Commit graph

299,176 commits

Author SHA1 Message Date
Bhanu Prakash Gollapudi
9be17fc43e [SCSI] bnx2fc: Reorganize cleanup code between interface_cleanup and if_destory
Move interface specific cleanup functionality to from bnx2fc_if_destroy to
bnx2fc_interface_cleanup. Do not access interface/hba in bnx2fc_if_destroy as
by the time this function is called interface may already be destroyed. This
patch is in preparation to handle NETDEV_UNREGISTER on a vlan device.

Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:35:33 -06:00
Bhanu Prakash Gollapudi
776cebcac6 [SCSI] bnx2fc: Change function names of bnx2fc_netdev_setup/bnx2fc_netdev_cleanup
Change them to bnx2fc_interface_setup/bnx2fc_interface_cleanup in preperation
for the patches to follow. Interface specific cleanup functionality will be
moved to bnx2fc_interface_cleanup.

Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:35:31 -06:00
Bhanu Prakash Gollapudi
cdf54668bc [SCSI] bnx2fc: Do not attempt destroying NPIV port twice
When NPIV ports are created/deleted rapidly there is a race condition between
bnx2fc_vport_destroy() from sysfs and bnx2fc_flogi_resp(), which could try to
delete the NPIV port from the list twice. Fix is to loop through the list of
NPIV ports to find a match, and only when it exists remove it.

Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:35:30 -06:00
Bhanu Prakash Gollapudi
b65d457913 [SCSI] bnx2fc: Remove erroneous kref_get on IO request
During sequence cleanup, an additional reference for an IO has been
taken. Because of this, the IO is never released into the free list.

Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:35:28 -06:00
Bhanu Prakash Gollapudi
e9a5289ca3 [SCSI] bnx2fc: Enable bsg_request support for bnx2fc
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:35:26 -06:00
Bhanu Prakash Gollapudi
3224876358 [SCSI] bnx2fc: Bug fixes in percpu_thread_create/destroy
Look up p->work_list to process cq completions, and correct the error check for
thread creation.

Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:35:25 -06:00
Bhanu Prakash Gollapudi
627e628f66 [SCSI] bnx2fc: Reset the max receive frame size
Reset max receive frame size every time before attempting FLOGI. Without this,
the stale MFS value will be used.

Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:35:23 -06:00
kashyap.desai@lsi.com
9ae89b0296 [SCSI] mpt2sas: Added missing mpt2sas_base_detach call from scsih_remove context
mpt2sas_base_detach() call was removed from _scsih_remove() while
doing some code shuffling.  Mainly when we work on adding code for
scsih_shutdown().  I have added back mpt2sas_base_detach() which will
get callled from _scsih_remove().

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:35:21 -06:00
kashyap.desai@lsi.com
731f56c845 [SCSI] mptfusion: Bump version 3.4.20
Upgrade version string.
Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:35:19 -06:00
kashyap.desai@lsi.com
a38ae37fdc [SCSI] mptfusion: Avoid out of order Event processing due to cpu migration
Driver will now schedule MPI event using "delay_work_queue_on" to
specify same CPU to be used to schedule work.  Earlier it used
"delay_work_queue" which can cause migration of work due to kernel'
timer migration feature.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:35:18 -06:00
kashyap.desai@lsi.com
e62cca19a9 [SCSI] mptfusion: Better handling of DEAD IOC PCI-E Link down error condition
Find Non-Operation IOC and remove it from OS: Detecting
dead(non-functional) ioc will be done reading doorbell register value
from fault reset thread, which has been called from work thread
context after each specific interval. If doorbell value is 0xFFFFFFFF,
it will be considered as IOC is non-operational and marked as dead
ioc.

Once Dead IOC has been detected, it will be removed at pci layer using
"pci_remove_bus_device" API.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:35:16 -06:00
kashyap.desai@lsi.com
3850b14e51 [SCSI] mptfusion: Set max sector count module parameter
The max_sector setting is currently hard-coded in the driver to 8192
sectors (4MB transfers). Using new module parameter, if max_sectors is
specified at load time, the default of 8192 will be overridden.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:35:14 -06:00
Dan Williams
4fcf812ca3 [SCSI] libsas: export sas_alloc_task()
Now that isci has added a 3rd open coded user of this functionality just
share the libsas version.

