.
Delete unneeded -O3.
https://github.com/llvm/llvm-project/pull/109165
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Driver][Sparc] Default to -mcpu=v9 for 32-bit Linux/sparc64 (PR #109278)
https://github.com/MaskRay approved this pull request. https://github.com/llvm/llvm-project/pull/109278 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [RISCV] Fix incorrect check prefix in riscv32-toolchain.c and riscv64-toolchain.c. NFC. (PR #109390)
https://github.com/MaskRay approved this pull request. You can use `[test]` instead of `. NFC.` `-SAME: ` is preferred when testing a string on the same line https://github.com/llvm/llvm-project/pull/109390 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Driver][Sparc] Default to -mcpu=v9 for 32-bit Linux/sparc64 (PR #109278)
MaskRay wrote: We try to restrict distribution differences to things like default linker options and library paths. Affecting `-mcpu=` seems very unintuitive. There are many Debian derivatives. It's weird that Debian uses -mcpu=v9 while others use -mcpu=v8. We should not increase `IsDebian` or `IsGentoo` use. If -mcpu=v9 seems the right thing for the majority of configurations, we can bump -mcpu=v9 for all Linux and ask other, older systems to use https://clang.llvm.org/docs/UsersManual.html#configuration-files to specify `-mcpu=v8`. https://github.com/llvm/llvm-project/pull/109278 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [llvm] [clang] Add cc1 --output-asm-variant= to set output syntax (PR #109360)
https://github.com/MaskRay edited https://github.com/llvm/llvm-project/pull/109360 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [llvm] [clang] Add cc1 --output-asm-variant= to set output syntax (PR #109360)
https://github.com/MaskRay created
https://github.com/llvm/llvm-project/pull/109360
2fcaa549a824efeb56e807fcf750a56bf985296b (2010) added cc1as option
`-output-asm-variant` (untested) to set the output syntax.
`clang -cc1as -filetype asm -output-asm-variant 1` allows AT&T input and
Intel output.
This patch renames the cc1as option (to avoid collision with -o) and
makes it available for cc1 to set output syntax. This allows different
input & output syntax:
```
echo 'asm("mov $1, %eax");' | clang -xc - -S -o - -Xclang --output-asm-variant=1
```
Note: `AsmWriterFlavor` (with a misleading name), used to initialize
MCAsmInfo::AssemblerDialect, is primarily used for assembly input, not
for output.
>From 7be5f9edad29769ae5bf2a49e861afed0c984014 Mon Sep 17 00:00:00 2001
From: Fangrui Song
Date: Thu, 19 Sep 2024 19:00:08 -0700
Subject: [PATCH] =?UTF-8?q?[=F0=9D=98=80=F0=9D=97=BD=F0=9D=97=BF]=20initia?=
=?UTF-8?q?l=20version?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Created using spr 1.3.5-bogner
---
clang/include/clang/Basic/CodeGenOptions.def | 1 +
clang/include/clang/Driver/Options.td | 5 ++--
clang/lib/CodeGen/BackendUtil.cpp | 2 ++
.../test/CodeGen/inline-asm-output-variant.c | 26 +++
clang/test/Misc/cc1as-output-asm-variant.c| 8 ++
llvm/include/llvm/MC/MCTargetOptions.h| 2 ++
llvm/lib/CodeGen/LLVMTargetMachine.cpp| 4 ++-
7 files changed, 45 insertions(+), 3 deletions(-)
create mode 100644 clang/test/CodeGen/inline-asm-output-variant.c
create mode 100644 clang/test/Misc/cc1as-output-asm-variant.c
diff --git a/clang/include/clang/Basic/CodeGenOptions.def
b/clang/include/clang/Basic/CodeGenOptions.def
index b600198998d85b..2893377e5a38be 100644
--- a/clang/include/clang/Basic/CodeGenOptions.def
+++ b/clang/include/clang/Basic/CodeGenOptions.def
@@ -96,6 +96,7 @@ CODEGENOPT(EmulatedTLS , 1, 0) ///< Set by default or
-f[no-]emulated-tls.
ENUM_CODEGENOPT(EmbedBitcode, EmbedBitcodeKind, 2, Embed_Off)
/// Inline asm dialect, -masm=(att|intel)
ENUM_CODEGENOPT(InlineAsmDialect, InlineAsmDialectKind, 1, IAD_ATT)
+CODEGENOPT(OutputAsmVariant, 2, 3) ///< Set the asm variant for output (3:
unspecified).
CODEGENOPT(ForbidGuardVariables , 1, 0) ///< Issue errors if C++ guard
variables
///< are required.
CODEGENOPT(FunctionSections , 1, 0) ///< Set when -ffunction-sections is
enabled.
diff --git a/clang/include/clang/Driver/Options.td
b/clang/include/clang/Driver/Options.td
index aa3ae92fb6ae78..2586782f3e3181 100644
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -7209,6 +7209,9 @@ def fuse_ctor_homing: Flag<["-"], "fuse-ctor-homing">,
def as_secure_log_file : Separate<["-"], "as-secure-log-file">,
HelpText<"Emit .secure_log_unique directives to this filename.">,
MarshallingInfoString>;
+def output_asm_variant : Joined<["--"], "output-asm-variant=">,
+ HelpText<"Select the asm variant (integer) to use for output">,
+ MarshallingInfoInt>;
} // let Visibility = [CC1Option, CC1AsOption]
@@ -8303,8 +8306,6 @@ def filetype : Separate<["-"], "filetype">,
HelpText<"Specify the output file type ('asm', 'null', or 'obj')">;
// Transliterate Options
-def output_asm_variant : Separate<["-"], "output-asm-variant">,
-HelpText<"Select the asm variant index to use for output">;
def show_encoding : Flag<["-"], "show-encoding">,
HelpText<"Show instruction encoding information in transliterate mode">;
def show_inst : Flag<["-"], "show-inst">,
diff --git a/clang/lib/CodeGen/BackendUtil.cpp
b/clang/lib/CodeGen/BackendUtil.cpp
index fa49763e312f13..916c92adb89309 100644
--- a/clang/lib/CodeGen/BackendUtil.cpp
+++ b/clang/lib/CodeGen/BackendUtil.cpp
@@ -509,6 +509,8 @@ static bool initTargetOptions(DiagnosticsEngine &Diags,
Options.MCOptions.X86RelaxRelocations = CodeGenOpts.X86RelaxRelocations;
Options.MCOptions.CompressDebugSections =
CodeGenOpts.getCompressDebugSections();
+ if (CodeGenOpts.OutputAsmVariant != 3) // 3 (default): not specified
+Options.MCOptions.OutputAsmVariant = CodeGenOpts.OutputAsmVariant;
Options.MCOptions.ABIName = TargetOpts.ABI;
for (const auto &Entry : HSOpts.UserEntries)
if (!Entry.IsFramework &&
diff --git a/clang/test/CodeGen/inline-asm-output-variant.c
b/clang/test/CodeGen/inline-asm-output-variant.c
new file mode 100644
index 00..5fc5c1cc09b016
--- /dev/null
+++ b/clang/test/CodeGen/inline-asm-output-variant.c
@@ -0,0 +1,26 @@
+// REQUIRES: x86-registered-target
+/// AT&T input
+// RUN: %clang_cc1 -triple x86_64 -S --output-asm-variant=0 %s -o - |
FileCheck --check-prefix=ATT %s
+// RUN: %clang_cc1 -triple x86_64 -S --output-asm-variant=1 %s -o - |
FileCheck --check-prefix=INTEL %s
+
+/// Intel input
+// RUN: %clang_cc1 -triple x86_64 -S -D INTEL -mllvm -x86-asm-syntax=intel
-inline-asm=intel %s -o -
[clang] [Driver] Add toolchain for X86_64 UEFI target (PR #76838)
https://github.com/MaskRay approved this pull request. https://github.com/llvm/llvm-project/pull/76838 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Driver] Add toolchain for X86_64 UEFI target (PR #76838)
@@ -0,0 +1,14 @@ +// RUN: %clang_cl -### %s --target=x86_64-unknown-uefi \ +// RUN: --sysroot=%S/platform -fuse-ld=lld -g 2>&1 \ +// RUN: | FileCheck -check-prefixes=CHECK %s +// CHECK: "-cc1" +// CHECK-SAME: "-triple" "x86_64-unknown-uefi" +// CHECK-SAME: "-mrelocation-model" "pic" "-pic-level" "2" +// CHECK-SAME: "-mframe-pointer=all" +// CHECK-NEXT: "-nologo" +// CHECK-SAME: "-subsystem:efi_application" +// CHECK-SAME: "-entry:EfiMain" +// CHECK-SAME: "-tsaware:no" +// CHECK-SAME: "-dll" +// CHECK-SAME: "-debug" + MaskRay wrote: delete trailing blank line https://github.com/llvm/llvm-project/pull/76838 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Driver] Add toolchain for X86_64 UEFI target (PR #76838)
@@ -22,6 +22,10 @@ // RUN: FileCheck --check-prefix=X86_64 %s // X86_64: target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128" +// RUN: %clang_cc1 -triple x86_64-unknown-uefi -emit-llvm -o - %s | \ MaskRay wrote: With clang/test/CodeGen/X86/uefi-data-layout.c, this is unneeded https://github.com/llvm/llvm-project/pull/76838 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [flang] [mlir] [clang][flang][mlir] Support -frecord-command-line option (PR #102975)
https://github.com/MaskRay approved this pull request. LGTM. As you are renaming functions, you can fix the function naming style `tools::function` instead of `tools::Function`. While the style is largely inconsistent, we usually fix the style while renaming functions. https://github.com/llvm/llvm-project/pull/102975 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Clang][Driver][AArch64] Add support for aarch64-amazon-linux triple (PR #109263)
https://github.com/MaskRay requested changes to this pull request. . https://github.com/llvm/llvm-project/pull/109263 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Clang][Driver][AArch64] Add support for aarch64-amazon-linux triple (PR #109263)
MaskRay wrote: > The behviour was diverged between X86 and aarch64: prior to this patch, a > standard clang with no configuration works on x86 but not aarch64 amazon > linux; the latter fails to find C++ headers and libraries. This is not an issue of the source code. The system needs to do `-DLLVM_DEFAULT_TARGET_TRIPLE=aarch64-amazon-linux`. The comment is here for reasons as it doesn't scale to encode every possible target triple. https://github.com/llvm/llvm-project/pull/109263 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [UEFI] X86_64 UEFI Clang Driver (PR #76838)
MaskRay wrote: `[Driver] Add toolchain for ...' might be a more conventional patch subject. https://github.com/llvm/llvm-project/pull/76838 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [UEFI] X86_64 UEFI Clang Driver (PR #76838)
@@ -0,0 +1,4 @@ +// RUN: %clang -target x86_64-unknown-uefi -S -emit-llvm -o - %s | \ +// RUN: FileCheck --check-prefix=X86_64_UEFI %s +// X86_64_UEFI: target datalayout = "e-m:w-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128" + MaskRay wrote: delete trailing blank line https://github.com/llvm/llvm-project/pull/76838 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [UEFI] X86_64 UEFI Clang Driver (PR #76838)
@@ -0,0 +1,88 @@ +//===--- UEFI.h - UEFI ToolChain Implementations --*- C++ -*-===// MaskRay wrote: first line is incorrect. ensure 80-char wide and omit `C++` for `.cpp` files. https://github.com/llvm/llvm-project/pull/76838 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [UEFI] X86_64 UEFI Clang Driver (PR #76838)
@@ -0,0 +1,4 @@ +// RUN: %clang -target x86_64-unknown-uefi -S -emit-llvm -o - %s | \ MaskRay wrote: test/Driver tests should generally only test `-###`, not `-emit-llvm -S` (codegen). Place such codegen tests to test/CodeGen https://github.com/llvm/llvm-project/pull/76838 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [UEFI] X86_64 UEFI Clang Driver (PR #76838)
@@ -0,0 +1,61 @@
+//===--- UEFI.h - UEFI ToolChain Implementations --*- 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
+//
+//===--===//
+
+#ifndef LLVM_CLANG_LIB_DRIVER_TOOLCHAINS_UEFI_H
+#define LLVM_CLANG_LIB_DRIVER_TOOLCHAINS_UEFI_H
+
+#include "clang/Driver/Tool.h"
+#include "clang/Driver/ToolChain.h"
+
+namespace clang {
MaskRay wrote:
`namespace clang::driver {`
https://github.com/llvm/llvm-project/pull/76838
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [libc] [libcxx] [Clang] Do not implicitly link C libraries for the GPU targets (PR #109052)
https://github.com/MaskRay approved this pull request. https://github.com/llvm/llvm-project/pull/109052 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [llvm] [llvm][Triple] Add `Environment` members and parsing for glibc/musl parity. (PR #107664)
https://github.com/MaskRay approved this pull request. https://github.com/llvm/llvm-project/pull/107664 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [flang] [mlir] [clang][flang][mlir] Support -frecord-command-line option (PR #102975)
