| Age | Commit message (Collapse) | Author |
|---|
|
This moves md5.c out of libkern and into sys/crypto where it belongs (as
requested by markus@). Note that md5.c is still mandatory (dev/rnd.c uses it).
Verified with IPsec + hmac-md5 and tcp md5sig. OK henning@ and hshoexer@
|
|
This matches our SLIST behaviour and NetBSD's SIMPLEQ as well.
ok millert krw deraadt
|
|
jason is being a slacker
|
|
tested on alpha, i386, powerpc, sparc64, m68k.
ok miod@
|
|
on ppc now.
|
|
(Now modexp runs for non-1k/2k operations)
|
|
parts of memory... picky picky ;)
- zero the bignum memory after copying out the result (one less copy of
sensitive material floating around...)... Why don't vendors provide a
bit you can toggle that does this?
- Oh, and modexp now works.
|
|
a snapshot of work in progress (doesn't work correctly yet).
|
|
|
|
minutes ago (could have sworn I hit ^X^S).
|
|
- fix interrupt printing
- make shared structure entries volatile to ensure ordering
- swap the key, iv, and mackey arguments appropriately
- treat particle descriptor as two 32 elements (necessary for swabbing)
- set the endian control byte correctly and flip off swabbing on the packet data
(now works on macppc)
|
|
map structure.
- pull in some of sam's debugging stuff so that #define SAFE_DEBUG will compile
- set the dst <- src when uio && nicealign && uniform
|
|
also: let safe_uniform() check all of the elements
|
|
this is a completely hacked up version, complaints should be to me not sam =)
|
