202128 Commits

Author SHA1 Message Date
markj
b497150163 Use the linker to perform relocations in the SUNW_dof section rather than
doing them in drti during startup. This fixes a number of problems with
using USDT probes in stripped executables and shared libraries, and with
USDT probes in static functions.

Reviewed by:	rpaulo
MFC after:	1 month
Sponsored by:	EMC / Isilon Storage Division
Phabric: 	D751
2014-09-11 01:04:56 +00:00
hrs
08cd0ead43 Fix a bug which could break extended attributes in a dump output.
This occurred when a file was >892kB long and had a large data (>1kB)
in the extended attributes.

Reported by:	Masashi Toriumi
Reviewed by:	mckusick
2014-09-10 22:37:20 +00:00
hrs
3a5e998607 Fix header output when -P is specified and (ncpus - 1) != maxid.
Reported by:	Hiroaki Shimizu
PR:		152738
2014-09-10 22:34:08 +00:00
jhb
fa53fae9e4 MFamd64: Use initializecpu() to set various model-specific registers on
AP startup and AP resume (it was already used for BSP startup and BSP
resume).
- Split code to do one-time probing of cache properties out of
  initializecpu() and into initializecpucache().  This is called once on
  the BSP during boot.
- Move enable_sse() into initializecpu().
- Call initializecpu() for AP startup instead of enable_sse() and
  manually frobbing MSR_EFER to enable PG_NX.
- Call initializecpu() when an AP resumes.  In theory this will now
  properly re-enable PG_NX in MSR_EFER when resuming a PAE kernel on
  APs.
2014-09-10 21:37:47 +00:00
jhb
6f8d6cd57b To workaround an errata on certain Pentium Pro CPUs, i386 disables
the local APIC in initializecpu() and re-enables it if the APIC code
decides to use the local APIC after all.  Rework this workaround
slightly so that initializecpu() won't re-disable the local APIC if
it is called after the APIC code re-enables the local APIC.
2014-09-10 21:25:54 +00:00
mav
a3e07f1c6a Extend UNMAP blacklist on all STEC SSD models.
None of existing STEC devices need UNMAP or even support it well, having
many limitations and even hanging sometimes executing those commands.
New devices that may use UNMAP going to be released under HGST name.

MFC after:	3 days
2014-09-10 21:24:15 +00:00
imp
ea3c3dd245 Add support for calling pcibios routines from the
bootloader. Implement the following routines:
	pcibios-device-count	count the number of instances of a devid
	pcibios-read-config	read pci config space
	pcibios-write-config	write pci config space
	pcibios-find-devclass	find the nth device with a given devclass
	pcibios-find-device	find the nth device with a given devid
	pcibios-locator		convert bus device function ti pcibios locator
These commands are thin wrappers over their PCI BIOS 2.1 counterparts. More
informaiton, such as it is, can be found in the standard.

Export a nunmber of pcibios.X variables into the environment to report
what the PCI IDENTIFY command returned.

Also implmenet a new command line primitive (pci-device-count), but don't
include it by default just yet, since it depends on the recently added
words and any errors here can render a system unbootable.

This is intended to allow the boot loader to do special things based
on the hardware it finds. This could be have special settings that are
optimized for the specific cards, or even loading special drivers. It
goes without saying that writing to pci config space should not be
done without a just cause and a sound mind.

Sponsored by:	Netflix
2014-09-10 21:07:00 +00:00
jhb
2a48d5d52c Move code to set various MSRs on AMD cpus out of printcpuinfo() and
into initalizecpu() instead.
2014-09-10 21:04:44 +00:00
mav
73b5b7048f Add PCI ID for Promise TX8660 8-port 3Gbps HBA.
This device reports RAID subclass, but appears to be AHCI compatible.

