Luigi Rizzo 6dba29a285 Two performance-related fixes: ... 1. as reported by Alexander Fiveg, the allocator was reporting half of the allocated memory. Fix this by exiting from the loop earlier (not too critical because this code is going away soon). 2. following a discussion on freebsd-current http://lists.freebsd.org/pipermail/freebsd-current/2012-January/031144.html turns out that (re)loading the dmamap was expensive and not optimized. This operation is in the critical path when doing zero-copy forwarding between interfaces. At least on netmap and i386/amd64, the bus_dmamap_load can be completely bypassed if the map is NULL, so we do it. The latter change gives an almost 3x improvement in forwarding performance, from the previous 9.5Mpps at 2.9GHz to the current line rate (14.2Mpps) at 1.733GHz. (this is for 64+4 byte packets, in other configurations the PCIe bus is a bottleneck).		2012-01-13 10:21:15 +00:00
..
aac
acpi_support
acpica	Further relax the strictness of enforcing allocations to only come from	2011-12-29 16:23:14 +00:00
adb
adlink
advansys
ae
age
agp
aha
ahb
ahci	Add 0x2826 device ID for C600 (Patsburg) SATA controller in RAID mode.	2012-01-06 00:22:55 +00:00
aic
aic7xxx
alc
ale
amdsbwd
amdtemp
amr
an	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
arcmsr
asmc
asr
ata	Add the CAM Target Layer (CTL).	2012-01-12 00:34:33 +00:00
ath	Re-enable the PHY radar error frames if sc_dodfs is set.	2012-01-11 00:18:33 +00:00
atkbdc
auxio
bce
bfe	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
bge
bktr
bm	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
buslogic
bwi
bwn
bxe	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
cardbus
cas
ce	Fix clang warnings in sys/dev/ce/if_ce.c and sys/dev/cp/if_cp.c,	2011-12-29 17:46:24 +00:00
cesa
cfe
cfi
ciss	Add the CAM Target Layer (CTL).	2012-01-12 00:34:33 +00:00
cm
cmx
coretemp
cp	Fix clang warnings in sys/dev/ce/if_ce.c and sys/dev/cp/if_cp.c,	2011-12-29 17:46:24 +00:00
cpuctl
cpufreq
cs
ct
ctau
cx
cxgb
cxgbe
cy
dc
dcons
de
digi
dpms
dpt
drm	In sys/dev/drm/radeon_state.c, use the correct printf length modifiers	2011-12-30 02:07:50 +00:00
e1000	clear the pointer after freeing the mbuf. Without that, we	2012-01-12 17:30:44 +00:00
ed
eisa
en
ep	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
esp
et	style. No functional changes.	2012-01-10 20:52:02 +00:00
ex	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
exca
fatm
fb
fdc
fdt	DEBUG is a kernel option. Don't define it here, as it breaks LINT.	2011-12-30 04:00:31 +00:00
fe
firewire	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
flash	Add a couple more SPI flash device IDs found in commercial	2012-01-04 04:17:03 +00:00
fxp
gem
glxiic
glxsb
gpio
hatm
hifn
hme
hpt27xx	Import the first release of HighPoint RocketRAID 27xx SAS 6Gb/s HBA card	2011-12-28 23:26:58 +00:00
hptiop
hptmv
hptrr
hwpmc	Update PMC events from October 2011 Intel documentation.	2012-01-04 07:58:36 +00:00
ic
ichsmb
ichwd	Remove use of explicit bus space tags and handles and use methods that	2012-01-05 16:27:32 +00:00
ida
ie	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
ieee488
if_ndis	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
iicbus
iir
io
ipmi
ips
ipw
iscsi/initiator
isp	Fix target mode compilation issues that arose after a change	2011-12-27 14:59:24 +00:00
ispfw
iwi
iwn	Don't rely on MCS7 being at index 7 while determining the amount	2012-01-03 09:42:31 +00:00
ixgb	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
ixgbe	small code cleanup in preparation for future modifications in	2012-01-10 19:57:23 +00:00
jme
joy
kbd
kbdmux
ksyms
le
led
lge	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
lindev
lmc
malo
mc146818
mca
mcd
md
mem
mfi
mge
mii
mk48txx
mlx
mly
mmc
mn
mps	Set maximum I/O size for mps(4) to MAXPHYS. Looking into the code, I see	2011-12-28 22:49:28 +00:00
mpt
mse
msk	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
mvs
mwl
mxge	Use strchr() and strrchr().	2012-01-02 12:12:10 +00:00
my	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
ncv
netmap	Two performance-related fixes:	2012-01-13 10:21:15 +00:00
nfe	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
nge
nmdm
nsp
null	Fix for PR 138526.	2012-01-11 15:00:16 +00:00
nve	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
nvram
nvram2env
nxge
ofw
patm
pbio
pccard
pccbb
pcf
pci
pcn
pdq
powermac_nvram
ppbus
ppc
pst	Pass the arguments to mtx_init() in the correct order. There should be	2012-01-11 21:38:05 +00:00
pty
puc
qlxgb	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
quicc
ral
random	Enable hardware RNG for VIA Nano processors.	2012-01-09 23:20:30 +00:00
rc
re
rndtest
rp
rt	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
safe
sbni
scc
scd
sdhci
sec
sf
sfxge	Update recently added drivers to use the if_*addr_r*lock() wrapper	2012-01-05 18:32:37 +00:00
sge
si
siba
siis
sio
sis	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
sk	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
smbus
smc
sn	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
snc	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
snp
sound	Replace GPL'd headers in the emu10kx snd driver code.	2012-01-11 21:17:14 +00:00
speaker
spibus
ste
stg
stge	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
streams
sym
syscons
tdfx
ti
tl	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
tpm
trm
tsec	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
twa
twe
tws	Don't forget to release queue lock when allocation of memory failed.	2012-01-03 20:17:35 +00:00
tx
txp
uart	Add PCI Id for the AMT SOL UART on 5 series Intel chipsets.	2012-01-11 17:46:08 +00:00
ubsec
usb	Correct use of USB 3.0 POWER bit in the port status register,	2012-01-13 07:28:34 +00:00
utopia
vge
viawd
virtio
vkbd
vr
vte
vx	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
vxge	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
watchdog
wb	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
wds
wi
wl
wpi
wtap	Introduce wtap, the beginnings of a net80211 wlan simulator.	2012-01-11 17:41:14 +00:00
xe
xen	ether_ifattach() sets if_mtu to ETHERMTU, don't bother set it again	2012-01-07 09:41:57 +00:00
xl