[llvm] 0915825 - [ThinLTO] Allow usage of all hardware threads in the system
Hans Wennborg via llvm-commits
llvm-commits at lists.llvm.org
Tue Apr 7 01:00:23 PDT 2020
I've dropped the number of threads to 100 in
717a460867633008520713614018f3c37503a5ca to unbreak the tests for now.
On Mon, Apr 6, 2020 at 7:19 PM Alexandre Ganea
<alexandre.ganea at ubisoft.com> wrote:
>
> Yeah, maybe it isn't wise to use "/opt:lldltojobs=1000" in a test. I wanted to make that $(($(nproc)*2)) to see if extra threads were equally dispatched on all CPU sockets.
> I'll try to repro first, and probably reduce that number somehow.
> Thanks for the heads up!
>
> -----Message d'origine-----
> De : Hans Wennborg <hans at chromium.org>
> Envoyé : April 6, 2020 12:52 PM
> À : Alexandre Ganea <alexandre.ganea at ubisoft.com>; Alexandre Ganea <llvmlistbot at llvm.org>
> Cc : llvm-commits <llvm-commits at lists.llvm.org>
> Objet : Re: [llvm] 0915825 - [ThinLTO] Allow usage of all hardware threads in the system
>
> This broke some LLD ThinLTO tests in 32-bit Windows builds. Can you take a look?
>
> I wonder if it's simply running out of the 32-bit address space when running so many threads (I have a lot of cores on my machine)..
>
> See the example output below:
>
> C:\src\llvm.monorepo>ninja -C build.32 lld opt llvm-nm && python build.32\bin\llvm-lit.py -v lld\test\COFF\thinlto.ll
> ninja: Entering directory `build.32'
> ninja: no work to do.
> llvm-lit.py: C:/src/llvm.monorepo/llvm\utils\lit\lit\llvm\config.py:342:
> note: using ld.lld: c:\src\llvm.monorepo\build.32\bin\ld.lld.exe
> llvm-lit.py: C:/src/llvm.monorepo/llvm\utils\lit\lit\llvm\config.py:342:
> note: using lld-link: c:\src\llvm.monorepo\build.32\bin\lld-link.exe
> llvm-lit.py: C:/src/llvm.monorepo/llvm\utils\lit\lit\llvm\config.py:342:
> note: using ld64.lld: c:\src\llvm.monorepo\build.32\bin\ld64.lld.exe
> llvm-lit.py: C:/src/llvm.monorepo/llvm\utils\lit\lit\llvm\config.py:342:
> note: using wasm-ld: c:\src\llvm.monorepo\build.32\bin\wasm-ld.exe
> -- Testing: 1 tests, 1 workers --
> FAIL: lld :: COFF/thinlto.ll (1 of 1)
> ******************** TEST 'lld :: COFF/thinlto.ll' FAILED ********************
> Script:
> --
> : 'RUN: at line 2'; rm -fr
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto
> : 'RUN: at line 3'; mkdir
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto
> : 'RUN: at line 4'; c:\src\llvm.monorepo\build.32\bin\opt.exe
> -thinlto-bc -o C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.obj
> C:\src\llvm.monorepo\lld\test\COFF\thinlto.ll
> : 'RUN: at line 5'; c:\src\llvm.monorepo\build.32\bin\opt.exe
> -thinlto-bc -o C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/foo.obj
> C:\src\llvm.monorepo\lld\test\COFF/Inputs/lto-dep.ll
> : 'RUN: at line 6'; c:\src\llvm.monorepo\build.32\bin\lld-link.exe
> /lldsavetemps /out:C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe
> /entry:main /subsystem:console
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.obj
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/foo.obj
> : 'RUN: at line 7'; c:\src\llvm.monorepo\build.32\bin\llvm-nm.exe
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe1.lto.obj
> | c:\src\llvm.monorepo\build.32\bin\filecheck.exe
> C:\src\llvm.monorepo\lld\test\COFF\thinlto.ll
> : 'RUN: at line 10'; c:\src\llvm.monorepo\build.32\bin\lld-link.exe
> /lldsavetemps /out:C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe
> /entry:main /subsystem:console
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.obj
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/foo.obj
> /opt:lldltojobs=1
> : 'RUN: at line 11'; c:\src\llvm.monorepo\build.32\bin\llvm-nm.exe
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe1.lto.obj
> | c:\src\llvm.monorepo\build.32\bin\filecheck.exe
> C:\src\llvm.monorepo\lld\test\COFF\thinlto.ll
> : 'RUN: at line 12'; c:\src\llvm.monorepo\build.32\bin\lld-link.exe
> /lldsavetemps /out:C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe
> /entry:main /subsystem:console
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.obj
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/foo.obj
> /opt:lldltojobs=all
> : 'RUN: at line 13'; c:\src\llvm.monorepo\build.32\bin\llvm-nm.exe
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe1.lto.obj
> | c:\src\llvm.monorepo\build.32\bin\filecheck.exe
> C:\src\llvm.monorepo\lld\test\COFF\thinlto.ll
> : 'RUN: at line 14'; c:\src\llvm.monorepo\build.32\bin\lld-link.exe
> /lldsavetemps /out:C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe
> /entry:main /subsystem:console
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.obj
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/foo.obj
> /opt:lldltojobs=1000
> : 'RUN: at line 15'; c:\src\llvm.monorepo\build.32\bin\llvm-nm.exe
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe1.lto.obj
> | c:\src\llvm.monorepo\build.32\bin\filecheck.exe
> C:\src\llvm.monorepo\lld\test\COFF\thinlto.ll
> : 'RUN: at line 16'; not
> c:\src\llvm.monorepo\build.32\bin\lld-link.exe /lldsavetemps /out:C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe
> /entry:main /subsystem:console
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.obj
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/foo.obj
> /opt:lldltojobs=foo 2>&1 |
> c:\src\llvm.monorepo\build.32\bin\filecheck.exe
> C:\src\llvm.monorepo\lld\test\COFF\thinlto.ll --check-prefix=BAD-JOBS
> : 'RUN: at line 21'; c:\src\llvm.monorepo\build.32\bin\lld-link.exe
> /lib C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/foo.obj
> /out:C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output\thinlto.ll.tmp.lib
> : 'RUN: at line 22'; c:\src\llvm.monorepo\build.32\bin\lld-link.exe
> /lldsavetemps /out:C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output\thinlto.ll.tmp.exe
> /entry:main /subsystem:console
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.obj
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output\thinlto.ll.tmp.lib
> --
> Exit Code: -529697949
>
> Command Output (stdout):
> --
