ƒ”<256776
159
223
151
252
176
2011
108
196
310
110
120
136
120
130
275
473
181
392
255
304
135
358
293
377
313
455
364
502
291
198
154
362
281
368
154
297
196
445
328
156
212
219
1524
142
367
306
436
550
129
551
256
251
171
198
308
829
196
227
379
145
205
183
106
148
251
151
124
521
1880
106
221
193
282
571
714
714
991
413
387
141
163
115
569
132
292
212
458
163
139
248
342
450
588
345
530
201
159
107
229
387
155
162
159
150
963
2097
106
1445
112
312
199
142
167
410
447
876
290
187
167
207
167
202
314
542
231
306
295
167
294
172
162
184
180
112
477
261
761
359
284
306
498
309
286
309
246
980
667
107
269
219
288
442
158
246
142
122
211
395
112
1330

K 10
svn:author
V 8
cperciva
K 8
svn:date
V 27
2013-10-19T21:55:42.381957Z
K 7
svn:log
V 125
Bump __FreeBSD_version to 1100001 for addition of support for "first boot"
rc.d scripts, so that ports can make use of this.

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-20T01:28:39.197522Z
K 7
svn:log
V 51
Allow hypervisor calls with more than 7 arguments.

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-20T01:31:09.409670Z
K 7
svn:log
V 151
Add initial driver for POWER hypervisor interpartition ethernet. This is
sufficient to pass traffic but needs some more work before merging to
STABLE.

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-20T01:40:59.639265Z
K 7
svn:log
V 76
Clean up some debug code that snuck in. Add a block comment on future work.

END
K 10
svn:author
V 5
andre
K 8
svn:date
V 27
2013-10-20T09:51:36.170316Z
K 7
svn:log
V 1915
Give ext_free way more latitude in handling of external mbuf storage
and associated refcounts.

The behavior is managed by using the m_ext.ext_flags to signal various
combinations of external storage and refcount management.  At every
step (*ext_free) can capture the whole mbuf freeing process by using
the return values EXT_FREE_CONT or EXT_FREE_DONE.

With EXT_FLAG_EMBREF no external refcount storage is allocated by the
mbuf system and a pointer must be supplied by the m_extadd() caller.
The location of the refcount storage must be inside the attached external
memory.  It is managed as usual by incrementing/decrementing the value
at the location.

With EXT_FLAG_EXTREF no external refcount storage is allocated by the
mbuf system and a pointer must be supplied by the m_extadd() caller.
The location of the refcount storage can be anywhere within KVM.  It
must not disappear until after the associated external storage has been
freed.  It is managed as usual by incrementing/decrementing the value
at the location.

With EXT_FLAG_REFCNT no external refcount storage is alloced by the
mbuf system.  No pointer needs to be supplied.  Every in/decrement
of the refcount is done through (*m_ext_free) and the new parameters
EXT_FREE_REFINC, EXT_FREE_REFDEC, and EXT_FREE_REFDEL.

The previous mbuf type EXT_EXTREF is removed and the m_ext.ext_flags
are used instead.

As long as the new features are not used the behavior remains the same
as before.

NB: This is API and implementation is still subject to changes and
refinements.

Other changes:

Simplify m_free() by removing m_tag_delete_chain() which is done
in the zone_mbuf dtor anyways.

Add typedef ext_free_t for (*)(struct mbuf *, void *, void *, int).

m_extadd() gains two more parameters: u_int refcnt, int ext_flags.

Users of the m_extadd()/MEXTADD() APIs in the generic kernel are
adjusted.  Within modules there is some fallout not yet fixed.

END
K 10
svn:author
V 5
andre
K 8
svn:date
V 27
2013-10-20T10:02:04.258074Z
K 7
svn:log
V 14
IFC @r256780.

END
K 10
svn:author
V 5
gavin
K 8
svn:date
V 27
2013-10-20T11:19:37.539657Z
K 7
svn:log
V 101
Add 26 new device IDs to uslcom(4).  This brings us in sync with Linux
v3.12 rc5.

MFC after:	3 days

END
K 10
svn:author
V 5
andre
K 8
svn:date
V 27
2013-10-20T11:55:39.346633Z
K 7
svn:log
V 215
Pimp m_print() to tell us everything about an mbuf, its headers and
contents.

Change m_print() to only print information about the one specified
mbuf.

Add m_printm() to print information on every mbuf in a chain.

END
K 10
svn:author
V 6
andrew
K 8
svn:date
V 27
2013-10-20T12:50:13.851525Z
K 7
svn:log
V 15
MFC to r256779

END
K 10
svn:author
V 6
andrew
K 8
svn:date
V 27
2013-10-20T12:55:23.064508Z
K 7
svn:log
V 25
Remove debugging left in

END
K 10
svn:author
V 5
andre
K 8
svn:date
V 27
2013-10-20T12:59:54.636259Z
K 7
svn:log
V 42
anity() to reflect the updated realities.

END
K 10
svn:author
V 7
antoine
K 8
svn:date
V 27
2013-10-20T13:41:42.515965Z
K 7
svn:log
V 24
Add more obsolete files

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-20T14:01:09.487185Z
K 7
svn:log
V 30
Add multicast filter control.

END
K 10
svn:author
V 6
andrew
K 8
svn:date
V 27
2013-10-20T14:52:14.206913Z
K 7
svn:log
V 179
The ARM EABI Run-time ABI defines the __aeabi_* floating-point functions to
pass their values in the integer registers. Update the hard-float version of
the functions to do this.

END
K 10
svn:author
V 6
andrew
K 8
svn:date
V 27
2013-10-20T15:13:32.733261Z
K 7
svn:log
V 377
Merge from projects/arm_eabi_vfp r255380:

Fix the VCVT instruction. It must round towards zero when converting from
a floating-point to an integer value. This was not the case causing issues
when printing certain values.

There is a VCVTR instruction that will round depending on the current
rounding mode. We don't yet support this instruction, or setting the
rounding mode.

END
K 10
svn:author
V 2
np
K 8
svn:date
V 27
2013-10-20T15:24:44.872075Z
K 7
svn:log
V 90
MFC r256459.

cxgbe(4): Update T4 and T5 firmwares to 1.9.12.0

Approved by:	re (glebius)

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-20T16:14:03.823021Z
K 7
svn:log
V 291
Unify the AIM and Book-E vm_machdep.c implementations, which previously
differed only with respect to the AIM version not following style(9) and
some additional features for 64-bit systems and machines with direct maps
in the AIM implementation that are no-ops on Book-E (at least for now).

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-20T16:37:03.860257Z
K 7
svn:log
V 154
Replace the two almost-exactly-identical AIM and Book-E clock.c
implementations with a single one after the application of a very small
amount of #ifdef.

END
K 10
svn:author
V 2
np
K 8
svn:date
V 27
2013-10-20T16:45:01.758930Z
K 7
svn:log
V 212
MFC r256477:

cxgbe(4): Store the log2 of the # of doorbells per BAR2 page for both
ingress and egress queues, and for both T4 and T5.  These values are
used by the T4/T5 iWARP driver.

Approved by:	re (glebius)

END
K 10
svn:author
V 5
andre
K 8
svn:date
V 27
2013-10-20T16:52:27.753435Z
K 7
svn:log
V 41
Replace bce_dump_mbuf() with m_printm().

END
K 10
svn:author
V 5
andre
K 8
svn:date
V 27
2013-10-20T17:47:40.517989Z
K 7
svn:log
V 263
Change lge(4) driver from managing its own jumbo mbuf pool to
the generic MJUM9BYTES jumbo buffers provided by the kernel.

