aganea created this revision.
aganea added reviewers: tejohnson, thakis, rnk, RobRich999.
Herald added subscribers: cfe-commits, dang, dexonsmith, mikhail.ramalho,
steven_wu, MaskRay, aheejin, hiraditya, arichardson, inglorion, sbc100, emaste.
Herald added a reviewer: espindola.
Herald added projects: clang, LLVM.
Before this patch, it wasn't possible to extend the ThinLTO threads to all
SMT/CMT/hyper-threads in the system. Only one thread per core was allowed,
instructed by usage of `llvm::heavyweight_hardware_concurrency()` in places in
the ThinLTO objects initialization.
Any number passed to the LLD flag `/opt:lldltojobs=...`, or any other
ThinLTO-specific flag, would be interpreted in the context of
`llvm::heavyweight_hardware_concurrency()`, which means SMT disabled.
After this patch, one can say in LLD: `/opt:lldltojobs=all` -- which means to
ignore the `llvm::heavyweight_hardware_concurrency()` requirement, and instead
use all SMT/CMT/hyper-threads in the system, which is equivalent to using
`llvm::hardware_concurrency()`.
All cmd-line flags and code paths that lead to initialize ThinLTO have been
modified by this patch: `-flto-jobs=...`, `-threads=`, `--thinlto-jobs=...`,
`-thinlto-threads=...`, `--plugin-opt=jobs=...`.
Summary
-------
To sum up:
`/opt:lldltojobs=0` -- use all threads available, but constrained by usage of
`heavyweight_hardware_concurrency()` or `hardware_concurrency()`.
`/opt:lldltojobs=N` -- limit usage to N threads, but constrained by usage of
`heavyweight_hardware_concurrency()` or `hardware_concurrency()`.
`/opt:lldltojobs=all` -- use all threads available, regardless of usage of
`heavyweight_hardware_concurrency()` or `hardware_concurrency()`.
Repository:
rG LLVM Github Monorepo
https://reviews.llvm.org/D75153
Files:
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/lib/LTO/LTO.cpp
llvm/test/Transforms/PGOProfile/thinlto_samplepgo_icp3.ll
llvm/tools/gold/gold-plugin.cpp
llvm/tools/llvm-lto2/llvm-lto2.cpp
Index: llvm/tools/llvm-lto2/llvm-lto2.cpp
===================================================================
--- llvm/tools/llvm-lto2/llvm-lto2.cpp
+++ llvm/tools/llvm-lto2/llvm-lto2.cpp
@@ -66,9 +66,10 @@
"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",
@@ -276,6 +277,16 @@
Conf.PTO.LoopVectorization = Conf.OptLevel > 1;
Conf.PTO.SLPVectorization = Conf.OptLevel > 1;
+ auto getStrategy = [](StringRef Num) {
+ if (Num == "all")
+ return llvm::hardware_concurrency();
+ if (Num.empty())
+ return ThreadPoolStrategy();
+ unsigned V;
+ Num.getAsInteger(10, V);
+ return llvm::heavyweight_hardware_concurrency(V);
+ };
+
ThinBackend Backend;
if (ThinLTODistributedIndexes)
Backend = createWriteIndexesThinBackend(/* OldPrefix */ "",
@@ -284,7 +295,7 @@
/* LinkedObjectsFile */ nullptr,
/* OnWrite */ {});
else
- Backend = createInProcessThinBackend(Threads);
+ Backend = createInProcessThinBackend(getStrategy(Threads));
LTO Lto(std::move(Conf), std::move(Backend));
bool HasErrors = false;
Index: llvm/tools/gold/gold-plugin.cpp
===================================================================
--- llvm/tools/gold/gold-plugin.cpp
+++ llvm/tools/gold/gold-plugin.cpp
@@ -139,6 +139,7 @@
static unsigned Parallelism = 0;
// Default regular LTO codegen parallelism (number of partitions).
