This patch adds support for the MSVC /HOTPATCH flag: https://docs.microsoft.com/sv-se/cpp/build/reference/hotpatch-create-hotpatchable-image?view=msvc-170&viewFallbackFrom=vs-2019
The flag is translated to a new -fms-hotpatch flag, which in turn adds a 'patchable-function' attribute for each function in the TU. This is then picked up by the PatchableFunction pass which would generate a TargetOpcode::PATCHABLE_OP of minsize = 2 (which means the target instruction must resolve to at least two bytes). TargetOpcode::PATCHABLE_OP is only implemented for x86/x64. When targetting ARM/ARM64, /HOTPATCH isn't required (instructions are always 2/4 bytes and suitable for hotpatching).
Additionally, when using /Z7, we generate a 'hot patchable' flag in the CodeView debug stream, in the S_COMPILE3 record. This flag is then picked up by LLD (or link.exe) and is used in conjunction with the linker /FUNCTIONPADMIN flag to generate extra space before each function, to accommodate for live patching long jumps. Please see: d703b92296/lld/COFF/Writer.cpp (L1298)
The outcome is that we can finally use Live++ or Recode along with clang-cl.
NOTE: It seems that MSVC cl.exe always enables /HOTPATCH on x64 by default, although if we did the same I thought we might generate sub-optimal code (if this flag was active by default). Additionally, MSVC always generates a .debug$S section and a S_COMPILE3 record, which Clang doesn't do without /Z7. Therefore, the following MSVC command-line "cl /c file.cpp" would have to be written with Clang such as "clang-cl /c file.cpp /HOTPATCH /Z7" in order to obtain the same result.
Depends on D43002, D80833 and D81301 for the full feature.
Differential Revision: https://reviews.llvm.org/D116511
This was intended to be fixed by D98856, but that only seemed to have
the desired behaviour when compiling to assembly using `-S`, not when
compiling into an object file or executable. Given that this was not
the intention of D98856, this patch fixes the behaviour.
Added warning for potential cases of
unaligned access when option
-mno-unaligned-access has been specified
Differential Revision: https://reviews.llvm.org/D116221
Early revisions of the VR4300 have a hardware bug where two consecutive
multiplications can produce an incorrect result in the second multiply.
This revision adds the `-mfix4300` flag to llvm (and clang) which, when
passed, provides a software fix for this issue.
More precise description of the "mulmul" bug:
```
mul.[s,d] fd,fs,ft
mul.[s,d] fd,fs,ft or [D]MULT[U] rs,rt
```
When the above sequence is executed by the CPU, if at least one of the
source operands of the first mul instruction happens to be `sNaN`, `0`
or `Infinity`, then the second mul instruction may produce an incorrect
result. This can happen both if the two mul instructions are next to each
other and if the first one is in a delay slot and the second is the first
instruction of the branch target.
Description of the fix:
This fix adds a backend pass to llvm which scans for mul instructions in
each basic block and inserts a nop whenever the following conditions are
met:
- The current instruction is a single or double-precision floating-point
mul instruction.
- The next instruction is either a mul instruction (any kind) or a branch
instruction.
Differential Revision: https://reviews.llvm.org/D116238
The Linux kernel has a make macro called cc-option that invokes the
compiler with an option in isolation to see if it is supported before
adding it to CFLAGS. The exit code of the compiler is used to determine
if the flag is supported and should be added to the compiler invocation.
A call to cc-option with '-mno-outline-atomics' was added to prevent
linking errors with newer GCC versions but this call succeeds with a
non-AArch64 target because there is no warning from clang with
'-mno-outline-atomics', just '-moutline-atomics'. Because the call
succeeds and adds '-mno-outline-atomics' to the compiler invocation,
there is a warning from LLVM because the 'outline-atomics target
feature is only supported by the AArch64 backend.
$ echo | clang -target x86_64 -moutline-atomics -Werror -x c -c -o /dev/null -
clang-14: error: The 'x86_64' architecture does not support -moutline-atomics; flag ignored [-Werror,-Woption-ignored]
$ echo $?
1
$ echo | clang -target x86_64 -mno-outline-atomics -Werror -x c -c -o /dev/null -
'-outline-atomics' is not a recognized feature for this target (ignoring feature)
$ echo $?
0
This does not match GCC's behavior, which errors when the flag is added
to a non-AArch64 target.
$ echo | gcc -moutline-atomics -x c -c -o /dev/null -
gcc: error: unrecognized command-line option ‘-moutline-atomics’; did you mean ‘-finline-atomics’?
$ echo | gcc -mno-outline-atomics -x c -c -o /dev/null -
gcc: error: unrecognized command-line option ‘-mno-outline-atomics’; did you mean ‘-fno-inline-atomics’?
