dragonfly.git
4 years agoMakefile.inc1: don't build aicasm
Simon Schubert [Sun, 13 Sep 2009 22:55:42 +0000 (00:55 +0200)]
Makefile.inc1: don't build aicasm

aicasm is built by the kernel Makefile, so there is no need to do this
from Makefile.inc1.

4 years agomake upgrade: Remove a file that was existing with a dir name for a while.
Sascha Wildner [Sun, 13 Sep 2009 22:28:12 +0000 (00:28 +0200)]
make upgrade: Remove a file that was existing with a dir name for a while.

mtree couldn't create the directory because of that.

This can't be fixed with Makefile_upgrade.inc so we need a bit of
additional code for such cases.

4 years agoSync LiveDVD kernel configs with GENERIC.
Sascha Wildner [Sun, 13 Sep 2009 22:01:44 +0000 (00:01 +0200)]
Sync LiveDVD kernel configs with GENERIC.

4 years agomtree: No longer create /usr/libdata/msdosfs.
Sascha Wildner [Sun, 13 Sep 2009 22:01:10 +0000 (00:01 +0200)]
mtree: No longer create /usr/libdata/msdosfs.

4 years agomake upgrade: Take back destroy_dev(9), it still exists.
Sascha Wildner [Sun, 13 Sep 2009 22:00:27 +0000 (00:00 +0200)]
make upgrade: Take back destroy_dev(9), it still exists.

4 years agoSwitch pkg_radd and pkg_search to appropriate CPU type path, i386 or amd64.
Justin C. Sherrill [Sun, 13 Sep 2009 19:55:37 +0000 (12:55 -0700)]
Switch pkg_radd and pkg_search to appropriate CPU type path, i386 or amd64.

4 years agoDEVFS - Wait an extra 2 seconds before looking for the root mount
Matthew Dillon [Sun, 13 Sep 2009 18:00:17 +0000 (11:00 -0700)]
DEVFS - Wait an extra 2 seconds before looking for the root mount

* Some drivers pickup devices asynchronously in a manner which the
  intitial CAM probe does not cover.  Try to give these drivers a little
  extra time to pickup devices before mounting root.  This is a hack.

4 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Sun, 13 Sep 2009 17:58:56 +0000 (10:58 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

4 years agoDEVFS - Fix unprintables in serial numbers reported by the device
Matthew Dillon [Sun, 13 Sep 2009 17:57:37 +0000 (10:57 -0700)]
DEVFS - Fix unprintables in serial numbers reported by the device

* Some device serial numbers contain weird punctuation and unprintable
  characters.  Replace unprintables and other weird characters with
  underscores.

Testing-by: Sascha Wildner <swildner@crater.dragonflybsd.org>
4 years ago/etc/devtab: Fix wording in comments.
Sascha Wildner [Sun, 13 Sep 2009 17:24:11 +0000 (19:24 +0200)]
/etc/devtab: Fix wording in comments.

4 years agoAdjust/cleanup fstab.5 a bit.
Sascha Wildner [Sun, 13 Sep 2009 15:47:00 +0000 (17:47 +0200)]
Adjust/cleanup fstab.5 a bit.

4 years agoAdd a tentative devtab(5) manual page.
Sascha Wildner [Sun, 13 Sep 2009 15:45:39 +0000 (17:45 +0200)]
Add a tentative devtab(5) manual page.

4 years agoGENERIC - add device kbdmux
Alex Hornung [Sun, 13 Sep 2009 07:52:57 +0000 (08:52 +0100)]
GENERIC - add device kbdmux

* kbdmux should be part of the GENERIC kernel, as it solves a lot of
  multi-keyboard issues, if not all.

4 years agokbdmux - Add missing opt_kbd.h include
Alex Hornung [Sun, 13 Sep 2009 07:50:50 +0000 (08:50 +0100)]
kbdmux - Add missing opt_kbd.h include

* Add missing opt_kbd.h include so that whenever KBD_INSTALL_CDEV is set
  in the kernel config, kbdmux also gets it kbdX installed.

