#
69289 |
| 08-May-1995 |
mckusick |
update from margo |
#
68674 |
| 30-Mar-1995 |
mckusick |
lfs_init moves to lfs_vfsops.c |
#
68550 |
| 21-Mar-1995 |
mckusick |
daddr_t => ufs_daddr_t |
#
68549 |
| 21-Mar-1995 |
mckusick |
convert from timespec to explicit types |
#
65241 |
| 30-Dec-1993 |
mckusick |
new queue structure |
#
64610 |
| 23-Sep-1993 |
bostic |
changes for 4.4BSD-Lite requested by USL |
#
64525 |
| 21-Sep-1993 |
bostic |
changes for 4.4BSD-Lite requested by USL |
#
64418 |
| 05-Sep-1993 |
bostic |
Cleanups for 4.4BSD-Lite |
#
63375 |
| 11-Jun-1993 |
bostic |
4.4BSD snapshot (revision 8.1); add 1993 to copyright |
#
57804 |
| 02-Feb-1993 |
mckusick |
vinvalbuf takes parameter to indicate interrupt |
#
56867 |
| 17-Nov-1992 |
margo |
use di_inumber not di_inum. |
#
56476 |
| 08-Oct-1992 |
margo |
Pass flags to vinvalbuf. Use new queue structures. |
#
56350 |
| 29-Sep-1992 |
margo |
Increment blocks available after truncate rather than decrement. |
#
56158 |
| 02-Sep-1992 |
margo |
lfs_truncate may be removing blocks that have been made dirty in the cache. Since there are no disk addresses for these blocks yet, they weren't being counted, and lfs_bavail was never being updat
lfs_truncate may be removing blocks that have been made dirty in the cache. Since there are no disk addresses for these blocks yet, they weren't being counted, and lfs_bavail was never being updated. Now we count them.
show more ...
|
#
56026 |
| 25-Aug-1992 |
bostic |
SEGDEC should look for 0 instead of UNASSIGNED disk addresses since it is reading the value off dinode or indirect page. |
#
55938 |
| 21-Aug-1992 |
bostic |
Formatting. Change UBWRITE to VOP_BWRITE, add inode accounting. Fix inode block accounting and free space accounting. |
#
55807 |
| 01-Aug-1992 |
bostic |
delete VERBOSE #ifdef's |
#
55786 |
| 29-Jul-1992 |
bostic |
check for negative bytes in seguse update; increment version number after blocks are free (for the cleaner) |
#
55589 |
| 23-Jul-1992 |
bostic |
Do block accounting in terms of disk sectors, not file system blocks. |
#
55549 |
| 22-Jul-1992 |
bostic |
truncate doesn't need to be synchronous add LFS_SYNC, so that "synchronous" operations don't cause deadlock |
#
55459 |
| 20-Jul-1992 |
bostic |
indirect block count should be fixed by revision 7.23 in lfs_segment.c |
#
55456 |
| 20-Jul-1992 |
bostic |
put in short symlink support; fix block counts make truncate not wait -- FFS truncate must wait because the in-core bit maps could be updated before the inode gets to disk which would be bad; LFS do
put in short symlink support; fix block counts make truncate not wait -- FFS truncate must wait because the in-core bit maps could be updated before the inode gets to disk which would be bad; LFS doesn't have this issue, and can't wait because the segment writer may deadlock with an VOP_REMOVE operation
show more ...
|
#
54765 |
| 07-Jul-1992 |
mckusick |
Chris sez time is volatile |
#
54694 |
| 05-Jul-1992 |
bostic |
vget becomes a VFS op (Kirk); describe ap (Kirk); use ap instead of passing arguments for stackable file systems (Heideman); |
#
54264 |
| 23-Jun-1992 |
bostic |
directory ops, unmount, minor cleanup; from Margo Seltzer |