summaryrefslogtreecommitdiffstats
path: root/sys
Commit message (Collapse)AuthorAgeFilesLines
* Fix Tx/Rx checksum offload ioctl handling and make Rx handler honoryongari2008-11-251-9/+26
| | | | | | | checksum offload configuration. Now checksum offload can be controlled by ifconfig(8). While I'm here add an additional check for interface capabilities before applying user's request.
* Make fxp(4) build with FXP_IP_CSUM_WAR.yongari2008-11-251-2/+2
|
* Sort head files and removed ununsed header file.yongari2008-11-251-13/+9
|
* Fix various whitespace botches, mostly having them at the end of a line.imp2008-11-251-15/+13
|
* Whitespace fix.yongari2008-11-253-25/+25
|
* Add support for Marvell PATA M88SX6121.rpaulo2008-11-252-0/+2
| | | | | | PR: 129154 Submitted by: João Barros <joao.barros at gmail.com> MFC after: 1 month
* Replace disclaimer with the one from COPYRIGHT. Joint authors aggreedimp2008-11-253-30/+33
| | | | to the change.
* Replace three magic constants with L1_S_SIZE, since that's what isimp2008-11-251-3/+3
| | | | really meant in those places.
* Tweak the disclaimer section of the license to match COPYRIGHT, forimp2008-11-2527-270/+298
| | | | | better or worse. Ok'd by folks that have additional copyrights to the files in cases where there's joint authorship.
* Correct typo in comment: thier -> theiremaste2008-11-241-1/+1
|
* Restore sfence semantics in mxge after the introductiongallatin2008-11-242-25/+25
| | | | of a global mfence based mb() in r185162
* Make sure arc4random(9) is properly seeded when /etc/rc.d/initrandom returns.cperciva2008-11-242-3/+24
| | | | | | Approved by: so (cperciva) Approved by: re (kensmith) Security: FreeBSD-SA-08:11.arc4random
* use consistent stylesam2008-11-241-10/+8
|
* Save the value read from TX queue CSR, so we don't test against ayongari2008-11-241-1/+1
| | | | | | stale one. Submitted by: sephe
* print the extended tx/rx descriptor for 5416 and later partssam2008-11-241-9/+32
|
* nuke special handling of RXORN interrupt; the hal marks the FATALsam2008-11-242-17/+4
| | | | | bit in the interrupt status when RXORN is hit and the chip requires a reset so our special handling was causing useless resets
* Remove DELAY() on hdac_audio_commit().mav2008-11-231-1/+0
| | | | I don't see the reason why it is needed.
* Eliminate an unused variable reported by coverityimp2008-11-231-4/+2
| | | | Submitted by: Ganbold
* It's possible that the dump device has gone away after it wasdwmalone2008-11-231-1/+1
| | | | | | | | configured, change the message to let people know this is a possibility. I've slightly changed the message from the one submitted by Pekka to keep the printf on one line. Submitted by: Pekka Savola <pekkas@netcore.fi>
* Make the new CCEQ macro a little more like the old one - first dodwmalone2008-11-231-1/+1
| | | | | | | | | the comparison between c and val and then compare val to _POSIX_VDISABLE. This avoids comparing c (which is usually of type char) to _POSIX_VDISABLE (which has value 0xff and may not be representable as a char). Reviewed by: ed
* Do not use soft PCM volume for digital outputs to allow SPDIF AC3 bypassmav2008-11-231-4/+8
| | | | to work. Soft PCM volume (and vchans) unable to bypass AC3 stream now.
* Don't use PRIV_ROOT. Here we check if user can share ZFS file system, sopjd2008-11-231-2/+1
| | | | | | PRIV_NFS_DAEMON seems best choice. Discussed with: rwatson
* Strictly differentiate digital and analog PCM devices according to codecmav2008-11-232-58/+91
| | | | | | | nodes capabilities. Add "Analog"/"Digital" marks to the pcm device names. I hope it will help new users easier accept concept of several PCM devices and understand exact purposes of that devices.
