]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
15 years agoTweak the oshmctl() compile fix: convert the K&R definition to ANSI.
John Baldwin [Thu, 25 Jun 2009 13:36:57 +0000 (13:36 +0000)]
Tweak the oshmctl() compile fix: convert the K&R definition to ANSI.

15 years ago- Make pprint print through fd 3, so it can be used in customisation
Nick Hibma [Thu, 25 Jun 2009 13:15:20 +0000 (13:15 +0000)]
- Make pprint print through fd 3, so it can be used in customisation
  functions to print something to the screen.
- Prefix each line with the running time (bikeshed).

Submitted by: Rick van der Zwet (Wireless Leiden)

15 years agoFix c194955 - somehow I managed all the new files, tripling their
Edward Tomasz Napierala [Thu, 25 Jun 2009 13:08:02 +0000 (13:08 +0000)]
Fix c194955 - somehow I managed all the new files, tripling their
contents.

15 years agoBump manual page timestamps.
Edward Tomasz Napierala [Thu, 25 Jun 2009 12:53:50 +0000 (12:53 +0000)]
Bump manual page timestamps.

15 years agoAdd NFSv4 ACL support to libc.
Edward Tomasz Napierala [Thu, 25 Jun 2009 12:46:59 +0000 (12:46 +0000)]
Add NFSv4 ACL support to libc.

This adds the following functions to the acl(3) API: acl_add_flag_np,
acl_clear_flags_np, acl_create_entry_np, acl_delete_entry_np,
acl_delete_flag_np, acl_get_extended_np, acl_get_flag_np, acl_get_flagset_np,
acl_set_extended_np, acl_set_flagset_np, acl_to_text_np, acl_is_trivial_np,
acl_strip_np, acl_get_brand_np.  Most of them are similar to what Darwin
does.  There are no backward-incompatible changes.

Approved by:    rwatson@

15 years agoRaise the default size of the EFI partition on ia64 from 100MB to 400MB.
John Baldwin [Thu, 25 Jun 2009 12:34:05 +0000 (12:34 +0000)]
Raise the default size of the EFI partition on ia64 from 100MB to 400MB.
A fresh install of a current 8.0 snapshot uses 156MB with a single kernel
and having the filesystem too small prevented the system from booting.

Reviewed by: marcel
MFC after: 1 week

15 years agoAdd a new global rwlock, in_ifaddr_lock, which will synchronize use of the
Robert Watson [Thu, 25 Jun 2009 11:52:33 +0000 (11:52 +0000)]
Add a new global rwlock, in_ifaddr_lock, which will synchronize use of the
in_ifaddrhead and INADDR_HASH address lists.

Previously, these lists were used unsynchronized as they were effectively
never changed in steady state, but we've seen increasing reports of
writer-writer races on very busy VPN servers as core count has gone up
(and similar configurations where address lists change frequently and
concurrently).

For the time being, use rwlocks rather than rmlocks in order to take
advantage of their better lock debugging support.  As a result, we don't
enable ip_input()'s read-locking of INADDR_HASH until an rmlock conversion
is complete and a performance analysis has been done.  This means that one
class of reader-writer races still exists.

MFC after:      6 weeks
Reviewed by:    bz

15 years agoInclude SMP support in the MPC85XX kernel by default.
Rafal Jaworowski [Thu, 25 Jun 2009 10:07:21 +0000 (10:07 +0000)]
Include SMP support in the MPC85XX kernel by default.

15 years agoEnable all populated TWSI (I2C) controllers on Marvell SOCs.
Rafal Jaworowski [Thu, 25 Jun 2009 10:03:51 +0000 (10:03 +0000)]
Enable all populated TWSI (I2C) controllers on Marvell SOCs.

Obtained from: Semihalf

15 years agoupdate the manpage to track existing options (including --iso),
Luigi Rizzo [Thu, 25 Jun 2009 09:28:04 +0000 (09:28 +0000)]
update the manpage to track existing options (including --iso),
and alphabetize them.

MFC after: 3 days

15 years agoSwitch cmd argument of ioctl to u_long as elsewhere in the kernel.
Roman Divacky [Thu, 25 Jun 2009 08:52:20 +0000 (08:52 +0000)]
Switch cmd argument of ioctl to u_long as elsewhere in the kernel.
Propagate this change down the callchain.

Approved by: kan (maintainer)
Approved by: ed (mentor)

15 years agoClean up reference management in in6_update_ifa and in6_unlink_ifa, and
Robert Watson [Thu, 25 Jun 2009 08:37:38 +0000 (08:37 +0000)]
Clean up reference management in in6_update_ifa and in6_unlink_ifa, and
in particular, add a reference for in6_ifaddrhead since we do remove a
reference for it when an IPv6 address is removed.  This fixes ifconfig
delete of an IPv6 alias.

