summaryrefslogtreecommitdiff
path: root/sys/miscfs
AgeCommit message (Expand)Author
2007-10-17Split I/O-related routines (getpages, putpages, etc.) which are heavilypooka
2007-10-11This was using mutex_exit where mutex_enter was required.ad
2007-10-10umapm_hashlock is a mutex.ad
2007-10-10Merge from vmlocking:ad
2007-10-08Merge run time accounting changes from the vmlocking branch. These makead
2007-10-08Merge brelse() changes from the vmlocking branch.ad
2007-10-08Merge file descriptor locking, cwdi locking and cross-call changesad
2007-10-07Update the file system copy-on-write handler.hannken
2007-09-01Make bioops a pointer and point it to the softdeps struct in softdeppooka
2007-08-03cleanup unused prototypepooka
2007-08-03ANSI-fypooka
2007-07-31* nuke the nameidata parameter from VFS_MOUNT(). Nobody on tech-kernpooka
2007-07-29It's not a good idea for device drivers to modify b_flags, as they don'tad
2007-07-29It's not a good idea for device drivers to modify b_flags, as they don'tad
2007-07-27whoops, forgot to commit this a while back: initialize new vnode sizepooka
2007-07-27vop_mmap parameter changepooka
2007-07-26Use eopnotsupp() instead of vfs_stdsuspendctl() and retire the latter.pooka
2007-07-23Initialize size and/or writesize when creating a vnode.pooka
2007-07-22Retire uvn_attach() - it abuses VXLOCK and its functionality,pooka
2007-07-22Don't allow getcwd() on procfs vnodes and provide "/" as the pathpooka
2007-07-21nuke homegrown getcwd_common() declpooka
2007-07-17Make set_statvfs_info() take a parameter for the vfs name insteadpooka
2007-07-12Change the VFS_MOUNT() interface so that the 'data' buffer passed to thedsl
2007-07-09Merge some of the less invasive changes from the vmlocking branch:ad
2007-07-08* allow unmount even if rootvp has a usecount > 1 provided thatpooka
2007-07-02I'm all for redundant and failsafe computing, but ...pooka
2007-06-30VOP_LOCK() doesn't handle LK_RETRY, call vn_lock() insteadpooka
2007-06-30Updates for changes prototype of kauth_cred_set/getgroups().dsl
2007-06-30Using POOL_INIT here makes no sense, since file systems always havepooka
2007-06-05improve post-ubc file overwrite performance in common cases.yamt
2007-05-26In /proc/<pid>/statm, avoid leaking buffer space if the attempt to getagc
2007-05-25Use a bit more common code for the MULTIPROCESSOR and !MULTIPROCESSORagc
2007-05-25Various changes for better Linux emulation:agc
2007-05-24use PRIu64, not llu, to unbork on 64-bit platforms.dogcow
2007-05-24Extend the Linux emulation of /proc to includeagc
2007-05-17Fstrans_start() always returns zero, so change its type to void.hannken
2007-05-13use a cached value of v_size. no functional changes.yamt
2007-04-24Split the VOP interface part of genfs_putpages() from the code. The newperseant
2007-04-16Don't expand RCS id of ancestor file. The id itself is actually copiedenami
2007-04-16define a pager flag PGO_RECLAIM, similar to FSYNC_RECLAIM, and use itchs
2007-04-08Remove now obsolete vn_start_write() and vn_finished_write() andhannken
2007-04-07Remove calls to now obsolete vn_start_write() and vn_finished_write().hannken
2007-04-04Unfortunately, missed procfs_proc_unlock() in previous.rmind
2007-04-04procfs_readlink: Handle a possible fail of fd_getfile(), also, wermind
2007-04-01Instead of reading and writing little by little, allocate memory andchristos
2007-04-01return a page less than the actual top of stack so that linux-java works.christos
2007-03-11Remove useless cast.ad
2007-03-09- Make the proclist_lock a mutex. The write:read ratio is unfavourable,ad
2007-03-04Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.christos
2007-03-03Don't prepend rootvnode to the path in non-NULL case for exe links.salo