Commit Graph

193494 Commits

Author SHA1 Message Date
Pyun YongHyeon
b624ef0a8e Add support for QAC AR816x/AR817x Gigabit/Fast Ethernet controllers.
These controllers seem to have the same feature of AR813x/AR815x and
improved RSS support(4 TX queues and 8 RX queues).  alc(4) supports
all hardware features except RSS.  I didn't implement RX checksum
offloading for AR816x/AR817x just because I couldn't get
confirmation from the Vendor whether AR816x/AR817x corrected its
predecessor's RX checksum offloading bug on fragmented packets.
This change adds supports for the following controllers.
 o AR8161 PCIe Gigabit Ethernet controller
 o AR8162 PCIe Fast Ethernet controller
 o AR8171 PCIe Gigabit Ethernet controller
 o AR8172 PCIe Fast Ethernet controller
 o Killer E2200 Gigabit Ethernet controller

Tested by:	Many
Relnotes:	yes
MFC after:	2 weeks
HW donated by:	Qualcomm Atheros Communications, Inc.
2014-10-08 05:47:01 +00:00
Pyun YongHyeon
b19487dff8 Add new quirk PCI_QUIRK_MSI_INTX_BUG to pci(4).
QAC AR816x/E2200 controller has a silicon bug that MSI interrupt
does not assert if PCIM_CMD_INTxDIS bit of command register is set.

Reviewed by:	jhb
2014-10-08 05:34:39 +00:00
Eitan Adler
920aa23dad don't reinvent the wheel: rely on basename(3)
Reviewed by:	nwhitehorn
2014-10-08 05:04:31 +00:00
Pyun YongHyeon
0999f75a2f Fix a long standing bug in MAC statistics register access. One
additional register was erroneously added in the MAC register set
such that 7 TX statistics counters were wrong.
2014-10-08 01:03:32 +00:00
Sean Bruno
f6f6703f27 Implement PLPMTUD blackhole detection (RFC 4821), inspired by code
from xnu sources.  If we encounter a network where ICMP is blocked
the Needs Frag indicator may not propagate back to us.  Attempt to
downshift the mss once to a preconfigured value.

Default this feature to off for now while we do not have a full PLPMTUD
implementation in our stack.

Adds the following new sysctl's for control:
net.inet.tcp.pmtud_blackhole_detection -- turns on/off this feature
net.inet.tcp.pmtud_blackhole_mss       -- mss to try for ipv4
net.inet.tcp.v6pmtud_blackhole_mss     -- mss to try for ipv6

Adds the following new sysctl's for monitoring:
-- Number of times the code was activated to attempt a mss downshift
net.inet.tcp.pmtud_blackhole_activated
-- Number of times the blackhole mss was used in an attempt to downshift
net.inet.tcp.pmtud_blackhole_min_activated
-- Number of times that we failed to connect after we downshifted the mss
net.inet.tcp.pmtud_blackhole_failed

Phabricator:	https://reviews.freebsd.org/D506
Reviewed by:	rpaulo bz
MFC after:	2 weeks
Relnotes:	yes
Sponsored by:	Limelight Networks
2014-10-07 21:50:28 +00:00
Navdeep Parhar
19abdd0654 cxgbe/tom: don't leak resources tied to an active open request that
cannot be sent to the chip because a prerequisite L2 resolution
failed.

Submitted by:	Hariprasad at chelsio dot com (original version)
MFC after:	2 weeks.
2014-10-07 21:26:22 +00:00
Jung-uk Kim
37417245bf Make kern.nswbuf tunable from loader.
MFC after:	1 week
2014-10-07 20:13:47 +00:00
Edward Tomasz Napierala
7ebeea9aa6 Remove call to access(2) which didn't serve any purpose, and make it more
tolerant to errors.

MFC after:	1 month
Sponsored by:	The FreeBSD Foundation
2014-10-07 19:08:02 +00:00
Gavin Atkinson
4c62148d7b Support the Vodafone R215 LET USB dongle, which is apparently a rebadged
E5372 with different product IDs.

