| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2000-06-26 | Instead of using an ugly hack with timeouts to wake up the process | Artur Grabowski | |
| waiting for resources, use the built-in facilities in tsleep. | |||
| 2000-06-23 | remove obsolete vtrace guts; art@ | Michael Shalayeff | |
| 2000-04-26 | Support for ext2fs rev. 1 (from NetBSD) | jasoni | |
| 2000-04-20 | Fix a problem that occurs when the filesystem fills up. | Artur Grabowski | |
| When the filesystem is able to allocate an indirect block but not the data block, it incorrectly unwinds the indirect block leaving a dangling pointer to a free block. This is said to fix the "freeing free block" panics people were seeing. From FreeBSD. | |||
| 2000-03-13 | typo | Artur Grabowski | |
| 2000-03-04 | Cut the number of dependencies to reduce problems on low-memory machines. | Artur Grabowski | |
| 2000-02-08 | prevent a panic when suspending on an idle system | Niklas Hallqvist | |
| 2000-02-07 | update to use new checkexp vfs operation | assar | |
| 2000-01-14 | Try to allocate the first indirect block in the same cg as the inode. | Artur Grabowski | |
| This improves read speed for files of moderate size (100-500k). From FreeBSD (mckusick@FreeBSD.ORG). | |||
| 2000-01-14 | rename DEBUGLOCKS to LOCKDEBUG to be consistent with the rest of the kernel | Artur Grabowski | |
| 1999-12-06 | Yet another solution to the mfs unmount/kill race (not ugly this time). | Artur Grabowski | |
| Require that the mount point is vfs_busy on entry to dounmount. | |||
| 1999-12-06 | When truncating directories by less than a block we shouldn't zero the | Artur Grabowski | |
| part of the block that we're deallocating. From NetBSD. | |||
| 1999-12-06 | ffs_init is a vfsop. Move it to ffs_vfsops. | Artur Grabowski | |
| 1999-12-06 | Always make sure that softdep mounts are not async. | Artur Grabowski | |
| 1999-12-06 | Don't compute values that we don't use later. | Artur Grabowski | |
| 1999-12-06 | #if 0 some broken unreached code. | Artur Grabowski | |
| 1999-12-05 | Integrate the latest soft updates code. | Artur Grabowski | |
| From mckusick@mckusick.com. | |||
| 1999-12-03 | Clean up ffs_mount. | Artur Grabowski | |
| 1999-12-02 | Opps. Back out a part of last commit that wasn't intended for commit yet. | Artur Grabowski | |
| 1999-11-26 | Yet another useless indentation fix (I don't like 'M' in my cvs updates). | Artur Grabowski | |
| 1999-11-17 | Avoid overflowing when calculating blocks at the third level of indirection. | Artur Grabowski | |
| From NetBSD. | |||
| 1999-11-17 | daddr_t vs. ufs_daddr_t cleanup | Artur Grabowski | |
| 1999-11-17 | Use u_int32_t for lfs inode number. | Artur Grabowski | |
| If we would ever change ino_t, the struct dinode would break. | |||
| 1999-10-15 | Yet another unmount/kill race. (I hate this) | Artur Grabowski | |
| 1999-09-10 | let the DOINGSOFTDEP macro always return 0 if FFS_SOFTUPDATES is not | Artur Grabowski | |
| defined. It will do that anyway, but this way the compiler can optimize better. | |||
| 1999-09-03 | Horrible workaround for hanging mount_mfs processes that received a signal. | Artur Grabowski | |
| After finding the third deadlock during dounmount, I decided that it would be less painful to let the unmounting be handled with a separate thread. XXX - this is a kludge. | |||
| 1999-08-30 | Kill mfsiom.h. We are not using anything from it. | Artur Grabowski | |
| 1999-08-17 | we're lucky that the return values from VOP_INACTIVE are ignored | Artur Grabowski | |
| 1999-06-01 | minor comment beautification | Todd C. Miller | |
| 1999-05-31 | New struct statfs with mount options. NOTE: this replaces statfs(2), | Todd C. Miller | |
