mav 33230ae6d4 MFC r277169: Reimplement TRIM throttling added in r248577. ... Previous throttling implementation approached problem from the wrong side. It significantly limited useful delaying of TRIM requests and aggregation potential, while not so much controlled TRIM burstiness under heavy load. With this change random 4K write benchmarks (probably the worst case for TRIM) show me IOPS increase by 20%, average latency reduction by 30%, peak TRIM bursts reduction by 3 times and same peak TRIM map size (memory usage). Also the new logic does not force map size down so heavily, really allowing to keep deleted data for 32 TXG or 30 seconds under moderate load. It was practically impossible with old throttling logic, which pushed map down to only 64 segments.		2015-01-28 02:55:20 +00:00
..
amd64	MFC r277055:	2015-01-19 11:07:29 +00:00
arm	MFC:	2015-01-13 07:45:16 +00:00
boot	MFC r264400,r265836:	2014-12-31 23:25:37 +00:00
bsm
cam	MFC r277647: Fix wrong LUN reference in XCOPY block-to-block operation.	2015-01-27 19:41:24 +00:00
cddl	MFC r277169: Reimplement TRIM throttling added in r248577.	2015-01-28 02:55:20 +00:00
compat	MFC r277322:	2015-01-25 13:15:12 +00:00
conf	Merge reaper facility.	2015-01-05 03:27:09 +00:00
contrib	MFC r275199.	2014-12-10 20:54:23 +00:00
crypto	MFC r264400,r265836:	2014-12-31 23:25:37 +00:00
ddb	MFC r272958, r273006:	2014-10-18 19:22:59 +00:00
dev	MFC: r277594	2015-01-26 19:27:30 +00:00
fs	MFC r277390:	2015-01-26 11:37:16 +00:00
gdb
geom	MFC r275503:	2014-12-19 01:13:41 +00:00
gnu	MFC r271143, r271151: Reimport dts from vendor branch after proper trim.	2014-10-26 23:45:33 +00:00
i386	MFC r277047:	2015-01-19 10:52:55 +00:00
ia64	Merge the fueword(9) and casueword(9). In particular,	2014-11-18 12:53:32 +00:00
isa
kern	MFC r277322:	2015-01-25 13:15:12 +00:00
kgssapi
libkern	Clean up our ARM assembly:	2014-12-14 16:28:53 +00:00
mips	MFC r274816:	2015-01-22 21:17:58 +00:00
modules	MFC r273683	2014-12-30 22:22:46 +00:00
net	MFC r277295:	2015-01-25 16:35:03 +00:00
net80211	MFC r263710, r273377, r273378, r273423 and r273455:	2014-10-27 14:38:00 +00:00
netatalk	Merge r262763, r262767, r262771, r262806 from head:	2014-03-21 15:15:30 +00:00
netgraph	MFC r263710, r273377, r273378, r273423 and r273455:	2014-10-27 14:38:00 +00:00
netinet	Fix SCTP SCTP_SS_VALUE kernel memory corruption and disclosure vulnerability	2015-01-27 19:36:08 +00:00
netinet6	MFC r272886:	2015-01-27 06:19:30 +00:00
netipsec	MFC r275393:	2014-12-16 11:45:33 +00:00
netipx	MFC r263710, r273377, r273378, r273423 and r273455:	2014-10-27 14:38:00 +00:00
netnatm
netpfil	Merge r274709 by eri@: deal with IPv6 same way as we IPv4 and calculate	2015-01-23 18:15:15 +00:00
netsmb	MFC r264494:	2014-05-02 21:30:59 +00:00
nfs	MFC r267479:	2014-06-22 21:26:46 +00:00
nfsclient	MFC r275897:	2015-01-01 10:44:20 +00:00
nfsserver	MFC r275897:	2015-01-01 10:44:20 +00:00
nlm	MFC r262991:	2014-03-26 23:57:09 +00:00
ofed	MFC r276749:	2015-01-13 17:07:30 +00:00
opencrypto	MFC r258622: dtrace sdt: remove the ugly sname parameter of SDT_PROBE_DEFINE	2014-01-17 10:58:59 +00:00
pc98	MFC: r272492	2014-12-23 12:13:31 +00:00
pci	MFC r263957:	2014-04-14 04:51:59 +00:00
powerpc	MFC r275616:	2014-12-15 10:46:07 +00:00
rpc	MFC r275745:	2014-12-27 00:55:14 +00:00
security	MFC r259885:	2014-08-16 13:11:59 +00:00
sparc64	Merge the fueword(9) and casueword(9). In particular,	2014-11-18 12:53:32 +00:00
sys	Bump __FreeBSD_version after r277789: Changes to the UDP tunneling callback	2015-01-27 06:24:33 +00:00
teken	MFC 259016,259019,259049,259071,259102,259110,259129,259130,259178,259179,	2014-03-06 18:30:56 +00:00
tools	MFC r262626, r262626, r264979, r265154, r265162, r265163, r267319, r270863,	2014-10-27 00:47:55 +00:00
ufs	MFC r276007:	2015-01-04 00:46:06 +00:00
vm	MFC r277055:	2015-01-19 11:07:29 +00:00
x86	MFC 272666: Fix build for i386 kernels with out 'I686_CPU'.	2015-01-21 17:59:32 +00:00
xdr
xen
Makefile