* Report card erase sector size as disk stripe size.mav2008-11-231-0/+2
|
* Improve detach handling: close races, flush queue.mav2008-11-231-13/+16
|
* Add backward compatibility ifdefs for non-multiq kernelskmacy2008-11-233-2/+29
|
* correct sortingsam2008-11-231-1/+1
|
* work around periodic leak on queue overrun by enabling coalescing of packets ↵kmacy2008-11-231-1/+1
| | | | | | in to work requests by default
* buf_ring_peek should return NULL if the ring is empty rather thankmacy2008-11-231-1/+4
| | | | whatever happened to be at cons_tail last time it was in use
* intr_machdep.h breaks build on some arches and is not neededkmacy2008-11-231-4/+0
|
* Unbreak previous commit.marcel2008-11-221-2/+0
|
* Include the QUICC backend in the kernel module.marcel2008-11-221-1/+2
| | | | PR: 127120
* Cast to uintptr_t before casting to void*. This allows themarcel2008-11-221-3/+3
| | | | | | | QUICC backend to be built on LP64 platforms. This makes it possible to include the QUICC backend in the kernel module. PR: 127120
* Remove unused variable.mav2008-11-221-3/+2
| | | | | Found with: Coverity Prevent(tm) CID: 3682
* Fix typo. Clear session stats instead of config and part of stats.mav2008-11-221-1/+1
| | | | | Found with: Coverity Prevent(tm) CID: 2472
* Remove unneeded NULL check. At first msg can't be null here and and at secondmav2008-11-221-2/+1
| | | | | | NG_FREE_MSG() also checks it. Found with: Coverity Prevent(tm)
* Whitespace cleanup for previous commit.mav2008-11-221-3/+3
|
* Handle device_get_children() errors in more correct way.mav2008-11-221-9/+12
| | | | | | | We shouldn't detach until all children are surely destroyed. Found with: Coverity Prevent(tm) CID: 2137
* Change check order to avoid array overrun.mav2008-11-221-2/+2
| | | | | Found with: Coverity Prevent(tm) CID: 2497
* Fix amd64 build and re-enable gptzfsboot.dfr2008-11-223-6/+16
|
* IFp4: Don't rely on disk IDs and always use vdev guids, which means always lookpjd2008-11-221-219/+69
| | | | | up for components by reading metadata. This might be slower when there are big number of disks in the system, but is definiately more reliable.
* IFp4: Finish implemnetation of chflags(2) for ZFS. While doing this I foundpjd2008-11-221-3/+54
| | | | | that zfs_access() can only handle VREAD, VWRITE and VEXEC, for the rest we need to use vaccess(9).
* IFp4: Don't free pathname too soon, debugging code is still using it.pjd2008-11-221-5/+4
|
* Busy ufs filesystem around block of code that does ".." lookup. Sincekib2008-11-221-1/+26
| | | | | | | | | | | | | | | | mnt_lock is before lock of any vnode on the mp, it uses LK_NOWAIT. Since MNTK_UNMOUNT may be transient, pdp lock is dropped when vfs_busy() failed, and operation is retried after some time. This way, ffs_vget() is not called on the mp that may be in the process of being destroyed by unmount. Check for the VI_DOOMED flag on pdp after its lock is reacquired, to better detect some situations where directory containing ".." entry is removed during the lookup. Reviewed by: tegge, attilio (previous version) Tested by: pho MFC after: 1 month
* Add sv_flags field to struct sysentvec with intention to provide descriptionkib2008-11-2216-20/+50
| | | | | | | | of the ABI of the currently executing image. Change some places to test the flags instead of explicit comparing with address of known sysentvec structures to determine ABI features. Discussed with: dchagin, imp, jhb, peter
* Unbreak LINT.jkoshy2008-11-224-0/+24
|
* Add flowid to mbuf to allow drivers to uniquelykmacy2008-11-221-0/+3
| | | | | identify connection flows to guarantee ordering across queues
* - enable multiple transmit queueskmacy2008-11-225-246/+109
| | | | | | | | | | | - invert sense of hw.cxgb.singleq tunable to hw.cxgb.multiq - don't wake up transmitting thread by default - add per tx queue ifaltq to handle ALTQ - remove several unused functions in cxgb_multiq.c - add several sysctls: multiq_tx_enable, coalesce_tx_enable, and wakeup_tx_thread - this obsoletes the hw.cxgb.snd_queue_len as ifq is replaced by a buf_ring
* convert calls to IFQ_HANDOFF to if_transmitkmacy2008-11-2210-15/+11
|
* Define mb(), rmb() and wmb() for real.marcel2008-11-221-3/+3
|
OpenPOWER on IntegriCloud