MaskRay wrote: > > Why do we need the two new files? > > We have clang/lib/Driver/ToolChains/CommonArgs.cpp and > > clang/lib/Driver/ToolChain.cpp for shared code. > > Since most of the code in `CommonArgs.cpp` had to do with functions in the > `clang::driver::tools` namespace, I wasn't sure that code that is only shared > between `Clang.cpp` and `Flang.cpp` ought to go there. The other function > that was moved is a utility function to escape arguments and was originally > declared `static` in `Clang.cpp`. > > If it is ok to use `CommandArgs.cpp` for limited sharing and utilities, I can > move the code there. Adding the shared code to `CommonArgs.cpp` LGTM. Shared functions there use `tools::`, though. `rg driver:: clang/lib/Driver` reveals what files define `driver::` member functions. `CommonArgs.cpp` doesn't define `driver::`. https://github.com/llvm/llvm-project/pull/102975 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang][driver] Don't warn when -S and -c are used together without -fsyntax-only (PR #104477)
https://github.com/MaskRay requested changes to this pull request. . https://github.com/llvm/llvm-project/pull/104477 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [llvm] [llvm][Triple] Add `Environment` members and parsing for glibc/musl parity. (PR #107664)
MaskRay wrote: Are these `musl*` environments blessed by upstream musl? @richfelker https://github.com/llvm/llvm-project/pull/107664 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [flang] [mlir] [clang][flang][mlir] Support -frecord-command-line option (PR #102975)
MaskRay wrote: Why do we need the two new files? (added) clang/lib/Driver/ToolChains/CommonUtils.cpp (+76) (added) clang/lib/Driver/ToolChains/CommonUtils.h (+44) We have clang/lib/Driver/ToolChains/CommonArgs.cpp and clang/lib/Driver/ToolChain.cpp for shared code. https://github.com/llvm/llvm-project/pull/102975 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [flang] [mlir] [clang][flang][mlir] Support -frecord-command-line option (PR #102975)
@@ -1980,9 +1980,11 @@ within a command line are combined with spaces; spaces and backslashes within an argument are escaped with backslashes. This format differs from the format of the equivalent section produced by GCC with the -frecord-gcc-switches flag. This option is currently only supported on ELF targets.}]>, - Group; + Group, MaskRay wrote: I think f_clang_Group is a legacy group for Clang options not in GCC. f_clang_Group has never gain significant adoption and it's used by any code. New -f options can just use f_Group. https://github.com/llvm/llvm-project/pull/102975 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Driver] Avoid repeated hash lookups (NFC) (PR #108676)
https://github.com/MaskRay approved this pull request. https://github.com/llvm/llvm-project/pull/108676 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [lld] [llvm] Deprecate the `-fbasic-block-sections=labels` option. (PR #107494)
@@ -64,14 +64,16 @@ static lto::Config createConfig() {
c.Options.BBAddrMap = config->ltoBBAddrMap;
// Check if basic block sections must be used.
- // Allowed values for --lto-basic-block-sections are "all", "labels",
+ // Allowed values for --lto-basic-block-sections are "all",
// "", or none. This is the equivalent
// of -fbasic-block-sections= flag in clang.
if (!config->ltoBasicBlockSections.empty()) {
if (config->ltoBasicBlockSections == "all") {
c.Options.BBSections = BasicBlockSection::All;
} else if (config->ltoBasicBlockSections == "labels") {
- c.Options.BBSections = BasicBlockSection::Labels;
+ c.Options.BBAddrMap = true;
MaskRay wrote:
suggest that you create a separate change for lld/ELF.
If this is deprecated, we need a test for the warning message.
Please check other warnings for the style. We don't add the full stop.
https://github.com/llvm/llvm-project/pull/107494
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [lld] [llvm] Deprecate the `-fbasic-block-sections=labels` option. (PR #107494)
https://github.com/MaskRay edited https://github.com/llvm/llvm-project/pull/107494 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [lld] [llvm] Deprecate the `-fbasic-block-sections=labels` option. (PR #107494)
https://github.com/MaskRay approved this pull request. (Thanks for the patch. I am on vacation and may be slow/inconvenient to respond.) https://github.com/llvm/llvm-project/pull/107494 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [lld] [llvm] Deprecate the `-fbasic-block-sections=labels` option. (PR #107494)
@@ -2369,14 +2369,16 @@ are listed below. $ cd $P/bar && clang -c -funique-internal-linkage-names name_conflict.c $ cd $P && clang foo/name_conflict.o && bar/name_conflict.o -.. option:: -fbasic-block-sections=[labels, all, list=, none] +.. option:: -f[no]-basic-block-address-map: + Emits a ``SHT_LLVM_BB_ADDR_MAP`` which includes address offsets for each MaskRay wrote: `SHT_LLVM_BB_ADDR_MAP section` https://github.com/llvm/llvm-project/pull/107494 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] fix kcfi doesn't take effect when callee function has no input parameter (PR #106677)
MaskRay wrote: > If a function is declared without a prototype (pre-C23), it doesn't mean it > has a void argument list; it means the argument list is undeclared. So you > can't just assume the argument list is void; you have to compute the argument > types from the actual arguments passed to the call. +1. I don't think we have take this patch. > Or you could just forbid unprototyped functions in KCFI mode, If you don't > want to deal with the edge cases here (i.e. -Werror=strict-prototypes). Seems fine. https://github.com/llvm/llvm-project/pull/106677 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang-tools-extra] [clangd] Construct SmallVector with ArrayRef (NFC) (PR #105829)
https://github.com/MaskRay approved this pull request. https://github.com/llvm/llvm-project/pull/105829 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] Deprecate -fheinous-gnu-extensions; introduce a new warning flag (PR #105821)
https://github.com/MaskRay approved this pull request. LGTM @zatrazz https://github.com/llvm/llvm-project/pull/105821 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang] force libc linked with --no-as-needed when using compiler-rt (PR #95848)
MaskRay wrote: > ```shell > ../build/bin/clang test.c -o libtest.so -fPIC -shared -v -Wl,--as-needed > -Wl,--trace-symbol=__cxa_finalize > ``` This command line does not close `--as-needed` with `--no-as-needed`, so `-lc` is linked in `as-needed` mode. The driver is not responsible for forcing `--no-as-needed`. I believe this PR should be closed. https://github.com/llvm/llvm-project/pull/95848 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang] force libc linked with --no-as-needed when using compiler-rt (PR #95848)
https://github.com/MaskRay closed https://github.com/llvm/llvm-project/pull/95848 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] Honor pragmas with -ffp-contract=fast, depecate fast-honor-pragmas (PR #105746)
@@ -30,8 +29,10 @@ // RUN: | FileCheck --check-prefix=CHECK-FPC-FAST %s // RUN: %clang -### -ffast-math -ffp-contract=fast-honor-pragmas -c %s 2>&1 \ -// RUN: | FileCheck --check-prefixes=CHECK-FPC-FAST-HONOR,WARN_FM_FHP %s -// CHECK-FPC-FAST-HONOR: "-ffp-contract=fast-honor-pragmas" +// RUN: | FileCheck --check-prefixes=CHECK-FPC-FAST-HONOR,WARN_FHP_DEPRECATED %s +// WARN_FHP_DEPRECATED: clang: warning: argument '-ffp-contract=fast-honor-pragmas' is deprecated, use '-ffp-contract=fast' instead [-Wdeprecated] MaskRay wrote: We generally don't check the string before "warning" (can be customized) https://github.com/llvm/llvm-project/pull/105746 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] Honor pragmas with -ffp-contract=fast, depecate fast-honor-pragmas (PR #105746)
https://github.com/MaskRay commented: Thanks for doing this! I support this direction as mentioned in #100453 (I'm starting a 3-week vacation this Friday and will have limited availability. Happy for other reviewers to approve.) https://github.com/llvm/llvm-project/pull/105746 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] Honor pragmas with -ffp-contract=fast, depecate fast-honor-pragmas (PR #105746)
https://github.com/MaskRay edited https://github.com/llvm/llvm-project/pull/105746 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang][Driver] Add a custom error option in multilib.yaml. (PR #105684)
@@ -130,7 +143,7 @@ class MultilibSet {
/// Select compatible variants, \returns false if none are compatible
bool select(const Multilib::flags_list &Flags,
MaskRay wrote:
For other parts of Driver, I think `const Driver &` is more commonly used as
the first parameter
https://github.com/llvm/llvm-project/pull/105684
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang][Driver] Add a custom error option in multilib.yaml. (PR #105684)
https://github.com/MaskRay commented: I'm starting a 3-week vacation this Friday and will have limited availability. Happy for other reviewers to approve. https://github.com/llvm/llvm-project/pull/105684 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang][Driver] Add a custom error option in multilib.yaml. (PR #105684)
@@ -0,0 +1,63 @@ +# REQUIRES: shell +# UNSUPPORTED: system-windows + +# RUN: rm -rf %T/baremetal_multilib MaskRay wrote: do not use `%T` https://llvm.org/docs/CommandGuide/lit.html#substitutions just use `rm -rf %t && mkdir %t` for what you do with `%T/baremetal_multilib` ? https://github.com/llvm/llvm-project/pull/105684 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang][Driver] Add a custom error option in multilib.yaml. (PR #105684)
https://github.com/MaskRay edited https://github.com/llvm/llvm-project/pull/105684 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] Allow installation of Clang .cfg files if they where generated. (PR #105580)
https://github.com/MaskRay commented: Apologies, I'm starting a 3-week vacation this Friday and will have limited availability. Happy for other reviewers to approve. https://github.com/llvm/llvm-project/pull/105580 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [compiler-rt] [XRay] Add support for instrumentation of DSOs on x86_64 (PR #90959)
@@ -2887,6 +2887,10 @@ def fxray_selected_function_group : HelpText<"When using -fxray-function-groups, select which group of functions to instrument. Valid range is 0 to fxray-function-groups - 1">, MarshallingInfoInt, "0">; +def fxray_enable_shared : Flag<["-"], "fxray-enable-shared">, Group, Visibility<[ClangOption, CC1Option]>, MaskRay wrote: Use `BoolFOption`. You can check some existing ones and how they determine whether `clang -cc1 -ffoo` or `clang -cc1 -fno-foo` is available. https://github.com/llvm/llvm-project/pull/90959 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Sparc] Add flags to enable errata workaround pass for GR712RC and UT700 (PR #104742)
https://github.com/MaskRay approved this pull request. https://github.com/llvm/llvm-project/pull/104742 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Sparc] Add flags to enable errata workaround pass for GR712RC and UT700 (PR #104742)
MaskRay wrote: This patch looks ok, but I expect justification why we need these -mfix-* options, which adds some, even if little, burden to the generic code. I asked because these are apparent workarounds. Sparc is a very old port that doesn't receive lots of development, and we should carefully balance needs and maintenance burden. https://github.com/llvm/llvm-project/pull/104742 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [PS5][Driver] Link main components with -pie by default (PR #102901)
@@ -1,3 +1,18 @@
+// Test that PIE is the default for main components
+
+// RUN: %clang --target=x86_64-scei-ps5 %s -### 2>&1 | FileCheck
--check-prefixes=CHECK-PIE %s
+
+// CHECK-PIE: {{ld(\.exe)?}}"
+// CHECK-PIE-SAME: "-pie"
+
+// RUN: %clang --target=x86_64-scei-ps5 -no-pie %s -### 2>&1 | FileCheck
--check-prefixes=CHECK-NO-PIE %s
+// RUN: %clang --target=x86_64-scei-ps5 -r %s -### 2>&1 | FileCheck
--check-prefixes=CHECK-NO-PIE %s
+// RUN: %clang --target=x86_64-scei-ps5 -shared %s -### 2>&1 | FileCheck
--check-prefixes=CHECK-NO-PIE %s
MaskRay wrote:
Suggest that the `-shared` RUN line additionally checks `-shared`.
