summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2002-07-06KNF, realloc fix, and clean usageTheo de Raadt
2002-07-06KNFTheo de Raadt
2002-07-06Add some missing checks for malloc() returning NULL.Todd C. Miller
Noticed by Lars J. Buitinck. deraadt@ OK
2002-07-06cope better if realloc fails; better fix than cloder@acm.orgTheo de Raadt
2002-07-05Free crp_opaque only after we've determined we're not going toAngelos D. Keromytis
re-submit it. From sam@errno.com
2002-07-05Also, return EAFNOSUPPORT instead of ENOBUFS. Both this and theAngelos D. Keromytis
previous commit where noted by sam@errno.com
2002-07-05Bump ipips_family in ipipstat in all cases where EAFNOSUPPORT is returned.Angelos D. Keromytis
2002-07-05Never call crp_callback directly, use crypto_done() insteadJason Wright
2002-07-05KNF (no space after casts)Jason Wright
2002-07-05add sys/time.h to SYNOPSIS; from bk@rt.fmNiels Provos
2002-07-05unbreak.Henning Brauer
2002-07-05another small bug I found while installing a -current pf firewall.Henning Brauer
we don't support pass/block in on ! <interface> (at least, not yet) let the parser complain instead of ignoring the '!' ok pb@, dhartmei@
2002-07-05regress for getaddrinfo/getnameinfoJun-ichiro itojun Hagino
2002-07-05document "set loginterface none"Henning Brauer
2002-07-05allow unsetting the statusinterface viaHenning Brauer
set loginterface none ok dhartmei@
2002-07-05fix a small bug I found while installing a -current pf firewall at aHenning Brauer
client some days ago: if you had a rulefile with "set loginterface <interface>" and loaded through pfctl -e -f /etc/pf.conf, pfctl -si didn't display the interface stats, because on DIOCSTART pf_status.ifname was cleared and enableing is done after loading the ruleset. similar for DIOCCLRSTATUS, remember pf_status.ifname there as well. added feature: On DIOCSETSTATUSIF unset the statusinterface if pi->ifname is empty. ok dhartmei@
2002-07-05volatile sig_atomic_t, suggested by Theo.Hakan Olsson
2002-07-05Typo; stobe -> strobeAaron Campbell
2002-07-05Remove an unused #define (belongs only in the PCI attachment source file).Aaron Campbell
2002-07-05gcc 3.1 nits. Pointed out by David Krause.Hakan Olsson
2002-07-05Sync the ethernet controllers list with GENERIC.Miod Vallat
2002-07-05attempt to make ahc smaller. change PCI interrupt handler.Steve Murphree
2002-07-05allocation mischecks by cloder@acm.org; lots of other cleanup by meTheo de Raadt
2002-07-05more refinementTheo de Raadt
2002-07-04repair a dmesg cleanupMichael Shalayeff
2002-07-04More cleanup/shrinkage/clarification.Kenneth R Westerback
a) Abstract selection list handling into a function do_selection_list() rather than repeating the code twice. b) Eliminate function get_selection() and put limited remaining functionality directly into only caller, get_get_file_list(). c) Remove unused parameters in glob_selection() and invocations of same. d) Remove unused parameters from install_get_files_list() invocations. e) Remove unnecessary local parameter '_file_list' in install_mounted_fs().
2002-07-04Back out the previous revisionbrian
Objected to by: Andre Oppermann <oppermann@pipeline.ch> After Andre's objection, I've re-examined rfc 2759 and noted that it says that the domain name shouldn't be used when generating the NT-Response field. So it looks like the bug is in freeradius rather than in ppp.
2002-07-04Do not assume we have an active exchange during payload validation.Hakan Olsson
2002-07-04Eliminate the separate display_* functions. A little moreKenneth R Westerback
consolidation reduces calls to these functions to either 0 (display_failure_msg()) or 1. Just put the here documents in place of the invocation.
2002-07-04support non-"objdir" build. From: David Krause <openbsd@davidkrause.com>Jun-ichiro itojun Hagino
2002-07-04ep at isapnp works (dma on isapnp is still non-working).Miod Vallat
And fix a comment while there.
2002-07-04Add a few scary notes about the upgrade process, and advise people toMiod Vallat
backup their data and reinstall when the release gap is too big.
2002-07-04Fix invalid .Xr's; from PR 2804Miod Vallat
2002-07-04Since the directory /var/run is now present during install/upgrade,Kenneth R Westerback
put dmesg.boot there rather than the non-normal location /tmp/dmesg.boot.
2002-07-04styleHakan Olsson
2002-07-04Fix set selection list logic so that when extracting more setsKenneth R Westerback
a) The selection list does not display default sets already transferred. b) Saying 'done' does not seem to mysteriously skip the sets erroneously shown as selected due to a). This problem was found by miod@. In the process a) Use _setsdone only to record sets that have successfully been extracted. b) Don't loop on fetching files if a transfer fails. Just let the user know, and let the normal logic offer them a chance to extract the failed sets from the same or different source. c) Eliminate some (newly) dead variables, especially the global _osetsdone.
2002-07-04correct udp-from-broadcast check. from <andre@ae-35.com>Jun-ichiro itojun Hagino
2002-07-04document setting options in pf.confHenning Brauer
ok dhartmei@
2002-07-04don't allocate, copy, and discard if there is not interested in the data; ok ↵Markus Friedl
deraadt@
2002-07-04document port allocationJakob Schlyter
2002-07-04use tcp high portrange for active data connections. ok deraadt@Jakob Schlyter
2002-07-04blah blah minor nothing as i read and re-read and re-read...Theo de Raadt
2002-07-04minor KNFTheo de Raadt
2002-07-04new deviceNathan Binkert
2002-07-04New product ID.Nathan Binkert
From NetBSD
2002-07-04syncNathan Binkert
2002-07-04new devicesNathan Binkert
sort some devices by product id from NetBSD
2002-07-04Fix off by one error found by aaron@; my fixTodd C. Miller
2002-07-04 correct handling of EOF on both tty input and network. for example,Vincent Labrecque
this allows stuff like nc -l 10101 < /dev/null to have nc exit automatically when it gets EOF from the network. ok ericj@
2002-07-04ansiTheo de Raadt
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-07 02:25:40 +0000