NB: The driver always uses jumbo mbufs, even when the MTU is
the normal 1500.  That issue is not fixed.

Untested due to lack of hardware.

END
K 10
svn:author
V 5
andre
K 8
svn:date
V 27
2013-10-20T18:13:06.246318Z
K 7
svn:log
V 198
Sorta fix the wb(4) driver to allocate complete mbuf+cluster
packets and remove the need for MEXTADD.

I don't believe the driver was in working condition before.

Untested due to lack of hardware.

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-20T18:38:19.587187Z
K 7
svn:log
V 276
Some nexus devices add wildcard children. Since fdtbus_probe returned
BUS_PROBE_DEFAULT, it would attach to them all, producing both many
fdtbus instances and preventing other devices from attaching. Instead
return BUS_PROBE_NOWILDCARD, which exists for exactly this purpose.

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-20T18:40:55.156489Z
K 7
svn:log
V 212
Since the PS3 port was committed, the AIM nexus device works perfectly fine
on all PowerPC platforms, whether or not they have Open Firmware. Remove
some more duplication and have there be only one nexus driver.

END
K 10
svn:author
V 6
jilles
K 8
svn:date
V 27
2013-10-20T20:10:31.287217Z
K 7
svn:log
V 359
pathchk: Ensure bytes >= 128 are considered non-portable characters.

This was not broken on architectures such as ARM where char is unsigned.

Also, remove the first non-portable character from the output. POSIX does
not require this, and printing the first byte may yield an invalid byte
sequence with UTF-8.

PR:		bin/165988
Reported by:	Nicolas Rachinsky

END
K 10
svn:author
V 8
mckusick
K 8
svn:date
V 27
2013-10-20T20:41:38.228098Z
K 7
svn:log
V 266
First of several cleanups to soft dependency implementation.
Convert three functions exported from ffs_softdep.c to static
functions as they are not used outside of ffs_softdep.c.

No functional change.

Tested by:    Peter Holm and Scott Long
Sponsored by: Netflix

END
K 10
svn:author
V 6
jilles
K 8
svn:date
V 27
2013-10-20T20:50:17.643271Z
K 7
svn:log
V 406
popen(): Try to prevent inappropriate fd passing even if 'e' is not used.

Even though not all race conditions can be fixed if the 'e' option is not
used, still fix some race conditions using pipe2():

* Prevent both ends of the pipe from leaking to a concurrent popen().

* Prevent the child process's end of the pipe from leaking to any concurrent
  fork and exec.

This change also simplifies the code.

END
K 10
svn:author
V 8
mckusick
K 8
svn:date
V 27
2013-10-20T20:52:07.236556Z
K 7
svn:log
V 193
Second of several cleanups to soft dependency implementation.
Delete two unused functions in ffs_sofdep.c.

No functional change.

Tested by:    Peter Holm and Scott Long
Sponsored by: Netflix

END
K 10
svn:author
V 3
ian
K 8
svn:date
V 27
2013-10-20T21:03:15.867180Z
K 7
svn:log
V 105
Switch to using the standard uart console driver instead of the special
driver for early boot debugging.

END
K 10
svn:author
V 6
jilles
K 8
svn:date
V 27
2013-10-20T21:04:44.885687Z
K 7
svn:log
V 59
syslog: Use SOCK_CLOEXEC instead of separate fcntl() call.

END
K 10
svn:author
V 3
ian
K 8
svn:date
V 27
2013-10-20T21:07:38.669866Z
K 7
svn:log
V 269
Add a driver for the Freescale Fast Ethernet Controller found on various
Freescale SoCs including the i.MX series.  This also works for the newer
SoCs with the ENET gigabit controller, but doesn't use any of the new
hardware features other than enabling gigabit speed.

END
K 10
svn:author
V 6
andrew
K 8
svn:date
V 27
2013-10-20T21:11:36.350786Z
K 7
svn:log
V 185
In the hard float version of the fenv functions:
 * Remove _FPUSW_SHIFT and _ENABLE_MASK as they are unneeded.
 * Correctly spell fpscr, the Floating-point Status and Control Register.

END
K 10
svn:author
V 8
mckusick
K 8
svn:date
V 27
2013-10-20T21:11:40.762655Z
K 7
svn:log
V 270
Third of several cleanups to soft dependency implementation.
Ensure that softdep_unmount() and softdep_setup_sbupdate()
only get called for filesystems running with soft dependencies.

No functional change.

Tested by:    Peter Holm and Scott Long
Sponsored by: Netflix

END
K 10
svn:author
V 3
ian
K 8
svn:date
V 27
2013-10-20T21:21:07.820043Z
K 7
svn:log
V 62
Add configuration for the Freescale i.MX53 Quick Start Board.

END
K 10
svn:author
V 6
alfred
K 8
svn:date
V 27
2013-10-20T21:21:50.627703Z
K 7
svn:log
V 201
Fix resource free.

The order of releasing resources in mlxen was wrong, which caused
panic on reload of the module.

MFC: 256682

Submitted by:	Shahar Klein (shahark at mellanox.com)
Approved by:	 re

END
K 10
svn:author
V 2
ae
K 8
svn:date
V 27
2013-10-20T21:23:52.903746Z
K 7
svn:log
V 104
* Initialize sin6_scope_id using specified ifnet.
* Don't add loopback routes for link-local addresses.

END
K 10
svn:author
V 8
mckusick
K 8
svn:date
V 27
2013-10-20T22:21:01.555177Z
K 7
svn:log
V 347
Fourth of several cleanups to soft dependency implementation.
Add KASSERTS that soft dependency functions only get called
for filesystems running with soft dependencies. Calling these
functions when soft updates are not compiled into the system
become panic's.

No functional change.

Tested by:    Peter Holm and Scott Long
Sponsored by: Netflix

END
K 10
svn:author
V 5
gibbs
K 8
svn:date
V 27
2013-10-20T22:50:57.674339Z
K 7
svn:log
V 233
The Xen userland event channel driver isn't complete.  Disabled it
from the kernel build until it is ready.

sys/conf/files:
	Remove the entry for xen/evtchn/evtchn_dev.c so it is not included
	in any kernel builds.

Noticed by:	smh

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-20T23:02:16.354668Z
K 7
svn:log
V 56
Avoid sign overflow if there are more than 2 GB of RAM.

END
K 10
svn:author
V 3
ian
K 8
svn:date
V 27
2013-10-20T23:40:16.357149Z
K 7
svn:log
V 119
Calculate the baud rate divisor rather than using a hard-coded value.

Submitted by:	Steven Lawrance <stl@koffein.net>

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-21T00:04:26.465021Z
K 7
svn:log
V 118
Allow the OFW interrupt mapping code to work with PCI devices not enumerated
by Open Firmware, as in the case of FDT.

END
K 10
svn:author
V 8
mckusick
K 8
svn:date
V 27
2013-10-21T00:28:02.489497Z
K 7
svn:log
V 1425
Restructuring of the soft updates code to set it up so that the
single kernel-wide soft update lock can be replaced with a
per-filesystem soft-updates lock. This per-filesystem lock will
allow each filesystem to have its own soft-updates flushing thread
rather than being limited to a single soft-updates flushing thread
for the entire kernel.

Move soft update variables out of the ufsmount structure and into
their own mount_softdeps structure referenced by ufsmount field
um_softdep.  Eventually the per-filesystem lock will be in this
structure. For now there is simply a pointer to the kernel-wide
soft updates lock.

Change all instances of ACQUIRE_LOCK and FREE_LOCK to pass the lock
pointer in the mount_softdeps structure instead of a pointer to the
kernel-wide soft-updates lock.

