89118 Commits

Author SHA1 Message Date
Andriy Gapon
c496727c54 vnode_if: fix locking protocol description for lookup and cachedlookup
Also remove the checks from vop_lookup_pre and vop_lookup_post, which
are now completely redundant (before this change they were partially
redundant).

Discussed with:	kib
MFC after:	10 days
2012-11-19 11:32:56 +00:00
Andriy Gapon
dbe922173c zfs_remove: assert that delete_now case is never true on FreeBSD
That case is specific to Solaris VFS and it would violate pretty
fundamental contracts of FreeBSD VFS.

Discussed with:	pjd
MFC after:	12 days
2012-11-19 11:30:08 +00:00
Andriy Gapon
7b069f7fee zfs_remove: set VV_NOSYNC flag if a node is unlinked
Suggested by:	kib
MFC after:	12 days
2012-11-19 11:25:20 +00:00
Juli Mallett
b3a6fcb3b7 Return port numbers for ATCA-7220 SPI interfaces in a different place for
consistency reasons, and to ensure that CRC addition is disabled on output.
With this, transmit seems to be working properly on the ATCA-7220.
2012-11-19 08:35:58 +00:00
Juli Mallett
fcd5eed4f0 Prevent hang on ATCA-7220 when transmitting packets < 60 bytes. 2012-11-19 08:30:29 +00:00
Juli Mallett
d0ebf478da Remove redundant printf of SDK version which already appears earlier in boot. 2012-11-19 08:29:53 +00:00
Juli Mallett
72d324fab6 Fix packet receive on the ATCA-7220 by disabling FCS-related checks, since the
FCS is stripped by the underlying hardware before it reaches the Octeon.
2012-11-19 05:24:33 +00:00
Juli Mallett
8aff4e5fdd Add basic support for the Radisys-specific PCI console mechanism found on the
Radisys ATCA-7220.
2012-11-19 01:58:20 +00:00
Edward Tomasz Napierala
ed1f69514a Fix build on powerpc.
Reviewed by:	nwhitehorn
2012-11-19 01:15:32 +00:00
Juli Mallett
8961aadb2b o) Do boot descriptor parsing before console setup so that we can use a console
other than UART 0 from the outset.
o) Print board information from sysinfo after consoles have been initialized
   rather than doing it during boot descriptor parsing.
o) Use cvmx_safe_printf and platform_reset rather than panic when doing very
   early boot descriptor parsing before the console is set up.
o) Get rid of the global octeon_bootinfo.
2012-11-19 00:19:27 +00:00
Edward Tomasz Napierala
213a71c68b Fix build of kdump(1). 2012-11-18 22:03:31 +00:00
Adrian Chadd
73f81b5b19 Remove this include, it isn't needed. 2012-11-18 20:41:46 +00:00
Edward Tomasz Napierala
1848286ada Add UFS writesuspension mechanism, designed to allow userland processes
to modify on-disk metadata for filesystems mounted for write.

Reviewed by:	kib, mckusick
Sponsored by:	FreeBSD Foundation
2012-11-18 18:57:19 +00:00
Hiroki Sato
3829d13564 Check if an extracted zoneid is equal to the non-zero sin6_scope_id only when
it is link-local or MC interface-local.
2012-11-18 16:06:51 +00:00
Mateusz Guzik
dd103d4d06 Fix possible fp reference leak in posix_openpt
Reviewed by:	ed
Approved by:	trasz (mentor)
MFC after:	3 days
2012-11-18 15:48:34 +00:00
Adrian Chadd
73e357ea5b Correctly use spaces here.
Pointed out by:	pjd
2012-11-18 14:05:28 +00:00
Andriy Gapon
62875a6b10 spa_import_rootpool: fall back to use configuration from zpool.cache...
if we fail to generate a proper root pool configuration based on disk
probing.  Currently we can not properly generate the configuration for
multi-vdev pools.  Make that explicit.