Reported by: tegge
MFC after: 6 weeks

15 years agoFix ibcs2_ipc.c build by adding missing limits.h include.
Robert Watson [Thu, 25 Jun 2009 07:25:39 +0000 (07:25 +0000)]
Fix ibcs2_ipc.c build by adding missing limits.h include.

Submitted by: keramida

15 years agooshmctl() now requires a sysv_shm.c-local function prototype.
Robert Watson [Thu, 25 Jun 2009 07:16:10 +0000 (07:16 +0000)]
oshmctl() now requires a sysv_shm.c-local function prototype.

15 years agoMake a DVD for ia64.
Marcel Moolenaar [Thu, 25 Jun 2009 04:25:26 +0000 (04:25 +0000)]
Make a DVD for ia64.

15 years agoupdates AMRR statistics with tx complete status that if not the tx rate
Weongyo Jeong [Thu, 25 Jun 2009 02:28:12 +0000 (02:28 +0000)]
updates AMRR statistics with tx complete status that if not the tx rate
always would be reduced.

15 years agouses ZYD_NOTIF_RETRYSTATUS info to count the number of retries.
Weongyo Jeong [Thu, 25 Jun 2009 02:14:47 +0000 (02:14 +0000)]
uses ZYD_NOTIF_RETRYSTATUS info to count the number of retries.

15 years ago - Use DPCPU for SCHED_STATS. This is somewhat awkward because the
Jeff Roberson [Thu, 25 Jun 2009 01:33:51 +0000 (01:33 +0000)]
 - Use DPCPU for SCHED_STATS.  This is somewhat awkward because the
   offset of the stat is not known until link time so we must emit a
   function to call SYSCTL_ADD_PROC rather than using SYSCTL_PROC
   directly.
 - Eliminate the atomic from SCHED_STAT_INC now that it's using per-cpu
   variables.  Sched stats are always incremented while we're holding
   a spinlock so no further protection is required.

Reviewed by: sam

15 years ago - Add a sysctl_dpcpu_long to support long typed pcpu stats.
Jeff Roberson [Thu, 25 Jun 2009 01:31:59 +0000 (01:31 +0000)]
 - Add a sysctl_dpcpu_long to support long typed pcpu stats.
 - Remove the #ifndef SMP case as the SMP code works on UP as well.

Reviewed by: sam

15 years agoFix two known problems in clnt_rc.c, plus issues w.r.t. smp noted
Rick Macklem [Thu, 25 Jun 2009 00:28:43 +0000 (00:28 +0000)]
Fix two known problems in clnt_rc.c, plus issues w.r.t. smp noted
during reading of the code. Change the code so that it never accesses
rc_connecting, rc_closed or rc_client when the rc_lock mutex is not held.
Also, it now performs the CLNT_CLOSE(client) and CLNT_RELEASE(client)
calls after the rc_lock mutex has been released, since those calls do
msleep()s with another mutex held. Change clnt_reconnect_call() so that
releasing the reference count is delayed until after the
"if (rc->rc_client == client)" check, so that rc_client cannot have been
recycled.

Tested by: pho
Reviewed by: dfr
Approved by: kib (mentor)

15 years ago - Add the right includes to use kmem_alloc(). This was broken by my
Jeff Roberson [Thu, 25 Jun 2009 00:14:27 +0000 (00:14 +0000)]
 - Add the right includes to use kmem_alloc().  This was broken by my
   DPCPU commit.
Reported by: bz

15 years agoLock around access to nc_file and netconfig_info ("ni"). The RPC
Xin LI [Wed, 24 Jun 2009 23:17:16 +0000 (23:17 +0000)]
Lock around access to nc_file and netconfig_info ("ni").  The RPC
part of libc is still not thread safe but this would at least
reduce the problems we have.

PR: threads/118544
Submitted by: Changming Sun <snnn119 gmail com>
MFC after: 2 weeks

15 years agoMake sysinstall search for /dev/daXa and register such devices as USB disks.
Colin Percival [Wed, 24 Jun 2009 23:17:00 +0000 (23:17 +0000)]
Make sysinstall search for /dev/daXa and register such devices as USB disks.
This covers the common case of unsliced USB drives, and makes it possible to
select them as installation source media.

PR: 61152, 115197, 135016
Submitted by: randi
MFC after: 1 month

15 years ago- fix dummynet 'fast' mode for WF2Q case.
Oleg Bulyzhin [Wed, 24 Jun 2009 22:57:07 +0000 (22:57 +0000)]
- fix dummynet 'fast' mode for WF2Q case.
- fix printing of pipe profile data.
- introduce new pipe parameter: 'burst' - how much data can be sent through
  pipe bypassing bandwidth limit.

15 years agoThis adds a new "stdio" mode to cxgbtool - it's an interactive mode
Navdeep Parhar [Wed, 24 Jun 2009 22:28:48 +0000 (22:28 +0000)]
This adds a new "stdio" mode to cxgbtool - it's an interactive mode
meant primarily for _non_ interactive use.  Scripts that run cxgbtool
repeatedly to perform register r/w or mdio will benefit from this.
Instead of fork/exec'ing a new cxgbtool for every regio/mdio you can
simply open a pair of pipes to/from cxgbtool and run cmds over them.

