summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* If the initial attempt to write ${entropy_file} to /dev/random failssheldonh2000-07-241-2/+10
| | | | | | | and the randomdev.ko module is not resident, try to load the module and perform the write again. Reviewed by: markm
* Document device tdfx and options TDFX_LINUX.asmodai2000-07-242-0/+6
|
* truncate.c:sheldonh2000-07-242-4/+13
| | | | | | | | | | | Do not include unused header files. Add rcsid. Change copyright. truncate.1: Add AUTHORS section. Submitted by: charnier
* Get rid of the stupid bug where diff sticks trash in your generatedgreen2000-07-241-3/+2
| | | | | | | | | | diff output; there being no trailing newline caused it to bitch and moan and ruin diffs. That's dumb, if I may say so myself. Yes, this file is coming off the vendor branch. It hasn't been updated in 6 years. Approved by: peter
* This patch corrects the first round of panics and hangs reportedmckusick2000-07-2424-103/+298
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | with the new snapshot code. Update addaliasu to correctly implement the semantics of the old checkalias function. When a device vnode first comes into existence, check to see if an anonymous vnode for the same device was created at boot time by bdevvp(). If so, adopt the bdevvp vnode rather than creating a new vnode for the device. This corrects a problem which caused the kernel to panic when taking a snapshot of the root filesystem. Change the calling convention of vn_write_suspend_wait() to be the same as vn_start_write(). Split out softdep_flushworklist() from softdep_flushfiles() so that it can be used to clear the work queue when suspending filesystem operations. Access to buffers becomes recursive so that snapshots can recursively traverse their indirect blocks using ffs_copyonwrite() when checking for the need for copy on write when flushing one of their own indirect blocks. This eliminates a deadlock between the syncer daemon and a process taking a snapshot. Ensure that softdep_process_worklist() can never block because of a snapshot being taken. This eliminates a problem with buffer starvation. Cleanup change in ffs_sync() which did not synchronously wait when MNT_WAIT was specified. The result was an unclean filesystem panic when doing forcible unmount with heavy filesystem I/O in progress. Return a zero'ed block when reading a block that was not in use at the time that a snapshot was taken. Normally, these blocks should never be read. However, the readahead code will occationally read them which can cause unexpected behavior. Clean up the debugging code that ensures that no blocks be written on a filesystem while it is suspended. Snapshots must explicitly label the blocks that they are writing during the suspension so that they do not cause a `write on suspended filesystem' panic. Reorganize ffs_copyonwrite() to eliminate a deadlock and also to prevent a race condition that would permit the same block to be copied twice. This change eliminates an unexpected soft updates inconsistency in fsck caused by the double allocation. Use bqrelse rather than brelse for buffers that will be needed soon again by the snapshot code. This improves snapshot performance.
* Revert the reversion of the default mtree behaviour. The mtreemarcel2000-07-242-12/+11
| | | | | | command now follows symlinks again. Requested by: jkh
* Using an atomic operation here won't help if nobody else uses them (forgreen2000-07-231-2/+3
| | | | this). Use the simple_lock() on v_interlock like elsewhere.
* Argh... mtree has been removed from the bootstrap-tools. In thatmarcel2000-07-231-2/+2
| | | | | | case we need to save it in installworld. That latter I forgot... Pointy hat: me
* Name all kernels 'kernel'. This fixes the incompatible behaviourmarcel2000-07-231-5/+3
| | | | | | | | | of the buildkernel and installkernel targets where the kernel was called after the config name. While here, fix the brokenness of the installkernel target. It used to use ${IMAKEENV}, but since that has a very restricted PATH, it couldn't find make(1). Use ${CROSSENV} instead.
* Add bounds checking to stackgap_alloc. Previously it was possiblemarcel2000-07-236-22/+28
| | | | | | | | | | | to construct a path that was long enough (ie longer than SPARE_USRSPACE bytes) and trash the stack. Note that SPARE_USRSPACE is much smaller than MAXPATHLEN so that the Linuxulator will now return ENAMETOOLONG even if the path is smaller than MAXPATHLEN. PR: 12749
* Remove mtree for bootstrap-tools. The -L switch change has beenmarcel2000-07-231-1/+1
| | | | backed out.
* Backout addition of -L switch to mtree. Using -L breaks themarcel2000-07-237-16/+16
| | | | | | build process in too many cases. Adding mtree to bootstrap-tools to solve this breaks the upgrade path because mtree needs a libc that has strtofflags and fflagstostr.
* MFPAO: Add support for AX88190, equipped in MELCO LPC3-TX.tanimura2000-07-234-76/+273
|
* Close file descriptors after use so as not to abuse the descriptorsheldonh2000-07-231-0/+2
| | | | | | table when a long argument list is given. :-) Reported by: Sven Agnew <afterhours80@hotmail.com>
* Add weak_mountd_authentication, which is examined in /etc/rc.network.asmodai2000-07-231-0/+1
| | | | | Setting this to YES instead of its default NO, causes mountd to be passed the -n flag, which allow non-root users mount requests to be served.
* Remove references to the (removed) kern.timecounter.method sysctl.markm2000-07-231-15/+1
|
* Clean this up with some BDE-inspired fixes.markm2000-07-237-73/+92
| | | | | | | | | | | o Make the comments KNF-compliant. o Use nanotime instead of getnanotime; the manpage lies about the kern.timecounter.method - it has been removed. o Fix the ENTROPYSOURCE const permanently. o Make variable names more consistent. o Make function prototypes more consistent. Some more needs to be done; to follow.
