summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* add -z option to zero driver statistics (but note we don't yet zerosam2009-02-133-3/+18
| | | | stats obtained from the hal)
* add stats no frags and missing stat for no dma bufferssam2009-02-131-1/+9
|
* backout r188327, it broke summary printingsam2009-02-133-53/+43
|
* add SIOCZATHSTATS ioctl to zero driver statisticssam2009-02-132-0/+7
|
* remove ic_stats; it was intended to accumulate stats from vaps as theysam2009-02-133-3/+0
| | | | were reaped but was never used and is inaccessible
* add driver stat to count tx drops due to insufficient frag bufferssam2009-02-132-3/+3
|
* strip M_FRAG & co mbufs coming down from ip_output, these flags aresam2009-02-131-1/+2
| | | | | re-used within net80211 to mark 802.11 frags so allowing them to leak through to the driver caused packets to be dropped in ath
* Restore some of the link handling code that was inadvertently removed inthompsa2009-02-131-12/+41
| | | | | | | r188412, this broke the GU-1000T adapters. Submitted by: yongari Pointy hat: me
* Add module dependency on miibus.thompsa2009-02-131-1/+3
| | | | Submitted by: yongari
* Receving VLAN oversized frames raise SIS_ISR_RX_ERR interrupt, soyongari2009-02-131-2/+3
| | | | | | | make sis_rxeof() handle that too. Previously it used to receive the frame and reset controller. PR: kern/131414
* Recognize AR5212_AR2317_REV2 in ar5312Probe()sam2009-02-131-0/+1
| | | | Submitted by: Pavel Roskin <proski@gnu.org>
* Check the exit flag at the start of the taskqueue loop rather than the end. Itthompsa2009-02-131-2/+2
| | | | | | | | is possible to tear down the taskqueue before the thread has run and the taskqueue loop would sleep forever. Reviewed by: sam MFC after: 1 week
* Remove unused ifaddr local variable in ioctl routine.rwatson2009-02-131-2/+0
| | | | MFC after: 3 days
* Remove unused ifaddr and ifreq local variables.rwatson2009-02-121-4/+0
| | | | MFC after: 3 days
* o replace ieee80211_node_reclaim by individual operations to release thesam2009-02-121-60/+2
| | | | | | | | | | references to iv_bss and the sta table; this is equivalent and causes direct reclaim of the old bss node when any references in packets inflight are reclaimed (previously the old node would sit in the bss table until the inactivity processing reclaimed it) o remove ieee80211_node_reclaim now that it's only use is gone Reviewed by: avatar, cbzimmer
* To prevent various race conditions in the RAS code, store and restore thecognet2009-02-124-4/+23
| | | | | | values in ARM_RAS_START and ARM_RAS_END at context switch time. MFC after: 1 week
* Do not set thread0.td_frame to a bogus value, as it's going to overwrite thecognet2009-02-121-1/+0
| | | | | | | | thread0 pcb, while the board-dependant code already set a good trapframe. Reported by: Mark Tinguely <tinguely at casselton d0t net> MFC after: 1 week
* Unbreak build by making svr4 conditional on i386.cognet2009-02-121-1/+2
|
* Properly invalidate highest pts number when calling setttyent().ed2009-02-121-5/+4
| | | | | | | | | When calling setttyent() after calling endttyent(), pts_valid will never be set to 1, because the readdir()-loop will likely never vind a pts that has a higher number than before. Simplify the code by removing pts_valid. We'll just set maxpts to -1 when we don't have a valid count yet.
* Remove pts(4) entries from /etc/ttys.ed2009-02-127-3584/+0
| | | | | | | | | | | | | | | Even though I increased the amount of pts(4) entries in /etc/ttys some time ago, I didn't realize back then those entries shouldn't have been there in the first place. I just looked at the getttyent() source code and it turns out when you call setttyent(), it walks through /dev/pts and looks for the device with the highest number. After you receive EOF's from getttyent(), it makes up entries for pts(4) devices. This means that adding entries for pts(4) is somewhat harmful, because if you now traverse the list, you get redundant entries, so just remove them.
