Previously an opt-in flag `-fopenmp-new-driver` was used to enable the
new offloading driver. After passing tests for a few months it should be
sufficiently mature to flip the switch and make it the default. The new
offloading driver is now enabled if there is OpenMP and OpenMP
offloading present and the new `-fno-openmp-new-driver` is not present.
The new offloading driver has three main benefits over the old method:
- Static library support
- Device-side LTO
- Unified clang driver stages
Depends on D122683
Differential Revision: https://reviews.llvm.org/D122831
The target profile option(/T) decide the shader model when compile hlsl.
The format is shaderKind_major_minor like ps_6_1.
The shader model is saved as llvm::Triple is clang/llvm like
dxil-unknown-shadermodel6.1-hull.
The main job to support the option is translating ps_6_1 into
shadermodel6.1-pixel.
That is done inside tryParseProfile at HLSL.cpp.
To integrate the option into clang Driver, a new DriverMode DxcMode is
created. When DxcMode is enabled, OSType for TargetTriple will be
forced into Triple::ShaderModel. And new ToolChain HLSLToolChain will
be created when OSType is Triple::ShaderModel.
In HLSLToolChain, ComputeEffectiveClangTriple is overridden to call
tryParseProfile when targetProfile option is set.
To make test work, Fo option is added and .hlsl is added for active
-xhlsl.
Reviewed By: beanz
Differential Revision: https://reviews.llvm.org/D122865
Patch by: Xiang Li <python3kgae@outlook.com>
This adds a PS5-specific ToolChain subclass, which defines some basic
PS5 driver behavior. Future patches will add more target-specific
driver behavior.
Add CSKY target toolchains to support csky in linux and elf environment.
It can leverage the basic universal Linux toolchain for linux environment, and only add some compile or link parameters.
For elf environment, add a CSKYToolChain to support compile and link.
Also add some parameters into basic codebase of clang driver.
Differential Revision: https://reviews.llvm.org/D121445
--overlay-platform-toolchain inserts a whole new toolchain path with
higher priority than system default, which could be achieved by
composing smaller options. We need to figure out alternative solution
and what is missing among these basic options.
In some cases, we need to set alternative toolchain path other than the
default with system (headers, libraries, dynamic linker prefix, ld path,
etc.), e.g., to pick up newer components, but keep sysroot at the same
time (to pick up extra packages).
This change introduces a new option --overlay-platform-toolchain to set
up such alternative toolchain path.
Reviewed By: hubert.reinterpretcast
Differential Revision: https://reviews.llvm.org/D121992
clang -extract-api should accept multiple headers and forward them to a
single CC1 instance. This change introduces a new ExtractAPIJobAction.
Currently API Extraction is done during the Precompile phase as this is
the current phase that matches the requirements the most. Adding a new
phase would need to change some logic in how phases are scheduled. If
the headers scheduled for API extraction are of different types the
driver emits a diagnostic.
Differential Revision: https://reviews.llvm.org/D121936
This path refactors the new driver to be less dependent on OpenMP. This
is done in preparation for the new driver to be able to handle other
offloading kinds and compile them together.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D120934
When both CUDA or HIP programs and C++ programs are passed
to clang driver without -c, C++ programs are treated as CUDA
or HIP program, which is incorrect.
This is because action builder sets the offloading kind of input
job actions to the linking action to be the union of offloading
kind of the input job actions, i.e. if there is one HIP or CUDA
input to the linker, then all the input to the linker is marked
as HIP or CUDA.
To fix this issue, the offload action builder tracks the originating
input argument of each host action, which allows it to determine
the active offload kind of each host action. Then the offload
kind of each input action to the linker can be determined
individually.
Reviewed by: Artem Belevich
Differential Revision: https://reviews.llvm.org/D120911
When both HIP and C++ programs are input files to clang
with -c, clang treats C++ programs as HIP programs,
which is incorrect.
This is due to action builder does not set correct
offloading kind for job actions for C++ programs.
Reviewed by: Artem Belevich
Differential Revision: https://reviews.llvm.org/D120910
When we are creating jobs for the new driver we first check the cache to
see if the job was already created as a part of the offloading
toolchain. This would sometimes fail if the bound architecture was set
for the host during offloading. We want to ingore this because it is not
relevant for looking up host actions. Previously it was set on some
machines and would cause the cache lookup to fail.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D118858
This patch implements the fist support for handling LTO in the
offloading pipeline. The flag `-foffload-lto` is used to control if
bitcode is embedded into the device. If bitcode is found in the device,
the extracted files will be sent to the LTO pipeline to be linked and
sent to the backend. This implementation does not separately link the
device bitcode libraries yet.