* Remove the awi(4) manual page from the sys/ hierarchy after a repo-sheldonh2000-07-231-154/+0
| | | | | copy to share/man/man4 . The file is still not connected to the build from that directory either.
* Clarification (forced commit):green2000-07-230-0/+0
| | | | | The immutability flag referred to in the previous revision is actually VTEXT, not VEXEC.
* Solve the problem where it is possible to get the kernel stuck ingreen2000-07-231-3/+31
| | | | | | | | | | | | | | | | | | a loop down in pmap_init_pt(). A subtraction causes the number of pages to become negative, that was assigned to an unsigned variable, and there is a lot of iteration. The bug is due to the ELF image activator not properly checking for its files being the correct size as specified by the ELF header. The solution is to check that the header doesn't ask for part of a file when that part of the file doesn't exist. Make sure to set VEXEC at the proper times to make the executables immutable (remove race conditions). Also, the ELF format specifiies header entries that allow embedding of other executables (hence how ld-elf.so.1 gets loaded, but not the same as loading shared libraries), so those executables need to be set VEXEC, too, so they're immutable. Reviewed by: peter
* First attempt to fix a buildworld from -stable machine. Explicitlyimp2000-07-231-2/+3
| | | | | include strtofflags.c from lib/libc/gen when building mtree until the current dependency mess with mtree is fixed.
* apm -d also needs a writable file descriptor.nsayer2000-07-231-1/+1
|
* Oops, the moused disable option is now 6 instead of 4. Catch the menu textjhb2000-07-233-3/+3
| | | | | | up to this. Submitted by: Valeriy E. Ushakov <uwe@ptc.spbu.ru>
* Don't save cp(1) in installworld. Its use has been replaced bymarcel2000-07-231-1/+1
| | | | a use of install.
* Expand the -r option to support hex. Though undocumented, the -rjwd2000-07-222-2/+3
| | | | | | | | option already supported octal. Add a comment to the -r option in the man page so it's a bit more specific. Discrepancy brought to my attention by: sasdrq@unx.sas.com Approved by: jkh
* Add transmit buffer limitation depending on actual bandwidth.phk2000-07-221-113/+160
| | | | | Add unframed E1 mode. Various cleanup.
* Remove the reference to KERN_UPDATEINTERVAL and syncer(4), sincesheldonh2000-07-221-6/+0
| | | | | | | KERN_UPDATEINTERVAL can't be used to control sched_sync(). In fact, there's no easy way to control the syncer with sysctls. Reported by: bde
* Add overflow testsse2000-07-221-3/+76
|
* Rework the FILES section to cover the list of common symbolic linkssheldonh2000-07-221-12/+53
| | | | | in the /dev directory. Also cover this driver's quirky relationship with the MAKEDEV script.
* Add the missing flags to ls(1)'s usage: -Bb.green2000-07-221-2/+2
|
* This is yet another attempt to fix "psmintr out of sync.." problems.yokota2000-07-222-26/+18
| | | | | | - Properly protect the watchdog timer routine. - Relocate the sync check statement so that we don't see too many error messages.
* When a connection is being dropped due to a listen queue overflow,jayanth2000-07-217-3/+65
| | | | | | | | | delete the cloned route that is associated with the connection. This does not exhaust the routing table memory when the system is under a SYN flood attack. The route entry is not deleted if there is any prior information cached in it. Reviewed by: Peter Wemm,asmodai
* Remove an unnecessary .PATH entry.jhb2000-07-211-1/+0
|
* If ${BINDIR} is not present, create it before installing to it.jkh2000-07-213-3/+15
|
* Add a terminal entry for xterm; a lot of people are using xtermsjkh2000-07-2112-9/+30
| | | | in the serial-installation of FreeBSD.
* Revert to 1.8. I misread the sentence and its context.chris2000-07-211-2/+1
|
* Properly document %.0f behavior.chris2000-07-211-1/+2
|
* Blah, need to add /usr/bin to the path also.obrien2000-07-211-1/+1
| | | | | | Of course this is a bug in that the dhclient script will not work properly if one has a local / and an NFS mounted /usr and needs to obtain its IP address via DHCP before being able to mount /usr.
* Add in the second half of the Slovenian FTP mirror entry.jhb2000-07-213-0/+3
|
* Fix a really stupid bug where I assumed sizeof(int) == 2. This resulted injhb2000-07-212-2/+2
| | | | | | | MBR's with a 4th slice failing the signature check and fdisk saying that they are invalid. Submitted by: bde
* Syncache2000-07-211-7/+7
|
* Somehow the module makefile got missed in the big commit.nsayer2000-07-211-0/+14
|
* Move the ef(4) manual page from sys/modules/if_ef to share/man/man4, assheldonh2000-07-213-104/+1
| | | | discussed with bp.
* Back out 1.15. It caused more harm than good. Something, however,nsayer2000-07-211-4/+4
| | | | | | must be done to fix this situation. Submitted by: bde
* cons: shorten "rs" again, syscons fixedache2000-07-211-1/+1
|
* Describe Resetache2000-07-211-0/+2
|
* Make Reset (ESC c) reset attributes too, not only clear screenache2000-07-211-1/+6
|
* Don't print summary in quiet mode.des2000-07-211-0/+3
|
* Add a STANDARDS section to discourage the use of this utilitysheldonh2000-07-211-0/+4
| | | | in environments where portability is a concern.
* Don't display a running count in quiet mode.des2000-07-211-2/+7
| | | | | Don't add the offset to the size; it's libfetch's job to report the correct size of the requested file.
OpenPOWER on IntegriCloud