summaryrefslogtreecommitdiff
path: root/sys/ufs
AgeCommit message (Expand)Author
2000-07-24Need string.h for memset() prototype.mycroft
2000-07-22change the lf_advlock() arguments fromjdolecek
2000-07-22ext2fs_reload(), ext2fs_mountfs(): do devvp locking same way as ffsjdolecek
2000-07-13XXX Use of hzto() return value needs to be double-checked here.thorpej
2000-07-06Fix so non-kernel code will compile (_LKM)perseant
2000-07-05Clean up accounting of lfs_uinodes (dirty but unwritten inodes).perseant
2000-07-05kern.maxvnodes (and hence desireddquot) depends more directly on NVNODE thanjdolecek
2000-07-04Fix errors observed while trying to fill the filesystem with yesterday'sperseant
2000-07-04Add missing second argument to tablefull call. I *think* the addedmjacob
2000-07-03Fix i_ffs_blocks in fragment extension case where fragment has not yetperseant
2000-07-03Correct typo in previous.fvdl
2000-07-03i_lfs_effnblks fixes. Put debugging printfs under #ifdef DEBUG_LFS.perseant
2000-07-03Allow the number of free segments reserved for the cleaner to beperseant
2000-07-01Move SET_ENDOP after vrele to avoid deactivating vnode twice, ifperseant
2000-06-30Rearrange code around getnewvnode as was already done for ffs, to avoidfvdl
2000-06-28remove include of <vm/vm.h> and <uvm/uvm_extern.h>mrg
2000-06-28<vm/vm.h> -> <uvm/uvm_extern.h>mrg
2000-06-27Due to popular demand, change vinsheadfree to ungetnewvnode to makefvdl
2000-06-27In ffs_vget, do not hold ufs_haslock across the call to getnewvnode.fvdl
2000-06-27Fixes associated with filling an LFS:perseant
2000-06-27From John Evans <jevans@cray.com>: use datosn() to convert to segmentperseant
2000-06-27We shouldn't be defining DEBUG and DIAGNOSTIC on our own; these may havepk
2000-06-22fix my own typo, grr....perseant
2000-06-22Read i_ffs_gen from the version number in the Ifile during lfs_valloc,perseant
2000-06-22Update lfs_vunref for the fact that now a vnode can be locked with noperseant
2000-06-22Moved here from gnu/sys/ufs/ffsfvdl
2000-06-22Copyright changed.fvdl
2000-06-16make it compile (fix typo)perseant
2000-06-16ignore the softdep flags when mounting and there's no softdep in the kernel.matt
2000-06-15Allow MNT_SOFTDEP to be passed in via the mount(2) system call, do notfvdl
2000-06-11Bitbucket MFS I/O after vfs_shutdown has started..sommerfeld
2000-06-06Don't try to inactivate dirop vnodes that are still in the middle ofperseant
2000-06-06Protect inode free list with seglock, instead of separate lock, so thatperseant
2000-05-31Make this build. (Balance parenthesis.fredb
2000-05-31update for IN_ACCESSED changesperseant
2000-05-30Adjust where IN_MODIFIED and IN_ACCESSED are cleared (as in the FFS code).mycroft
2000-05-30Back out previous kluge.mycroft
2000-05-30In ffs_update():mycroft
2000-05-30Mark an inode as changed after a rename. It wasn't before in the softdepfvdl
2000-05-30Don't try to "correct" accounting for fragments being extended but whichperseant
2000-05-29According to Frank, buffers with dependencies *are* left on v_dirtyblks, somycroft
2000-05-29Pull in IN_ACCESSED changes and some MNT_LAZY `bug fixes' from FFS to EXT2FS.mycroft
2000-05-29Pull in IN_ACCESSED changes and some MNT_LAZY `bug fixes' from FFS.mycroft
2000-05-29Use LIST_{FIRST,NEXT,EMPTY}().mycroft
2000-05-29Add a new inode flags called IN_ACCESSED. This used in place of IN_MODIFIEDmycroft
2000-05-29Never call softdep_sync_metadata() in the FSYNC_RECLAIM case. Any pendingmycroft
2000-05-29In ffs_fsync(), remove the FSYNC_RECLAIM special case, so that it properlymycroft
2000-05-29MNT_WAIT -> FSYNC_WAITmycroft
2000-05-28Pull in indirect block unwind code from FFS.mycroft
2000-05-28DTRT when unwinding multiple levels.mycroft