Submitted by:	Yuri Perejilin <yuri@rivera.ru>
MFC after:	1 week
2014-09-10 19:53:31 +00:00
asomers
c3952a5fba Regen after 271401.
Sponsored by:	Spectra Logic
2014-09-10 19:22:14 +00:00
asomers
aebc4f8d52 Conditionalize build of etcupdate(8) on MK_RCS. Since etcupdate calls
merge(1), which is part of the RCS package, it must not be installed if
WITHOUT_RCS update is set. Otherwise, it will produce confusing errors.

CR:		https://reviews.freebsd.org/D691
MFC after:	1 week
Sponsored by:	Spectra Logic
2014-09-10 19:00:17 +00:00
ae
1576b695b6 Add scope zone id to the in_endpoints and hc_metrics structures.
A non-global IPv6 address can be used in more than one zone of the same
scope. This zone index is used to identify to which zone a non-global
address belongs.

Also we can have many foreign hosts with equal non-global addresses,
but from different zones. So, they can have different metrics in the
host cache.

Obtained from:	Yandex LLC
Sponsored by:	Yandex LLC
2014-09-10 16:26:18 +00:00
andrew
249f5cbad9 Unify interrupts bit definition and usage. While here remove PSR_C_bit.
Submitted by:	Svatopluk Kraus <onwahe at gmail.com>,
		Michal Meloun <meloun at miracle.cz>
Differential Revision: https://reviews.freebsd.org/D754
2014-09-10 15:25:15 +00:00
asomers
4dbc32b6e1 Abort the create_socket test if socket creation fails.
MFC after:	1 week
Reported by:	Coverity
CID:		1232756
2014-09-10 14:57:26 +00:00
ae
c8498c6b7f Add additional checks for IPV6_PKTINFO handling (RFC 3542):
* Return ENETDOWN when interface specified by ipi6_ifindex is not
  enabled for IPv6 use.
* Return EADDRNOTAVAIL when ipi6_ifindex specifies an interface, but the
  address ipi6_addr is not available for use on that interface.
* Return EINVAL when ipi6_addr is multicast address.

Obtained from:	Yandex LLC
Sponsored by:	Yandex LLC
2014-09-10 14:32:07 +00:00
trasz
d8aaa0480a Make sure we handle less than zero timeouts in iSCSI initiator and target
in a reasonable way.

Sponsored by:	The FreeBSD Foundation
2014-09-10 14:04:10 +00:00
andrew
a363841a18 Add more register values to armreg.h and remove CPU_CONTROL_32BP_ENABLE
from asm.h as they were already defined in armreg.h.

Submitted by:	Michal Meloun <meloun at miracle.cz>
2014-09-10 13:38:52 +00:00
trasz
6cacb7cf77 Make it possible to disable NOP-In PDUs by the iSCSI initiator by setting
kern.cam.ctl.iscsi.ping_timeout to 0.  This fixes interoperability with
some initiators that don't properly support NOP-Ins, namely iPXE/gPXE.

Submitted by:	Chen Wen <pokkys@gmail.com>
MFC after:	1 week
Sponsored by:	The FreeBSD Foundation
2014-09-10 13:34:27 +00:00
ae
9186ea6a59 Make in6_pcblookup_hash_locked and in6_pcbladdr static.
Obtained from:	Yandex LLC
Sponsored by:	Yandex LLC
2014-09-10 13:17:35 +00:00
andrew
924d84a44c Stop accessing the saved stack pointer by looking past the end of the
array of registers.

Submitted by:	Michal Meloun <meloun at miracle.cz>
2014-09-10 13:07:01 +00:00
glebius
5939c729a8 Remove unused arguments for VOP_GETPAGES(), VOP_PUTPAGES(). 2014-09-10 12:36:41 +00:00
ae
82d0b71937 Introduce INP6_PCBHASHKEY macro. Replace usage of hardcoded part of
IPv6 address as hash key in all places.

