Commit graph

291,781 commits

Author SHA1 Message Date
Fabrice Jouhaud
d44651d0f9 ext4: fix ext4 so it works without CONFIG_PROC_FS
This fixes a bug which was introduced in dd68314ccf.  The problem
came from the test of the return value of proc_mkdir which is always
false without procfs, and this would initialization of ext4.

Signed-off-by: Fabrice Jouhaud <yargil@free.fr>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-10-08 16:26:03 -04:00
Arnd Bergmann
aab3a70e98 Merge branch 'omap/fixes-for-3.1' into fixes 2011-10-08 22:21:07 +02:00
Arnd Bergmann
589a66da99 Merge branch 'fixes' into next/fixes 2011-10-08 22:18:39 +02:00
Tao Ma
6ee3b21224 ext4: use le32_to_cpu for ext4_extent_idx.ei_block in ext4_ext_search_left()
ext4_extent_idx.e_block is __le32, so use le32_to_cpu() in
ext4_ext_search_left().

Signed-off-by: Tao Ma <boyu.mt@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-10-08 16:08:34 -04:00
Tao Ma
7fd59c83b0 ext4: remove the obsolete/broken EXT4_IOC_WAIT_FOR_READONLY ioctl
There are no users of the EXT4_IOC_WAIT_FOR_READONLY ioctl, and it is
also broken.  No one sets the set_ro_timer, no one wakes up us and our
state is set to TASK_INTERRUPTIBLE not RUNNING.  So remove it.

Signed-off-by: Tao Ma <boyu.mt@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-10-08 15:56:35 -04:00
Tao Ma
df3ab17072 ext4: fix the comment describing ext4_ext_search_right()
The comment describing what ext4_ext_search_right() does is incorrect.
We return 0 in *phys when *logical is the 'largest' allocated block,
not smallest.  

Fix a few other typos while we're at it.

Cc: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Tao Ma <boyu.mt@taobao.com>
2011-10-08 15:53:49 -04:00
Arnd Bergmann
a3849a4c03 Merge branch 'stericsson/fixes' into next/cleanup
Conflicts:
	arch/arm/mach-ux500/cpu.c
2011-10-08 21:47:06 +02:00
Arnd Bergmann
32d80f97fe Merge branch 'imx/fixes' into next/fixes 2011-10-08 21:43:22 +02:00
Arnaud Patard (Rtp)
100aafc992 ARM: mx5: fix clock usage for suspend
While suspending, we're enabling a clock in ->suspend() but we're in atomic
context, leading to this :