(I'm starting a 3-week vacation this Friday and will have limited availability.)
https://github.com/llvm/llvm-project/pull/102901
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [PS5][Driver] Link main components with -pie by default (PR #102901)
@@ -237,7 +237,10 @@ void tools::PS5cpu::Linker::ConstructJob(Compilation &C,
const JobAction &JA,
if (!D.SysRoot.empty())
CmdArgs.push_back(Args.MakeArgString("--sysroot=" + D.SysRoot));
- if (Args.hasArg(options::OPT_pie))
+ // Default to PIE for non-static main components.
MaskRay wrote:
While the generic ABI refers to executable and shared object files as
"components", "main component" isn't used.
main executable or just executable
https://github.com/llvm/llvm-project/pull/102901
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Driver] Use llvm::make_range(std::pair) (NFC) (PR #105470)
https://github.com/MaskRay approved this pull request. https://github.com/llvm/llvm-project/pull/105470 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [llvm] Fix KCFI types for generated functions with integer normalization (PR #104826)
https://github.com/MaskRay approved this pull request. https://github.com/llvm/llvm-project/pull/104826 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang] Support -Wa, options -mmsa and -mno-msa (PR #99615)
https://github.com/MaskRay closed https://github.com/llvm/llvm-project/pull/99615 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang] Support -Wa, options -mmsa and -mno-msa (PR #99615)
https://github.com/MaskRay approved this pull request. https://github.com/llvm/llvm-project/pull/99615 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang] Support -Wa, options -mmsa and -mno-msa (PR #99615)
@@ -0,0 +1,8 @@ +// RUN: %clang -### -c --target=mips64el-unknown-linux-gnuabi64 \ MaskRay wrote: We don't need -fno-integrated-as tests. They are just forwarded. The -mmsa test needs a NOT. I'll push a commit to improve the test https://github.com/llvm/llvm-project/pull/99615 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang] Support -Wa, options -mmsa and -mno-msa (PR #99615)
https://github.com/MaskRay updated
https://github.com/llvm/llvm-project/pull/99615
>From 7c839979164229ecbbdafb684107a6017b354df8 Mon Sep 17 00:00:00 2001
From: Ying Huang
Date: Fri, 19 Jul 2024 04:19:09 -0400
Subject: [PATCH 1/2] [clang] Support -Wa, options -mmsa and -mno-msa
---
clang/include/clang/Basic/CodeGenOptions.def | 1 +
clang/include/clang/Driver/Options.td| 4 +++-
clang/lib/Driver/ToolChains/Clang.cpp| 13 -
clang/test/Driver/mips-msa.c | 19 +++
4 files changed, 35 insertions(+), 2 deletions(-)
create mode 100644 clang/test/Driver/mips-msa.c
diff --git a/clang/include/clang/Basic/CodeGenOptions.def
b/clang/include/clang/Basic/CodeGenOptions.def
index 09e892d6d4defe..ecea476abe3232 100644
--- a/clang/include/clang/Basic/CodeGenOptions.def
+++ b/clang/include/clang/Basic/CodeGenOptions.def
@@ -176,6 +176,7 @@ CODEGENOPT(MergeAllConstants , 1, 1) ///< Merge identical
constants.
CODEGENOPT(MergeFunctions, 1, 0) ///< Set when -fmerge-functions is
enabled.
CODEGENOPT(NoCommon , 1, 0) ///< Set when -fno-common or C++ is
enabled.
CODEGENOPT(NoExecStack , 1, 0) ///< Set when -Wa,--noexecstack is
enabled.
+CODEGENOPT(MipsMsa , 1, 0) ///< Set when -Wa,-mmsa is enabled.
CODEGENOPT(FatalWarnings , 1, 0) ///< Set when -Wa,--fatal-warnings is
///< enabled.
CODEGENOPT(NoWarn, 1, 0) ///< Set when -Wa,--no-warn is enabled.
diff --git a/clang/include/clang/Driver/Options.td
b/clang/include/clang/Driver/Options.td
index c66e035a259b3f..625a8303b9bf15 100644
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -5383,7 +5383,9 @@ def mmadd4 : Flag<["-"], "mmadd4">,
Group,
def mno_madd4 : Flag<["-"], "mno-madd4">, Group,
HelpText<"Disable the generation of 4-operand madd.s, madd.d and related
instructions.">;
def mmsa : Flag<["-"], "mmsa">, Group,
- HelpText<"Enable MSA ASE (MIPS only)">;
+ Visibility<[ClangOption, CC1Option, CC1AsOption]>,
+ HelpText<"Enable MSA ASE (MIPS only)">,
+ MarshallingInfoFlag>;
def mno_msa : Flag<["-"], "mno-msa">, Group,
HelpText<"Disable MSA ASE (MIPS only)">;
def mmt : Flag<["-"], "mmt">, Group,
diff --git a/clang/lib/Driver/ToolChains/Clang.cpp
b/clang/lib/Driver/ToolChains/Clang.cpp
index 6180c6eeb6f07a..9fd8a0fcf50270 100644
--- a/clang/lib/Driver/ToolChains/Clang.cpp
+++ b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -2556,6 +2556,7 @@ static void CollectArgsForIntegratedAssembler(Compilation
&C,
bool Crel = false, ExperimentalCrel = false;
bool UseRelaxRelocations = C.getDefaultToolChain().useRelaxRelocations();
bool UseNoExecStack = false;
+ std::optional Msa;
const char *MipsTargetFeature = nullptr;
StringRef ImplicitIt;
for (const Arg *A :
@@ -2665,7 +2666,14 @@ static void
CollectArgsForIntegratedAssembler(Compilation &C,
CmdArgs.push_back("-soft-float");
continue;
}
-
+if (Value == "-mmsa") {
+ Msa = true;
+ continue;
+}
+if (Value == "-mno-msa") {
+ Msa = false;
+ continue;
+}
MipsTargetFeature = llvm::StringSwitch(Value)
.Case("-mips1", "+mips1")
.Case("-mips2", "+mips2")
@@ -2685,6 +2693,7 @@ static void CollectArgsForIntegratedAssembler(Compilation
&C,
.Default(nullptr);
if (MipsTargetFeature)
continue;
+break;
}
if (Value == "-force_cpusubtype_ALL") {
@@ -2777,6 +2786,8 @@ static void CollectArgsForIntegratedAssembler(Compilation
&C,
<< "-Wa,--crel" << D.getTargetTriple();
}
}
+ if (Msa && *Msa)
+CmdArgs.push_back("-mmsa");
if (!UseRelaxRelocations)
CmdArgs.push_back("-mrelax-relocations=no");
if (UseNoExecStack)
diff --git a/clang/test/Driver/mips-msa.c b/clang/test/Driver/mips-msa.c
new file mode 100644
index 00..78ff58288bd8a8
--- /dev/null
+++ b/clang/test/Driver/mips-msa.c
@@ -0,0 +1,19 @@
+// RUN: %clang -### -c --target=mips64el-unknown-linux-gnuabi64 \
+// RUN: -Wa,-mmsa %s -Werror 2>&1 | FileCheck %s --check-prefix=CHECK-MMSA
+// CHECK-MMSA: "-cc1" {{.*}}"-mmsa"
+
+// RUN: %clang -### -c --target=mips64el-unknown-linux-gnuabi64 \
+// RUN: -Wa,-mno-msa,-mmsa %s -Werror 2>&1 | FileCheck %s
--check-prefix=CHECK-MMSA
+
+// RUN: %clang -### -c --target=mips64el-unknown-linux-gnuabi64 \
+// RUN: -Wa,-mmsa,-mno-msa %s -Werror 2>&1 | FileCheck %s
--check-prefix=CHECK-NOMMSA
+// CHECK-NOMMSA: "-cc1"
+
+// RUN: %clang -### -c --target=mips64el-unknown-linux-gnuabi64 \
+// RUN:-fno-integrated-as -Wa,-mmsa %s -Werror 2>&1 | FileCheck %s
--check-prefix=MIPS-MSA
+// MIPS-MSA: as{{(.exe)?}}" "-march" "mips64r2" "-mabi" "64" "-EL" "-KPIC"
"-mmsa"
+
+// RUN: %clang -### -c --target=mips64el-unknown-linux-gnuabi64 \
+// RUN:-fn
[clang] [clang] Support -Wa, options -mmsa and -mno-msa (PR #99615)
https://github.com/MaskRay updated
https://github.com/llvm/llvm-project/pull/99615
>From 7c839979164229ecbbdafb684107a6017b354df8 Mon Sep 17 00:00:00 2001
From: Ying Huang
Date: Fri, 19 Jul 2024 04:19:09 -0400
Subject: [PATCH 1/2] [clang] Support -Wa, options -mmsa and -mno-msa
---
clang/include/clang/Basic/CodeGenOptions.def | 1 +
clang/include/clang/Driver/Options.td| 4 +++-
clang/lib/Driver/ToolChains/Clang.cpp| 13 -
clang/test/Driver/mips-msa.c | 19 +++
4 files changed, 35 insertions(+), 2 deletions(-)
create mode 100644 clang/test/Driver/mips-msa.c
diff --git a/clang/include/clang/Basic/CodeGenOptions.def
b/clang/include/clang/Basic/CodeGenOptions.def
index 09e892d6d4defe..ecea476abe3232 100644
--- a/clang/include/clang/Basic/CodeGenOptions.def
+++ b/clang/include/clang/Basic/CodeGenOptions.def
@@ -176,6 +176,7 @@ CODEGENOPT(MergeAllConstants , 1, 1) ///< Merge identical
constants.
CODEGENOPT(MergeFunctions, 1, 0) ///< Set when -fmerge-functions is
enabled.
CODEGENOPT(NoCommon , 1, 0) ///< Set when -fno-common or C++ is
enabled.
CODEGENOPT(NoExecStack , 1, 0) ///< Set when -Wa,--noexecstack is
enabled.
+CODEGENOPT(MipsMsa , 1, 0) ///< Set when -Wa,-mmsa is enabled.
CODEGENOPT(FatalWarnings , 1, 0) ///< Set when -Wa,--fatal-warnings is
///< enabled.
