Commit graph

299,176 commits

Author SHA1 Message Date
Eric Dumazet
e7e5a4033f niu: fix skb truesize underestimation
Add a 'truesize' argument to niu_rx_skb_append(), filled with rcr_size
by the caller to properly account frag sizes in skb->truesize

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-13 22:25:21 -04:00
Eric Dumazet
5e6c355c47 vmxnet3: fix skb truesize underestimation
vmxnet3 allocates a page per skb fragment. We must account
PAGE_SIZE increments on skb->truesize, not the actual frag length.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Shreyas Bhatewara <sbhatewara@vmware.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-13 22:25:21 -04:00
Eric Dumazet
5935f81c59 ftgmac100: fix skb truesize underestimation
ftgmac100 allocates a page per skb fragment. We must account
PAGE_SIZE increments on skb->truesize, not the actual frag length.

If frame is under 64 bytes, page is freed, and truesize adjusted.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Po-Yu Chuang <ratbert@faraday-tech.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-13 22:25:20 -04:00
Pavel Shilovsky
d59dad2be0 CIFS: Move byte range lock list from fd to inode
that let us do local lock checks before requesting to the server.

Signed-off-by: Pavel Shilovsky <piastry@etersoft.ru>
Signed-off-by: Steve French <smfrench@gmail.com>
2011-10-13 19:52:47 -05:00
Jean Delvare
bf164c58e5 hwmon: (w83627ehf) Properly report thermal diode sensors
The w83627ehf driver is improperly reporting thermal diode sensors as
type 2, instead of 3. This caused "sensors" and possibly other
monitoring tools to report these sensors as "transistor" instead of
"thermal diode".

Furthermore, diode subtype selection (CPU vs. external) is only
supported by the original W83627EHF/EHG. All later models only support
CPU diode type, and some (NCT6776F) don't even have the register in
question so we should avoid reading from it.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: stable@kernel.org
Signed-off-by: Guenter Roeck <guenter.roeck@ericsson.com>
2011-10-13 16:51:29 -07:00
Jeff Layton
fe11e4ccb8 cifs: clean up check_rfc1002_header
Rename it for better clarity as to what it does and have the caller pass
in just the single type byte. Turn the if statement into a switch and
optimize it by placing the most common message type at the top. Move the
header length check back into cifs_demultiplex_thread in preparation
for adding a new receive phase and normalize the cFYI messages.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <smfrench@gmail.com>
2011-10-13 18:44:40 -05:00
Pavel Shilovsky
03776f4516 CIFS: Simplify byte range locking code
Split cifs_lock into several functions and let CIFSSMBLock get pid
as an argument.

Signed-off-by: Pavel Shilovsky <piastry@etersoft.ru>
Signed-off-by: Steve French <smfrench@gmail.com>
2011-10-13 17:16:28 -05:00
Olof Johansson
ecb7b0e33e ARM: tegra: update defconfig
Refresh tegra_defconfig:

New options enabled: RTC, SPI, USB and USB_STORAGE together with
corresponding tegra drivers. Also enable some of the common usb ethernet
adapters.

Enable Tegra ALSA/ASoC/sound support, including drivers for TrimSlice,
and WM8903-based platforms such as Harmony and Seaboard.

Finally, enable new merged boards (Ventana) and the generic devicetree board.

Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Stephen Warren <swarren@nvidia.com>
2011-10-13 15:07:40 -07:00
Stephen Warren
9eb4b91e69 arm/tegra: Harmony: Configure PMC for low-level interrupts
Without this, the PMC continually detects an interrupt when the PMU_IRQ
line is high, causing the tps6686x IRQ handler thread to hog an entire
CPU.

This change was originally written by Wei Ni <wni@nvidia.com> for Seaboard
in the ChromeOS kernel.

Long-term, this should probably be moved into some kind of PMU driver,
or perhaps integrated into the GPIO/IRQ/pinmux system?

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Olof Johansson <olof@lixom.net>
2011-10-13 15:04:55 -07:00
Peter De Schrijver
add29e61d4 arm/tegra: device tree support for ventana board
Signed-off-by: Peter De Schrijver <pdeschrijver@nvidia.com>
Acked-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Olof Johansson <olof@lixom.net>
2011-10-13 15:04:54 -07:00
Peter De Schrijver
21fb1ccc92 arm/tegra: add support for ventana pinmuxing
Add support for ventana pinmuxing as a seaboard derivative. This is a cut down
version of work done by Jong Kim <jongk@nvidia.com>.