[30803.667305] PM: late suspend of devices complete after 1.237 msecs
[30803.667449] BUG: sleeping function called from invalid context at kernel/mutex.c:271
[30803.667464] in_atomic(): 0, irqs_disabled(): 128, pid: 4941, name: pm-suspend
[30803.667474] INFO: lockdep is turned off.
[30803.667483] irq event stamp: 0
[30803.667489] hardirqs last  enabled at (0): [<  (null)>]   (null)
[30803.667503] hardirqs last disabled at (0): [<c001e4a8>] copy_process.part.48+0x1e0/0xa7c
[30803.667543] softirqs last  enabled at (0): [<c001e4a8>] copy_process.part.48+0x1e0/0xa7c
[30803.667562] softirqs last disabled at (0): [<  (null)>]   (null)
[30803.667574] Backtrace:
[30803.667611] [<c0010e00>] (dump_backtrace+0x0/0x110) from [<c0424c00>] (dump_stack+0x18/0x1c)
[30803.667624]  r6:00000000 r5:00000000 r4:d9648000 r3:d9648000
[30803.667652] [<c0424be8>] (dump_stack+0x0/0x1c) from [<c0424eec>] (__might_sleep.part.101+0x90/0xa8)
[30803.667673] [<c0424e5c>] (__might_sleep.part.101+0x0/0xa8) from [<c001a5c4>] (__might_sleep+0x80/0x94)
[30803.667686]  r4:c05d58c0
[30803.667705] [<c001a544>] (__might_sleep+0x0/0x94) from [<c043230c>] (mutex_lock_nested+0x2c/0x30c)
[30803.667735] [<c04322e0>] (mutex_lock_nested+0x0/0x30c) from [<c0017834>] (clk_enable+0x28/0x50)
[30803.667747]  r8:c0438464 r7:00000003 r6:00000000 r5:00000000 r4:c05d2e98
[30803.667780] [<c001780c>] (clk_enable+0x0/0x50) from [<c0017260>] (mx5_suspend_enter+0x1c/0x98)
[30803.667792]  r4:00000003 r3:c060dfe0
[30803.667821] [<c0017244>] (mx5_suspend_enter+0x0/0x98) from [<c0059b48>] (suspend_enter+0xec/0x15c)
[30803.667833]  r4:00000003 r3:c0017244
[30803.667856] [<c0059a5c>] (suspend_enter+0x0/0x15c) from [<c0059c4c>] (suspend_devices_and_enter+0x94/0x130)
[30803.667868]  r6:00000000 r5:00000003 r4:c0c0af00 r3:00002710
[30803.667897] [<c0059bb8>] (suspend_devices_and_enter+0x0/0x130) from [<c0059db0>] (enter_state+0xc8/0x130)
[30803.667909]  r6:00000000 r5:00000003 r4:c05286e4
[30803.667934] [<c0059ce8>] (enter_state+0x0/0x130) from [<c00592d4>] (state_store+0xac/0xc0)
[30803.667945]  r6:00000003 r5:00000003 r4:df39d000 r3:00000003
[30803.667978] [<c0059228>] (state_store+0x0/0xc0) from [<c01b0100>] (kobj_attr_store+0x1c/0x28)
[30803.668009] [<c01b00e4>] (kobj_attr_store+0x0/0x28) from [<c00e435c>] (sysfs_write_file+0x88/0xbc)
[30803.668032] [<c00e42d4>] (sysfs_write_file+0x0/0xbc) from [<c0091fa0>] (vfs_write+0xbc/0x138)
[30803.668051] [<c0091ee4>] (vfs_write+0x0/0x138) from [<c0092204>] (sys_write+0x44/0x70)
[30803.668062]  r8:00000000 r7:00000004 r6:00000003 r5:002694d0 r4:d966acc0
[30803.668094] [<c00921c0>] (sys_write+0x0/0x70) from [<c000db00>] (ret_fast_syscall+0x0/0x3c)
[30803.668106]  r9:d9648000 r8:c000dcc4 r6:00000001 r5:002694d0 r4:00000003
[30803.669927] PM: early resume of devices complete after 0.972 msecs

Just move the clk_enable/disable in ->prepare() and ->finish()

Signed-off-by: Arnaud Patard <arnaud.patard@rtp-net.org>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
2011-10-08 21:43:06 +02:00
Arnd Bergmann
71f2c15375 Merge branch 'depends/rmk/devel-stable' into next/cleanup 2011-10-08 21:07:42 +02:00
Lukas Czerner
4113c4caa4 ext4: remove deprecated oldalloc
For a long time now orlov is the default block allocator in the
ext4. It performs better than the old one and no one seems to claim
otherwise so we can safely drop it and make oldalloc and orlov mount
option deprecated.

This is a part of the effort to reduce number of ext4 options hence the
test matrix.

Signed-off-by: Lukas Czerner <lczerner@redhat.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-10-08 14:34:47 -04:00
Theodore Ts'o
af909a57fd ext4: documentation: remove acl and user_xattr mount options
Acl and user_xattr mount options are no longer needed since those
features are enabled by default if configured in (seee commit
ea66333694). We can not easily deprecate
mount options itself (since it is probably too early), but we can
remove it from documentation first.

Signed-off-by: Lukas Czerner <lczerner@redhat.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-10-08 14:01:08 -04:00
Mark Brown
7ca3a18b05 ASoC: Assign power_check when we allocate DAPM widgets
This ensures none of the rest of the code ever encounters a widget which
does not have a power check function.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
2011-10-08 17:55:55 +01:00
Mark Brown
2dc00213b0 ASoC: Ensure all DAPM widgets are instantiated with the card
Specifically for the widgets added by machine driver late probe functions.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
2011-10-08 17:55:51 +01:00
Mark Brown
87bea31c7b ASoC: Remove redundant snd_soc_dapm_sync() calls from machine drivers
The core will sync DAPM as part of the card initialization, there is no
need for machine drivers to do so during their setup.

