Add "memory" as a clobber for bswap inline assembly.

This had been causing Camellia (the only cipher that uses these
macros) to fail when compiling "out-of-the-box" with gcc version
"4.3.3-5ubuntu4".  I think because the compiler had no idea any memory
access was going on in these macros.

Adding "memory" as a clobber solves the problem, but is probably
overkill.  I suspect that if we specify the constraint for y
differently, we could get rid of both "memory" and __volatile__, which
would allow the compiler to optimize much more.

Also, in gcc versions that support it, we should probably use the
bswap builtins instead.
This commit is contained in:
Patrick Pelletier 2011-09-25 17:11:25 -07:00 committed by Steffen Jaeckel
parent ee7c031ddf
commit cefff85550

View File

@ -105,13 +105,13 @@ asm __volatile__ ( \
"bswapq %0 \n\t" \
"movq %0,(%1)\n\t" \
"bswapq %0 \n\t" \
::"r"(x), "r"(y));
::"r"(x), "r"(y): "memory");
#define LOAD64H(x, y) \
asm __volatile__ ( \
"movq (%1),%0\n\t" \
"bswapq %0\n\t" \
:"=r"(x): "r"(y));
:"=r"(x): "r"(y): "memory");
#else