$ echo | aarch64-linux-gnu-gcc -moutline-atomics -x c -c -o /dev/null -
$ echo | aarch64-linux-gnu-gcc -mno-outline-atomics -x c -c -o /dev/null -
To get closer to GCC's behavior, issue a warning when
'-mno-outline-atomics' is used without an AArch64 triple and do not add
'{-,+}outline-atomic" to the list of target features in these cases.
Link: https://github.com/ClangBuiltLinux/linux/issues/1552
Reviewed By: melver, nickdesaulniers
Differential Revision: https://reviews.llvm.org/D116128
This patch adds a toolchain (TC) for SPIR-V along with the
following changes in Driver and base ToolChain and Tool.
This is required to provide a mechanism in clang to bypass
SPIR-V backend in LLVM for SPIR-V until it lands in LLVM and
matures.
The SPIR-V code is generated by the SPIRV-LLVM translator tool
named 'llvm-spirv' that is sought in 'PATH'.
The compilation phases/actions should be bound for SPIR-V in
the meantime as following:
compile -> tools::Clang
backend -> tools::SPIRV::Translator
assemble -> tools::SPIRV::Translator
However, Driver’s ToolSelector collapses compile-backend-assemble
and compile-backend sequences to tools::Clang. To prevent this,
added new {use,has}IntegratedBackend properties in ToolChain and
Tool to which the ToolSelector reacts on, and which SPIR-V TC
overrides.
Linking of multiple input files is currently not supported but
can be added separately.
Differential Revision: https://reviews.llvm.org/D112410
Co-authored-by: Henry Linjamäki <henry.linjamaki@parmance.com>
Reland integrates build fixes & further review suggestions.
Thanks to @zturner for the initial S_OBJNAME patch!
Differential Revision: https://reviews.llvm.org/D43002
Also revert all subsequent fixes:
- abd1cbf5e5 [Clang] Disable debug-info-objname.cpp test on Unix until I sort out the issue.
- 00ec441253 [Clang] debug-info-objname.cpp test: explictly encode a x86 target when using %clang_cl to avoid falling back to a native CPU triple.
- cd407f6e52 [Clang] Fix build by restricting debug-info-objname.cpp test to x86.
This patch refactors the HIP tool chain for new HIP tool chain, HIPSPV
tool chain, which is added in the follow up patch part 2.
Rename HIPToolChain to HIPAMDToolChain and Renames HIP.* files to HIPAMD.*.
Introduce HIPUtility.* file where common HIP utilities, shared among HIP
tool chain implementations, are placed in.
Move constructHIPFatbinCommand() and
constructGenerateObjFileFromHIPFatBinary() to HIPUtility. HIPSPV tool
chain is going to use them.
Tweak bundle target ID in constructHIPFatbinCommand(): extra dashes are
dropped if the Target ID is empty and 'hip' offload kind is made default
for non-AMD targets.
Patch by: Henry Linjamäki
Reviewed by: Yaxun Liu, Artem Belevich, Eric Christopher
Differential Revision: https://reviews.llvm.org/D110549
When this pass was originally implemented, the fix pass was enabled
using a llvm command-line flag. This works fine, except in the case of
LTO, where the flag is not passed into the linker plugin in order to
enable the function pass in the LTO backend.
Now LTO exists, the expectation now is to use target features rather
than command-line arguments to control code generation, as this ensures
that different command-line arguments in different files are correctly
represented, and target-features always get to the LTO plugin as they
are encoded into LLVM IR.
The fall-out of this change is that the fix pass has to always be added
to the backend pass pipeline, so now it makes no changes if the function
does not have the right target feature to enable it. This should make a
minimal difference to compile time.
One advantage is it's now much easier to enable when compiling for a
Cortex-A53, as CPUs imply their own individual sets of target-features,
in a more fine-grained way. I haven't done this yet, but it is an
option, if the fix should be enabled in more places.
Existing tests of the user interface are unaffected, the changes are to
reflect that the argument is now turned into a target feature.
Reviewed By: tmatheson
Differential Revision: https://reviews.llvm.org/D114703
The new runtime is currently broken for AMD offloading. This patch makes
the default the old runtime only for the AMD target.
Reviewed By: ronlieb
Differential Revision: https://reviews.llvm.org/D114965
This patch changes the `-fopenmp-target-new-runtime` option which controls if
the new or old device runtime is used to be true by default. Disabling this to
use the old runtime now requires using `-fno-openmp-target-new-runtime`.
Reviewed By: JonChesterfield, tianshilei1992, gregrodgers, ronlieb
Differential Revision: https://reviews.llvm.org/D114890
Handle branch protection option on the commandline as well as a function
attribute. One patch for both mechanisms, as they use the same underlying
parsing mechanism.
