On Sun, Jul 12, 2015 at 11:41 PM, Mobile Mouse <[email protected]> wrote:
> Your patch works. Compiled with GCC-5.1 (macports version, in turn built by
> Xcode-6.4 on Yosemite), "cryptest.exe v” now succeeds.
>
Yeah, at -O3, GCC uses SSE instructions and vectorizes the XOR. The
vectoriztion includes vmovdqu, vinsertf128, vmovdqa, and vxorps.
The problem appears to be, vmovdqa has a 128-bit alignment
requirement, but the code only provides a 64-bit alignment guarantee.
I'm researching how we can make it play well with GCC now and moving forward.
*****
void xorbuf(byte *buf, const byte *mask, size_t count)
{
size_t i;
if (IsAligned<word32>(buf) && IsAligned<word32>(mask))
{
if (!CRYPTOPP_BOOL_SLOW_WORD64 && IsAligned<word64>(buf) &&
IsAligned<word64>(mask))
{
for (i=0; i<count/8; i++)
((word64*)buf)[i] ^= ((word64*)mask)[i];
count -= 8*i;
if (!count)
return;
buf += 8*i;
mask += 8*i;
}
for (i=0; i<count/4; i++)
((word32*)buf)[i] ^= ((word32*)mask)[i];
count -= 4*i;
if (!count)
return;
buf += 4*i;
mask += 4*i;
}
for (i=0; i<count; i++)
buf[i] ^= mask[i];
}
--
--
You received this message because you are subscribed to the "Crypto++ Users"
Google Group.
To unsubscribe, send an email to [email protected].
More information about Crypto++ and this group is available at
http://www.cryptopp.com.
---
You received this message because you are subscribed to the Google Groups
"Crypto++ Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
