summaryrefslogtreecommitdiff
path: root/lib
AgeCommit message (Collapse)Author
2002-06-18unbreak sshd with privsep: open /dev/crypto, keep fd, and callMarkus Friedl
CRIOGET per EVP_Init(); ok niklas@, miod@
2002-06-18per-evp state is now sizeof(struct dev_crypto_state) instead sizeof(struct ↵Markus Friedl
session_op)
2002-06-18Rewrite to get proper Copyright.Marc Espie
Should even be slightly better than the old one. Checked and fixed by miod@, my initial version was not 100% right (writing m68k loops blindly doesn't always work).
2002-06-17Make the stupid warnings go away.Angelos D. Keromytis
2002-06-17make fabs use asm, and license repairTheo de Raadt
2002-06-17sync with libc_r (NOPOLL no longer needed)Marco S Hyman
2002-06-16match man page to source; from Jean-Francois Brousseau <krapht@secureops.com>Todd T. Fries
ok deraadt@
2002-06-15enable libeventNiels Provos
2002-06-15build static and profiled.Niels Provos
2002-06-15import libevent an event notification abstraction on top of select or kqueueNiels Provos
supporting timeout and signal callsbacks, too. okay deraadt@ millert@
2002-06-14spelling; from Brian Poole <raj@cerias.purdue.edu>Todd T. Fries
2002-06-13keep a FD per EVP_init, use a global FD for all asym operations;Markus Friedl
ok beck@
2002-06-12__indr_reference() and the definition in the same compilation unit areFederico G. Schwindt
a bad idea. use __weak_alias(), but for now put the definition at the bottom, since AUX_FUNC is not properly propagated to the weak definition, and will cause grief to the linker. should fix vax. tested by hin@ via LDSTATIC=-static, millert@ ok.
2002-06-11KNFTheo de Raadt
2002-06-11add "dsa_dsa_mod_exp" - This mimics the software dsa_mod_exp funtionBob Beck
using two mod_exp operations - otherwise we use BN_mod_exp2 entirely in software, which makes dsa verifications glacially slow while signatures, (which use mod_exp) are fast. This lets cards that can only do bn_mod_exp decently offload most of dsa.
2002-06-11Make DSA work now... at least for things that can do bn_mod_exp.Bob Beck
2002-06-11Make asymmetric crypto work in userlandBob Beck
this will only be used if you both have a card that supports it with a working driver and you set sysctl kern.userasymcrypto=1
2002-06-11Document kern.userasymcryptoAngelos D. Keromytis
2002-06-11random cleanup that's cheaper to commit than to remove from my tree.Artur Grabowski
2002-06-11strike newline from err(3) functions; ok nate@Jason Peel
2002-06-10correct comment.Federico G. Schwindt
2002-06-10no more _NO_WEAK_ALIASES here.Federico G. Schwindt
2002-06-10put back in the assert changes for ISO C 99 (put __func__ in it)Marc Espie
2002-06-09some -Wall and spaces cleanup, scsi.c left.Federico G. Schwindt
some brave soul should look at it.
2002-06-09spelling; moritz@jodeit.orgTheo de Raadt
2002-06-09spelling; moritz@jodeit.orgTheo de Raadt
2002-06-09Document KERN_MBSTATAngelos D. Keromytis
2002-06-09use strchr() instead of index()Theo de Raadt
2002-06-09Pass the right arguments for RSA, DSA, and modexp operations. Fix theAngelos D. Keromytis
translation between the crypto framework's format and the BN structure.
2002-06-09another \n removed from syslog()Hans Insulander
2002-06-08Document proc arg sysctls.Artur Grabowski
2002-06-08Document NFILES, TTYCOUNT (to come) and NUMVNODES (to come).Angelos D. Keromytis
2002-06-08If possible get arg and env information with sysctl.Artur Grabowski
2002-06-08Add a flag - KVM_NO_FILES which tells kvm_openfiles to notArtur Grabowski
open any files, just return an kvm handle that we can use to some of the functions.
2002-06-08ELF architectures all support weak aliases, enable them for this part.Dale Rahn
ok fgsch@
2002-06-08After much horrible and painful slogging through asn1 code,Bob Beck
this fixes the source of connection problems with ssl/tls connections between sparc64 and other things. The punchline, we just found a bug in floating point emulation on sparc64 when this script produces off-by-one output on sparc64. This fix is annoyingly easy for the effort expended.
2002-06-08use weak aliases whenever is possible; millert@ ok.Federico G. Schwindt
2002-06-07do not propose IDEA cipher on SSL connection. tested by beckJun-ichiro itojun Hagino
noticed by Sverre Froyen <sverre@viewmark.com>
2002-06-07document net.inet6.ip6.v6onlyJun-ichiro itojun Hagino
2002-06-07typo - I am a luser and a moron.Bob Beck
2002-06-07Merge OpenSSL 0.9.7-stable-20020605,Bob Beck
correctly autogenerate obj_mac.h
2002-06-07support scoped IPv6 address.Jun-ichiro itojun Hagino
no visible API change, old config files work just fine. now you can use expressions like "ALL: [fe80::%lo0/64]". theo ok
2002-06-07func prototypeJun-ichiro itojun Hagino
2002-06-07more strict prototypeJun-ichiro itojun Hagino
2002-06-07snprintfJun-ichiro itojun Hagino
2002-06-06Implement _atomic_lock in assembler to make it more readable.Artur Grabowski
plus some various assembler improvements. Tested by various people a few months ago and in my tree for almost 6 months now.
2002-06-06Fix *documentation*: one arg of the ktrace(2) is pid_t not int, found whileMike Pechkin
have fun w/ pid_t cleanups. millert@ ok
2002-06-05one more level of indirection for dlopen structure.Marc Espie
Allows dlopen to be used much earlier, e.g., in dynamic constructors. From NetBSD. okay art@
2002-06-05snprintfTheo de Raadt
2002-06-05If send() returns ENOBUFS, sleep for one microsecond and retry.Todd C. Miller
deraadt@ OK
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-28 13:06:03 +0000