CODEGENOPT(NoWarn, 1, 0) ///< Set when -Wa,--no-warn is enabled.
diff --git a/clang/include/clang/Driver/Options.td
b/clang/include/clang/Driver/Options.td
index c66e035a259b3f..625a8303b9bf15 100644
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -5383,7 +5383,9 @@ def mmadd4 : Flag<["-"], "mmadd4">,
Group,
def mno_madd4 : Flag<["-"], "mno-madd4">, Group,
HelpText<"Disable the generation of 4-operand madd.s, madd.d and related
instructions.">;
def mmsa : Flag<["-"], "mmsa">, Group,
- HelpText<"Enable MSA ASE (MIPS only)">;
+ Visibility<[ClangOption, CC1Option, CC1AsOption]>,
+ HelpText<"Enable MSA ASE (MIPS only)">,
+ MarshallingInfoFlag>;
def mno_msa : Flag<["-"], "mno-msa">, Group,
HelpText<"Disable MSA ASE (MIPS only)">;
def mmt : Flag<["-"], "mmt">, Group,
diff --git a/clang/lib/Driver/ToolChains/Clang.cpp
b/clang/lib/Driver/ToolChains/Clang.cpp
index 6180c6eeb6f07a..9fd8a0fcf50270 100644
--- a/clang/lib/Driver/ToolChains/Clang.cpp
+++ b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -2556,6 +2556,7 @@ static void CollectArgsForIntegratedAssembler(Compilation
&C,
bool Crel = false, ExperimentalCrel = false;
bool UseRelaxRelocations = C.getDefaultToolChain().useRelaxRelocations();
bool UseNoExecStack = false;
+ std::optional Msa;
const char *MipsTargetFeature = nullptr;
StringRef ImplicitIt;
for (const Arg *A :
@@ -2665,7 +2666,14 @@ static void
CollectArgsForIntegratedAssembler(Compilation &C,
CmdArgs.push_back("-soft-float");
continue;
}
-
+if (Value == "-mmsa") {
+ Msa = true;
+ continue;
+}
+if (Value == "-mno-msa") {
+ Msa = false;
+ continue;
+}
MipsTargetFeature = llvm::StringSwitch(Value)
.Case("-mips1", "+mips1")
.Case("-mips2", "+mips2")
@@ -2685,6 +2693,7 @@ static void CollectArgsForIntegratedAssembler(Compilation
&C,
.Default(nullptr);
if (MipsTargetFeature)
continue;
+break;
}
if (Value == "-force_cpusubtype_ALL") {
@@ -2777,6 +2786,8 @@ static void CollectArgsForIntegratedAssembler(Compilation
&C,
<< "-Wa,--crel" << D.getTargetTriple();
}
}
+ if (Msa && *Msa)
+CmdArgs.push_back("-mmsa");
if (!UseRelaxRelocations)
CmdArgs.push_back("-mrelax-relocations=no");
if (UseNoExecStack)
diff --git a/clang/test/Driver/mips-msa.c b/clang/test/Driver/mips-msa.c
new file mode 100644
index 00..78ff58288bd8a8
--- /dev/null
+++ b/clang/test/Driver/mips-msa.c
@@ -0,0 +1,19 @@
+// RUN: %clang -### -c --target=mips64el-unknown-linux-gnuabi64 \
+// RUN: -Wa,-mmsa %s -Werror 2>&1 | FileCheck %s --check-prefix=CHECK-MMSA
+// CHECK-MMSA: "-cc1" {{.*}}"-mmsa"
+
+// RUN: %clang -### -c --target=mips64el-unknown-linux-gnuabi64 \
+// RUN: -Wa,-mno-msa,-mmsa %s -Werror 2>&1 | FileCheck %s
--check-prefix=CHECK-MMSA
+
+// RUN: %clang -### -c --target=mips64el-unknown-linux-gnuabi64 \
+// RUN: -Wa,-mmsa,-mno-msa %s -Werror 2>&1 | FileCheck %s
--check-prefix=CHECK-NOMMSA
+// CHECK-NOMMSA: "-cc1"
+
+// RUN: %clang -### -c --target=mips64el-unknown-linux-gnuabi64 \
+// RUN:-fno-integrated-as -Wa,-mmsa %s -Werror 2>&1 | FileCheck %s
--check-prefix=MIPS-MSA
+// MIPS-MSA: as{{(.exe)?}}" "-march" "mips64r2" "-mabi" "64" "-EL" "-KPIC"
"-mmsa"
+
+// RUN: %clang -### -c --target=mips64el-unknown-linux-gnuabi64 \
+// RUN:-fn
[clang] [clang] Support -Wa, options -mmsa and -mno-msa (PR #99615)
https://github.com/MaskRay updated
https://github.com/llvm/llvm-project/pull/99615
>From cab152b6306bf486daf478709fe5eaa3a0bb79d2 Mon Sep 17 00:00:00 2001
From: Ying Huang
Date: Fri, 19 Jul 2024 04:19:09 -0400
Subject: [PATCH 1/2] [clang] Support -Wa, options -mmsa and -mno-msa
---
clang/include/clang/Basic/CodeGenOptions.def | 1 +
clang/include/clang/Driver/Options.td| 4 +++-
clang/lib/Driver/ToolChains/Clang.cpp| 13 -
clang/test/Driver/mips-msa.c | 19 +++
4 files changed, 35 insertions(+), 2 deletions(-)
create mode 100644 clang/test/Driver/mips-msa.c
diff --git a/clang/include/clang/Basic/CodeGenOptions.def
b/clang/include/clang/Basic/CodeGenOptions.def
index 09e892d6d4defe..ecea476abe3232 100644
--- a/clang/include/clang/Basic/CodeGenOptions.def
+++ b/clang/include/clang/Basic/CodeGenOptions.def
@@ -176,6 +176,7 @@ CODEGENOPT(MergeAllConstants , 1, 1) ///< Merge identical
constants.
CODEGENOPT(MergeFunctions, 1, 0) ///< Set when -fmerge-functions is
enabled.
CODEGENOPT(NoCommon , 1, 0) ///< Set when -fno-common or C++ is
enabled.
CODEGENOPT(NoExecStack , 1, 0) ///< Set when -Wa,--noexecstack is
enabled.
+CODEGENOPT(MipsMsa , 1, 0) ///< Set when -Wa,-mmsa is enabled.
CODEGENOPT(FatalWarnings , 1, 0) ///< Set when -Wa,--fatal-warnings is
///< enabled.
CODEGENOPT(NoWarn, 1, 0) ///< Set when -Wa,--no-warn is enabled.
diff --git a/clang/include/clang/Driver/Options.td
b/clang/include/clang/Driver/Options.td
index c66e035a259b3f..625a8303b9bf15 100644
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -5383,7 +5383,9 @@ def mmadd4 : Flag<["-"], "mmadd4">,
Group,
def mno_madd4 : Flag<["-"], "mno-madd4">, Group,
HelpText<"Disable the generation of 4-operand madd.s, madd.d and related
instructions.">;
def mmsa : Flag<["-"], "mmsa">, Group,
- HelpText<"Enable MSA ASE (MIPS only)">;
+ Visibility<[ClangOption, CC1Option, CC1AsOption]>,
+ HelpText<"Enable MSA ASE (MIPS only)">,
+ MarshallingInfoFlag>;
def mno_msa : Flag<["-"], "mno-msa">, Group,
HelpText<"Disable MSA ASE (MIPS only)">;
def mmt : Flag<["-"], "mmt">, Group,
diff --git a/clang/lib/Driver/ToolChains/Clang.cpp
b/clang/lib/Driver/ToolChains/Clang.cpp
index 6180c6eeb6f07a..9fd8a0fcf50270 100644
--- a/clang/lib/Driver/ToolChains/Clang.cpp
+++ b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -2556,6 +2556,7 @@ static void CollectArgsForIntegratedAssembler(Compilation
&C,
bool Crel = false, ExperimentalCrel = false;
bool UseRelaxRelocations = C.getDefaultToolChain().useRelaxRelocations();
bool UseNoExecStack = false;
+ std::optional Msa;
const char *MipsTargetFeature = nullptr;
StringRef ImplicitIt;
for (const Arg *A :
@@ -2665,7 +2666,14 @@ static void
CollectArgsForIntegratedAssembler(Compilation &C,
CmdArgs.push_back("-soft-float");
continue;
}
-
+if (Value == "-mmsa") {
+ Msa = true;
+ continue;
+}
+if (Value == "-mno-msa") {
+ Msa = false;
+ continue;
+}
MipsTargetFeature = llvm::StringSwitch(Value)
.Case("-mips1", "+mips1")
.Case("-mips2", "+mips2")
@@ -2685,6 +2693,7 @@ static void CollectArgsForIntegratedAssembler(Compilation
&C,
.Default(nullptr);
if (MipsTargetFeature)
continue;
+break;
}
if (Value == "-force_cpusubtype_ALL") {
@@ -2777,6 +2786,8 @@ static void CollectArgsForIntegratedAssembler(Compilation
&C,
<< "-Wa,--crel" << D.getTargetTriple();
}
}
+ if (Msa && *Msa)
+CmdArgs.push_back("-mmsa");
if (!UseRelaxRelocations)
CmdArgs.push_back("-mrelax-relocations=no");
if (UseNoExecStack)
diff --git a/clang/test/Driver/mips-msa.c b/clang/test/Driver/mips-msa.c
new file mode 100644
index 00..78ff58288bd8a8
--- /dev/null
+++ b/clang/test/Driver/mips-msa.c
@@ -0,0 +1,19 @@
+// RUN: %clang -### -c --target=mips64el-unknown-linux-gnuabi64 \
+// RUN: -Wa,-mmsa %s -Werror 2>&1 | FileCheck %s --check-prefix=CHECK-MMSA
+// CHECK-MMSA: "-cc1" {{.*}}"-mmsa"
+
+// RUN: %clang -### -c --target=mips64el-unknown-linux-gnuabi64 \
+// RUN: -Wa,-mno-msa,-mmsa %s -Werror 2>&1 | FileCheck %s
--check-prefix=CHECK-MMSA
+
+// RUN: %clang -### -c --target=mips64el-unknown-linux-gnuabi64 \
+// RUN: -Wa,-mmsa,-mno-msa %s -Werror 2>&1 | FileCheck %s
--check-prefix=CHECK-NOMMSA
+// CHECK-NOMMSA: "-cc1"
+
+// RUN: %clang -### -c --target=mips64el-unknown-linux-gnuabi64 \
+// RUN:-fno-integrated-as -Wa,-mmsa %s -Werror 2>&1 | FileCheck %s
--check-prefix=MIPS-MSA
+// MIPS-MSA: as{{(.exe)?}}" "-march" "mips64r2" "-mabi" "64" "-EL" "-KPIC"
"-mmsa"
+
+// RUN: %clang -### -c --target=mips64el-unknown-linux-gnuabi64 \
+// RUN:-fn
[libunwind] [libunwind] Detect cycles of length 1 (PR #103476)
https://github.com/MaskRay closed https://github.com/llvm/llvm-project/pull/103476 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[libunwind] [libunwind] Detect cycles of length 1 (PR #103476)
MaskRay wrote: > Are you able to insert a `.cfi_undefined ` in the generated > assembly? That would tell libunwind to stop unwinding and is cleaner than > this workaround IMO. > > I just noticed a similar problem in FreeBSD's startup files but I plan to fix > those in FreeBSD instead of working around them here. Thanks for weighing in. I'll close this PR. I agree that sending a patch to musl is worth trying: > (And, IIUC, musl people generally don't want programs to be able to unwind > their own stack for debugging purposes. If I send a fix to musl, I imagine > the answer may be "gdb already works, libunwind already doesn't work, and > that's exactly how we want it", but may be worth a try.) https://github.com/llvm/llvm-project/pull/103476 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang] Add support for omitting only global destructors (PR #104899)
https://github.com/MaskRay approved this pull request. LGTM! (I'm starting a 3-week vacation this Friday and will have limited availability.) https://github.com/llvm/llvm-project/pull/104899 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang] Add support for omitting only global destructors (PR #104899)
@@ -7967,8 +7967,9 @@ void Clang::ConstructJob(Compilation &C, const JobAction &JA, options::OPT_fno_keep_persistent_storage_variables); Args.addOptInFlag(CmdArgs, options::OPT_fcomplete_member_pointers, options::OPT_fno_complete_member_pointers); - Args.addOptOutFlag(CmdArgs, options::OPT_fcxx_static_destructors, - options::OPT_fno_cxx_static_destructors); + if (Arg *A = Args.getLastArg(options::OPT_cxx_static_destructors_EQ)) +CmdArgs.push_back( MaskRay wrote: `A->render(Args, CmdArgs);` https://github.com/llvm/llvm-project/pull/104899 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang] Add support for omitting only global destructors (PR #104899)
@@ -0,0 +1,9 @@ +// RUN: %clang -### -c -fc++-static-destructors=all %s 2>&1 | FileCheck --check-prefix ALL %s MaskRay wrote: thx for adding a driver test. The previous `-fc++-static-destructors` patch https://reviews.llvm.org/D50994 did not come with a driver test https://github.com/llvm/llvm-project/pull/104899 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang] Add support for omitting only global destructors (PR #104899)
https://github.com/MaskRay edited https://github.com/llvm/llvm-project/pull/104899 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang] Add support for omitting only global destructors (PR #104899)
@@ -0,0 +1,9 @@ +// RUN: %clang -### -c -fc++-static-destructors=all %s 2>&1 | FileCheck --check-prefix ALL %s +// RUN: %clang -### -c -fc++-static-destructors %s 2>&1 | FileCheck --check-prefix ALL %s +// RUN: %clang -### -c -fc++-static-destructors=none %s 2>&1 | FileCheck --check-prefix NONE %s MaskRay wrote: You can add -fc++-static-destructors before -fc++-static-destructors=none to enhance the test to show we use the `getLastArg` semantics https://github.com/llvm/llvm-project/pull/104899 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Sparc] Add flags to enable errata workaround pass for GR712RC and UT700 (PR #104742)
@@ -0,0 +1,12 @@ +// RUN: %clang -target sparc -mfix-gr712rc -### %s 2> %t +// RUN: FileCheck --check-prefix=CHECK-FIX-GR712RC < %t %s +// CHECK-FIX-GR712RC: "-target-feature" "+fix-tn0009" +// CHECK-FIX-GR712RC: "-target-feature" "+fix-tn0011" MaskRay wrote: If these are adjacent, you can place them on the same line. If you don't that, use `-SAME:` https://github.com/llvm/llvm-project/pull/104742 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Sparc] Add flags to enable errata workaround pass for GR712RC and UT700 (PR #104742)
@@ -0,0 +1,12 @@ +// RUN: %clang -target sparc -mfix-gr712rc -### %s 2> %t MaskRay wrote: `--target=` for new tests. `-target ` has been deprecated since around Clang 3.4 https://github.com/llvm/llvm-project/pull/104742 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [llvm] [Driver] Default -msmall-data-limit= to 0 and clean up code (PR #83093)
https://github.com/MaskRay updated
https://github.com/llvm/llvm-project/pull/83093
>From dbc9ddd9a9b1b73d56f8aab5265db36a44e6fcaf Mon Sep 17 00:00:00 2001
From: Fangrui Song
Date: Mon, 26 Feb 2024 17:58:55 -0800
Subject: [PATCH] =?UTF-8?q?[=F0=9D=98=80=F0=9D=97=BD=F0=9D=97=BF]=20initia?=
=?UTF-8?q?l=20version?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Created using spr 1.3.4
---
clang/lib/Driver/ToolChains/Clang.cpp | 41 ++-
.../CodeGen/RISCV/riscv-sdata-module-flag.c | 4 +-
clang/test/Driver/riscv-sdata-warning.c | 4 --
clang/test/Driver/riscv-sdata.c | 5 +++
llvm/lib/Target/RISCV/RISCVTargetObjectFile.h | 2 +-
5 files changed, 12 insertions(+), 44 deletions(-)
delete mode 100644 clang/test/Driver/riscv-sdata-warning.c
create mode 100644 clang/test/Driver/riscv-sdata.c
diff --git a/clang/lib/Driver/ToolChains/Clang.cpp
b/clang/lib/Driver/ToolChains/Clang.cpp
index 6e1b7e8657d0dc..9b9eab3990607f 100644
--- a/clang/lib/Driver/ToolChains/Clang.cpp
+++ b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -2029,42 +2029,6 @@ void Clang::AddPPCTargetArgs(const ArgList &Args,
}
}
-static void SetRISCVSmallDataLimit(const ToolChain &TC, const ArgList &Args,
- ArgStringList &CmdArgs) {
- const Driver &D = TC.getDriver();
- const llvm::Triple &Triple = TC.getTriple();
- // Default small data limitation is eight.