Acked-by: Jack Wang <jack_wang@usish.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:35:13 -06:00
Chad Dupuis
7ca3c803e8 [SCSI] qla2xxx: Update version number to 8.03.07.07-k.
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:23:28 -06:00
Saurav Kashyap
51cc9a8e5f [SCSI] qla2xxx: Set the task attributes after memsetting fcp cmnd.
The memset of the fcp_cmnd struct needs to be moved so that it will not
zero-out valid data.

Signed-off-by: Saurav Kashyap <saurav.kashyap@qlogic.com>
Signed-off-by: Chad Dupuis <chad.dupuis@qlogic.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:22:09 -06:00
Andrew Vasquez
58b4857696 [SCSI] qla2xxx: Correct inadvertent loop state transitions during port-update handling.
Transitioning to a LOOP_UPDATE loop-state could cause the driver
to miss normal link/target processing.  LOOP_UPDATE is a crufty
artifact leftover from at time the driver performed it's own
internal command-queuing.  Safely remove this state.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: Chad Dupuis <chad.dupuis@qlogic.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:21:26 -06:00
istvan_v@mailbox.hu
84b271488d [media] cx88: notch filter control fixes
This patch reduces the number of available choices for the notch filter type control
so that the standard-specific filter types cannot be selected. It is now limited to
being either 0 (4xFsc, the default) or 1 (square pixel optimized).
The patch also removes the initialization of this control from cx88_reset(), since
that is already done by init_controls(), which is called by cx8800_initdev().

Signed-off-by: Istvan Varga <istvan_v@mailbox.hu>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 11:21:16 -03:00
Joe Perches
2b50763a0b [media] tda18212: Use standard logging, remove tda18212_priv.h
Use the more current logging styles with pr_fmt.
Remove now unnecessary private include.

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 11:20:45 -03:00
Saurav Kashyap
3553d343e7 [SCSI] qla2xxx: Save and restore irq in the response queue interrupt handler.
Signed-off-by: Saurav Kashyap <saurav.kashyap@qlogic.com>
Signed-off-by: Chad Dupuis <chad.dupuis@qlogic.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:20:18 -06:00
Chad Dupuis
bc91ade9b7 [SCSI] qla2xxx: Double check for command completion if abort mailbox command fails.
Close a small window where we could falsely fail an abort request if the mailbox
command fails but the command was returned during interrupt context.

Signed-off-by: Chad Dupuis <chad.dupuis@qlogic.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:19:56 -06:00
Joe Perches
be85fefecb [media] tda18271: Use printk extension %pV
Deduplicate printk formats to save ~20KB text.

$ size drivers/media/common/tuners/tda18271*o.*
   text	   data	    bss	    dec	    hex	filename
  10747	     56	   1920	  12723	   31b3	drivers/media/common/tuners/tda18271-common.o.new
  18889	     56	   3112	  22057	   5629	drivers/media/common/tuners/tda18271-common.o.old
  20561	    204	   4264	  25029	   61c5	drivers/media/common/tuners/tda18271-fe.o.new
  31093	    204	   6000	  37297	   91b1	drivers/media/common/tuners/tda18271-fe.o.old
   3681	   6760	    440	  10881	   2a81	drivers/media/common/tuners/tda18271-maps.o.new
   5631	   6760	    680	  13071	   330f	drivers/media/common/tuners/tda18271-maps.o.old

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 11:19:43 -03:00
Saurav Kashyap
7594206493 [SCSI] qla2xxx: Acquire hardware lock while manipulating dsd list.
The dsd list shouldn't be manipulated without taking the per host hardware
lock to prevent multiple callers from trampling upon one another.

Signed-off-by: Saurav Kashyap <saurav.kashyap@qlogic.com>
Signed-off-by: Chad Dupuis <chad.dupuis@qlogic.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:19:37 -06:00
Chad Dupuis
42cd4f5dc2 [SCSI] qla2xxx: Fix qla24xx revision check while enabling interrupts.
Since we enable interrupts before initializing the firmware, use the chip
revision from PCI config space directly to perform the chip revision check.
Also remove the unnecessary firmware attributes test.