> $ ":" "RUN: at line 2"
> $ "rm" "-fr" "C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto"
> $ ":" "RUN: at line 3"
> $ "mkdir" "C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto"
> $ ":" "RUN: at line 4"
> $ "c:\src\llvm.monorepo\build.32\bin\opt.exe" "-thinlto-bc" "-o"
> "C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.obj"
> "C:\src\llvm.monorepo\lld\test\COFF\thinlto.ll"
> $ ":" "RUN: at line 5"
> $ "c:\src\llvm.monorepo\build.32\bin\opt.exe" "-thinlto-bc" "-o"
> "C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/foo.obj"
> "C:\src\llvm.monorepo\lld\test\COFF/Inputs/lto-dep.ll"
> $ ":" "RUN: at line 6"
> $ "c:\src\llvm.monorepo\build.32\bin\lld-link.exe" "/lldsavetemps"
> "/out:C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe"
> "/entry:main" "/subsystem:console"
> "C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.obj"
> "C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/foo.obj"
> $ ":" "RUN: at line 7"
> $ "c:\src\llvm.monorepo\build.32\bin\llvm-nm.exe"
> "C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe1.lto.obj"
> $ "c:\src\llvm.monorepo\build.32\bin\filecheck.exe"
> "C:\src\llvm.monorepo\lld\test\COFF\thinlto.ll"
> $ ":" "RUN: at line 10"
> $ "c:\src\llvm.monorepo\build.32\bin\lld-link.exe" "/lldsavetemps"
> "/out:C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe"
> "/entry:main" "/subsystem:console"
> "C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.obj"
> "C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/foo.obj"
> "/opt:lldltojobs=1"
> $ ":" "RUN: at line 11"
> $ "c:\src\llvm.monorepo\build.32\bin\llvm-nm.exe"
> "C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe1.lto.obj"
> $ "c:\src\llvm.monorepo\build.32\bin\filecheck.exe"
> "C:\src\llvm.monorepo\lld\test\COFF\thinlto.ll"
> $ ":" "RUN: at line 12"
> $ "c:\src\llvm.monorepo\build.32\bin\lld-link.exe" "/lldsavetemps"
> "/out:C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe"
> "/entry:main" "/subsystem:console"
> "C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.obj"
> "C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/foo.obj"
> "/opt:lldltojobs=all"
> $ ":" "RUN: at line 13"
> $ "c:\src\llvm.monorepo\build.32\bin\llvm-nm.exe"
> "C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe1.lto.obj"
> $ "c:\src\llvm.monorepo\build.32\bin\filecheck.exe"
> "C:\src\llvm.monorepo\lld\test\COFF\thinlto.ll"
> $ ":" "RUN: at line 14"
> $ "c:\src\llvm.monorepo\build.32\bin\lld-link.exe" "/lldsavetemps"
> "/out:C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe"
> "/entry:main" "/subsystem:console"
> "C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.obj"
> "C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/foo.obj"
> "/opt:lldltojobs=1000"
> # command stderr:
> PLEASE submit a bug report to https://bugs.llvm.org/ and include the crash backtrace.
> Stack dump:
> 0. Program arguments:
> c:\src\llvm.monorepo\build.32\bin\lld-link.exe /lldsavetemps /out:C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.exe
> /entry:main /subsystem:console
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/main.obj
> C:\src\llvm.monorepo\build.32\tools\lld\test\COFF\Output/thinlto/foo.obj
> /opt:lldltojobs=1000
> 0x73F647B2 (0xE06D7363 0x00000001 0x00000003 0x04FEE500),
> RaiseException() + 0x62 bytes(s)
> 0x72CC7A36 (0x04FEE524 0x6AA43570 0x004ED908 0x004B7A50),
> _CxxThrowException() + 0x66 bytes(s)
> 0x6AA183F6 (0x00000006 0x004B7A40 0x04FEE680 0x004B7A40),
> ?_Throw_Cpp_error at std@@YAXH at Z() + 0x36 bytes(s)
> 0x02BA8A79 (0x000003E8 0x00000001 0x004B7A40 0x04FEE680)
> 0x01E86C79 (0x00000188 0x00000190 0x0046035C 0x04FEE664)
> 0x7740CFB7 (0x04FEE680 0x00000000 0x00000188 0x00000000),
> RtlAllocateHeap() + 0x307 bytes(s)
> 0x741CF1F6 (0x00000188 0x004B6DD8 0x01E85E3D 0x00000000),
> _malloc_base() + 0x26 bytes(s)
> 0x02B83056 (0x00000000 0x00000000 0x004935D8 0x02BF3280) 0x01E85E3D (0x00000000 0x00000000 0x04FEE6FC 0x02B83077)
> 0x741CDAE8 (0x00000000 0x04FEE6D8 0x01E922FE 0x00000000), free() + 0x18 bytes(s)
> 0x02B83077 (0x00000000 0x00000000 0x00000000 0x004935D8) 0x01E922FE (0x04FEE760 0x004B6DB4 0x004B69B0 0x004B6DD8)
>
> error: command failed with exit status: 0xe06d7363
>
> --
>
> On Fri, Mar 27, 2020 at 3:21 PM Alexandre Ganea via llvm-commits <llvm-commits at lists.llvm.org> wrote:
> >
> >
> > Author: Alexandre Ganea
> > Date: 2020-03-27T10:20:58-04:00
> > New Revision: 09158252f777c2e2f06a86b154c44abcbcf9bb74
> >
> > URL:
> > https://github.com/llvm/llvm-project/commit/09158252f777c2e2f06a86b154
> > c44abcbcf9bb74
> > DIFF:
> > https://github.com/llvm/llvm-project/commit/09158252f777c2e2f06a86b154
> > c44abcbcf9bb74.diff
> >
> > LOG: [ThinLTO] Allow usage of all hardware threads in the system
> >
> > Before this patch, it wasn't possible to extend the ThinLTO threads to all SMT/CMT threads in the system. Only one thread per core was allowed, instructed by usage of llvm::heavyweight_hardware_concurrency() in the ThinLTO code. Any number passed to the LLD flag /opt:lldltojobs=..., or any other ThinLTO-specific flag, was previously interpreted in the context of llvm::heavyweight_hardware_concurrency(), which means SMT disabled.
> >
> > One can now say in LLD:
> > /opt:lldltojobs=0 -- Use one std::thread / hardware core in the system (no SMT). Default value if flag not specified.
> > /opt:lldltojobs=N -- Limit usage to N threads, regardless of usage of heavyweight_hardware_concurrency().
> > /opt:lldltojobs=all -- Use all hardware threads in the system. Equivalent to /opt:lldltojobs=$(nproc) on Linux and /opt:lldltojobs=%NUMBER_OF_PROCESSORS% on Windows. When an affinity mask is set for the process, threads will be created only for the cores selected by the mask.
> >
> > When N > number-of-hardware-threads-in-the-system, the threads in the thread pool will be dispatched equally on all CPU sockets (tested only on Windows).
> > When N <= number-of-hardware-threads-on-a-CPU-socket, the threads will remain on the CPU socket where the process started (only on Windows).
