glibc/sysdeps
Chris Metcalf a329844ff8 Make sure tilepro uses kernel atomics fo atomic_store
It's not legal for raw stores to be mixed with atomic operations
on tilepro, since the atomics are managed by kernel fast syscalls.
It's possible for a hardware store and a kernel fast atomic to race
with each other in such a way that the hardware store is lost.

Suppose you have an initial zero value, and you race with a store
of 2 and a kernel cmpxchg from 0 to 1.  The legal output is only 2:
either the store hit first and the cmpxchg failed, or the cmpxchg
hit first and succeeded, then was overwritten by the 2.  But if
the kernel cmpxchg starts first and loads the zero, then the store
hits and sets the value to 2, the cmpxchg will still decide it was
successful and write the 1, leaving the value illegally set to 1.

Using atomic_exchange variants to implement atomic_store fixes this
problem for tilepro.
2016-11-10 20:08:24 -05:00
..
aarch64 Regenerate ULPs for aarch64 2016-11-10 16:52:35 +05:30
alpha Fix alpha sqrt fegetenv namespace (bug 20768). 2016-11-04 17:19:13 +00:00
arm nptl: Document the reason why __kind in pthread_mutex_t is part of the ABI 2016-11-07 20:24:32 +01:00
generic Refactor some libm type-generic macros. 2016-11-10 21:41:56 +00:00
gnu
hppa nptl: Document the reason why __kind in pthread_mutex_t is part of the ABI 2016-11-07 20:24:32 +01:00
i386
ia64 nptl: Document the reason why __kind in pthread_mutex_t is part of the ABI 2016-11-07 20:24:32 +01:00
ieee754 Refactor some libm type-generic macros. 2016-11-10 21:41:56 +00:00
init_array
m68k nptl: Document the reason why __kind in pthread_mutex_t is part of the ABI 2016-11-07 20:24:32 +01:00
mach
microblaze nptl: Document the reason why __kind in pthread_mutex_t is part of the ABI 2016-11-07 20:24:32 +01:00
mips nptl: Document the reason why __kind in pthread_mutex_t is part of the ABI 2016-11-07 20:24:32 +01:00
nacl
nios2 nptl: Document the reason why __kind in pthread_mutex_t is part of the ABI 2016-11-07 20:24:32 +01:00
nptl
posix
powerpc
pthread
s390 nptl: Document the reason why __kind in pthread_mutex_t is part of the ABI 2016-11-07 20:24:32 +01:00
sh nptl: Document the reason why __kind in pthread_mutex_t is part of the ABI 2016-11-07 20:24:32 +01:00
sparc nptl: Document the reason why __kind in pthread_mutex_t is part of the ABI 2016-11-07 20:24:32 +01:00
tile Make sure tilepro uses kernel atomics fo atomic_store 2016-11-10 20:08:24 -05:00
unix Fix sh4 build with __ASSUME_ST_INO_64_BIT redefinition 2016-11-10 18:03:37 -02:00
wordsize-32
wordsize-64
x86 nptl: Document the reason why __kind in pthread_mutex_t is part of the ABI 2016-11-07 20:24:32 +01:00
x86_64