OMAP drivers are omitted as I know Peter already has patches for them.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
2011-10-08 14:10:54 +01:00
Arnd Bergmann
97b09da4ee ARM: pxa: use correct __iomem annotations
This tries to clear up the confusion between integers and iomem pointers
in the marvell pxa platform. MMIO addresses are supposed to be __iomem*
values, in order to let the Linux type checking work correctly. This
patch moves the cast to __iomem as far back as possible, to the place
where the MMIO virtual address windows are defined.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
2011-10-08 21:03:07 +08:00
Arnd Bergmann
7272889d3f ARM: pxa: sharpsl pm needs SPI
The sharpsl code selects the max1111 spi driver, so it must also
ensure that SPI itself is enabled. The platform does not work
without max1111.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
2011-10-08 21:03:06 +08:00
Arnd Bergmann
41f91d90ec ARM: pxa: centro and treo680 need palm27x
The two platforms are part of the palm27x family and use the same
code. Select the palm27x symbol to make sure they can be built
standalone.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
2011-10-08 21:03:05 +08:00
Eric Miao
c9d1917c63 ARM: pxa: make pxafb_smart_*() empty when not enabled
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
2011-10-08 21:03:04 +08:00
Arnd Bergmann
8582643a4b ARM: pxa: select POWER_SUPPLY on raumfeld
The raumfeld platform code calls power_supply_set_battery_charged
which is part of the power supply layer, so that always has
to be enabled.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
2011-10-08 21:03:03 +08:00
Arnd Bergmann
59efd93f64 ARM: pxa: pxa95x is incompatible with earlier pxa
We cannot support ARMv5 and ARMv7 based boards in a single kernel,
so introduce a new option in mach-pxa to select between the two.

The PJ4 (ARMv7) based boards are now only visible when
CONFIG_ARCH_PXA_V7 is set, the other boards are only visible
when it's not set.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
2011-10-08 21:03:02 +08:00
Arnd Bergmann
ca7d156e30 ARM: pxa: CPU_FREQ_TABLE is needed for CPU_FREQ
The pxa specific cpufreq code is based on the cpu_freq_table
module, so we have to select that.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
2011-10-08 21:03:01 +08:00
Eric Miao
654a60e0b4 ARM: pxa: pxa95x/saarb depends on pxa3xx code
saarb uses pxa3xx_map_io and pxa3xx_handle_irq, which are part
of the pxa3xx code. This makes sure the necessary header and
implementation is used when building the board file.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
2011-10-08 21:03:00 +08:00
Arnd Bergmann
0c82cc5d72 ARM: pxa: allow selecting just one of TREO680/CENTRO
There are two variants of the palmtreo machine, the 680 and the
centro, and Kconfig allows selecting one or both of them.
This changes the board file so that it's actually possible
to build all configurations that are allowed.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: Marek Vasut <marek.vasut@gmail.com>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
2011-10-08 21:02:59 +08:00
Arnd Bergmann
145580388b ARM: pxa: export symbols from pxa3xx-ulpi
The pxa3xx_u2d_start_hc/pxa3xx_u2d_stop_hc symbols are used by the
ohci-pxa27x driver, which can be a module.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: Igor Grinberg <grinberg@compulab.co.il>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
2011-10-08 21:02:58 +08:00
Arnd Bergmann
ec13592060 ARM: pxa: make zylonite_pxa*_init declaration match code
When support for zylonite is disabled, the zylonite_pxa300_init
and zylonite_pxa320_init functions are not there, but the declaration
is still kept around if any other boards for the same soc are
enabled. This changes the declaration to be conditional on the
same symbol as the code.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
2011-10-08 21:02:57 +08:00
Eric Miao
ba4a90a6d8 ARM: pxa/z2: fix building error of pxa27x_cpu_suspend() no longer available
Cc: Vasily Khoruzhick <anarsoul@gmail.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
2011-10-08 21:02:44 +08:00
Malcolm Priestley
e305288547 [media] it913x-fe changes to power up and down of tuner
Currently the tuner is constantly powered causing these effects.
1. Remembering last tune channel causing corruptions of changing channel.
2. Causing corruption on other frontend.
3. Higher current in standby of demodulator with clock running.