Obtained from:	Yandex LLC
2014-09-10 12:35:42 +00:00
ray
7c22eb535c Fix one more spelling mistake.
Pointed by:	danfe
2014-09-10 11:48:13 +00:00
ray
25f116b73a spelling fixes
Submitted by:	"Sam Fourman Jr." <sfourman@gmail.com>
MFC after:	1 week
2014-09-10 11:27:33 +00:00
ray
a95d8f6006 o Add sysctls to enable/disable potentially dengerous key combinations, like
reboot/halt/debug.
o Add support for most key combinations supported by syscons(4).

Reviewed by:	dumbbell, emaste (prev revision of D747)
MFC after:	5 days
Sponsored by:	The FreeBSD Foundation
2014-09-10 11:13:13 +00:00
andrew
ae8d0bb549 Move if_smc_fdt.c to live in sys/dev/smc. It's not specific to the ARM
Versatile hardware.
2014-09-10 10:59:17 +00:00
rwatson
da0f8310e7 Replace local copy-and-paste implementations of printmbuf() in several
device drivers with calls to the centralised m_print() implementation.
While the formatting and output details differ a little, the content
is essentially the same, and it is unlikely anyone has used this
debugging output in some time.

This change reduces awareness of mbuf cluster allocation (and,
especially, the M_EXT flag) outside of the mbuf allocator, which will
make it easier to refine the external storage mechanism without
disrupting drivers in the future.

Style bugs are preserved.

Reviewed by:	bz, glebius
MFC after:	3 days
Sponsored by:	EMC / Isilon Storage Division
2014-09-10 09:57:32 +00:00
mav
eb9cf8a129 Fix memory leak, reported by Coverity.
CID:		1229996
2014-09-10 08:06:31 +00:00
ngie
4cc9e0c39f Remove many false positives with make checkdpadd
- Reduce DPADD and LDADD in checkdpadd to -l<foo>
- Skip over -Wl,[es]*-group because -Wl,--end-group and
  -Wl,--start-group might be required to properly link objects (see
  usr.bin/clang/lldb as an example)

This caveat has been present for a while with some components of
the build. However, these false positives were made more more apparent
after r269648.

Phabric: D635
Reviewed by: jmmv (an earlier version)
PR: 192730
MFC after: 2 weeks
2014-09-10 07:55:51 +00:00
mav
645f3f53cf Make ctl_port_mask an array to support more then 32 ports.
Overflow reported by Coverity.

CID:		1229894
MFC after:	3 days
2014-09-10 07:16:17 +00:00
mav
20136fb37c Remove uninitialized and unused variable, reported by Coverity.
CID:		1230015
2014-09-10 07:00:36 +00:00
mav
4db7e51eb2 Fix array overrun, reported by Coverity.
CID:		1229970
2014-09-10 06:56:45 +00:00
mav
a209c226c5 Fix couple off-by-one range check errors, reported by Coverity.
CID:		1007837
2014-09-10 06:35:00 +00:00
mav
8633346319 Fix memory leak on error, reported by Coverity.
CID:		1007773
2014-09-10 06:29:31 +00:00
mav
aa30f2d2de Fix minor buffer overflow reported by Coverity.
CID:		1006781
2014-09-10 06:25:18 +00:00
alc
996a2e8d68 Fix a boundary case error in vm_reserv_alloc_contig(): If a reservation
isn't being allocated for the last of the requested pages, because a
reservation won't fit in the gap between allocated pages, then the
reservation structure shouldn't be initialized.

While I'm here, improve the nearby comments.

Reported by:	jeff, pho
MFC after:	1 week
Sponsored by:	EMC / Isilon Storage Division
2014-09-10 05:52:30 +00:00
grehan
f30a88f9b1 Fix issue with nmdm and leading zeros in device name.
The nmdm code enforces a number between the 'nmdm' and 'A|B' portions
of the device name. This is then used as a unit number, and sprintf'd
back into the tty name. If leading zeros were used in the name,
the created device name is different than the string used for the
clone-open (e.g. /dev/nmdm0001A will result in /dev/nmdm1A).