Signed-off-by: Chad Dupuis <chad.dupuis@qlogic.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:18:47 -06:00
Antti Palosaari
f68baeff45 [media] af9015: use logic or instead of sum numbers
Style issue.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 11:14:43 -03:00
Antti Palosaari
4c0cfa24f7 [media] af9015: map remote for Leadtek WinFast DTV2000DS
Thanks to Thomas Gutzler for reporting this.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Cc: Thomas Gutzler <thomas.gutzler@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 11:14:32 -03:00
Antti Palosaari
e36454376a [media] em28xx: use MFE lock for PCTV nanoStick T2 290e
Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 11:14:22 -03:00
Julia Lawall
eb28dc39d3 [media] drivers/media/dvb/dvb-usb/usb-urb.c: adjust array index
Convert array index from the loop bound to the loop index.

A simplified version of the semantic patch that fixes this problem is as

// <smpl>
@@
expression e1,e2,ar;
@@

for(e1 = 0; e1 < e2; e1++) { <...
  ar[
- e2
+ e1
  ]
  ...> }
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 11:14:15 -03:00
Arun Easi
e02587d777 [SCSI] qla2xxx: T10 DIF - Fix incorrect error reporting.
This fix:
    - Disables app tag peeking; correct tag check will be added when the
      SCSI API is available.
    - Always derive ref_tag from scsi_get_lba()
    - Removes incorrect swap of FCP_LUN in FCP_CMND
    - Moves app-tag error check before ref-tag check. The reason being,
      currently there is no interface in SCSI to retrieve the app-tag
      for protection I/Os, so driver puts zero for app-tag in the
      firmware interface, but requests not to validate it, but when a
      ref-tag error is detected by firmware, it would put
      expected/actual tags for all the protection tags (guard/app/ref).
      As driver checks for app tag error first, a ref-tag error is
      incorrectly flagged as app-tag error.
    - Convert HBA specific checks to capability based.

Signed-off-by: Arun Easi <arun.easi@qlogic.com>
Signed-off-by: Chad Dupuis <chad.dupuis@qlogic.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 08:13:52 -06:00
Arun Easi
8cb2049c74 [SCSI] qla2xxx: T10 DIF - Handle uninitalized sectors.
Driver needs to update protection bytes for uninitialized sectors as they are
not DMA-d.

Signed-off-by: Arun Easi <arun.easi@qlogic.com>
Reviewed-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: Chad Dupuis <chad.dupuis@qlogic.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-27 07:55:46 -06:00
Julian Scheel
93165b7774 [media] Add support for new revision of KNC 1 DVB-C cards. Using tda10024 instead of tda10023, which is compatible to tda10023 driver
Signed-off-by: Julian Scheel <julian@jusst.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 10:41:46 -03:00
Benjamin Larsson
3188d5485e [media] get_dvb_firmware: Firmware extraction for IT9135 based devices
Signed-off-by: Benjamin Larsson <benjamin@southpole.se>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 10:37:56 -03:00
Malcolm Priestley
f6d8735493 [media] it913x: Driver for Kworld UB499-2T (id 1b80:e409) v1.05
Driver for Kworld UB499-2T (id 1b80:e409)

The device driver has been named it913x, so that support for other family members
can be added later.

TODOs
Firmware support for other it913x devices.
Remote control support, there are two known types.

Signed-off-by: Malcolm Priestley <tvboxspy@gmail.com>
[mchehab@redhat.com: Fix a merge conflict]
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 10:35:21 -03:00
Malcolm Priestley
52cb3e2757 [media] it9137: Fimrware retrival information for Kworld UB499-2T T09 (id 1b80:e409)
Firmware information for Kworld UB499-2T T09 based on IT913x series. This device
uses file dvb-usb-it9137-01.fw.

Signed-off-by: Malcolm Priestley <tvboxspy@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 10:30:43 -03:00
Malcolm Priestley
3dbbf82f49 [media] it913x_fe: frontend and tuner driver v1.05
Fronted and Tuner Driver for ITE IT913x Series with inital support for
IT9137 integrated demodulator and tuner device.

The driver is loosely based on AF9035 series. However, support is not intended for
this device specificity.