- const char *SmallDataLimit = "8";
- // Get small data limitation.
- if (Args.getLastArg(options::OPT_shared, options::OPT_fpic,
- options::OPT_fPIC)) {
-// Not support linker relaxation for PIC.
-SmallDataLimit = "0";
-if (Args.hasArg(options::OPT_G)) {
- D.Diag(diag::warn_drv_unsupported_sdata);
-}
- } else if (Args.getLastArgValue(options::OPT_mcmodel_EQ)
- .equals_insensitive("large") &&
- (Triple.getArch() == llvm::Triple::riscv64)) {
-// Not support linker relaxation for RV64 with large code model.
-SmallDataLimit = "0";
-if (Args.hasArg(options::OPT_G)) {
- D.Diag(diag::warn_drv_unsupported_sdata);
-}
- } else if (Triple.isAndroid()) {
-// GP relaxation is not supported on Android.
-SmallDataLimit = "0";
-if (Args.hasArg(options::OPT_G)) {
- D.Diag(diag::warn_drv_unsupported_sdata);
-}
- } else if (Arg *A = Args.getLastArg(options::OPT_G)) {
-SmallDataLimit = A->getValue();
- }
- // Forward the -msmall-data-limit= option.
- CmdArgs.push_back("-msmall-data-limit");
- CmdArgs.push_back(SmallDataLimit);
-}
-
void Clang::AddRISCVTargetArgs(const ArgList &Args,
ArgStringList &CmdArgs) const {
const llvm::Triple &Triple = getToolChain().getTriple();
@@ -2073,7 +2037,10 @@ void Clang::AddRISCVTargetArgs(const ArgList &Args,
CmdArgs.push_back("-target-abi");
CmdArgs.push_back(ABIName.data());
- SetRISCVSmallDataLimit(getToolChain(), Args, CmdArgs);
+ if (Arg *A = Args.getLastArg(options::OPT_G)) {
+CmdArgs.push_back("-msmall-data-limit");
+CmdArgs.push_back(A->getValue());
+ }
if (!Args.hasFlag(options::OPT_mimplicit_float,
options::OPT_mno_implicit_float, true))
diff --git a/clang/test/CodeGen/RISCV/riscv-sdata-module-flag.c
b/clang/test/CodeGen/RISCV/riscv-sdata-module-flag.c
index 9bd69e7995877e..e91b2e7f225f52 100644
--- a/clang/test/CodeGen/RISCV/riscv-sdata-module-flag.c
+++ b/clang/test/CodeGen/RISCV/riscv-sdata-module-flag.c
@@ -30,14 +30,14 @@
void test(void) {}
-// RV32-DEFAULT: !{i32 8, !"SmallDataLimit", i32 8}
+// RV32-DEFAULT: !{i32 8, !"SmallDataLimit", i32 0}
// RV32-G4: !{i32 8, !"SmallDataLimit", i32 4}
// RV32-S0: !{i32 8, !"SmallDataLimit", i32 0}
// RV32-S2G4:!{i32 8, !"SmallDataLimit", i32 4}
// RV32-T16: !{i32 8, !"SmallDataLimit", i32 16}
// RV32-PIC: !{i32 8, !"SmallDataLimit", i32 0}
-// RV64-DEFAULT: !{i32 8, !"SmallDataLimit", i32 8}
+// RV64-DEFAULT: !{i32 8, !"SmallDataLimit", i32 0}
// RV64-G4: !{i32 8, !"SmallDataLimit", i32 4}
// RV64-S0: !{i32 8, !"SmallDataLimit", i32 0}
// RV64-S2G4:!{i32 8, !"SmallDataLimit", i32 4}
diff --git a/clang/test/Driver/riscv-sdata-warning.c
b/clang/test/Driver/riscv-sdata-warning.c
deleted file mode 100644
index ace9a32ee116fd..00
--- a/clang/test/Driver/riscv-sdata-warning.c
+++ /dev/null
@@ -1,4 +0,0 @@
-// REQUIRES: riscv-registered-target
-// RUN: %clang -S --target=riscv32-unknown-elf -fpic -msmall-data-limit=8 %s
2>&1 \
-// RUN: | FileCheck -check-prefix=CHECK-PIC-SDATA %s
-// CHECK-PIC-SDATA: warning: ignoring '-msmall-data-limit=' with
-mcmodel=large for -fpic or RV64
diff --git a/clang/test/Driver/riscv-sdata.c b/clang/test/Driver/riscv-sdata.c
new file mode 100644
index 00..84cd109813db4c
--- /dev/null
+++ b/clang/test/Driver/riscv-sdata.c
@@ -0,0 +1,5 @@
+// RUN: %clang -### -S --target=riscv64 %s 2>&1 |
[clang] [clang][NFC] Split invalid-cpu-note tests (PR #104601)
https://github.com/MaskRay approved this pull request. https://github.com/llvm/llvm-project/pull/104601 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Driver] Introduce ffp-model=aggressive (PR #100453)
MaskRay wrote: > > > @MaskRay Are you OK with this change? > > > > > > Sorry for the delay. I've read through > > [discourse.llvm.org/t/making-ffp-model-fast-more-user-friendly/78402](https://discourse.llvm.org/t/making-ffp-model-fast-more-user-friendly/78402), > > how we ended up with `-ffp-contract=fast-honor-pragmas` beside `=fast` > > (GCC compatibility, sigh), some notes that ICC/GCC have behavior > > differences. A different, safer mode for `-ffp-model` (absent from GCC) > > makes sense to me. > > GCC doesn't support the pragma, and ICC doesn't have a mode that only allows > contraction within a single expression (as the C standard requires), so I > don't think compatibility with the behavior of the pragma in those compilers > makes sense. > > I personally feel very strongly that even our -ffast-math handling should > honor pragmas by default and that -ffp-contract=fast should honor pragmas > (perhaps with the addition of -ffp-contract=fast-no-honor-pragmas if anyone > really needs that). Someone else was surprised by this behavior earlier this > year (#88633). On the other hand, we've got a bug in the front end where we > crash with "#pragma STDC FP_CONTRACT DEFAULT" if you've used > "-ffp-contract=fast-honor-pragmas" > ([godbolt.org/z/EqE6Kn6Wf](https://godbolt.org/z/EqE6Kn6Wf)) so maybe I > should hold the fp-contract change back for a future patch. I've read https://discourse.llvm.org/t/fp-contract-fast-and-pragmas/58529 and I agree that `fast`/`fast-honor-pragmas` could be merged and if needed, `-fast-no-honor-pragmas` could be added (I suspect that there is no such need...) https://github.com/llvm/llvm-project/pull/100453 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang][rtsan] Introduce realtime sanitizer codegen and driver (PR #102622)
@@ -0,0 +1,9 @@ +// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -fsanitize=realtime -emit-llvm -o - %s | FileCheck %s MaskRay wrote: this should be merged with `at_entry.c`. https://github.com/llvm/llvm-project/pull/102622 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang][rtsan] Introduce realtime sanitizer codegen and driver (PR #102622)
@@ -0,0 +1,14 @@ +// RUN: %clang -target x86_64-unknown-linux -fsanitize=realtime %s -S -emit-llvm -o - | FileCheck %s MaskRay wrote: Thx for letting me know the existing asan.c/tsan. I'll try to clean up the two tests. For this driver test which will use `-###`, `-O` is entirely uninteresting. Sometimes `-O1` or above may be interesting for CodeGen tests but probably not for this. https://github.com/llvm/llvm-project/pull/102622 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang][rtsan] Introduce realtime sanitizer codegen and driver (PR #102622)
@@ -0,0 +1,9 @@ +// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -fsanitize=realtime -emit-llvm -o - %s | FileCheck %s + +// __rtsan_realtime_exit should be inserted at all function returns MaskRay wrote: End all full sentence comments with a `.` https://github.com/llvm/llvm-project/pull/102622 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang][rtsan] Introduce realtime sanitizer codegen and driver (PR #102622)
@@ -0,0 +1,9 @@ +// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -fsanitize=realtime -emit-llvm -o - %s | FileCheck %s MaskRay wrote: `*-unknown-linux-gnu` is more popular. `none` could also be used for the OS component and is slightly misleading. https://github.com/llvm/llvm-project/pull/102622 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang][rtsan] Introduce realtime sanitizer codegen and driver (PR #102622)
@@ -845,6 +845,15 @@ void CodeGenFunction::StartFunction(GlobalDecl GD,
QualType RetTy,
if (SanOpts.has(SanitizerKind::ShadowCallStack))
Fn->addFnAttr(llvm::Attribute::ShadowCallStack);
+ if (SanOpts.has(SanitizerKind::Realtime)) {
MaskRay wrote:
remove braces per
https://llvm.org/docs/CodingStandards.html#don-t-use-braces-on-simple-single-statement-bodies-of-if-else-loop-statements
https://github.com/llvm/llvm-project/pull/102622
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang][rtsan] Introduce realtime sanitizer codegen and driver (PR #102622)
https://github.com/MaskRay approved this pull request. https://github.com/llvm/llvm-project/pull/102622 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang][rtsan] Introduce realtime sanitizer codegen and driver (PR #102622)
@@ -0,0 +1,14 @@ +// RUN: %clang -target x86_64-unknown-linux -fsanitize=realtime %s -S -emit-llvm -o - | FileCheck %s MaskRay wrote: Driver tests should only use `-###` to test CC1 options. We don't test different `-O` combinations. It's not useful to test `-flto` test/CodeGen tests `%clang_cc1` We generally don't add end-to-end tests. https://github.com/llvm/llvm-project/pull/102622 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[libunwind] [libunwind] Detect cycles of length 1 (PR #103476)
https://github.com/MaskRay requested changes to this pull request. . https://github.com/llvm/llvm-project/pull/103476 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[libunwind] [libunwind] Detect cycles of length 1 (PR #103476)
MaskRay wrote: > This happens with musl libc in __clone(), which has incorrect DWARF unwind > information. Can you elaborate this? How is it incorrect? (tools/add-cfi.x86_64.awk on `musl/src/thread/x86_64/clone.s`, I assume) I guess this workaround does address your issue, but for the upstream we really need more principled fix, instead of adding a hack. https://github.com/llvm/llvm-project/pull/103476 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [llvm] [Driver] Default -msmall-data-limit= to 0 and clean up code (PR #83093)