These are recorded in a set of LLVM IR module-level attributes like we do for
AArch64 PAC/BTI (see https://reviews.llvm.org/D85649):
- command-line options are "translated" to module-level LLVM IR
attributes (metadata).
- functions have PAC/BTI specific attributes iff the
__attribute__((target("branch-protection=...))) was used in the function
declaration.
- command-line option -mbranch-protection to armclang targeting Arm,
following this grammar:
branch-protection ::= "-mbranch-protection=" <protection>
protection ::= "none" | "standard" | "bti" [ "+" <pac-ret-clause> ]
| <pac-ret-clause> [ "+" "bti"]
pac-ret-clause ::= "pac-ret" [ "+" <pac-ret-option> ]
pac-ret-option ::= "leaf" ["+" "b-key"] | "b-key" ["+" "leaf"]
b-key is simply a placeholder to make it consistent with AArch64's
version. In Arm, however, it triggers a warning informing that b-key is
unsupported and a-key will be selected instead.
- Handle _attribute_((target(("branch-protection=..."))) for AArch32 with the
same grammer as the commandline options.
This patch is part of a series that adds support for the PACBTI-M extension of
the Armv8.1-M architecture, as detailed here:
https://community.arm.com/arm-community-blogs/b/architectures-and-processors-blog/posts/armv8-1-m-pointer-authentication-and-branch-target-identification-extension
The PACBTI-M specification can be found in the Armv8-M Architecture Reference
Manual:
https://developer.arm.com/documentation/ddi0553/latest
The following people contributed to this patch:
- Momchil Velikov
- Victor Campos
- Ties Stuij
Reviewed By: vhscampos
Differential Revision: https://reviews.llvm.org/D112421
Allow toggling of -fnew-infallible so last instance takes precedence
Testing:
ninja check-all
Reviewed By: bruno
Differential Revision: https://reviews.llvm.org/D113523
[NFC] As part of using inclusive language within the llvm project, this patch
renames master flag to main flag in these comments.
Reviewed By: ZarkoCA
Differential Revision: https://reviews.llvm.org/D114090
From GCC's manpage:
-fplugin-arg-name-key=value
Define an argument called key with a value of value for the
plugin called name.
Since we don't have a key-value pair similar to gcc's plugin_argument
struct, simply accept key=value here anyway and pass it along as-is to
plugins.
This translates to the already existing '-plugin-arg-pluginname arg'
that clang cc1 accepts.
There is an ambiguity here because in clang, both the plugin name
as well as the option name can contain dashes, so when e.g. passing
-fplugin-arg-foo-bar-foo
it is not clear whether the plugin is foo-bar and the option is foo,
or the plugin is foo and the option is bar-foo. GCC solves this by
interpreting all dashes as part of the option name. So dashes can't be
part of the plugin name in this case.
Differential Revision: https://reviews.llvm.org/D113250
AMD64 ABI mandates caller to specify the number of used SSE registers
when passing variable arguments.
GCC also provides option -mskip-rax-setup to skip the setup of rax when
SSE is disabled. This helps to reduce the code size, see pr23258.
Reviewed By: nickdesaulniers
Differential Revision: https://reviews.llvm.org/D112413
ld.lld used by Android ignores .note.GNU-stack and defaults to noexecstack,
so the `-z noexecstack` linker option is unneeded.
The `--noexecstack` assembler option is unneeded because AsmPrinter.cpp
prints `.section .note.GNU-stack,"",@progbits` (when `llvm.init.trampoline` is unused),
so the assembler won't synthesize an executable .note.GNU-stack.
Reviewed By: danalbert
Differential Revision: https://reviews.llvm.org/D113840
With this,
void f() { __asm__("mov eax, ebx"); }
now compiles with clang with -masm=intel.
This matches gcc.
The flag is not accepted in clang-cl mode. It has no effect on
MSVC-style `__asm {}` blocks, which are unconditionally in intel
mode both before and after this change.
One difference to gcc is that in clang, inline asm strings are
"local" while they're "global" in gcc. Building the following with
-masm=intel works with clang, but not with gcc where the ".att_syntax"
from the 2nd __asm__() is in effect until file end (or until a
".intel_syntax" somewhere later in the file):
__asm__("mov eax, ebx");
__asm__(".att_syntax\nmovl %ebx, %eax");
__asm__("mov eax, ebx");
This also updates clang's intrinsic headers to work both in
-masm=att (the default) and -masm=intel modes.
The official solution for this according to "Multiple assembler dialects in asm
templates" in gcc docs->Extensions->Inline Assembly->Extended Asm
is to write every inline asm snippet twice:
bt{l %[Offset],%[Base] | %[Base],%[Offset]}
This works in LLVM after D113932 and D113894, so use that.
