Commit Graph

106189 Commits

Author SHA1 Message Date
Tim J. Robbins
c5fe2733ea This commit was generated by cvs2svn to compensate for changes in r133543,
which included commits to RCS files with non-trunk default branches.
2004-08-12 05:37:46 +00:00
Tim J. Robbins
9da8305ad3 Import of GNU sort from coreutils CVS (trimmed) 2004-08-12 05:37:46 +00:00
Scott Long
dc9efde5b7 Destroy mutexes on detach. 2004-08-12 05:31:17 +00:00
Scott Long
ba1d57e7d9 Change FREE() to free() 2004-08-12 05:09:17 +00:00
Scott Long
bb6fe253cc Remove the AAC_LOCK macros. They no longer abstract anything and only
obfuscate the code.  No functional differences.
2004-08-12 05:05:06 +00:00
Scott Long
3576af8fd7 Release the sync fib after the controller has been shut down. This also
releases the I/O lock instead of just leaking it.
2004-08-12 04:47:32 +00:00
Robert Watson
52bc12b3da Set IFF_NEEDSGIANT for fwe network interface since the firewire
framework isn't yet MPSAFE.

Approved by:	simokawa
2004-08-12 03:02:16 +00:00
Andrey A. Chernov
5e29d1ab0a Back in the USSR (for me) 2004-08-12 02:23:24 +00:00
Alexander Kabaev
db20bc12fe This commit was generated by cvs2svn to compensate for changes in r133534,
which included commits to RCS files with non-trunk default branches.
2004-08-12 02:21:58 +00:00
Alexander Kabaev
025ee8b5ee Import one more file for Gcc 3.4.2 20040728. 2004-08-12 02:21:58 +00:00
Nate Lawson
310953d935 Fix the PRT entry code in acpi_pci_link to always add the entry, even if
there is no irq link.  Since we now use the stored copy of PRT, not the
one that used to be passed into acpi_pcib_route_interrupt(), we need it in
the list. [1]

Fix a bug in acpi_pci_find_prt() where we weren't checking the bus, thus
choosing the wrong PRT entry to use for routing the link.  Also, add a
printf for the case where the PRT entry is not found as this should not
happen.

Tested by:	marcel [1]
2004-08-12 02:06:19 +00:00
Robert Watson
c19c5239a6 When udp_send() fails, make sure to free the control mbufs as well as
the data mbuf.  This was done in most error cases, but not the case
where the inpcb pointer is surprisingly NULL.
2004-08-12 01:34:27 +00:00
John-Mark Gurney
4f34d8d132 add comment about needing to upgrade loader.rc on i386 systems that have
problems loading modules...

Solution by:	ru
2004-08-12 00:00:15 +00:00
Pawel Jakub Dawidek
887c9fd564 MFp4: Simplify code a bit:
- Remove kern.geom.mirror.sync_block_size sysctl. It is quite obvious that we
  want to use the biggest size possible.
- Do not use UMA zone for sync data allocations. There could be only one
  synchronization request per synchronized disk at a time, so allocate memory
  for one request on whole synchronization process related to one disk.

Tested by synchronizing one component (out of three) and by synchronizing
two components (out of three) in parallel.
2004-08-11 23:41:53 +00:00
David Xu
91574a97b7 Mark end of frames. 2004-08-11 23:23:05 +00:00
Pawel Jakub Dawidek
445a4b68f2 Actually, HARDCODED flag isn't stored in metadata, so don't bother
dumping it.
2004-08-11 22:16:42 +00:00
Pawel Jakub Dawidek
2def749bb1 - Fix typo.
- Dump HARDCODED flag.
2004-08-11 22:12:44 +00:00
Warner Losh
13dbaef46a Be explicit that a driver may assume softc is preserved when returning 0 2004-08-11 21:52:31 +00:00
David E. O'Brien
55dfe3e212 Add more aliases for common CPUTYPE spellings 2004-08-11 21:51:27 +00:00
Don Lewis
414ce15cb3 Deorbit the fxp tuning hint (hint.fxp.UNIT_NUMBER.ipcbxmit_disable)
introduced in if_fxp.c revision 1.180.  The bug fix committed in
revision 1.180 fixed the packet truncation problem.
2004-08-11 21:41:25 +00:00
Ruslan Ermilov
aaea4b16e1 Fix the country of birth.
Requested by:	dima
2004-08-11 21:23:42 +00:00
Marius Strobl
490a46bf68 Unbreak after struct resource was hidden. Tested with EBus front-end on
Sun AXe board.
2004-08-11 21:19:31 +00:00
Marius Strobl
a4954466bd - Use the rman_get_* functions instead of reaching into struct resource.
- Remove __RMAN_RESORUCE_VISIBLE again. It's no longer required either
  because of the above change or because struct rman is no longer hidden.

Reviewed by:	grehan
Tested by:	cross-compile on i386
2004-08-11 21:09:40 +00:00
Julian Elischer
0f54f48225 Properly keep track of how many kses are on the system run queue(s). 2004-08-11 20:54:48 +00:00
Nate Lawson
8f24713075 Add some more verbose warning/error messages to help with users reporting
problems with irq routing.
2004-08-11 20:37:24 +00:00
Justin T. Gibbs
28606ea437 Correctly export the size of our softc to newbus in our EISA attachement.
This avoids a panic upon first softc field reference.
2004-08-11 20:35:55 +00:00
Andre Oppermann
420a281164 Backout removal of UMA_ZONE_NOFREE flag for all zones which are established
for structures with timers in them.  It might be that a timer might fire
even when the associated structure has already been free'd.  Having type-
stable storage in this case is beneficial for graceful failure handling and
debugging.