Replace the five hash tables used by soft updates with per-filesystem
copies of these tables allocated in the mount_softdeps structure.

Several functions that flush dependencies when too many are allocated
in the kernel used to operate across all filesystems. They are now
parameterized to flush dependencies from a specified filesystem.
For now, we stick with the round-robin flushing strategy when the
kernel as a whole has too many dependencies allocated.

While there are many lines of changes, there should be no functional
change in the operation of soft updates.

Tested by:    Peter Holm and Scott Long
Sponsored by: Netflix

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-21T00:58:35.063878Z
K 7
svn:log
V 42
Fix 80-column line wrapping in a comment.

END
K 10
svn:author
V 2
np
K 8
svn:date
V 27
2013-10-21T01:10:37.874862Z
K 7
svn:log
V 275
MFC r256694, r256713, r256714.

r256694:
iw_cxgbe: iWARP driver for Chelsio T4/T5 chips.  This is a straight port
of the iw_cxgb4 found in OFED distributions.

r256713:
iw_cxgbe should have a dependency on t4nex.

r256714:
Fix typo in previous commit.

Approved by:	re (hrs)

END
K 10
svn:author
V 2
np
K 8
svn:date
V 27
2013-10-21T01:15:05.832928Z
K 7
svn:log
V 214
MFC r256467:

Delete all of the old RDMA code (except krping, which was switched to
use sys/ofed some time back).  This has been sitting around as dead code
in the tree for a very long time.

Approved by:	re (hrs)

END
K 10
svn:author
V 5
gibbs
K 8
svn:date
V 27
2013-10-21T04:00:23.763077Z
K 7
svn:log
V 341
MFC 256813:

The Xen userland event channel driver isn't complete.  Disabled it
from the kernel build until it is ready.

sys/conf/files:
	Remove the entry for xen/evtchn/evtchn_dev.c so it is not included
	in any kernel builds.

Noticed by:	smh
Approved by:	re (hrs)
------------------------------------------------------------------------

END
K 10
svn:author
V 5
markj
K 8
svn:date
V 27
2013-10-21T04:15:55.325501Z
K 7
svn:log
V 455
When fetching function arguments out of a frame on amd64, explicitly select
the register based on the argument index rather than relying on the fields
in struct reg to be in the right order. This assumption is incorrect on
FreeBSD and generally led to bogus argument values for the sixth argument
of PID and USDT probes; the first five are passed directly to dtrace_probe()
via the fasttrap trap handler and so were correctly handled.

MFC after:	2 weeks

END
K 10
svn:author
V 6
jasone
K 8
svn:date
V 27
2013-10-21T05:10:46.279577Z
K 7
svn:log
V 34
Update jemalloc to version 3.4.1.

END
K 10
svn:author
V 7
glebius
K 8
svn:date
V 27
2013-10-21T05:14:00.847470Z
K 7
svn:log
V 454
Provide a working example line for an interface with 1 address running
with CARP.

Currently, we've got a problem that interface isn't IFF_UP at the time
we assign it a redundant address, and the latter gets stuck in INIT state.
Additional SIOCSIFFLAGS from ifconfig(8) kicks it to a working state.

A proper fix is kernel side and appeared to be non-trivial, not to be
checked in before 10.0-RELEASE.

Submitted by:	Ole Myhre <ole.myhre dataoppdrag.no>

END
K 10
svn:author
V 7
yongari
K 8
svn:date
V 27
2013-10-21T05:57:36.931440Z
K 7
svn:log
V 159
MFC r256038:
  Fix clearing MAC stats registers. Previously it cleared every
  fourth register.

  Submitted by:	Paul A. Patience <paul-a.patience@polymtl.ca>

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-21T06:04:39.918687Z
K 7
svn:log
V 158
Partial MFproject/camlock r256671:
Fix several target mode SIMs to not blindly clear ccb_h.flags field of
ATIO CCBs.  Not all CCB flags there belong to them.

END
K 10
svn:author
V 7
yongari
K 8
svn:date
V 27
2013-10-21T06:22:20.399841Z
K 7
svn:log
V 75
Correct MAC revision bits. Previously it always cleared bit 20 and
bit 21.

END
K 10
svn:author
V 7
yongari
K 8
svn:date
V 27
2013-10-21T06:27:20.009693Z
K 7
svn:log
V 101
Add preliminary support for RTL8106E PCIe FastEthernet.

H/W donated by:	RealTek Semiconductor Corp.

END
K 10
svn:author
V 2
np
K 8
svn:date
V 27
2013-10-21T06:31:56.285420Z
K 7
svn:log
V 216
MFC r256470:

Update krping to the latest upstream code.  Move all the FreeBSD
specific parts to krping_dev.c, which leaves the other files as
close to their upstream versions as possible.

Approved by:	re (glebius)

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-21T06:44:55.237571Z
K 7
svn:log
V 736
MFprojects/camlock r256619:
Restore BIO_UNMAPPED and BIO_TRANSIENT_MAPPING in biodonne() when unmapping
temporary mapped buffer.  That fixes double unmap if biodone() called twice
for the same BIO (but with different done methods).

Move mapping removal before calling bio_done() method.  I believe that it is
very wrong to do anything to BIO after reporting completion.  kib@ thinks
it was done for some forgotten now case when bio_done() method needed mapped
buffer. But 1) if BIO was sent as unmapped, then IMO done() should be called
in the same way; 2) IMO there is no guatantee that buffer will be mapped at
this point at all, for example, if all underlying stack supports unmapped
I/O, so bio_done() handler can not expect that.

END
K 10
svn:author
V 3
pho
K 8
svn:date
V 27
2013-10-21T06:55:07.114933Z
K 7
svn:log
V 103
Added test scenario for per-filesystem soft-updates lock.

Sponsored by:	EMC / Isilon storage division

END
K 10
svn:author
V 2
ae
K 8
svn:date
V 27
2013-10-21T07:49:36.844892Z
K 7
svn:log
V 135
Add a note that lacp_compose_key() should be updated, when new media
types will be added.

Submitted by:	melifaro
X-MFC after:	r256689

END
K 10
svn:author
V 7
delphij
K 8
svn:date
V 27
2013-10-21T07:58:37.162689Z
K 7
svn:log
V 282
Drop cm_lock before calling mapper_close, which in turn could call
_citrus_mapper_close again and result in a deadlock otherwise.

This is similar to NetBSD PR/24023 (fixed in their r1.5 of this file).

PR:		bin/182994
Submitted by:	Fabian Keil <fk fabiankeil de>
MFC after:	3 days

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-21T08:37:17.370569Z
K 7
svn:log
V 53
Don't access BIO pointer after we called bio_done().

END
K 10
svn:author
V 3
hrs
K 8
svn:date
V 27
2013-10-21T08:49:46.878931Z
K 7
svn:log
V 112
- Fix jail_parallel_start="YES".
- Fix ip[46].addr when interface parameter is not defined.

Spotted by:	rpaulo

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-21T08:57:27.742675Z
K 7
svn:log
V 91
MFprojects/camlock:
Remove hard limit on number of BIOs handled with one ATA TRIM request.

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-21T09:03:30.937655Z
K 7
svn:log
V 14
MFC @ r256836

END
K 10
svn:author
V 5
trasz
K 8
svn:date
V 27
2013-10-21T09:12:04.555985Z
K 7
svn:log
V 54
Don't test arrays for being NULL.

MFC after:	1 month

END
K 10
svn:author
V 7
ganbold
K 8
svn:date
V 27
2013-10-21T09:34:04.890213Z
K 7
svn:log
V 154
Move and rename dwc otg driver to more
generic one as it appears to work
for rk3188 SoC based board too.