Signed-off-by: Peter De Schrijver <pdeschrijver@nvidia.com>
Acked-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Olof Johansson <olof@lixom.net>
2011-10-13 15:04:54 -07:00
Peter De Schrijver
e146245453 arm/tegra: prepare Seaboard pinmux code for derived boards
This patch splits out the common part of pinmux and GPIO initialization for
seaboard and derived boards. This code is based on work done by Jong Kim
<jongk@nvidia.com>.

Signed-off-by: Peter De Schrijver <pdeschrijver@nvidia.com>
Acked-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Olof Johansson <olof@lixom.net>
2011-10-13 15:04:53 -07:00
Stephen Warren
48f2eceefb arm/tegra: pinmux: ioremap registers
Use ioremap to obtain access to registers instead of using static
mappings. This reduces the number of users of the static mappings, which
will eventually allow them to be removed.

Note that on Tegra30, the number of register "banks" will decrease to 2,
and the packing of specific bits into registers will change significantly.
That's why this change adds the "*_bank" fields to the pingroup tables,
rather than implementing some more hard-coded scheme.

Also, completely remove the implementation of suspend/resume; Tegra doesn't
yet support suspend/resume, and the implementation is complex for the
general pinmux driver:

* Not all registers are used within each bank, so we probably shouldn't
  just iterate over every register in the bank, and save/restore it,
  since that would mean touching undefined registers.

* Registers are shared between pingroups, so we can't simply iterate over
  each pingroup, and save/restore the registers it uses.

It'd probably be best have probe() calculate a bitmask of actually-used
registers for each bank, and have suspend/resume iterate over those
bitmaps.

Oh, and Real Soon Now, I should be looking into converting this driver to
the new pinmux/pinctrl subsystem, so I didn't want to put too much work
into the current incarnation.

v2: s/space/bank/ to match comments on reg_* fields in pinmux.h.
    Re-order bank/reg parameters to pg_readl/pg_writel.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Olof Johansson <olof@lixom.net>
2011-10-13 15:04:53 -07:00
Stephen Warren
88d8951e58 gpio/tegra: Convert to a platform device
v3: Make regs variable static. Remove empty init of tegra_gpio_banks.

v2: Retrieve IRQ and memory addresses from resources instead of hard-
coding them. Add back initialization of tegra_gpio_chip.of_node.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
[olof: switched probe routine to __devinit]
Signed-off-by: Olof Johansson <olof@lixom.net>
2011-10-13 15:04:52 -07:00
Stephen Warren
1ebc8496e8 arm/tegra: Convert pinmux driver to a platform device
Signed-off-by: Stephen Warren <swarren@nvidia.com>
[olof: switch probe function to __devinit]
Signed-off-by: Olof Johansson <olof@lixom.net>
2011-10-13 15:04:38 -07:00
Eric Dumazet
7ae60b3f3b sky2: fix skb truesize underestimation
sky2 allocates a page per skb fragment. We must account
PAGE_SIZE increments on skb->truesize, not the actual frag length.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-13 17:12:46 -04:00
Olof Johansson
d8e9c00e38 ARM: tegra: devices.c should include devices.h
Resolves lots of sparse warnings:

arch/arm/mach-tegra/devices.c:102:24: warning: symbol 'tegra_i2c_device1' was not declared. Should it be static?
arch/arm/mach-tegra/devices.c:112:24: warning: symbol 'tegra_i2c_device2' was not declared. Should it be static?
arch/arm/mach-tegra/devices.c:122:24: warning: symbol 'tegra_i2c_device3' was not declared. Should it be static?
arch/arm/mach-tegra/devices.c:132:24: warning: symbol 'tegra_i2c_device4' was not declared. Should it be static?
[...]

Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Stephen Warren <swarren@nvidia.com>
2011-10-13 14:08:31 -07:00
Olof Johansson
6686c733b8 ARM: tegra: cpu-tegra: unexport two functions
Two static functions that are not exported:

arch/arm/mach-tegra/cpu-tegra.c:59:5: warning: symbol 'tegra_verify_speed' was not declared. Should it be static?
arch/arm/mach-tegra/cpu-tegra.c:64:14: warning: symbol 'tegra_getspeed' was not declared. Should it be static?

Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
2011-10-13 14:08:30 -07:00
Olof Johansson
fdb684ac1c ARM: tegra: cpu-tegra: sparse type fix
Type fix:
arch/arm/mach-tegra/cpu-tegra.c:144:14: warning: incorrect type in argument 5 (different signedness)
arch/arm/mach-tegra/cpu-tegra.c:144:14:    expected unsigned int *index
arch/arm/mach-tegra/cpu-tegra.c:144:14:    got int *<noident>

Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
2011-10-13 14:08:30 -07:00
Olof Johansson
cf28cba0ab ARM: tegra: dma: staticify some tables and functions
None of them are used externally.

Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
2011-10-13 14:08:29 -07:00
Olof Johansson
87c6e46a2c ARM: tegra: tegra2_clocks: don't export some tables
Not used externally, and certainly don't need to be exported.

Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
2011-10-13 14:08:29 -07:00
Olof Johansson
784278e116 ARM: tegra: tegra_powergate_is_powered should be static
Not exported and not used externally.

Also, fix return type. Due to new return type, errors can't be returned
so WARN_ON instead of returning error if a bad parameter is specified.

Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
2011-10-13 14:08:28 -07:00
Olof Johansson
b28fba2a4e ARM: tegra: tegra_rtc_read_ms should be static
Not exported and not used externally.

Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
2011-10-13 14:08:28 -07:00
Olof Johansson
74ae6c3cc5 ARM: tegra: tegra_init_cache should be static
Not exported and not used externally.

Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
2011-10-13 14:08:28 -07:00
Olof Johansson
31e660685d ARM: tegra: pcie: 0 -> NULL changes
Fixes:

arch/arm/mach-tegra/pcie.c:465:10: warning: Using plain integer as NULL pointer

Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
2011-10-13 14:08:27 -07:00
Olof Johansson
3ead5137c2 ARM: tegra: pcie: include board.h
Fixes:

arch/arm/mach-tegra/pcie.c:908:12: warning: symbol 'tegra_pcie_init' was not declared. Should it be static?

Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
2011-10-13 14:08:27 -07:00
Olof Johansson
efaa19a582 ARM: tegra: pcie: don't cast __iomem pointers
Fixes a lot of:

arch/arm/mach-tegra/pcie.c:678:8: warning: cast removes address space of expression
arch/arm/mach-tegra/pcie.c:678:8: warning: incorrect type in argument 1 (different base types)
arch/arm/mach-tegra/pcie.c:678:8:    expected void const volatile [noderef] <asn:2>*<noident>
arch/arm/mach-tegra/pcie.c:678:8:    got unsigned int
arch/arm/mach-tegra/pcie.c:678:8: warning: cast removes address space of expression

Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
2011-10-13 14:08:26 -07:00
Olof Johansson
45fba21862 ARM: tegra: tegra2_clocks: 0 -> NULL changes
Fixes a lot of:

arch/arm/mach-tegra/tegra2_clocks.c:921:34: warning: Using plain integer as NULL pointer
arch/arm/mach-tegra/tegra2_clocks.c:1462:4: warning: Using plain integer as NULL pointer
arch/arm/mach-tegra/tegra2_clocks.c:1864:4: warning: Using plain integer as NULL pointer

Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
2011-10-13 14:08:26 -07:00
Olof Johansson
d395935f55 ARM: tegra: tegra2_clocks: don't cast __iomem pointers
Fixes a lot of:

arch/arm/mach-tegra/tegra2_clocks.c:180:2: warning: cast removes address space of expression
arch/arm/mach-tegra/tegra2_clocks.c:180:2: warning: incorrect type in argument 1 (different base types)
arch/arm/mach-tegra/tegra2_clocks.c:180:2:    expected void const volatile [noderef] <asn:2>*<noident>
arch/arm/mach-tegra/tegra2_clocks.c:180:2:    got unsigned int
arch/arm/mach-tegra/tegra2_clocks.c:180:2: warning: cast removes address space of expression

Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
2011-10-13 14:08:25 -07:00
Olof Johansson
75d711662f ARM: tegra: timer: don't cast __iomem pointers
Fixes a lot of:

arch/arm/mach-tegra/timer.c:75:2: warning: cast removes address space of expression
arch/arm/mach-tegra/timer.c:75:2: warning: incorrect type in argument 1 (different base types)
arch/arm/mach-tegra/timer.c:75:2:    expected void const volatile [noderef] <asn:2>*<noident>
arch/arm/mach-tegra/timer.c:75:2:    got unsigned int
arch/arm/mach-tegra/timer.c:75:2: warning: cast removes address space of expression

Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
2011-10-13 14:08:25 -07:00
Olof Johansson
e748b7310f ARM: tegra: annotate IO_*_VIRT pointers
Provide __iomem annotation for IO_*_VIRT pointers, which will propagate
up through IO_TO_VIRT(). Also fixes a 0 to NULL conversion of the base
case to silence sparse.

