summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* - In kseq_choose(), don't recalculate slice values for processes with ajeff2004-12-141-11/+25
| | | | | | | | | | | | | nice of 0. Doing so can cause an infinite loop because they should be running, but a nice -20 process could prevent them from doing so. - Add a new flag KEF_PRIOELEV to flag a thread that has had its priority elevated due to priority propagation. If a thread has had its priority elevated, we assume that it must go on the current queue and it must get a slice. - In sched_userret() if our priority was elevated and we shouldn't have a timeslice, yield here until we should. Found/Tested by: glebius
* Be a bit more assertive about vnode bypass.phk2004-12-141-16/+17
|
* Minor mdoc(7) tweaks, excellent work Nate!ru2004-12-141-14/+18
|
* Add a new kind of reference count (fd_holdcnt) to struct filedescphk2004-12-142-17/+47
| | | | | | | | | | | | | | | | | | | | | | | which holds on to just the data structure and the mutex. (The existing refcount (fd_refcnt) holds onto the open files in the descriptor.) The fd_holdcnt is protected by fdesc_mtx, fd_refcnt by FILEDESC_LOCK. Add fdhold(struct proc *) which gets a hold on the filedescriptors of the specified proc.. Add fddrop(struct filedesc *) which drops the fd_holdcnt and if zero destroys the mutex and frees the memory. Initialize the fd_holdcnt to one in fdinit(). Normal operations on the filedesc structure will not change it. In fdfree() use fddrop() to dispose of the mutex and structure. Hold the FILEDESC_LOCK() until we have cleaned out the contents and carefully set the fields to null values during cleanup. Use fdhold()/fddrop() in mountcheckdirs() and sysctl_kern_file().
* mdoc(7) nit: Removed redundant .Pp call.ru2004-12-141-1/+0
|
* Make fdesc_mtx private to kern_descrip.c now that the flock has come home.phk2004-12-142-3/+1
|
* Move the checkdirs() function from vfs_mount.c to kern_descrip.c andphk2004-12-143-52/+52
| | | | call it mountcheckdirs().
* - Use ng_callout() instead of timeout.glebius2004-12-141-13/+7
| | | | | | | | - Schedule next timeout *after* finishing job of the current one. - Remove spl(9) calls. Tested by: ru Reviewed by: julian
* Make LINT compile.phk2004-12-141-1/+1
| | | | | | | When leaving functions for ddb use don't make them static: it makes gcc think they are unused. Shouldn't this be in #ifdef DDB anyway ?
* Add new function fdunshare() which encapsulates the necessary light magicphk2004-12-144-22/+23
| | | | | | | | for ensuring that a process' filedesc is not shared with anybody. Use it in the two places which previously had private implmentations. This collects all fd_refcnt handling in kern_descrip.c
* A bunch more whitespace and formatting diff reductions for NetBSD.julian2004-12-141-17/+19
| | | | | Obtained from: NetBSD MFC after: 1 week
* Add manual page for snd_audiocs(4).yongari2004-12-142-1/+77
| | | | | Reminded by: trhodes MFC after: 1 week
* Don't abandon ship just because the number of companions doesn't seem correct.julian2004-12-141-3/+6
| | | | | Obtained from: NetBSD MFC after: 1 week
* Slightly reorganise part of the ohci_softintr() functionjulian2004-12-141-15/+26
| | | | | | | | | | | to better keep track of the total amoutn transferred during a transfer. Seems similar to some code in the NetBSD version. I notice they have incorporated matches from him so I don't know which direction it went. Submitted by: damien.bergamini@free.fr Obtained from: patches to make the ueagle driver work MFC after: 1 week
* Small formatting change..julian2004-12-141-1/+1
| | | | | | | Move a declaration to the same place as in NetBSD. Obtained from: NetBSD MFC after: 1 week
* Checkpoint reworking the man page to be more useful to non-developers,njl2004-12-141-230/+197
| | | | document some debug output variables, etc. Mdoc assistance welcome.
* Unify ACPI_DEBUG support for all OEM drivers under ACPI_OEM. Since more thannjl2004-12-137-4/+16
| | | | | | | one will never be supported on the same platform, this does not hurt debugging. MFC after: 3 days
* The format of the MLINKS entries is pairwise, not MLINKS += wise. Soimp2004-12-131-2/+6
| | | | | list kernel_mount.9 the required number of times to allow make installworld to complete.
