comphelper/qa/unit/threadpooltest.cxx | 5 +++-- comphelper/source/misc/threadpool.cxx | 19 +++++++++++-------- include/comphelper/threadpool.hxx | 5 +++-- include/o3tl/safeint.hxx | 10 ++++++++++ package/inc/ZipOutputStream.hxx | 3 ++- package/source/zipapi/ZipOutputStream.cxx | 6 +++--- package/source/zippackage/ZipPackageStream.cxx | 3 ++- 7 files changed, 34 insertions(+), 17 deletions(-)
New commits: commit b4163877e722298b42f6e96831bd8ffef7785a20 Author: Stephan Bergmann <sberg...@redhat.com> AuthorDate: Thu Jun 2 15:14:56 2022 +0200 Commit: Stephan Bergmann <sberg...@redhat.com> CommitDate: Thu Jun 2 17:13:17 2022 +0200 Use more appropriate return type for ThreadPool::getPreferredConcurrency All call sites already effectively asked for an unsigned return type, including: * The ThreadPool ctor took an nWorkers argument of type sal_Int32, but internally stores that as std::size_t mnMaxWorkers. * ZipOutputStream::reduceScheduledThreadTasksToGivenNumberOrLess apparently benefits from an unsigned nThreadTasks parameter, getting rid of various casts in its implementation that were necessary to silence signed vs. unsigned comparison warnings. The only drawback is that comphelper::ThreadPool::getPreferredConcurrency() * 4 in package/source/zippackage/ZipPackageStream.cxx would now silently wrap around instead of causing UB on overflow (which could be detected with appropriate tools). Ideally, it would use some o3tl::saturating_mul if we had that, so add a TODO comment for now. While std::thread::hardware_concurrency returns unsigned, it looked more natural to go with std::size_t here, as some call sites already used that (see above), so the implementation of ThreadPool::getPreferredConcurrency is a natural place to hide clamping std::thread::hardware_concurrency() to std::size_t (in the unlikely case that std::size_t is of smaller rank than unsigned). This required addition of o3tl::clamp_to_unsigned in o3tl/safeint.hxx. Change-Id: I0a04a8b32e63ebfeb39f924c4b38520455a6fb38 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/135309 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sberg...@redhat.com> diff --git a/comphelper/qa/unit/threadpooltest.cxx b/comphelper/qa/unit/threadpooltest.cxx index 14da59988ef4..13eaf210a18b 100644 --- a/comphelper/qa/unit/threadpooltest.cxx +++ b/comphelper/qa/unit/threadpooltest.cxx @@ -17,6 +17,7 @@ #include <stdlib.h> #include <atomic> +#include <cstddef> #include <thread> #include <mutex> @@ -42,7 +43,7 @@ void ThreadPoolTest::testPreferredConcurrency() { // Check default. auto nThreads = comphelper::ThreadPool::getPreferredConcurrency(); - sal_Int32 nExpected = 4; // UTs are capped to 4. + std::size_t nExpected = 4; // UTs are capped to 4. CPPUNIT_ASSERT_MESSAGE("Expected no more than 4 threads", nExpected >= nThreads); #ifndef _WIN32 @@ -51,7 +52,7 @@ void ThreadPoolTest::testPreferredConcurrency() setenv("MAX_CONCURRENCY", std::to_string(nThreads).c_str(), true); nThreads = comphelper::ThreadPool::getPreferredConcurrency(); CPPUNIT_ASSERT_MESSAGE("Expected no more than hardware threads", - nThreads <= static_cast<sal_Int32>(std::thread::hardware_concurrency())); + nThreads <= std::thread::hardware_concurrency()); // Revert and check. Again, nothing should change. unsetenv("MAX_CONCURRENCY"); diff --git a/comphelper/source/misc/threadpool.cxx b/comphelper/source/misc/threadpool.cxx index 00fee7d8eb59..f0a71eb05168 100644 --- a/comphelper/source/misc/threadpool.cxx +++ b/comphelper/source/misc/threadpool.cxx @@ -11,6 +11,7 @@ #include <com/sun/star/uno/Exception.hpp> #include <config_options.h> +#include <o3tl/safeint.hxx> #include <sal/config.h> #include <sal/log.hxx> #include <salhelper/thread.hxx> @@ -18,6 +19,7 @@ #include <memory> #include <thread> #include <chrono> +#include <cstddef> #include <comphelper/debuggerinfo.hxx> #include <utility> @@ -91,7 +93,7 @@ public: } }; -ThreadPool::ThreadPool(sal_Int32 nWorkers) +ThreadPool::ThreadPool(std::size_t nWorkers) : mbTerminate(true) , mnMaxWorkers(nWorkers) , mnBusyWorkers(0) @@ -116,7 +118,7 @@ std::shared_ptr< ThreadPool >& GetStaticThreadPool() static std::shared_ptr< ThreadPool > POOL = []() { - const sal_Int32 nThreads = ThreadPool::getPreferredConcurrency(); + const std::size_t nThreads = ThreadPool::getPreferredConcurrency(); return std::make_shared< ThreadPool >( nThreads ); }(); return POOL; @@ -129,21 +131,22 @@ ThreadPool& ThreadPool::getSharedOptimalPool() return *GetStaticThreadPool(); } -sal_Int32 ThreadPool::getPreferredConcurrency() +std::size_t ThreadPool::getPreferredConcurrency() { - static sal_Int32 ThreadCount = []() + static std::size_t ThreadCount = []() { - const sal_Int32 nHardThreads = std::max(std::thread::hardware_concurrency(), 