Interestingly, the standard E5372 IDs (12d1:1506) are currently listed in
u3g.c and are the same as the E3131.  However, the R215/E5372 is an NCM
device and works well with cdce(4) whereas the E3131 isn't.  More work
may be needed to better identify the other device IDs.

MFC after:	1 week
2014-10-07 19:07:50 +00:00
Aleksandr Rybalko
c32f462f28 Allow vt(4) to disable terminal bell with sysctl kern.vt.bell_enable=0,
similar as syscons(4) do.

Submitted by:	Tiwei Bie <btw@mail.ustc.edu.cn>
MFC after:	2 weeks
Sponsored by:	The FreeBSD Foundation
2014-10-07 18:47:53 +00:00
Bjoern A. Zeeb
c896236487 Since introducing the extra mapping in r250103 for architectural performance
events we have actually counted 'Branch Instruction Retired' when people
asked for 'Unhalted core cycles' using the 'unhalted-core-cycles' event mask
mnemonic.

Reviewed by:	jimharris
Discussed with:	gnn, rwatson
MFC after:	3 days
Sponsored by:	DARPA/AFRL
2014-10-07 18:00:34 +00:00
Ruslan Bukin
41709d23c4 Add driver for Synopsys DesignWare Mobile Storage Host Controller.
Sponsored by:	DARPA, AFRL
2014-10-07 17:39:30 +00:00
Neel Natu
8ccb28efcd Implement the FLUSH operation in the virtio-block emulation.
This gets rid of the following error message during FreeBSD guest bootup:
"vtbd0: hard error cmd=flush fsbn 0"

Reported by:	rodrigc
Reviewed by:	grehan
2014-10-07 17:08:53 +00:00
Andriy Gapon
c3d1d2e104 l2arc_write_buffers: reduce headroom value
FreeBSD has ARC_BUFC_NUMMETADATALISTS metadata lists and ARC_BUFC_NUMDATALISTS
data lists (currently both are 16) while illumos has just a single list
of each kind.

headroom determines how much data is scanned on a single list
during each run of the l2arc feed thread.
Because FreeBSD has more lists we proportionally decrease the limit.

Reviewed by:	Brendan Gregg (earlier version)
MFC after:	2 weeks
Sponsored by:	HybridCluster
2014-10-07 16:08:21 +00:00
Andriy Gapon
9f96723ec5 revert r272702: wrong (earlier) change was committed 2014-10-07 16:06:10 +00:00
Michael Tuexen
5558cc334d Fix a bug introduced in
https://svnweb.freebsd.org/base?view=revision&revision=272347

MFC after: 3 days
2014-10-07 16:01:17 +00:00
Andriy Gapon
4c3b02bfce reduce L2ARC_WRITE_SIZE on FreeBSD
FreeBSD has ARC_BUFC_NUMMETADATALISTS metadata lists and ARC_BUFC_NUMDATALISTS
data lists (currently both are 16) while illumos has just a single list
of each kind.

L2ARC_WRITE_SIZE determines the default value of l2arc_write_max which
defines limits on how much data is scanned and written to a cache device
during each run of the l2arc feed thread.  The limits are applied on the
per buffer list basis.
Because FreeBSD has more lists we proportionally reduce the limits.

Reviewed by:	Brendan Gregg (earlier version)
MFC after:	2 weeks
Sponsored by:	HybridCluster
2014-10-07 14:30:24 +00:00
Andriy Gapon
ab26525af2 make userland __assfail from opensolaris compat honor 'aok' variable
This should allow zdb -A option to actually make difference.

MFC after:	2 weeks
2014-10-07 14:15:50 +00:00
Andrey V. Elsukov
9ef268219a Our packet filters use mbuf's rcvif pointer to determine incoming interface.
Change mbuf's rcvif to enc0 and restore it after pfil processing.