Reported by:	madpilot, Bartosz Stec <bartosz.stec@it4pro.pl>
Tested by:	madpilot, Bartosz Stec <bartosz.stec@it4pro.pl>
MFC after:	4 days
2012-11-18 11:47:25 +00:00
Adrian Chadd
d60ec817ea Fix up a compile time warning if INET6 isn't defined. 2012-11-18 04:51:46 +00:00
Juli Mallett
2d7499b141 Remove one wholly-unused and buggy routine and some nearby alternative symbols.
While here, also correct a comment that seems to imply that this file is
NetBSD's all-singing, all-dancing locore.S, rather than our conservative set of
assembly support routines.
2012-11-17 23:53:12 +00:00
Andrew Turner
7dc6ac7b74 Clean up the two i80321 copies of initarm to be closer to one another. 2012-11-17 23:06:00 +00:00
Adrian Chadd
b4d68cd536 Just compile the whole ath chipset support in. 2012-11-17 21:55:49 +00:00
Hiroki Sato
6bbfef9004 Fill sin6_scope_id in sockaddr_in6 before passing it from the kernel to
userland via routing socket or sysctl.  This eliminates the following
KAME-specific sin6_scope_id handling routine from each userland utility:

 sin6.sin6_scope_id = ntohs(*(u_int16_t *)&sin6.sin6_addr.s6_addr[2]);

This behavior can be controlled by net.inet6.ip6.deembed_scopeid.  This is
set to 1 by default (sin6_scope_id will be filled in the kernel).

Reviewed by:	bz
2012-11-17 20:19:00 +00:00
Michael Tuexen
3a51a2647a Add support for SCTP/UDP/IPV6.
This completes the support of
http://tools.ietf.org/html/draft-ietf-tsvwg-sctp-udp-encaps

MFC after: 1 week
2012-11-17 20:04:04 +00:00
Alexander Motin
6d36c821ad Summarize information about connected analog pins and EDID-like data received
from HDMI/DisplayPort devices in form of general connection status and
sound(4)-style channel matrix.  Now that information is only reported in
readable form to verbose logs, but potentially could be used by sound(4)
to correctly choose default devices and configure vchans.

Fix rear and side channels swap on analog 7.1 outputs.  As soon as there is
a huge mess in industry about naming and using of these channels, duplicate
rear channels of 4 and 5.1 streams to both read and side speakers.
2012-11-17 06:33:13 +00:00
Adrian Chadd
f447c9bf87 Ensure hwpmc support is correctly included. 2012-11-17 04:11:57 +00:00
Adrian Chadd
54491754cb Make MIPS24k PMC optional on "hwpmc_mips24k."
Requested by:	juli
2012-11-17 04:10:42 +00:00
Adrian Chadd
1766e26f55 make the MIPS24k HWPMC support require hwpmc_mips24k as well as hwpmc.
This way non-MIPS24k platforms (eg Octeon) can have separate PMC support.

Requested by:	juli
2012-11-17 04:09:28 +00:00
Adrian Chadd
c612af968c Migrate the AR71xx UART (an 8250 derivative) to hide behind uart_ar71xx.
The AR9330/AR9331 UART is a totally different thing, so having it included
with 'uart' is not going to work out.
2012-11-17 04:05:46 +00:00
Alan Cox
969a0af09d Update a comment to reflect the elimination of the hold queue in r242300. 2012-11-17 04:00:19 +00:00
Adrian Chadd
875d039e0a Correctly populate the RTS field.
Tested:
	* AR5210, STA mode, RTS enabled
2012-11-17 02:39:37 +00:00
Adrian Chadd
3a4d0022de * Remove ah_desc.h, it's not needed
* Add some shifts that I'm using in userspace (athalq.)

However, this exposes a fun little bug..
2012-11-17 02:39:09 +00:00
Adrian Chadd
93cad1bdbd .. include ah_desc.h here now. 2012-11-17 02:02:36 +00:00
Adrian Chadd
f8af1be8f8 Remove the ah_desc.h reference; it's not needed.
I'm using these descriptor header files in userland and I'm trying to
avoid populating a compatibility ah_desc.h file.
2012-11-17 02:00:33 +00:00
Adrian Chadd
69f33b13d1 I'm not sure why ah_desc.h was required here, but it doesn't _need_
to be. So, just toss it.

There's no options or ah_desc fields in here.

Whilst I'm here, fix up the #ifdef and #define to mach.
2012-11-16 20:04:45 +00:00
Adrian Chadd
e3f0668803 * Remove a duplicate TX ALQ post routine!
* For CABQ traffic, I -can- chain them together using the next pointer
  and just push that particular chain head to the CABQ.  However, this
  doesn't magically make EDMA TX CABQ work - I have to do some further
  hoop jumping.
2012-11-16 19:58:15 +00:00
Adrian Chadd
bb327d284b ALQ logging enhancements:
* upon setup, tell the alq code what the chip information is.
* add TX/RX path logging for legacy chips.
* populate the tx/rx descriptor length fields with a best-estimate.
  It's overly big (96 bytes when AH_SUPPORT_AR5416 is enabled)
  but it'll do for now.