static unsigned ParallelCodeGenParallelismLevel = 1;
+ static bool ParallelismHeavyWeight = true;
#ifdef NDEBUG
static bool DisableVerify = true;
#else
@@ -270,7 +271,10 @@
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))
+ StringRef Num(opt_ + 5);
+ if (Num == "all")
+ ParallelismHeavyWeight = false;
+ else if (Num.getAsInteger(10, Parallelism))
message(LDPL_FATAL, "Invalid parallelism level: %s", opt_ + 5);
} else if (opt.startswith("lto-partitions=")) {
if (opt.substr(strlen("lto-partitions="))
@@ -875,14 +879,18 @@
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 {
+ ThreadPoolStrategy S =
+ options::ParallelismHeavyWeight
+ ? llvm::heavyweight_hardware_concurrency(options::Parallelism)
+ : llvm::hardware_concurrency();
+ Backend = createInProcessThinBackend(S);
}
Conf.OverrideTriple = options::triple;
Index: llvm/test/Transforms/PGOProfile/thinlto_samplepgo_icp3.ll
===================================================================
--- llvm/test/Transforms/PGOProfile/thinlto_samplepgo_icp3.ll
+++ llvm/test/Transforms/PGOProfile/thinlto_samplepgo_icp3.ll
@@ -7,6 +7,10 @@
; Test to make sure importing and dead stripping works in the
; case where the target is a local function that also indirectly calls itself.
; RUN: llvm-lto2 run -thinlto-threads=1 -save-temps -o %t3 %t.bc %t2.bc -r %t.bc,fptr,plx -r %t.bc,main,plx -r %t2.bc,_Z6updatei,pl -r %t2.bc,fptr,l -print-imports 2>&1 | FileCheck %s --check-prefix=IMPORTS
+
+; Also test with all threads on
+; RUN: llvm-lto2 run -thinlto-threads=all -save-temps -o %t3 %t.bc %t2.bc -r %t.bc,fptr,plx -r %t.bc,main,plx -r %t2.bc,_Z6updatei,pl -r %t2.bc,fptr,l -print-imports 2>&1 | FileCheck %s --check-prefix=IMPORTS
+
; Make sure we import the promted indirectly called target
; IMPORTS: Import _ZL3foov.llvm.0
Index: llvm/lib/LTO/LTO.cpp
===================================================================
--- llvm/lib/LTO/LTO.cpp
+++ llvm/lib/LTO/LTO.cpp
@@ -477,7 +477,8 @@
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 @@
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 @@
};
} // 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);
};
}
Index: llvm/include/llvm/LTO/LTO.h
===================================================================
--- llvm/include/llvm/LTO/LTO.h
+++ llvm/include/llvm/LTO/LTO.h
@@ -228,7 +228,7 @@
/// 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
Index: lld/wasm/LTO.cpp
===================================================================
--- lld/wasm/LTO.cpp
+++ lld/wasm/LTO.cpp
@@ -63,10 +63,11 @@
if (config->saveTemps)
checkError(c.addSaveTemps(config->outputFile.str() + ".",
/*UseInputModulePath*/ true));
-
- lto::ThinBackend backend;
- if (config->thinLTOJobs != -1U)
- backend = lto::createInProcessThinBackend(config->thinLTOJobs);
+ ThreadPoolStrategy S =
+ config->thinLTOJobsHeavyWeightThreads
+ ? llvm::heavyweight_hardware_concurrency(config->thinLTOJobs)
+ : llvm::hardware_concurrency();
+ lto::ThinBackend backend = lto::createInProcessThinBackend(S);
return std::make_unique<lto::LTO>(std::move(c), backend,
config->ltoPartitions);
}
Index: lld/wasm/Driver.cpp
===================================================================
--- lld/wasm/Driver.cpp
+++ lld/wasm/Driver.cpp
@@ -342,7 +342,15 @@
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);
+
+ if (auto *arg = args.getLastArgNoClaim(OPT_thinlto_jobs)) {
+ StringRef s = arg->getValue();
+ if (s == "all") {
+ config->thinLTOJobsHeavyWeightThreads = false;
+ arg->claim();
+ } else
+ config->thinLTOJobs = args::getInteger(args, OPT_thinlto_jobs, 0);
+ }
errorHandler().verbose = args.hasArg(OPT_verbose);
LLVM_DEBUG(errorHandler().verbose = true);
threadsEnabled = args.hasFlag(OPT_threads, OPT_no_threads, true);
@@ -395,8 +403,6 @@
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 (config->pie && config->shared)