| fstatfs(2), and getfsstat(2) so you will need to build a new kernel before doing a "make build" or you will get "unimplemented syscall" errors. The new struct statfs has the following featuires: o Has a u_int32_t flags field--now softdep can have a real flag. o Uses u_int32_t instead of longs (nicer on the alpha). Note: the man page used to lie about setting invalid/unused fields to -1. SunOS does that but our code never has. o Gets rid of f_type completely. It hasn't been used since NetBSD 0.9 and having it there but always 0 is confusing. It is conceivable that this may cause some old code to not compile but that is better than silently breaking. o Adds a mount_info union that contains the FSTYPE_args struct. This means that "mount" can now tell you all the options a filesystem was mounted with. This is especially nice for NFS. Other changes: o The linux statfs emulation didn't convert between BSD fs names and linux f_type numbers. Now it does, since the BSD f_type number is useless to linux apps (and has been removed anyway) o FreeBSD's struct statfs is different from our (both old and new) and thus needs conversion. Previously, the OpenBSD syscalls were used without any real translation. o mount(8) will now show extra info when invoked with no arguments. However, to see *everything* you need to use the -v (verbose) flag. | |||
| 1999-05-27 | Use lblktosize(fs, foo) instead of foo*fs->fs_bsize | Artur Grabowski | |
| 1999-04-28 | zap the newhashinit hack. | Artur Grabowski | |
| Add an extra flag to hashinit telling if it should wait in malloc. update all calls to hashinit. | |||
| 1999-04-25 | Reading large file from ext2fs caused vm_fault; gluk@ecsc.mipt.ru | Todd C. Miller | |
| Approved by csapuntz@openbsd.org | |||
| 1999-03-17 | If an mfs io is successful set b_resid to 0. | Artur Grabowski | |
| From Chuck Silvers <chs@netbsd>. | |||
| 1999-03-09 | POSIX says: | Artur Grabowski | |
| If O_TRUNC is set and the file did previously exist, upon successful completion the open() function shall mark for update the st_ctime and st_mtime fields of the file. | |||
| 1999-03-09 | workaround deadlocks in the unmount/kill mount_mfs race. | Artur Grabowski | |
| XXX - This is wrong. I have to rethink this/rewrite mfs. | |||
| 1999-02-26 | ext2fs is missing a LINK_MAX check too | Todd C. Miller | |
| 1999-02-26 | Better fix for i_nlink oflow from imp@openbsd.org | Todd C. Miller | |
| Move the check up with the other sanity checks and unlock the vnode properly and jump to abortit, not bad when >= LINK_MAX. | |||
| 1999-02-26 | compat with uvm vnode pager | Artur Grabowski | |
| 1999-02-26 | compat with uvm vnode pager | Artur Grabowski | |
| 1999-02-26 | add compatibility with uvm vnode pagers | Artur Grabowski | |
| 1999-02-25 | Add missing check for i_ffs_nlink > LINK_MAX; marcs@znep.com | Todd C. Miller | |
| 1999-01-22 | I'm stupid. Fix a panic when not DIAGNOSTIC. | Artur Grabowski | |
| 1999-01-16 | Do not lose st_blocks when files > 2GB | Niklas Hallqvist | |
| 1999-01-12 | fixed unused variable warnings when DIAGNOSTIC not defined | Todd C. Miller | |
| 1999-01-11 | panic prints a newline for you, don't do it in the panic string | Todd C. Miller | |
| 1999-01-10 | fix panic messages and put some code in DIAGNOSTIC | Artur Grabowski | |
| 1998-12-05 | There's still a good bit of lock debugging that needs to happen. Don't | Constantine Sapuntzakis | |
| turn it on by default or people will be getting too many panics. | |||
| 1998-12-05 | Finally, enable lock checking for FFS, which has full-blown lock support. | Constantine Sapuntzakis | |
| 1998-12-05 | Make sure the VLOCKSWORK diagnostic flag is correctly maintained. | Constantine Sapuntzakis | |
 |
index : src | |
| OpenBSD base system |
| summaryrefslogtreecommitdiff |