summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2010-04-01WARNINGS -> CAVEATS, and a little neccessary cleanup;Jason McIntyre
2010-04-01Track blackhole and reject flags since the MIB knows about them.Claudio Jeker
Remove an unneeded stack storage as well. OK jsg@
2010-04-01Minor cleanup figured out while syncing kroute with snmpd's.Claudio Jeker
OK dlg@
2010-04-01After syncing kroute with bgpd sync now against ospfd's so that this codeClaudio Jeker
uses only one rtm parser instead of three different ones. Fixes problems with priorities that were not handled everywhere correctly. Found by and OK jsg@
2010-04-01Allow for multiple CPUs by moving to an array of cpu_info.Joel Sing
ok kettenis@
2010-04-01typoTheo de Raadt
2010-04-01document that setting the time of the machine is unsafe. explain why,Theo de Raadt
and suggest workarounds ok guenther
2010-04-01tweak previous; ok deraadtJason McIntyre
2010-04-01Allow top to accept tab characters as input in interactive modelum
while using the kill and renice commands. This removes a bug where process numbers that came after a tab character in the input string would not be affected by the corresponding command. ok tedu@ otto@
2010-03-31make ProxyPass exclude directives only exclude on match, instead of alwaysAlexander Hall
noted by Vasiliy Kiryanov, thanks! ok stephan@
2010-03-31More carefully manage PG_PMAP_EXE bit and cache flushing on pmap_protectDale Rahn
operations, where X or W is taken away. ok deraadt@ kettenis@
2010-03-31Make sure the boundaries of uvm_pglistalloc() calls are set up with lowMiod Vallat
being page-aligned, and high being end of page (i.e. high & PAGE_MASK == PAGE_MASK) everywhere, for consistency. Future code will depend on this. ok deraadt@
2010-03-31Add bnx firmware on installation media; this causes a media oflow, soMiod Vallat
enlarge bsd.rd filesystem; this in turn causes a cd oflow, so enlarge cd filesystem. ok jsing@ deraadt@
2010-03-31Bring /dev/apm support in line with apm(4). Make the suspend button sendMark Kettenis
a suspend request event to apmd(8) instead of suspending immediately. Also keep track of whether /dev/apm and /dev/apmctl are currently open, such that we can still suspend immediately if apmd(8) isn't running. ok deraadt@, marco@, pirofti@, jsing@, oga@
2010-03-31network static and network connected have been superseded by network inetClaudio Jeker
static and network inet connected a long time ago. It is time to remove the old compat code.
2010-03-31Don't accept keys with modifiers as input. Fixes crash reported by BrianNicholas Marriott
R Landy.
2010-03-31put date(1) onto the media, it fitsTheo de Raadt
2010-03-31Make a -DSMALL version of date(1) for the mediaTheo de Raadt
2010-03-31if bthub defines are needed by conf.c, they must come from files.bluetoothTheo de Raadt
2010-03-31fix typoTheo de Raadt
2010-03-31Radeon HD2400 XT works. Tested by Dennis den Brok; thanks!Owain Ainsworth
2010-03-31Track interface addresses in a AF independent way. This is needed to supportClaudio Jeker
the AF independent MIBs like RFC4292 in the future. Additionally count number of routes (also needed by some MIBs) and do not track the state of routes since the MIB do not require it and it is expensive. OK jsg@
2010-03-31update the inteldrm GEM interface to the newer version.Owain Ainsworth
This adds a couple of new ioctls and remove the I915_GEM_EXECBUFFER ioctl in exchange for EXECBUFFER2 which supports better tiling parameters. the original interface is removed since our GEM support will not support it. This temporarily breaks the build of libdrm_intel, this will be fixed in the next commit. discussed with matthieu@
2010-03-30regenMatthieu Herrb
2010-03-30Add missing bthub device entries on arches that support bluetooth.Matthieu Herrb
ok miod@, deraadt@, todd@.
