summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Rename _thr_enter_cancellation_point to _thr_cancel_enter, renamedavidxu2003-12-0960-272/+296
| | | | | | | | | | | | | | | | | | | | | _thr_leave_cancellation_point to _thr_cancel_leave, add a parameter to _thr_cancel_leave to indicate whether cancellation point should be checked, this gives us an option to not check cancallation point if a syscall successfully returns to avoid any leaks, current I have creat(), open() and fcntl(F_DUPFD) to not check cancellation point after they sucessfully returned. Replace some members in structure kse with bit flags to same some memory. Conditionally compile THR_ASSERT to nothing if _PTHREAD_INVARIANTS is not defined. Inline some small functions in thr_cancel.c. Use __predict_false in thr_kern.c for some executed only once code. Reviewd by: deischen
* Put on some factor 30+ bikeshed repellent and export the internalpeter2003-12-091-1/+7
| | | | | | | | | | Makefile.inc1 underscore targets with a big warning that they are intentionally undocumented, subject to change without notice and might poison your dog etc. If you don't know what they are, then they are not meant for you to use. I've added these by hand to so many many trees that I've lost count. I find them rather useful.
* More reliably check timeout for pthread_mutex_timedlock.davidxu2003-12-092-2/+2
|
* Fix a problem where 'make installworld' will fail and leave thegad2003-12-091-13/+7
| | | | | | | | | | | | | | system in a messy state *if* the user is upgrading from a system which has no /libexec to a system which builds a DYNAMICROOT, and if that user has set DISTDIR (as documented for ports, but it turns out that the same variable name is used for a completely unrelated purpose in 'make release'). There are other possible fixes for this issue, and ru@ may later decide to commit one of those fixes. I just wanted some fix in ASAP, and this is the fix that I have tested. Reviewed by: bde, imp, and ru
* Set __mbrtowc and __wcrtomb correctly when changing to the C/POSIX locale.tjr2003-12-081-0/+14
| | | | | | | Save __mbrtowc and __wcrtomb and restore them when changing back to the cached locale. Reported by: perky
* Correct the READ/WRITE CDB direction definition.ps2003-12-081-2/+2
| | | | Submitted by: John Cagle <john.cagle@hp.com>
* First pass at attempted debrucification:peter2003-12-082-7/+7
| | | | | | | | | | - sort the -E switch into the right place. - add previously missing -p pid in usage (from the last few commits). - add -E to usage. - consistently use trfile in the man page. I knew I shouldn't have touched the man page. If I commit to a man page, it just makes people suspicious. :-)
* Add hints about nForce3 issuespeter2003-12-081-1/+16
|
* Add a small program to test/measure with the RFC 2783 API for timingphk2003-12-083-0/+231
| | | | external signals.
* Hook boot0sio to the build. Use 'boot0cfg -b /boot/boot0sio' to use.obrien2003-12-083-15/+14
|
* Add a serial console capable version of the FreeBSD boot manager. This hasbms2003-12-083-1/+883
| | | | | | | been lying around my tree(s) for the past year or so. It could do with TLC. Requested by: obrien Sponsored by: Weyland-Yutani Corporation
* Define RAID 5+1 and RAID ADG fault tolerancesps2003-12-082-0/+6
|
* Go back to using rev 1.18 where thread locks are used instead of KSEdeischen2003-12-082-34/+32
| | | | | | | locks for [libc] spinlock implementation. This was previously backed out because it exposed a bug in ia64 implementation. OK'd by: marcel
* - changed the logic in nd6_is_addr_neighbor(); check on-link prefixesume2003-12-081-17/+23
| | | | | | | | | | | | | | (not interface addresses) to see if a given address is on-link. - skip offlink prefixes in neighbor determination in nd6_is_addr_neighbor. - in nd6_is_addr_neighbor, regarded every address as on-link when the default router list is empty. otherwise, we'd not be able make a neighbor cache for the address. this algorithm is applied to hosts only. - in nd6_is_addr_neighbor, check if the default interface is equal to the interface in question in addition to check if the default router list is empty. Obtained from: KAME
* Try to catch up with device name changes due to GEOM'ification. Removeobrien2003-12-081-14/+19
| | | | | | note about requirement of operating on 'c' partition, GEOM removed this. Mention RAIDframe, don't mention DPT hardware RAID as a good alternative.