(Just putting `.att_syntax` at the start of the snippet works in some but not
all cases: When LLVM interpolates in parameters like `%0`, it uses at&t or
intel syntax according to the inline asm snippet's flavor, so the `.att_syntax`
within the snippet happens to late: The interpolated-in parameter is already
in intel style, and then won't parse in the switched `.att_syntax`.)
It might be nice to invent a `#pragma clang asm_dialect push "att"` /
`#pragma clang asm_dialect pop` to be able to force asm style per snippet,
so that the inline asm string doesn't contain the same code in two variants,
but let's leave that for a follow-up.
Fixes PR21401 and PR20241.
Differential Revision: https://reviews.llvm.org/D113707
Change the error message to use ignorelist, and changed some variable and function
names in related code and test.
Reviewed By: thakis
Differential Revision: https://reviews.llvm.org/D113189
The driver uses class SanitizerArgs to store parsed sanitizer arguments. It keeps a cached
SanitizerArgs object in ToolChain and uses it for different jobs. This does not work if
the sanitizer options are different for different jobs, which could happen when an
offloading toolchain translates the options for different jobs.
To fix this, SanitizerArgs should be created by using the actual arguments passed
to jobs instead of the original arguments passed to the driver, since the toolchain
may change the original arguments. And the sanitizer arguments should be diagnose
once.
This patch also fixes HIP toolchain for handling -fgpu-sanitize: a warning is emitted
for GPU's not supporting sanitizer and skipped. This is for backward compatibility
with existing -fsanitize options. -fgpu-sanitize is also turned on by default.
Reviewed by: Artem Belevich, Evgenii Stepanov
Differential Revision: https://reviews.llvm.org/D111443
Implement support for loading the stack canary from a memory location held in
the TLS register, with an optional offset applied. This is used by the Linux
kernel to implement per-task stack canaries, which is impossible on SMP systems
when using a global variable for the stack canary.
Reviewed By: nickdesaulniers
Differential Revision: https://reviews.llvm.org/D112768
Add new triple and target info for ‘spirv32’ and ‘spirv64’ and,
thus, enabling clang (LLVM IR) code emission to SPIR-V target.
The target for SPIR-V is mostly reused from SPIR by derivation
from a common base class since IR output for SPIR-V is mostly
the same as SPIR. Some refactoring are made accordingly.
Added and updated tests for parts that are different between
SPIR and SPIR-V.
Patch by linjamaki (Henry Linjamäki)!
Differential Revision: https://reviews.llvm.org/D109144
Trying to update some options that don't at least have an inclusive language version.
This patch adds `objcmt-allowlist-dir-path` as a default alternative.
Reviewed By: akyrtzi
Differential Revision: https://reviews.llvm.org/D112591
This reverts commit 2d7fba5f95.
The patch was reverted because it caused regression with rocThrust
due to ambiguity of template specialization.
For details please see https://reviews.llvm.org/D109496
A resolution to the ambiguity issues created by P0522, which is a DR solving
CWG 150, did not come as expected, so we are just going to accept the change,
and watch how users digest it.
For now we deprecate the flag with a warning, and make it on by default.
We don't remove the flag completely in order to give users a chance to
work around any problems by disabling it.
Signed-off-by: Matheus Izvekov <mizvekov@gmail.com>
Reviewed By: rsmith
Differential Revision: https://reviews.llvm.org/D109496
This patch splits the existing SveVectorBits LangOpt into VScaleMin and
VScaleMax LangOpts such that we can represent such an option. The cc1
option has also been split into -mvscale-{min,max}=<n> options so that the
cc1 arguments better reflect the vscale_range IR attribute.
Differential Revision: https://reviews.llvm.org/D111790
Representation of the file's last modification time depends on the file
system and isn't guaranteed to be in seconds. Cast to seconds explicitly
and tighten the test case to check the magnitude of the calculated
value, so we can catch passing milliseconds or nanoseconds.
rdar://83915615
Differential Revision: https://reviews.llvm.org/D111205
This patch ensures that we always tune for a given CPU on AArch64
targets when the user specifies the "-mtune=xyz" flag. In the
AArch64Subtarget if the tune flag is unset we use the CPU value
instead.
I've updated the release notes here:
llvm/docs/ReleaseNotes.rst
and added tests here:
clang/test/Driver/aarch64-mtune.c
Differential Revision: https://reviews.llvm.org/D110258
By default clang emits complete contructors as alias of base constructors if they are the same.
The backend is supposed to emit symbols for the alias, otherwise it causes undefined symbols.