* Make ttyslot(3) work with pts(4) devices.ed2009-02-121-6/+5
| | | | | | | It seems ttyslot() calls rindex(), to strip the device name to the last slash, but this is obviously invalid. /dev/pts/0 should be stripped until pts/0. Because /etc/ttys only supports TTY names in /dev/, just strip this piece of the pathname.
* Add a ieee80211_waitfor_parent() function that will wait for all deferredthompsa2009-02-124-2/+18
| | | | | | | | parent interface tasks to complete. This had been added to the ioctl path but it is also need elsewhere like detach so its safe to teardown. Reported by: Hans Petter Selasky Submitted by: sam
* Remove obsolete C preprocessor assertions.rdivacky2009-02-121-1/+1
| | | | Approved by: kib (mentor)
* turn off ttyv*; not aware of any arm-based systems with sysconssam2009-02-121-8/+8
| | | | Reviewed by: imp
* Don't add a bwrite() symbol, it breaks the build when building newfscognet2009-02-122-16/+16
| | | | | | statically. Instead, bring in a stripped down version of sbwrite(), and add the offset to every bwrite() calls.
* Fix a few problems related to building modules in /sys/modules (notn_hibma2009-02-126-4/+7
| | | | | | | | | | checked whether this applies to builds in /sys/*/compile/* as well): - Create empty opt_*.h files were missing - Hook up svr4 to the build. It compiles fine here, so no reason to disconnect it in the Makefile. were missing - Hook up svr4 to the build. It compiles fine here, so no reason to disconnect it in the Makefile.
* Fix spelling a bit.mav2009-02-121-3/+3
|
* Add support for the Sunix SUN1889-based dual parallel port card.kevlo2009-02-121-0/+6
| | | | | PR: kern/128219 Submitted by: Thinker K.F. Li <thinker at branda dot to>
* s/succeded/succeeded/ in comments.mav2009-02-121-3/+3
|
* Enhance debug messages and attempt to unify them into one format where possible.sbruno2009-02-121-10/+12
| | | | Reviewed by: scottl
* Update comments around various structs.sbruno2009-02-123-9/+27
| | | | Add speeds S800, S1600 and S3200
* Rewrite get_pv_entry() to match expectations of the rest of theimp2009-02-121-11/+78
| | | | | | | | | | kernel. Rather than just kick off the page daemon, we actively retire more mappings. The inner loop now looks a lot like the inner loop of pmap_remove_all. Also, get_pv_entry can't return NULL now, so remove panic if it did. Reviewed by: alc@
* pmap_kenrel() was recently deleted from pmap.h. It was still usedimp2009-02-121-2/+2
| | | | | here, but inconsistently. Change all instances of it to kernel_pmap to match rest of code.
* fix ARM_USE_SMALL_ALLOC after memory layout changessam2009-02-111-7/+10
| | | | Submitted by: cognet
* fix both instances of namesam2009-02-111-1/+1
| | | | Pointy hat: sam
* Reorder dacleanup() and cdcleanup() slightly so that the sysctl context isjhb2009-02-112-8/+9
| | | | | | | freed while the periph lock is not held. While here, wait until after freeing the softc before reacquiring the periph lock. Tested by: sbruno
* - Consolidate error handling in the cd9660 and udf mount routines.jhb2009-02-112-24/+26
| | | | | | | | | - Always read the character device pointer while the associated devfs vnode is locked. Also, use dev_ref() to obtain a new reference on the vnode for the mountpoint. This reference is released on unmount. This mirrors the earlier fix to FFS. Reviewed by: kib
* - If the g_access() call for the initial root mount fails, then fullyjhb2009-02-111-6/+6
| | | | | | | | cleanup. Before the GEOM consumer would not have been closed. - Bump the reference on the character device being mounted while the associated devfs vnode is locked. Reviewed by: kib
* fix typo in AH_CHIP definitionsam2009-02-111-1/+1
| | | | Submitted by: Pavel Roskin <proski@gnu.org>
* gcc 4.3.2 examines getLowerUpperIndex() and concludes that it's notsam2009-02-111-0/+1
| | | | | | | | | | | guaranteed to initialize its two last arguments. Therefore, there is a warning in the subsequent caller ar5416FillVpdTable(), which doesn't initialize those arguments. Change getLowerUpperIndex() to assign values to indexL and indexR even in the case of assertion failure. Submitted by: Pavel Roskin <proski@gnu.org>
