<feed xmlns='http://www.w3.org/2005/Atom'>
<title>wireguard-openbsd/lib/libc/arch/amd64/string/Makefile.inc, branch jd/queueboosts</title>
<subtitle>WireGuard implementation for the OpenBSD kernel</subtitle>
<id>https://git.zx2c4.com/wireguard-openbsd/atom/lib/libc/arch/amd64/string/Makefile.inc?h=jd%2Fqueueboosts</id>
<link rel='self' href='https://git.zx2c4.com/wireguard-openbsd/atom/lib/libc/arch/amd64/string/Makefile.inc?h=jd%2Fqueueboosts'/>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/wireguard-openbsd/'/>
<updated>2014-12-09T15:10:39Z</updated>
<entry>
<title>Import new amd64 assembly versions of strchr/index, strrchr/rindex,</title>
<updated>2014-12-09T15:10:39Z</updated>
<author>
<name>reyk</name>
<email>reyk@openbsd.org</email>
</author>
<published>2014-12-09T15:10:39Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/wireguard-openbsd/commit/?id=c2a49701f6bfcd31d0b97ecc55a0a2af126dbb7d'/>
<id>urn:sha1:c2a49701f6bfcd31d0b97ecc55a0a2af126dbb7d</id>
<content type='text'>
and strlen that provide a significantly faster performance than our
previous .c or .S implementations.  Based on NetBSD's code.

Tested with different amd64 CPUs.

ok deraadt@ mikeb@
</content>
</entry>
<entry>
<title>restructure libc/string + libc/arch/*/string coperation regarding</title>
<updated>2014-11-30T19:43:54Z</updated>
<author>
<name>deraadt</name>
<email>deraadt@openbsd.org</email>
</author>
<published>2014-11-30T19:43:54Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/wireguard-openbsd/commit/?id=5b859c19fe53bbea08f5c342e0a4470e99f883e1'/>
<id>urn:sha1:5b859c19fe53bbea08f5c342e0a4470e99f883e1</id>
<content type='text'>
(potentially) MD versions (function dependent, not filename dependent)
split out memcpy/memmove/bcopy and strchr/index/strrchr/rindex
Bring back amd64 .S versions

And the final touch: switch all architectures temporarily to MI
memcpy.c, which contains syslog + abort for overlapping copies.  A nice
harsh undefined behaviour.  We will clean the entire userland of the
remaining issues in this catagory, then switch to the optimised memcpy
which skips the memmove check.

I tried to cut this change into pieces, but testing each sub-step on
every architecture is too time consuming and mindnumbing.
ok miod
</content>
</entry>
<entry>
<title>One of these optimized stubs creates some incredibly subtle damage,</title>
<updated>2014-11-20T21:51:02Z</updated>
<author>
<name>deraadt</name>
<email>deraadt@openbsd.org</email>
</author>
<published>2014-11-20T21:51:02Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/wireguard-openbsd/commit/?id=d1a57517ee8d16db5fa9ffc7f8a82afb73782cbe'/>
<id>urn:sha1:d1a57517ee8d16db5fa9ffc7f8a82afb73782cbe</id>
<content type='text'>
causing as(1) to create a wrong nop-sled for text segment aligns.
Revert, until it is found and fixed.
</content>
</entry>
<entry>
<title>Add amd64 assembler versions of some standard functions to libc.  The</title>
<updated>2014-11-20T14:33:00Z</updated>
<author>
<name>reyk</name>
<email>reyk@openbsd.org</email>
</author>
<published>2014-11-20T14:33:00Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/wireguard-openbsd/commit/?id=71cd5ab55175e4ba4e758f03a22a8b3c34a8d5a9'/>
<id>urn:sha1:71cd5ab55175e4ba4e758f03a22a8b3c34a8d5a9</id>
<content type='text'>
code is already used in the kernel and the files are unmodified copies
from src/sys/lib/libkern/arch/amd64/.  Depending on the function, this
gives us some speedup in userland.

ok deraadt@, no objections from miod@
</content>
</entry>
<entry>
<title>remove lint leftovers; ok guenther@</title>
<updated>2012-09-04T03:10:42Z</updated>
<author>
<name>okan</name>
<email>okan@openbsd.org</email>
</author>
<published>2012-09-04T03:10:42Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/wireguard-openbsd/commit/?id=ac9b4aacc1da35008afea06a5d23c2f2dea9b93e'/>
<id>urn:sha1:ac9b4aacc1da35008afea06a5d23c2f2dea9b93e</id>
<content type='text'>
</content>
</entry>
<entry>
<title>Reimplement mutexes, condvars, and rwlocks to eliminate bugs,</title>
<updated>2012-01-17T02:34:18Z</updated>
<author>
<name>guenther</name>
<email>guenther@openbsd.org</email>
</author>
<published>2012-01-17T02:34:18Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/wireguard-openbsd/commit/?id=2aa8ea07027e21f7e3a39dbbc13d2a3af24af722'/>
<id>urn:sha1:2aa8ea07027e21f7e3a39dbbc13d2a3af24af722</id>
<content type='text'>
particularly the "consume the signal you just sent" hang, and putting
the wait queues in userspace.

Do cancellation handling in pthread_cond_*wait(), pthread_join(),
and sem_wait().

Add __ prefix to thr{sleep,wakeup,exit,sigdivert}() syscalls; add
'abort" argument to thrsleep to close cancellation race; make
thr{sleep,wakeup} return errno values via *retval to avoid touching
userspace errno.
</content>
</entry>
<entry>
<title>fix some lint 'xxx used, but not defined' warnings by properly adding</title>
<updated>2007-05-15T18:42:31Z</updated>
<author>
<name>otto</name>
<email>otto@openbsd.org</email>
</author>
<published>2007-05-15T18:42:31Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/wireguard-openbsd/commit/?id=94fd4554194a14f126fba33b837cc68a1df42468'/>
<id>urn:sha1:94fd4554194a14f126fba33b837cc68a1df42468</id>
<content type='text'>
stuff to LSRCS
</content>
</entry>
<entry>
<title>things for amd64; from art@</title>
<updated>2004-01-28T01:44:44Z</updated>
<author>
<name>mickey</name>
<email>mickey@openbsd.org</email>
</author>
<published>2004-01-28T01:44:44Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/wireguard-openbsd/commit/?id=118f6189edb47d41d3a0badafe95d9585c0d1c57'/>
<id>urn:sha1:118f6189edb47d41d3a0badafe95d9585c0d1c57</id>
<content type='text'>
</content>
</entry>
</feed>