> >
> > Differential Revision: https://reviews.llvm.org/D75153
> >
> > Added:
> >
> >
> > Modified:
> > clang/lib/Driver/ToolChains/CommonArgs.cpp
> > clang/lib/Driver/ToolChains/CommonArgs.h
> > clang/lib/Driver/ToolChains/Darwin.cpp
> > lld/COFF/Config.h
> > lld/COFF/Driver.cpp
> > lld/COFF/LTO.cpp
> > lld/ELF/Config.h
> > lld/ELF/Driver.cpp
> > lld/ELF/LTO.cpp
> > lld/test/COFF/thinlto.ll
> > lld/test/ELF/basic.s
> > lld/test/ELF/lto/thinlto.ll
> > lld/test/wasm/lto/thinlto.ll
> > lld/wasm/Config.h
> > lld/wasm/Driver.cpp
> > lld/wasm/LTO.cpp
> > llvm/include/llvm/LTO/LTO.h
> > llvm/include/llvm/Support/Threading.h
> > llvm/lib/LTO/LTO.cpp
> > llvm/lib/Support/Threading.cpp
> > llvm/lib/Support/Unix/Threading.inc
> > llvm/lib/Support/Windows/Threading.inc
> > llvm/tools/gold/gold-plugin.cpp
> > llvm/tools/llvm-lto2/llvm-lto2.cpp
> >
> > Removed:
> >
> >
> >
> > ######################################################################
> > ########## diff --git a/clang/lib/Driver/ToolChains/CommonArgs.cpp
> > b/clang/lib/Driver/ToolChains/CommonArgs.cpp
> > index 51b2e5f3c34d..303ced52454c 100644
> > --- a/clang/lib/Driver/ToolChains/CommonArgs.cpp
> > +++ b/clang/lib/Driver/ToolChains/CommonArgs.cpp
> > @@ -50,6 +50,7 @@
> > #include "llvm/Support/Program.h"
> > #include "llvm/Support/ScopedPrinter.h"
> > #include "llvm/Support/TargetParser.h"
> > +#include "llvm/Support/Threading.h"
> > #include "llvm/Support/VirtualFileSystem.h"
> > #include "llvm/Support/YAMLParser.h"
> >
> > @@ -338,14 +339,14 @@ std::string tools::getCPUName(const ArgList &Args, const llvm::Triple &T,
> > }
> > }
> >
> > -unsigned tools::getLTOParallelism(const ArgList &Args, const Driver
> > &D) {
> > - unsigned Parallelism = 0;
> > +llvm::StringRef tools::getLTOParallelism(const ArgList &Args, const
> > +Driver &D) {
> > Arg *LtoJobsArg = Args.getLastArg(options::OPT_flto_jobs_EQ);
> > - if (LtoJobsArg &&
> > - StringRef(LtoJobsArg->getValue()).getAsInteger(10, Parallelism))
> > - D.Diag(diag::err_drv_invalid_int_value) << LtoJobsArg->getAsString(Args)
> > - << LtoJobsArg->getValue();
> > - return Parallelism;
> > + if (!LtoJobsArg)
> > + return {};
> > + if (!llvm::get_threadpool_strategy(LtoJobsArg->getValue()))
> > + D.Diag(diag::err_drv_invalid_int_value)
> > + << LtoJobsArg->getAsString(Args) << LtoJobsArg->getValue();
> > + return LtoJobsArg->getValue();
> > }
> >
> > // CloudABI uses -ffunction-sections and -fdata-sections by default.
> > @@ -410,7 +411,8 @@ void tools::addLTOOptions(const ToolChain &ToolChain, const ArgList &Args,
> > if (IsThinLTO)
> > CmdArgs.push_back("-plugin-opt=thinlto");
> >
> > - if (unsigned Parallelism = getLTOParallelism(Args,
> > ToolChain.getDriver()))
> > + StringRef Parallelism = getLTOParallelism(Args,
> > + ToolChain.getDriver()); if (!Parallelism.empty())
> > CmdArgs.push_back(
> > Args.MakeArgString("-plugin-opt=jobs=" +
> > Twine(Parallelism)));
> >
> >
> > diff --git a/clang/lib/Driver/ToolChains/CommonArgs.h
> > b/clang/lib/Driver/ToolChains/CommonArgs.h
> > index 984f3ee98af1..c94b2b828c9b 100644
> > --- a/clang/lib/Driver/ToolChains/CommonArgs.h
> > +++ b/clang/lib/Driver/ToolChains/CommonArgs.h
> > @@ -88,7 +88,8 @@ llvm::opt::Arg *getLastProfileSampleUseArg(const
> > llvm::opt::ArgList &Args);
> >
> > bool isObjCAutoRefCount(const llvm::opt::ArgList &Args);
> >
> > -unsigned getLTOParallelism(const llvm::opt::ArgList &Args, const
> > Driver &D);
> > +llvm::StringRef getLTOParallelism(const llvm::opt::ArgList &Args,
> > + const Driver &D);
> >
> > bool areOptimizationsEnabled(const llvm::opt::ArgList &Args);
> >
> >
> > diff --git a/clang/lib/Driver/ToolChains/Darwin.cpp
> > b/clang/lib/Driver/ToolChains/Darwin.cpp
> > index d1aec649cdc7..451d0d206d07 100644
> > --- a/clang/lib/Driver/ToolChains/Darwin.cpp
> > +++ b/clang/lib/Driver/ToolChains/Darwin.cpp
> > @@ -23,6 +23,7 @@
> > #include "llvm/Support/Path.h"
> > #include "llvm/Support/ScopedPrinter.h"
> > #include "llvm/Support/TargetParser.h"
> > +#include "llvm/Support/Threading.h"
> > #include "llvm/Support/VirtualFileSystem.h"
> > #include <cstdlib> // ::getenv
> >
> > @@ -605,10 +606,12 @@ void darwin::Linker::ConstructJob(Compilation
> > &C, const JobAction &JA,
> >
> > getMachOToolChain().addProfileRTLibs(Args, CmdArgs);
> >
> > - if (unsigned Parallelism =
> > - getLTOParallelism(Args, getToolChain().getDriver())) {
> > + StringRef Parallelism = getLTOParallelism(Args,
> > + getToolChain().getDriver()); if (!Parallelism.empty()) {
> > CmdArgs.push_back("-mllvm");
> > - CmdArgs.push_back(Args.MakeArgString("-threads=" + Twine(Parallelism)));
> > + unsigned NumThreads =
> > + llvm::get_threadpool_strategy(Parallelism)->compute_thread_count();
> > + CmdArgs.push_back(Args.MakeArgString("-threads=" +
> > + Twine(NumThreads)));
> > }
> >
> > if (getToolChain().ShouldLinkCXXStdlib(Args))
> >
> > diff --git a/lld/COFF/Config.h b/lld/COFF/Config.h index
> > e376ea51f133..93b732ea6821 100644
> > --- a/lld/COFF/Config.h
> > +++ b/lld/COFF/Config.h
> > @@ -144,7 +144,7 @@ struct Configuration {
> > unsigned ltoo = 2;
> >
> > // Used for /opt:lldltojobs=N
> > - unsigned thinLTOJobs = 0;
> > + std::string thinLTOJobs;
> > // Used for /opt:lldltopartitions=N
> > unsigned ltoPartitions = 1;
> >
> >
> > diff --git a/lld/COFF/Driver.cpp b/lld/COFF/Driver.cpp index
> > 8b3ba1cdf24b..7d99f798a322 100644