No objections from: hselasky@
Reviewed by: ray@

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-21T09:51:06.213614Z
K 7
svn:log
V 59
Collect some garbage left after several code refactorings.

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-21T10:08:39.908052Z
K 7
svn:log
V 32
Includes are not needed either.

END
K 10
svn:author
V 8
bdrewery
K 8
svn:date
V 27
2013-10-21T10:09:48.527565Z
K 7
svn:log
V 423
Fix 'make delete-old-libs' and 'make check-libs' to delete .debug
files created by WITH_DEBUG_FILES. Also cleanup .symbols files from
the period between r244236 when .symbols were supported and r251512
when they were renamed to .debug.

Only propose to delete a .debug file if the corresponding library
itself was deleted already.

Reported by:	des
Reviewed by:	emaste (earlier version)
Approved by:	bapt
MFC after:	3 days

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-21T12:00:26.561794Z
K 7
svn:log
V 1786
Merge CAM locking changes from the projects/camlock branch to radically
reduce lock congestion and improve SMP scalability of the SCSI/ATA stack,
preparing the ground for the coming next GEOM direct dispatch support.

Replace big per-SIM locks with bunch of smaller ones:
 - per-LUN locks to protect device and peripheral drivers state;
 - per-target locks to protect list of LUNs on target;
 - per-bus locks to protect reference counting;
 - per-send queue locks to protect queue of CCBs to be sent;
 - per-done queue locks to protect queue of completed CCBs;
 - remaining per-SIM locks now protect only HBA driver internals.

While holding LUN lock it is allowed (while not recommended for performance
reasons) to take SIM lock.  The opposite acquisition order is forbidden.
All the other locks are leaf locks, that can be taken anywhere, but should
not be cascaded.  Many functions, such as: xpt_action(), xpt_done(),
xpt_async(), xpt_create_path(), etc. are no longer require (but allow) SIM
lock to be held.

To keep compatibility and solve cases where SIM lock can't be dropped, all
xpt_async() calls in addition to xpt_done() calls are queued to completion
threads for async processing in clean environment without SIM lock held.

Instead of single CAM SWI thread, used for commands completion processing
before, use multiple (depending on number of CPUs) threads.  Load balanced
between them using "hash" of the device B:T:L address.

HBA drivers that can drop SIM lock during completion processing and have
sufficient number of completion threads to efficiently scale to multiple
CPUs can use new function xpt_done_direct() to avoid extra context switch.
Make ahci(4) driver to use this mechanism depending on hardware setup.

Sponsored by:	iXsystems, Inc.
MFC after:	2 months

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-21T12:12:17.316187Z
K 7
svn:log
V 14
MFC @ r256843

END
K 10
svn:author
V 8
mckusick
K 8
svn:date
V 27
2013-10-21T13:09:09.288368Z
K 7
svn:log
V 123
Fix build problem on ARM (which defaults to building without soft updates).

Reported by:  Tinderbox
Sponsored by: Netflix

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-21T13:57:34.697534Z
K 7
svn:log
V 93
Provide an interface for PCI bus drivers that need some of ofw_pci's
metadata during attach.

END
K 10
svn:author
V 3
kib
K 8
svn:date
V 27
2013-10-21T16:17:46.048829Z
K 7
svn:log
V 189
Print more useful information about the transfer that trigger the assertion.
Other data is available with ddb command 'show pginfo'.

Sponsored by:	The FreeBSD Foundation
MFC after:	1 week

END
K 10
svn:author
V 3
kib
K 8
svn:date
V 27
2013-10-21T16:22:51.242684Z
K 7
svn:log
V 478
Reset function on SandyBridge holds the gt_lock for the whole duration
already.  Also, according to the specs, GDRST register is not in the
power well, so the forcewake for reset status read is excessive for
this reason.

Use plain register read for waiting of the reset completion
notification, to avoid gt_lock recursion.  Linux upstream did the
similar change, but their code was already restructured.

Reported by:	ray
Sponsored by:	The FreeBSD Foundation
MFC after:	1 week

END
K 10
svn:author
V 3
kib
K 8
svn:date
V 27
2013-10-21T16:44:53.020539Z
K 7
svn:log
V 621
Add a resource limit for the total number of kqueues available to the
user.  Kqueue now saves the ucred of the allocating thread, to
correctly decrement the counter on close.

Under some specific and not real-world use scenario for kqueue, it is
possible for the kqueues to consume memory proportional to the square
of the number of the filedescriptors available to the process.  Limit
allows administrator to prevent the abuse.

This is kernel-mode side of the change, with the user-mode enabling
commit following.

Reported and tested by:	pho
Discussed with:	jmg
Sponsored by:	The FreeBSD Foundation
MFC after:	2 weeks

END
K 10
svn:author
V 3
kib
K 8
svn:date
V 27
2013-10-21T16:46:12.863307Z
K 7
svn:log
V 621
Add a resource limit for the total number of kqueues available to the
user.  Kqueue now saves the ucred of the allocating thread, to
correctly decrement the counter on close.

Under some specific and not real-world use scenario for kqueue, it is
possible for the kqueues to consume memory proportional to the square
of the number of the filedescriptors available to the process.  Limit
allows administrator to prevent the abuse.

This is kernel-mode side of the change, with the user-mode enabling
commit following.

Reported and tested by:	pho
Discussed with:	jmg
Sponsored by:	The FreeBSD Foundation
MFC after:	2 weeks

END
K 10
svn:author
V 6
emaste
K 8
svn:date
V 27
2013-10-21T18:20:20.275581Z
K 7
svn:log
V 895
MFC r244437: Support restrict qualifier in dwarf debug info

  Newer clang/llvm emit DW_TAG_restrict_type, which wasn't handled by gdb.
  Import support from Apple's gdb-1822:

  |  2009-03-24  Jason Molenda  (jmolenda@apple.com)
  |
  |    * gdbtypes.c (make_cv_type): Rename this function to make_cvr_type to
  |    also handle restrict qualifiers.
  |    (check_typedef): Handle TYPE_RESTRICT.
  |    * gdbtypes.h (TYPE_FLAG_RESTRICT, TYPE_RESTRICT): New.
  |    * hpread.c (hpread_type_lookup): Update to use make_cvr_type.
  |    * stabsread.c (read_type): Pass the restrict qualifiers along.
  |    * parse.c (follow_types): Pass the restrict qualifiers along.
  |    * dwarf2read.c (read_tag_const_type): Call make_cvr_type.
  |    (read_tag_volatile_type): Same.
  |    (read_tag_restrict_type): New function.
  |    (read_type_die): Handle DW_TAG_restrict_type.

  Obtained from: Apple

END
K 10
svn:author
V 6
emaste
K 8
svn:date
V 27
2013-10-21T18:23:16.917895Z
K 7
svn:log
V 317
MFC r251665: Handle DW_TAG_rvalue_reference_type

  2012-05-21  Jim Ingham <jingham@apple.com>

    * dwarf2read.c (process_die): Handle DW_TAG_rvalue_reference_type.
      (read_type_die): Ditto.
      (dwarf_tag_name): Ditto.

    * elf/dwarf2.h: Add DW_TAG_rvalue_reference_type.

  Obtained from:	Apple, gdb-1820

END
K 10
svn:author
V 6
emaste
K 8
svn:date
V 27
2013-10-21T18:25:36.074412Z
K 7
svn:log
V 291
MFC r251667: Handle DW_TAG_unspecified_type

  2011-11-14  Jim Ingham  <jingham@apple.com>

  * dwarf2read.c (read_type_die): Handle DW_TAG_unspecified_type.
    (read_tag_unspecified_type): New function, add a type for the
    DW_TAG_unspecified_type die.