* More errata fixing for the SiI3112A disaster chip:sos2003-12-081-1/+3
| | | | | | | | | | Serialize access to the SATA channels, the chip messes up if both channels are used at the same time. The SiI3112 hereby takes the price as the most crappy SATA chip in existance by a significant amount. My advise to our userbase is to avoid this chip like the plague...
* Make msdosfs long filenames matching case insensitive again.fjoe2003-12-0810-33/+66
| | | | | PR: 59765 Submitted by: Ryuichiro Imura <imura@ryu16.org>
* Fix more problems with the Silicon Image 3112A:sos2003-12-081-1/+17
| | | | | | | | | | | | | | | | | | Setup decent transfer mode defaults as some BIOS's seem to put in things that it *knows* doesn't work. (Note to BIOS writers: stop doing that nonsense, we will get things working with your crappy HW anyways, and then recommend users to buy someone else's products that "just works", thankyou.. ) Limit the device transfer mode to ATA100/UDMA5 on generic SATA. Since we dont know if the user is using a pure SATA device or an old PATA drive with a SATA converter dongle, we need to limit the speed used here to cover up the problems with Marvell ATA-SATA bridges used in lots of SATA products. This workaround is enabled for all detectable SATA controllers as they seem to have semilar problems here. One notable exception is all the Promise pdc2037x chips which just always work (cudos to Promise!).
* Move the bktr(4) <arch>/include/ioctl_{bt848,meteor}.h files to dev/bktrobrien2003-12-087-19/+18
| | | | | | as these ioctl's aren't MD. This also means they are installed in /usr/include/dev/bktr now. Also provide compatability wrappers for where these headers lived in 4.x.
* Unbreak vmstat -i on ia64:marcel2003-12-081-9/+12
| | | | | | | o nintr and inamlen must by of type size_t, not int, o Remove now unnecessary casts, o Handle the aflag differently, because the intr. names have a fixed width and almost always have trailing spaces.
* Don't use caddr_t in mchash(). Also use C99 spellings over BSD ones.obrien2003-12-0828-140/+142
| | | | Requested by: bde,imp
* Move the bktr(4) <arch>/include/ioctl_{bt848,meteor}.h files to dev/bktrobrien2003-12-0813-1378/+101
| | | | | | as these ioctl's aren't MD. This also means they are installed in /usr/include/dev/bktr now. Also provide compatability wrappers for where these headers lived in 4.x.
* Fix sort order.obrien2003-12-081-1/+1
|
* Repo copied from sys/i386/include. All the <arch>/machine ones areobrien2003-12-080-0/+0
| | | | going to share this copy.
* Move the ciss quirk to the right section, also update the commentps2003-12-081-8/+7
| | | | | stating that the controllers do not support SYNC CACHE since ciss only supports a small subset of the scsi spec.
* Catch up with i386 GENERIC rev 1.389.obrien2003-12-081-6/+8
|
* ciss doesn't like scsi SYNC CACHE. turn it offps2003-12-081-0/+8
|
* Forced commit:imp2003-12-080-0/+0
| | | | | | Last commit moved info about the aha-1640 card. It also added notes about the fact that the A card is busted (and likely will never be fixed) and the B card doesn't work well on heavy load.
