This one time, at band camp, Falk Hueffner said: > Hi, > > is there a problem with this patch? Or why hasn't it been applied?
Well, this part looks good:
> --- libclamav/others.c~ 2005-04-20 01:33:17.000000000 +0200
> +++ libclamav/others.c 2005-05-07 15:45:11.000000000 +0200
> -#if WORDS_BIGENDIAN == 0
> - ret = *(int32_t *) buff;
> +#if WORDS_BIGENDIAN == 0 && defined(__GNUC__)
> + ret = ((struct { int32_t x; } __attribute__((packed)) *) buff)->x;
But I am not so sure about this one:
> --- libclamav/upx.c~ 2005-04-27 23:53:57.000000000 +0200
> +++ libclamav/upx.c 2005-05-07 16:14:46.000000000 +0200
> -#if WORDS_BIGENDIAN == 1
> +#ifndef __i386__
> -#if WORDS_BIGENDIAN == 0
> +#ifdef __i386__
Other architectures besides i386 are little endian, and I'm not sure
just swapping them is the right approach here. If the problem to be
fixed is on one arch or 32 vs. 64 bit only, probably that should be
tested for instead.
Take care,
--
-----------------------------------------------------------------
| ,''`. Stephen Gran |
| : :' : [EMAIL PROTECTED] |
| `. `' Debian user, admin, and developer |
| `- http://www.debian.org |
-----------------------------------------------------------------
signature.asc
Description: Digital signature