  Obtained from:	Apple, gdb-1752

END
K 10
svn:author
V 6
emaste
K 8
svn:date
V 27
2013-10-21T18:26:22.782135Z
K 7
svn:log
V 46
MFC r251844: Include die tag in error message

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-21T18:47:02.105263Z
K 7
svn:log
V 63
Use standard ofw_bus helpers instead of reinventing the wheel.

END
K 10
svn:author
V 3
sjg
K 8
svn:date
V 27
2013-10-21T18:47:29.025660Z
K 7
svn:log
V 23
Fix botched sed script

END
K 10
svn:author
V 8
andreast
K 8
svn:date
V 27
2013-10-21T19:11:15.212667Z
K 7
svn:log
V 471
MFC: 256555

Move the resource allocation from the ata_*_probe section to the ata_*_attach
section. This prevents a boot crash on nearly all iMacs and PowerMacs/Books.

The allocation in the probe section was working before because ata_probe was
returning 0 which did not invoke a second DEVICE_PROBE. Now it returns
a BUS_PROBE_DEFAULT which can invoke a second DEVICE_PROBE which results in
a "failed to reserve resource" exit.

PR:	powerpc/182978
Approved by:	re(gjb)

END
K 10
svn:author
V 7
glebius
K 8
svn:date
V 27
2013-10-21T19:21:36.727694Z
K 7
svn:log
V 36
Start lection on memory management.

END
K 10
svn:author
V 6
emaste
K 8
svn:date
V 27
2013-10-21T20:38:02.681282Z
K 7
svn:log
V 196
Don't force 64-bit DWARF2 on MIPS

64-bit debug data is only necessary for objects with greater than 4GB of
debug data, and is not used on other 64-bit FreeBSD targets.

Sponsored by:	DARPA, AFRL

END
K 10
svn:author
V 6
brooks
K 8
svn:date
V 27
2013-10-21T20:51:08.809671Z
K 7
svn:log
V 116
Allow kernels without options SOFTUPDATES to build.  This should fix the
embedded tinderboxes.

Reviewed by:	emaste

END
K 10
svn:author
V 6
brooks
K 8
svn:date
V 27
2013-10-21T21:13:01.298460Z
K 7
svn:log
V 362
MFP4: 223121 (FDT infrastructure portion)

Implement support for interrupt-parent nodes in simplebus.  The current
implementation requires that device declarations have an interrupt-parent
node and that it point to a device that has registered itself as a
interrupt controller in fdt_ic_list_head and implements the fdt_ic
interface.

Sponsored by:   DARPA/AFRL

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-21T21:16:50.517873Z
K 7
svn:log
V 71
Add comments that taskqueue_enqueue_locked() returns without the lock.

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-21T21:32:24.172447Z
K 7
svn:log
V 47
Add and use constants for g_collectstats bits.

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-21T22:25:54.043431Z
K 7
svn:log
V 147
Make hard-wired TLB allocations be at minimum one page. This is required by
some implementations, most notably (in my case) QEMU's e500 emulation.

END
K 10
svn:author
V 6
brooks
K 8
svn:date
V 27
2013-10-21T22:43:38.823898Z
K 7
svn:log
V 246
Remove the isf(4) driver.  It was created by accident and is subset of
the cfi(4) driver.  It remained in the tree longer than would be ideal
due to the time required to bring cfi(4) to feature parity.

Sponsored by:	DARPA/AFRL
MFC after:	3 days

END
K 10
svn:author
V 6
sbruno
K 8
svn:date
V 27
2013-10-21T22:55:56.261187Z
K 7
svn:log
V 354
Resolve clang warning about a use of syslog. By using a proper enforcement
of string format in a call so syslog

/usr/src/gnu/lib/libssp/../../../contrib/gcclibs/libssp/ssp.c:137:23:
warning: format string is not a string literal (potentially insecure)
      [-Wformat-security]
    syslog (LOG_CRIT, msg1);
                      ^~~~

Reviewed by:	dim@

END
K 10
svn:author
V 4
neel
K 8
svn:date
V 27
2013-10-21T23:46:37.706017Z
K 7
svn:log
V 494
The ASID allocation in SVM is incorrect because it allocates a single ASID for
all vcpus belonging to a guest. This means that when different vcpus belonging
to the same guest are executing on the same host cpu there may be "leakage"
in the mappings created by one vcpu to another.

The proper fix for this is being worked on and will be committed shortly.

In the meantime workaround this bug by flushing the guest TLB entries on every
VM entry.

Submitted by:	Anish Gupta (akgupt3@gmail.com)

END
K 10
svn:author
V 2
bz
K 8
svn:date
V 27
2013-10-22T00:50:53.558303Z
K 7
svn:log
V 253
Make netback compile without INET support in the kernel.

This shuld have been a problem since r230587.  Not exactly sure why it
was not detected the last weeks with the tinderbox.  I would assume
r255744 is what started to cause it.

MFC after:	1 week

END
K 10
svn:author
V 4
neel
K 8
svn:date
V 27
2013-10-22T00:58:51.354661Z
K 7
svn:log
V 436
MFC r256645.

Add a new capability, VM_CAP_ENABLE_INVPCID, that can be enabled to expose
'invpcid' instruction to the guest. Currently bhyve will try to enable this
capability unconditionally if it is available.

Consolidate code in bhyve to set the capabilities so it is no longer
duplicated in BSP and AP bringup.

Add a sysctl 'vm.pmap.invpcid_works' to display whether the 'invpcid'
instruction is available.

Approved by:	re (hrs)

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-22T02:39:56.594615Z
K 7
svn:log
V 100
Return standards-compliant code from OF_nextprop() with FDT when no
properties remain on this node.

END
K 10
svn:author
V 5
gonzo
K 8
svn:date
V 27
2013-10-22T05:19:42.980681Z
K 7
svn:log
V 65
- Implement watchdog function and register it with watchdog list

END
K 10
svn:author
V 7
glebius
K 8
svn:date
V 27
2013-10-22T05:22:25.496504Z
K 7
svn:log
V 11
Fix build.

END
K 10
svn:author
V 5
gonzo
K 8
svn:date
V 27
2013-10-22T05:22:46.118144Z
K 7
svn:log
V 134
Make watchdog function conform watchdog(9):
Set error to 0 when watchdog is armed and disable it when timeout
is too large to be set.

END
K 10
svn:author
V 3
hrs
K 8
svn:date
V 27
2013-10-22T06:02:40.650542Z
K 7
svn:log
V 294
MFC 256716,256835:

- Fix jail_parallel_start="YES".
- Fix ip[46].addr when interface parameter is not defined.
- Fix a bug which prevented jails from starting when $jail_conf was used and
  no jail name was specified.
- Display error messages when start/stop fails.

Approved by:	re (glebius)

END
K 10
svn:author
V 3
des
K 8
svn:date
V 27
2013-10-22T06:53:01.699607Z
K 7
svn:log
V 63
Start rtsold if necessary.

Reviewed by:	hrs
MFC after:	3 days

END
K 10
svn:author
V 3
pho
K 8
svn:date
V 27
2013-10-22T07:16:20.582093Z
K 7
svn:log
V 70
Added a regression test.

Sponsored by:	EMC / Isilon storage division

END
K 10
svn:author
V 3
pho
K 8
svn:date
V 27
2013-10-22T07:17:22.815712Z
K 7
svn:log
V 67
Fix compiler warning.

Sponsored by:	EMC / Isilon storage division

END
K 10
svn:author
V 3
des
K 8
svn:date
V 27
2013-10-22T07:44:26.285011Z
K 7
svn:log
V 58
Add a reload command.