Approved by: gnn (mentor)

15 years agoMerge from p4: CH154790,154793,154874
Bjoern A. Zeeb [Wed, 24 Jun 2009 22:21:30 +0000 (22:21 +0000)]
Merge from p4: CH154790,154793,154874

  Import if_epair(4), a virtual cross-over Ethernet-like interface pair.

Note these files are 1:1 from p4 and not yet connected to the build
not knowing about the new netisr interface.

Sponsored by: The FreeBSD Foundation

15 years agoMake CPU bind call architecture specific to satisfy LINT
Jack F Vogel [Wed, 24 Jun 2009 22:17:40 +0000 (22:17 +0000)]
Make CPU bind call architecture specific to satisfy LINT

15 years agoneed to make intr_bind call architecture specific for
Jack F Vogel [Wed, 24 Jun 2009 22:16:02 +0000 (22:16 +0000)]
need to make intr_bind call architecture specific for
global builds (failing sun4v lint build)

15 years ago- Apply the same naming rules of LVM names as done in the LVM code itself.
Ulf Lilleengen [Wed, 24 Jun 2009 22:09:30 +0000 (22:09 +0000)]
- Apply the same naming rules of LVM names as done in the LVM code itself.

PR: kern/135874

15 years agoWrap a PR_VNET inside "#ifdef VIMAGE" since that the only place it applies.
Jamie Gritton [Wed, 24 Jun 2009 22:06:56 +0000 (22:06 +0000)]
Wrap a PR_VNET inside "#ifdef VIMAGE" since that the only place it applies.
bz wants the blame for this.

Noticed by: rwatson
Approved by: bz (mentor)

15 years agoDesignate special builtins as such in command -V and type.
Jilles Tjoelker [Wed, 24 Jun 2009 22:04:04 +0000 (22:04 +0000)]
Designate special builtins as such in command -V and type.
Also document various properties of special builtins that we implement.

Approved by: ed (mentor) (implicit)

15 years agoVarious ifmedia related fixes in cxgb(4), including:
Navdeep Parhar [Wed, 24 Jun 2009 21:56:05 +0000 (21:56 +0000)]
Various ifmedia related fixes in cxgb(4), including:

- build ifmedia list based on phy->caps, not string comparisons.
- rebuild media list when a transceiver change is detected.
- return EOPNOTSUPP instead of ENXIO in cxgb_media_status.

Approved by: gnn (mentor)
MFC after: 2 weeks.

15 years agoBump __FreeBSD_version for SYSVIPC ABI change (along with other changes
John Baldwin [Wed, 24 Jun 2009 21:55:18 +0000 (21:55 +0000)]
Bump __FreeBSD_version for SYSVIPC ABI change (along with other changes
today).

15 years agoRegen.
John Baldwin [Wed, 24 Jun 2009 21:54:08 +0000 (21:54 +0000)]
Regen.

15 years agoAdd 10Gbase-T to known ethernet media types.
Navdeep Parhar [Wed, 24 Jun 2009 21:53:25 +0000 (21:53 +0000)]
Add 10Gbase-T to known ethernet media types.

Approved by: gnn (mentor)
MFC after: 1 week.

15 years agoAbout to add 10Gbase-T to known media types, this is just a whitespace
Navdeep Parhar [Wed, 24 Jun 2009 21:51:42 +0000 (21:51 +0000)]
About to add 10Gbase-T to known media types, this is just a whitespace
cleanup before that commit.  No functional impact.

Approved by: gnn (mentor)

15 years agoSync with NetBSD:
Xin LI [Wed, 24 Jun 2009 21:51:02 +0000 (21:51 +0000)]
Sync with NetBSD:
 - gzip -n does not store timestamp; [1]
 - Reduce diff against NetBSD by moving some casts in our local
   versions.

PR: bin/134955
Obtained from: NetBSD
MFC after: 1 month

15 years agoIn case of prisons with their own network stack, permit
Jamie Gritton [Wed, 24 Jun 2009 21:39:50 +0000 (21:39 +0000)]
In case of prisons with their own network stack, permit
additional privileges as well as not restricting the type of
sockets a user can open.

Note: the VIMAGE/vnet fetaure of of jails is still considered
      experimental and cannot guarantee that privileged users
      can be kept imprisoned if enabled.

Reviewed by: rwatson
Approved by: bz (mentor)

15 years agoUse queue(9) instead of hand-crafted link lists for the global netatalk
Robert Watson [Wed, 24 Jun 2009 21:36:09 +0000 (21:36 +0000)]
Use queue(9) instead of hand-crafted link lists for the global netatalk
address list.  Generally follow the style and convention of similar parts
in netinet.

