adrian 29059de593 Make sure the final descriptor in an aggregate has rate control information. ... This was broken by me when merging the 802.11n aggregate descriptor chain setup with the default descriptor chain setup, in preparation for supporting AR9380 NICs. The corner case here is quite specific - if you queue an aggregate frame with >1 frames in it, and the last subframe has only one descriptor making it up, then that descriptor won't have the rate control information copied into it. Look at what happens inside ar5416FillTxDesc() if both firstSeg and lastSeg are set to 1. Then when ar5416ProcTxDesc() goes to fill out ts_rate based on the transmit index, it looks at the rate control fields in that descriptor and dutifully sets it to be 0. It doesn't happen for non-aggregate frames - if they have one descriptor, the first descriptor already has rate control info. I removed the call to ath_hal_setuplasttxdesc() when I migrated the code to use the "new" style aggregate chain routines from the HAL. But I missed this particular corner case. This is a bit inefficient with MIPS boards as it involves a few redundant writes into non-cachable memory. I'll chase that up when it matters. Tested: * AR9280 STA mode, TCP iperf traffic * Rui Paulo <rpaulo@> first reported this and has verified it on his AR9160 based AP. PR: kern/173636		2012-11-15 03:00:49 +00:00
..
aac	Give panic format string to pacify clang warning.	2012-11-09 13:58:52 +00:00
acpi_support	add support for newer Lenovo ThinkPads to acpi_ibm	2012-10-29 10:22:00 +00:00
acpica	Update MAX_TASKS to scale a bit based on MAXCPU	2012-11-12 18:09:25 +00:00
adb	Use uiomove return value instead of returning 0.	2012-11-13 06:03:43 +00:00
adlink
advansys
ae
age	TSO engine of L1 requires a separate DMA descriptor for TCP	2012-10-30 07:55:03 +00:00
agp
aha
ahb
ahci	Announce diagnostic page 7 (Element Descriptor) support.	2012-11-07 22:53:46 +00:00
aic
aic7xxx	Fix a bug in aicasm_gram.y, noted by a newer clang 3.2 snapshot: it	2012-11-14 18:54:08 +00:00
alc
ale	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
altera
amdsbwd
amdtemp
amr
an
arcmsr
asmc	Give device_printf format string to pacify clang warning.	2012-11-09 14:58:26 +00:00
asr
ata	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
ath	Make sure the final descriptor in an aggregate has rate control information.	2012-11-15 03:00:49 +00:00
atkbdc
auxio
bce	- Remove BCE_JUMBO_HDRSPLIT kernel option which was forgotten in r218423.	2012-10-30 13:22:39 +00:00
bfe
bge	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
bktr	Fix typo; s/ouput/output	2012-11-07 07:00:59 +00:00
bm
buslogic
bwi
bwn
bxe
cardbus
cas	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
ce
cesa
cfe
cfi
ciss	Increase the driver-side limit on the number of logical volumes that	2012-10-25 18:46:02 +00:00
cm
cmx
coretemp
cp
cpuctl	cpuctl_do_cpuid: explicitly use ecx=0 for cpuid call	2012-11-04 13:46:28 +00:00
cpufreq
cs
ct	Use ANSI prototype to fix build with clang.	2012-11-10 14:58:06 +00:00
ctau
cx
cxgb
cxgbe	Make sure the inp hasn't been dropped before trying to access its socket	2012-11-06 20:22:39 +00:00
cy
dc	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
dcons
de
digi
dpms
dpt
drm	Cast VIA_READ to (void) where the result is unused.	2012-11-09 14:46:23 +00:00
drm2	- If DRM_DEBUG_DEFAULT_ON is defined, then initialize drm_debug_flagi to	2012-11-02 05:26:33 +00:00
e1000
ed
eisa
en
ep
esp
et
etherswitch	Disable automatic attachment of arswitch. It can't be auto-detected (like PHYs	2012-11-07 22:43:09 +00:00