@yaxunl observed that this issue is related to the llvm options `-amdgpu-early-inline-all=true`
and `-amdgpu-function-calls=false`. This issue is resolved by only inlining global values
with internal linkage. The `getCalleeFunction()` in AMDGPUResourceUsageAnalysis also had
to be extended to support aliases to functions. inline-calls.ll was corrected appropriately.
Reviewed By: yaxunl, #amdgpu
Differential Revision: https://reviews.llvm.org/D109707
By default clang emits complete contructors as alias of base constructors if they are the same.
The backend is supposed to emit symbols for the alias, otherwise it causes undefined symbols.
@yaxunl observed that this issue is related to the llvm options `-amdgpu-early-inline-all=true`
and `-amdgpu-function-calls=false`. This issue is resolved by only inlining global values
with internal linkage. The `getCalleeFunction()` in AMDGPUResourceUsageAnalysis also had
to be extended to support aliases to functions. inline-calls.ll was corrected appropriately.
Reviewed By: yaxunl, #amdgpu
Differential Revision: https://reviews.llvm.org/D109707
This reland commit 1131b1eb35, which
adds support to __attribute__((availability)) annotation for Fuchsia
platform. This patch also adds '-ffuchsia-api-level' to allow specify
Fuchsia API level from the command line.
Differential Revision: https://reviews.llvm.org/D108592
This patch adds support to __attribute__((availability)) annotation for
Fuchsia platform. This patch also adds '-ffuchsia-api-level' to allow
specify Fuchsia API level from the command line.
Differential Revision: https://reviews.llvm.org/D108592
An archive containing device code object files can be passed to
clang command line for linking. For each given offload target
it creates a device specific archives which is either passed to llvm-link
if the target is amdgpu, or to clang-nvlink-wrapper if the target is
nvptx. -L/-l flags are used to specify these fat archives on the command
line. E.g.
clang++ -fopenmp -fopenmp-targets=nvptx64 main.cpp -L. -lmylib
It currently doesn't support linking an archive directly, like:
clang++ -fopenmp -fopenmp-targets=nvptx64 main.cpp libmylib.a
Linking with x86 offload also does not work.
Reviewed By: ye-luo
Differential Revision: https://reviews.llvm.org/D105191
This patch adds two flags to be supported for the new runtime. The flags
are `-fopenmp-assume-threads-oversubscription` and
-fopenmp-assume-teams-oversubscription`. These add global values that
can be checked by the work sharing runtime functions to make better
judgements about how to distribute work between the threads.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D111348
An archive containing device code object files can be passed to
clang command line for linking. For each given offload target
it creates a device specific archives which is either passed to llvm-link
if the target is amdgpu, or to clang-nvlink-wrapper if the target is
nvptx. -L/-l flags are used to specify these fat archives on the command
line. E.g.
clang++ -fopenmp -fopenmp-targets=nvptx64 main.cpp -L. -lmylib
It currently doesn't support linking an archive directly, like:
clang++ -fopenmp -fopenmp-targets=nvptx64 main.cpp libmylib.a
Linking with x86 offload also does not work.
Reviewed By: ye-luo
Differential Revision: https://reviews.llvm.org/D105191
clang-cl maps /wdNNNN to -Wno-flags for a few warnings that map
cleanly from cl.exe concepts to clang concepts.
This patch adds support for the same numbers to
`#pragma warning(disable : NNNN)`. It also lets
`#pragma warning(push)` and `#pragma warning(pop)` have an effect,
since these are used together with `warning(disable)`.
The optional numeric argument to `warning(push)` is ignored,
as are the other non-`disable` `pragma warning()` arguments.
(Supporting `error` would be easy, but we also don't support
`/we`, and those should probably be added together.)
The motivating example is that a bunch of code (including in LLVM)
uses this idiom to locally disable warnings about calls to deprecated
functions in Windows-only code, and 4996 maps nicely to
-Wno-deprecated-declarations:
#pragma warning(push)
#pragma warning(disable: 4996)
f();
#pragma warning(pop)
Implementation-wise:
- Move `/wd` flag handling from Options.td to actual Driver-level code
- Extract the function mapping cl.exe IDs to warning groups to the
new file clang/lib/Basic/CLWarnings.cpp
- Create a diag::Group enum so that CLWarnings.cpp can refer to
existing groups by ID (and give DllexportExplicitInstantiationDecl
a named group), and add a function to map a diag::Group to the
spelling of it's associated commandline flag
- Call that new function from PragmaWarningHandler
Differential Revision: https://reviews.llvm.org/D110668
On AIX, we relied on LTO to merge the csects for profiling data/counter
sections.
AIX binder now get the namedcsect support to support the merging,
so now we can enable PGO without LTO with the new binder.