MFC after: 6 weeks

15 years agoFix CARP build.
Robert Watson [Wed, 24 Jun 2009 21:34:38 +0000 (21:34 +0000)]
Fix CARP build.

Reported by: bz

15 years agoFix lint issue.
Jack F Vogel [Wed, 24 Jun 2009 21:32:51 +0000 (21:32 +0000)]
Fix lint issue.

15 years agoChange the ABI of some of the structures used by the SYSV IPC API:
John Baldwin [Wed, 24 Jun 2009 21:10:52 +0000 (21:10 +0000)]
Change the ABI of some of the structures used by the SYSV IPC API:
- The uid/cuid members of struct ipc_perm are now uid_t instead of unsigned
  short.
- The gid/cgid members of struct ipc_perm are now gid_t instead of unsigned
  short.
- The mode member of struct ipc_perm is now mode_t instead of unsigned short
  (this is merely a style bug).
- The rather dubious padding fields for ABI compat with SV/I386 have been
  removed from struct msqid_ds and struct semid_ds.
- The shm_segsz member of struct shmid_ds is now a size_t instead of an
  int.  This removes the need for the shm_bsegsz member in struct
  shmid_kernel and should allow for complete support of SYSV SHM regions
  >= 2GB.
- The shm_nattch member of struct shmid_ds is now an int instead of a
  short.
- The shm_internal member of struct shmid_ds is now gone.  The internal
  VM object pointer for SHM regions has been moved into struct
  shmid_kernel.
- The existing __semctl(), msgctl(), and shmctl() system call entries are
  now marked COMPAT7 and new versions of those system calls which support
  the new ABI are now present.
- The new system calls are assigned to the FBSD-1.1 version in libc.  The
  FBSD-1.0 symbols in libc now refer to the old COMPAT7 system calls.
- A simplistic framework for tagging system calls with compatibility
  symbol versions has been added to libc.  Version tags are added to
  system calls by adding an appropriate __sym_compat() entry to
  src/lib/libc/incldue/compat.h. [1]

PR: kern/16195 kern/113218 bin/129855
Reviewed by: arch@, rwatson
Discussed with: kan, kib [1]

15 years agoAdd a dying flag to prevent races at detach.
Andrew Gallatin [Wed, 24 Jun 2009 21:09:56 +0000 (21:09 +0000)]
Add a dying flag to prevent races at detach.

I tried re-ordering ether_ifdetach(), but this created a new race
where sometimes, when under heavy receive load (>1Mpps) and running
tcpdump, the machine would panic.  At panic, the ithread was still in
the original (not dead) if_input() path, and was accessing stale BPF
data structs.  By using a dying flag, I can close the interface prior
to if_detach() to be certain the interface cannot send packets up in
the middle of ether_ifdetach.

15 years agoFix typo.
Olivier Houchard [Wed, 24 Jun 2009 21:03:59 +0000 (21:03 +0000)]
Fix typo.

15 years agoConvert netinet6 to using queue(9) rather than hand-crafted linked lists
Robert Watson [Wed, 24 Jun 2009 21:00:25 +0000 (21:00 +0000)]
Convert netinet6 to using queue(9) rather than hand-crafted linked lists
for the global IPv6 address list (in6_ifaddr -> in6_ifaddrhead).  Adopt
the code styles and conventions present in netinet where possible.

Reviewed by: gnn, bz
MFC after: 6 weeks (possibly not MFCable?)

15 years agoFix typo.
Olivier Houchard [Wed, 24 Jun 2009 21:00:13 +0000 (21:00 +0000)]
Fix typo.

15 years agoUse queue(9) instead of hand-crafted link lists for the global IPX
Robert Watson [Wed, 24 Jun 2009 20:57:50 +0000 (20:57 +0000)]
Use queue(9) instead of hand-crafted link lists for the global IPX
address list (ipx_ifaddr -> ipx_ifaddrhead), and generally adopt the
naming and usage conventions found in netinet.

MFC after: 6 weeks

15 years ago- Change this driver to do taskqueue(9) based TX and interrupt
Marius Strobl [Wed, 24 Jun 2009 20:56:06 +0000 (20:56 +0000)]
- Change this driver to do taskqueue(9) based TX and interrupt
  handling in order to reduce interrupt overhead which results in
  better performance.
- Call ether_ifdetach(9) before stopping the controller and the
  callouts detach in order to prevent active BPF listeners to clear
  promiscuous mode which may lead to the tick callout being restarted
  which will trigger a panic once it's actually gone.
- Add explicit IFF_DRV_RUNNING checking in order to prevent extra
  link up/down events when using dhclient(8).
- Use the correct macro for deciding whether 2/3 of the available TX
  descriptors are used.
- Wrap the RX fault printing in #ifdef CAS_DEBUG in order to not
  unnecessarily frighten users and as debugging was the actual
  intention. Real errors caused by these faults still will be
  accumulated as input errors. It might be a good idea to later on
  add driver specific counters for the faults though.