> > --- a/lld/COFF/Driver.cpp
> > +++ b/lld/COFF/Driver.cpp
> > @@ -1417,9 +1417,9 @@ void LinkerDriver::link(ArrayRef<const char *> argsArr) {
> > error("/opt:lldlto: invalid optimization level: " + optLevel);
> > } else if (s.startswith("lldltojobs=")) {
> > StringRef jobs = s.substr(11);
> > - if (jobs.getAsInteger(10, config->thinLTOJobs) ||
> > - config->thinLTOJobs == 0)
> > + if (!get_threadpool_strategy(jobs))
> > error("/opt:lldltojobs: invalid job count: " + jobs);
> > + config->thinLTOJobs = jobs.str();
> > } else if (s.startswith("lldltopartitions=")) {
> > StringRef n = s.substr(17);
> > if (n.getAsInteger(10, config->ltoPartitions) ||
> >
> > diff --git a/lld/COFF/LTO.cpp b/lld/COFF/LTO.cpp index
> > 247331b16928..94906e500593 100644
> > --- a/lld/COFF/LTO.cpp
> > +++ b/lld/COFF/LTO.cpp
> > @@ -101,8 +101,9 @@ BitcodeCompiler::BitcodeCompiler() {
> > std::string(config->thinLTOPrefixReplace.first),
> > std::string(config->thinLTOPrefixReplace.second),
> > config->thinLTOEmitImportsFiles, indexFile.get(),
> > OnIndexWrite);
> > - } else if (config->thinLTOJobs != 0) {
> > - backend = lto::createInProcessThinBackend(config->thinLTOJobs);
> > + } else {
> > + backend = lto::createInProcessThinBackend(
> > + llvm::heavyweight_hardware_concurrency(config->thinLTOJobs));
> > }
> >
> > ltoObj = std::make_unique<lto::LTO>(createConfig(), backend,
> >
> > diff --git a/lld/ELF/Config.h b/lld/ELF/Config.h index
> > 9f9f0d696f63..0263708634cb 100644
> > --- a/lld/ELF/Config.h
> > +++ b/lld/ELF/Config.h
> > @@ -246,7 +246,7 @@ struct Configuration {
> > unsigned ltoPartitions;
> > unsigned ltoo;
> > unsigned optimize;
> > - unsigned thinLTOJobs;
> > + StringRef thinLTOJobs;
> > unsigned timeTraceGranularity;
> > int32_t splitStackAdjustSize;
> >
> >
> > diff --git a/lld/ELF/Driver.cpp b/lld/ELF/Driver.cpp index
> > f9a1340e25d3..95276cd33420 100644
> > --- a/lld/ELF/Driver.cpp
> > +++ b/lld/ELF/Driver.cpp
> > @@ -974,7 +974,7 @@ static void readConfigs(opt::InputArgList &args) {
> > config->thinLTOIndexOnly = args.hasArg(OPT_thinlto_index_only) ||
> > args.hasArg(OPT_thinlto_index_only_eq);
> > config->thinLTOIndexOnlyArg =
> > args.getLastArgValue(OPT_thinlto_index_only_eq);
> > - config->thinLTOJobs = args::getInteger(args, OPT_thinlto_jobs,
> > -1u);
> > + config->thinLTOJobs = args.getLastArgValue(OPT_thinlto_jobs);
> > config->thinLTOObjectSuffixReplace =
> > getOldNewOptions(args, OPT_thinlto_object_suffix_replace_eq);
> > config->thinLTOPrefixReplace =
> > @@ -1040,8 +1040,8 @@ static void readConfigs(opt::InputArgList &args) {
> > error("invalid optimization level for LTO: " + Twine(config->ltoo));
> > if (config->ltoPartitions == 0)
> > error("--lto-partitions: number of threads must be > 0");
> > - if (config->thinLTOJobs == 0)
> > - error("--thinlto-jobs: number of threads must be > 0");
> > + if (!get_threadpool_strategy(config->thinLTOJobs))
> > + error("--thinlto-jobs: invalid job count: " +
> > + config->thinLTOJobs);
> >
> > if (config->splitStackAdjustSize < 0)
> > error("--split-stack-adjust-size: size must be >= 0");
> >
> > diff --git a/lld/ELF/LTO.cpp b/lld/ELF/LTO.cpp index
> > ba943283706a..f4e04a82f0c2 100644
> > --- a/lld/ELF/LTO.cpp
> > +++ b/lld/ELF/LTO.cpp
> > @@ -146,8 +146,9 @@ BitcodeCompiler::BitcodeCompiler() {
> > std::string(config->thinLTOPrefixReplace.first),
> > std::string(config->thinLTOPrefixReplace.second),
> > config->thinLTOEmitImportsFiles, indexFile.get(),
> > onIndexWrite);
> > - } else if (config->thinLTOJobs != -1U) {
> > - backend = lto::createInProcessThinBackend(config->thinLTOJobs);
> > + } else {
> > + backend = lto::createInProcessThinBackend(
> > + llvm::heavyweight_hardware_concurrency(config->thinLTOJobs));
> > }
> >
> > ltoObj = std::make_unique<lto::LTO>(createConfig(), backend,
> >
> > diff --git a/lld/test/COFF/thinlto.ll b/lld/test/COFF/thinlto.ll
> > index 517be330c710..a8390721cc42 100644
> > --- a/lld/test/COFF/thinlto.ll
> > +++ b/lld/test/COFF/thinlto.ll
> > @@ -6,6 +6,16 @@
> > ; RUN: lld-link /lldsavetemps /out:%T/thinlto/main.exe /entry:main
> > /subsystem:console %T/thinlto/main.obj %T/thinlto/foo.obj ; RUN:
> > llvm-nm %T/thinlto/main.exe1.lto.obj | FileCheck %s
> >
> > +; Test various possible options for /opt:lldltojobs ; RUN: lld-link
> > +/lldsavetemps /out:%T/thinlto/main.exe /entry:main /subsystem:console
> > +%T/thinlto/main.obj %T/thinlto/foo.obj /opt:lldltojobs=1 ; RUN:
> > +llvm-nm %T/thinlto/main.exe1.lto.obj | FileCheck %s ; RUN: lld-link
> > +/lldsavetemps /out:%T/thinlto/main.exe /entry:main /subsystem:console
> > +%T/thinlto/main.obj %T/thinlto/foo.obj /opt:lldltojobs=all ; RUN:
> > +llvm-nm %T/thinlto/main.exe1.lto.obj | FileCheck %s ; RUN: lld-link
> > +/lldsavetemps /out:%T/thinlto/main.exe /entry:main /subsystem:console
> > +%T/thinlto/main.obj %T/thinlto/foo.obj /opt:lldltojobs=1000 ; RUN:
> > +llvm-nm %T/thinlto/main.exe1.lto.obj | FileCheck %s ; RUN: not
> > +lld-link /lldsavetemps /out:%T/thinlto/main.exe /entry:main
> > +/subsystem:console %T/thinlto/main.obj %T/thinlto/foo.obj
> > +/opt:lldltojobs=foo 2>&1 | FileCheck %s --check-prefix=BAD-JOBS ;
> > +BAD-JOBS: error: /opt:lldltojobs: invalid job count: foo
> > +
> > ; This command will store full path to foo.obj in the archive %t.lib
> > ; Check that /lldsavetemps is still usable in such case.