* Let stat(1) use fdevname(3).ed2009-02-111-12/+20
| | | | | | | | | | | | | | Because we now have a reliable library function that converts file descriptors to character device names, let stat(1) use this. This means it can now do the following: $ stat -f %N /dev/pts/0 I've changed main() to set file properly, so output() is never called with file set to NULL. Approved by: dougb (older version, still used devname)
* Add two new routines: fdevname() and fdevname_r().ed2009-02-117-20/+90
| | | | | | | | | | | | | | A more elegant way of obtaining a name of a character device by its file descriptor on FreeBSD, is to use the FIODGNAME ioctl. Because a valid file descriptor implies a file descriptor is visible in /dev, it will always resolve a valid device name. I'm adding a more friendly wrapper for this ioctl, called fdevname(). It is a lot easier to use than devname() and also has better error handling. When a device name cannot be resolved, it will just return NULL instead of a generated device name that makes no sense. Discussed with: kib
* improve key index map reclaim debug msgsam2009-02-111-6/+5
|
* oops, fix octal escape sequencesam2009-02-111-1/+1
|
* - Use the correct argument when determining the buffer size.lulf2009-02-111-1/+1
| | | | | PR: kern/131575 MFC after: 2 days
* remove bogus reference reclaim on the slave-side bss node on beacon misssam2009-02-111-1/+0
|
* fix mis-numbering of IEEE80211_MSG_BITSsam2009-02-111-4/+4
|
* Serialize write() calls on TTYs.ed2009-02-114-6/+37
| | | | | | | | | | | | | | | | | | | | Just like the old TTY layer, the current MPSAFE TTY layer does not make any attempt to serialize calls of write(). Data is copied into the kernel in 256 (TTY_STACKBUF) byte chunks. If a write() call occurs at the same time, the data may interleave. This is especially likely when the TTY starts blocking, because the output queue reaches the high watermark. I've implemented this by adding a new flag, TTY_BUSY_OUT, which is used to mark a TTY as having a thread stuck in write(). Because I don't want non-blocking processes to be possibly blocked by a sleeping thread, I'm still allowing it to bypass the protection. According to this message, the Linux kernel returns EAGAIN in such cases, but I think that's a little too restrictive: http://kerneltrap.org/index.php?q=mailarchive/linux-kernel/2007/5/2/85418/thread PR: kern/118287
* Modify fdcopy() so that, during fork(2), it won't copy file descriptorsrwatson2009-02-111-1/+2
| | | | | | | | | | | | | | | | | | | | | | | from the parent to the child process if they have an operation vector of &badfileops. This narrows a set of races involving system calls that allocate a new file descriptor, potentially block for some extended period, and then return the file descriptor, when invoked by a threaded program that concurrently invokes fork(2). Similar approches are used in both Solaris and Linux, and the wideness of this race was introduced in FreeBSD when we moved to a more optimistic implementation of accept(2) in order to simplify locking. A small race necessarily remains because the fork(2) might occur after the finit() in accept(2) but before the system call has returned, but that appears unavoidable using current APIs. However, this race is vastly narrower. The fix can be validated using the newfileops_on_fork regression test. PR: kern/130348 Reported by: Ivan Shcheklein <shcheklein at gmail dot com> Reviewed by: jhb, kib MFC after: 1 week
* Fix lptopen() and lptclose() to not trash the state of the HAVEBUS flagjhb2009-02-111-11/+8
| | | | | | | | in 'sc_state'. This allows the lpt_release_ppbus() calls in those two routines to actually release the ppbus and thus fixes the hangs noticed with the lpt(4) driver since the recent ppbus changes. The old lpt(4) driver didn't actually check the HAVEBUS flag in lpt_release_ppbus() which is why these bugs weren't noticed before.
OpenPOWER on IntegriCloud