about summary refs log tree commit diff
Commit message (Collapse)AuthorAgeFilesLines
* SSE4.2 strstr/strcasestr for x86-64.H.J. Lu2009-07-208-5/+551
| | | | | This patch implements SSE4.2 strstr/strcasestr, using Knuth-Morris-Pratt string searching algorithm.
* Use correct index for _NL_CTYPE_NONASCII_CASE.Ulrich Drepper2009-07-202-5/+5
|
* Check generated locale for non-ASCII 8-bit characters with case conversion.Ulrich Drepper2009-07-205-4/+47
| | | | | | If a locale does not have 8-bit characters with case conversion which are different from the ASCII conversion (±0x20) then we can perform some optimizations. These will follow later.
* Minor optimizations of last x86-64 condvar changes.Ulrich Drepper2009-07-203-23/+22
|
* Support requeueing for condvars using PI mutex. x86-64 only.Ulrich Drepper2009-07-196-36/+177
| | | | | Add support for the new FUTEX_WAIT_REQUEUE_PI and FUTEX_CMP_REQUEUE_PI options of futex.
* Make x86-64 pthread_cond_timedwait more robust.Ulrich Drepper2009-07-192-0/+7
| | | | | | It just happens that __pthread_enable_asynccancel doesn't modify the $rdi register. But this isn't guaranteed. Hence we reload the register after the calls.
* Extend x86-64 __lll_robust_timedlock_wait to use futex syscall with absolute ↵Ulrich Drepper2009-07-192-2/+80
| | | | timeout.
* Add more sem_timedwait tests.Ulrich Drepper2009-07-183-1/+20
|
* Extend x86-64 pthread_rwlock_timedrdlock to use futex syscall with absolute ↵Ulrich Drepper2009-07-182-18/+57
| | | | timeout.
* Pretty printing last change.Ulrich Drepper2009-07-181-6/+6
|
* Extend x86-64 pthread_rwlock_timedwrlock to use futex syscall with absolute ↵Ulrich Drepper2009-07-182-12/+53
| | | | timeout.
* Remove leftover cfi.Ulrich Drepper2009-07-181-4/+1
|
* Extend pthread_cond_timedwait tests.Ulrich Drepper2009-07-182-1/+18
|
* Extend x86-64 pthread_cond_timedwait to use futex syscall with absolute timeout.Ulrich Drepper2009-07-182-126/+296
|
* Optimize x86-64 pthread_cond_timedwait.Ulrich Drepper2009-07-182-144/+190
| | | | | Instead of actively registering an unwind buffer we now use the exception handling functionality of the gcc runtime.
* Optimize x86-64 pthread_cond_wait.Ulrich Drepper2009-07-183-144/+315
| | | | | Instead of actively registering an unwind buffer we now use the exception handling functionality of the gcc runtime.
* Extend x86-64 sem_timedwait to use futex syscall with absolute timeout.Ulrich Drepper2009-07-172-46/+120
|
* Consistently use symbolic constants.Ulrich Drepper2009-07-171-4/+3
|
* Replace hand-coded unwind tables from x86-64 sem_timedwait.Ulrich Drepper2009-07-172-122/+31
|
* Optimize x86-64 sem_wait for uncontested semaphore.Ulrich Drepper2009-07-172-24/+30
|
* Replace hand-coded unwind tables from x86-64 sem_wait.Ulrich Drepper2009-07-172-101/+26
|
* Replace hand-coded unwind tables from x86-64 pthread_once.Ulrich Drepper2009-07-174-113/+65
|
* Replace hand-coded unwind tables from x86-64 pthread_cond_wait.Ulrich Drepper2009-07-172-76/+24
|
* Revert "Fix lock handling in memory hander of nscd."Ulrich Drepper2009-07-178-16/+27
| | | | | | | | This reverts commit 137028b4d7e50f71906c1656c27079eac5a1d085. Conflicts: ChangeLog
* Add missing BZ number in ChangeLog.Ulrich Drepper2009-07-171-0/+1
|
* Handle overly large answer buffers in resolver.Ulrich Drepper2009-07-162-1/+7
| | | | | | In EDNS0 records the maximum result size is transmitted in a 16 bit value. Large buffer sizes were handled incorrectly by using only the low 16 bits. Fix this by limiting the size to 0xffff.