> > ; RUN: lld-link /lib %T/thinlto/foo.obj /out:%t.lib
> >
> > diff --git a/lld/test/ELF/basic.s b/lld/test/ELF/basic.s index
> > b0509a2dc123..5300c6efa2ef 100644
> > --- a/lld/test/ELF/basic.s
> > +++ b/lld/test/ELF/basic.s
> > @@ -249,9 +249,21 @@ _start:
> > # RUN: not ld.lld %t --plugin-opt=lto-partitions=0 2>&1 | FileCheck
> > --check-prefix=NOTHREADS %s # NOTHREADS: --lto-partitions: number of
> > threads must be > 0
> >
> > -# RUN: not ld.lld %t --thinlto-jobs=0 2>&1 | FileCheck
> > --check-prefix=NOTHREADSTHIN %s -# RUN: not ld.lld %t
> > --plugin-opt=jobs=0 2>&1 | FileCheck --check-prefix=NOTHREADSTHIN %s
> > -# NOTHREADSTHIN: --thinlto-jobs: number of threads must be > 0
> > +# RUN: ld.lld %t --thinlto-jobs=0 -verbose 2>&1 | FileCheck
> > +--check-prefix=THREADSTHIN %s # RUN: ld.lld %t --thinlto-jobs=1
> > +-verbose 2>&1 | FileCheck --check-prefix=THREADSTHIN %s # RUN: ld.lld
> > +%t --thinlto-jobs=2 -verbose 2>&1 | FileCheck
> > +--check-prefix=THREADSTHIN %s # RUN: ld.lld %t --thinlto-jobs=all
> > +-verbose 2>&1 | FileCheck --check-prefix=THREADSTHIN %s # RUN: ld.lld
> > +%t --thinlto-jobs=1000 -verbose 2>&1 | FileCheck
> > +--check-prefix=THREADSTHIN %s # THREADSTHIN: basic.s.tmp # RUN: not
> > +ld.lld %t --thinlto-jobs=foo -verbose 2>&1 | FileCheck
> > +--check-prefix=BADTHREADSTHIN %s # BADTHREADSTHIN: error:
> > +--thinlto-jobs: invalid job count: foo
> > +
> > +# RUN: ld.lld %t --plugin-opt=jobs=0 -verbose 2>&1 | FileCheck
> > +--check-prefix=THREADSTHIN %s # RUN: ld.lld %t --plugin-opt=jobs=1
> > +-verbose 2>&1 | FileCheck --check-prefix=THREADSTHIN %s # RUN: ld.lld
> > +%t --plugin-opt=jobs=2 -verbose 2>&1 | FileCheck
> > +--check-prefix=THREADSTHIN %s # RUN: ld.lld %t --plugin-opt=jobs=all
> > +-verbose 2>&1 | FileCheck --check-prefix=THREADSTHIN %s # RUN: ld.lld
> > +%t --plugin-opt=jobs=1000 -verbose 2>&1 | FileCheck
> > +--check-prefix=THREADSTHIN %s # RUN: not ld.lld %t
> > +--plugin-opt=jobs=foo -verbose 2>&1 | FileCheck
> > +--check-prefix=BADTHREADSTHIN %s
> >
> > # RUN: not ld.lld %t -z ifunc-noplt -z text 2>&1 | FileCheck
> > --check-prefix=NOIFUNCPLTNOTEXTREL %s # NOIFUNCPLTNOTEXTREL: -z text
> > and -z ifunc-noplt may not be used together
> >
> > diff --git a/lld/test/ELF/lto/thinlto.ll
> > b/lld/test/ELF/lto/thinlto.ll index 2a7ffb90f6ff..ce37433f0bf4 100644
> > --- a/lld/test/ELF/lto/thinlto.ll
> > +++ b/lld/test/ELF/lto/thinlto.ll
> > @@ -16,8 +16,25 @@
> > ; RUN: llvm-nm %t31.lto.o | FileCheck %s --check-prefix=NM1 ; RUN:
> > llvm-nm %t32.lto.o | FileCheck %s --check-prefix=NM2
> >
> > -; Then check without --thinlto-jobs (which currently default to
> > hardware_concurrency) -; RUN: ld.lld -shared %t1.o %t2.o -o %t3
> > +; Test with all threads, on all cores, on all CPU sockets ; RUN: rm
> > +-f %t31.lto.o %t32.lto.o ; RUN: ld.lld -save-temps --thinlto-jobs=all
> > +-shared %t1.o %t2.o -o %t3 ; RUN: llvm-nm %t31.lto.o | FileCheck %s
> > +--check-prefix=NM1 ; RUN: llvm-nm %t32.lto.o | FileCheck %s
> > +--check-prefix=NM2
> > +
> > +; Test with many more threads than the system has ; RUN: rm -f
> > +%t31.lto.o %t32.lto.o ; RUN: ld.lld -save-temps --thinlto-jobs=1000
> > +-shared %t1.o %t2.o -o %t3 ; RUN: llvm-nm %t31.lto.o | FileCheck %s
> > +--check-prefix=NM1 ; RUN: llvm-nm %t32.lto.o | FileCheck %s
> > +--check-prefix=NM2
> > +
> > +; Test with a bad value
> > +; RUN: rm -f %t31.lto.o %t32.lto.o
> > +; RUN: not ld.lld -save-temps --thinlto-jobs=foo -shared %t1.o %t2.o
> > +-o %t3 2>&1 | FileCheck %s --check-prefix=BAD-JOBS ; BAD-JOBS: error:
> > +--thinlto-jobs: invalid job count: foo
> > +
> > +; Then check without --thinlto-jobs (which currently defaults to
> > +heavyweight_hardware_concurrency, meanning one thread per hardware
> > +core -- not SMT) ; RUN: ld.lld -shared -save-temps %t1.o %t2.o -o %t3
> > ; RUN: llvm-nm %t31.lto.o | FileCheck %s --check-prefix=NM1 ; RUN:
> > llvm-nm %t32.lto.o | FileCheck %s --check-prefix=NM2
> >
> >
> > diff --git a/lld/test/wasm/lto/thinlto.ll
> > b/lld/test/wasm/lto/thinlto.ll index 062da1a33b89..460452905597 100644
> > --- a/lld/test/wasm/lto/thinlto.ll
> > +++ b/lld/test/wasm/lto/thinlto.ll
> > @@ -14,8 +14,26 @@
> > ; RUN: llvm-nm %t31.lto.o | FileCheck %s --check-prefix=NM1 ; RUN:
> > llvm-nm %t32.lto.o | FileCheck %s --check-prefix=NM2
> >
> > -; Check without --thinlto-jobs (which currently default to
> > hardware_concurrency) -; RUN: wasm-ld -r %t1.o %t2.o -o %t3
> > +; Test with all threads, on all cores, on all CPU sockets ; RUN: rm
> > +-f %t31.lto.o %t32.lto.o ; RUN: wasm-ld -r -save-temps
> > +--thinlto-jobs=all %t1.o %t2.o -o %t3 ; RUN: llvm-nm %t31.lto.o |
> > +FileCheck %s --check-prefix=NM1 ; RUN: llvm-nm %t32.lto.o | FileCheck
> > +%s --check-prefix=NM2
> > +
> > +; Test with many more threads than the system has ; RUN: rm -f
> > +%t31.lto.o %t32.lto.o ; RUN: wasm-ld -r -save-temps
> > +--thinlto-jobs=1000 %t1.o %t2.o -o %t3 ; RUN: llvm-nm %t31.lto.o |
> > +FileCheck %s --check-prefix=NM1 ; RUN: llvm-nm %t32.lto.o | FileCheck
> > +%s --check-prefix=NM2
> > +
> > +; Test with a bad value
> > +; RUN: rm -f %t31.lto.o %t32.lto.o
> > +; RUN: not wasm-ld -r -save-temps --thinlto-jobs=foo %t1.o %t2.o -o
> > +%t3 2>&1 | FileCheck %s --check-prefix=BAD-JOBS ; BAD-JOBS: error:
> > +--thinlto-jobs: invalid job count: foo
> > +
> > +; Check without --thinlto-jobs (which currently defaults to
> > +heavyweight_hardware_concurrency, meanning one thread per hardware
> > +core -- not SMT) ; RUN: rm -f %t31.lto.o %t32.lto.o ; RUN: wasm-ld -r
> > +-save-temps %t1.o %t2.o -o %t3