Reviewed By: Whitney
Differential Revision: https://reviews.llvm.org/D110671
This matches GCC.
Change the CC1 option to encode the unwind table level (1: needed by exceptions,
2: asynchronous) so that we can support two modes in the future.
This is to build the foundation of a new debug info feature to use only
the base name of template as its debug info name (eg: "t1" instead of
the full "t1<int>"). The intent being that a consumer can still retrieve
all that information from the DW_TAG_template_*_parameters.
So gno-simple-template-names is business as usual/previously ("t1<int>")
=simple is the simplified name ("t1")
=mangled is a special mode to communicate the full information, but
also indicate that the name should be able to be simplified. The data
is encoded as "_STNt1|<int>" which will be matched with an
llvm-dwarfdump --verify feature to deconstruct this name, rebuild the
original name, and then try to rebuild the simple name via the DWARF
tags - then compare the latter and the former to ensure that all the
data necessary to fully rebuild the name is present.
Summary:
Introduce a new frontend flag `-fswift-async-fp={auto|always|never}`
that controls how code generation sets the Swift extended async frame
info bit. There are three possibilities:
* `auto`: which determines how to set the bit based on deployment target, either
statically or dynamically via `swift_async_extendedFramePointerFlags`.
* `always`: default, always set the bit statically, regardless of deployment
target.
* `never`: never set the bit, regardless of deployment target.
Differential Revision: https://reviews.llvm.org/D109451
This patch introduces the flags `-fopenmp-target-debug` and
`-fopenmp-target-debug=` to set the value of a global in the device.
This will be used to enable or disable debugging features statically in
the device runtime library.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D109544
- Make flto an alias of flto=full.
- Make foffload-lto an alias of foffload-lto=full.
- Make flto_EQ_jobserver, flto_EQ_auto aliases of flto=full,
since they are being treated as full lto right now.
- Clean up the code for parseLTOMode and setLTOMode.
- Replace uses of OPT_flto with OPT_flto_EQ since they alias now.
Differential Revision: https://reviews.llvm.org/D108881
Change-Id: I5d867db83a680434fba5c8d85c9a83135d3b81ee
- Make flto an alias of flto=full.
- Make foffload-lto an alias of foffload-lto=full.
- Make flto_EQ_jobserver, flto_EQ_auto aliases of flto=full,
since they are being treated as full lto right now.
- Clean up the code for parseLTOMode and setLTOMode.
- Replace uses of OPT_flto with OPT_flto_EQ since they alias now.
Change-Id: Iea5338c20cb800b43529b20745e92600e2cfd2b1
Earlier BundleEntryID used to be <OffloadKind>-<Triple>-<GPUArch>.
This used to work because the clang-offload-bundler didn't need
GPUArch explicitly for any bundling/unbundling action. With
unbundleArchive it needs GPUArch to ensure compatibility between
device specific code objects. D93525 enforced triples to have
separators for all 4 components irrespective of number of
components, like "amdgcn-amd-amdhsa--". It was required to
to correctly parse a possible 4th environment component or a GPU.
But, this condition is breaking backward compatibility with
archive libraries compiled with compilers older than D93525.
This patch allows triples to have any number of components with
and without extra separator for empty environment field. Thus,
both the following bundle entry IDs are same:
openmp-amdgcn-amd-amdhsa--gfx906
openmp-amdgcn-amd-amdhsa-gfx906
Reviewed By: yaxunl, grokos
Differential Revision: https://reviews.llvm.org/D106809
Now prints the list of known archs. This requires plumbing a Driver
arg through a few functions.
Also add two more convenience insert() overlods to StringMap.
Differential Revision: https://reviews.llvm.org/D109105
The intent of this patch is to add support of -fp-model=[source|double|extended] to allow
the compiler to use a wider type for intermediate floating point calculations. As a side
effect to that, the value of FLT_EVAL_METHOD is changed according to the pragma
float_control.
Unfortunately some issue was uncovered with this change in preprocessing. See details in
https://reviews.llvm.org/D93769 . We are therefore reverting this patch until we find a way
to reconcile the value of FLT_EVAL_METHOD, the pragma and the -E flow.
This reverts commit 66ddac22e2.
This patch implements Clang support for an original OpenMP extension
we have developed to support OpenACC: the `ompx_hold` map type
modifier. The next patch in this series, D106510, implements OpenMP
runtime support.
Consider the following example:
```
#pragma omp target data map(ompx_hold, tofrom: x) // holds onto mapping of x
{
foo(); // might have map(delete: x)
#pragma omp target map(present, alloc: x) // x is guaranteed to be present
printf("%d\n", x);
}
```
The `ompx_hold` map type modifier above specifies that the `target
data` directive holds onto the mapping for `x` throughout the
associated region regardless of any `target exit data` directives
executed during the call to `foo`. Thus, the presence assertion for
`x` at the enclosed `target` construct cannot fail. (As usual, the
standard OpenMP reference count for `x` must also reach zero before
the data is unmapped.)