Submitted by: yongari (original patch)

15 years ago- Remove unused variables. [1]
Marius Strobl [Wed, 24 Jun 2009 20:52:59 +0000 (20:52 +0000)]
- Remove unused variables. [1]
- Remove redundant zeroing of tmf_req which Coverity Prevent(tm) complains
  about. [2]

Submitted by: Christoph Mallon [1]
Found with:     Coverity Prevent(tm) [2]
CID:            2496 [2]
MFC after: 2 weeks

15 years agoMFp4:
Alexander Motin [Wed, 24 Jun 2009 20:52:37 +0000 (20:52 +0000)]
MFp4:
Define several ATA capabilies bits.

15 years agoo merge from amd64:
Marius Strobl [Wed, 24 Jun 2009 20:49:02 +0000 (20:49 +0000)]
o merge from amd64:
  - r187144: Add a reference to the config(5) manpage and
    to the "env" kernel config option.
  - Add/enable the default USB drivers. Originally the USB
    controller and keyboard drivers were disabled as these
    interacted badly with the Open Firmware console driver,
    i.e. caused the keyboard to not work with ofw_console(4).
    Even when switch to uart(4) and the frame buffer drivers
    most of the USB drivers still were kept disabled as
    several of them, amongst others all of the drivers for
    USB Ethernet controllers, weren't endian clean. With the
    new USB stack these problem should be gone now so there's
    no longer a reason to not include the same set of USB
    drivers amd64 does.
o Remove the commented out device ofw_console; apart from it
  being currently broken by some TTY changes one really needs
  to know how to actually enable and make it work correctly.

15 years agoMFp4:
Alexander Motin [Wed, 24 Jun 2009 20:43:51 +0000 (20:43 +0000)]
MFp4:
Remove unused ATAPI definitions, conflicting with ata.h.

Submitted by: scottl

15 years agoAdd test for r190698.
Jilles Tjoelker [Wed, 24 Jun 2009 20:22:54 +0000 (20:22 +0000)]
Add test for r190698.

Submitted by: Eygene Ryabinkin
Approved by: ed (mentor) (implicit)

15 years agoRemove kernel SLIP and PPP privileges, since they are no longer used.
Robert Watson [Wed, 24 Jun 2009 20:06:16 +0000 (20:06 +0000)]
Remove kernel SLIP and PPP privileges, since they are no longer used.

Suggested by: bz

15 years agoDeprecate the msgsys(), semsys(), and shmsys() system calls by moving
John Baldwin [Wed, 24 Jun 2009 20:01:13 +0000 (20:01 +0000)]
Deprecate the msgsys(), semsys(), and shmsys() system calls by moving
them under COMPAT_FREEBSD[4567].  Starting with FreeBSD 5.0 the SYSV IPC
API was implemented via direct system calls (e.g. msgctl(), msgget(), etc.)
rather than indirecting through the var-args *sys() system calls.  The
shmsys() system call was already effectively deprecated for all but
COMPAT_FREEBSD4 already as its implementation for the !COMPAT_FREEBSD4 case
was to simply invoke nosys().

15 years agoMFp4:
Alexander Motin [Wed, 24 Jun 2009 19:49:18 +0000 (19:49 +0000)]
MFp4:
Reduce default PCI ATA drivers priorities from absolute to default,
to allow them been overriden. It was so before modularization.

15 years agoDrop the defunct FDOPT_NOERRLOG option from all the floppy utilities.
Joerg Wunsch [Wed, 24 Jun 2009 19:47:53 +0000 (19:47 +0000)]
Drop the defunct FDOPT_NOERRLOG option from all the floppy utilities.
The kernel does not log floppy media errors anymore.

In fdcontrol, do always open the file descriptor in read-only mode so
it can operate on read-only media, as there is no longer a separate
control device to operate on.

15 years agoWith the fdc control device disappearing some 5 years ago, it is no
Joerg Wunsch [Wed, 24 Jun 2009 19:30:31 +0000 (19:30 +0000)]
With the fdc control device disappearing some 5 years ago, it is no
longer useful for the FD_STYPE and FD_SOPTS ioctls to insist on being
issued on a writable file descriptor.  Otherwise, there's no longer a
chance to set the drive type or options when a read-only medium is
present in the drive, as there is no way to obtain a writable fd then.

15 years agoDocument new hint.atapci.X.msi and hint.ata.X.pm_level tunables.
Alexander Motin [Wed, 24 Jun 2009 19:25:47 +0000 (19:25 +0000)]
Document new hint.atapci.X.msi and hint.ata.X.pm_level tunables.

15 years agoWhitespace fix.
John Baldwin [Wed, 24 Jun 2009 19:16:48 +0000 (19:16 +0000)]
Whitespace fix.