Whilst I'm here, add CURVNET_RESTORE() here during probe/attach as a
partial solution to fixing crashes during attach when the attach fails.
There are other attach failures that I have to deal with; those'll come
later.
2012-11-16 19:57:16 +00:00
Adrian Chadd
956d4fb965 ath(4) ALQ logging improvements.
* Add a new method which allows the driver to push the MAC/phy/hal info
  into the logging stream.
* Add a new ALQ logging entry which logs the mac/phy/hal information.
* Modify the ALQ startup path to log the MAC/phy/hal information
  so the decoder knows which HAL/chip is generating this information.
* Convert the header and mac/phy/hal information to use be32, rather than
  host order.  I'd like to make this stuff endian-agnostic so I can
  decode MIPS generated logs on a PC.

This requires some further driver modifications to correctly log the
right initial chip information.

Also - although noone bar me is currently using this, I've shifted the
debug bitmask around a bit.  Consider yourself warned!
2012-11-16 19:39:29 +00:00
Michael Tuexen
325c8c46b1 Get the accounting working. We now have counters how many
chunks for each SCTP outgoing stream are in the send and
sent queue.
While there, improve the naming of NR-SACK related constants
recently introduced.

MFC after: 1 week
2012-11-16 19:39:10 +00:00
Gleb Smirnoff
716963cb5d Update comment. 2012-11-16 14:00:54 +00:00
Andrey V. Elsukov
73cb2f38f2 Reduce the overhead of locking, use IF_AFDATA_RLOCK() when we are doing
simple lookups.

Sponsored by:	Yandex LLC
MFC after:	1 week
2012-11-16 12:12:02 +00:00
Konstantin Belousov
134eb42e24 In pget(9), if PGET_NOTWEXIT flag is not specified, also search the
zombie list for the pid. This allows several kern.proc sysctls to
report useful information for zombies.

Hold the allproc_lock around all searches instead of relocking it.
Remove private pfind_locked() from the new nfs client code.

Requested and reviewed by:	pjd
Tested by:	pho
MFC after:	3 weeks
2012-11-16 08:25:06 +00:00
Alexander Motin
1fc2a61440 For some more completeness add matrixes for 3.1 and 7.0 channel setups. 2012-11-16 07:05:57 +00:00
Konstantin Belousov
ea293f3f1d Restore the proper handling of the pid 0 for waitpid(2).
Fix the style around.

Reported and reviewed by:	bde (previous version)
MFC after:	28 days
2012-11-16 06:32:38 +00:00
Konstantin Belousov
e2d55f4866 Move the definition of the idtype_t from sys/types.h to sys/wait.h.
Fix the bug, use #if __BSD_VISIBLE instead of #if defined(__BSD_VISIBLE),
since __BSD_VISIBLE is always defined.
Reformat the comments from the Solaris style to KNF.

Reported and reviewed by:	bde
MFC after:	28 days
2012-11-16 06:29:52 +00:00
Konstantin Belousov
43bdcf9335 Alphabetically reorder the forward-declarations of the structures.
Add the declaration for enum idtype, to be used later.

Reported and reviewed by:	bde
MFC after:	28 days
2012-11-16 06:25:20 +00:00
Konstantin Belousov
a2a8559624 Style fixes for r242958.
Reported and reviewed by:	bde
MFC after:	28 days
2012-11-16 06:22:14 +00:00
Konstantin Belousov
43f48b65c0 Move the declaration of vm_phys_paddr_to_vm_page() from vm/vm_page.h
to vm/vm_phys.h, where it belongs.

Requested and reviewed by:	alc
MFC after:	2 weeks
2012-11-16 05:55:56 +00:00
Konstantin Belousov
962b064afe Explicitely state that M_USE_RESERVE requires M_NOWAIT, using assertion.
Reviewed by:	alc
MFC after:	2 weeks
2012-11-16 05:49:56 +00:00
Devin Teske
b98ba21ac2 + This patch does not change user experience or functionality
+ Cleanup syntax, slim-down code, and make things more readable
+ Introduce new +c! operator and ilk to reduce heap usage/allocations
+ Add safemode_enabled? safemode_enable and safemode_disable functions
+ Add singleuser_enabled? singleuser_enable singleuser_disable functions
+ Add verbose_enabled? verbose_enable and verbose_disable functions
+ Centralize strings (also to reduce heap usage)

Reviewed by:	peterj, adrian (co-mentor)
Approved by:	adrian (co-mentor)
2012-11-16 01:24:11 +00:00
Navdeep Parhar
e1e0aa8d3e cxgbe/tom: Plug mbuf leak.
MFC after:	3 days
2012-11-16 00:21:54 +00:00