summaryrefslogtreecommitdiffstats
path: root/sys
Commit message (Collapse)AuthorAgeFilesLines
* Back out previous commit, obrien doesn't seem to be watching. The problempeter1998-04-191-98/+318
| | | | | | is that the previous commit spammed a hacked 2.2-stable onto -current, deleting the DMA support etc. (I guess that's one way of minimizing diffs between -current and -stable.. :-] )
* Return EINVAL and do not change file pointer if resulting offset is negative.des1998-04-182-10/+14
| | | | PR: kern/6184
* MFC (reduce diff w/-CUR, with hopes of getting DMA support into -STABLE)obrien1998-04-181-318/+98
| | | | | | | | | 1.128 (FUNCTIONAL: better sleep handling) 1.131 (STYLE: labels w/o stmt) 1.132 (STYLE: remove unused #includes) 1.141 (FUNCTIONAL: devfs bug, 2nd controler not showing) 1.144 (STYLE: dont shadow other vars) 1.150 (FUNCTIONAL: fix search bug)
* In vfs_msync(), test to see if the vnode being examined is "interesting"peter1998-04-182-6/+20
| | | | | | | | | | | | | (ie: it has a vm_object attached and is marked as OBJ_MIGHTBEDIRTY) before attempting to lock it. This should reduce the cpu hit that is incurred when doing a sync(2) and when the syncer process is doing the 30-second writeback of dirty mmap() data to disk. Skip this speedup if we are doing an unmount() to be sure to get everything - we can afford to occasionally miss a msync while the system is running, but not at unmount. I'm not sure about the VXLOCK and MNT_WAIT case, it seems a bit odd to skip doing a page_clean at unmount time just because a vnode is VXLOCKed, but that's what was being done before...
* Sync with sys/i386/conf/majors.i386 revision 1.37.kato1998-04-181-1/+3
|
* Sync with sys/i386/isa/syscons.c revision 1.260.kato1998-04-181-1/+7
|
* Sync with sys/i386/conf/Makefile.i386 revision 1.109.kato1998-04-182-4/+4
|
* Support compiling with `gcc -ansi'.kato1998-04-182-2/+6
|
* Reviewed by: Amancioahasty1998-04-187-50/+217
| | | | | | | | | | | Submitted by: Roger Hardiman <roger@cs.strath.ac.uk> Roger Hardiman <roger@cs.strath.ac.uk> : Revised autodetection code to correctly handle both old and new VideoLogic Captivator PCI cards. Added tsleep of 2 seconds to initialistion code for PAL users. Corrected clock selection code on format change. --- Amancio
* Seventy-odd "its" / "it's" typos in comments fixed as per kern/6108.des1998-04-1784-170/+170
|
* Reserve a number for the ALTQ control devices (96)julian1998-04-172-2/+6
|
* Stop the screensaver before switching consoles.des1998-04-173-3/+21
|
* Undo my last change.dima1998-04-173-6/+6
|
* Really finish supporting compiling with `gcc -ansi'.bde1998-04-172-4/+4
|
* Sync with sys/i386/isa/syscons.c revision 1.259.kato1998-04-161-8/+11
|
* Sync with sys/i386/isa/npx.c revision 1.59.kato1998-04-161-6/+6
|
* Sync with sys/i386/conf/Makefile.i386 revision 1.108.kato1998-04-162-4/+4
|
* Sync with sys/i386/isa/pcaudio.c revision 1.40.kato1998-04-161-2/+2
|
* Sync with sys/i386/isa/lpt.c revision 1.68.kato1998-04-161-4/+4
|
* Sync with sys/i386/isa/if_ed.c revision 1.139.kato1998-04-161-8/+8
|
* Sync with sys/i386/i386/trap.c revision 1.126.kato1998-04-161-4/+4
|
* Attempt to fix BOUNCE_BUFFERS. I cannot test these drivers, but theypeter1998-04-162-2/+54
| | | | | seem to compile OK with the bounce buffer mods. I have only visually checked for missing bounce buffer support, I could have missed some.
* Attempt to get bounce_buffers working again for per-process kernel stackpeter1998-04-161-10/+40
| | | | | | data targets. At least st0 works for me again.... Also, scsi_scsi_cmd() looks like it's been exiting without a biodone() on an attached buffer in a number of error cases, leading to locked buffers.
* fix an obvious broken BOUNCE_BUFFER case.peter1998-04-161-1/+9
|
* Swapped "should we just return" and "should we stop the screensaver"des1998-04-163-24/+33
| | | | in scrn_timer() to avoid deadlock with graphical screensavers.
