aboutsummaryrefslogtreecommitdiff
path: root/sys/kern/vfs_bio.c
Commit message (Expand)AuthorAgeFilesLines
* Eliminate inconsistency in the setting of the B_DONE flag. Specifically,Alan Cox2005-07-201-1/+0
* - Add and enhance asserts related to the wrong bufobj panic.Jeff Roberson2005-06-141-1/+1
* - Split one KASSERT in bremfree() into two to aid in debugging.Jeff Roberson2005-06-131-1/+3
* Fix a serious deadlock with the NFS client. Given a large enoughBrian Feldman2005-06-101-1/+1
* - My sub-par public school education has been exposed. s/sentinal/sentinel/Jeff Roberson2005-06-091-4/+4
* - Under heavy IO load the buf daemon can run for many hundereds ofJeff Roberson2005-06-081-7/+29
* - Add bufobj_wrefl() to add a write ref to a bufobj that is already locked.Jeff Roberson2005-05-301-0/+9
* - Remove long dead splbio() calls and comments relating to the oldJeff Roberson2005-04-301-71/+4
* - Don't acquire Giant before calling b_biodone, individual consumers areJeff Roberson2005-04-301-6/+0
* - Add two KASSERTs to prevent us from recycling a buf that is still on aJeff Roberson2005-04-221-0/+6
* - Add information about the buf lock to db_show_buffer.Jeff Roberson2005-03-251-0/+18
* - Lock access to the buffer_map with the vm_map lock. In 4.x this wasJeff Roberson2005-03-081-0/+5
* Make various vnode related functions staticPoul-Henning Kamp2005-02-101-3/+3
* - Add more information to the getnewbuf() recycling KTR.Jeff Roberson2005-02-101-2/+4
* - Remove an invalid KASSERT added in recent background write reshuffling.Jeff Roberson2005-02-081-3/+0
* Background writes are entirely an FFS/Softupdates thing.Poul-Henning Kamp2005-02-081-126/+5
* - Don't release BKGRDINPROG until after we've bufdone'd the copy.Jeff Roberson2005-02-051-14/+15
* - Don't drop the wref on the bufobj until after bufdone() has completed.Jeff Roberson2005-01-281-7/+10
* Don't use VOP_GETVOBJECT, use vp->v_object directly.Poul-Henning Kamp2005-01-251-10/+8
* Kill the VV_OBJBUF and test the v_object for NULL instead.Poul-Henning Kamp2005-01-241-4/+3
* - Add CTR calls to trace the lifecycle of a buffer.Jeff Roberson2005-01-241-79/+84
* Add BO_SYNC() and add a default which uses the secret vnode pointerPoul-Henning Kamp2005-01-111-0/+8
* Remove the unused credential argument from VOP_FSYNC() and VFS_SYNC().Poul-Henning Kamp2005-01-111-1/+1
* - Eliminate the acquisition and release of the bqlock in bremfree() byJeff Roberson2004-11-181-38/+66
* Retire b_magic now, we have the bufobj containing the same hint.Poul-Henning Kamp2004-11-041-1/+0
* Change buf->b_object to buf->b_bufobj->bo_objectPoul-Henning Kamp2004-11-041-14/+16
* whitespacePoul-Henning Kamp2004-11-041-4/+2
* Remove buf->b_dev field.Poul-Henning Kamp2004-11-041-7/+2
* The synchronization provided by vm object locking has eliminated theAlan Cox2004-11-031-2/+0
* Remove the last call in the system to VOP_SPECSTRATEGY(): We can noPoul-Henning Kamp2004-10-291-7/+2
* Give dev_strategy() an explict cdev argument in preparation for removingPoul-Henning Kamp2004-10-291-13/+49
* Lock bp->b_bufobj->b_object instead of bp->b_objectPoul-Henning Kamp2004-10-281-10/+10
* The island council met and voted buf_prewrite() home.Poul-Henning Kamp2004-10-261-5/+3
* Put the I/O block size in bufobj->bo_bsize.Poul-Henning Kamp2004-10-261-8/+1
* Hold the lock on the containing vm object when callingAlan Cox2004-10-261-0/+2
* Remove vnode->v_bsize. This was a dead-end.Poul-Henning Kamp2004-10-251-7/+0
* Use VM_ALLOC_NOBUSY to eliminate vm_page_wakeup() calls and the acquisitionAlan Cox2004-10-251-10/+3
* Collapse vnode->v_object and buf->b_object into bufobj->bo_object.Poul-Henning Kamp2004-10-251-3/+6
* Move the buffer method vector (buf->b_op) to the bufobj.Poul-Henning Kamp2004-10-241-37/+18
* Add b_bufobj to struct buf which eventually will eliminate the need for b_vp.Poul-Henning Kamp2004-10-221-56/+66
* Move the VI_BWAIT flag into no bo_flag element of bufobj and call it BO_WWAITPoul-Henning Kamp2004-10-211-8/+66
* use dev_re[fl]thread() rather than home rolled versions.Poul-Henning Kamp2004-09-241-8/+2
* Eliminate DEV_STRATEGY() macro: call dev_strategy() directly.Poul-Henning Kamp2004-09-231-3/+12
* Do not refcount the cdevsw, but rather maintain a cdev->si_threadcountPoul-Henning Kamp2004-09-231-5/+11
* Remove unused B_WRITEINPROG flagPoul-Henning Kamp2004-09-151-1/+1
* undent some functions a bit.Poul-Henning Kamp2004-09-151-159/+161
* stylistic polishing.Poul-Henning Kamp2004-09-151-37/+57
* Remove the buffercache/vnode side of BIO_DELETE processing inPoul-Henning Kamp2004-09-131-8/+1
* Eliminate unused second argument to reassignbuf() and simplify itPoul-Henning Kamp2004-07-251-2/+2
* Neuter this warning for now, I think I know the remaining issues.Poul-Henning Kamp2004-07-251-1/+2