MaskRay wrote: Thanks! I'll land this in a few days. I believe it's now up to the gp default users to justify the default and why they cannot user a Clang configuration file (this is difficult to justify, as we've many many decisions to Clang configuration files). I think this is important to ensure that future sdata/sbss development like #87040 not cause disruption to users who don't use the feature. https://github.com/llvm/llvm-project/pull/83093 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Driver] Introduce ffp-model=aggressive (PR #100453)
MaskRay wrote: > @MaskRay Are you OK with this change? Sorry for the delay. I've read through https://discourse.llvm.org/t/making-ffp-model-fast-more-user-friendly/78402, how we ended up with `-ffp-contract=fast-honor-pragmas` beside `=fast` (GCC compatibility, sigh), some notes that ICC/GCC have behavior differences. A different, safer mode for `-ffp-model` (absent from GCC) makes sense to me. https://github.com/llvm/llvm-project/pull/100453 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Driver] Introduce ffp-model=aggressive (PR #100453)
@@ -163,23 +183,41 @@ // CHECK-FEB-IGNORE: "-fno-rounding-math" // CHECK-FEB-IGNORE: "-ffp-exception-behavior=ignore" -// RUN: %clang -### -nostdinc -Werror -ffast-math -ffp-model=fast -c %s 2>&1 \ +// RUN: %clang -### -nostdinc -Werror -ffast-math -ffp-model=aggressive -c %s 2>&1 \ +// RUN: | FileCheck --check-prefix=CHECK-FASTMATH-FPM-AGGR %s +// CHECK-FASTMATH-FPM-AGGR: "-cc1" MaskRay wrote: perhaps pack some options on the same line to show that they are adjacent (no other option in between) If this extra check isn't important, I expect at least `-SAME:` https://github.com/llvm/llvm-project/pull/100453 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Driver] Introduce ffp-model=aggressive (PR #100453)
@@ -109,20 +112,37 @@
// CHECK-TRAP: "-cc1"
// CHECK-TRAP: "-ffp-exception-behavior=strict"
+// RUN: %clang -### -nostdinc -ffp-model=aggressive -c %s 2>&1 \
+// RUN: | FileCheck --check-prefix=CHECK-FPM-AGGR %s
+// CHECK-FPM-AGGR: "-cc1"
+// CHECK-FPM-AGGR: "-menable-no-infs"
MaskRay wrote:
see my other comment: pack some options (to show they are adjacent) or use
`-SAME: {{^}} "-menable-no-nans"` (clang/test/Driver/linux-cross.cpp style)
https://github.com/llvm/llvm-project/pull/100453
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Driver] Introduce ffp-model=aggressive (PR #100453)
@@ -3033,8 +3039,8 @@ static void RenderFloatingPointOptions(const ToolChain
&TC, const Driver &D,
SignedZeros = true;
StringRef Val = A->getValue();
- if (OFastEnabled && Val != "fast") {
-// Only -ffp-model=fast is compatible with OFast, ignore.
+ if (OFastEnabled && Val != "aggressive") {
+// Only -ffp-model=aggressive is compatible with OFast, ignore.
MaskRay wrote:
`-Ofast`.
There isn't a variable named `OFast`.
https://github.com/llvm/llvm-project/pull/100453
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Driver] Introduce ffp-model=aggressive (PR #100453)
https://github.com/MaskRay approved this pull request. fp-model= => -ffp-model= LGTM. I expect that the subject explicit mentions -ffp-model=fast behavior difference, even if it becomes less aggressive and unlikely breaks users. With "Introduce ...", nobody would expect behavior differences. https://github.com/llvm/llvm-project/pull/100453 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Driver] Introduce ffp-model=aggressive (PR #100453)
@@ -2879,10 +2879,31 @@ static void RenderFloatingPointOptions(const ToolChain
&TC, const Driver &D,
std::string ComplexRangeStr = "";
std::string GccRangeComplexOption = "";
+ auto setComplexRange = [&](LangOptions::ComplexRangeKind NewRange) {
+// Warn if user expects to perform full implementation of complex
+// multiplication or division in the presence of nan or ninf flags.
MaskRay wrote:
typo: nan => nnan ?
https://github.com/llvm/llvm-project/pull/100453
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Driver] Introduce ffp-model=aggressive (PR #100453)
@@ -2,13 +2,13 @@ // and other floating point options get a warning diagnostic. // -// RUN: %clang -### -ffp-model=fast -ffp-contract=off -c %s 2>&1 \ +// RUN: %clang -### -ffp-model=aggressive -ffp-contract=off -c %s 2>&1 \ MaskRay wrote: It's still useful to test the behavior of `-ffp-model=fast -ffp-contract=off` https://github.com/llvm/llvm-project/pull/100453 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [Driver] Introduce ffp-model=aggressive (PR #100453)
https://github.com/MaskRay edited https://github.com/llvm/llvm-project/pull/100453 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang][NFC] Split invalid-cpu-note tests (PR #104601)
@@ -0,0 +1,90 @@
+// Use --implicit-check-not to ensure no additional CPUs are in this list
+
+// RUN: not %clang_cc1 -triple arm64--- -target-cpu not-a-cpu -fsyntax-only %s
2>&1 | FileCheck %s --implicit-check-not={{[a-zA-Z0-9]}}
+// RUN: not %clang_cc1 -triple arm64--- -tune-cpu not-a-cpu -fsyntax-only %s
2>&1 | FileCheck %s --implicit-check-not={{[a-zA-Z0-9]}}
+
+// CHECK: error: unknown target CPU 'not-a-cpu'
+// CHECK-NEXT: note: valid target CPU values are:
+// CHECK-SAME: a64fx,
MaskRay wrote:
Instead of having every CPU one the same line, you can group them by families
like:
```
{{^}}, ampere1, ampere1a, ampere1b
{{^}}, fooX, fooY, fooZ
```
which still looks quite good.
Word diff tools are quite robust, so I think strictly avoiding packing is not
necessary.
https://github.com/llvm/llvm-project/pull/104601
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang][NFC] Split invalid-cpu-note tests (PR #104601)
@@ -0,0 +1,90 @@
+// Use --implicit-check-not to ensure no additional CPUs are in this list
+
+// RUN: not %clang_cc1 -triple arm64--- -target-cpu not-a-cpu -fsyntax-only %s
2>&1 | FileCheck %s --implicit-check-not={{[a-zA-Z0-9]}}
+// RUN: not %clang_cc1 -triple arm64--- -tune-cpu not-a-cpu -fsyntax-only %s
2>&1 | FileCheck %s --implicit-check-not={{[a-zA-Z0-9]}}
+
+// CHECK: error: unknown target CPU 'not-a-cpu'
+// CHECK-NEXT: note: valid target CPU values are:
+// CHECK-SAME: a64fx,
MaskRay wrote:
clang/test/Driver/linux-cross.cpp uses the `ARCH_I686-SAME: {{^}} xxx` pattern
to check the content on the same line. You can use the same pattern. I think
it's useful to pack some elements on the same line, e.g. `{{^}}, ampere1,
ampere1a, ampere1b`
https://github.com/llvm/llvm-project/pull/104601
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] f861e33 - [Driver] Reject -Wa, -mrelax-relocations= for non-ELF
Author: Fangrui Song
Date: 2024-08-15T18:01:10-07:00
New Revision: f861e33912732d3de05369cdff852968fcdcb3a0
URL:
https://github.com/llvm/llvm-project/commit/f861e33912732d3de05369cdff852968fcdcb3a0
DIFF:
https://github.com/llvm/llvm-project/commit/f861e33912732d3de05369cdff852968fcdcb3a0.diff
LOG: [Driver] Reject -Wa,-mrelax-relocations= for non-ELF
Added:
Modified:
clang/lib/Driver/ToolChains/Clang.cpp
clang/test/Driver/relax.c
Removed:
diff --git a/clang/lib/Driver/ToolChains/Clang.cpp
b/clang/lib/Driver/ToolChains/Clang.cpp
index 2b7241964a345d..daffa7cf96978a 100644
--- a/clang/lib/Driver/ToolChains/Clang.cpp
+++ b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -2582,6 +2582,7 @@ static void CollectArgsForIntegratedAssembler(Compilation
&C,
bool TakeNextArg = false;
const llvm::Triple &Triple = C.getDefaultToolChain().getTriple();
+ bool IsELF = Triple.isOSBinFormatELF();
bool Crel = false, ExperimentalCrel = false;
bool UseRelaxRelocations = C.getDefaultToolChain().useRelaxRelocations();
bool UseNoExecStack = false;
@@ -2621,10 +2622,16 @@ static void
CollectArgsForIntegratedAssembler(Compilation &C,
continue; // LLVM handles bigobj automatically
auto Equal = Value.split('=');
- auto checkArg = [&](std::initializer_list Set) {
-if (!llvm::is_contained(Set, Equal.second))
+ auto checkArg = [&](bool ValidTarget,
+ std::initializer_list Set) {
+if (!ValidTarget) {
+ D.Diag(diag::err_drv_unsupported_opt_for_target)
+ << (Twine("-Wa,") + Equal.first + "=").str()
+ << Triple.getTriple();
+} else if (!llvm::is_contained(Set, Equal.second)) {
D.Diag(diag::err_drv_unsupported_option_argument)
<< (Twine("-Wa,") + Equal.first + "=").str() << Equal.second;
+}
};
switch (C.getDefaultToolChain().getArch()) {
default:
@@ -2634,7 +2641,7 @@ static void CollectArgsForIntegratedAssembler(Compilation
&C,
if (Equal.first == "-mrelax-relocations" ||
Equal.first == "--mrelax-relocations") {
UseRelaxRelocations = Equal.second == "yes";
- checkArg({"yes", "no"});
+ checkArg(IsELF, {"yes", "no"});
continue;
}
if (Value == "-msse2avx") {
@@ -2656,7 +2663,7 @@ static void CollectArgsForIntegratedAssembler(Compilation
&C,
if (Equal.first == "-mimplicit-it") {
// Only store the value; the last value set takes effect.