Justification for inclusion in Clang and LLVM's OpenMP runtime:
* The `ompx_hold` modifier supports OpenACC functionality (structured
reference count) that cannot be achieved in standard OpenMP, as of
5.1.
* The runtime implementation for `ompx_hold` (next patch) will thus be
used by Flang's OpenACC support.
* The Clang implementation for `ompx_hold` (this patch) as well as the
runtime implementation are required for the Clang OpenACC support
being developed as part of the ECP Clacc project, which translates
OpenACC to OpenMP at the directive AST level. These patches are the
first step in upstreaming OpenACC functionality from Clacc.
* The Clang implementation for `ompx_hold` is also used by the tests
in the runtime implementation. That syntactic support makes the
tests more readable than low-level runtime calls can. Moreover,
upstream Flang and Clang do not yet support OpenACC syntax
sufficiently for writing the tests.
* More generally, the Clang implementation enables a clean separation
of concerns between OpenACC and OpenMP development in LLVM. That
is, LLVM's OpenMP developers can discuss, modify, and debug LLVM's
extended OpenMP implementation and test suite without directly
considering OpenACC's language and execution model, which can be
handled by LLVM's OpenACC developers.
* OpenMP users might find the `ompx_hold` modifier useful, as in the
above example.
See new documentation introduced by this patch in `openmp/docs` for
more detail on the functionality of this extension and its
relationship with OpenACC. For example, it explains how the runtime
must support two reference counts, as specified by OpenACC.
Clang recognizes `ompx_hold` unless `-fno-openmp-extensions`, a new
command-line option introduced by this patch, is specified.
Reviewed By: ABataev, jdoerfert, protze.joachim, grokos
Differential Revision: https://reviews.llvm.org/D106509
-fstack-clash-protection was added in Clang commit e67cbac812 but was
enabled only on Linux. Allow it on FreeBSD as well, as it works fine.
Reviewed By: serge-sans-paille
Differential Revision: https://reviews.llvm.org/D108571
Previoulsy debug-info-for-profiling and pseudo-probe-for-profiling are mutual exclusive because they compete the dwarf discrimnator for callsites on the IR. This changes allows to use the two switches together. The side effect is that callsite discriminators will be taken by pseudo probe, while discriminators for other instructions are still available for AutoFDO use. This is less than ideal, however, it still allows us a chance to smoothly transition from AutoFDO to CSSPGO, by collecting both profiles from a CSSPGO binary.
Reviewed By: wenlei, wmi
Differential Revision: https://reviews.llvm.org/D107876
Temporary files created by the offloading device toolchain are not removed
after compilation when using a two-step compilation. The offload-bundler uses a
different filename for the device binary than the `.o` file present in the
Job's input list. This is not listed as a temporary file so it is never
removed. This patch explicitly adds the device binary as a temporary file to
consume it. This fixes PR50336.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D107668
GCC supports multiple forms of -falign-loops=.
-falign-loops= is currently ignored in Clang.
This patch implements the simplest but the most useful form where N is a
power of 2.
The underlying implementation uses a `llvm::TargetOptions` option for now.
Bitcode generation ignores this option.
Differential Revision: https://reviews.llvm.org/D106701
The declaration for the global new function in C++ is generated in the compiler front-end. When examining exception propagation, we found that this is the largest root throw site propagator requiring unwind code to be generated for callers up the stack. Allowing this to be handled immediately with termination stops upward propagation and leads to significantly less landing pads generated. This in turns leads to a performance and .text size win.
With `-fnew-infallible` this annotates the declaration with `throw()` and `__attribute__((returns_nonnull))`. `throw()` allows the compiler to assume exceptions do not propagate out of new and eliminate it as a root throw site. Note that the definition of global new is user-replaceable so users should ensure that the one used follows these semantics.
Measuring internally, we're seeing at 0.5% CPU win in one of our large internal FB workload. Measuring on clang self-build (cd0a1226b5) we get:
thinlto/
"dwarfehprepare.NumCleanupLandingPadsRemaining": 153494,
"dwarfehprepare.NumNoUnwind": 26309,
thinlto_newinfallible/
"dwarfehprepare.NumCleanupLandingPadsRemaining": 143660,
"dwarfehprepare.NumNoUnwind": 28744,
a 1-143660/153494 = 6.4% reduction in landing pads and a 28744/26309 = 9.3% increase in the number of nounwind functions.