The IT9137 tuner has been tested on UHF bands, but VHF has only been simulated.

Possible TODO the tuner sections may be separated from the main driver. All future devices
should use the it913x_fe_script_loader for other tuner devices.

Signed-off-by: Malcolm Priestley <tvboxspy@gmail.com>
[mchehab@redhat.com: Fix an issue at the Kconfig help]
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 10:29:20 -03:00
Doron Cohen
ea3709435c [media] siano: apply debug flag to module level
Siano modules already had sms_dbg flag which is a module parameter which
sets the debug mode so module prints messages to dmesg for debugging.
The variable was static therefore apply only to the file which defines
the module. In modules as smsmdtv.ko that contain a few files, the debug
flag applied only for functions in that main file.
flag was changed to be non-static and therefore can be accessed by all
module files (although it is still not exported out of the module).

Signed-off-by: Doron Cohen <doronc@siano-ms.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 09:56:06 -03:00
Magnus Damm
d0168fdc7a ARM: mach-shmobile: sh7372 LCDC1 suspend fix V2 (incremental)
This patch updates the recently submitted
"Associate the HDMI clock together with LCDC1 on sh7372"
to V2 with the following change:
 - Use lcdc1_device on AP4EVB to build properly.

Signed-off-by: Magnus Damm <damm@opensource.se>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
2011-08-27 14:21:00 +02:00
Jarod Wilson
fda516b72a [media] mceusb: update version, copyright, author
Add note about recent updates coming from Microsoft's publicly available
specs on Windows Media Center remotes and receivers/transmitters.

Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 08:57:01 -03:00
Jarod Wilson
e217fb43c4 [media] mceusb: report actual tx frequencies
Rather than dumping out hex values, lets print the actual calculated
frequency and period the hardware has been configured for. After this

[ 2643.276215] mceusb 3-1:1.0: tx data: 9f 07 (length=2)
[ 2643.276218] mceusb 3-1:1.0: Get carrier mode and freq
[ 2643.277206] mceusb 3-1:1.0: rx data: 9f 06 01 42 (length=4)
[ 2643.277209] mceusb 3-1:1.0: Got carrier of 37037 Hz (period 27us)

Matches up perfectly with the table in Microsoft's docs.

Of course, I've noticed on one of my devices that the MS-recommended
default value of 1 for carrier pre-scaler and 66 for carrier period was
butchered, and instead of converting 66 to hex (0x42 like above), they
put in 0x66, so the hardware reports a default carrier of 24390Hz.
Fortunately, I guess, this particular device is rx-only, but I wouldn't
put it past other hw to screw up here too.

Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 08:56:39 -03:00
Jarod Wilson
b71969bee2 [media] mceusb: flash LED (emu v2+ only) to signal end of init
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 08:55:33 -03:00
Jarod Wilson
a411e83944 [media] mceusb: get misc port data from hardware
According to the specs, you can read the number of tx ports, number of
rx sensors, which tx ports have cables plugged into them, and which rx
sensors are active. In practice, most of my devices do seem to report
sane values for tx ports and rx sensors (but not all -- one without any
tx ports reports having them), and most report the active sensor
correctly, but only one of eight reports cabled tx ports correctly. So
for the most part, this is just for informational purposes.

Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 08:55:02 -03:00
Jarod Wilson
ab1072eba9 [media] mceusb: query device for firmware emulator version
Supposedly, there are essentially three different classes of devices
that are compatible with Microsoft's specs. First are the "legacy"
devices, which are built using Microsoft-provided hardware specs and
firmware. Second are "emulator" devices, which are built using custom
hardware and firmware, written to emulate Microsoft's firmware. Third
are "port" devices, which have their own device driver and firmware,
which provides compatible data to higher levels of the stack.

>From what I can tell, things like nuvoton-cir and fintek-cir are
essentially "port" devices -- their raw IR buffer format is very similar
to that of the mceusb devices. Now, within the mceusb driver, we have
three different "generations", which at first, seemed like maybe they
mapped to emulator versions. Unfortuantely, every single device I have
responds "illegal command" to the query to get firmware emulator version
from the hardware, which means they're either all emulator version 1, or
they're legacy devices, and our different "generations" aren't at all
related here. Though in theory, its possible the gen1 devices are
"legacy" devices and the rest are emulator v1. There are some useful
features of the v2 interface I was hoping to play with, but alas...

Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 08:53:15 -03:00
Jarod Wilson
4840b788ad [media] mceusb: issue device resume cmd when needed
According to MS docs, the device firmware may halt after receiving an
unknown instruction, but that it should be possible to tell the firmware
to continue running by simply sending a device resume command. So lets
do that.

Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 08:51:27 -03:00
Jarod Wilson
fa3348980a [media] mceusb: set wakeup bits for IR-based resume
Its not uncommon for folks to force these bits enabled, because people
do want to wake their htpc kit via their remote. Lets just set the bits
for 'em.

Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 08:50:31 -03:00
Jarod Wilson
417c0a23b7 [media] mceusb: give hardware time to reply to cmds
Sometimes the init routine is blasting commands out to the hardware
faster than it can reply. Throw a brief delay in there to give the
hardware a chance to reply before we send the next command.

Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 08:49:30 -03:00
Jarod Wilson
a6fbd3b77a [media] mceusb: command/response updates from MS docs
I was recently pointed to the document titled
Windows-Media-Center-RC-IR-Collection-Green-Button-Specification-03-08-2011-V2.pdf
which as of this writing, is publicly available from
download.microsoft.com. It covers a LOT of the gaps in the mceusb
driver, which to this point, was written almost entirely by
reverse-engineering. First up, I'm updating the defines for all the MCE
commands and responses to match their names in the spec. More to come...

Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-08-27 08:48:05 -03:00
Shaohua Li
136333d104 slub: explicitly document position of inserting slab to partial list
Adding slab to partial list head/tail is sensitive to performance.
So explicitly uses DEACTIVATE_TO_TAIL/DEACTIVATE_TO_HEAD to document
it to avoid we get it wrong.

Acked-by: Christoph Lameter <cl@linux.com>
Signed-off-by: Shaohua Li <shli@kernel.org>
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
2011-08-27 11:59:00 +03:00
Shaohua Li
130655ef09 slub: add slab with one free object to partial list tail
The slab has just one free object, adding it to partial list head doesn't make
sense. And it can cause lock contentation. For example,
1. CPU takes the slab from partial list
2. fetch an object
3. switch to another slab
4. free an object, then the slab is added to partial list again
In this way n->list_lock will be heavily contended.
In fact, Alex had a hackbench regression. 3.1-rc1 performance drops about 70%
against 3.0. This patch fixes it.

Acked-by: Christoph Lameter <cl@linux.com>
Reported-by: Alex Shi <alex.shi@intel.com>
Signed-off-by: Shaohua Li <shli@kernel.org>
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
2011-08-27 11:58:59 +03:00
Jeff Kirsher
88f07484cc drivers/net/ethernet/*: Enabled vendor Kconfig options
Based on finds for Stephen Rothwell, where current defconfig's
enable a ethernet driver and it is not compiled due to the newly
added NET_VENDOR_* component of Kconfig.

This patch enables all the "new" Kconfig options so that current
defconfig's will continue to compile the expected drivers.  In
addition, by enabling all the new Kconfig options does not add
any un-expected options.

CC: Stephen Rothwll <sfc@canb.auug.org.au>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-27 00:58:56 -07:00
Jeff Kirsher
88491d8103 drivers/net: Kconfig & Makefile cleanup
The is does a general cleanup of the drivers/net/ Kconfig and
Makefile.  This patch create a "core" option and places all
the networking core drivers into this option (default is yes
for this option).  In addition, it alphabitizes the Kconfig
driver options.

As a side cleanup, found that the arcnet, token ring, and PHY
Kconfig options were a tri-state option and should have been
a bool option.

Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-27 00:58:49 -07:00
Jeff Kirsher
330278cde6 com20020_cs: Move the PCMCIA Arcnet driver
Move the COM20020 PCMICA Arcnet driver into drivers/net/arcnet/ with
the other Arcnet drivers.  Made the necessary Kconfig and Makefile
changes as well.

Since this was the "last" PCMCIA driver in drivers/net/pcmcia/, this patch
also cleans up the references to drivers/net/pcmcia.

CC: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-27 00:58:44 -07:00