Sorry, I just discovered a problem that didn't show up on initial testing. Luckily it showed up on my PPC machine.
If you're building on a big-endian machine, compilation will fail on md4.c due to missing definition of htole32. Or at least it does on Linux. Here's the patch, already comitted to CVS and will be in -pre2. Index: md4.c =================================================================== RCS file: /source/radiusd/src/lib/md4.c,v retrieving revision 1.5 diff -r1.5 md4.c 36a37,39 > * Add htole32 define from > http://www.squid-cache.org/mail-archive/squid-dev/200307/0130.html > * (The bswap32 definition in the patch.) > * This is only used on BIG_ENDIAN systems, so we can always swap the > bits. 68a72,77 > #define htole32(x) \ > (((((uint32_t)x) & 0xff000000) >> 24) | \ > ((((uint32_t)x) & 0x00ff0000) >> 8) | \ > ((((uint32_t)x) & 0x0000ff00) << 8) | \ > ((((uint32_t)x) & 0x000000ff) << 24)) > I'm test-building it now, but I'm confident it'll work. The only risk is if we're clashing with an existing definition... -- Paul "TBBle" Hampson, on an alternate email client. - List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html