* Add the meteor driver man page back. It has moved up a level from theimp2003-12-083-5/+886
| | | | | | | | | | | | man4.i386. It documents that meteor no longer works, but keeps the extensive documentation on the meteor interface, which the bktr driver implements also. This should be merged into tha man page, but such a merging seems to be planned by others. # we really need something like video4bsd to define these sorts of # things for all video capture drivers. Requested by: rwatson and obrien
* Remove trailing whitespace.kan2003-12-081-7/+7
|
* Diff to vendor reduction.obrien2003-12-081-0/+1
|
* The last argument to mtx_init() should be MTX_DEF, not 0. This is not atruckman2003-12-082-5/+10
| | | | functional change since MTX_DEF happens to be defined as 0.
* The last argument to mtx_init() should be MTX_DEF, not 0. This is not atruckman2003-12-082-2/+2
| | | | functional change since MTX_DEF happens to be defined as 0.
* Addendum to revision 1.174: In the case where vm_pager_allocate() is calledalc2003-12-081-2/+6
| | | | | | | to create a vnode-backed object, the vnode lock must be held by the caller. Reported by: truckman Discussed with: kan
* Copy the old dgb man page to digi.4, plus very minor hacking. Add aimp2003-12-082-0/+383
| | | | | | note that says that this man page is sub-optimal. Bruce Mah should be happier about this, but someone that groks the cards supported by the digi driver is encouraged to make this man page suck less.
* Lock and unlock sched_lock when walking through thread list, current wedavidxu2003-12-071-1/+11
| | | | | insert kse upcall thread into thread list at mi_switch time, process lock is not enough.
* Correct usage of mtx_init() API. This is not a functional change sincetruckman2003-12-071-1/+1
| | | | | | the code happened to work because MTX_DEF and NULL are both defined as 0. Reviewed by: phk
* Correct usage of mtx_init() API. This is not a functional change sincetruckman2003-12-072-3/+3
| | | | | | the code happened to work because MTX_DEF and NULL are both defined as 0. Reviewed by: phk
* Try a little harder to catch when fixate finishes.sos2003-12-071-2/+4
| | | | | On the <QSI CD-RW/DVD-ROM SBW-242> that failed causing burncd to wait forever...
* Urk, add arch attributes to items from my previous commit.bmah2003-12-072-8/+8
|
* Style changetrhodes2003-12-072-2/+2
|
* New release notes: Record the demise of the lmc, dgb, hea, meteor,bmah2003-12-072-6/+28
| | | | and loran drivers.
* Don't remove the virtual-to-physical mapping when an sf_buf is freed.alc2003-12-072-9/+11
| | | | | | | | | | | | | | | Instead, allow the mapping to persist, but add the sf_buf to a free list. If a later sendfile(2) or zero-copy send resends the same physical page, perhaps with the same or different contents, then the mapping overhead is avoided and the sf_buf is simply removed from the free list. In other words, the i386 sf_buf implementation now behaves as a cache of virtual-to-physical translations using an LRU replacement policy on inactive sf_bufs. This is similar in concept to a part of http://www.cs.princeton.edu/~yruan/debox/ patch, but much simpler in implementation. Note: none of this is required on alpha, amd64, or ia64. They now use their direct virtual-to-physical mapping to avoid any emphemeral mapping overheads in their sf_buf implementations.
* The lmc, meteor, and loran drivers have been removed from the sourcebmah2003-12-072-20/+8
| | | | | | | | tree, so remove them from the hardware notes. The dgb driver has been replaced by the digi driver. The hea driver has been replaced by the en driver.
* Bump various entities for 5.2-CURRENT, trim release notes and errata.bmah2003-12-074-1620/+34
|
* Pass MTX_DEF as the last argument to mtx_init() instead of 0. Thistruckman2003-12-072-3/+3
| | | | | is not a functional change. The code happened to work properly only because MTX_DEF is defined as 0.
* bandaid LOR in rt_setgate; a proper fix requires code refactoringsam2003-12-071-4/+7
|
* add install rulesam2003-12-072-0/+11
|
* o report 0 instead of -1 for xmit rate when in transitionsam2003-12-071-4/+39
| | | | | o report rssi in running display o futz with column widths in running display
* fix commentsam2003-12-071-1/+1
|
OpenPOWER on IntegriCloud