Discussed with:	bosko, tegge, rwatson
2004-08-11 20:30:08 +00:00
David E. O'Brien
495b92b24e The last commit had one too many libs. 2004-08-11 19:19:26 +00:00
Andre Oppermann
2dc1d58164 Convert the routing table to use an UMA zone for rtentries. The zone is
called "rtentry".

This saves a considerable amount of kernel memory.  R_Zmalloc previously
used 256 byte blocks (plus kmalloc overhead) whereas UMA only needs 132
bytes.

Idea from:	OpenBSD
2004-08-11 17:26:56 +00:00
Warner Losh
a9a74ddc37 Remove badly broken pcic driver for NEWCARD 2004-08-11 17:24:42 +00:00
Warner Losh
1b868fa919 Remove pcic for NEWCARD 2004-08-11 17:23:32 +00:00
Warner Losh
215e417c4d Remove references to pcic for newcard from NOTES. 2004-08-11 17:22:37 +00:00
Andre Oppermann
4efb805c0c Remove the UMA_ZONE_NOFREE flag to all uma_zcreate() calls in the IP and
TCP code.  This flag would have prevented giving back excessive free slabs
to the global pool after a transient peak usage.
2004-08-11 17:08:31 +00:00
Hartmut Brandt
0d5467be9e Link to the build again. The problem was that we need a new
gensnmptree for the .def file to parse. This was fixed in
Makefile.inc1.
2004-08-11 16:47:12 +00:00
Hartmut Brandt
88e299900c Ensure that a new gensnmptree is built. This should fix the problems
with the snmp_atm module.
2004-08-11 16:45:20 +00:00
Dmitry Morozovsky
06967ced14 Add my birthday.
Approved by:	den (mentor)
2004-08-11 16:18:18 +00:00
Hartmut Brandt
6c6e198147 Again something bad is happening so comment out the ATM module. 2004-08-11 16:01:46 +00:00
Oliver Eikemeier
f6404ebe03 typo
Approved by:	ru
2004-08-11 15:47:13 +00:00
Oliver Eikemeier
a08b3a047a FWIIW, make the `test' target in pkg_install/version actually work. 2004-08-11 15:46:10 +00:00
Nate Lawson
e4116e931c Re-work ACPI PCI IRQ routing (_PRT, link devices). The old approach was
incomplete in that the PRT routing was not aware of link programming.
Fix this by doing all routing through the link devices.  The new algorithm
for setting up links is:

1. Read _CRS to get current setting.  If invalid (not in _PRS), then set
   to 0.
2. Attempt to call _DIS on the link.  If successful, mark the link as not
   routed.  Otherwise, assume it still is.

Then when a routing request occurs:

3. Update weights for all IRQs
4. Attempt to route the initial IRQ if valid
5. If that fails, walk through the sorted list, attempting to route IRQs.
6. Configure the trigger/polarity based on _PRS.

Other changes:
* Add acpi_pci_find_prt() to look up the PRT entry for a given device and
  acpi_pci_link_route() to select/route the best IRQ for it.
* Remove duplicated code in acpi_pcib_route_interrupt() that picked the
  first IRQ from _PRS.
* Remove unneeded arguments from acpi_pcib_resume() and friends.
* Ignore _STA on link devices but report if it seems strange.
* Add a prt_source handle to the PRT structure since the ACPI struct
  ACPI_PCI_ROUTING_TABLE uses a fixed-size entry for it.  We'll need to
  dynamically size this object if we want to use it the same way ACPI-CA
  does.  Null-terminate the source.

Tested by:	Luo Hong <luohong99_at_mails.tsinghua.edu.cn>,
		Jeffrey Katcher <jmkatcher_at_yahoo.com>
Info from:	jhb, Len Brown (Intel)
2004-08-11 14:52:50 +00:00
John-Mark Gurney
e8c9966d55 fix description of the macros for BUS_DMASYNC opertions. It's from the CPU's
point of view...  Add a reference to a better description of what the ops
do...

Reviewed by:	scottl
2004-08-11 14:52:33 +00:00
Pawel Jakub Dawidek
a5ef629f10 Increase default kern.geom.stripe.maxmem to 50 elements. 2004-08-11 12:57:17 +00:00
Andre Oppermann
67d0b24ed1 Make use of in_localip() function and replace previous direct LIST_FOREACH
loops over INADDR_HASH.
2004-08-11 12:32:10 +00:00
Hartmut Brandt
d6024e35bf Now, after the import of NgATM-1.1, re-enable the atm module. 2004-08-11 12:27:31 +00:00
Hartmut Brandt
63fa1c3303 Exclusion list for easier import. 2004-08-11 12:25:09 +00:00
Hartmut Brandt
f3e103df9f This commit was generated by cvs2svn to compensate for changes in r133492,
which included commits to RCS files with non-trunk default branches.
2004-08-11 12:21:36 +00:00
Hartmut Brandt
e42ab7cde1 Virgin import of NgATM shared kernel/user part 1.1 2004-08-11 12:21:36 +00:00
Hartmut Brandt
0943e320d0 Exclusion list for easier import. 2004-08-11 12:19:32 +00:00
Hartmut Brandt
36919bb12f Virgin import of NgATM user part 1.1 2004-08-11 12:17:54 +00:00
Hartmut Brandt
4b61cb7ef9 This commit was generated by cvs2svn to compensate for changes in r133488,
which included commits to RCS files with non-trunk default branches.
2004-08-11 12:17:54 +00:00