1U); - sal_Int32 nThreads = nHardThreads; + const std::size_t nHardThreads = o3tl::clamp_to_unsigned<std::size_t>( + std::max(std::thread::hardware_concurrency(), 1U)); + std::size_t nThreads = nHardThreads; const char *pEnv = getenv("MAX_CONCURRENCY"); if (pEnv != nullptr) { // Override with user/admin preference. - nThreads = rtl_str_toInt32(pEnv, 10); + nThreads = o3tl::clamp_to_unsigned<std::size_t>(rtl_str_toInt32(pEnv, 10)); } nThreads = std::min(nHardThreads, nThreads); - return std::max<sal_Int32>(nThreads, 1); + return std::max<std::size_t>(nThreads, 1); }(); return ThreadCount; diff --git a/include/comphelper/threadpool.hxx b/include/comphelper/threadpool.hxx index a3c3091ce483..84f9dc9284f6 100644 --- a/include/comphelper/threadpool.hxx +++ b/include/comphelper/threadpool.hxx @@ -15,6 +15,7 @@ #include <comphelper/comphelperdllapi.h> #include <mutex> #include <condition_variable> +#include <cstddef> #include <vector> #include <memory> @@ -55,9 +56,9 @@ public: /// limit to avoid spawning an unnecessarily /// large number of threads on high-core boxes. /// MAX_CONCURRENCY env. var. controls the cap. - static sal_Int32 getPreferredConcurrency(); + static std::size_t getPreferredConcurrency(); - ThreadPool( sal_Int32 nWorkers ); + ThreadPool( std::size_t nWorkers ); ~ThreadPool(); /// push a new task onto the work queue diff --git a/include/o3tl/safeint.hxx b/include/o3tl/safeint.hxx index 85c61f8c33d3..027d283342b5 100644 --- a/include/o3tl/safeint.hxx +++ b/include/o3tl/safeint.hxx @@ -208,6 +208,16 @@ make_unsigned(T value) return value; } +template<typename T1, typename T2> constexpr std::enable_if_t<std::is_unsigned_v<T1>, T1> +clamp_to_unsigned(T2 value) { + if constexpr (std::is_unsigned_v<T2>) { + return value <= std::numeric_limits<T1>::max() ? value : std::numeric_limits<T1>::max(); + } else { + static_assert(std::is_signed_v<T2>); + return value < 0 ? 0 : clamp_to_unsigned<T1>(make_unsigned(value)); + } +} + // An implicit conversion from T2 to T1, useful in places where an explicit conversion from T2 to // T1 is needed (e.g., in list initialization, if the implicit conversion would be narrowing) but // tools like -fsanitize=implicit-conversion should still be able to detect truncation: diff --git a/package/inc/ZipOutputStream.hxx b/package/inc/ZipOutputStream.hxx index aeb7ab1f86e8..f55ef59a8880 100644 --- a/package/inc/ZipOutputStream.hxx +++ b/package/inc/ZipOutputStream.hxx @@ -25,6 +25,7 @@ #include "ByteChucker.hxx" #include <comphelper/threadpool.hxx> +#include <cstddef> #include <vector> struct ZipEntry; @@ -85,7 +86,7 @@ private: public: void reduceScheduledThreadTasksToGivenNumberOrLess( - sal_Int32 nThreadTasks); + std::size_t nThreadTasks); const std::shared_ptr<comphelper::ThreadTaskTag>& getThreadTaskTag() const { return mpThreadTaskTag; } }; diff --git a/package/source/zipapi/ZipOutputStream.cxx b/package/source/zipapi/ZipOutputStream.cxx index a4d67fcfd54a..33321627b6f9 100644 --- a/package/source/zipapi/ZipOutputStream.cxx +++ b/package/source/zipapi/ZipOutputStream.cxx @@ -143,13 +143,13 @@ void ZipOutputStream::consumeFinishedScheduledThreadTaskEntries() m_aEntries = aNonFinishedEntries; } -void ZipOutputStream::reduceScheduledThreadTasksToGivenNumberOrLess(sal_Int32 nThreadTasks) +void ZipOutputStream::reduceScheduledThreadTasksToGivenNumberOrLess(std::size_t nThreadTasks) { - while(static_cast< sal_Int32 >(m_aEntries.size()) > nThreadTasks) + while(m_aEntries.size() > nThreadTasks) { consumeFinishedScheduledThreadTaskEntries(); - if(static_cast< sal_Int32 >(m_aEntries.size()) > nThreadTasks) + if(m_aEntries.size() > nThreadTasks) { osl::Thread::wait(std::chrono::microseconds(100)); } diff --git a/package/source/zippackage/ZipPackageStream.cxx b/package/source/zippackage/ZipPackageStream.cxx index c4f94806b9b8..b63bf24eaf4f 100644 --- a/package/source/zippackage/ZipPackageStream.cxx +++ b/package/source/zippackage/ZipPackageStream.cxx @@ -57,6 +57,7 @@ #include <PackageConstants.hxx> #include <algorithm> +#include <cstddef> using namespace com::sun::star::packages::zip::ZipConstants; using namespace com::sun::star::packages::zip; @@ -771,7 +772,7 @@ bool ZipPackageStream::saveChild( // cores and allow 4-times the amount for having the queue well filled. The // 2nd parameter is the time to wait between cleanups in 10th of a second. // Both values may be added to the configuration settings if needed. - static sal_Int32 nAllowedTasks(comphelper::ThreadPool::getPreferredConcurrency() * 4); + static std::size_t nAllowedTasks(comphelper::ThreadPool::getPreferredConcurrency() * 4); //TODO: overflow rZipOut.reduceScheduledThreadTasksToGivenNumberOrLess(nAllowedTasks); // Start a new thread task deflating this zip entry