Reported-by: Sascha Wildner
4 years agonrelease - generate git repos for pkgsrc and (if a gui build) src
Matthew Dillon [Sun, 13 Sep 2009 03:50:27 +0000 (20:50 -0700)]
nrelease - generate git repos for pkgsrc and (if a gui build) src

* Ship a git repo for /usr/pkgsrc and (for the gui build) /usr/src.
  Installations from the ISO can then do an incremental update using
  the git repo as a base right off the bat, saving users a lot of
  time and effort.

4 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Sun, 13 Sep 2009 02:25:06 +0000 (19:25 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

4 years agonrelease - Redo /usr/Makefile to use only git repos.
Matthew Dillon [Sun, 13 Sep 2009 02:11:26 +0000 (19:11 -0700)]
nrelease - Redo /usr/Makefile to use only git repos.

* Make and make help are now context sensitive based on what files
  already exist in /usr/pkgsrc and /usr/src (if any).

* Use git for the pkgsrc repo.

* Use a remote setup instead of git-clone to make it easier to add
  additional remotes.

4 years agolibiconv: change mount_ntfs to use libkiconv.
Alexander Polakov [Sat, 12 Sep 2009 23:32:47 +0000 (03:32 +0400)]
libiconv: change mount_ntfs to use libkiconv.

Drop -W, add -C.

4 years agolibiconv: convert NTFS to libiconv.
Alexander Polakov [Sat, 12 Sep 2009 23:22:09 +0000 (03:22 +0400)]
libiconv: convert NTFS to libiconv.

Partly-taken-from: FreeBSD

4 years agomount_msdos: Remove some /usr/libdata/msdosfs remains.
Sascha Wildner [Sat, 12 Sep 2009 18:43:30 +0000 (20:43 +0200)]
mount_msdos: Remove some /usr/libdata/msdosfs remains.

4 years agomount_cd9660.8: Reduce differences to FreeBSD a bit.
Sascha Wildner [Sat, 12 Sep 2009 18:26:58 +0000 (20:26 +0200)]
mount_cd9660.8: Reduce differences to FreeBSD a bit.

4 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Sat, 12 Sep 2009 17:44:21 +0000 (10:44 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

4 years agomount_cd9660: change to FreeBSD-like behaviour.
Alexander Polakov [Sat, 12 Sep 2009 17:09:17 +0000 (21:09 +0400)]
mount_cd9660: change to FreeBSD-like behaviour.

 * fix a segfault
 * change -L to -C
 * add -C to manpage

Reported-by: Simon Schubert
Dragonfly-bug: http://bugs.dragonflybsd.org/issue1501

4 years agobackgammon(6)/hack(6)/larn(6): replace sgtty with termios
Sascha Wildner [Sat, 12 Sep 2009 15:37:36 +0000 (17:37 +0200)]
backgammon(6)/hack(6)/larn(6): replace sgtty with termios

Courtesy of FreeBSD ports.

Also does some additional cleanup.

Submitted-by: Ulrich Spoerlein <uqs@spoerlein.net>
Dragonfly-bugs: <http://bugs.dragonflybsd.org/issue1491>
                <http://bugs.dragonflybsd.org/issue1492>
                <http://bugs.dragonflybsd.org/issue1493>

4 years agodevfs.5: chmod 644 :(
Sascha Wildner [Sat, 12 Sep 2009 14:57:28 +0000 (16:57 +0200)]
devfs.5: chmod 644 :(

4 years agoatomic.h: use system namespace for function arguments
Simon Schubert [Sat, 12 Sep 2009 14:41:25 +0000 (16:41 +0200)]
atomic.h: use system namespace for function arguments

We shouldn't use non-system arguments for function arguments in headers,
since it might collide with other symbols.  In this case it collided
with exp(3) from math.h, both included by top(1) (with indirections).

Reported-by: swildner@
4 years agogcc44/gmp: remove Makefile hack, it is fixed proper in bsd.dep.mk
Simon Schubert [Sat, 12 Sep 2009 09:20:52 +0000 (11:20 +0200)]
gcc44/gmp: remove Makefile hack, it is fixed proper in bsd.dep.mk

4 years agobsd.dep.mk: fix race condition with beforedepend
Simon Schubert [Sat, 12 Sep 2009 08:58:37 +0000 (10:58 +0200)]
bsd.dep.mk: fix race condition with beforedepend

The recent depend changes introduced a possible race between running
mkdep and beforedepend, since the .ORDER was only specified for
beforedepend and ${DEPENDFILE}, but not for the intermediate
${_DEPENDFILES}.

