summaryrefslogtreecommitdiff
path: root/src
Commit message (Expand)AuthorAgeFilesLines
...
* Fix store-load ordering in AO_stack_pop_explicit_aux_acquire (PowerPC)Hans Boehm2016-08-151-1/+2
* gcc/generic-small.h: Add TODO itemsHans Boehm2016-05-122-5/+60
* Basic support of TILE-Gx and TILEPro CPUsChris Metcalf2016-05-113-0/+54
* Use GCC atomic intrinsics for x86 and x64 (gcc 4.8+ and clang 3.4+)Ivan Maidanski2016-04-253-22/+66
* Do not force GCC atomic intrinsics usage if AO_PREFER_BUILTIN_ATOMICSIvan Maidanski2016-04-212-6/+2
* Remove AO_T_IS_INT definition for GCC atomic intrinsics caseIvan Maidanski2016-04-183-9/+4
* Fix AO_TS_VAL_t used for __atomic_test_and_set (gcc)Ivan Maidanski2016-04-156-21/+8
* Fix typo in gcc/arm.h (ARMv6)Ivan Maidanski2016-04-151-1/+1
* Avoid __atomic_and/or/xor_fetch if unsupported (clang/arm-v5te)atomics-expIvan Maidanski2016-04-143-338/+448
* Use GCC atomic intrinsics for ARM (gcc 4.8+ and clang 3.5+)Ivan Maidanski2016-04-123-85/+151
* Use GCC atomic intrinsics for MIPS (GCC 4.9+ and clang 3.5+)Ivan Maidanski2016-04-081-14/+38
* Implement CAS_acquire/release/full using __atomic_compare_exchange_n (gcc)Ivan Maidanski2016-04-083-31/+404
* Avoid atomic_compare_exchange_n if no __GCC_HAVE_SYNC_COMPARE_AND_SWAP_nIvan Maidanski2016-04-074-96/+144
* Eliminate redundant lwsync 2nd call in CAS_full on fail (gcc/PowerPC)Ivan Maidanski2016-04-061-2/+4
* Define AO_TS_SET as __GCC_ATOMIC_TEST_AND_SET_TRUEVAL if applicableIvan Maidanski2016-04-044-6/+28
* New macro AO_PREFER_BUILTIN_ATOMICS to rely on C11 atomics fully (AArch64)Ivan Maidanski2016-04-012-17/+25
* Add initial nios2 architecture supportMarek Vasut2016-03-153-0/+21
* Initial ibmc/powerpc (xlc) supportFabrizio Fabbri2016-03-041-8/+63
* Eliminate 'signed-to-unsigned value extension' compiler warning in AO_mallocIvan Maidanski2016-02-081-5/+7
* Support NaCl/armIvan Maidanski2015-10-141-12/+71
* Fix GCC 5.x compatibility for AArch64 double-wide primitivesFrank Schaefer2015-06-141-2/+2
* Support n32 ABI for mips64Frank Schaefer2015-05-131-1/+1
* Minor fix of code alignment in mips AO_compare_and_swapJames Cowgill2015-01-081-1/+1
* Remove inclusion of acquire_release_volatile.h on mipsJames Cowgill2015-01-081-2/+0
* Use LLD and SCD instructions on mips64James Cowgill2015-01-081-24/+30
* Relax shareability domain for dmb st in AO_nop_write (ARM/AArch64)Steve Capper2014-10-302-2/+12
* Fix missing output folder on making auto-generated headers (Automake)Kochin Chang2014-10-251-0/+19
* Fix makefile preventing AO_pause undefined in libatomic_ops_gplPavel Raiskup2014-05-181-0/+1
* Restore contribution info in ChangeLog for authors not listed in git logIvan Maidanski2014-05-041-3/+0
* Bump atomic_ops versionIvan Maidanski2014-01-311-2/+2
* [7.4.0]libatomic_ops-7_4_0Ivan Maidanski2013-11-171-2/+9
* Do not use LDREXD/STREXD for Clang3.2/arm (and earlier)Ivan Maidanski2013-08-171-4/+8
* Fix ARM char/short fetch_and_add and double-CAS operands width (GCC/Clang)Ivan Maidanski2013-08-171-8/+8
* Fix ARMv7 LDREXD/STREXD double-wide operand specification (GCC/Clang)Ivan Maidanski2013-08-171-5/+5
* Support ARMv8 target (gcc/arm)Ivan Maidanski2013-08-171-2/+6
* Fix typo in comment of generalize-small.templateIvan Maidanski2013-07-302-5/+5
* Add comment about store_full implementation for x86Ivan Maidanski2013-05-261-0/+4
* Add non-generalized double-CAS-full implementation for AArch64Ivan Maidanski2013-05-261-1/+24
* Enable build using toolchain without pthreadsGilles Talis2013-05-261-6/+7
* Add comment about double-wide load/store on x86_64 (GCC)Ivan Maidanski2013-04-081-1/+12
* Fix generalized store_release_write for double-wide type on x64 (GCC)Ivan Maidanski2013-04-082-42/+42
* Generalize double_load/store for x86_64 (GCC)Ivan Maidanski2013-03-291-0/+6
* Fix fetch_compare_and_swap asm constraints in gcc/x86.hIvan Maidanski2013-03-291-1/+1
* Reformat asm instructions of sunc/x86.hIvan Maidanski2013-03-291-13/+19
* Fix asm constraints of primitives in sunc/x86.hIvan Maidanski2013-03-291-18/+18
* Merge branch 'add-aarch64-support'Ivan Maidanski2013-03-1611-3/+1487
|\
| * Always use 'mfence' for nop_full if target CPU supports SSE2 (gcc/x86)add-aarch64-supportIvan Maidanski2013-03-091-3/+2
| * Fix assembly comment prefix and identify memory clobbers for AArch64Yvan Roux2013-03-091-44/+35
| * Use AO_EXPECT_FALSE for AArch64 double_load/storeIvan Maidanski2013-03-051-14/+20
| * Workaround missing built-in double-wide primitives for AArch64 in gcc-4.7Yvan Roux2013-03-052-40/+199
View Lorry Controller Status