> > ; RUN: llvm-nm %t31.lto.o | FileCheck %s --check-prefix=NM1 ; RUN:
> > llvm-nm %t32.lto.o | FileCheck %s --check-prefix=NM2
> >
> >
> > diff --git a/lld/wasm/Config.h b/lld/wasm/Config.h index
> > 3b3f2daf576c..b396bda93d58 100644
> > --- a/lld/wasm/Config.h
> > +++ b/lld/wasm/Config.h
> > @@ -53,7 +53,7 @@ struct Configuration {
> > unsigned ltoPartitions;
> > unsigned ltoo;
> > unsigned optimize;
> > - unsigned thinLTOJobs;
> > + llvm::StringRef thinLTOJobs;
> >
> > llvm::StringRef entry;
> > llvm::StringRef outputFile;
> >
> > diff --git a/lld/wasm/Driver.cpp b/lld/wasm/Driver.cpp index
> > 73b330ae1470..b38e980004ad 100644
> > --- a/lld/wasm/Driver.cpp
> > +++ b/lld/wasm/Driver.cpp
> > @@ -362,7 +362,7 @@ static void readConfigs(opt::InputArgList &args) {
> > config->thinLTOCachePolicy = CHECK(
> > parseCachePruningPolicy(args.getLastArgValue(OPT_thinlto_cache_policy)),
> > "--thinlto-cache-policy: invalid cache policy");
> > - config->thinLTOJobs = args::getInteger(args, OPT_thinlto_jobs,
> > -1u);
> > + config->thinLTOJobs = args.getLastArgValue(OPT_thinlto_jobs);
> > errorHandler().verbose = args.hasArg(OPT_verbose);
> > LLVM_DEBUG(errorHandler().verbose = true);
> > threadsEnabled = args.hasFlag(OPT_threads, OPT_no_threads, true);
> > @@ -415,8 +415,8 @@ static void checkOptions(opt::InputArgList &args) {
> > error("invalid optimization level for LTO: " + Twine(config->ltoo));
> > if (config->ltoPartitions == 0)
> > error("--lto-partitions: number of threads must be > 0");
> > - if (config->thinLTOJobs == 0)
> > - error("--thinlto-jobs: number of threads must be > 0");
> > + if (!get_threadpool_strategy(config->thinLTOJobs))
> > + error("--thinlto-jobs: invalid job count: " +
> > + config->thinLTOJobs);
> >
> > if (config->pie && config->shared)
> > error("-shared and -pie may not be used together");
> >
> > diff --git a/lld/wasm/LTO.cpp b/lld/wasm/LTO.cpp index
> > 0b7d8cb12d0e..28b62022f29f 100644
> > --- a/lld/wasm/LTO.cpp
> > +++ b/lld/wasm/LTO.cpp
> > @@ -63,10 +63,8 @@ static std::unique_ptr<lto::LTO> createLTO() {
> > if (config->saveTemps)
> > checkError(c.addSaveTemps(config->outputFile.str() + ".",
> > /*UseInputModulePath*/ true));
> > -
> > - lto::ThinBackend backend;
> > - if (config->thinLTOJobs != -1U)
> > - backend = lto::createInProcessThinBackend(config->thinLTOJobs);
> > + lto::ThinBackend backend = lto::createInProcessThinBackend(
> > + llvm::heavyweight_hardware_concurrency(config->thinLTOJobs));
> > return std::make_unique<lto::LTO>(std::move(c), backend,
> > config->ltoPartitions); }
> >
> > diff --git a/llvm/include/llvm/LTO/LTO.h
> > b/llvm/include/llvm/LTO/LTO.h index df2853345682..e93439819415 100644
> > --- a/llvm/include/llvm/LTO/LTO.h
> > +++ b/llvm/include/llvm/LTO/LTO.h
> > @@ -228,7 +228,7 @@ using ThinBackend =
> > std::function<std::unique_ptr<ThinBackendProc>(
> >
> > /// This ThinBackend runs the individual backend jobs in-process.
> > /// The default value means to use one job per hardware core (not hyper-thread).
> > -ThinBackend createInProcessThinBackend(unsigned ParallelismLevel =
> > 0);
> > +ThinBackend createInProcessThinBackend(ThreadPoolStrategy
> > +Parallelism);
> >
> > /// This ThinBackend writes individual module indexes to files,
> > instead of /// running the individual backend jobs. This backend is
> > for distributed builds
> >
> > diff --git a/llvm/include/llvm/Support/Threading.h
> > b/llvm/include/llvm/Support/Threading.h
> > index 24a0789e0964..f133c790ad96 100644
> > --- a/llvm/include/llvm/Support/Threading.h
> > +++ b/llvm/include/llvm/Support/Threading.h
> > @@ -166,8 +166,20 @@ void llvm_execute_on_thread_async(
> > /// sockets. \p ThreadPoolNum represents a number bounded by [0,
> > /// compute_thread_count()).
> > void apply_thread_strategy(unsigned ThreadPoolNum) const;
> > +
> > + /// Finds the CPU socket where a thread should go. Returns 'None' if the
> > + /// thread shall remain on the actual CPU socket.
> > + Optional<unsigned> compute_cpu_socket(unsigned ThreadPoolNum)
> > + const;
> > };
> >
> > + /// Build a strategy from a number of threads as a string provided in \p Num.
> > + /// When Num is above the max number of threads specified by the \p
> > + Default /// strategy, we attempt to equally allocate the threads on all CPU sockets.
> > + /// "0" or an empty string will return the \p Default strategy.
> > + /// "all" for using all hardware threads.
> > + Optional<ThreadPoolStrategy>
> > + get_threadpool_strategy(StringRef Num, ThreadPoolStrategy Default =
> > + {});
> > +
> > /// Returns a thread strategy for tasks requiring significant memory or other
> > /// resources. To be used for workloads where hardware_concurrency() proves to
> > /// be less efficient. Avoid this strategy if doing lots of I/O.
> > Currently @@ -182,6 +194,18 @@ void llvm_execute_on_thread_async(
> > return S;
> > }
> >
> > + /// Like heavyweight_hardware_concurrency() above, but builds a
> > + strategy /// based on the rules described for get_threadpool_strategy().
> > + /// If \p Num is invalid, returns a default strategy where one
> > + thread per /// hardware core is used.
> > + inline ThreadPoolStrategy heavyweight_hardware_concurrency(StringRef Num) {
> > + Optional<ThreadPoolStrategy> S =
> > + get_threadpool_strategy(Num, heavyweight_hardware_concurrency());
> > + if (S)
> > + return *S;
> > + return heavyweight_hardware_concurrency();
> > + }
> > +
> > /// Returns a default thread strategy where all available hardware ressources
> > /// are to be used, except for those initially excluded by an affinity mask.