4 years agoAdd a devfs(5) manual page (based on FreeBSD's).
Sascha Wildner [Sat, 12 Sep 2009 14:55:04 +0000 (16:55 +0200)]
Add a devfs(5) manual page (based on FreeBSD's).

4 years agokbdmux.4: chmod 644
Sascha Wildner [Sat, 12 Sep 2009 14:28:25 +0000 (16:28 +0200)]
kbdmux.4: chmod 644

4 years agokbdcontrol.1: Reduce differences with FreeBSD a bit.
Sascha Wildner [Sat, 12 Sep 2009 14:27:27 +0000 (16:27 +0200)]
kbdcontrol.1: Reduce differences with FreeBSD a bit.

4 years agoAdd kbdmux to LINT.
Sascha Wildner [Sat, 12 Sep 2009 14:27:10 +0000 (16:27 +0200)]
Add kbdmux to LINT.

4 years agoAdd a kbdmux(4) manual page (taken from FreeBSD).
Sascha Wildner [Sat, 12 Sep 2009 14:26:57 +0000 (16:26 +0200)]
Add a kbdmux(4) manual page (taken from FreeBSD).

4 years agopfi: Adjust a bit better for devfs.
Sascha Wildner [Sat, 12 Sep 2009 13:41:32 +0000 (15:41 +0200)]
pfi: Adjust a bit better for devfs.

Only check for pfi.conf on devices which exist in /dev. This reduces some
error messages upon LiveCD boot.

4 years agoinstaller: Add dumpdev support.
Sascha Wildner [Sat, 12 Sep 2009 12:31:04 +0000 (14:31 +0200)]
installer: Add dumpdev support.

* Use the configured swap with dumpon for the installation.

* Automatically add a dumpdev entry in the installed system's /etc/rc.conf.

4 years agodevfs - change default GID to wheel
Alex Hornung [Sat, 12 Sep 2009 09:43:54 +0000 (10:43 +0100)]
devfs - change default GID to wheel

* Change the default GID (used for every non-device node) to wheel.

Suggested-by: Sascha Wildner
4 years agoptsname,posix_openpt - move to libc/stdlib
Alex Hornung [Sat, 12 Sep 2009 07:58:21 +0000 (08:58 +0100)]
ptsname,posix_openpt - move to libc/stdlib

* As the unix98 pty functions ptsname, grantpt, unlockpt and
  posix_openpt are part of stdlib, it makes more sense having them in
  stdlib/ rather than in gen/.

4 years agoptsname - Minor fixes
Alex Hornung [Fri, 11 Sep 2009 22:11:49 +0000 (23:11 +0100)]
ptsname - Minor fixes

* Return NULL and not error on __is* checks.

* Add ending newline.

Reported-by: Sascha Wildner
4 years agodiskmbr - Don't use 'tempname', use dsname()
Alex Hornung [Fri, 11 Sep 2009 21:59:02 +0000 (22:59 +0100)]
diskmbr - Don't use 'tempname', use dsname()

* sname was initially assigned "tempname" and then never changed
  using dsname as intended. Now we properly call dsname before
  printing sname.

Reported-By: Sascha Wildner
4 years agoHAMMER - Update hammer.5 manual page
Matthew Dillon [Fri, 11 Sep 2009 19:18:42 +0000 (12:18 -0700)]
HAMMER - Update hammer.5 manual page

* Add a description of how the bioq sysctls can be adjusted to give
  reads an even higher priority over writes.

4 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Fri, 11 Sep 2009 17:20:47 +0000 (10:20 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

4 years agoHAMMER - HAMMERIOC_GETHISTORY ioctl fails on directories in PFSs
Matthew Dillon [Fri, 11 Sep 2009 17:19:39 +0000 (10:19 -0700)]
HAMMER - HAMMERIOC_GETHISTORY ioctl fails on directories in PFSs

* The HAMMERIOC_GETHISTORY ioctl on a directory inside a PFS other then
  the root PFS would fail due to an incorrect localization assignment.
  Fix the assignment.