Reviewed by:	hrs
MFC after:	3 days

END
K 10
svn:author
V 5
andre
K 8
svn:date
V 27
2013-10-22T07:59:14.910088Z
K 7
svn:log
V 868
The TCP delayed ACK logic isn't aware of LRO passing up large aggregated
segments thinking it received only one segment. This causes it to enable
the delay the ACK for 100ms to wait for another segment which may never
come because all the data was received already.

Doing delayed ACK for LRO segments is bogus for two reasons: a) it pushes
us further away from acking every other packet; b) it introduces additional
delay in responding to the sender.  The latter is especially bad because it
is in the nature of LRO to aggregated all segments of a burst with no more
coming until an ACK is sent back.

Change the delayed ACK logic to detect LRO segments by being larger than
the MSS for this connection and issuing an immediate ACK for them to keep
the ACK clock ticking without interruption.

Reported by:	julian, cperciva
Tested by:	cperciva
Reviewed by:	(pending)

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-22T08:22:19.904182Z
K 7
svn:log
V 2003
Merge GEOM direct dispatch changes from the projects/camlock branch.

When safety requirements are met, it allows to avoid passing I/O requests
to GEOM g_up/g_down thread, executing them directly in the caller context.
That allows to avoid CPU bottlenecks in g_up/g_down threads, plus avoid
several context switches per I/O.

The defined now safety requirements are:
 - caller should not hold any locks and should be reenterable;
 - callee should not depend on GEOM dual-threaded concurency semantics;
 - on the way down, if request is unmapped while callee doesn't support it,
   the context should be sleepable;
 - kernel thread stack usage should be below 50%.

To keep compatibility with GEOM classes not meeting above requirements
new provider and consumer flags added:
 - G_CF_DIRECT_SEND -- consumer code meets caller requirements (request);
 - G_CF_DIRECT_RECEIVE -- consumer code meets callee requirements (done);
 - G_PF_DIRECT_SEND -- provider code meets caller requirements (done);
 - G_PF_DIRECT_RECEIVE -- provider code meets callee requirements (request).
Capable GEOM class can set them, allowing direct dispatch in cases where
it is safe.  If any of requirements are not met, request is queued to
g_up or g_down thread same as before.

Such GEOM classes were reviewed and updated to support direct dispatch:
CONCAT, DEV, DISK, GATE, MD, MIRROR, MULTIPATH, NOP, PART, RAID, STRIPE,
VFS, ZERO, ZFS::VDEV, ZFS::ZVOL, all classes based on g_slice KPI (LABEL,
MAP, FLASHMAP, etc).

To declare direct completion capability disk(9) KPI got new flag equivalent
to G_PF_DIRECT_SEND -- DISKFLAG_DIRECT_COMPLETION.  da(4) and ada(4) disk
drivers got it set now thanks to earlier CAM locking work.

This change more then twice increases peak block storage performance on
systems with manu CPUs, together with earlier CAM locking changes reaching
more then 1 million IOPS (512 byte raw reads from 16 SATA SSDs on 4 HBAs to
256 user-level threads).

Sponsored by:	iXsystems, Inc.
MFC after:	2 months

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-22T09:43:14.257400Z
K 7
svn:log
V 14
MFC @ r256880

END
K 10
svn:author
V 7
glebius
K 8
svn:date
V 27
2013-10-22T09:58:31.451889Z
K 7
svn:log
V 1347
Merge r255143 from head:
  Merge 1.12 of pf_lb.c from OpenBSD, with some changes. Original commit:

    date: 2010/02/04 14:10:12;  author: sthen;  state: Exp;  lines: +24 -19;
    pf_get_sport() picks a random port from the port range specified in a
    nat rule. It should check to see if it's in-use (i.e. matches an existing
    PF state), if it is, it cycles sequentially through other ports until
    it finds a free one. However the check was being done with the state
    keys the wrong way round so it was never actually finding the state
    to be in-use.

    - switch the keys to correct this, avoiding random state collisions
    with nat. Fixes PR 6300 and problems reported by robert@ and viq.

    - check pf_get_sport() return code in pf_test(); if port allocation
    fails the packet should be dropped rather than sent out untranslated.

    Help/ok claudio@.

  Some additional changes to 1.12:

  - We also need to bzero() the key to zero padding, otherwise key
    won't match.
  - Collapse two if blocks into one with ||, since both conditions
    lead to the same processing.
  - Only naddr changes in the cycle, so move initialization of other
    fields above the cycle.
  - s/u_intXX_t/uintXX_t/g

  PR:             kern/181690
  Submitted by:   Olivier Cochard-Labbé <olivier cochard.me>
  Sponsored by:   Nginx, Inc.

END
K 10
svn:author
V 7
dchagin
K 8
svn:date
V 27
2013-10-22T10:10:34.600700Z
K 7
svn:log
V 16
Sync with HEAD.

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-22T10:21:20.818632Z
K 7
svn:log
V 219
Remove Giant-locked drivers support (DISKFLAG_NEEDSGIANT flag) from disk(9).

Since at least FreeBSD 7 we had only four of them in the base tree, and
in head branch, thanks to jhb@, we have no any for more then a year.

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-22T10:40:26.199447Z
K 7
svn:log
V 106
Remove global device lock acquisition from dev_relthread(), replacing it
with atomics on per-device data.

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-22T11:56:46.394048Z
K 7
svn:log
V 50
Update ahci(4), respecting recent driver changes.

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-22T12:42:49.151453Z
K 7
svn:log
V 75
Fix memory and references leak due to unfreed path.

Coverity CID:	1109817

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-22T12:58:22.216560Z
K 7
svn:log
V 317
Unconditionally acquire periph reference on CCB allocation failure.

cam_periph_acquire() can return error if periph already invalidated, but
that may be unacceptable and cause deadlock if the invalidated periph can't
be destroyed without "executing" the scheduled request.

Coverity CID:	1109822
MFC after:	2 months

END
K 10
svn:author
V 3
smh
K 8
svn:date
V 27
2013-10-22T13:31:36.066626Z
K 7
svn:log
V 354
Use the vdev's ashift to calculate the supported min block size passed to
zio_compress_data(..) when compressing l2arc buffers.

This eliminates l2arc I/O errors, which resulted in very poor performance on
vdev's configured with block size greater than 512b due to compression
assuming a smaller min block size than the vdev supports.

MFC after:	2 days

END
K 10
svn:author
V 5
andre
K 8
svn:date
V 27
2013-10-22T13:45:03.462235Z
K 7
svn:log
V 781
Add gross proof of concept hack to bypass the entire lower stack
for outbound tcp sends for certain very high performance uses.

Bypassing means packets going directly to the wire and skip over
all firewalls or other processing done in ip_output() and
ether_output().

This very first version only does one interface and MAC address
lookup at the beginning of a connection.  It doesn't detect any
changes afterwards and will grind into a void if it has to.  On
certain errors it also leaks a bit of memory.

The point of this patch is to allow for performance analysis of
a) the normal vs. bypass path; b) to profile and optimize the
normal path to bring it as close as possible to the bypass path.

This patch is not intended to be merged to HEAD in its current
form (or at all).

END
K 10
svn:author
V 3
ray
K 8
svn:date
V 27
2013-10-22T13:47:25.851326Z
K 7
svn:log
V 197
o Add simplified structure describing framebuffer object.
o Add EVENTHANDLER wrapper to register/unregister framebuffer w/o depend on new
framebuffer methods.

Sponsored by:	The FreeBSD Foundation

END
K 10
svn:author
V 3
ray
K 8
svn:date
V 27
2013-10-22T13:50:54.430918Z
K 7
svn:log
V 95
Declare framebuffer's register/unregister EVENTHANDLERs.

