d/freebsd-dev
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
232,777 Commits 72 Branches 135 Tags 3.2 GiB
ac4a7f30d2
Commit Graph

16065 Commits

Author SHA1 Message Date
Mark Johnston
13679ebac9 Don't pass a section cookie to CK for non-preemptible epoch sections. 
They're only useful when multiple threads may share an epoch record,
and that can't happen with non-preemptible sections.

Reviewed by:	mmacy
Differential Revision:	https://reviews.freebsd.org/D15507
 2018-05-21 16:03:51 +00:00
Matt Macy
9725c9cef7 AF_UNIX gc unused label 
...sigh
 2018-05-20 21:37:34 +00:00
Matt Macy
cb8f450b94 AF_UNIX: Don't unlock unp/unp2 if they're not locked 
Reported by:	mjg
 2018-05-20 21:20:26 +00:00
Matt Macy
e10ef65d23 AF_UNIX: fix LOR introduced by the locking rewrite 2018-05-20 05:50:53 +00:00
Matt Macy
7118990962 Add additional preinitialized cap_rights 2018-05-20 05:13:12 +00:00
Mateusz Guzik
2186ee6e72 vfs: simplify vop_stdlock/unlock 
The interlock pointer is non-NULL by definition and the compiler see through
that and eliminates the NULL checks. Just remove them from the code as they
play no role.

No difference in generated assembly.
 2018-05-20 04:45:05 +00:00
Matt Macy
d95253403f AF_UNIX: make unpcb lock name line up with what's in witness 2018-05-20 04:32:48 +00:00
Warner Losh
f344fb0b4b Restore the all rights reserved language. Put it on each of the prior 
two copyrights. The line originated with the Berkeely Regents, who
we have not approached about removing it (it's honestly too trivial
to be worth that fight). Restore it to rwatson's line as well. He
can decide if he wants it or not on his own. Matt clearly doesn't
want it, per project preference and his own statements on IRC.

Noticed by: rgrimes@
 2018-05-19 17:29:57 +00:00
Ed Maste
1b30e10e48 Remove duplicate cap_no_rights from r333874 
Archs using in-tree gcc were broken with `warning: redundant
redeclaration of 'cap_no_rights' [-Wredundant-decls]`.

Sponsored by:	The FreeBSD Foundation
 2018-05-19 11:37:02 +00:00
Matt Macy
f6a1a10613 Unbreak BeagleBone Black boot by collapsing 29 SYSINITs in to 1 
Reported by:	ilya at bakulin.de
 2018-05-19 07:31:35 +00:00
Matt Macy
fc2e87be2b intr unbreak KTR/LINT build 2018-05-19 07:04:43 +00:00
Matt Macy
4b06dee1e5 AF_UNIX: switch to annotations to avoid warnings 2018-05-19 05:37:58 +00:00
Matt Macy
acbde29858 capsicum: propagate const correctness 2018-05-19 05:14:05 +00:00
Matt Macy
ba3f7276c0 intr: eliminate / annotate unused stack locals 2018-05-19 05:12:18 +00:00
Matt Macy
7fd6841438 sendfile: annotate unused value and ensure that npages is actually initialized 2018-05-19 05:10:51 +00:00
Matt Macy
e1a92f058f umtx: don't call umtxq_getchain unless the value is needed 2018-05-19 05:09:10 +00:00
Matt Macy
a6c7423a92 cpuset: revert and annotate instead 2018-05-19 05:07:31 +00:00
Matt Macy
6fa5abfdda conf: revert last change and annotate unused var instead 2018-05-19 05:07:03 +00:00
Matt Macy
1c0336c1c1 kevent: annotate unused stack local 2018-05-19 05:06:18 +00:00
Matt Macy
788390df0a lockf: annotate LOCKF_DEBUG only var 2018-05-19 05:04:38 +00:00
Matt Macy
d1230b1159 capsicum: annotate variable only used by debug 2018-05-19 05:02:40 +00:00
Matt Macy
3adccf38e3 turnstile / sleepqueue: annotate variables only used by debug builds 2018-05-19 05:00:16 +00:00
Matt Macy
84482abd21 vfs: annotate variables only used by debug builds as __unused 2018-05-19 04:59:39 +00:00
Matt Macy
a2bb4e080e tty: use __unused annotation instead to silence warnings 2018-05-19 04:48:26 +00:00
Matt Macy
5072a5f465 malloc: avoid possibly returning stack garbage if MALLOC_DEBUG is defined 2018-05-19 04:43:49 +00:00
Matt Macy
39eef2f45a cpuset_thread0: avoid unused assignment on non debug build 2018-05-19 04:14:00 +00:00
Matt Macy
926cfdb8da make_dev: avoid unused assignments on non debug builds 2018-05-19 04:13:20 +00:00
Matt Macy
4949ad7264 mqueue: avoid unused variables 2018-05-19 04:10:53 +00:00
Matt Macy
cd6ba3f086 physio: avoid uninitialized variables 2018-05-19 04:09:58 +00:00
Matt Macy
e9b1074bc7 cache_lookup remove unused variable and initialize used 2018-05-19 04:08:11 +00:00
Matt Macy
ec8d23352b filt_timerdetach: only assign to old if we're going to check it in 
a KASSERT
 2018-05-19 04:07:00 +00:00
Matt Macy
5cc2d25a2b getnextevent: put variable only used by KTR under ifdef KTR 2018-05-19 04:05:36 +00:00
Matt Macy
bfd0eacb02 simplify control flow so that gcc knows we never pass save to curthread_pflags_restore 
without initializing
 2018-05-19 04:04:44 +00:00
Matt Macy
3ef78c9c96 tty: conditionally assign to ret value only used by MPASS statement 2018-05-19 04:02:29 +00:00
Matt Macy
02fe8a2409 remove unused locked variable in lockmgr_unlock_fast_path 2018-05-19 03:58:40 +00:00
Matt Macy
ddd4d15ecd signotify: don't create a stack local that isn't used on non-debug builds 2018-05-19 03:57:41 +00:00
Matt Macy
46117e1f0c sysv_msg initialize saved_msgsz 2018-05-19 03:56:39 +00:00
Matt Macy
11d4f748d7 remove unused variable 2018-05-19 03:55:42 +00:00
Matt Macy
1dce110f63 fix uninitialized variable warning in reader locks 2018-05-19 03:52:55 +00:00
Matt Macy
b203713694 fix uninitialized variable warning 2018-05-19 03:49:36 +00:00
Matt Macy
ac8b2d5cb1 sys_process.c fix set but not used warning 2018-05-19 03:48:35 +00:00
Matt Macy
e339e43685 subr_epoch.c fix unused variable warnings 2018-05-19 03:47:37 +00:00
Matt Macy
ae6be8e6f7 pidctrl Actually use the variables that we assign to as seatbelts to prevent divide 
by zero