> > /// This function takes affinity into consideration. Returns 1 when
> > LLVM is
> >
> > diff --git a/llvm/lib/LTO/LTO.cpp b/llvm/lib/LTO/LTO.cpp index
> > f8affcb20cef..f652bd628770 100644
> > --- a/llvm/lib/LTO/LTO.cpp
> > +++ b/llvm/lib/LTO/LTO.cpp
> > @@ -477,7 +477,8 @@ LTO::RegularLTOState::RegularLTOState(unsigned
> > ParallelCodeGenParallelismLevel, LTO::ThinLTOState::ThinLTOState(ThinBackend Backend)
> > : Backend(Backend), CombinedIndex(/*HaveGVs*/ false) {
> > if (!Backend)
> > - this->Backend = createInProcessThinBackend();
> > + this->Backend =
> > +
> > + createInProcessThinBackend(llvm::heavyweight_hardware_concurrency())
> > + ;
> > }
> >
> > LTO::LTO(Config Conf, ThinBackend Backend, @@ -1090,13 +1091,12 @@
> > class InProcessThinBackend : public ThinBackendProc {
> > public:
> > InProcessThinBackend(
> > const Config &Conf, ModuleSummaryIndex &CombinedIndex,
> > - unsigned ThinLTOParallelismLevel,
> > + ThreadPoolStrategy ThinLTOParallelism,
> > const StringMap<GVSummaryMapTy> &ModuleToDefinedGVSummaries,
> > AddStreamFn AddStream, NativeObjectCache Cache)
> > : ThinBackendProc(Conf, CombinedIndex, ModuleToDefinedGVSummaries),
> > - BackendThreadPool(
> > - heavyweight_hardware_concurrency(ThinLTOParallelismLevel)),
> > - AddStream(std::move(AddStream)), Cache(std::move(Cache)) {
> > + BackendThreadPool(ThinLTOParallelism), AddStream(std::move(AddStream)),
> > + Cache(std::move(Cache)) {
> > for (auto &Name : CombinedIndex.cfiFunctionDefs())
> > CfiFunctionDefs.insert(
> >
> > GlobalValue::getGUID(GlobalValue::dropLLVMManglingEscape(Name)));
> > @@ -1192,13 +1192,13 @@ class InProcessThinBackend : public
> > ThinBackendProc { }; } // end anonymous namespace
> >
> > -ThinBackend lto::createInProcessThinBackend(unsigned
> > ParallelismLevel) {
> > +ThinBackend lto::createInProcessThinBackend(ThreadPoolStrategy
> > +Parallelism) {
> > return [=](const Config &Conf, ModuleSummaryIndex &CombinedIndex,
> > const StringMap<GVSummaryMapTy> &ModuleToDefinedGVSummaries,
> > AddStreamFn AddStream, NativeObjectCache Cache) {
> > return std::make_unique<InProcessThinBackend>(
> > - Conf, CombinedIndex, ParallelismLevel, ModuleToDefinedGVSummaries,
> > - AddStream, Cache);
> > + Conf, CombinedIndex, Parallelism, ModuleToDefinedGVSummaries, AddStream,
> > + Cache);
> > };
> > }
> >
> >
> > diff --git a/llvm/lib/Support/Threading.cpp
> > b/llvm/lib/Support/Threading.cpp index de5adaddd9d3..39de05892981
> > 100644
> > --- a/llvm/lib/Support/Threading.cpp
> > +++ b/llvm/lib/Support/Threading.cpp
> > @@ -84,16 +84,34 @@ void llvm::llvm_execute_on_thread_async(
> > int computeHostNumHardwareThreads();
> >
> > unsigned llvm::ThreadPoolStrategy::compute_thread_count() const {
> > + if (ThreadsRequested > 0)
> > + return ThreadsRequested;
> > +
> > int MaxThreadCount = UseHyperThreads ? computeHostNumHardwareThreads()
> > : sys::getHostNumPhysicalCores();
> > if (MaxThreadCount <= 0)
> > MaxThreadCount = 1;
> > + return MaxThreadCount;
> > +}
> >
> > - // No need to create more threads than there are hardware threads,
> > it would
> > - // uselessly induce more context-switching and cache eviction.
> > - if (!ThreadsRequested || ThreadsRequested > (unsigned)MaxThreadCount)
> > - return MaxThreadCount;
> > - return ThreadsRequested;
> > +Optional<ThreadPoolStrategy>
> > +llvm::get_threadpool_strategy(StringRef Num, ThreadPoolStrategy
> > +Default) {
> > + if (Num == "all")
> > + return llvm::hardware_concurrency();
> > + if (Num.empty())
> > + return Default;
> > + unsigned V;
> > + if (Num.getAsInteger(10, V))
> > + return None; // malformed 'Num' value
> > + if (V == 0)
> > + return Default;
> > +
> > + // Do not take the Default into account. This effectively disables
> > + // heavyweight_hardware_concurrency() if the user asks for any
> > + number of // threads on the cmd-line.
> > + ThreadPoolStrategy S = llvm::hardware_concurrency();
> > + S.ThreadsRequested = V; return S;
> > }
> >
> > namespace {
> >
> > diff --git a/llvm/lib/Support/Unix/Threading.inc
> > b/llvm/lib/Support/Unix/Threading.inc
> > index 8cacaa83e961..307a24469144 100644
> > --- a/llvm/lib/Support/Unix/Threading.inc
> > +++ b/llvm/lib/Support/Unix/Threading.inc
> > @@ -273,7 +273,7 @@ SetThreadPriorityResult
> > llvm::set_thread_priority(ThreadPriority Priority) { int
> > computeHostNumHardwareThreads() { #if defined(HAVE_SCHED_GETAFFINITY) && defined(HAVE_CPU_COUNT)
> > cpu_set_t Set;
> > - if (sched_getaffinity(0, sizeof(Set), &Set))
> > + if (sched_getaffinity(0, sizeof(Set), &Set) == 0)
> > return CPU_COUNT(&Set);
> > #endif
> > // Guard against std::thread::hardware_concurrency() returning 0.
> >
> > diff --git a/llvm/lib/Support/Windows/Threading.inc
> > b/llvm/lib/Support/Windows/Threading.inc
> > index 50c94068b395..296e87b77695 100644
> > --- a/llvm/lib/Support/Windows/Threading.inc
> > +++ b/llvm/lib/Support/Windows/Threading.inc
> > @@ -131,6 +131,10 @@ struct ProcessorGroup {
> > unsigned UsableThreads;
> > unsigned ThreadsPerCore;
> > uint64_t Affinity;
> > +
> > + unsigned useableCores() const {
> > + return std::max(1U, UsableThreads / ThreadsPerCore); }
> > };
> >
> > template <typename F>
> > @@ -232,33 +236,41 @@ int computeHostNumHardwareThreads() {
> > return Threads;
> > }
> >
> > -// Assign the current thread to a more appropriate CPU socket or CPU
> > group -void llvm::ThreadPoolStrategy::apply_thread_strategy(
> > - unsigned ThreadPoolNum) const {
> > +// Finds the proper CPU socket where a thread number should go. Returns 'None'
> > +// if the thread shall remain on the actual CPU socket.
> > +Optional<unsigned>
> > +llvm::ThreadPoolStrategy::compute_cpu_socket(unsigned ThreadPoolNum)
> > +const {
> > ArrayRef<ProcessorGroup> Groups = getProcessorGroups();
> > + // Only one CPU socket in the system or process affinity was set,
> > + no need to // move the thread(s) to another CPU socket.
