Simon Josefsson <[EMAIL PROTECTED]> writes:
> Ok to install patch below?
There weren't any objections, so I take that as a "yes". Installed:
> 2005-10-23 Simon Josefsson <[EMAIL PROTECTED]>
>
> * md5.h, md5.c: Simplify buffer handling visavi alignment, same as
> recent md4 change.
Stepan Kasal <[EMAIL PROTECTED]> writes:
> wouldn't the following be more readable?
Yes.
> Do I understand correctly that this is only executed for the last block,
> so that it's not necessary to be so careful about each tick?
Yes.
Paul Eggert <[EMAIL PROTECTED]> writes:
> Also, in some other
Hello,
On Sun, Oct 23, 2005 at 10:34:18PM -0700, Paul Eggert wrote:
> uint32_t *p = (uint32_t *) ((char *) ctx->buffer + bytes + pad);
> p[0] = SWAP (ctx->total[0] << 3);
> p[1] = SWAP ((ctx->total[1] << 3) | (ctx->total[0] >> 29));
wouldn't the following be more readable?
uint32_t bytes
Simon Josefsson <[EMAIL PROTECTED]> writes:
> - *(uint32_t *) &ctx->buffer[bytes + pad] = SWAP (ctx->total[0] << 3);
> - *(uint32_t *) &ctx->buffer[bytes + pad + 4] = SWAP ((ctx->total[1] << 3) |
> - (ctx->total[0] >> 29));
> + ctx->buffer[(by
How about this? Similar to the md4 fix.
Caveat: I had md5 self test failures on x86-solaris recently, but
could not figure out what the reason was. I replaced the md5.? files
with the files from coreutils 4.5.5 to see if any changes we have
applied to gnulib recently caused the problem, but I g