error("-shared and -pie may not be used together");
Index: lld/wasm/Config.h
===================================================================
--- lld/wasm/Config.h
+++ lld/wasm/Config.h
@@ -46,6 +46,7 @@
bool stripDebug;
bool stackFirst;
bool trace;
+ bool thinLTOJobsHeavyWeightThreads = true;
uint32_t globalBase;
uint32_t initialMemory;
uint32_t maxMemory;
@@ -53,7 +54,7 @@
unsigned ltoPartitions;
unsigned ltoo;
unsigned optimize;
- unsigned thinLTOJobs;
+ unsigned thinLTOJobs = 0;
llvm::StringRef entry;
llvm::StringRef outputFile;
Index: lld/test/wasm/lto/thinlto.ll
===================================================================
--- lld/test/wasm/lto/thinlto.ll
+++ lld/test/wasm/lto/thinlto.ll
@@ -14,7 +14,13 @@
; 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)
+; 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
+
+; Check without --thinlto-jobs (which currently defaults to heavyweight_hardware_concurrency, meanning one thread per hardware core -- not SMT)
; RUN: wasm-ld -r %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
Index: lld/test/ELF/lto/thinlto.ll
===================================================================
--- lld/test/ELF/lto/thinlto.ll
+++ lld/test/ELF/lto/thinlto.ll
@@ -16,7 +16,13 @@
; 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)
+; 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
+
+; Then check without --thinlto-jobs (which currently defaults to heavyweight_hardware_concurrency, meanning one thread per hardware core -- not SMT)
; RUN: ld.lld -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
Index: lld/test/ELF/basic.s
===================================================================
--- lld/test/ELF/basic.s
+++ lld/test/ELF/basic.s
@@ -249,9 +249,19 @@
# 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
+# THREADSTHIN: basic.s.tmp
+# RUN: not ld.lld %t --thinlto-jobs=1- -verbose 2>&1 | FileCheck --check-prefix=BADTHREADSTHIN %s
+# BADTHREADSTHIN: error: --{{.*}}jobs=1-: number expected, but got '1-'
+
+# 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: not ld.lld %t --plugin-opt=jobs=1- -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
Index: lld/test/COFF/thinlto.ll
===================================================================
--- lld/test/COFF/thinlto.ll
+++ lld/test/COFF/thinlto.ll
@@ -6,6 +6,11 @@
; 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
+; 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
+
; CHECK-NOT: U foo
target datalayout = "e-m:w-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
Index: lld/ELF/LTO.cpp
===================================================================
--- lld/ELF/LTO.cpp
+++ lld/ELF/LTO.cpp
@@ -146,8 +146,12 @@
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 {
+ ThreadPoolStrategy S =
+ config->thinLTOJobsHeavyWeightThreads
+ ? llvm::heavyweight_hardware_concurrency(config->thinLTOJobs)
+ : llvm::hardware_concurrency();
+ backend = lto::createInProcessThinBackend(S);
}
ltoObj = std::make_unique<lto::LTO>(createConfig(), backend,
Index: lld/ELF/Driver.cpp
===================================================================
--- lld/ELF/Driver.cpp
+++ lld/ELF/Driver.cpp
@@ -977,7 +977,6 @@
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->thinLTOObjectSuffixReplace =
getOldNewOptions(args, OPT_thinlto_object_suffix_replace_eq);
config->thinLTOPrefixReplace =
@@ -1034,6 +1033,15 @@
for (auto *arg : args.filtered(OPT_plugin_opt))
parseClangOption(arg->getValue(), arg->getSpelling());
+ if (auto *arg = args.getLastArgNoClaim(OPT_thinlto_jobs)) {
+ StringRef s = arg->getValue();
+ if (s == "all") {
+ config->thinLTOJobsHeavyWeightThreads = false;
+ arg->claim();
+ } else
+ config->thinLTOJobs = args::getInteger(args, OPT_thinlto_jobs, 0);
+ }
+
// Parse -mllvm options.