Since unit numbers are no longer required with the updated tty
code, there seems to be no reason to force the string to be a
number. The fix is to allow an arbitrary string between
'nmdm' and 'A|B', within the constraints of devfs names. This allows
all existing user of numeric strings to continue to work, and also
allows more meaningful names to be used, such as bhyve VM names.

Tested on amd64, i386 and ppc64.

Reported by:	Dave Smith
PR:		192281
Reviewed by:	neel, glebius
Phabric:	D729
MFC after:	3 days
2014-09-10 05:44:15 +00:00
ngie
cbe58136b8 Add diagnostic printfs and disable test # 4 on i386
Reviewed by: jmmv, rpaulo
MFC after: 3 days
Phabric: D749
PR: 191676
Sponsored by: EMC / Isilon Storage Division
2014-09-10 03:54:57 +00:00
gjb
bd9dfb26f5 Bump __FreeBSD_version after SA-14:18
Approved by:	re (implicit)
Sponsored by:	The FreeBSD Foundation
2014-09-10 00:19:33 +00:00
grehan
c7ae60cb63 Allow vtnet operation without merged rx buffers.
NetBSD's virtio-net implementation doesn't negotiate
the merged rx-buffers feature. To support this, check
to see if the feature was negotiated, and then adjust
the operation of the receive path accordingly by using
a larger iovec, and a smaller rx header.
In addition, ignore writes to the (read-only) status byte.

Tested with NetBSD/amd64 5.2.2, 6.1.4 and 7-beta.

Reviewed by:	neel, tychon
Phabric:	D745
MFC after:	3 days
2014-09-09 22:35:02 +00:00
ngie
10e2253fbd Expand the tests structure in test_small(..) to workaround the
"initializer not constant" warning with gcc

Approved by: jmmv (mentor)
MFC after: 3 days
Phabric: D744
Sponsored by: EMC / Isilon Storage Division
2014-09-09 22:14:15 +00:00
gjb
cea3546fa6 Add an arch.powerpc64 entity.
Sponsored by:	The FreeBSD Foundation
2014-09-09 19:58:55 +00:00
gjb
ef9091f941 Change how the recommended mailing list to track is
added to the footer of the release/doc/ pages by
moving a hard-coded value (that is subject to human
error to change) to release.ent where other values
are regularly changed, and adding parsing logic to
release.xsl.

Approved by:	re (implicit)
MFC after:	3 days
Sponsored by:	The FreeBSD Foundation
2014-09-09 19:51:57 +00:00
np
d63cc6f68a Whitespace nit.
MFC after:	1 week
2014-09-09 18:36:00 +00:00
bdrewery
ec56ad408f Don't cross mount boundaries when cleaning tmp files.
Mounting something in /tmp such as a build jail with nullfs mounts for
some directories can result in very surprising results the next day.

MFC after:	2 weeks
Relnotes:	yes
2014-09-09 17:03:58 +00:00
trasz
b7774f528d Use keys_add_int() where appropriate. No functional changes.
Sponsored by:	The FreeBSD Foundation
2014-09-09 16:57:02 +00:00
trasz
7fd1fb1a76 Fix ctld(8) to not forget to send TargetPortalGroupTag and TargetAlias
when the initiator skips security negotiation.  This fixes interoperability
with Xtend SAN initiator.

PR:		193021
MFC after:	1 week
Sponsored by:	The FreeBSD Foundation
2014-09-09 16:45:36 +00:00
trasz
76b4ee1f43 Avoid unlocking unlocked mutex in RCTL jail code. Specific test case
is attached to PR.

PR:		193457
MFC after:	1 week
Sponsored by:	The FreeBSD Foundation
2014-09-09 16:05:33 +00:00
mav
d6b90c0b3c Report that DPO and FUA bits are supported after r271311. 2014-09-09 15:19:38 +00:00
joel
ecdd8a906d Minor mdoc nit. 2014-09-09 14:34:54 +00:00