Unfortunately map_desc takes an unsigned long for the pointer instead of
a void __iomem *. For now, cast explicitly for those cases.

v2: change define to use IOMEM() like many other mach platforms per
comment from Russell.

Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
2011-10-13 14:07:38 -07:00
Bartlomiej Zolnierkiewicz
0ab3d8b321 cy82c693: fix PCI device selection
Wrong PCI device may be selected by cy82c693_set_pio_mode() if modular
IDE host drivers are used and there are additional IDE PCI devices
installed in the system.  Fix it.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-13 17:04:50 -04:00
Bartlomiej Zolnierkiewicz
acc8dbe7f4 icside: DMA support fix
Fix problem introduced by commit 5e37bdc ("ide: add struct
ide_dma_ops (take 3)"): d.dma_ops shouldn't be cleared if
we are going to use DMA.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-13 17:04:50 -04:00
Mikulas Patocka
839e7306e5 IDE: Don't powerdown Compaq Triflex IDE device on suspend
Don't powerdown Compaq Triflex IDE device on suspend

This fixes APM suspend on Compaq Armada 7400.
APM BIOS doesn't suspend if IDE is powered down when suspending.
The Triflex controller is found only on old Compaq boards,
so this patch will hopefully have no side effects.

This patch fixes a suspend regression introduced in
feb22b7f8e ("ide: add proper PCI PM
support (v2)").

Signed-off-by: Mikulas Patocka <mikulas@artax.karlin.mff.cuni.cz>
[bart: add commit's summary in parens]
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-13 17:04:49 -04:00
Andrei Emeltchenko
928abaa777 Bluetooth: AMP: read local amp info HCI command
Implementation of Read Local AMP Info Command

Signed-off-by: Andrei Emeltchenko <andrei.emeltchenko@intel.com>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
2011-10-13 17:34:16 -03:00
Pavel Shilovsky
94443f4340 CIFS: Fix incorrect max RFC1002 write size value
..the length field has only 17 bits.

Cc: <stable@kernel.org>
Acked-by: Jeff Layton <jlayton@samba.org>
Signed-off-by: Pavel Shilovsky <piastry@etersoft.ru>
Signed-off-by: Steve French <smfrench@gmail.com>
2011-10-13 15:20:14 -05:00
Andrei Emeltchenko
f89cef09ce Bluetooth: EFS: add efs option in L2CAP conf req
Add Extended Flow Specification option when building L2CAP
Configuration Request. EFS is added if both the local and
remote L2CAP entities have indicated support for the
Extended Flow Specification for BR/EDR.

...
< ACL data: handle 1 flags 0x00 dlen 10
    L2CAP(s): Info req: type 2
> ACL data: handle 1 flags 0x02 dlen 16
    L2CAP(s): Info rsp: type 2 result 0
      Extended feature mask 0x01f8
        Enhanced Retransmission mode
        Streaming mode
        FCS Option
        Extended Flow Specification
        Fixed Channels
        Extended Window Size
...
< ACL data: handle 1 flags 0x00 dlen 45
    L2CAP(s): Config req: dcid 0x0040 flags 0x00 clen 33
      RFC 0x03 (Enhanced Retransmission, TxWin 63, MaxTx 3, RTo 0, MTo 0, MPS 498)
      EFS (Id 0x01, SerType Best Effort, MaxSDU 0xffff, SDUitime 0xffffffff,
          AccLat 0xffffffff, FlushTO 0x0000ffff)
...