for (auto *arg : args.filtered(OPT_mllvm))
parseClangOption(arg->getValue(), arg->getSpelling());
@@ -1042,8 +1050,6 @@
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 (config->splitStackAdjustSize < 0)
error("--split-stack-adjust-size: size must be >= 0");
Index: lld/ELF/Config.h
===================================================================
--- lld/ELF/Config.h
+++ lld/ELF/Config.h
@@ -191,6 +191,7 @@
bool trace;
bool thinLTOEmitImportsFiles;
bool thinLTOIndexOnly;
+ bool thinLTOJobsHeavyWeightThreads = true;
bool timeTraceEnabled;
bool tocOptimize;
bool undefinedVersion;
@@ -244,7 +245,7 @@
unsigned ltoPartitions;
unsigned ltoo;
unsigned optimize;
- unsigned thinLTOJobs;
+ unsigned thinLTOJobs = 0;
unsigned timeTraceGranularity;
int32_t splitStackAdjustSize;
Index: lld/COFF/LTO.cpp
===================================================================
--- lld/COFF/LTO.cpp
+++ lld/COFF/LTO.cpp
@@ -102,8 +102,12 @@
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 {
+ ThreadPoolStrategy S =
+ config->thinLTOJobsHeavyWeightThreads
+ ? llvm::heavyweight_hardware_concurrency(config->thinLTOJobs)
+ : llvm::hardware_concurrency();
+ backend = lto::createInProcessThinBackend(S);
}
ltoObj = std::make_unique<lto::LTO>(createConfig(), backend,
Index: lld/COFF/Driver.cpp
===================================================================
--- lld/COFF/Driver.cpp
+++ lld/COFF/Driver.cpp
@@ -1416,8 +1416,10 @@
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 (jobs == "all")
+ config->thinLTOJobsHeavyWeightThreads = false;
+ else if (jobs.getAsInteger(10, config->thinLTOJobs) ||
+ config->thinLTOJobs == 0)
error("/opt:lldltojobs: invalid job count: " + jobs);
} else if (s.startswith("lldltopartitions=")) {
StringRef n = s.substr(17);
Index: lld/COFF/Config.h
===================================================================
--- lld/COFF/Config.h
+++ lld/COFF/Config.h
@@ -230,6 +230,8 @@
bool swaprunNet = false;
bool thinLTOEmitImportsFiles;
bool thinLTOIndexOnly;
+ // Used for /opt:lldltojobs=all
+ bool thinLTOJobsHeavyWeightThreads = true;
};
extern Configuration *config;
Index: clang/lib/Driver/ToolChains/Darwin.cpp
===================================================================
--- clang/lib/Driver/ToolChains/Darwin.cpp
+++ clang/lib/Driver/ToolChains/Darwin.cpp
@@ -605,8 +605,8 @@
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)));
}
Index: clang/lib/Driver/ToolChains/CommonArgs.h
===================================================================
--- clang/lib/Driver/ToolChains/CommonArgs.h
+++ clang/lib/Driver/ToolChains/CommonArgs.h
@@ -88,7 +88,8 @@
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);
Index: clang/lib/Driver/ToolChains/CommonArgs.cpp
===================================================================
--- clang/lib/Driver/ToolChains/CommonArgs.cpp
+++ clang/lib/Driver/ToolChains/CommonArgs.cpp
@@ -338,14 +338,17 @@
}
}
-unsigned tools::getLTOParallelism(const ArgList &Args, const Driver &D) {
+llvm::StringRef tools::getLTOParallelism(const ArgList &Args, const Driver &D) {
unsigned Parallelism = 0;
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 (StringRef(LtoJobsArg->getValue()) == "all")
+ return LtoJobsArg->getValue();
+ if (StringRef(LtoJobsArg->getValue()).getAsInteger(10, Parallelism))
+ 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 +413,8 @@
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)));
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