Depends on D116675
Differential Revision: https://reviews.llvm.org/D116975
This patch introduces a linker wrapper tool that allows us to preprocess
files before they are sent to the linker. This adds a dummy action and
job to the driver stage that builds the linker command as usual and then
replaces the command line with the wrapper tool.
Depends on D116543
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D116544
This patch introduces the `-fopenmp-new-driver` option which instructs
the compiler to use a new driver scheme for producing offloading code.
In this scheme we create a complete offloading object file and then pass
it as input to the host compilation phase. This will allow us to embed
the object code in the backend phase.
This is the start of a series of commits to rework the OpenMP offloading driver
pipeline. The goal of this is to simplify the steps required for creating an
offloading program. This patch changes the driver's configuration to simply pass
the device file back to the host as an input so it can be embedded as an LLVM IR
global during the backend, then simply passes that object file to the linker.
This driver implementation will currently create the following phases,
```
$ clang input.c -fopenmp -fopenmp-targets=nvptx64 -fopenmp-new-driver -ccc-print-phases
+- 0: input, "input.c", c, (host-openmp)
+- 1: preprocessor, {0}, cpp-output, (host-openmp)
+- 2: compiler, {1}, ir, (host-openmp)
| | +- 3: input, "input.c", c, (device-openmp)
| | +- 4: preprocessor, {3}, cpp-output, (device-openmp)
| |- 5: compiler, {4}, ir, (device-openmp)
| +- 6: offload, "host-openmp (x86_64-unknown-linux-gnu)" {2}, "device-openmp (nvptx64)" {5}, ir
| +- 7: backend, {6}, assembler, (device-openmp)
|- 8: assembler, {7}, object, (device-openmp)
+- 9: offload, "host-openmp (x86_64-unknown-linux-gnu)" {2}, "device-openmp (nvptx64)" {8}, ir
+- 10: backend, {9}, assembler, (host-openmp)
+- 11: assembler, {10}, object, (host-openmp)
12: clang-linker-wrapper, {11}, image, (host-openmp)
```
Which will map to the following bindings
```
# "x86_64-unknown-linux-gnu" - "clang", inputs: ["input.c"], output: "/tmp/input-bae62e.bc"
# "nvptx64" - "clang", inputs: ["input.c", "/tmp/input-bae62e.bc"], output: "/tmp/input-76784e.s"
# "nvptx64" - "NVPTX::Assembler", inputs: ["/tmp/input-76784e.s"], output: "/tmp/input-8f29db.o"
# "x86_64-unknown-linux-gnu" - "clang", inputs: ["/tmp/input-bae62e.bc", "/tmp/input-8f29db.o"], output: "/tmp/input-545450.o"
# "x86_64-unknown-linux-gnu" - "Offload::Linker", inputs: ["/tmp/input-545450.o"], output: "a.out"
```
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D116541
This patch builds on the change in D117634 that expanded the short
triples when passed in by the user. This patch adds the same
functionality for the `-Xopenmp-target=` flag. Previously it was
unintuitive that passing `-fopenmp-targets=nvptx64
-Xopenmp-target=nvptx64 <arg>` would not forward the arg because the
triples did not match on account of `nvptx64` being expanded to
`nvptx64-nvidia-cuda`.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D118495
The --offload option was added in D110622 to "override the default
device target". When it landed it supported only HIP. This patch
extends that option to support SPIR-V targets for CUDA.
Reviewed By: tra
Differential Revision: https://reviews.llvm.org/D117137
The OpenMP offloading libraries are built with fixed triples and linked
in during compile time. This would cause un-helpful errors if the user
passed in the wrong expansion of the triple used for the bitcode
library. because we only support these triples for OpenMP offloading we
can normalize them to the full verion used in the bitcode library.
Reviewed By: jdoerfert, JonChesterfield
Differential Revision: https://reviews.llvm.org/D117634
Add support of linking files compiled into SPIR-V objects
using spirv-link.
Command line inteface examples:
clang --target=spirv64 test1.cl test2.cl
clang --target=spirv64 test1.cl -o test1.o
clang --target=spirv64 test1.o test2.cl -o test_app.out
This works independently from the SPIR-V generation method
(via an external tool or an internal backend) and applies
to either approach that is being used.
Differential Revision: https://reviews.llvm.org/D116266
When passing a set of flags to configure defaults for a specific
target (similar to the cmake settings `CLANG_DEFAULT_RTLIB`,
`CLANG_DEFAULT_UNWINDLIB`, `CLANG_DEFAULT_CXX_STDLIB` and
`CLANG_DEFAULT_LINKER`, but without hardcoding them in the binary),
some of the flags may cause warnings (e.g. `-stdlib=` when compiling C
code). Allow requesting selectively ignoring unused arguments among
some of the arguments on the command line, without needing to resort
to `-Qunused-arguments` or `-Wno-unused-command-line-argument`.