> > + if (Groups.size() <= 1)
> > + return None;
> > +
> > + // We ask for less threads than there are hardware threads per CPU
> > + socket, no // need to dispatch threads to other CPU sockets.
> > + unsigned MaxThreadsPerSocket =
> > + UseHyperThreads ? Groups[0].UsableThreads :
> > + Groups[0].useableCores(); if (compute_thread_count() <= MaxThreadsPerSocket)
> > + return None;
> >
> > assert(ThreadPoolNum < compute_thread_count() &&
> > "The thread index is not within thread strategy's range!");
> >
> > - // In this mode, the ThreadNumber represents the core number, not
> > the
> > - // hyper-thread number. Assumes all NUMA groups have the same
> > amount of
> > - // hyper-threads.
> > - if (!UseHyperThreads)
> > - ThreadPoolNum *= Groups[0].ThreadsPerCore;
> > -
> > - unsigned ThreadRangeStart = 0;
> > - for (unsigned I = 0; I < Groups.size(); ++I) {
> > - const ProcessorGroup &G = Groups[I];
> > - if (ThreadPoolNum >= ThreadRangeStart &&
> > - ThreadPoolNum < ThreadRangeStart + G.UsableThreads) {
> > -
> > - GROUP_AFFINITY Affinity{};
> > - Affinity.Group = G.ID;
> > - Affinity.Mask = G.Affinity;
> > - SetThreadGroupAffinity(GetCurrentThread(), &Affinity, nullptr);
> > - }
> > - ThreadRangeStart += G.UsableThreads;
> > - }
> > + // Assumes the same number of hardware threads per CPU socket.
> > + return (ThreadPoolNum * Groups.size()) / compute_thread_count(); }
> > +
> > +// Assign the current thread to a more appropriate CPU socket or CPU
> > +group void llvm::ThreadPoolStrategy::apply_thread_strategy(
> > + unsigned ThreadPoolNum) const {
> > + Optional<unsigned> Socket = compute_cpu_socket(ThreadPoolNum);
> > + if (!Socket)
> > + return;
> > + ArrayRef<ProcessorGroup> Groups = getProcessorGroups();
> > + GROUP_AFFINITY Affinity{};
> > + Affinity.Group = Groups[*Socket].ID;
> > + Affinity.Mask = Groups[*Socket].Affinity;
> > + SetThreadGroupAffinity(GetCurrentThread(), &Affinity, nullptr);
> > }
> >
> > llvm::BitVector llvm::get_thread_affinity_mask() {
> >
> > diff --git a/llvm/tools/gold/gold-plugin.cpp
> > b/llvm/tools/gold/gold-plugin.cpp index 2db571617181..383f972b8636
> > 100644
> > --- a/llvm/tools/gold/gold-plugin.cpp
> > +++ b/llvm/tools/gold/gold-plugin.cpp
> > @@ -28,6 +28,7 @@
> > #include "llvm/Support/MemoryBuffer.h"
> > #include "llvm/Support/Path.h"
> > #include "llvm/Support/TargetSelect.h"
> > +#include "llvm/Support/Threading.h"
> > #include "llvm/Support/raw_ostream.h"
> > #include <list>
> > #include <map>
> > @@ -134,11 +135,9 @@ namespace options {
> > };
> > static OutputType TheOutputType = OT_NORMAL;
> > static unsigned OptLevel = 2;
> > - // Default parallelism of 0 used to indicate that user did not specify.
> > - // Actual parallelism default value depends on implementation.
> > // Currently only affects ThinLTO, where the default is the max
> > cores in the
> > - // system.
> > - static unsigned Parallelism = 0;
> > + // system. See llvm::get_threadpool_strategy() for acceptable values.
> > + static std::string Parallelism;
> > // Default regular LTO codegen parallelism (number of partitions).
> > static unsigned ParallelCodeGenParallelismLevel = 1; #ifdef NDEBUG
> > @@ -272,8 +271,10 @@ namespace options {
> > message(LDPL_FATAL, "Optimization level must be between 0 and 3");
> > OptLevel = opt[1] - '0';
> > } else if (opt.startswith("jobs=")) {
> > - if (StringRef(opt_ + 5).getAsInteger(10, Parallelism))
> > - message(LDPL_FATAL, "Invalid parallelism level: %s", opt_ + 5);
> > + StringRef Num(opt_ + 5);
> > + if (!get_threadpool_strategy(Num))
> > + message(LDPL_FATAL, "Invalid parallelism level: %s", Num.data());
> > + Parallelism = Num;
> > } else if (opt.startswith("lto-partitions=")) {
> > if (opt.substr(strlen("lto-partitions="))
> > .getAsInteger(10, ParallelCodeGenParallelismLevel)) @@
> > -877,14 +878,15 @@ static std::unique_ptr<LTO> createLTO(IndexWriteCallback OnIndexWrite,
> > Conf.PTO.LoopVectorization = options::OptLevel > 1;
> > Conf.PTO.SLPVectorization = options::OptLevel > 1;
> >
> > - if (options::Parallelism)
> > - Backend = createInProcessThinBackend(options::Parallelism);
> > if (options::thinlto_index_only) {
> > std::string OldPrefix, NewPrefix;
> > getThinLTOOldAndNewPrefix(OldPrefix, NewPrefix);
> > Backend = createWriteIndexesThinBackend(OldPrefix, NewPrefix,
> > options::thinlto_emit_imports_files,
> > LinkedObjectsFile,
> > OnIndexWrite);
> > + } else {
> > + Backend = createInProcessThinBackend(
> > +
> > + llvm::heavyweight_hardware_concurrency(options::Parallelism));
> > }
> >
> > Conf.OverrideTriple = options::triple;
> >
> > diff --git a/llvm/tools/llvm-lto2/llvm-lto2.cpp
> > b/llvm/tools/llvm-lto2/llvm-lto2.cpp
> > index dda596be0ce6..6d78082f980e 100644
> > --- a/llvm/tools/llvm-lto2/llvm-lto2.cpp
> > +++ b/llvm/tools/llvm-lto2/llvm-lto2.cpp
> > @@ -68,9 +68,10 @@ static cl::opt<bool>
> > "distributed backend case"));
> >
> > // Default to using all available threads in the system, but using
> > only one -// thread per core, as indicated by the usage of -//
> > heavyweight_hardware_concurrency() in the InProcessThinBackend constructor.
> > -static cl::opt<int> Threads("thinlto-threads", cl::init(0));
> > +// thread per core (no SMT).
> > +// Use -thinlto-threads=all to use hardware_concurrency() instead,
> > +which means // to use all hardware threads or cores in the system.
> > +static cl::opt<std::string> Threads("thinlto-threads");
> >
> > static cl::list<std::string> SymbolResolutions(
> > "r",
> > @@ -286,7 +287,8 @@ static int run(int argc, char **argv) {
> > /* LinkedObjectsFile */ nullptr,
> > /* OnWrite */ {});
> > else
> > - Backend = createInProcessThinBackend(Threads);
> > + Backend = createInProcessThinBackend(
> > + llvm::heavyweight_hardware_concurrency(Threads));
> > LTO Lto(std::move(Conf), std::move(Backend));
> >
> > bool HasErrors = false;
> >
> >
> >
> > _______________________________________________
> > llvm-commits mailing list
> > llvm-commits at lists.llvm.org
> > https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits
More information about the llvm-commits
mailing list