ex
exca
fatm
fb
fdc
fdt
fe
filemon	A little bit easier to read.	2012-10-26 20:24:13 +00:00
firewire
flash	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
fxp	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
gem	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
glxiic
glxsb
gpio
gxemul
hatm
hifn
hme
hpt27xx
hptiop
hptmv
hptrr
hwpmc	Fixup r240246: hwpmc needs to retain the pinning until ASTs are not	2012-10-30 15:10:50 +00:00
ic
ichsmb
ichwd
ida
ie
ieee488
if_ndis
iicbus	Check the return value of uiomove(9).	2012-11-13 05:58:52 +00:00
iir
io
ipmi
ips
ipw
isci
iscsi/initiator
isf
isp	Don't allow for more than one segment for the control space since	2012-11-02 14:38:57 +00:00
ispfw
iwi
iwn
ixgb
ixgbe	A few important fixes:	2012-10-31 23:50:36 +00:00
jme
joy
kbd
kbdmux
ksyms
le
led
lge	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
lindev
lmc
malo
mc146818
mca
mcd
md	Zero the newly allocated md(4) swap-backed page to prevent random	2012-11-08 03:17:41 +00:00
mem
mfi	Add support for SCSI pass through devices to be attached and	2012-11-08 00:32:36 +00:00
mge
mii	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
mk48txx
mlx
mly
mmc	Add BCM2835 SDHCI driver and enable it in Raspberry Pi config	2012-10-29 17:23:45 +00:00
mn	Remove unused static mn_timeout.	2012-11-09 14:14:51 +00:00
mps
mpt
mse
msk
mvs
mwl
mxge
my
nand
ncv
netmap
nfe
nge	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
nmdm
nsp
null	Provide a device name in the sysctl tree for programs to query the	2012-11-01 17:01:05 +00:00
nvd
nve	Fix all the inconsistent nve_os* function declarations and definitions	2012-11-10 16:02:12 +00:00
nvme	Use callout_reset_curcpu to allow the callout to be handled by the	2012-10-31 23:44:19 +00:00
nvram
nvram2env
nxge	Fix typo; s/ouput/output	2012-11-07 07:00:59 +00:00
oce
ofw
patm
pbio
pccard
pccbb
pcf
pci	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
pcn
pdq
powermac_nvram
ppbus
ppc
pst
pty
puc	Add support for Advantech PCI-1602 RS-485/RS-422 serial card	2012-11-09 01:51:06 +00:00
qlxgb
quicc
ral
random
rc
re	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
rndtest
rp
rt
safe
sbni
scc
scd
sdhci	Add new quirks:	2012-10-29 17:21:58 +00:00
sec
sf
sfxge
sge
si
siba
siis
sio	Add support for CIR1000 - Cirrus Logic V34 to the sio driver	2012-11-13 20:38:55 +00:00
sis	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
sk
smbus
smc
sn
snc	Use ANSI prototype to fix build with clang.	2012-11-10 14:58:06 +00:00
snp
sound	Fix typo; s/ouput/output	2012-11-07 07:00:59 +00:00
speaker
spibus
ste	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
stg
stge	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
streams
sym
syscons	Add tty_set_winsize().	2012-11-03 22:21:37 +00:00
tdfx
terasic
ti	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
tl	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
tpm
trm
tsec
twa	Give tw_osl_dbg_printf format string to pacify clang warning.	2012-11-09 15:29:52 +00:00
twe
tws
tx
txp
uart	Add support for CIR1000 - Cirrus Logic V34 to the uart driver	2012-11-13 21:04:35 +00:00
ubsec
usb	Add new USB IDs.	2012-11-12 07:25:51 +00:00
utopia
vge
viawd
virtio
vkbd
vr	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
vte
vx
vxge
watchdog
wb	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00
wbwd
wds
wi
wl
wpi
wtap
xe
xen	Redo r242889, now using the method from projects/amd64_xen_pv r240747.	2012-11-12 22:28:32 +00:00
xl	Remove duplicate const specifiers in many drivers (I hope I got all of	2012-11-05 19:16:27 +00:00