Testing:
ninja check-all
new test case to make sure these attributes are added correctly to global new.
Reviewed By: urnathan
Differential Revision: https://reviews.llvm.org/D105225
With this patch, OpenMP on AMDGCN will use the math functions
provided by ROCm ocml library. Linking device code to the ocml will be
done in the next patch.
Reviewed By: JonChesterfield, jdoerfert, scchan
Differential Revision: https://reviews.llvm.org/D104904
Definition of `__cpp_threadsafe_static_init` macro is controlled by
language option Opts.ThreadsafeStatics. This patch sets language
option to false by default in OpenCL mode, resulting in macro
`__cpp_threadsafe_static_init` being undefined. Default value can be
overridden using command line option -fthreadsafe-statics.
Change is supposed to address portability because not all OpenCL
vendors support thread safe implementation of static initialization.
Fixes llvm.org/PR48012
Differential Revision: https://reviews.llvm.org/D107163
With this patch, OpenMP on AMDGCN will use the math functions
provided by ROCm ocml library. Linking device code to the ocml will be
done in the next patch.
Reviewed By: JonChesterfield, jdoerfert, scchan
Differential Revision: https://reviews.llvm.org/D104904
The Intel compiler ICC supports the option "-fp-model=(source|double|extended)"
which causes the compiler to use a wider type for intermediate floating point
calculations. Also supported is a way to embed this effect in the source
program with #pragma float_control(source|double|extended).
This patch extends pragma float_control syntax, and also adds support
for a new floating point option "-ffp-eval-method=(source|double|extended)".
source: intermediate results use source precision
double: intermediate results use double precision
extended: intermediate results use extended precision
Reviewed By: Aaron Ballman
Differential Revision: https://reviews.llvm.org/D93769
Change the ffp-model=precise to enables -ffp-contract=on (previously
-ffp-model=precise enabled -ffp-contract=fast). This is a follow-up
to Andy Kaylor's comments in the llvm-dev discussion "Floating Point
semantic modes". From the same email thread, I put Andy's distillation
of floating point options and floating point modes into UsersManual.rst
Also fixes bugs.llvm.org/show_bug.cgi?id=50222
I had to revert this a few times because of failures on the x86-64
buildbot but I think we finally have that fixed by LNT/79f2b03c51.
Reviewed By: rjmccall, andrew.kaylor
Differential Revision: https://reviews.llvm.org/D74436
Moving `InputInfo.h` from `lib/Driver/` into `include/Driver` to be able to expose it in an API consumed from outside of `clangDriver`.
Reviewed By: dexonsmith
Differential Revision: https://reviews.llvm.org/D106787
Constructor homing reduces the amount of class type info that is emitted
by emitting conmplete type info for a class only when a constructor for
that class is emitted.
This will mainly reduce the amount of duplicate debug info in object
files. In Chrome enabling ctor homing decreased total build directory sizes
by about 30%.
It's also expected that some class types (such as unused classes)
will no longer be emitted in the debug info. This is fine, since we wouldn't
expect to need these types when debugging.
In some cases (e.g. libc++, https://reviews.llvm.org/D98750), classes
are used without calling the constructor. Since this is technically
undefined behavior, enabling constructor homing should be fine.
However Clang now has an attribute
`__attribute__((standalone_debug))` that can be used on classes to
ignore ctor homing.
Bug: https://bugs.llvm.org/show_bug.cgi?id=46537
Differential Revision: https://reviews.llvm.org/D106084
This patch adds the -fminimize-whitespace with the following effects:
* If combined with -E, remove as much non-line-breaking whitespace as
possible.
* If combined with -E -P, removes as much whitespace as possible,
including line-breaks.
The motivation is to reduce the amount of insignificant changes in the
preprocessed output with source files where only whitespace has been
changed (add/remove comments, clang-format, etc.) which is in particular
useful with ccache.
A patch for ccache for using this flag has been proposed to ccache as well:
https://github.com/ccache/ccache/pull/815, which will use
-fnormalize-whitespace when clang-13 has been detected, and additionally
uses -P in "unify_mode". ccache already had a unify_mode in an older
version which was removed because of problems that using the
preprocessor itself does not have (such that the custom tokenizer did
not recognize C++11 raw strings).
This patch slightly reorganizes which part is responsible for adding
newlines that are required for semantics. It is now either
startNewLineIfNeeded() or MoveToLine() but never both; this avoids the
ShouldUpdateCurrentLine workaround and avoids redundant lines being
inserted in some cases. It also fixes a mandatory newline not inserted
after a _Pragma("...") that is expanded into a #pragma.
Reviewed By: aaron.ballman
Differential Revision: https://reviews.llvm.org/D104601
Alexandre Ganea