Power sequence now follows;
Power Up
Tuner on -> Frontend suspend off -> Tuner clk on
Power Down
Frontend suspend on -> Tuner clk off -> Tuner off

Signed-off-by: Malcolm Priestley <tvboxspy@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-10-08 08:03:27 -03:00
Javier Martinez Canillas
64933337e3 [media] tvp5150: Add video format registers configuration values
The tvp5150 video decoder has two operation modes to configure the video
standard used.

If auto-switch mode is enable, the device can sense the signal and detect which
video standard the device is operating. Also the device can be forced to use a
user defined video standard.

Each operation mode uses a different register and the bitmask values to
represent each standard is different.

So we add video standard constants for both autoswitch and no-autoswitch mode.

Signed-off-by: Javier Martinez Canillas <martinez.javier@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-10-08 08:01:52 -03:00
Paul Gortmaker
86caf81138 [media] drivers/media: fix dependencies in video mt9t001/mt9p031
Both mt9t001.c and mt9p031.c have two identical issues, those
being that they will need module.h inclusion for the upcoming
cleanup going on there, and that their dependencies don't limit
selection of configs that will fail to compile as follows:

The related config options are CONFIG_MEDIA_CONTROLLER and
CONFIG_VIDEO_V4L2_SUBDEV_API.  Looking at the code, it appears
that the driver was never intended to work without these enabled,
so add a dependency on CONFIG_VIDEO_V4L2_SUBDEV_API, which in
turn already has a dependency on CONFIG_MEDIA_CONTROLLER.

Reported-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Acked-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-10-08 08:01:46 -03:00
Mauro Carvalho Chehab
d56ae6fbf3 [media] em28xx: Add VIDIOC_QUERYSTD support
Allow subdevs to return the detected standards

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-10-08 08:01:40 -03:00
Mauro Carvalho Chehab
5d1ed98683 [media] msp3400: Add standards detection to the driver
As msp3400 allows standards detection, add support for it. That
efectivelly means that devices with msp3400 can now implement
VIDIOC_QUERYSTD, and it will provide very good detection for
the standard, specially if combined with a video decoder detection.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-10-08 08:01:34 -03:00
Mauro Carvalho Chehab
8d6c0b216f [media] videodev2: Reorganize standard macros and add a few more macros
Reorganize the standards macro and add a few more, that will be
used on msp3400 in order to allow it to detect the audio standard.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-10-08 08:01:06 -03:00
Mauro Carvalho Chehab
a5abdb6044 [media] pvrusb2: initialize standards mask before detecting standard
Acked-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-10-08 08:00:54 -03:00
Mauro Carvalho Chehab
e41567a61d [media] saa7115: Trust that V4L2 core will fill the mask
Instead of using V4L2_STD_ALL when no standard is detected,
trust that the maximum allowed standards are already filled by
the V4L2 core. It is better this way, as the bridge and/or the audio
decoder may have some extra restrictions to some video standards.

This also allow other devices like audio and tuners to contribute to
standards detection, when they support such feature.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-10-08 08:00:44 -03:00
Mauro Carvalho Chehab
8715e6cbae [media] v4l2-ioctl: Fill the default value for VIDIOC_QUERYSTD
According with the V4L2 API spec:

	"When detection is not possible or fails, the set must contain
	 all standards supported by the current video input or output."

The V4L2 core has the mask with all supported standards already. So,
apply it. Driver and subdevs can then just remove standards from the
mask, as they're able of detecting audio, video and frames frequency.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-10-08 08:00:35 -03:00
Mauro Carvalho Chehab
7383a47395 [media] pvrusb2: implement VIDIOC_QUERYSTD
Acked-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-10-08 08:00:26 -03:00
Mauro Carvalho Chehab
b728666ba5 [media] saa7115: Fix standards detection
There are several bugs at saa7115 standards detection:

After the fix, the driver is returning the proper standards,
as tested with 3 different broadcast sources:

On an invalid channel (without any TV signal):
[ 4394.931630] saa7115 15-0021: Status byte 2 (0x1f)=0xe0
[ 4394.931635] saa7115 15-0021: detected std mask = 00ffffff

With a PAL/M signal:
[ 4410.836855] saa7115 15-0021: Status byte 2 (0x1f)=0xb1
[ 4410.837727] saa7115 15-0021: Status byte 1 (0x1e)=0x82
[ 4410.837731] saa7115 15-0021: detected std mask = 00000900

With a NTSC/M signal:
[ 4422.383893] saa7115 15-0021: Status byte 2 (0x1f)=0xb1
[ 4422.384768] saa7115 15-0021: Status byte 1 (0x1e)=0x81
[ 4422.384772] saa7115 15-0021: detected std mask = 0000b000

Tests were done with a WinTV PVR USB2 Model 29xx card.