Sponsored by:	The FreeBSD Foundation

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-22T13:52:20.059638Z
K 7
svn:log
V 75
Fix memory and references leak due to unfreed path.

Coverity CID:	1109815

END
K 10
svn:author
V 3
ray
K 8
svn:date
V 27
2013-10-22T13:53:58.010020Z
K 7
svn:log
V 114
Don't forget to clean terminal window memory. But colors makes more smiles.

Sponsored by:	The FreeBSD Foundation

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2013-10-22T13:56:30.630727Z
K 7
svn:log
V 75
Fix memory and references leak due to unfreed path.

Coverity CID:	1054773

END
K 10
svn:author
V 3
ray
K 8
svn:date
V 27
2013-10-22T13:56:42.882118Z
K 7
svn:log
V 109
Hardware specific drivers have bigger priority (over VGA for example).

Sponsored by:	The FreeBSD Foundation

END
K 10
svn:author
V 3
ray
K 8
svn:date
V 27
2013-10-22T14:00:46.565476Z
K 7
svn:log
V 221
Add new terminal method terminal_set_winsize_blank. Same as terminal_set_winsize,
but with optional blank. That will allow us to see early messages after attach
more specific driver.

Sponsored by:	The FreeBSD Foundation

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-22T14:07:57.460742Z
K 7
svn:log
V 441
Standards-conformance and code deduplication:
- Use bus reference phandles in place of FDT offsets as IRQ domain keys
- Unify the identical macio/fdt/mambo OpenPIC drivers into one
- Be more forgiving (following ePAPR) about what we need from the device
  tree to identify an OpenPIC
- Correctly map all IRQs into an interrupt domain
- Set IRQ_*_CONFORM for interrupts on an unknown PIC type instead of
  failing attachment for that device.

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-22T14:08:57.344976Z
K 7
svn:log
V 130
Allow lots of interrupts (useful on multi-domain platforms) and do not
set device_quiet() on all devices attached under nexus(4).

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-22T14:10:00.488480Z
K 7
svn:log
V 205
Set BUS_PROBE_NOWILDCARD on this attachment as a stopgap. Unconditionally
poking at registers in unknown devices is not the best probe mechanism.
This should be reverted and a better solution found later.

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-22T14:11:16.282263Z
K 7
svn:log
V 194
Catch up on 6 years of improvements in Open Firmware nexus devices by
importing the sparc64 one. At least 90% of this code is MI and will be
moved into /sys/dev/ofw at some point in the future.

END
K 10
svn:author
V 3
ray
K 8
svn:date
V 27
2013-10-22T14:32:42.890733Z
K 7
svn:log
V 75
Reverse priority. Bigger now better.

Sponsored by:	The FreeBSD Foundation

END
K 10
svn:author
V 3
ray
K 8
svn:date
V 27
2013-10-22T14:37:20.599999Z
K 7
svn:log
V 201
o Do not blank screen on window resize.
o Force resize for first window (usually console).
o Inform user on driver changes.
o Reverse priority. Bigger now better.

Sponsored by:	The FreeBSD Foundation

END
K 10
svn:author
V 3
ray
K 8
svn:date
V 27
2013-10-22T14:39:32.385426Z
K 7
svn:log
V 80
Add VT(9) driver for generic framebuffer.

Sponsored by:	The FreeBSD Foundation

END
K 10
svn:author
V 3
ray
K 8
svn:date
V 27
2013-10-22T14:45:35.238259Z
K 7
svn:log
V 70
Add generic/simple framebuffer.

Sponsored by:	The FreeBSD Foundation

END
K 10
svn:author
V 3
ray
K 8
svn:date
V 27
2013-10-22T14:47:09.388654Z
K 7
svn:log
V 92
Enable to build vt_fb driver and generic framebuffer.

Sponsored by:	The FreeBSD Foundation

END
K 10
svn:author
V 3
ray
K 8
svn:date
V 27
2013-10-22T14:50:28.753394Z
K 7
svn:log
V 88
Turn imx51_ipuv3 into simple frambuffer provider.

Sponsored by:	The FreeBSD Foundation

END
K 10
svn:author
V 7
glebius
K 8
svn:date
V 27
2013-10-22T14:55:04.807351Z
K 7
svn:log
V 16
More on memory.

END
K 10
svn:author
V 5
andre
K 8
svn:date
V 27
2013-10-22T15:06:24.830447Z
K 7
svn:log
V 382
Make struct ifnet comprehensible by grouping related fields
and variables together.

Quantify the structure size and how bloated it has become.

This is only a mechanical change.

Obvious first candidates for slimming by externalizing are:
 - char if_xname[IFNAMSIZ]
 - struct ifaltq if_snd
 - struct if_data if_data
 - void *if_afdata[AF_MAX]

Sponsored by:	The FreeBSD Foundation

END
K 10
svn:author
V 5
andre
K 8
svn:date
V 27
2013-10-22T15:09:31.931766Z
K 7
svn:log
V 166
Prepare to move all interface queue related structures, macros
and definitions from net/if_var.h to its own new net/ifq.h home.

Sponsored by:	The FreeBSD Foundation

END
K 10
svn:author
V 6
brooks
K 8
svn:date
V 27
2013-10-22T15:29:59.271195Z
K 7
svn:log
V 665
MFP4: 223121 (PIC portion), 225861, 227822, 229692 (PIC only), 229693,
230523, 1123614

Implement a driver for Robert Norton's PIC as an FDT interrupt
controller. Devices whose interrupt-parent property points to a beripic
device will have their interrupt allocation, activation , and setup
operations routed through the IC rather than down the traditional bus
hierarchy.

This driver largely abstracts the underlying CPU away allowing the
PIC to be implemented on CPU's other than BERI. Due to insufficient
abstractions a small amount of MIPS specific code is currently required
in fdt_mips.c and to implement counters.

MFC after:	3 days
Sponsored by:	DARPA/AFRL

END
K 10
svn:author
V 6
brooks
K 8
svn:date
V 27
2013-10-22T15:45:31.499985Z
K 7
svn:log
V 263
Sync BERI kernel configs with P4:

Switch the majority of device configuration to FDT from hints.

Add BERI_*_BASE configs to reduce duplication in the MDROOT and SDROOT
kernels.

Add NFS and GSSAPI support by default.

MFC after:	3 days
Sponsored by:	DARPA/AFRL

END
K 10
svn:author
V 5
andre
K 8
svn:date
V 27
2013-10-22T15:46:21.900499Z
K 7
svn:log
V 189
Remove all interface queue related structures, macros and
definitions from net/if_var.h and the other way around in
net/ifq.h.

Discussed with:	glebius
Sponsored by:	The FreeBSD Foundation

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-22T15:47:13.982277Z
K 7
svn:log
V 205
Ignore registers on devices where the reg property is malformed. Issue a
warning if this happens under bootverbose. This prevents some
strange-looking entries in dmesg for SMU devices on Apple G5 systems.

END
K 10
svn:author
V 6
brooks
K 8
svn:date
V 27
2013-10-22T15:53:29.316064Z
K 7
svn:log
V 402
Stop conflating WITHOUT_CLANG with WITHOUT_CLANG_IS_CC.  This allows
bootstrapping a copy of clang without building clang for the base system
which is useful for nanobsd and similar setups.  It's still probably
wrong to conflate what is installed as /usr/bin/cc with the selection
of a bootstrap compiler under WITH*_CLANG_IS_CC, but that's for another
day.