PR:		110959
Sponsored by:	Yandex LLC
2014-10-07 13:31:04 +00:00
Hans Petter Selasky
15983afb04 Fix compile warning when compiling with GCC.
MFC after:	3 days
Sponsored by:	Mellanox Technologies
2014-10-07 10:04:25 +00:00
Andrey A. Chernov
097c578c1e 1) Fix the case we have less arguments for format string than we expected.
2) Return error on unsupported format specs.
(both according to POSIX)

PR:     93197
2014-10-07 06:34:05 +00:00
Andrey A. Chernov
e15d3f3c09 Back out timegm error check from r272562.
POSIX treats negative time_t as undefined (i.e. may be valid too,
depends on system's policy we don't have) and we don't set EOVERFLOW
in mktime/timegm as POSIX requires to surely distinguish -1 return
as valid negative time from -1 as error return.
2014-10-07 06:02:08 +00:00
Xin LI
af1360206c Add MLINK for explicit_bzero(3) and bump .Dd date.
MFC after:	2 weeks
2014-10-07 04:59:11 +00:00
Xin LI
0ef99c7017 Bump __FreeBSD_version for the addition of explicit_bzero(3). 2014-10-07 04:54:47 +00:00
Xin LI
78b59024b5 Add explicit_bzero(3) and its kernel counterpart.
Obtained from:	OpenBSD
MFC after:	2 weeks
2014-10-07 04:54:11 +00:00
Mark Johnston
84e874bd6c Treat D keywords as identifiers in certain postfix expressions. This allows
one to, for example, access the "provider" field of a struct g_consumer,
even though "provider" is a D keyword.

PR:		169657
MFC after:	2 months
Discussed with:	Bryan Cantrill
Sponsored by:	EMC / Isilon Storage Division
2014-10-06 21:52:40 +00:00
Neel Natu
65145c7f50 Inject #UD into the guest when it executes either 'MONITOR' or 'MWAIT'.
The hypervisor hides the MONITOR/MWAIT capability by unconditionally setting
CPUID.01H:ECX[3] to 0 so the guest should not expect these instructions to
be present anyways.

Discussed with:	grehan
2014-10-06 20:48:01 +00:00
John Baldwin
23c7e79ad8 Properly set the timeout in a query_state. The global query_timeout
configuration value is an integer count of seconds, it is not a timeval.
Using memcpy() to copy a timeval from it put garbage into the tv_usec
field.

PR:		194025
Submitted by:	David Shane Holden <dpejesh@yahoo.com>
MFC after:	1 week
2014-10-06 18:16:45 +00:00
John Baldwin
a2f67d9087 Fix build for i386 kernels with out 'I686_CPU'.
PR:		193660
Submitted by:	holger@freyther.de
2014-10-06 18:11:05 +00:00
Luigi Rizzo
86f82ef946 Add netmap support to libpcap. Tcpdump and other native pcap application can now
run directly on netmap ports using netmap:foo or valeXX:YY device names.
Modifications to existing code are small and trivial, the netmap-specific
code is all in a new file.

Please be aware that in netmap mode the physical interface is disconnected from
the host stack, so libpcap will steal the traffic not just make a copy.

For the full version of the code (including linux and autotools support) see
	https://code.google.com/p/netmap-libpcap/

MFC after:	3 days
2014-10-06 15:48:28 +00:00
John Baldwin
4a872f3d4f Call initializecpucache() on the BSP for i386 in the !XEN case. This was
my bug in r271409 that I noticed while reviewing r272492.
2014-10-06 15:43:57 +00:00
Baptiste Daroussin
11fce282db Import byacc 20141005 from vendor 2014-10-06 15:12:59 +00:00
Luigi Rizzo
2d4dfaf58b Fix comment and sort rights by name
MFC after:	3 days
2014-10-06 15:03:08 +00:00
Baptiste Daroussin
f78bee35d7 import byacc 20141005 2014-10-06 14:55:38 +00:00
Alexander Motin
204ca472bf Set CAM_SIM_QUEUED flag before calling ctl_queue() to avoid race.
PR:		194128
Submitted by:	Scott M. Ferris <smferris@gmail.com>
MFC after:	3 days
Sponsored by:	EMC/Isilon Storage Division
2014-10-06 14:52:04 +00:00
Craig Rodrigues
ab7bf3d49d MFV:
use calloc in get_line() when allocating line to ensure it is fully initialized,
 fixes a later uninitialized value in copy_param() (FreeBSD #193499).