* Make code MPSAFE.rik2004-12-133-181/+768
| | | | | | | | You could turn this off by debug.mpsafenet=0 for full network stack or via debug.{cp|cx|ctau}.mpsafenet for cp(4), cx(4) and ctau(4) accordingly. MFC after: 10 days
* Make kttcp work in -stable (and -current)gallatin2004-12-132-537/+37
| | | | | | | | | | | | | | | | | | | - Removed kttcp_sosend() and kttcp_soreceive() in favor of using sosend() / soreceive() with UIO_NOCOPY. The locking changes in the socket layer make merging the kttcp_so* and the so* routines a real pain. It was a lot easier to just use the integrated routines. - Eliminated KTTCP_MAX_XMIT restrictions on send and receive sizes because I encountered no problems with streams larger than MAX_INT. MAX_INT bytes is only good for a few seconds at 4Gb/sec ;) - Removed Giant from send / recv routines. - Fixed character device protos, cdevsw init, etc, to work in 5.x / 6.x Reviewed by: sam MFC after: 1 week
* Add kernel_mount.9 and appropriate MLINKS to the build.trhodes2004-12-131-0/+3
|
* Add a manual page to document phk's mount work.trhodes2004-12-131-0/+208
| | | | Reviewed by: phk (content), brueffer (grammar and markup)
* correct module dependencysam2004-12-131-1/+1
|
* correct module dependencysam2004-12-131-1/+1
|
* Add ``enable echo''.brian2004-12-134-8/+8
|
* LINT defines things which compile in code that as referring to the oldimp2004-12-131-2/+2
| | | | | | | a_desc element. change this to the new a_gen.a_desc to reflect changes to vnode_if.h generation. Noticed by: tinderbox, phk
* Mention the necessity for ``enable echo'' in ppp(8).brian2004-12-131-0/+7
|
* - If delivering a signal will result in killing a process that has ajeff2004-12-131-0/+2
| | | | | | nice value above 0, set it to 0 so that it may proceed with haste. This is especially important on ULE, where adjusting the priority does not guarantee that a thread will be granted a greater time slice.
* Exporting of NTFS filesystem broke in rev 1.70. Fix it.ssouhlal2004-12-131-7/+4
| | | | Approved by: phk, grehan (mentor)
* Back the warnings down from 9 to 6, since that is the highest we have. Ifobrien2004-12-131-1/+1
| | | | we added a WARNS 7 level, it shouldn't be assumed this code will pass it.
* Handle select() returning -1/EINTR - fairly common in these KSE days.brian2004-12-131-1/+5
| | | | | PR: 74972 MFC after: 3 days
* - Take up a 'slot' while we're on the assigned queue, waiting to bejeff2004-12-131-16/+16
| | | | | posted to another processor. Otherwise, kern_switch() gets confused and tries to sched_add(NULL).
* bioq_insert_head() function is already in subr_disk.c.pjd2004-12-131-10/+0
|
* Add bioq_insert_head() function.pjd2004-12-132-0/+8
| | | | OK'd by: phk
* Implement an ``enable/disable echo'' option, defaults to off.brian2004-12-135-35/+109
| | | | | | | | | This allows LCP ECHOs to be enabled independently of LQR reports. Note: This introduces a change in the default behaviour (search for lqr and echo in the man page). I'll update UPDATING to reflect this. PR: 74821
* Add myself to the committers' birthday calendarniels2004-12-131-0/+1
| | | | Approved by nectar (mentor)
* Don't forget to bypass vnodes in corner cases.phk2004-12-131-8/+6
| | | | | Found by: kkenn and ports/shell/zsh Thanks to: jeffr
* Might as well get the right $NetBSD$ string while we are at it.julian2004-12-131-1/+1
| | | | MFC after: 1 week
* Diff reduction to NetBSD.julian2004-12-131-67/+59
| | | | | | | | | Now only things that are different between us and NetBSD show up. Means that these files are more of NetBSD style in some places but since thay are NetBSD files, um, that's ok. Obtained from: NetBSD MFC after: 1 week
* MFNetBSDjulian2004-12-131-43/+44
| | | | | | | | Whitespace diff reduction, formatting fixes and one actual arithmetic error that NetBSD have fixed. Obtained from: NetBSD MFC after: 1 week
* Another FNONBLOCK -> O_NONBLOCK.phk2004-12-131-2/+2
| | | | | Don't unconditionally set IO_UNIT to device drivers in write: nobody checks it, and since it was always set it did not carry information anyway.
* Use O_NONBLOCK instead of FNONBLOCK alias.phk2004-12-131-1/+1
|
* Rather than casting through a (void *) to avoid the aliasing warning,imp2004-12-131-3/+3
| | | | | | | | | do things correctly from an aliasing perspective. Put the vop_generic_args element as the first element for all the vop_*_args and adjust the code to take the address of that instead of the structure. OK'd based on a vague description by: phk
* Explicit panic in vop_read/vop_write for devicesphk2004-12-131-0/+2
|
* We don't need vnode.h, but do need uio.h in these files. vnode.himp2004-12-134-4/+4
| | | | | | shouldn't be included in drivers, generally, so remove them. OK'd by: phk
* Remove a stray critical_exit().scottl2004-12-131-1/+0
| | | | Submitted by: johan
* Add sys/uio.h explicitly, and move sys/vnode.h include to be moreimp2004-12-134-4/+8
| | | | | | | alphabetical. # vnode.h should not be included here, but it is required for proper decoding # of the flags args. This may change in the future...
* Explicitly panic vop_read/vop_write on fifos.phk2004-12-131-0/+2
|
* Re-enable link programming on resume. It appears the previous boundsnjl2004-12-131-2/+0
| | | | error had caused the hang and it has been corrected now.
* sys/vnode.h is inappropriate for a driver. Use sys/uio.h instead.imp2004-12-131-1/+1
|
OpenPOWER on IntegriCloud