MFC after:	1 week
Sponsored by:	DARPA/AFRL

END
K 10
svn:author
V 8
cperciva
K 8
svn:date
V 27
2013-10-22T16:09:44.197457Z
K 7
svn:log
V 211
MFC r256775,r256776:
  Add support for "first boot" rc.d scripts.

  Document this new functionality in rc.conf(5) and rc(8).

  Bump __FreeBSD_version so that ports can make use of this.

Approved by:	re (gjb)

END
K 10
svn:author
V 8
cperciva
K 8
svn:date
V 27
2013-10-22T16:13:15.428395Z
K 7
svn:log
V 188
MFC r256775,r256776:
  Add support for "first boot" rc.d scripts.

  Document this new functionality in rc.conf(5) and rc(8).

  Bump __FreeBSD_version so that ports can make use of this.

END
K 10
svn:author
V 3
jhb
K 8
svn:date
V 27
2013-10-22T18:07:49.766729Z
K 7
svn:log
V 216
Please welcome Oded Shanoon (odeds@) as a new src committer.  Oded
leads the OFED-FreeBSD team at Mellanox and will be maintaining our
OFED stack.  Alfred Perlstein and I will serve as co-mentors.

Approved by:	core

END
K 10
svn:author
V 3
ian
K 8
svn:date
V 27
2013-10-22T18:14:06.990276Z
K 7
svn:log
V 153
Mask out non-address bits in the mac address register, for proper
detection of an all-zeroes address.  Also remove a misplaced return.

Reviewed by:	br@

END
K 10
svn:author
V 5
andre
K 8
svn:date
V 27
2013-10-22T18:24:34.725129Z
K 7
svn:log
V 885
The TCP delayed ACK logic isn't aware of LRO passing up large aggregated
segments thinking it received only one segment. This causes it to enable
the delay the ACK for 100ms to wait for another segment which may never
come because all the data was received already.

Doing delayed ACK for LRO segments is bogus for two reasons: a) it pushes
us further away from acking every other packet; b) it introduces additional
delay in responding to the sender.  The latter is especially bad because it
is in the nature of LRO to aggregated all segments of a burst with no more
coming until an ACK is sent back.

Change the delayed ACK logic to detect LRO segments by being larger than
the MSS for this connection and issuing an immediate ACK for them to keep
the ACK clock ticking without interruption.

Reported by:	julian, cperciva
Tested by:	cperciva
Reviewed by:	lstewart
MFC after:	3 days

END
K 10
svn:author
V 8
cperciva
K 8
svn:date
V 27
2013-10-22T18:36:39.260645Z
K 7
svn:log
V 569
Thou shalt not leak build host state into the system being compiled.

The VERSION variable is encoded into the SUNW_ctf sections of the kernel
and every kernel module when dtrace is enabled; starting with 9.2-RELEASE
(when dtrace was turned on in GENERIC) this means that different host kernels
will result in very different kernel binaries being generated.  This tripped
up freebsd-update builds after the build boxes were updated from 9.x to 10.x.

MFC after:	3 days (stable/9)
X-MFC after:	0 days (stable/10)
Security:	Rendered two members of so@ temporarily insane

END
K 10
svn:author
V 5
andre
K 8
svn:date
V 27
2013-10-22T18:39:43.853701Z
K 7
svn:log
V 13
IFC @256921.

END
K 10
svn:author
V 5
gavin
K 8
svn:date
V 27
2013-10-22T18:55:52.258942Z
K 7
svn:log
V 174
Rename the "alfred" command to "auto", and document it.  Retain support
for "portsnap alfred" for now.

Discussed:	many times, most recently on svn-src-all
MFC after:	1 week

END
K 10
svn:author
V 8
ambrisko
K 8
svn:date
V 27
2013-10-22T19:49:19.116156Z
K 7
svn:log
V 121
MFC r255869:

Add PCI device ID for MegaRAID Invader cards.  This was the only
change needed to make it work on my card.

END
K 10
svn:author
V 4
tijl
K 8
svn:date
V 27
2013-10-22T19:53:52.884754Z
K 7
svn:log
V 194
Add a dummy statement to the beginning of the pthread_cleanup_pop() macro
to allow a call of the macro to be labelled as in:

label:
  pthread_cleanup_pop();

Reviewed by:	imp
MFC after:	3 days

END
K 10
svn:author
V 6
grehan
K 8
svn:date
V 27
2013-10-22T19:55:04.382762Z
K 7
svn:log
V 346
Fix AHCI ATAPI emulation when backed with /dev/cd0

- remove assumption that the backing file/device had
  512-byte sectors
- fix incorrect iovec size variable that would result
  in a buffer overrun when an o/s issued an i/o request
  with more s/g elements than the blockif api

Reviewed by:	Zhixiang Yu (zxyu.core@gmail.com)
MFC after:	3 days

END
K 10
svn:author
V 3
jhb
K 8
svn:date
V 27
2013-10-22T20:10:38.222047Z
K 7
svn:log
V 66
Add a manpage for the getenv() family of functions in the kernel.

END
K 10
svn:author
V 3
ray
K 8
svn:date
V 27
2013-10-22T20:27:30.609055Z
K 7
svn:log
V 153
o Update drivers according to fb_info structure changes.
o Switch drm_fb_helper to be simple framebuffer provider.

Sponsored by:	The FreeBSD Foundation

END
K 10
svn:author
V 3
ray
K 8
svn:date
V 27
2013-10-22T20:30:54.856244Z
K 7
svn:log
V 50
Enable cache early to speed up memory operations.

END
K 10
svn:author
V 3
ray
K 8
svn:date
V 27
2013-10-22T20:31:59.485889Z
K 7
svn:log
V 30
Remove syscons related flags.

END
K 10
svn:author
V 6
brooks
K 8
svn:date
V 27
2013-10-22T20:50:41.433539Z
K 7
svn:log
V 115
Enable ATSE_CFI_HACK in BERI configs, stable MAC addresses are useful.

MFC after:	3 days
Sponsored by:	DARPA/AFRL

END
K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2013-10-22T20:57:24.761999Z
K 7
svn:log
V 294
Add a new function (OF_getencprop()) that undoes the transformation applied
by encode-int. Specifically, it takes a set of 32-bit cell values and
changes them to host byte order. Most non-string instances of OF_getprop()
should be using this function, which is a no-op on big-endian platforms.

END
K 10
svn:author
V 7
glebius
K 8
svn:date
V 27
2013-10-22T20:58:23.667880Z
K 7
svn:log
V 16
More on memory.

END
K 10
svn:author
V 6
brooks
K 8
svn:date
V 27
2013-10-22T21:06:27.550535Z
K 7
svn:log
V 1233
MFP4:
Change 221534 by rwatson@rwatson_zenith_cl_cam_ac_uk on 2013/01/27 16:05:30

        FreeBSD/mips stores page-table entries in a near-identical format
        to MIPS TLB entries -- only it overrides certain "reserved" bits
        in the MIPS-defined EntryLo register to hold software-defined bits
        (swbits) to avoid significantly increasing the page table memory
        footprint.  On n32 and n64, these bits were (a) colliding with
        MIPS64r2 physical memory extensions and (b) being improperly
        cleared.

        Attempt to fix both of these problems by pushing swbits further
        along 64-bit EntryLo registers into the reserved space, and
        improving consistency between C-based and assembly-based clearing
        of swbits -- in particular, to use the same definition.  This
        should stop swbits from leaking into TLB entries -- while ignored
        by most current MIPS hardware, this would cause a problem with
        (much) larger physical memory sizes, and also leads to confusing
        hardware-level tracing as physical addresses contain unexpected
        (and inconsistent) higher bits.

        Discussed with: imp, jmallett

MFC after:	3 days
Sponsored by:	DARPA/AFRL

END