Reported-by: YONETANI Tomokazu <qhwt+dfly@les.ath.cx>
4 years agogetcwd: Return proper error codes.
Stathis Kamperis [Fri, 11 Sep 2009 14:56:57 +0000 (17:56 +0300)]
getcwd: Return proper error codes.

Dragonfly-bug: http://bugs.dragonflybsd.org/issue1394
Reported-by: hasso@
4 years agodevfs - populate statfs f_files properly
Alex Hornung [Fri, 11 Sep 2009 08:59:30 +0000 (09:59 +0100)]
devfs - populate statfs f_files properly

* Now we set statfs f_files to the total number of currently allocated
  devfs_nodes.

Reported-by: Sascha Wildner
4 years agokbdmux - Bind into module build Makefile
Alex Hornung [Fri, 11 Sep 2009 08:36:59 +0000 (09:36 +0100)]
kbdmux - Bind into module build Makefile

4 years agoFix MLINKS.
Sascha Wildner [Fri, 11 Sep 2009 08:20:24 +0000 (10:20 +0200)]
Fix MLINKS.

4 years agoconf.h - Fix buildworld
Alex Hornung [Fri, 11 Sep 2009 07:51:03 +0000 (08:51 +0100)]
conf.h - Fix buildworld

* Don't use cdev_t in struct cdev, but rather use struct cdev itself.

Reported-by: Hasso Tepper
4 years agokbdcontrol - Add support for kbdmux
Alex Hornung [Thu, 10 Sep 2009 10:04:46 +0000 (11:04 +0100)]
kbdcontrol - Add support for kbdmux

* Add support to kbdcontrol to attach and detach keyboards from the
  keyboard multiplexer (kbdmux).

Obtained-from: FreeBSD

4 years agokbdmux - make it work for us
Alex Hornung [Wed, 9 Sep 2009 13:54:18 +0000 (14:54 +0100)]
kbdmux - make it work for us

* port kbdmux to DragonFly.

* change syscons to use kbdmux, if available.

* change kbd to use kbdmux, if available.

* cleanup kbd.

* use a simple buffer instead of clists for kbd.

* move kbd_ macros from syscons.h to kbdregs.h

* use kbd_ macros everywhere instead of unreadable stuff.

* sync ukbd with FreeBSD's latest ukbd before their switch to
  usb4bsd.

Partially-Obtained-from: FreeBSD

4 years agokbdmux - Initial import from FreeBSD
Alex Hornung [Wed, 9 Sep 2009 07:23:44 +0000 (08:23 +0100)]
kbdmux - Initial import from FreeBSD

* Import unmodified kbdmux from FreeBSD revision 193752.

Obtained-from: FreeBSD

4 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Fri, 11 Sep 2009 06:54:22 +0000 (23:54 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

4 years agoFix buildworld issues - conditionalize thread.h in mountctl.h
Matthew Dillon [Fri, 11 Sep 2009 06:52:00 +0000 (23:52 -0700)]
Fix buildworld issues - conditionalize thread.h in mountctl.h

* mountctl.h only needs thread.h if being compiled by the kernel.

4 years agounix98 ptys - Add man pages
Alex Hornung [Fri, 11 Sep 2009 06:46:00 +0000 (07:46 +0100)]
unix98 ptys - Add man pages

* Add man pages for ptsname, grantpt, unlockpt and posix_openpt

Reported-By: Sascha Wildner
4 years agomake_dev(9) - Update man page for make_dev_covering
Alex Hornung [Fri, 11 Sep 2009 06:25:12 +0000 (07:25 +0100)]
make_dev(9) - Update man page for make_dev_covering

4 years agodisk - use make_dev_covering instead of make_dev
Alex Hornung [Fri, 11 Sep 2009 06:13:27 +0000 (07:13 +0100)]
disk - use make_dev_covering instead of make_dev

* Switch subr_disk to use the new make_dev_covering so proper major
  numbers will appear.
  This was especially an issue because there could be major/minor
  collisions.

Reported-By: Sascha Wildner
4 years agodevfs - Add new make_dev_covering
Alex Hornung [Fri, 11 Sep 2009 06:10:09 +0000 (07:10 +0100)]
devfs - Add new make_dev_covering

* Add a new make_dev_covering which is used by systems such as the disk
  subsystem, that make_dev for an underlying device.