Reviewed-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2011-10-08 08:00:17 -03:00
Mark Brown
4f4c007222 ASoC: Suppress early calls to snd_soc_dapm_sync()
Ensure we only have one sync during the initial startup of the card by
making snd_soc_dapm_sync() a noop on non-instantiated cards. This avoids
any bounces due to things like jacks reporting their initial state on
partially initialised cards. The callers that don't also get called at
runtime should just be removed.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
2011-10-08 11:47:32 +01:00
Will Deacon
a26bce1220 ARM: 7127/1: hw_breakpoint: skip v7-specific reset on v6 cores
ARMv6 cores do not implement the DBGOSLAR register, so we don't need to
try and clear it on boot. Furthermore, the VCR is zeroed out of reset,
so we don't need to zero it explicitly when a CPU comes online.

Tested-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-10-08 10:05:34 +01:00
Alexander Duyck
077887c386 igb: add support for NETIF_F_RXHASH
This patch adds support for Rx hashing.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by:  Aaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-07 22:56:22 -07:00
Alexander Duyck
6d095fa8cb igb: move TX hang check flag into ring->flags
This change moves the Tx hang check into the ring flags.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by:  Aaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-07 22:54:36 -07:00
Alexander Duyck
5faf030c9b igb: fix recent VLAN changes that would leave VLANs disabled after reset
This patch cleans up several issues with VLANs on igb after the recent
changes that were meant to leave the VLANs enabled/disable via the
netdev->features flags.

Specifically the Rx VLAN settings were being dropped after reset due to the
fact that they were not being restored correctly.  In addition I removed
the IRQ disable/enable since those were in place to protect the setting of
vlgrp.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by:  Aaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-07 22:50:57 -07:00
Alexander Duyck
3ceb90fd48 igb: leave staterr in place and instead us a helper function to check bits
Instead of doing a byte swap on the staterr bits in the Rx descriptor we can
save ourselves a bit of space and some CPU time by instead just testing for
the various bits out of the Rx descriptor directly.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by:  Aaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-07 22:49:19 -07:00
Alexander Duyck
294e7d78f5 igb: retire the RX_CSUM flag and use the netdev flag instead
Since the netdev now has its' own checksum flag to indicate if Rx checksum
is enabled we might as well use that instead of using the ring flag.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by:  Aaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-07 22:47:25 -07:00
Alexander Duyck
4be000c874 igb: cleanup IVAR configuration
This change is meant to cleanup some of the IVAR register configuration.
igb_assign_vector had become pretty large with multiple copies of the same
general code for setting the IVAR. This change consolidates most of that
code by adding the igb_write_ivar function which allows us just to compute
the index and offset and then use that information to setup the IVAR.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by:  Aaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-07 22:45:46 -07:00
Alexander Duyck
0ba829943c igb: Move ITR related data into work container within the q_vector
This change moves information related to interrupt throttle rate
configuration into a separate q_vector sub-structure called a work
container. A similar change has already been made for ixgbe and this work
is based off of that.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by:  Aaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-07 22:44:05 -07:00
Alexander Duyck
866cff0690 igb: Consolidate all of the ring feature flags into a single value
This change moves all of the ring flags into a single value.  The advantage
to this is that there is one central area for all of these flags and they
can all make use of the set/test bit operations.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by:  Aaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-07 22:37:24 -07:00
Alexander Duyck
6ad4edfcd7 igb: avoid unnecessary conversions from u16 to int
There are a number of places where we have values that are stored as u16
but are being converted to int unnecessarily.  In order to avoid that we
should convert all variables that deal with the next_to_clean, next_to_use,
and count to u16 values.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by:  Aaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-07 22:35:50 -07:00
Alexander Duyck
81c2fc2232 igb: Use node specific allocations for the q_vectors and rings
This change is meant to update the ring and vector allocations so that they
are per node instead of allocating everything on the node that
ifconfig/modprobe is called on.  By doing this we can cut down
significantly on cross node traffic.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by:  Aaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-07 22:34:13 -07:00