* When the softdep conversion took place, the periodic vfs_msync() frompeter1998-04-162-2/+4
| | | | | | | | | update got lost. This is responsible for ensuring that dirty mmap() pages get periodically written to disk. Without it, long time mmap's might not have their dirty pages written out at all of the system crashes or isn't cleanly shut down. This could be nasty if you've got a long-running writing via mmap(), dirty pages used to get written to disk within 30 seconds or so.
* Finish supporting compiling with `gcc -ansi'. Fix missing `volatile'sbde1998-04-153-18/+18
| | | | in __asm() statements while I'm here.
* Unlock mountlist_slock if the mount point was busy (unmount in progress)tegge1998-04-152-4/+8
| | | | during the attempt at lazy fsync.
* Use 0440 root.kmem kermissions on kernel instead ofdima1998-04-153-6/+6
| | | | | | 0555 root.wheel. Reviewed by: freebsd-security
* Support compiling with `gcc -ansi'.bde1998-04-1578-366/+378
|
* Support compiling with `gcc -ansi'. Fix missing `volatile's in __asm()bde1998-04-152-13/+17
| | | | statements while I'm here.
* In pre-devfs days, /dev/psm0 had always been a non-blockingphk1998-04-151-6/+6
| | | | | | | | | | | device. But with devfs, currently, /dev/psm0 is the blocking device and /dev/npsm0 is the non-blocking one. DEVFS must stay consistent with the older behaviour. PR: 6260 Reviewed by: phk Submitted by: Kapil Chowksey <kchowksey@hss.hns.com>
* Fixed bitrot in the non-softdep case of ufs_dirremove():bde1998-04-151-24/+9
| | | | | | | | | | | - restored async mount support. The first entry in a block is still always written synchronously, although it probably shouldn't be in the async case. - restored use of BWRITE() instead of bowrite() for the DOWHITEOUT case, although bowrite() is probably better. Broken by: merge of softdep changes (rev.1.22). Found by: lmbench2 delete-file benchmarks.
* Fixed breakage of fork accounting in previous commit. A fork benchmarkbde1998-04-153-6/+6
| | | | | reported about 15 times as much sys time as real time. getmicroruntime() is confusing name.
* Add a missing LK_RETRY.dt1998-04-152-10/+6
| | | | | | Noticed by: Bruce (almost 2 monts ago) Remove a debugging printf.
* Fix page fault panic by probing NE200 compatible PCI card.kato1998-04-142-15/+17
| | | | Submitted by: chi@bd.mbn.or.jp (Chiharu Shibata)
* Fix a minor mbuf leak created by the previous change.phk1998-04-141-2/+4
| | | | | Reviewed by: phk Submitted by: pb@fasterix.freenix.org (Pierre Beyssac)
* Added EIDRM & ENOMSG errno in translation table.sos1998-04-131-0/+2
|
* Added EIDRM & ENOMSG to errno translation table.sos1998-04-132-4/+4
|
* Don't use INTR when only one device supports it.msmith1998-04-133-1/+12
| | | | Submitted by: Satoh Junichi <junichi@astec.co.jp>
* Add EIDRM errno (PR 176), ENOMSG errno (myself) for prober sysv_ipc.sos1998-04-131-2/+5
|
* Wrong header length used for certain reassembled IP packets.phk1998-04-131-1/+3
| | | | | | PR: 6177 Reviewed by: phk, wollman Submitted by: Eric Sprinkle <eric@ennovatenetworks.com>
* Fixed autonegotiation. Card registers are now accessed via memorysemenu1998-04-133-315/+480
| | | | not i/o space.
* Sync with sys/i386/isa/wd.c revision 1.156.kato1998-04-131-2/+2
|
* Aic APM driver support for e.g. Toshiba T2400 notebook with internalphk1998-04-121-2/+59
| | | | | | | | Adaptec SCSI (aic6360/aha152x) controller. PR: 5382 Reviewed by: phk Submitted by: Oliver Breuninger <ob@seicom.NET>
* Sync with sys/i386/isa/wd.c revision 1.155.kato1998-04-121-41/+85
|
* Fix the problem when SCSI ID is not contiguous.kato1998-04-122-14/+20
| | | | Submitted by: URATA Shuichiro <s-urata@nmit.tmg.nec.co.jp>
* Add aio_error decl.dyson1998-04-121-1/+6
|
* setsockopt() transports user option data in an mbuf. if the userphk1998-04-111-2/+7
| | | | | | | | | data is greater than MLEN, setsockopt is unable to pass it onto the protocol handler. Allocate a cluster in such case. PR: 2575 Reviewed by: phk Submitted by: Julian Assange proff@iq.org
* Minor typo in the wd driver. The manifestation of this bugdyson1998-04-111-2/+2
| | | | | | | is a tremendous perf decrease due to the disabling of advanced features such as DMA, Ultra DMA, and 32bit mode. This patch might have been reported by someone else (I seem to remember it.)
OpenPOWER on IntegriCloud