2010-03-30several missing closedir() callszinovik
regress test run fine with this changes ok deraadt@, drahn@
2010-03-30revert revision 1.53.Owain Ainsworth
Now the kernel does a proper vt switch for us this hack is no longer needed. ok miod@
2010-03-30Prevent the apmd/x races for good.Owain Ainsworth
When we hit suspend time, go through all wsdisplays on the system. if they are in mode MAPPED, but not MODE_DUMBFB then if possible do a full vt switch to a !mapped vt, and prevent switching back until resume time. This has to be called from MD code because this involves userland running so that X can run the vt switch signal handler. This way, any case where we are using the "poke registers from userland" model, we will not be on the hardware when we go down, so the kernel can actually handle thing properly. Tested on several acpi laptops (by kettenis@ and ian@), x40 (me and beck@ at LEAST) and zaurus (me). Maybe others, but if so I forgot who at this time.. Idea from deraadt somewhere over the Faroe Islands (I thought of a similar thing myself a while ago). Much prompting from him. Ok and comments miod@
2010-03-30After snprintf(3) call `dir' is no longer used so free it also on normalzinovik
code path. ok deraadt@
2010-03-30This diff synchronizes rcs parser code with OpenRCS. It shrinks code by 3zinovik
lines, but does exactly the same thing. It also removes unneeded variable `buf' ok deraadt@, xsa@
2010-03-30Hmpf, aid2af() leftover from the time I did the transition that should notClaudio Jeker
be there. network X set nexthop Y failed because of this.
2010-03-30Since we use a direct mapping for the area where we save a process' floatingMark Kettenis
point registers, we no longer need to flush/purge the cache.
2010-03-30Support Radeon Mobility HD3670; James Thackray. Thanks!Owain Ainsworth
2010-03-30syncOwain Ainsworth
2010-03-30Mobility Radeon HD3670, from James Thackray (first.last@gmail.com)Owain Ainsworth
2010-03-30Have bge_start check if there are less than 16 free send BDs and if notChristian Weisgerber
set IFF_OACTIVE and try again later. Previously bge(4) reserved 16 send BDs after loading DMA maps but the hardware only requires one reserved send BD. From FreeBSD, via Brad
2010-03-30Log non-informational messages at LOG_WARNING priority instead of LOG_INFO.Jacek Masiulaniec
OK millert@
2010-03-30syncTheo de Raadt
2010-03-30Do not print ac97 link rate message unless debug is enabled.David Coppa
Adapted from NetBSD OK jakemsr@
2010-03-30describe better what KERN_PROC_ARGV and KERN_PROC_ENV return; ok jmc@Otto Moerbeek
2010-03-30We will never condtionalize ficache on PTE_PROT(TLB_EXECUTE) because itTheo de Raadt
is risky. Delete the comment suggesting we might.
2010-03-29Initialize various uninitialized variables. Found by jsg@ via Clang.Kenneth R Westerback
Feedback from miod@ and kettenis@. ok beck@
2010-03-29merge 1.9.19, keeping local patchesIngo Schwarze
* scoping fixes for roff instructions * accept apostroph in place of dot as a macro control character * accept tabs between the control character and the macro name * check that man(7) .TH titles use capital letters
2010-03-29Initialize 'flags' field of tree_cache stack variables to 0. Paranoia.Kenneth R Westerback
2010-03-29Always put server id in NAK's to requests. Some router/switchesKenneth R Westerback
need this, as reported in PR#6339. Fix tested by submitter, Marcel Widget, who also supplied the diff that I tweaked. ok beck@
2010-03-29PMAP_CANFAIL for bus_dmamem_map on all other architectures (and someOwain Ainsworth
whitespace tweaks on i386 so that it matches). ok kettenis@
2010-03-29Add Radeon HD360 Mobility.Owain Ainsworth
Info from Mitja.
2010-03-29Sync kroute.c code with bgpd so that snmpd is able to track INET6 addrs asClaudio Jeker
well. First big sync, the interface address tree still needs to be made AF independent but that will be done in a second step. OK jsg@
2010-03-29Improve the way we handle spoofed paritions for Resize. Now any autoOtto Moerbeek
allocated can be treated. We still check if no non-native partition overlaps the openbsd area, to avoid surprised. ok krw@ lum@
2010-03-29I'm going to need community_ext_conv() outside of rde_attr.c soon, so moveClaudio Jeker
the prototype to rde.h