2.6.28-stable review patch. If anyone has any objections, please let us know. ------------------ From: Paul Mackerras upstream commit: 306a82881b14d950d59e0b59a55093a07d82aa9a Richard Henderson pointed out that the powerpc __futex_atomic_op has a bug: it will write the wrong value if the stwcx. fails and it has to retry the lwarx/stwcx. loop, since 'oparg' will have been overwritten by the result from the first time around the loop. This happens because it uses the same register for 'oparg' (an input) as it uses for the result. This fixes it by using separate registers for 'oparg' and 'ret'. Cc: stable@kernel.org Signed-off-by: Paul Mackerras Signed-off-by: Chris Wright Signed-off-by: Greg Kroah-Hartman --- arch/powerpc/include/asm/futex.h | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) --- a/arch/powerpc/include/asm/futex.h +++ b/arch/powerpc/include/asm/futex.h @@ -27,7 +27,7 @@ PPC_LONG "1b,4b,2b,4b\n" \ ".previous" \ : "=&r" (oldval), "=&r" (ret) \ - : "b" (uaddr), "i" (-EFAULT), "1" (oparg) \ + : "b" (uaddr), "i" (-EFAULT), "r" (oparg) \ : "cr0", "memory") static inline int futex_atomic_op_inuser (int encoded_op, int __user *uaddr) @@ -47,19 +47,19 @@ static inline int futex_atomic_op_inuser switch (op) { case FUTEX_OP_SET: - __futex_atomic_op("", ret, oldval, uaddr, oparg); + __futex_atomic_op("mr %1,%4\n", ret, oldval, uaddr, oparg); break; case FUTEX_OP_ADD: - __futex_atomic_op("add %1,%0,%1\n", ret, oldval, uaddr, oparg); + __futex_atomic_op("add %1,%0,%4\n", ret, oldval, uaddr, oparg); break; case FUTEX_OP_OR: - __futex_atomic_op("or %1,%0,%1\n", ret, oldval, uaddr, oparg); + __futex_atomic_op("or %1,%0,%4\n", ret, oldval, uaddr, oparg); break; case FUTEX_OP_ANDN: - __futex_atomic_op("andc %1,%0,%1\n", ret, oldval, uaddr, oparg); + __futex_atomic_op("andc %1,%0,%4\n", ret, oldval, uaddr, oparg); break; case FUTEX_OP_XOR: - __futex_atomic_op("xor %1,%0,%1\n", ret, oldval, uaddr, oparg); + __futex_atomic_op("xor %1,%0,%4\n", ret, oldval, uaddr, oparg); break; default: ret = -ENOSYS; -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/