15 years agoRevert the part of r194763 which added a dying flag and instead
Marius Strobl [Wed, 24 Jun 2009 19:04:08 +0000 (19:04 +0000)]
Revert the part of r194763 which added a dying flag and instead
call ether_ifdetach(9) before stopping the controller and the
callouts. The consensus is that the latter is now safe to do and
should also solve the problem of active BPF listeners clearing
promiscuous mode can result in the tick callout being restarted
which in turn will trigger a panic once it's actually gone.

15 years ago- Similar to the previous commit, but for CURRENT: Fix a bug where a FIFO vnode
Ulf Lilleengen [Wed, 24 Jun 2009 18:44:38 +0000 (18:44 +0000)]
- Similar to the previous commit, but for CURRENT: Fix a bug where a FIFO vnode
  use count was increased twice, but only decreased once.

15 years agoDon't use sys/nfs/rpcv2.h - it is part of the old kernel RPC implementation
Doug Rabson [Wed, 24 Jun 2009 18:42:21 +0000 (18:42 +0000)]
Don't use sys/nfs/rpcv2.h - it is part of the old kernel RPC implementation
and will be removed.

15 years ago- Fix a bug where a FIFO vnode use count was increased twice, but only
Ulf Lilleengen [Wed, 24 Jun 2009 18:38:51 +0000 (18:38 +0000)]
- Fix a bug where a FIFO vnode use count was increased twice, but only
  decreased once.

MFC after: 1 week

15 years agoIf the initial attempt to refresh credentials in the RPCSEC_GSS client
Rick Macklem [Wed, 24 Jun 2009 18:30:14 +0000 (18:30 +0000)]
If the initial attempt to refresh credentials in the RPCSEC_GSS client
side fails, the entry in the cache is left with no valid context
(gd_ctx == GSS_C_NO_CONTEXT). As such, subsequent hits on the cache
will result in persistent authentication failure, even after the user has
done a kinit or similar and acquired a new valid TGT. This patch adds a test
for that case upon a cache hit and calls rpc_gss_init() to make another
attempt at getting valid credentials. It also moves the setting of gc_proc
to before the import of the principal name to ensure that, if that case
fails, it will be detected as a failure after going to "out:".

Reviewed by: dfr
Approved by: kib (mentor)

15 years agoUpdate for the Intel 10G driver, this adds support for
Jack F Vogel [Wed, 24 Jun 2009 18:27:07 +0000 (18:27 +0000)]
Update for the Intel 10G driver, this adds support for
newest hardware, adds multiqueue tx interface, infrastructure
cleanup to allow up to 32 MSIX vectors on newer Nehalem systems.
Bug fixes, etc.

15 years agoadd a link named athpeek since my fingers keep typing it
Sam Leffler [Wed, 24 Jun 2009 18:24:37 +0000 (18:24 +0000)]
add a link named athpeek since my fingers keep typing it

15 years agoread back the written value and display
Sam Leffler [Wed, 24 Jun 2009 18:24:20 +0000 (18:24 +0000)]
read back the written value and display

15 years agoAdd the "vnet" and "-vnet" options, to allow moving interfaces between
Jamie Gritton [Wed, 24 Jun 2009 18:21:37 +0000 (18:21 +0000)]
Add the "vnet" and "-vnet" options, to allow moving interfaces between
jails with VIMAGE.

Approved by: bz (mentor)

15 years agoAdd libjail, a (somewhat) simpler interface to the jail_set and jail_get
Jamie Gritton [Wed, 24 Jun 2009 18:19:55 +0000 (18:19 +0000)]
Add libjail, a (somewhat) simpler interface to the jail_set and jail_get
system calls and the security.jail.param sysctls.

Approved by: bz (mentor)

15 years agoAdd libjail, a (somewhat) simpler interface to the jail_set and jail_get
Jamie Gritton [Wed, 24 Jun 2009 18:18:35 +0000 (18:18 +0000)]
Add libjail, a (somewhat) simpler interface to the jail_set and jail_get
system calls and the security.jail.param sysctls.

Approved by: bz (mentor)

15 years agoUpdates for both the em and igb drivers, add support
Jack F Vogel [Wed, 24 Jun 2009 17:41:29 +0000 (17:41 +0000)]
Updates for both the em and igb drivers, add support
for multiqueue tx, shared code updates, new device
support, and some bug fixes.

15 years agoRemove usb. The need to have core@ approve major changes to usb has
Warner Losh [Wed, 24 Jun 2009 17:23:10 +0000 (17:23 +0000)]
Remove usb.  The need to have core@ approve major changes to usb has
passed now that the new usb stack is in the tree.  The coordination
issues that necessitated this entry are now OBE.

Approved by: core (rwatson)

15 years agoSome DMA related changes:
Alexander Motin [Wed, 24 Jun 2009 17:03:06 +0000 (17:03 +0000)]
Some DMA related changes:
 - honor parent DMA tag limitations, as man page requires,
 - allow data buffer to be allocated within full 64bit address range, when
   support is announced by hardware,
 - add quirk, disabling 64bit addresses for broken chips, use it for MCP78.