ImplicitIt = Equal.second;
- checkArg({"always", "never", "arm", "thumb"});
+ checkArg(true, {"always", "never", "arm", "thumb"});
continue;
}
if (Value == "-mthumb")
diff --git a/clang/test/Driver/relax.c b/clang/test/Driver/relax.c
index 9315a0045f1f43..93249cac7f40f8 100644
--- a/clang/test/Driver/relax.c
+++ b/clang/test/Driver/relax.c
@@ -8,3 +8,6 @@
// RUN: not %clang -### --target=aarch64 -c -Wa,-mrelax-relocations=no %s 2>&1
| FileCheck %s --check-prefix=ERR2
// ERR2: error: unsupported argument '-mrelax-relocations=no' to option '-Wa,'
+
+// RUN: not %clang -### --target=x86_64-apple-darwin -c
-Wa,-mrelax-relocations=no %s 2>&1 | FileCheck %s --check-prefix=ERR3
+// ERR3: error: unsupported option '-Wa,-mrelax-relocations=' for target
'x86_64-apple-darwin'
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] 812e049 - [Driver] Correctly handle -Wa,--crel -Wa,--no-crel
Author: Fangrui Song
Date: 2024-08-15T17:32:16-07:00
New Revision: 812e049ac1c2424f96746ac4c453d6255b0a6ca5
URL:
https://github.com/llvm/llvm-project/commit/812e049ac1c2424f96746ac4c453d6255b0a6ca5
DIFF:
https://github.com/llvm/llvm-project/commit/812e049ac1c2424f96746ac4c453d6255b0a6ca5.diff
LOG: [Driver] Correctly handle -Wa,--crel -Wa,--no-crel
Follow-up to #97378
Added:
Modified:
clang/lib/Driver/ToolChains/CommonArgs.cpp
clang/test/Driver/crel.c
Removed:
diff --git a/clang/lib/Driver/ToolChains/CommonArgs.cpp
b/clang/lib/Driver/ToolChains/CommonArgs.cpp
index 1cba3e1220264..320d2901da06e 100644
--- a/clang/lib/Driver/ToolChains/CommonArgs.cpp
+++ b/clang/lib/Driver/ToolChains/CommonArgs.cpp
@@ -1143,8 +1143,8 @@ void tools::addLTOOptions(const ToolChain &ToolChain,
const ArgList &Args,
addMachineOutlinerArgs(D, Args, CmdArgs, ToolChain.getEffectiveTriple(),
/*IsLTO=*/true, PluginOptPrefix);
+ bool Crel = false;
for (const Arg *A : Args.filtered(options::OPT_Wa_COMMA)) {
-bool Crel = false;
for (StringRef V : A->getValues()) {
if (V == "--crel")
Crel = true;
@@ -1154,13 +1154,13 @@ void tools::addLTOOptions(const ToolChain &ToolChain,
const ArgList &Args,
continue;
A->claim();
}
-if (Crel) {
- if (Triple.isOSBinFormatELF() && !Triple.isMIPS()) {
-CmdArgs.push_back(Args.MakeArgString(Twine(PluginOptPrefix) +
"-crel"));
- } else {
-D.Diag(diag::err_drv_unsupported_opt_for_target)
-<< "-Wa,--crel" << D.getTargetTriple();
- }
+ }
+ if (Crel) {
+if (Triple.isOSBinFormatELF() && !Triple.isMIPS()) {
+ CmdArgs.push_back(Args.MakeArgString(Twine(PluginOptPrefix) + "-crel"));
+} else {
+ D.Diag(diag::err_drv_unsupported_opt_for_target)
+ << "-Wa,--crel" << D.getTargetTriple();
}
}
}
diff --git a/clang/test/Driver/crel.c b/clang/test/Driver/crel.c
index a47e7ebfeedd8..e1b3f58c6ea5b 100644
--- a/clang/test/Driver/crel.c
+++ b/clang/test/Driver/crel.c
@@ -24,8 +24,10 @@
/// The --allow-experimental-crel error check doesn't apply to LTO.
// RUN: %clang -### --target=x86_64-linux -Werror -flto -Wa,--crel %s 2>&1 |
FileCheck %s --check-prefix=LTO
+// RUN: %clang -### --target=x86_64-linux -Werror -flto -Wa,--crel
-Wa,--no-crel %s 2>&1 | FileCheck %s --check-prefix=LTO-NO
// LTO: "-plugin-opt=-crel"
+// LTO-NO-NOT: "-plugin-opt=-crel"
// RUN: touch %t.o
// RUN: not %clang -### --target=mips64-linux-gnu -flto -Wa,--crel %t.o 2>&1 |
FileCheck %s --check-prefix=ERR
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] 9d63a09 - [Driver] Improve error message for -Wa,-x=unknown
Author: Fangrui Song
Date: 2024-08-15T17:04:41-07:00
New Revision: 9d63a09b452b641e3cc5d88066464b8250bd2bf7
URL:
https://github.com/llvm/llvm-project/commit/9d63a09b452b641e3cc5d88066464b8250bd2bf7
DIFF:
https://github.com/llvm/llvm-project/commit/9d63a09b452b641e3cc5d88066464b8250bd2bf7.diff
LOG: [Driver] Improve error message for -Wa,-x=unknown
Added:
Modified:
clang/lib/Driver/ToolChains/Clang.cpp
clang/test/Driver/arm-target-as-mimplicit-it.s
clang/test/Driver/relax.c
Removed:
diff --git a/clang/lib/Driver/ToolChains/Clang.cpp
b/clang/lib/Driver/ToolChains/Clang.cpp
index c26ebd1ba56ba..2b7241964a345 100644
--- a/clang/lib/Driver/ToolChains/Clang.cpp
+++ b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -2621,6 +2621,11 @@ static void
CollectArgsForIntegratedAssembler(Compilation &C,
continue; // LLVM handles bigobj automatically
auto Equal = Value.split('=');
+ auto checkArg = [&](std::initializer_list Set) {
+if (!llvm::is_contained(Set, Equal.second))
+ D.Diag(diag::err_drv_unsupported_option_argument)
+ << (Twine("-Wa,") + Equal.first + "=").str() << Equal.second;
+ };
switch (C.getDefaultToolChain().getArch()) {
default:
break;
@@ -2629,8 +2634,8 @@ static void CollectArgsForIntegratedAssembler(Compilation
&C,
if (Equal.first == "-mrelax-relocations" ||
Equal.first == "--mrelax-relocations") {
UseRelaxRelocations = Equal.second == "yes";
- if (llvm::is_contained({"yes", "no"}, Equal.second))
-continue;
+ checkArg({"yes", "no"});
+ continue;
}
if (Value == "-msse2avx") {
CmdArgs.push_back("-msse2avx");
@@ -2651,8 +2656,8 @@ static void CollectArgsForIntegratedAssembler(Compilation
&C,
if (Equal.first == "-mimplicit-it") {
// Only store the value; the last value set takes effect.
ImplicitIt = Equal.second;
- if (CheckARMImplicitITArg(Equal.second))
-continue;
+ checkArg({"always", "never", "arm", "thumb"});
+ continue;
}
if (Value == "-mthumb")
// -mthumb has already been processed in ComputeLLVMTriple()
diff --git a/clang/test/Driver/arm-target-as-mimplicit-it.s
b/clang/test/Driver/arm-target-as-mimplicit-it.s
index d30433ab374d5..e40206dfc3a29 100644
--- a/clang/test/Driver/arm-target-as-mimplicit-it.s
+++ b/clang/test/Driver/arm-target-as-mimplicit-it.s
@@ -30,7 +30,7 @@
/// Test invalid input.
// RUN: not %clang -target arm-linux-gnueabi -### -Wa,-mimplicit-it=foo %s
2>&1 | FileCheck %s --check-prefix=INVALID
-// RUN: not %clang -target arm-linux-gnueabi -### -Xassembler
-mimplicit-it=foo %s 2>&1 | FileCheck %s --check-prefix=XINVALID
+// RUN: not %clang -target arm-linux-gnueabi -### -Xassembler
-mimplicit-it=foo %s 2>&1 | FileCheck %s --check-prefix=INVALID
// RUN: not %clang -target arm-linux-gnueabi -### -Wa,-mimplicit-it=always
-Wa,-mimplicit-it=foo %s 2>&1 | FileCheck %s --check-prefix=INVALID
// RUN: not %clang -target arm-linux-gnueabi -###
-Wa,-mimplicit-it=always,-mimplicit-it=foo %s 2>&1 | FileCheck %s
--check-prefix=INVALID
@@ -47,5 +47,4 @@
// NEVER-NOT: "-arm-implicit-it={{.*}}"
// ARM: "-mllvm" "-arm-implicit-it=arm"
// THUMB: "-mllvm" "-arm-implicit-it=thumb"
-// INVALID: error: unsupported argument '-mimplicit-it=foo' to option '-Wa,'
-// XINVALID: error: unsupported argument '-mimplicit-it=foo' to option
'-Xassembler'
+// INVALID: error: unsupported argument 'foo' to option '-Wa,-mimplicit-it='
diff --git a/clang/test/Driver/relax.c b/clang/test/Driver/relax.c
index 4771b9e3af111..9315a0045f1f4 100644
--- a/clang/test/Driver/relax.c
+++ b/clang/test/Driver/relax.c
@@ -4,7 +4,7 @@
// CHECK: "-mrelax-relocations=no"
// RUN: not %clang -### --target=x86_64 -c -Wa,-mrelax-relocations=x %s 2>&1 |
FileCheck %s --check-prefix=ERR
-// ERR: error: unsupported argument '-mrelax-relocations=x' to option '-Wa,'
+// ERR: error: unsupported argument 'x' to option '-Wa,-mrelax-relocations='
// RUN: not %clang -### --target=aarch64 -c -Wa,-mrelax-relocations=no %s 2>&1
| FileCheck %s --check-prefix=ERR2
// ERR2: error: unsupported argument '-mrelax-relocations=no' to option '-Wa,'
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] d156a5a - [Driver] Reject -Wa, -mrelax-relocations= for non-x86
Author: Fangrui Song
Date: 2024-08-15T16:28:42-07:00
New Revision: d156a5a1cb3b3405a6c1e941d112f262c29e15cf
URL:
https://github.com/llvm/llvm-project/commit/d156a5a1cb3b3405a6c1e941d112f262c29e15cf
DIFF:
https://github.com/llvm/llvm-project/commit/d156a5a1cb3b3405a6c1e941d112f262c29e15cf.diff
LOG: [Driver] Reject -Wa,-mrelax-relocations= for non-x86
Similar to other target-specific -Wa, options.
Added:
Modified:
clang/lib/Driver/ToolChains/Clang.cpp
clang/test/Driver/ohos.c
clang/test/Driver/relax.c
Removed:
diff --git a/clang/lib/Driver/ToolChains/Clang.cpp
b/clang/lib/Driver/ToolChains/Clang.cpp
index 96aa930ea28612..c26ebd1ba56bab 100644
--- a/clang/lib/Driver/ToolChains/Clang.cpp
+++ b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -2620,11 +2620,18 @@ static void
CollectArgsForIntegratedAssembler(Compilation &C,
Value == "-mbig-obj")
continue; // LLVM handles bigobj automatically
+ auto Equal = Value.split('=');
switch (C.getDefaultToolChain().getArch()) {
default:
break;
case llvm::Triple::x86:
case llvm::Triple::x86_64:
+if (Equal.first == "-mrelax-relocations" ||
+Equal.first == "--mrelax-relocations") {
+ UseRelaxRelocations = Equal.second == "yes";
+ if (llvm::is_contained({"yes", "no"}, Equal.second))
+continue;
+}
if (Value == "-msse2avx") {
CmdArgs.push_back("-msse2avx");
continue;
@@ -2641,10 +2648,10 @@ static void
CollectArgsForIntegratedAssembler(Compilation &C,
case llvm::Triple::thumbeb:
case llvm::Triple::arm:
case llvm::Triple::armeb:
-if (Value.starts_with("-mimplicit-it=")) {
+if (Equal.first == "-mimplicit-it") {
// Only store the value; the last value set takes effect.
