提交 bafe00cc 编写于 作者: M Martin Schwidefsky 提交者: Linus Torvalds

[PATCH] s390: fix in-user atomic futex operation.

From: Martin Schwidefsky <schwidefsky@de.ibm.com>

__futex_atomic_op needs to do an atomic operation in the user address space,
not the kernel address space.  Add the missing sacf 256/sacf 0 to switch to
the secondary mode before doing the compare-and-swap.  In addition add
another fixup for catch specification exceptions if the compare-and-swap
address is not aligned.
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
上级 71601e2b
......@@ -11,23 +11,24 @@
#define __futex_atomic_fixup \
".section __ex_table,\"a\"\n" \
" .align 4\n" \
" .long 0b,2b,1b,2b\n" \
" .long 0b,4b,2b,4b,3b,4b\n" \
".previous"
#else /* __s390x__ */
#define __futex_atomic_fixup \
".section __ex_table,\"a\"\n" \
" .align 8\n" \
" .quad 0b,2b,1b,2b\n" \
" .quad 0b,4b,2b,4b,3b,4b\n" \
".previous"
#endif /* __s390x__ */
#define __futex_atomic_op(insn, ret, oldval, newval, uaddr, oparg) \
asm volatile(" l %1,0(%6)\n" \
"0: " insn \
" cs %1,%2,0(%6)\n" \
"1: jl 0b\n" \
asm volatile(" sacf 256\n" \
"0: l %1,0(%6)\n" \
"1: " insn \
"2: cs %1,%2,0(%6)\n" \
"3: jl 1b\n" \
" lhi %0,0\n" \
"2:\n" \
"4: sacf 0\n" \
__futex_atomic_fixup \
: "=d" (ret), "=&d" (oldval), "=&d" (newval), \
"=m" (*uaddr) \
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册