15 years agoMove programming info from usb(4) to usbdi(9) and update for the usb stack
Andrew Thompson [Wed, 24 Jun 2009 17:01:17 +0000 (17:01 +0000)]
Move programming info from usb(4) to usbdi(9) and update for the usb stack
changeover. Needs much more content still.

15 years agoFix end-of-line issues that can come up when `lpq' reads information
Garance A Drosehn [Wed, 24 Jun 2009 16:57:33 +0000 (16:57 +0000)]
Fix end-of-line issues that can come up when `lpq' reads information
about a queue from a remote host.  That remote host may use \r, \r\n,
or \n\r as the line-ending character.  In some cases the remote host
will write a single line of information without *any* EOL sequence.

Translate all the non-unix EOL's to the standard newline, and make
sure the final line includes a terminating newline.  Logic is also
added to translate all unprintable characters to '?', but that is
#if-ed out for now.

PR: bin/104731
MFC after: 3 weeks

15 years agoUnbreak sparc64 after the swap accounting changes: mark kernel_map
Konstantin Belousov [Wed, 24 Jun 2009 16:52:30 +0000 (16:52 +0000)]
Unbreak sparc64 after the swap accounting changes: mark kernel_map
entries allocated for translations in pmap_init() as MAP_NOFAULT. This
prevents vm_map_insert from trying to account the entries for swap
usage, that is both wrong and too early to work.

While there, change FALSE to VMFS_NO_SPACE.

Reported and tested by: Florian Smeets <flo at kasimir com>
Reviewed by: marius

15 years agoRework locking and reference counting in ipx_control to be consistent with
Robert Watson [Wed, 24 Jun 2009 16:52:23 +0000 (16:52 +0000)]
Rework locking and reference counting in ipx_control to be consistent with
the model used in in_control().

MFC after: 6 weeks

15 years agofw_state ad cur_state are holding unsigned bitfields, so declare then
Scott Long [Wed, 24 Jun 2009 16:11:29 +0000 (16:11 +0000)]
fw_state ad cur_state are holding unsigned bitfields, so declare then
as unsigned as well.

Submitted by: rdivacky

15 years agodtrace/amd64: fix virtual address checks
Andriy Gapon [Wed, 24 Jun 2009 16:03:57 +0000 (16:03 +0000)]
dtrace/amd64: fix virtual address checks

On amd64 KERNBASE/kernbase does not mean start of kernel memory.
This should fix a KASSERT panic in dtrace_copycheck when copyin*()
is used in D program.
Also make checks for user memory a bit stricter.

Reported by: Thomas Backman <serenity@exscape.org>
Submitted by: wxs (kaddr part)
Tested by: Thomas Backman (prototype), wxs
Reviewed by: alc (concept), jhb, current@
Aprroved by: jb (concept)
MFC after: 2 weeks
PR: kern/134408

15 years agoMore precise description of the DS1553 driver.
Rafal Jaworowski [Wed, 24 Jun 2009 15:48:20 +0000 (15:48 +0000)]
More precise description of the DS1553 driver.

Pointed out by: stas

15 years agoIntroduce ata(4) support for Marvell integrated SATA controllers (found on
Rafal Jaworowski [Wed, 24 Jun 2009 15:41:18 +0000 (15:41 +0000)]
Introduce ata(4) support for Marvell integrated SATA controllers (found on
88F5xxx, 88F6xxx and MV78xxx system on chip devices).

Reviewed by: stas
Obtained from: Semihalf

15 years agoMove non-PCI prototypes from ata-pci.h -> ata-all.h.
Rafal Jaworowski [Wed, 24 Jun 2009 15:38:17 +0000 (15:38 +0000)]
Move non-PCI prototypes from ata-pci.h -> ata-all.h.

This removes unnecessary PCI #includes dependency for systems with ATA
controllers living at non-PCI buses.

Submitted by: Piotr Ziecik
Obtained from: Semihalf

15 years agoClean up struct prison, with the recent fields in more logical places,
Jamie Gritton [Wed, 24 Jun 2009 15:32:57 +0000 (15:32 +0000)]
Clean up struct prison, with the recent fields in more logical places,
and room for future expansion.

Approved by: bz (mentor)

15 years agoFix a race in vi_if_move, where a vnet is used after the prison that
Jamie Gritton [Wed, 24 Jun 2009 15:29:36 +0000 (15:29 +0000)]
Fix a race in vi_if_move, where a vnet is used after the prison that
referred to it has been released.

Approved by: bz (mentor)

15 years agoAdd missing unlock of if_addr_mtx when an unmatched ARP packet is received.
Robert Watson [Wed, 24 Jun 2009 14:49:26 +0000 (14:49 +0000)]
Add missing unlock of if_addr_mtx when an unmatched ARP packet is received.