- ImplicitIt = Value.split("=").second;
- if (CheckARMImplicitITArg(ImplicitIt))
+ ImplicitIt = Equal.second;
+ if (CheckARMImplicitITArg(Equal.second))
continue;
}
if (Value == "-mthumb")
@@ -2719,12 +2726,6 @@ static void
CollectArgsForIntegratedAssembler(Compilation &C,
Crel = false;
} else if (Value == "--allow-experimental-crel") {
ExperimentalCrel = true;
- } else if (Value == "-mrelax-relocations=yes" ||
- Value == "--mrelax-relocations=yes") {
-UseRelaxRelocations = true;
- } else if (Value == "-mrelax-relocations=no" ||
- Value == "--mrelax-relocations=no") {
-UseRelaxRelocations = false;
} else if (Value.starts_with("-I")) {
CmdArgs.push_back(Value.data());
// We need to consume the next argument if the current arg is a plain
diff --git a/clang/test/Driver/ohos.c b/clang/test/Driver/ohos.c
index 8de4e6de57f7ff..af8218fb232e60 100644
--- a/clang/test/Driver/ohos.c
+++ b/clang/test/Driver/ohos.c
@@ -7,7 +7,6 @@
// RUN: --sysroot=%S/Inputs/ohos_native_tree/sysroot -fuse-ld=ld
-march=armv7-a -mcpu=cortex-a7 -mfloat-abi=soft 2>&1 \
// RUN: | FileCheck -check-prefixes=CHECK,CHECK-ARM-A7-SOFT %s
// CHECK: {{.*}}clang{{.*}}" "-cc1"
-// CHECK-NOT: "--mrelax-relocations"
// CHECK-NOT: "-munwind-tables"
// CHECK: "-resource-dir" "[[RESOURCE_DIR:[^"]+]]"
// CHECK: "-isysroot" "[[SYSROOT:[^"]+]]"
diff --git a/clang/test/Driver/relax.c b/clang/test/Driver/relax.c
index 48d22fede12d76..4771b9e3af1116 100644
--- a/clang/test/Driver/relax.c
+++ b/clang/test/Driver/relax.c
@@ -1,4 +1,10 @@
-// RUN: %clang -### -c -integrated-as -Wa,--mrelax-relocations=no %s 2>&1 |
FileCheck %s
+// RUN: %clang -### --target=x86_64 -c -Wa,--mrelax-relocations=no %s 2>&1 |
FileCheck %s
// CHECK: "-cc1"
// CHECK: "-mrelax-relocations=no"
+
+// RUN: not %clang -### --target=x86_64 -c -Wa,-mrelax-relocations=x %s 2>&1 |
FileCheck %s --check-prefix=ERR
+// ERR: error: unsupported argument '-mrelax-relocations=x' to option '-Wa,'
+
+// RUN: not %clang -### --target=aarch64 -c -Wa,-mrelax-relocations=no %s 2>&1
| FileCheck %s --check-prefix=ERR2
+// ERR2: error: unsupported argument '-mrelax-relocations=no' to option '-Wa,'
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] 3333ec1 - [Driver] Make CodeGenOptions name match MCTargetOptions names
Author: Fangrui Song
Date: 2024-08-15T15:52:20-07:00
New Revision: ec1183803fa5a2353e710b8b22db6a72e292
URL:
https://github.com/llvm/llvm-project/commit/ec1183803fa5a2353e710b8b22db6a72e292
DIFF:
https://github.com/llvm/llvm-project/commit/ec1183803fa5a2353e710b8b22db6a72e292.diff
LOG: [Driver] Make CodeGenOptions name match MCTargetOptions names
* Initialize `X86RelaxRelocations`.
* Fix #96860 test to actually test -Wa,-msse2avx for non-x86.
Added:
Modified:
clang/include/clang/Basic/CodeGenOptions.def
clang/include/clang/Driver/Options.td
clang/lib/CodeGen/BackendUtil.cpp
clang/test/Driver/msse2avx.c
clang/tools/driver/cc1as_main.cpp
Removed:
diff --git a/clang/include/clang/Basic/CodeGenOptions.def
b/clang/include/clang/Basic/CodeGenOptions.def
index 5728ab8d86702..09e892d6d4def 100644
--- a/clang/include/clang/Basic/CodeGenOptions.def
+++ b/clang/include/clang/Basic/CodeGenOptions.def
@@ -37,8 +37,6 @@ VALUE_CODEGENOPT(Name, Bits, Default)
CODEGENOPT(DisableIntegratedAS, 1, 0) ///< -no-integrated-as
CODEGENOPT(Crel, 1, 0) ///< -Wa,--crel
-CODEGENOPT(RelaxELFRelocations, 1, 1) ///< -Wa,-mrelax-relocations={yes,no}
-CODEGENOPT(SSE2AVX , 1, 0) ///< -msse2avx
CODEGENOPT(AsmVerbose, 1, 0) ///< -dA, -fverbose-asm.
CODEGENOPT(PreserveAsmComments, 1, 1) ///< -dA, -fno-preserve-as-comments.
CODEGENOPT(AssumeSaneOperatorNew , 1, 1) ///< implicit __attribute__((malloc))
operator new
@@ -194,6 +192,8 @@ CODEGENOPT(HIPSaveKernelArgName, 1, 0) ///< Set when
-fhip-kernel-arg-name is en
CODEGENOPT(UniqueInternalLinkageNames, 1, 0) ///< Internal Linkage symbols get
unique names.
CODEGENOPT(SplitMachineFunctions, 1, 0) ///< Split machine functions using
profile information.
CODEGENOPT(PPCUseFullRegisterNames, 1, 0) ///< Print full register names in
assembly
+CODEGENOPT(X86RelaxRelocations, 1, 1) ///< -Wa,-mrelax-relocations={yes,no}
+CODEGENOPT(X86Sse2Avx, 1, 0) ///< -Wa,-msse2avx
/// When false, this attempts to generate code as if the result of an
/// overflowing conversion matches the overflowing behavior of a target's
native
diff --git a/clang/include/clang/Driver/Options.td
b/clang/include/clang/Driver/Options.td
index 40df91dc3fe0e..c9ed08c20fc04 100644
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -5231,7 +5231,7 @@ let Flags = [TargetSpecific] in {
def msse2avx : Flag<["-"], "msse2avx">, Group,
Visibility<[ClangOption, CC1Option, CC1AsOption]>,
HelpText<"Specify that the assembler should encode SSE instructions with VEX
prefix">,
- MarshallingInfoFlag>;
+ MarshallingInfoFlag>;
} // let Flags = [TargetSpecific]
defm zvector : BoolFOption<"zvector",
@@ -7133,7 +7133,7 @@ def crel : Flag<["--"], "crel">,
MarshallingInfoFlag>;
def mrelax_relocations_no : Flag<["-"], "mrelax-relocations=no">,
HelpText<"Disable x86 relax relocations">,
-MarshallingInfoNegativeFlag>;
+MarshallingInfoNegativeFlag>;
def msave_temp_labels : Flag<["-"], "msave-temp-labels">,
HelpText<"Save temporary labels in the symbol table. "
"Note this may change .s semantics and shouldn't generally be used "
diff --git a/clang/lib/CodeGen/BackendUtil.cpp
b/clang/lib/CodeGen/BackendUtil.cpp
index 81e6702d5de66..34c08818dbb9a 100644
--- a/clang/lib/CodeGen/BackendUtil.cpp
+++ b/clang/lib/CodeGen/BackendUtil.cpp
@@ -471,7 +471,7 @@ static bool initTargetOptions(DiagnosticsEngine &Diags,
Options.MCOptions.Dwarf64 = CodeGenOpts.Dwarf64;
Options.MCOptions.PreserveAsmComments = CodeGenOpts.PreserveAsmComments;
Options.MCOptions.Crel = CodeGenOpts.Crel;
- Options.MCOptions.X86RelaxRelocations = CodeGenOpts.RelaxELFRelocations;
+ Options.MCOptions.X86RelaxRelocations = CodeGenOpts.X86RelaxRelocations;
Options.MCOptions.CompressDebugSections =
CodeGenOpts.getCompressDebugSections();
Options.MCOptions.ABIName = TargetOpts.ABI;
diff --git a/clang/test/Driver/msse2avx.c b/clang/test/Driver/msse2avx.c
index a63ac9a6c8668..829b1b2b79360 100644
--- a/clang/test/Driver/msse2avx.c
+++ b/clang/test/Driver/msse2avx.c
@@ -1,7 +1,7 @@
-// RUN: %clang -### -c -target x86_64 -march=x86-64 -Xassembler -msse2avx %s
2>&1 | FileCheck %s
-// RUN: %clang -### -c -target x86_64 -march=x86-64 -x assembler -Xassembler
-msse2avx %s 2>&1 | FileCheck %s
+// RUN: %clang -### -c --target=x86_64 -march=x86-64 -Wa,-msse2avx %s 2>&1 |
FileCheck %s
+// RUN: %clang -### -c --target=x86_64 -march=x86-64 -x assembler -Xassembler
-msse2avx %s 2>&1 | FileCheck %s
// CHECK: "-msse2avx"
-// RUN: not %clang -### -c -target aarch64 -march=armv8a -msse2avx %s 2>&1 |
FileCheck --check-prefix=ERR %s
-// ERR: error: unsupported option '-msse2avx' for target 'aarch64'
+// RUN: not %clang -### -c --target=aarch64 -march=armv8a -Wa,-msse2avx %s
2>&1 | FileCheck --check-prefix=ERR %s
+//
[clang] [clang] Support -Wa, options -mmsa and -mno-msa (PR #99615)
https://github.com/MaskRay edited https://github.com/llvm/llvm-project/pull/99615 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang] Support -Wa, options -mmsa and -mno-msa (PR #99615)
@@ -2630,6 +2631,10 @@ static void
CollectArgsForIntegratedAssembler(Compilation &C,
CmdArgs.push_back("-massembler-no-warn");
} else if (Value == "--noexecstack") {
UseNoExecStack = true;
+ } else if (Value == "-mmsa") {
MaskRay wrote:
This code block should be moved above to the `switch`. There is a `case
llvm::Triple::mips:` for MIPS.
https://github.com/llvm/llvm-project/pull/99615
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] 9931357 - [-Wunsafe-buffer-usage] Fix warning after #102953
Author: Fangrui Song
Date: 2024-08-15T14:17:26-07:00
New Revision: 99313575b6bb7d5acbe437c2022b6e5dc7f297f1
URL:
https://github.com/llvm/llvm-project/commit/99313575b6bb7d5acbe437c2022b6e5dc7f297f1
DIFF:
https://github.com/llvm/llvm-project/commit/99313575b6bb7d5acbe437c2022b6e5dc7f297f1.diff
LOG: [-Wunsafe-buffer-usage] Fix warning after #102953
Added:
Modified:
clang/lib/Analysis/UnsafeBufferUsage.cpp
Removed:
diff --git a/clang/lib/Analysis/UnsafeBufferUsage.cpp
b/clang/lib/Analysis/UnsafeBufferUsage.cpp
index 95c288bfe3fd2f..da7446913f7c87 100644
--- a/clang/lib/Analysis/UnsafeBufferUsage.cpp
+++ b/clang/lib/Analysis/UnsafeBufferUsage.cpp
@@ -2797,9 +2797,8 @@ fixVariable(const VarDecl *VD, FixitStrategy::Kind K,
return {};
}
case FixitStrategy::Kind::Array: {
-if (VD->isLocalVarDecl())
- if (auto CAT = Ctx.getAsConstantArrayType(VD->getType()))
-return fixVariableWithArray(VD, Tracker, Ctx, Handler);
+if (VD->isLocalVarDecl() && Ctx.getAsConstantArrayType(VD->getType()))
+ return fixVariableWithArray(VD, Tracker, Ctx, Handler);
DEBUG_NOTE_DECL_FAIL(VD, " : not a local const-size array");
return {};
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[clang] [clang][driver] Don't warn when -S and -c are used together without -fsyntax-only (PR #104477)
MaskRay wrote: > Fixes https://github.com/llvm/llvm-project/issues/104347 I made a comment on #104347 . The warning for `-S -c` works as intended. https://github.com/llvm/llvm-project/pull/104477 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