Based upon haijun.liu <haijun.liu@atheros.com> series of patches
(sent Sun, 22 Aug 2010)

Signed-off-by: Andrei Emeltchenko <andrei.emeltchenko@intel.com>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
2011-10-13 17:09:44 -03:00
Andrei Emeltchenko
8f7975b153 Bluetooth: EFS: assign default values in chan add
Assign default EFS values when creating L2CAP channel

Signed-off-by: Andrei Emeltchenko <andrei.emeltchenko@intel.com>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
2011-10-13 17:09:08 -03:00
Andrei Emeltchenko
5a9e7057c5 Bluetooth: EFS: definitions and headers
Define Extended Flow Specification structures and default values.
Based upon haijun.liu <haijun.liu@atheros.com> series of patches
(sent Sun, 22 Aug 2010)

Signed-off-by: Andrei Emeltchenko <andrei.emeltchenko@intel.com>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
2011-10-13 17:09:02 -03:00
Eric Dumazet
98a045d7e4 e1000e: fix skb truesize underestimation
e1000e allocates a page per skb fragment. We must account
PAGE_SIZE increments on skb->truesize, not the actual frag length.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-13 16:05:08 -04:00
Eric Dumazet
9813064677 ixgbe: fix skb truesize underestimation
ixgbe allocates half a page per skb fragment. We must account
PAGE_SIZE/2 increments on skb->truesize, not the actual frag length.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-13 16:05:08 -04:00
Eric Dumazet
95b9c1dfb7 igb: fix skb truesize underestimation
e1000 allocates half a page per skb fragment. We must account
PAGE_SIZE/2 increments on skb->truesize, not the actual frag length.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-13 16:05:07 -04:00
Eric Dumazet
ed64b3cc11 e1000: fix skb truesize underestimation
e1000 allocates a full page per skb fragment. We must account PAGE_SIZE
increments on skb->truesize, not the actual frag length.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-13 16:05:07 -04:00
Eric Dumazet
a1f4e8bcbc bnx2: fix skb truesize underestimation
bnx2 allocates a full page per fragment. We must account PAGE_SIZE
increments on skb->truesize, not the actual frag length.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-13 16:05:07 -04:00
Eric Dumazet
bdb28a97f4 be2net: fix truesize errors
Fix skb truesize underestimations of this driver.

Each frag truesize is exactly rx_frag_size bytes. (2048 bytes per
default)

A driver should not use "sizeof(struct sk_buff)" at all.

Signed-off-by: Eric Dumazet <eric.dumazet>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-13 16:05:07 -04:00
Eric Dumazet
87fb4b7b53 net: more accurate skb truesize
skb truesize currently accounts for sk_buff struct and part of skb head.
kmalloc() roundings are also ignored.

Considering that skb_shared_info is larger than sk_buff, its time to
take it into account for better memory accounting.

This patch introduces SKB_TRUESIZE(X) macro to centralize various
assumptions into a single place.

At skb alloc phase, we put skb_shared_info struct at the exact end of
skb head, to allow a better use of memory (lowering number of
reallocations), since kmalloc() gives us power-of-two memory blocks.

Unless SLUB/SLUB debug is active, both skb->head and skb_shared_info are
aligned to cache lines, as before.

Note: This patch might trigger performance regressions because of
misconfigured protocol stacks, hitting per socket or global memory
limits that were previously not reached. But its a necessary step for a
more accurate memory accounting.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Andi Kleen <ak@linux.intel.com>
CC: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-13 16:05:07 -04:00
Andrei Emeltchenko
d43cb289b0 Bluetooth: EWS: define L2CAP header sizes
Adds definitins for L2CAP header sizes to be uses when calculating
payload size instead of magic numbers.

Signed-off-by: Andrei Emeltchenko <andrei.emeltchenko@intel.com>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
2011-10-13 16:52:37 -03:00
Andrei Emeltchenko
e4ca6d9854 Bluetooth: EWS: recalculate L2CAP header size
Recalculate length of L2CAP header based on Control field length.

Signed-off-by: Andrei Emeltchenko <andrei.emeltchenko@intel.com>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
2011-10-13 16:50:14 -03:00
Hartmut Knaack
25fcf2b7f1 gpio-pca953x: fix gpio_base
gpio_base was set to 0 if no system platform data or open firmware
platform data was provided. This led to conflicts, if any other gpiochip
with a gpiobase of 0 was instantiated already. Setting it to -1 will
automatically use the first one available.

Signed-off-by: Hartmut Knaack <knaack.h@gmx.de>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
2011-10-13 13:49:37 -06:00
Andrei Emeltchenko
e37817353b Bluetooth: EWS: rewrite handling POLL (P) bit
Handle POLL (P) bit in L2CAP ERTM using information about control field type.

Signed-off-by: Andrei Emeltchenko <andrei.emeltchenko@intel.com>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
2011-10-13 16:45:39 -03:00