Reported by: lstewart
MFC after: 6 weeks

15 years agoAllow admin to specify the initial mtu upon driver load
Andrew Gallatin [Wed, 24 Jun 2009 14:47:32 +0000 (14:47 +0000)]
Allow admin to specify the initial mtu upon driver load
for mxge.

15 years agoClear 'ia' after iterating if_addrhead for unicast address matching: since
Robert Watson [Wed, 24 Jun 2009 14:29:40 +0000 (14:29 +0000)]
Clear 'ia' after iterating if_addrhead for unicast address matching: since
'ifa' was used as the TAILQ_FOREACH() iterator argument, and 'ia' was just
derived form it, it could be left non-NULL which confused later
conditional freeing code.  This could cause kernel panics if multicast IP
packets were received.  [1]

Call 'struct in_ifaddr *' in ip_rtaddr() 'ia', not 'ifa' in keeping with
normal conventions.

When 'ipstealth' is enabled returns from ip_input early, properly release
the 'ia' reference.

Reported by: lstewart, sam [1]
MFC after: 6 weeks

15 years agoAdd a new COMPAT7 flag for FreeBSD 7.x compatibility system calls.
John Baldwin [Wed, 24 Jun 2009 13:36:37 +0000 (13:36 +0000)]
Add a new COMPAT7 flag for FreeBSD 7.x compatibility system calls.

15 years ago- Move syscall function argument structure types to be just above the
John Baldwin [Wed, 24 Jun 2009 13:35:38 +0000 (13:35 +0000)]
- Move syscall function argument structure types to be just above the
  relevenat system call function.
- Whitespace fixes.

15 years agoFix copy-and-paste-o's from kinfo_getfile.3 in kinfo_getvmmap.3.
Robert Watson [Wed, 24 Jun 2009 12:52:59 +0000 (12:52 +0000)]
Fix copy-and-paste-o's from kinfo_getfile.3 in kinfo_getvmmap.3.

MFC after: 3 days

15 years agoAdd stack_print_short() and stack_print_short_ddb() interfaces to
Robert Watson [Wed, 24 Jun 2009 12:06:15 +0000 (12:06 +0000)]
Add stack_print_short() and stack_print_short_ddb() interfaces to
stack(9), which generate a more compact rendition of a stack trace
via the kernel's printf.

MFC after: 1 week

15 years agoUse proper form of gnu designated initalizers. This lets
Roman Divacky [Wed, 24 Jun 2009 12:01:10 +0000 (12:01 +0000)]
Use proper form of gnu designated initalizers. This lets
clang compile this files.

Approved by: ed (mentor)
Silence from: harti (maintainer?)

15 years agoPrintf fewer warnings when adding a route to an atalk address fails;
Robert Watson [Wed, 24 Jun 2009 10:46:03 +0000 (10:46 +0000)]
Printf fewer warnings when adding a route to an atalk address fails;
userspace will print the error.

MFC after: 3 days

15 years agoIn if_setlladdr(), use IF_ADDR_LOCK() and ifaddr references to improve
Robert Watson [Wed, 24 Jun 2009 10:36:48 +0000 (10:36 +0000)]
In if_setlladdr(), use IF_ADDR_LOCK() and ifaddr references to improve
the safety of link layer address manipulation.

MFC after: 6 weeks

15 years agoIn ARP input, more consistently acquire and release ifaddr references.
Robert Watson [Wed, 24 Jun 2009 10:33:35 +0000 (10:33 +0000)]
In ARP input, more consistently acquire and release ifaddr references.

MFC after: 6 weeks

15 years agoBreak at_ifawithnet() into two variants:
Robert Watson [Wed, 24 Jun 2009 10:32:44 +0000 (10:32 +0000)]
Break at_ifawithnet() into two variants:

- at_ifawithnet(), which acquires an locks it needs and returns an
  at_ifaddr reference.
- at_ifawithnet_locked(), which relies on the caller locking
  at_ifaddr_list, and returns a pointer rather than a reference.

Update various consumers to prefer one or the other, including ether
and fddi output, to properly release at_ifaddr references.

Rework at_control() to manage locking and references in a manner
identical to in_control().

MFC after: 6 weeks

15 years agoReduce debugging output for netatalk routing events.
Robert Watson [Wed, 24 Jun 2009 10:28:30 +0000 (10:28 +0000)]
Reduce debugging output for netatalk routing events.

MFC after: 3 days

15 years agoInitialize the uip to silence gcc warning that seems to sneak in in some
Konstantin Belousov [Wed, 24 Jun 2009 09:26:33 +0000 (09:26 +0000)]
Initialize the uip to silence gcc warning that seems to sneak in in some
build environments.

Reported by: alc, bf1783 at googlemail com

15 years agoLock if_addrhead when iterating, and where necessary acquire and release
Robert Watson [Wed, 24 Jun 2009 08:53:23 +0000 (08:53 +0000)]
Lock if_addrhead when iterating, and where necessary acquire and release
ifadr references in if_sppp.

MFC after: 6 weeks