projects
/
ghc-hetmet.git
/ commitdiff
commit
grep
author
committer
pickaxe
?
search:
re
summary
|
shortlog
|
log
|
commit
| commitdiff |
tree
raw
|
patch
| inline |
side by side
(parent:
e3029b1
)
NCG: Use sync instead of msync for a memory barrier for powerpc
author
Ben.Lippmeier@anu.edu.au
<unknown>
Fri, 13 Feb 2009 00:49:10 +0000
(
00:49
+0000)
committer
Ben.Lippmeier@anu.edu.au
<unknown>
Fri, 13 Feb 2009 00:49:10 +0000
(
00:49
+0000)
Darwin 9.6.0 + GCC 4.0.1 doesn't understand "msync".
I think "sync" means the same thing.
includes/SMP.h
patch
|
blob
|
history
diff --git
a/includes/SMP.h
b/includes/SMP.h
index
c851054
..
223f158
100644
(file)
--- a/
includes/SMP.h
+++ b/
includes/SMP.h
@@
-195,7
+195,7
@@
store_load_barrier(void) {
#elif x86_64_HOST_ARCH
__asm__ __volatile__ ("lock; addq $0,0(%%rsp)" : : : "memory");
#elif powerpc_HOST_ARCH
- __asm__ __volatile__ ("msync" : : : "memory");
+ __asm__ __volatile__ ("sync" : : : "memory");
#elif sparc_HOST_ARCH
/* Sparc in TSO mode does not require write/write barriers. */
__asm__ __volatile__ ("membar" : : : "memory");