4 years agokgdb - Add -D_KERNEL_STRUCTURES to access kernel structures properly.
Matthew Dillon [Fri, 11 Sep 2009 05:54:06 +0000 (22:54 -0700)]
kgdb - Add -D_KERNEL_STRUCTURES to access kernel structures properly.

Reported-by: Stathis Kamperis
4 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Fri, 11 Sep 2009 02:16:49 +0000 (19:16 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

4 years agoKERNEL - Implement a poor man's ioscheduler using sys/iosched.h
Matthew Dillon [Fri, 11 Sep 2009 02:12:08 +0000 (19:12 -0700)]
KERNEL - Implement a poor man's ioscheduler using sys/iosched.h

* Move struct iosched_data from the LWP to the TD.

* Expand bd_wait()'s wakeup array from 128 entries to 16384 entries
  to cover a larger portion of the buffer cache's dirty space.

* bwillwrite() now adds to iosched_data->iowbytes and the ticks
  global is used to decay it.

  Total iowbytes is kept in per-cpu structures.  Each cpu does not
  necessarily match the threads assigned to it, but in aggregate
  the sum of iowbytes for all cpus will match the sum of iowbytes
  for all threads in the system.

* bwillwrite() now calculates a relative I/O write load for each thread
  in the system and passes a fractional calculation to bd_wait().

  This causes processes with low I/O loads to have a higher priority
  (as in, NOT stall in bd_wait() anywhere near as long), and processes
  with high I/O loads to have a lower priority (stall longer in bd_wait()),
  when the buffer cache is saturated with dirty data.

4 years agoHAMMER - Try to improve backend stalls due to heavy write activity.
Matthew Dillon [Fri, 11 Sep 2009 02:05:00 +0000 (19:05 -0700)]
HAMMER - Try to improve backend stalls due to heavy write activity.

* Call hammer_inode_waitreclaims() instead of hammer_inode_waithard(),
  and disable hammer_inode_waithard().

  hammer_inode_waithard() waits for an entire flush cycle but can get
  called by fairly innocuous VNOPS, causing a large number of processes
  to stall for no good reason.

* Change hammer_inode_waitreclaims() to wait for only one reclaim to occur
  before returning, instead of 2.

4 years agoIPV6 - Attempt to fix tcp46 compatibility listen sockets
Matthew Dillon [Thu, 10 Sep 2009 04:52:01 +0000 (21:52 -0700)]
IPV6 - Attempt to fix tcp46 compatibility listen sockets

* When creating the accepted socket via the syncache, if the
  listen socket is INET6 but allows compatibility connections
  the inp created via sonewconn() will probably set IN6P_IPV6_V6ONLY.

  This flag must be cleared if the incoming connection is
  actually IPV4, otherwise tcp6_soport() will not properly
  fall through to tcp_soport().

* NOTE: Even though the connection is IPV4 the socket's protocol
  is copied from the listen socket which is INET6.  This seems
  to work ok but it isn't ideal.

* When connecting make sure that IPV6 connections use cpu 0 for
  now (we don't have a proper hash for IPV6 addresses yet).

Reported-by: corecode
4 years agogcc44/gmp: fix last commit
Simon Schubert [Wed, 9 Sep 2009 19:25:55 +0000 (21:25 +0200)]
gcc44/gmp: fix last commit

4 years agoKernel - Enable the TCP inflight limiter by default, but with generous values.
Matthew Dillon [Wed, 9 Sep 2009 16:34:13 +0000 (09:34 -0700)]
Kernel - Enable the TCP inflight limiter by default, but with generous values.

The inflight limiter is now turned on by default, but with generous values
which should allow maximal bandwidth.  In particular, the slop defaults to
50 (5 packets), meaning that the TCP connection will allow up to N+5 packets
in-flight instead of just N, where N is the number of packets calculated
for the bandwidth-delay product of the connection.

The reason for doing this is that the limiter is the only mechanism we
have which seems to do a really good job preventing receiver RX rings
on network interfaces from getting blown out.  Even though GigE/10GigE
is supposed to flow control it looks like either it doesn't actually
do it or Open Source drivers do not properly enable it.  Large packet
bursts appear to be able to blow out the receiver's rx ring.

People using the limiter to reduce bottlenecks on slower WAN connections
should set the slop to 20 (2 packets).

