summaryrefslogtreecommitdiff
path: root/sbin/newfs/mkfs.c
AgeCommit message (Collapse)Author
2006-03-09Remove option -n from newfs as well as all references to fs_postbl()Pedro Martelletto
Various testing for a while, okay krw@
2005-12-19Add and use a few more constants and macros needed by UFS2Pedro Martelletto
No functional change
2005-08-08More informative error message when trying to create a filesystemOtto Moerbeek
larger than 1TB. ok deraadt@
2005-04-14handle snprintf out of range conditions; otto okTheo de Raadt
2005-01-06validate mfs size to avoid wrapping. ok miod@Otto Moerbeek
2004-10-14Do not generate either a too large super block or a negative numberOtto Moerbeek
of inodes per cylinder group for large block or fragment sizes. This allows for creation of filesystems with any legal block/fragment size combination. ok tedu@.
2004-09-10Bail out on errors when making very small filesystems. ok deraadt@Otto Moerbeek
2004-08-12Prevent core dump on large block sizes by checking block size againstOtto Moerbeek
MAXBSIZE. ok millert@
2004-07-06delint: offset arg to mmap is off_t and remove dead codeOtto Moerbeek
ok millert@
2004-06-26For all the mfs lovers:Otto Moerbeek
- Make the root of an mfs file system inherit the permission and owner/group of the mountpoint. Work by david@. - Add the ability to populate an mfs file system with the contents of a directory or a file system immediately after creation, controlled by the new -P mount option. Discussed with and tested by many, ok djm@ david@
2004-06-22acg.cg_nextfreeoff is already relative to &acg so don't subtractTodd C. Miller
&acg.cg_firstfield. Fixes a bogus "panic: cylinder group too big" I see sometimes when using mmap malloc. OK tholo@
2004-05-13provide a siginfo handler for -q run; idea from art's todo; millert@ okMichael Shalayeff
2004-01-13quiet warnings. better diff suggested by tom@ ok deraadt@Ted Unangst
2003-12-05madvise() the mfs storage area as randomly accessed as mfs does it's own ↵Michael Shalayeff
clustering; tedu@ ok
2003-11-06don't write magic until the file system is really done. preventsTed Unangst
fsck/mount unhappiness. hint from freebsd-arch ml. ok deraadt@
2003-11-03NULL for NULL pointers, not 0Ted Unangst
2003-11-03combine ugly printf(), exit() combo statements into err() or errx().Ted Unangst
mostly from tom cosgrove.
2003-08-25rename struct dinode to ufs1_dinode. clears the namespace and makesTed Unangst
way for some future work. no function changes yet. help testing otto@ and markus@
2003-07-16remove default limit of 16 cylinders per group. it's now set toTed Unangst
as many as fit with the other parameters given. change default frag size to 2048, bumping block size to 16k. from freebsd. ok deraadt@ and co.
2003-07-02missing protosTheo de Raadt
2003-06-25proto repairsTheo de Raadt
2003-06-11ansificationTheo de Raadt
2003-06-02Remove the advertising clause in the UCB license which BerkeleyTodd C. Miller
rescinded 22 July 1999. Proofed by myself and Theo.
2003-05-06remove FSIRAND define and make it non-optional. ok millert@Ted Unangst
2003-05-06after switch to mmap, we need to zero other malloc'd data. use calloc.Ted Unangst
2003-05-03The 'memleft' calcuations assume that the data address starts atTodd C. Miller
0, which is no longer the case. So instead of the hokey private malloc() implementation, just use mmap. From diffs I ran on anoncvs.usa 4 years ago.
2003-04-16fix race between mount(2) and exit(2) for mount_mfs.Markus Friedl
from netbsd (drochner, newfs.c 1.30, mkfs.c 1.29, no copyright changes) ok tedu@
2003-02-19Use pread/pwrite where sensible; also remove some bogus casts inTodd C. Miller
printf to long and long long for things that are int32_t.
2002-06-09rm trailing whitespaceTodd T. Fries
2002-05-29more snprintfTheo de Raadt
2001-11-05kill more registers;Mike Pechkin
millert@ ok
2001-07-07major -Wall cleanup, almost completeTheo de Raadt
2001-04-19Cleanup dirpref and synhronize superblock with freebsd.Grigoriy Orlov
From FreeBSD with little changes. art@ ok.
2001-04-13Fix overflow in superblock. From FreeBSD.Grigoriy Orlov
FreeBSD's log: > The ffs superblock includes a 128-byte region for use by temporary > in-core pointers to summary information. An array in this region > (fs_csp) could overflow on filesystems with a very large number of > cylinder groups (~16000 on i386 with 8k blocks). When this happens, > other fields in the superblock get corrupted, and fsck refuses to > check the filesystem. > > Solve this problem by replacing the fs_csp array in 'struct fs' > with a single pointer, and add padding to keep the length of the > 128-byte region fixed. Update the kernel and userland utilities > to use just this single pointer. > > With this change, the kernel no longer makes use of the superblock > fields 'fs_csshift' and 'fs_csmask'. Add a comment to newfs/mkfs.c > to indicate that these fields must be calculated for compatibility > with older kernels. art@ ok.
2001-04-04bye bye -U (soft updates).Grigoriy Orlov
2001-01-16_exit() in signal handlerTheo de Raadt
2000-09-22If a cylinder group won't fit inside of one block, adjust secpercylTodd C. Miller
and ntracks until it does, if possible. This allows one to newfs disks with funky (usually BIOS-generated) geometries without cranking the block size. This is really just a hack and a better solution would be to fake the geometries of zone-recorded disks since secpercyl is never accurate in that case anyway...
1999-12-03Allow enabling soft updates (with -U) on a new filesystem.Artur Grabowski
Allow soft updates on mfs filesystems (even if it's stupid).
1999-04-10Add -q flag to newfs and use it in the installTodd C. Miller
1997-11-17Work around a stackframe bug in gcc alpha, who cannot deal with frames > 64kNiklas Hallqvist
1997-06-03Make mkfs.c malloc its large buffers, which saves about 200K of bss sizegrr
when newfs is used in instbin and is otherwise harmless. Also add a little pro-forma malloc fail parnoia, though it's not really needed when the local malloc routine is used.
1997-06-03number of unallocated sectors message was based on modified fssize, make it ↵grr
deal with reality
1997-05-29mkfs() should not play with fssize; grr, PR#189Theo de Raadt
1997-03-27if lost+found generation code is enabled, create it with a nice modeTheo de Raadt
1997-02-23Inline fsirand(8) (at no noticable cost in speed).Todd C. Miller
1997-02-21Don't expose private copies of malloc(), calloc(), free(). Ewwww.Todd C. Miller
1997-02-11Remove occurences of headswitch and trackseek, they no longerTodd C. Miller
exist in struct fs.
1996-08-02proper column management for superblock backups; from freebsd, joerg/satoshiTheo de Raadt
1996-06-23update rcsidTheo de Raadt
1995-10-18initial import of NetBSD treeTheo de Raadt