PR: 193499
Submitted by: Thomas E. Dickey  <tom@invisible-island.net>
2014-10-06 14:43:02 +00:00
Alexander V. Chernikov
bbd5a84297 Improve r272609 (O_TCPOPTS).
MFC after:	3 dayes
2014-10-06 12:29:06 +00:00
Alexander Motin
04f10b6c96 Add support for MaxBurstLength and Expected Data transfer Length parameters.
Before this change target could send R2T request for write transfer of any
size, that could violate iSCSI RFC, which allows initiator to limit maximum
R2T size by negotiating MaxBurstLength connection parameter.

Also report an error in case of write underflow, when initiator provides
less data than initiator expects.  Previously in such case our target
sent R2T request for non-existing data, violating the RFC, and confusing
some initiators.  SCSI specs don't explicitly define how write underflows
should be handled and there are different oppinions, but reporting error
is hopefully better then violating iSCSI RFC with unpredictable results.

MFC after:	2 weeks
2014-10-06 12:20:46 +00:00
Alexander V. Chernikov
3615981425 Fix O_TCPOPTS processing.
Obtained from:	luigi
2014-10-06 11:15:11 +00:00
Alexander V. Chernikov
9bb47e7068 Improve "reserved keywords" hack:
we can't easily predict (in current parsing model)
if the keyword is ipfw(8) reserved keyword or port name.
Checking proto database via getprotobyname() consumes a lot of
CPU and leads to tens of seconds for parsing large ruleset.
Use list of reserved keywords and check them as pre-requisite
before doing getprotobyname().

Obtained from:	Yandex LLC
2014-10-06 11:00:47 +00:00
Alexander Motin
88971a900d Use r271207 optimization only for MSI-enabled HBAs.
It was found that VirtualBox' AHCI does not allow nterrupt to be cleared
before the interrupt status register is read, causing interrupt storm.

AHCI specification allows to skip this register use when multi-vector MSI
is enabled and so interrupting port is known.  For single-vector MSI that
is not stated explicitly, but if the port is only one, it is obviously
known too.
2014-10-06 10:58:54 +00:00
Andrew Turner
9677c48e30 Disable generating vfp and NEON instructions in the arm kernel. 2014-10-06 09:52:28 +00:00
Ganbold Tsagaankhuu
7f29c69aee Use documented compat string for msm uart.
Whilst here use tab instead of spaces.

Approved by:    stas (mentor)
2014-10-06 09:00:53 +00:00
Xin LI
1b5bcb8425 MFV r272591:
Use loaned ARC buffer for zfs receive to avoid copy.

Illumos issue:
    5162 zfs recv should use loaned arc buffer to avoid copy

MFC after:	2 weeks
2014-10-06 07:29:17 +00:00
Mateusz Guzik
3a222fe000 devfs: tidy up after 272596
This moves a var to an if statement, no functional changes.

MFC after:	1 week
2014-10-06 07:22:48 +00:00
Xin LI
9ee8d90af5 MFV r272588:
Handle old format deadlist.

Illumos issue:
    5178 zdb -vvvvv on old-format pool fails in dump_deadlist()

MFC after:	2 weeks
2014-10-06 07:09:47 +00:00
Xin LI
8fb26f5aef MFV r272585:
Split the godfather zio into CPU number's to reduce lock
contention.

Illumos issue:
    5176 lock contention on godfather zio

MFC after:	2 weeks
2014-10-06 07:03:17 +00:00
Alexander Motin
5554eb9bfc Fix length of Extended INQUIRY Data VPD page.
MFC after:	3 days
2014-10-06 07:01:32 +00:00
Mateusz Guzik
1d1b55fbba devfs: don't take proctree_lock unconditionally in devfs_close
MFC after:	1 week
2014-10-06 06:20:35 +00:00
Hans Petter Selasky
b228e6bf57 Minor code styling.
Suggested by:	glebius @
2014-10-06 06:19:54 +00:00