4 years agoMBUF - Do not limit an objcache common pool to 10 magazines.
Matthew Dillon [Wed, 9 Sep 2009 16:17:57 +0000 (09:17 -0700)]
MBUF - Do not limit an objcache common pool to 10 magazines.

* An objcache's common magazine pool was limited to 10 in order to return
  memory to the system sooner rather then later.

  This created a situation with the mbuf allocator where high-rate
  mbuf producers (GigE, 10GigE) would sometimes not be able to allocate
  a large enough batch of mbufs and drop packets.  10 magazines x 64 is
  only 640 mbufs.

* Increase the limit from 10 to 1000.

Tested-with: 10GigE setup from Aggelos Economopoulos <aoiko@cc.ece.ntua.gr>

4 years agoMBUF - Account for failed allocations so netstat -m reports properly.
Matthew Dillon [Wed, 9 Sep 2009 16:17:18 +0000 (09:17 -0700)]
MBUF - Account for failed allocations so netstat -m reports properly.

4 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Wed, 9 Sep 2009 16:13:22 +0000 (09:13 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

4 years agoNFE - Change default RX ring size from 128 -> 256, Adjust moderation timer.
Matthew Dillon [Wed, 9 Sep 2009 16:10:58 +0000 (09:10 -0700)]
NFE - Change default RX ring size from 128 -> 256, Adjust moderation timer.

* NFE does not seem to be able to flow control the phy.  TCP reception
  using window scaling and large socket bufs can blow out the receiver ring
  before the interrupt has a chance to drain it.

  This results in the receiver dropping a packet and the transmitter
  seeing bursts of duplicate TCP acks and having to go into recovery.

* Shorten the moderation timer from 500uS to 250uS, increasing its
  frequency from 2000Hz to 4000Hz.  Note that the moderation timer
  only kicks in when the discrete interrupt rate exceeds the frequency.

* The problem still occurs with the larger RX ring and shorter moderation
  timer, but not nearly as often.

4 years agogcc44/gmp: make directories before populating them
Simon Schubert [Wed, 9 Sep 2009 15:23:39 +0000 (17:23 +0200)]
gcc44/gmp: make directories before populating them

The snapshot build seems to have uncovered a race in the depend stage.
This is a tentative, unverified fix.

4 years agoRevert "Move isonum_ declaration to iso.h"
Alexander Polakov [Wed, 9 Sep 2009 09:43:08 +0000 (13:43 +0400)]
Revert "Move isonum_ declaration to iso.h"

This reverts commit f5ff88dca23b1b8f4b9287fb7c59959eb976a808. It
broke sysutils/hal and maybe something else.

Requested-by: Hasso Tepper
4 years agonfs: fix real/effective id mismatch in nfs_access
Nicolas Thery [Wed, 9 Sep 2009 07:39:02 +0000 (09:39 +0200)]
nfs: fix real/effective id mismatch in nfs_access

When access(2) or faccessat(2) without AT_EACCESS are called on
NFS file, use real uid/gid.

When faccessat(2) with AT_EACCESS set is called on NFS file, use
effective uid/gid.

4 years agopowernow - only build for i386, BIOS macros currently broken for 64-bit.
Matthew Dillon [Wed, 9 Sep 2009 04:03:05 +0000 (21:03 -0700)]
powernow - only build for i386, BIOS macros currently broken for 64-bit.

4 years agoUSB - Reintroduce OHCI fixes without the larger polling infrastructure fixes
Matthew Dillon [Tue, 8 Sep 2009 23:52:33 +0000 (16:52 -0700)]
USB - Reintroduce OHCI fixes without the larger polling infrastructure fixes

* Fix the OHCI interrupt livelock issue.

* Don't fix races with turning polling on and off yet as there appear
  to be other issues with OHCI that break whene we do.

4 years agoNFS - Fix panic if the readdir base offset is beyond the directory EOF
Matthew Dillon [Tue, 8 Sep 2009 21:31:20 +0000 (14:31 -0700)]
NFS - Fix panic if the readdir base offset is beyond the directory EOF

* A calculation could cause an signed underflow into unsigned if
  the base offset in a getdirentries call exceeds the directory
  EOF offset.