* Fix lock handling in memory hander of nscd.Petr Baudis2009-07-168-27/+16
| | | | | | | | | | The commit 20e498bd removes the pthread_mutex_rdlock() calls, but not the corresponding pthread_mutex_unlock() calls. Also, the database lock is never unlocked in one branch of the mempool_alloc() if. I think unreproducible random assert(dh->usable) crashes in prune_cache() were caused by this. But an easy way to make nscd threads hang with the broken locking was.
* Use correct release semantic in list update.Andreas Schwab2009-07-162-1/+7
| | | | | nscd uses lockfree lists and we need to ensure the correct release semantics is used when adding to the list.
* Fix race in corruption check.Ulrich Drepper2009-07-162-2/+29
| | | | | | With atomic fastbins the checks performed can race with concurrent modifications of the arena. If we detect a problem re-do the test after getting the lock.
* Use rel semantics of cas instead of acq semantics with full barrier before ↵Jakub Jelinek2009-07-163-13/+35
| | | | | | | | | | | it in _int_free The following patch fixes catomic_compare_and_exchange_*_rel definitions (which were never used and weren't correct) and uses catomic_compare_and_exchange_val_rel in _int_free. Comparing to the pre-2009-07-02 --enable-experimental-malloc state the generated code should be identical on all arches other than ppc/ppc64 and on ppc/ppc64 should use lwsync instead of isync barrier.
* Remove warning and little optimization.Ulrich Drepper2009-07-163-0/+9
| | | | | The prototype for _dl_higher_prime_number was missing. While at it, the function is now marked with internal_function.
* Optimize restoring of ymm registers on x86-64.Ulrich Drepper2009-07-162-43/+39
| | | | The patch mainly reduces the code size but also avoids some jumps.
* Fix up whitespaces in new memcmp for x86-64.Ulrich Drepper2009-07-161-42/+42
|
* memcmp implementation for x86-64 using SSE2.H.J. Lu2009-07-162-0/+363
|
* Fix thinko in AVX audit patch.Ulrich Drepper2009-07-151-20/+4
| | | | Don't use AVX instructions too often.
* Fix typo in last change.Ulrich Drepper2009-07-151-1/+1
|
* Secure AVX changes for auditing code.Ulrich Drepper2009-07-153-326/+299
| | | | | | | | The original AVX patch used a function pointer to handle the difference between machines with and without AVX support. This is insecure. A well-placed memory exploit could lead to redirection of the execution. Using a variable and several tests is a bit slower but cannot be exploited in this way.
* Fix build issue with modules for audit test on machines != x86-64.Ulrich Drepper2009-07-152-3/+10
|
* Add AVX support to ld.so auditing for x86-64.H.J. Lu2009-07-1019-126/+1018
|
* Remove do-lookup.h.Ulrich Drepper2009-07-104-388/+369
| | | | | It is not necessary/useful anymore to have the content separate from dl-lookup.c.
* Fix comment in Linux's <sys/epoll.h>.Ulrich Drepper2009-07-102-2/+6
|
* Implement STB_GNU_UNIQUE handling.Ulrich Drepper2009-07-0916-12/+375
| | | | | | | | | Some symbols have to be identified process-wide by their name. This is particularly important for some C++ features (e.g., class local static data and static variables in inline functions). This cannot completely be implemented with ELF functionality so far. The STB_GNU_UNIQUE binding helps by ensuring the dynamic linker will always use the same definition for all symbols with the same name and this binding.
* Define STB_GNU_UNIQUE.Ulrich Drepper2009-07-072-0/+3
|
* Clean up code for hash table handling in ld.so.Ulrich Drepper2009-07-073-69/+74
|
* Optimize test for valid ELF symbol types in lookup function. ibm/masterUlrich Drepper2009-07-062-7/+11
|
* Fix wrong PPC_FEATURE_* values.Andreas Schwab2009-07-052-2/+7
| | | | | Nothing uses these wrong values yet, but it fixes a warning due to conflicting definitions in <asm/cputable.h>.
* Minor cleanups in recently added files.Ulrich Drepper2009-07-033-79/+60
|
* Align functions to 16-byte boundary.Ulrich Drepper2009-07-035-0/+12
| | | | | | Some of the new multi-arch string functions for x86-64 were not aligned to 16 byte boundarie,s possibly creating unnecessary cache line misses and delays.
* Add SSE4.2 support for strcspn, strpbrk, and strspn on x86-64.H.J. Lu2009-07-0314-7/+875
|
* Fix whitespace in last patch.Ulrich Drepper2009-07-022-2/+2
|