Reviewed by:	jeffr
 2018-05-19 02:17:18 +00:00
Matt Macy
c0874c3468 fix gcc8 unused variable and set but not used variable in unix sockets 
add copyright from lock rewrite while here
 2018-05-19 02:15:40 +00:00
Mateusz Guzik
10391db530 lockmgr: avoid atomic on unlock in the slow path 
The code is pretty much guaranteed not to be able to unlock.

This is a minor nit. The code still performs way too many reads.
The altered exclusive-locked condition is supposed to be always
true as well, to be cleaned up at a later date.
 2018-05-18 22:57:52 +00:00
Matt Macy
d7c5a620e2 ifnet: Replace if_addr_lock rwlock with epoch + mutex 
Run on LLNW canaries and tested by pho@

gallatin:
Using a 14-core, 28-HTT single socket E5-2697 v3 with a 40GbE MLX5
based ConnectX 4-LX NIC, I see an almost 12% improvement in received
packet rate, and a larger improvement in bytes delivered all the way
to userspace.

When the host receiving 64 streams of netperf -H $DUT -t UDP_STREAM -- -m 1,
I see, using nstat -I mce0 1 before the patch:

InMpps OMpps  InGbs  OGbs err TCP Est %CPU syscalls csw     irq GBfree
4.98   0.00   4.42   0.00 4235592     33   83.80 4720653 2149771   1235 247.32
4.73   0.00   4.20   0.00 4025260     33   82.99 4724900 2139833   1204 247.32
4.72   0.00   4.20   0.00 4035252     33   82.14 4719162 2132023   1264 247.32
4.71   0.00   4.21   0.00 4073206     33   83.68 4744973 2123317   1347 247.32
4.72   0.00   4.21   0.00 4061118     33   80.82 4713615 2188091   1490 247.32
4.72   0.00   4.21   0.00 4051675     33   85.29 4727399 2109011   1205 247.32
4.73   0.00   4.21   0.00 4039056     33   84.65 4724735 2102603   1053 247.32

After the patch

InMpps OMpps  InGbs  OGbs err TCP Est %CPU syscalls csw     irq GBfree
5.43   0.00   4.20   0.00 3313143     33   84.96 5434214 1900162   2656 245.51
5.43   0.00   4.20   0.00 3308527     33   85.24 5439695 1809382   2521 245.51
5.42   0.00   4.19   0.00 3316778     33   87.54 5416028 1805835   2256 245.51
5.42   0.00   4.19   0.00 3317673     33   90.44 5426044 1763056   2332 245.51
5.42   0.00   4.19   0.00 3314839     33   88.11 5435732 1792218   2499 245.52
5.44   0.00   4.19   0.00 3293228     33   91.84 5426301 1668597   2121 245.52

Similarly, netperf reports 230Mb/s before the patch, and 270Mb/s after the patch

Reviewed by:	gallatin
Sponsored by:	Limelight Networks
Differential Revision:	https://reviews.freebsd.org/D15366
 2018-05-18 20:13:34 +00:00
Matt Macy
20ba6811e6 epoch(9): assert that epoch is allocated post-configure 2018-05-18 18:27:17 +00:00
Ed Maste
891cf3ed44 Use NULL for SYSINIT's last arg, which is a pointer type 
Sponsored by:	The FreeBSD Foundation
 2018-05-18 17:58:09 +00:00
Matt Macy
70398c2f86 epoch(9): Make epochs non-preemptible by default 
There are risks associated with waiting on a preemptible epoch section.
Change the name to make them not be the default and document the issue
under CAVEATS.

Reported by:	markj
 2018-05-18 17:29:43 +00:00
Matt Macy
60b7b90d65 epoch: actually allocate the counters we've assigned sysctls too 
Approved by:	sbruno
 2018-05-18 02:57:39 +00:00