4 years agoSILI - Merge Alex's ATA pass-through ioctls from AHCI driver into SILI driver.
Matthew Dillon [Tue, 8 Sep 2009 21:00:11 +0000 (14:00 -0700)]
SILI - Merge Alex's ATA pass-through ioctls from AHCI driver into SILI driver.

* Basically an exact copy from the AHCI code.

4 years agoahci(4) - Add support for ATA passthrough
Alex Hornung [Tue, 8 Sep 2009 07:48:30 +0000 (08:48 +0100)]
ahci(4) - Add support for ATA passthrough

* Add support for ATA_PASS_{12,16} cam commands to pass through
  an ATA command directly to the device.
  Incidentally, this will make smartmontools work again when used
  with device type 'sat'.

Dragonfly-bug: http://bugs.dragonflybsd.org/issue1412
Reported-By: Hasso Tepper
4 years agoiscontrol.8: A little cleanup.
Sascha Wildner [Tue, 8 Sep 2009 16:39:51 +0000 (18:39 +0200)]
iscontrol.8: A little cleanup.

4 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Tue, 8 Sep 2009 18:36:25 +0000 (11:36 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

4 years agoPIPE - Fix a panic introduced by the last commit.
Matthew Dillon [Tue, 8 Sep 2009 18:28:14 +0000 (11:28 -0700)]
PIPE - Fix a panic introduced by the last commit.

* Forgot to release a token on the Re-tests.

Reported-by: Tobias Weingartner <weingart@tepid.org>
4 years agopowernow: AMD K8 frequency control.
Alexander Polakov [Tue, 8 Sep 2009 18:15:36 +0000 (22:15 +0400)]
powernow: AMD K8 frequency control.

4 years agoISCSI - Minor module handling
Antonio Huete [Fri, 4 Sep 2009 22:18:21 +0000 (00:18 +0200)]
ISCSI - Minor module handling

  - Rename iscsi module name to iscsi_initiator
  - iscontrol now loads automatically iscsi_initiator module

4 years agoSync zoneinfo database with tzdata2009m from elsie.
Sascha Wildner [Tue, 8 Sep 2009 14:19:30 +0000 (16:19 +0200)]
Sync zoneinfo database with tzdata2009m from elsie.

asia:           8.36 -> 8.40
australasia:    8.12 -> 8.13

* Palestine will go back to winter time on Sep 4, 2009.

* Samoa will start DST on Oct 4, 2009 and end on Mar 28, 2010.

4 years agocd9660_util.c: fix a typo.
Alexander Polakov [Tue, 8 Sep 2009 13:11:10 +0000 (17:11 +0400)]
cd9660_util.c: fix a typo.

4 years agocd9660_util.c: ANSIfy, no functional changes.
Alexander Polakov [Tue, 8 Sep 2009 12:20:35 +0000 (16:20 +0400)]
cd9660_util.c: ANSIfy, no functional changes.

4 years agosound: mmap returns page number, not byte offset
Simon Schubert [Tue, 8 Sep 2009 11:58:28 +0000 (13:58 +0200)]
sound: mmap returns page number, not byte offset

Noticed-by: hasso@
4 years agomodules: remove KMODDEPS, it is not used anymore
Simon Schubert [Tue, 8 Sep 2009 10:29:04 +0000 (12:29 +0200)]
modules: remove KMODDEPS, it is not used anymore

4 years agomodules: add/correct module versions and depends
Simon Schubert [Tue, 8 Sep 2009 10:15:13 +0000 (12:15 +0200)]
modules: add/correct module versions and depends

4 years agosys/module.h: include sys/linker_set.h
Simon Schubert [Tue, 8 Sep 2009 10:13:49 +0000 (12:13 +0200)]
sys/module.h: include sys/linker_set.h

4 years agoig_hal: add module version
Simon Schubert [Tue, 8 Sep 2009 09:44:22 +0000 (11:44 +0200)]
ig_hal: add module version

4 years agoadd tools/chkmoddeps
Simon Schubert [Tue, 8 Sep 2009 09:40:17 +0000 (11:40 +0200)]
add tools/chkmoddeps

chkmoddeps is a tool to check for unsatisfied module dependencies.
invoke with pathnames to be searched for kernel and modules.