Fix up the existing diagnostics.c testcase. It was added in
response to PR12181 to fix handling of
`-Werror=unused-command-line-argument`, but the command line option
in the test (`-fzyzzybalubah`) now triggers "error: unknown argument"
instead of the intended warning. Change it into a linker input
(`-lfoo`) which triggers the intended diagnostic. Extend the
existing test case to check more cases and make sure that it keeps
testing the intended case.
Add testing of the new option to this existing test.
Differential Revision: https://reviews.llvm.org/D116503
Currently when -fgpu-rdc is specified, HIP toolchain always does host linking even
if --cuda-device-only is specified.
This patch fixes that. Only device linking is performed when --cuda-device-only
is specified.
Reviewed by: Artem Belevich
Differential Revision: https://reviews.llvm.org/D116840
Clang searches for runtimes (e.g. libclang_rt*) first in a
subdirectory named for the target triple (corresponding to
LLVM_ENABLE_PER_TARGET_RUNTIME_DIR=ON), then if it's not found uses
.../lib/<os>/libclang_rt* with a suffix corresponding to the arch and
environment name.
Android triples optionally include an API level indicating the minimum
Android version to be run on
(e.g. aarch64-unknown-linux-android21). When compiler-rt is built with
LLVM_ENABLE_PER_TARGET_RUNTIME_DIR=ON this API level is part of the
output path.
Linking code built for a later API level against a runtime built for
an earlier one is safe. In projects with several API level targets
this is desireable to avoid re-building the same runtimes many
times. This is difficult with the current runtime search method: if
the API levels don't exactly match Clang gives up on the per-target
runtime directory path.
To enable this more simply, this change tries target triple without
the API level before falling back on the old layout.
Another option would be to try every API level in the triple,
e.g. check aarch-64-unknown-linux-android21, then ...20, then ...19,
etc.
Differential Revision: https://reviews.llvm.org/D115049
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>
This patch enables SPIR-V binary emission for HIP device code via the
HIPSPV tool chain.
‘--offload’ option, which is envisioned in [1], is added for specifying
offload targets. This option is used to override default device target
(amdgcn-amd-amdhsa) for HIP compilation for emitting device code as
SPIR-V binary. The option is handled in getHIPOffloadTargetTriple().
getOffloadingDeviceToolChain() function (based on the design in the
SYCL repository) is added to select HIPSPVToolChain when HIP offload
target is ‘spirv64’.
The HIPActionBuilder is modified to produce LLVM IR at the backend
phase. HIPSPV tool chain expects to receive HIP device code as LLVM
IR so it can run external LLVM passes over them. HIPSPV TC is also
responsible for emitting the SPIR-V binary.
A Cuda GPU architecture ‘generic’ is added. The name is picked from
the LLVM SPIR-V Backend. In the HIPSPV code path the architecture
name is inserted to the bundle entry ID as target ID. Target ID is
expected to be always present so a component in the target triple
is not mistaken as target ID.
Tests are added for checking the HIPSPV tool chain.
[1]: https://lists.llvm.org/pipermail/cfe-dev/2020-December/067362.html
Patch by: Henry Linjamäki
Reviewed by: Yaxun Liu, Artem Belevich, Alexey Bader
Differential Revision: https://reviews.llvm.org/D110622
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
This enables Intel intrinsics support on FreeBSD.
Thanks to @pkubaj who noticed this feature was missing
Reviewed By: jsji
Differential Revision: https://reviews.llvm.org/D113451
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
Use the new sys::path::is_style_posix() and is_style_windows() in a few
places that need to detect the system's native path style.
In llvm/lib/Support/Path.cpp, this patch removes most uses of the
private `real_style()`, where is_style_posix() and is_style_windows()
are just a little tidier.
Elsewhere, this removes `_WIN32` macro checks. Added a FIXME to a
FileManagerTest that seemed fishy, but maintained the existing
behaviour.
Differential Revision: https://reviews.llvm.org/D112289
Previously if you passed a `-Wl,-foo` _before_ the source filename, the
first `InputInfos`, which is used for the base input name would be an
`InputArg` kind, which would never have a base input name. Now we use
that by default, but pick the first `InputInfo` that is of kind
`Filename` to get the name from if there is one.
Differential Revision: https://reviews.llvm.org/D112767
For x86, most contempory mingw toolchains use i686 as 32 bit
x86 arch target.
As long as the target triple is set to the right form, this works
fine, either as the compiler's default target, or via e.g.
a triple prefix like i686-w64-mingw32-clang.
However, if the unprefixed toolchain targets x86_64, but the user
tries to switch it to target 32 bit by adding the -m32 option, the
computeTargetTriple function in Clang, together with
Triple::get32BitArchVariant, sets the arch to i386. This causes
the right sysroot to not be found.
When targeting an arch where there are potential spelling ambiguities
with respect to the sysroots (i386 and arm), check if the driver can
find a sysroot with the arch name - if not, try a couple other
candidates.
Differential Revision: https://reviews.llvm.org/D111952
When building a multiarch MachO binary, previously the intermediate
output file names would contain random characters. On macOS this
filename, since it's used when linking, ended up being used as a
stable-ish identifier for the adhoc codesignature of the binary, leading
to non-reproducible binaries. This change uses the architecture, when
available, to create a stable, but unique, basename for the file.
Differential Revision: https://reviews.llvm.org/D111269
This moves the registry higher in the LLVM library dependency stack.
Every client of the target registry needs to link against MC anyway to
actually use the target, so we might as well move this out of Support.
This allows us to ensure that Support doesn't have includes from MC/*.
Differential Revision: https://reviews.llvm.org/D111454
When clang crashes, it writes a standalone source file and shell script
to reproduce the crash.
The Driver used to set `Mode = CPPMode` in generateCompilationDiagnostics()
to force preprocessing mode. This has the side effect of making
IsCLMode() return false, which in turn meant Clang::AddClangCLArgs()
didn't get called when creating the standalone source file, which meant
the stand-alone file was preprocessed with the gcc driver's defaults
In particular, exceptions default to on with the gcc driver, but to
off with the cl driver. The .sh script did use the original command
line, so in the reproducer for a clang-cl crash, the standalone source
file could contain exception-using code after preprocessing that the
compiler invocation in the shell script would then complain about.
This patch removes the `Mode = CPPMode;` line and instead additionally
checks for `CCGenDiagnostics` in most places that check `CCCIsCPP().
This also matches the strategy Clang::ConstructJob() uses to add
-frewrite-includes for creating the standalone source file for a crash
report.
Fixes PR52007.
Differential Revision: https://reviews.llvm.org/D110783
Call Driver::getFinalPhase() instead of duplicating it.
https://reviews.llvm.org/D65993 added the duplication, then
02e35832c3 maded it more obviously a copy of getFinalPhase().
The only difference is that getCompilationPhases() used to use
LastPhase / IfsMerge where getFinalPhase() used Link. Adapt
getFinalPhase() to return IfsMerge when needed.
No intentional behavior change.
Differential Revision: https://reviews.llvm.org/D110770
We used to put the canonical spelling of flags after alias processing
on that line. For clang-cl in particular, that meant that we put flags
on that line that the clang-cl driver doesn't even accept, and the
"Driver args:" line wasn't usable.
Differential Revision: https://reviews.llvm.org/D110458
Diagnose -fopenmp-targets for HIP programs since
dual HIP and OpenMP offloading in the same compilation
is currently not supported by HIP toolchain.
Reviewed by: Artem Belevich
Differential Revision: https://reviews.llvm.org/D109718
When nonexistent linker inputs are passed to the driver, the linker
now errors out, instead of the compiler. If the linker does not run,
clang now emits a "warning: linker input unused" instead of an error
for nonexistent files.
The motivation for this change is that I noticed that
`clang-cl /winsysroot sysroot main.cc ole32.lib` emitted a
"ole32.lib not found" error, even though the linker finds it just fine when
I run `clang-cl /winsysroot sysroot main.cc /link ole32.lib`.
The same problem occurs if running `clang-cl main.cc ole32.lib` in a
non-MSVC shell.
The problem is that DiagnoseInputExistence() only looked for libs in %LIB%,
but MSVCToolChain uses much more involved techniques.
For this particular problem, we could make DiagnoseInputExistence() ask
the toolchain to see if it can find a .lib file, but in general the
driver can't know what the linker will do to find files, so it shouldn't
try. For example, if we implement PR24616, lld-link will look in the
registry to determine a good default for %LIB% if it isn't set.
This is less or a problem for the gcc driver, since .a paths there are
either passed via -l flags (which honor -L), or via a qualified path
(that doesn't honor -L) -- but for example ld.lld's --chroot flag
can also trigger this problem. Without this patch,
`clang -fuse-ld=lld -Wl,--chroot,some/dir /file.o` will complain that
`/file.o` doesn't exist, even though
`clang -fuse-ld=lld -Wl,--chroot,some/dir -Wl,/file.o` succeeds just fine.
This implements rnk's suggestion on the old bug PR27234.
Differential Revision: https://reviews.llvm.org/D109624
Joseph Huber