public inbox for libstdc++-cvs@sourceware.org
help / color / mirror / Atom feed
* [gcc r14-7146] libstdc++: Prefer posix_memalign for aligned-new [PR113258]
@ 2024-01-11 17:55 Jonathan Wakely
  0 siblings, 0 replies; only message in thread
From: Jonathan Wakely @ 2024-01-11 17:55 UTC (permalink / raw)
  To: gcc-cvs, libstdc++-cvs

https://gcc.gnu.org/g:f50f2efae9fb0965d8ccdb62cfdb698336d5a933

commit r14-7146-gf50f2efae9fb0965d8ccdb62cfdb698336d5a933
Author: Jonathan Wakely <jwakely@redhat.com>
Date:   Tue Jan 9 15:22:46 2024 +0000

    libstdc++: Prefer posix_memalign for aligned-new [PR113258]
    
    As described in PR libstdc++/113258 there are old versions of tcmalloc
    which replace malloc and related APIs, but do not repalce aligned_alloc
    because it didn't exist at the time they were released. This means that
    when operator new(size_t, align_val_t) uses aligned_alloc to obtain
    memory, it comes from libc's aligned_alloc not from tcmalloc. But when
    operator delete(void*, size_t, align_val_t) uses free to deallocate the
    memory, that goes to tcmalloc's replacement version of free, which
    doesn't know how to free it.
    
    If we give preference to the older posix_memalign instead of
    aligned_alloc then we're more likely to use a function that will be
    compatible with the replacement version of free. Because posix_memalign
    has been around for longer, it's more likely that old third-party malloc
    replacements will also replace posix_memalign alongside malloc and free.
    
    libstdc++-v3/ChangeLog:
    
            PR libstdc++/113258
            * libsupc++/new_opa.cc: Prefer to use posix_memalign if
            available.

Diff:
---
 libstdc++-v3/libsupc++/new_opa.cc | 26 +++++++++++++++-----------
 1 file changed, 15 insertions(+), 11 deletions(-)

diff --git a/libstdc++-v3/libsupc++/new_opa.cc b/libstdc++-v3/libsupc++/new_opa.cc
index 8326b7497fe..35606e1c1b3 100644
--- a/libstdc++-v3/libsupc++/new_opa.cc
+++ b/libstdc++-v3/libsupc++/new_opa.cc
@@ -46,12 +46,12 @@ using std::bad_alloc;
 using std::size_t;
 extern "C"
 {
-# if _GLIBCXX_HAVE_ALIGNED_ALLOC
+# if _GLIBCXX_HAVE_POSIX_MEMALIGN
+  void *posix_memalign(void **, size_t alignment, size_t size);
+# elif _GLIBCXX_HAVE_ALIGNED_ALLOC
   void *aligned_alloc(size_t alignment, size_t size);
 # elif _GLIBCXX_HAVE__ALIGNED_MALLOC
   void *_aligned_malloc(size_t size, size_t alignment);
-# elif _GLIBCXX_HAVE_POSIX_MEMALIGN
-  void *posix_memalign(void **, size_t alignment, size_t size);
 # elif _GLIBCXX_HAVE_MEMALIGN
   void *memalign(size_t alignment, size_t size);
 # else
@@ -63,13 +63,10 @@ extern "C"
 #endif
 
 namespace __gnu_cxx {
-#if _GLIBCXX_HAVE_ALIGNED_ALLOC
-using ::aligned_alloc;
-#elif _GLIBCXX_HAVE__ALIGNED_MALLOC
-static inline void*
-aligned_alloc (std::size_t al, std::size_t sz)
-{ return _aligned_malloc(sz, al); }
-#elif _GLIBCXX_HAVE_POSIX_MEMALIGN
+// Prefer posix_memalign if available, because it's older than aligned_alloc
+// and so more likely to be provided by replacement malloc libraries that
+// predate the addition of aligned_alloc. See PR libstdc++/113258.
+#if _GLIBCXX_HAVE_POSIX_MEMALIGN
 static inline void*
 aligned_alloc (std::size_t al, std::size_t sz)
 {
@@ -83,6 +80,12 @@ aligned_alloc (std::size_t al, std::size_t sz)
     return ptr;
   return nullptr;
 }
+#elif _GLIBCXX_HAVE_ALIGNED_ALLOC
+using ::aligned_alloc;
+#elif _GLIBCXX_HAVE__ALIGNED_MALLOC
+static inline void*
+aligned_alloc (std::size_t al, std::size_t sz)
+{ return _aligned_malloc(sz, al); }
 #elif _GLIBCXX_HAVE_MEMALIGN
 static inline void*
 aligned_alloc (std::size_t al, std::size_t sz)
@@ -128,7 +131,8 @@ operator new (std::size_t sz, std::align_val_t al)
   if (__builtin_expect (sz == 0, false))
     sz = 1;
 
-#if _GLIBCXX_HAVE_ALIGNED_ALLOC
+#if _GLIBCXX_HAVE_POSIX_MEMALIGN
+#elif _GLIBCXX_HAVE_ALIGNED_ALLOC
 # if defined _AIX || defined __APPLE__
   /* AIX 7.2.0.0 aligned_alloc incorrectly has posix_memalign's requirement
    * that alignment is a multiple of sizeof(void*).

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2024-01-11 17:55 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2024-01-11 17:55 [gcc r14-7146] libstdc++: Prefer posix_memalign for aligned-new [PR113258] Jonathan Wakely

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).