4 years agoRevert "usbd - Do not start moused by default when a usb mouse is connected"
Simon Schubert [Tue, 8 Sep 2009 06:59:46 +0000 (08:59 +0200)]
Revert "usbd - Do not start moused by default when a usb mouse is connected"

This reverts commit 409ca8073a5e4b42c5f75f3fa5d9be5cdab83c2e, as
discussed on the mailing lists.

4 years agoRevert "USB - Formalize polling mode and fix ohci interrupt storm"
Matthew Dillon [Tue, 8 Sep 2009 06:59:18 +0000 (23:59 -0700)]
Revert "USB - Formalize polling mode and fix ohci interrupt storm"

This reverts commit 727afe75ef752829158e906c5b86991478fece53.  It broke
too many things, so the OHCI storm will have to be fixed another way.

4 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Tue, 8 Sep 2009 02:23:48 +0000 (19:23 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

4 years agoRemove nearly all MP contention from ip_randomid()
Matthew Dillon [Tue, 8 Sep 2009 02:21:34 +0000 (19:21 -0700)]
Remove nearly all MP contention from ip_randomid()

* Instead of using two global spinlocks split the id space up into
  ncpus_fit independant sections and shuffle within each section.

* Aggregate the random number generation into blocks of 32, on a
  per-cpu basis, to reduce cpu contention there.

Using-test-output-by: Aggelos Economopoulos <aoiko@cc.ece.ntua.gr>
4 years agoAdd the make_autoclone_dev(9) manual page.
Sascha Wildner [Mon, 7 Sep 2009 14:56:21 +0000 (16:56 +0200)]
Add the make_autoclone_dev(9) manual page.

Submitted-by: alexh
4 years agosio(4) - Fix double make_dev if used as console
Alex Hornung [Mon, 7 Sep 2009 11:34:45 +0000 (12:34 +0100)]
sio(4) - Fix double make_dev if used as console

* sio was calling another make_dev for ttyd0 if it was being used as
  console. This is no longer valid with devfs, so we look for the
  existing ttyd0, as it must exist already if the device has been
  attached.

Dragonfly-bug: http://bugs.dragonflybsd.org/issue1485
Reported-by: Hasso Tepper, Jordan Gordeev
4 years agoUSB - Formalize polling mode and fix ohci interrupt storm
Matthew Dillon [Mon, 7 Sep 2009 06:05:09 +0000 (23:05 -0700)]
USB - Formalize polling mode and fix ohci interrupt storm

* Formalize entering and leaving polling mode via usb_set_polling().

* Call bus->methods->do_poll() once when leaving polling mode to
  catch-up any lost interrupts.

* OHCI now acknowledges interrupts from the interrupt handler when
  it finds itself in polling mode to prevent the interrupt from
  storming.  The interrupts are recorded for later use.

4 years agoUSB - Preallocate umass buffers to avoid interrupt deadlock
Matthew Dillon [Mon, 7 Sep 2009 02:37:55 +0000 (19:37 -0700)]
USB - Preallocate umass buffers to avoid interrupt deadlock

* Try to avoid usb allocations made from the interrupt servicer.
  These can deadlock the interrupt handler and any other device
  sharing that irq as well as cause USB panics due to the broken
  critical section.

* This is only a temporary fix.  Another unaddressed issue is
  contigmalloc()'s inability to allocate large physically contiguous
  buffers after the machine has been up a while.

Reported-by: Rumko and multiple others over time
4 years agoUSB - Avoid unnecessary reprobes.
Matthew Dillon [Sun, 6 Sep 2009 21:53:51 +0000 (14:53 -0700)]
USB - Avoid unnecessary reprobes.

* Do not unconditionally setdiskinfo when a removable DA* device is
  opened.  It takes time for the kernel to reprobe the drive and this
  seriously messes up mountroot during kernel boot or attempts to
  access partitions in slices (instant open + access partition).

  If the size hasn't changed, do not call setdiskinfo again.

* This is a bad hack.  A better solution (detect actual device
  replacement) is needed.

4 years agoLINT: Mention KTR_CTXSW.
Sascha Wildner [Sun, 6 Sep 2009 18:19:21 +0000 (20:19 +0200)]
LINT: Mention KTR_CTXSW.

4 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Sat, 5 Sep 2009 23:17:59 +0000 (16:17 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly