sfrench/cifs-2.6.git
14 years agoUCC Ether driver: kmalloc casting cleanups
Ahmed S. Darwish [Sat, 27 Jan 2007 08:00:02 +0000 (00:00 -0800)]
UCC Ether driver: kmalloc casting cleanups

A kmalloc casting cleanup patch.

Signed-off-by: Ahmed Darwish <darwish.07@gmail.com>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agosky2: handle network device allocation failure
shemminger@linux-foundation.org [Fri, 26 Jan 2007 19:38:36 +0000 (11:38 -0800)]
sky2: handle network device allocation failure

If alloc_etherdev() failed, then sky2_init_netdev will return NULL,
and sky2_probe would end up returning 0 instead of -ENOMEM.

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agosky2: software rx/tx stats
shemminger@linux-foundation.org [Fri, 26 Jan 2007 19:38:39 +0000 (11:38 -0800)]
sky2: software rx/tx stats

Maintain packet statistics in software rather than hardware.
This is slightly slower, but allows easier debugging of problems
where packets are still being received by PHY but not being handled
by hardware.

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agosky2: version 1.11.1
shemminger@linux-foundation.org [Fri, 26 Jan 2007 19:38:40 +0000 (11:38 -0800)]
sky2: version 1.11.1

Version update to 1.11.1.

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoPEGASUS: Fix typo in Corega products.
YOSHIFUJI Hideaki / 吉藤英明 [Fri, 26 Jan 2007 13:59:01 +0000 (22:59 +0900)]
PEGASUS: Fix typo in Corega products.

s/FEter/FEther/.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoASIX: Add IO-DATA ETG-US2 Support.
YOSHIFUJI Hideaki / 吉藤英明 [Fri, 26 Jan 2007 13:57:38 +0000 (22:57 +0900)]
ASIX: Add IO-DATA ETG-US2 Support.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoforcedeth: statistics optimization
Ayaz Abdulla [Tue, 23 Jan 2007 17:27:21 +0000 (12:27 -0500)]
forcedeth: statistics optimization

This patch optimizes the data paths that can support hw counters. It
removes the sw counted statistics.

This is the last patch for the optimization set. Bumping up version of
driver.

Signed-Off-By: Ayaz Abdulla <aabdulla@nvidia.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoforcedeth: statistics supported
Ayaz Abdulla [Tue, 23 Jan 2007 17:27:00 +0000 (12:27 -0500)]
forcedeth: statistics supported

This patch introduces hw statistics for older devices that supported it.
It breaks up the counters supported into separate versions.

Signed-Off-By: Ayaz Abdulla <aabdulla@nvidia.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoforcedeth: tx max work
Ayaz Abdulla [Tue, 23 Jan 2007 17:00:56 +0000 (12:00 -0500)]
forcedeth: tx max work

This patch adds a limit to how much tx work can be done in each
iteration of tx processing. If the max limit is reached, remaining tx
completions will be handled by timer interrupt.

Signed-Off-By: Ayaz Abdulla <aabdulla@nvidia.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoforcedeth: irq data path optimization
Ayaz Abdulla [Sun, 21 Jan 2007 23:10:57 +0000 (18:10 -0500)]
forcedeth: irq data path optimization

This patch optimizes the irq data paths and cleans up the code.

Signed-Off-By: Ayaz Abdulla <aabdulla@nvidia.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoforcedeth: rx data path optimization
Ayaz Abdulla [Sun, 21 Jan 2007 23:10:52 +0000 (18:10 -0500)]
forcedeth: rx data path optimization

This patch optimizes the rx data paths and cleans up the code.

Signed-Off-By: Ayaz Abdulla <aabdulla@nvidia.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoforcedeth: tx data path optimization
Ayaz Abdulla [Sun, 21 Jan 2007 23:10:47 +0000 (18:10 -0500)]
forcedeth: tx data path optimization

This patch optimizes the tx data paths and cleans up the code (removes
vlan from descr1/2 since only valid for desc3, changes to make code
easier to read, etc).

Signed-Off-By: Ayaz Abdulla <aabdulla@nvidia.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoforcedeth: tx limiting
Ayaz Abdulla [Sun, 21 Jan 2007 23:10:42 +0000 (18:10 -0500)]
forcedeth: tx limiting

This patch optimizes the logic for tx limiting. It adds a flag to check
on the completion side instead of recalculating the number of empty
slots. Also, it removes the fields that were previous used for limiting
since they have no value.

Signed-Off-By: Ayaz Abdulla <aabdulla@nvidia.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoforcedeth: optimized routines
Ayaz Abdulla [Sun, 21 Jan 2007 23:10:37 +0000 (18:10 -0500)]
forcedeth: optimized routines

This patch breaks up the routines into two versions, one for legacy
descriptor versions (ver 1 and ver 2) and one for desc ver 3. This will
make the new desc functions more leaner and further reductions will be
made in next few patches.

Signed-Off-By: Ayaz Abdulla <aabdulla@nvidia.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agobonding: update version
Jay Vosburgh [Sat, 20 Jan 2007 02:15:56 +0000 (18:15 -0800)]
bonding: update version

Update version number to reflect recent changes.

Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agobonding: modify sysfs support to permit multiple loads
Jay Vosburgh [Sat, 20 Jan 2007 02:15:47 +0000 (18:15 -0800)]
bonding: modify sysfs support to permit multiple loads

The existing code would blindly attempt to create the
bonding_masters file (in /sys/class/net) every time the module was
loaded.  When the module is loaded multiple times (which is the
historical method used by initscripts and sysconfig to create multiple
bonding interfaces), this caused load failure of the second module load
attempt, as the creation request would fail.

This changes the code to note the failure, arrange to not remove
the bonding_masters file upon module exit, and then return success.

Bonding interfaces created by the second or subsequent loads of
the module will not exist in bonding_masters.  This is not a significant
change, as previously only the interfaces from the most recent load of
the module would be listed.  Both situations are less than optimal, but
this case permits compatibility with existing distro configuration
scripts, and is consistent.

Note that previously, the sysfs create request would overwrite
the exsting bonding_masters file and succeed, allowing multiple loads of
the module.  The sysfs code has recently changed to return an error if
the file being created already exists.

Patrick McHardy <kaber@trash.net>, who reported this problem,
observed crashes on the old kernel (before sysfs checked for
duplicates).  I did not experience such crashes, but this change should
resolve them.

Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agobonding: fix error check in sysfs creation
Jay Vosburgh [Sat, 20 Jan 2007 02:15:38 +0000 (18:15 -0800)]
bonding: fix error check in sysfs creation

The existing code did not correctly handle failures to create
the per-interface sysfs group for bonding.

Modified code to notice errors, and correctly unwind.

Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agobonding: fix device name allocation error
Jay Vosburgh [Sat, 20 Jan 2007 02:15:31 +0000 (18:15 -0800)]
bonding: fix device name allocation error

The code to select names for the bonding interfaces was, for the
non-sysfs creation case, always using a hard-coded set of bond0, bond1,
etc, up to max_bonds.  This caused conflicts for the second or
subsequent loads of the module.

Changed the code to obtain device names from dev_alloc_name().

Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agonet: use bitrev8
Akinobu Mita [Tue, 19 Dec 2006 21:09:08 +0000 (13:09 -0800)]
net: use bitrev8

Use bitrev8 for bmac, mace, macmace, macsonic, and skfp drivers.

[akpm@osdl.org: use the API, not the array]
Cc: Jeff Garzik <jgarzik@pobox.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Mirko Lindner <mlindner@syskonnect.de>
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agobonding.h: "extern inline" -> "static inline"
Adrian Bunk [Tue, 12 Dec 2006 16:24:39 +0000 (17:24 +0100)]
bonding.h: "extern inline" -> "static inline"

"extern inline" generates a warning with -Wmissing-prototypes and I'm
currently working on getting the kernel cleaned up for adding this to
the CFLAGS since it will help us to avoid a nasty class of runtime
errors.

If there are places that really need a forced inline, __always_inline
would be the correct solution.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoqla3xxx: Add support for Qlogic 4032 chip.
Ron Mercer [Thu, 4 Jan 2007 00:26:08 +0000 (16:26 -0800)]
qla3xxx: Add support for Qlogic 4032 chip.

Qlogic 4032 chip is an incremental change from the 4022.
Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoremove the broken OAKNET driver
Adrian Bunk [Thu, 4 Jan 2007 18:53:30 +0000 (19:53 +0100)]
remove the broken OAKNET driver

The OAKNET driver:
- has been marked as BROKEN for more than two years and
- is still marked as BROKEN.

Drivers that had been marked as BROKEN for such a long time seem to be
unlikely to be revived in the forseeable future.

But if anyone wants to ever revive this driver, the code is still
present in the older kernel releases.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agochelsio: more rx speedup
Stephen Hemminger [Mon, 8 Jan 2007 19:26:12 +0000 (11:26 -0800)]
chelsio: more rx speedup

Cleanup receive processing some more:
   * do the reserve padding of skb during setup
   * don't pass constants to get_packet
   * do smart prefetch of skb
   * make copybreak a module parameter

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agochelsio: NAPI speed improvement
Stephen Hemminger [Mon, 8 Jan 2007 19:26:30 +0000 (11:26 -0800)]
chelsio: NAPI speed improvement

Speedup and cleanup the receive processing by eliminating the
mmio read and a lock round trip.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoAdd support for the latest 1G/10G Chelsio adapter, T3.
Divy Le Ray [Fri, 19 Jan 2007 03:04:14 +0000 (22:04 -0500)]
Add support for the latest 1G/10G Chelsio adapter, T3.

This driver is required by the Chelsio T3 RDMA driver posted by
Steve Wise.

Signed-off-by: Divy Le Ray <divy@chelsio.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agomake hdlc_setup() static again
Adrian Bunk [Thu, 11 Jan 2007 13:48:59 +0000 (14:48 +0100)]
make hdlc_setup() static again

hdlc_setup was exported, but this export was never used.

If a driver using it actually shows up it can still be exported again.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: Krzysztof Halasa <khc@pm.waw.pl>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoforcedeth: rx skb recycle
Ayaz Abdulla [Tue, 9 Jan 2007 18:30:13 +0000 (13:30 -0500)]
forcedeth: rx skb recycle

This patch removes the code that recycled the skb on error. This will
help in reducing the branches in the main data paths.

Signed-Off-By: Ayaz Abdulla <aabdulla@nvidia.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoforcedeth: tx locking
Ayaz Abdulla [Tue, 9 Jan 2007 18:30:10 +0000 (13:30 -0500)]
forcedeth: tx locking

This patch reduces the amount of code within the lock to only the
critical sections.

Signed-Off-By: Ayaz Abdulla <aabdulla@nvidia.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoforcedeth: ring access
Ayaz Abdulla [Tue, 9 Jan 2007 18:30:07 +0000 (13:30 -0500)]
forcedeth: ring access

This patch modifys ring access by using pointers. This avoids computing
the current index and avoids accessing the base address of the rings.

Signed-Off-By: Ayaz Abdulla <aabdulla@nvidia.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoforcedeth: dma access
Ayaz Abdulla [Tue, 9 Jan 2007 18:30:02 +0000 (13:30 -0500)]
forcedeth: dma access

This patch allows the hardware to fetch the tx and rx ring descriptors
with 64 bytes per access instead of 32 bytes.

Signed-Off-By: Ayaz Abdulla <aabdulla@nvidia.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoremove the broken SKMC driver
Adrian Bunk [Tue, 19 Dec 2006 21:08:48 +0000 (13:08 -0800)]
remove the broken SKMC driver

The SKMC driver has:
- already been marked as BROKEN in 2.6.0 three years ago and
- is still marked as BROKEN.

Drivers that had been marked as BROKEN for such a long time seem to be
unlikely to be revived in the forseeable future.

But if anyone wants to ever revive this driver, the code is still
present in the older kernel releases.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agodriver for Silan SC92031 netdev
Cesar Eduardo Barros [Tue, 19 Dec 2006 21:08:47 +0000 (13:08 -0800)]
driver for Silan SC92031 netdev

This is a driver for the Silan SC92031/Rsltek 8139D NIC chip.

This chip is found on at least one counterfeit Encore ENL832-TX-RENT NIC
[1], which came with a mini-CD with the 2.4 driver.  A slightly older
version of the driver was found at [2].  The main difference between them
is that the newer one has a small bugfix in the RX path, a lot of
gratuitous renaming of functions, all the printable strings changed to show
as a "Rsltek 8139D" [sic], and a PCI ID of 8139 instead of 2031.  The
driver on this patch is a rewrite of the vendor drivers (based mostly on
the older one).

Changes from the previous patch sent to netdev:
- Use MMIO instead of PIO
- Changed TX bounce buffers allocation
- Use skb_copy_and_csum_dev
- Several small bug fixes
- Tested for more than just a few minutes each time

[1] See http://www.encore-usa.com/faq.php under ENL832-TX-RENT for more
    information
[2] Look for SL_LINUX.ZIP (which is really a .tar.gz) at
    http://broadbandforum.in/dataone_Intex_LAN_cardlinux-t4207-s15.html
[3] To compile on 2.6.17, simply add back the last argument to the
    interrupt handler in two places, and copy the boolean declarations
    from 2.6.19

[akpm@osdl.org: build fixes]
Signed-off-by: Cesar Eduardo Barros <cesarb@cesarb.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agosky2: better power state management
Stephen Hemminger [Wed, 20 Dec 2006 21:06:36 +0000 (13:06 -0800)]
sky2: better power state management

Improve power management and error handling by using pci_set_power_state(),
instead of driver doing PCI PM register changes in the driver.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoremove NETIF_F_TSO ifdefery
Arjan van de Ven [Tue, 12 Dec 2006 13:06:23 +0000 (14:06 +0100)]
remove NETIF_F_TSO ifdefery

Remove the NETIF_F_TSO #ifdef-ery in drivers/net; this was
for old-old-2.4 compat (even current 2.4 has NETIF_F_TSO)
but it's time to get rid of it by now.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoSpidernet RX Debugging printout
Linas Vepstas [Wed, 13 Dec 2006 21:23:56 +0000 (15:23 -0600)]
Spidernet RX Debugging printout

Add some debugging and error printing.

The show_rx_chain() prints out the status of the rx chain,
which shows that the status of the descriptors gets
messed up after the second & subsequent RX ramfulls.

Print out contents of bad packets if error occurs.

Signed-off-by: Linas Vepstas <linas@austin.ibm.com>
Cc: James K Lewis <jklewis@us.ibm.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoSpidernet Avoid possible RX chain corruption
Linas Vepstas [Wed, 13 Dec 2006 21:23:01 +0000 (15:23 -0600)]
Spidernet Avoid possible RX chain corruption

Delete possible source of chain corruption; the hardware
already knows the location of the tail, and writing it
again is likely to mess it up.

Signed-off-by: Linas Vepstas <linas@austin.ibm.com>
Cc: James K Lewis <jklewis@us.ibm.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoSpidernet Memory barrier
Linas Vepstas [Wed, 13 Dec 2006 21:22:04 +0000 (15:22 -0600)]
Spidernet Memory barrier

Add memory barrier to make sure that the rest of the
RX descriptor state is flushed to memory before we tell
the hardware that its ready to go.

Signed-off-by: Linas Vepstas <linas@austin.ibm.com>
Cc: James K Lewis <jklewis@us.ibm.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoSpidernet RX Chain tail
Linas Vepstas [Wed, 13 Dec 2006 21:20:59 +0000 (15:20 -0600)]
Spidernet RX Chain tail

Tell the hardware the location of the rx ring tail.
More punctuation cleanup.

Signed-off-by: Linas Vepstas <linas@austin.ibm.com>
Cc: James K Lewis <jklewis@us.ibm.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoSpidernet Remove unused variable
Linas Vepstas [Wed, 13 Dec 2006 21:19:54 +0000 (15:19 -0600)]
Spidernet Remove unused variable

Remove unused variable; this makes code easier to read.
Tweak commentary.

Signed-off-by: Linas Vepstas <linas@austin.ibm.com>
Cc: James K Lewis <jklewis@us.ibm.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoSpidernet RX Refill
Linas Vepstas [Wed, 13 Dec 2006 21:18:52 +0000 (15:18 -0600)]
Spidernet RX Refill

The invocation of the rx ring refill routine is haphazard,
it can be called from a central location.

Signed-off-by: Linas Vepstas <linas@austin.ibm.com>
Cc: James K Lewis <jklewis@us.ibm.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoSpidernet Cleanup return codes
Linas Vepstas [Wed, 13 Dec 2006 21:17:39 +0000 (15:17 -0600)]
Spidernet Cleanup return codes

Simplify the somewhat convoluted use of return codes
in the rx buffer handling.

Signed-off-by: Linas Vepstas <linas@austin.ibm.com>
Cc: James K Lewis <jklewis@us.ibm.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoSpidernet another skb mem leak
Linas Vepstas [Wed, 13 Dec 2006 21:16:18 +0000 (15:16 -0600)]
Spidernet another skb mem leak

Another skb leak in an error branch. Fix this by adding
call to dev_kfree_skb_irq() after moving to a more
appropriate spot.

Signed-off-by: Linas Vepstas <linas@austin.ibm.com>
Cc: James K Lewis <jklewis@us.ibm.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoSpidernet RX skb mem leak
Linas Vepstas [Wed, 13 Dec 2006 21:15:15 +0000 (15:15 -0600)]
Spidernet RX skb mem leak

One of the unlikely error branches has an skb memory leak.
Fix this by handling the error conditions consistently.

Signed-off-by: Linas Vepstas <linas@austin.ibm.com>
Cc: James K Lewis <jklewis@us.ibm.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoSpidernet cleanup un-needed API
Linas Vepstas [Wed, 13 Dec 2006 21:12:26 +0000 (15:12 -0600)]
Spidernet cleanup un-needed API

There is no need to pass a flag into spider_net_decode_one_descr()
so remove this, and perform some othre minor cleanup.

Signed-off-by: Linas Vepstas <linas@austin.ibm.com>
Cc: James K Lewis <jklewis@us.ibm.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoSpidernet remove rxramfull tasklet
Linas Vepstas [Wed, 13 Dec 2006 21:10:06 +0000 (15:10 -0600)]
Spidernet remove rxramfull tasklet

Get rid of the rxramfull tasklet, and let the NAPI poll routine
deal with this situation. (The rxramfull interrupt is simply
stating that the h/w has run out of room for incoming packets).

Signed-off-by: Linas Vepstas <linas@austin.ibm.com>
Cc: James K Lewis <jklewis@us.ibm.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoSpidernet add net_ratelimit to suppress long output
Linas Vepstas [Wed, 13 Dec 2006 21:08:25 +0000 (15:08 -0600)]
Spidernet add net_ratelimit to suppress long output

This patch adds net_ratelimit to many of the printks in order to
limit extraneous warning messages (created in response to Bug 28554).
This patch supercedes all previous ratelimit patches.
This has been tested, please apply.

From: James K Lewis <jklewis@us.ibm.com>
Signed-off-by: James K Lewis <jklewis@us.ibm.com>
Signed-off-by: Linas Vepstas <jlinas@austin.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoSpidernet DMA coalescing
Linas Vepstas [Wed, 13 Dec 2006 21:06:59 +0000 (15:06 -0600)]
Spidernet DMA coalescing

The current driver code performs 512 DMA mappings of a bunch of
32-byte ring descriptor structures. This is silly, as they are
all in contiguous memory. This patch changes the code to
dma_map_coherent() each rx/tx ring as a whole.

Signed-off-by: Linas Vepstas <linas@austin.ibm.com>
Cc: James K Lewis <jklewis@us.ibm.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agochelsio: tabulate the update of the statistic counters
Francois Romieu [Mon, 11 Dec 2006 23:13:48 +0000 (00:13 +0100)]
chelsio: tabulate the update of the statistic counters

Let's try to avoid some code duplication.

- cxgb2
  The data are contiguous. Use plain memcpy.

- ixf1010/pm3393/vsc7326
  The cast of &mac->stats to (u64 *) is not wonderful but it is not clear
  if it is worth to add an ad-hoc union under the struct cmac_statistics.

vsc7326_reg.h suggests that more statistics could be available.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
14 years agochelsio: misc cleanups in sge
Francois Romieu [Tue, 5 Dec 2006 22:57:41 +0000 (23:57 +0100)]
chelsio: misc cleanups in sge

- duplicated code in sge::free_cmdQ_buffers ;
- NET_IP_ALIGN is already defined in (included) <linux/skbuff.h> ;
- pci_alloc_consistent() returns void * ;
- pci_alloc_consistent() returns a zeroed chunk of memory ;
- early return in restart_tx_queues.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
14 years agochelsio: useless test in cxgb2::remove_one
Francois Romieu [Tue, 5 Dec 2006 22:19:06 +0000 (23:19 +0100)]
chelsio: useless test in cxgb2::remove_one

pci_get_drvadata() is necessarily distinct from NULL if
cxgb2::init_one succeeded. cxgb2::remove_one is solely
issued through the PCI device callback.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
14 years agochelsio: useless curly braces
Francois Romieu [Mon, 11 Dec 2006 22:49:13 +0000 (23:49 +0100)]
chelsio: useless curly braces

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
14 years agochelsio: spaces, tabs and friends
Francois Romieu [Mon, 11 Dec 2006 22:47:00 +0000 (23:47 +0100)]
chelsio: spaces, tabs and friends

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
14 years agochelsio: the return statement is not a function
Francois Romieu [Mon, 11 Dec 2006 22:41:36 +0000 (23:41 +0100)]
chelsio: the return statement is not a function

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
14 years agochelsio: move return, break and continue statements on their own line
Francois Romieu [Tue, 5 Dec 2006 21:38:00 +0000 (22:38 +0100)]
chelsio: move return, break and continue statements on their own line

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
14 years ago[PATCH] ipw2200: add iwconfig rts/frag auto support
Zhu Yi [Thu, 11 Jan 2007 09:32:54 +0000 (17:32 +0800)]
[PATCH] ipw2200: add iwconfig rts/frag auto support

This patch add ipw2200 support for iwconfig rts/frag auto.

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years ago[PATCH] zd1211rw: Add ID for ZyXEL ZyAIR G-220 v2
Daniel Drake [Thu, 4 Jan 2007 03:33:54 +0000 (03:33 +0000)]
[PATCH] zd1211rw: Add ID for ZyXEL ZyAIR G-220 v2

Tested by Marijn Schouten
zd1211b chip 0586:340f v4810 high 00-13-49 AL2230_RF pa0 g---
FCC ID: I88G220V2

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years ago[PATCH] softmac: avoid assert in ieee80211softmac_wx_get_rate
John W. Linville [Wed, 3 Jan 2007 02:22:05 +0000 (21:22 -0500)]
[PATCH] softmac: avoid assert in ieee80211softmac_wx_get_rate

Unconfigured bcm43xx device can hit an assert() during wx_get_rate
queries.  This is because bcm43xx calls ieee80211softmac_start late
(i.e. during open instead of probe).

   bcm43xx_net_open ->
      bcm43xx_init_board ->
         bcm43xx_select_wireless_core ->
            ieee80211softmac_start

Fix is to check that device is running before completing
ieee80211softmac_wx_get_rate.

Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years ago[PATCH] prism54: add ethtool -i interface
Kai Engert [Tue, 12 Dec 2006 20:09:41 +0000 (21:09 +0100)]
[PATCH] prism54: add ethtool -i interface

Add support for "ethtool -i" to prism54 driver.

ethtool -i queries the specified device for
associated driver information.

This helps tools like Fedora's system-config-network to
provide GUI management of network devices.
I learned how to write this patch by reading the ipw2100
driver code.

Signed-off-by: Kai Engert <kengert@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years ago[PATCH] bcm43xx: Interrogate hardware-enable switch and update LEDs
Larry Finger [Sun, 31 Dec 2006 05:30:32 +0000 (23:30 -0600)]
[PATCH] bcm43xx: Interrogate hardware-enable switch and update LEDs

The current bcm43xx driver ignores any wireless-enable switches on mini-PCI
and mini-PCI-E cards. This patch implements a new routine to interrogate the
radio hardware enabled bit in the interface, logs the initial state and any
changes in the switch (if debugging enabled), activates the LED to show the
state, and changes the periodic work handler to provide 1 second response
to switch changes and to account for changes in the periodic work specs.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years ago[PATCH] zd1211rw: Add ID for Linksys WUSBF54G
Daniel Drake [Sat, 30 Dec 2006 22:38:23 +0000 (22:38 +0000)]
[PATCH] zd1211rw: Add ID for Linksys WUSBF54G

Tested by Henrik Hjelte
zd1211b chip 13b1:0024 v4802 high 00-14-bf AL2230_RF pa0 ----

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years ago[PATCH] Update Prism54 MAINTAINERS entry
Michael Buesch [Fri, 15 Dec 2006 20:32:44 +0000 (21:32 +0100)]
[PATCH] Update Prism54 MAINTAINERS entry

prism54-private@prism54.org bounces with

    SMTP error from remote mailer after RCPT TO:<prism54-private@prism54.org>:
    host mx1.tuxfamily.net [212.85.158.8]: 550 unknown user

developers@islsm.org seems to be the new mailing list.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years ago[PATCH] zd1211rw: Remove addressing abstraction
Daniel Drake [Tue, 12 Dec 2006 01:26:11 +0000 (01:26 +0000)]
[PATCH] zd1211rw: Remove addressing abstraction

Instead of passing our own custom 32-bit addresses around and
translating them, this patch makes all our register address constants
absolute and removes the translation.

There are two ugly parts:
 - fw_reg_addr() is needed to compute addresses of firmware registers, as this
   is dynamic based upon firmware
 - inc_addr() needs a small hack to handle byte vs word addressing

However, both of those are only small, and we don't use fw_regs a whole
lot anyway.

The bonuses here include simplicity and improved driver readability. Also, the
fact that registers are now referenced by 16-bit absolute addresses (as
opposed to 32-bit pseudo addresses) means that over 2kb compiled code size has
been shaved off.

Includes some touchups and sparse fixes from Ulrich Kunitz.

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years ago[PATCH] zd1211rw: Consistency for address space constants
Daniel Drake [Tue, 12 Dec 2006 01:25:52 +0000 (01:25 +0000)]
[PATCH] zd1211rw: Consistency for address space constants

The zd1211rw address space has confused me once too many times. This
patch introduces the following naming notation:

Memory space is split into segments (cr, fw, eeprom) and segments may
contain components (e.g. boot code inside eeprom). These names are
arbitrary and only for the description below:

x_START: Absolute address of segment start
(previously these were named such as CR_BASE_OFFSET, but they weren't
really offsets unless you were considering them as an offset to 0)

x_LEN: Segment length

x_y_LEN: Length of component y of segment x

x_y_OFFSET: Relative address of component y into segment x. The absolute
address for this component is (x_START + x_y_OFFSET)

I also renamed EEPROM registers to EEPROM data. These 'registers' can't
be written to using standard I/O and really represent predefined data
from the vendor.

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years ago[PATCH] zd1211rw: 2 new ZD1211B device ID's
Daniel Drake [Tue, 12 Dec 2006 01:25:37 +0000 (01:25 +0000)]
[PATCH] zd1211rw: 2 new ZD1211B device ID's

Philips SNU5600, tested by unibrow
zd1211b chip 0471:1236 v4810 high 00-12-bf AL2230_RF pa0 g--

SMC Ez Connect 802.11g (SMCWUSB-G), tested by Victorino Sanz Prat
zd1211b chip 083a:4505 v4810 full 00-13-f7 AL2230_RF pa0 g--N

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years ago[PATCH] zd1211rw: Generic HMAC initialization
Daniel Drake [Tue, 12 Dec 2006 01:25:13 +0000 (01:25 +0000)]
[PATCH] zd1211rw: Generic HMAC initialization

Many of the registers written during ZD1211 HMAC initialization are
duplicated exactly for ZD1211B. Move the identical ones into a generic
part, and write the hardware-specific ones separately.

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoe1000: update version to 7.3.20-k2
Auke Kok [Thu, 18 Jan 2007 17:25:33 +0000 (09:25 -0800)]
e1000: update version to 7.3.20-k2

Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
14 years agoe1000: tune our dynamic itr transmit packet accounting
Jesse Brandeburg [Thu, 18 Jan 2007 17:25:31 +0000 (09:25 -0800)]
e1000: tune our dynamic itr transmit packet accounting

The driver was still mis-calculating the number of bytes sent during
transmit, now the driver computes what appears to be exactly 100%
correct byte counts (not including CRC) when figuring out how many
bytes and frames were sent during the current transmit packet.

14 years agoe1000: clear ip csum info from context descriptor
Bruce Allan [Thu, 18 Jan 2007 17:25:28 +0000 (09:25 -0800)]
e1000: clear ip csum info from context descriptor

Since the driver sets the IP checksum insertion bit (IXSM in Status
field) in transmit context descriptors, it should clear the IP checksum
bits of any garbage so as not to confuse the hardware.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
14 years agoe1000: display flow control of link status at link up
Auke Kok [Thu, 18 Jan 2007 17:25:26 +0000 (09:25 -0800)]
e1000: display flow control of link status at link up

Print RX/TX flow control setting at link up time to display the
actual link FC properties instead of the advertised values.

Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
14 years agoe1000: fix NAPI performance on 4-port adapters
Jesse Brandeburg [Thu, 18 Jan 2007 17:25:23 +0000 (09:25 -0800)]
e1000: fix NAPI performance on 4-port adapters

This fix attempts to solve a customer (IBM) reported issue with NAPI
enabled e1000 having bad performance when transmitting simultaneously
on four ports.  The issue comes down to an interaction between NAPI,
hardware interrupt balancing, and the driver rescheduling poll on
the same processor.  Try to fix by allowing the driver to re-enable
interrupts sooner instead of polling one more time, when there was
recently all the work completed in cleanup.

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
14 years agoe1000: Fix MSI only interrupt handler routine
Jesse Brandeburg [Thu, 18 Jan 2007 17:25:21 +0000 (09:25 -0800)]
e1000: Fix MSI only interrupt handler routine

Unfortunately the read-free MSI interrupt handler needs to flush write
the icr register and thus we can't be read-free. Our MSI irq routine
thus becomes a lot more simpler since we don't need to track link state
anymore.

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
14 years agoe1000: clean up debug output defines
Auke Kok [Thu, 18 Jan 2007 17:25:18 +0000 (09:25 -0800)]
e1000: clean up debug output defines

Remove unused MSGOUT macro and add "\n" to function debug output.

Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
14 years agoe1000: simplify case handling gigabit at half duplex
Jesse Brandeburg [Thu, 18 Jan 2007 17:25:15 +0000 (09:25 -0800)]
e1000: simplify case handling gigabit at half duplex

Remvoe duplicate code handling erraneous user supplied wrong case
of gigabit speed with half duplex.

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
14 years agoLinux 2.6.20 v2.6.20
Linus Torvalds [Sun, 4 Feb 2007 18:44:54 +0000 (10:44 -0800)]
Linux 2.6.20

14 years ago[PATCH] EFI x86: pass firmware call parameters on the stack
Frédéric Riss [Tue, 30 Jan 2007 20:41:17 +0000 (21:41 +0100)]
[PATCH] EFI x86: pass firmware call parameters on the stack

When calling into the EFI firmware, the parameters need to be passed on
the stack. The recent change to use -mregparm=3 breaks x86 EFI support.
This patch is needed to allow the new Intel-based Macs to suspend to ram
(efi.get_time is called during the suspend phase).

Signed-off-by: Frederic Riss <frederic.riss@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years ago[PATCH] fix rtl8150
Al Viro [Sun, 4 Feb 2007 03:02:17 +0000 (03:02 +0000)]
[PATCH] fix rtl8150

That code doesn't do what its author apparently thought it would do...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6
Linus Torvalds [Sat, 3 Feb 2007 19:26:39 +0000 (11:26 -0800)]
Merge /linux/kernel/git/jejb/scsi-rc-fixes-2.6

* master.kernel.org:/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6:
  [SCSI] sd: udev accessing an uninitialized scsi_disk field results in a crash
  [SCSI] st: A MTIOCTOP/MTWEOF within the early warning will cause the file number to be incorrect
  [SCSI] qla4xxx: bug fixes
  [SCSI] Fix scsi_add_device() for async scanning

14 years ago[PATCH] x86-64: define dma noncoherent API functions
Jeff Garzik [Sat, 3 Feb 2007 09:14:03 +0000 (01:14 -0800)]
[PATCH] x86-64: define dma noncoherent API functions

x86-64 is missing these:

Signed-off-by: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years ago[PATCH] Altix: more ACPI PRT support
John Keller [Sat, 3 Feb 2007 09:14:02 +0000 (01:14 -0800)]
[PATCH] Altix: more ACPI PRT support

The SN Altix platform does not conform to the IOSAPIC IRQ routing model.
Add code in acpi_unregister_gsi() to check if (acpi_irq_model ==
ACPI_IRQ_MODEL_PLATFORM) and return.

Due to an oversight, this code was not added previously when
similar code was added to acpi_register_gsi().

http://marc.theaimsgroup.com/?l=linux-acpi&m=116680983430121&w=2

Signed-off-by: John Keller <jpk@sgi.com>
Acked-by: Len Brown <lenb@kernel.org>
Cc: "Luck, Tony" <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years ago[PATCH] revert blockdev direct io back to 2.6.19 version
Andrew Morton [Sat, 3 Feb 2007 09:14:01 +0000 (01:14 -0800)]
[PATCH] revert blockdev direct io back to 2.6.19 version

Andrew Vasquez is reporting as-iosched oopses and a 65% throughput
slowdown due to the recent special-casing of direct-io against
blockdevs.  We don't know why either of these things are occurring.

The patch minimally reverts us back to the 2.6.19 code for a 2.6.20
release.

Cc: Andrew Vasquez <andrew.vasquez@qlogic.com>
Cc: Ken Chen <kenchen@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years ago[PATCH] alpha: fix epoll syscall enumerations
Mike Frysinger [Sat, 3 Feb 2007 09:13:55 +0000 (01:13 -0800)]
[PATCH] alpha: fix epoll syscall enumerations

We went and named them __NR_sys_foo instead of __NR_foo.

It may be too late to change this, but we can at least add the proper names
now.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years ago[PATCH] net/smc911x: match up spin lock/unlock
Peter Korsgaard [Sat, 3 Feb 2007 09:13:50 +0000 (01:13 -0800)]
[PATCH] net/smc911x: match up spin lock/unlock

smc911x_phy_configure's error handling unconditionally unlocks the
spinlock even if it wasn't locked. Patch fixes it.

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Cc: Jeff Garzik <jeff@garzik.org>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years ago[PATCH] kexec: Avoid migration of already disabled irqs (ia64)
Magnus Damm [Sat, 3 Feb 2007 09:13:48 +0000 (01:13 -0800)]
[PATCH] kexec: Avoid migration of already disabled irqs (ia64)

This patch fixes up ia64 kexec support for HP rx2620 hardware.  It does
this by skipping migration of already disabled irqs.  This is most likely a
problem on other ia64 platforms as well, but I've only been able to
reproduce it on one machine so far.

The full story is that handle_bad_irq() gets invoked before starting the
new kernel without this patch.  This seems to happen when fixup_irqs()
calls generic_handle_irq() on already migrated (and disabled) irqs.  So by
avoiding migration of disabled irqs we stay away of handle_bad_irq().

The code has been tested on three different ia64 machines, all with good
results.  It is possible to trigger the same bug by offlining a processor
using echo 0 > /sys/devices/system/cpu/cpuX/online.

More detailed information is available in the following mail thread:
http://lists.osdl.org/pipermail/fastboot/2007-January/thread.html#5774

Signed-off-by: Magnus Damm <magnus@valinux.co.jp>
Acked-by: Simon Horman <horms@verge.net.au>
Acked-by: Zou, Nanhai <nanhai.zou@intel.com>
Acked-by: Jay Lan <jlan@sgi.com>
Acked-by: "Luck, Tony" <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years ago[PATCH] aio: fix buggy put_ioctx call in aio_complete - v2
Ken Chen [Sat, 3 Feb 2007 09:13:45 +0000 (01:13 -0800)]
[PATCH] aio: fix buggy put_ioctx call in aio_complete - v2

An AIO bug was reported that sleeping function is being called in softirq
context:

BUG: warning at kernel/mutex.c:132/__mutex_lock_common()
Call Trace:
     [<a000000100577b00>] __mutex_lock_slowpath+0x640/0x6c0
     [<a000000100577ba0>] mutex_lock+0x20/0x40
     [<a0000001000a25b0>] flush_workqueue+0xb0/0x1a0
     [<a00000010018c0c0>] __put_ioctx+0xc0/0x240
     [<a00000010018d470>] aio_complete+0x2f0/0x420
     [<a00000010019cc80>] finished_one_bio+0x200/0x2a0
     [<a00000010019d1c0>] dio_bio_complete+0x1c0/0x200
     [<a00000010019d260>] dio_bio_end_aio+0x60/0x80
     [<a00000010014acd0>] bio_endio+0x110/0x1c0
     [<a0000001002770e0>] __end_that_request_first+0x180/0xba0
     [<a000000100277b90>] end_that_request_chunk+0x30/0x60
     [<a0000002073c0c70>] scsi_end_request+0x50/0x300 [scsi_mod]
     [<a0000002073c1240>] scsi_io_completion+0x200/0x8a0 [scsi_mod]
     [<a0000002074729b0>] sd_rw_intr+0x330/0x860 [sd_mod]
     [<a0000002073b3ac0>] scsi_finish_command+0x100/0x1c0 [scsi_mod]
     [<a0000002073c2910>] scsi_softirq_done+0x230/0x300 [scsi_mod]
     [<a000000100277d20>] blk_done_softirq+0x160/0x1c0
     [<a000000100083e00>] __do_softirq+0x200/0x240
     [<a000000100083eb0>] do_softirq+0x70/0xc0

See report: http://marc.theaimsgroup.com/?l=linux-kernel&m=116599593200888&w=2

flush_workqueue() is not allowed to be called in the softirq context.
However, aio_complete() called from I/O interrupt can potentially call
put_ioctx with last ref count on ioctx and triggers bug.  It is simply
incorrect to perform ioctx freeing from aio_complete.

The bug is trigger-able from a race between io_destroy() and aio_complete().
A possible scenario:

cpu0                               cpu1
io_destroy                         aio_complete
  wait_for_all_aios {                __aio_put_req
     ...                                 ctx->reqs_active--;
     if (!ctx->reqs_active)
        return;
  }
  ...
  put_ioctx(ioctx)

                                     put_ioctx(ctx);
                                        __put_ioctx
                                          bam! Bug trigger!

The real problem is that the condition check of ctx->reqs_active in
wait_for_all_aios() is incorrect that access to reqs_active is not
being properly protected by spin lock.

This patch adds that protective spin lock, and at the same time removes
all duplicate ref counting for each kiocb as reqs_active is already used
as a ref count for each active ioctx.  This also ensures that buggy call
to flush_workqueue() in softirq context is eliminated.

Signed-off-by: "Ken Chen" <kenchen@google.com>
Cc: Zach Brown <zach.brown@oracle.com>
Cc: Suparna Bhattacharya <suparna@in.ibm.com>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Cc: <stable@kernel.org>
Acked-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years ago[NETFILTER]: nf_conntrack_h323: fix compile error with CONFIG_IPV6=m, CONFIG_NF_CONNT...
Adrian Bunk [Sat, 3 Feb 2007 03:33:52 +0000 (19:33 -0800)]
[NETFILTER]: nf_conntrack_h323: fix compile error with CONFIG_IPV6=m, CONFIG_NF_CONNTRACK_H323=y

Fix this by letting NF_CONNTRACK_H323 depend on (IPV6 || IPV6=n).

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years ago[NETFILTER]: ctnetlink: fix compile failure with NF_CONNTRACK_MARK=n
Patrick McHardy [Sat, 3 Feb 2007 03:33:11 +0000 (19:33 -0800)]
[NETFILTER]: ctnetlink: fix compile failure with NF_CONNTRACK_MARK=n

  CC      net/netfilter/nf_conntrack_netlink.o
net/netfilter/nf_conntrack_netlink.c: In function 'ctnetlink_conntrack_event':
net/netfilter/nf_conntrack_netlink.c:392: error: 'struct nf_conn' has no member named 'mark'
make[3]: *** [net/netfilter/nf_conntrack_netlink.o] Error 1

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years ago[SCSI] sd: udev accessing an uninitialized scsi_disk field results in a crash
Nagendra Singh Tomar [Fri, 2 Feb 2007 12:04:56 +0000 (17:34 +0530)]
[SCSI] sd: udev accessing an uninitialized scsi_disk field results in a crash

sd_probe() calls class_device_add() even before initializing the
sdkp->device variable. class_device_add() eventually results in the user mode
udev program to be called. udev program can read the the allow_restart
attribute of the newly created scsi device. This is resulting in a crash as
the show function for allow_restart (i.e sd_show_allow_restart) returns the
attribute value by reading the sdkp->device->allow_restart variable. As the
sdkp->device is not initialized before calling the user mode hotplug helper,
this results in a crash.
The patch below solves it by calling class_device_add() only after the
necessary fields in the scsi_disk structure are initialized properly.

Signed-off-by: Nagendra Singh Tomar <nagendra_tomar@adaptec.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
14 years agoMerge branch 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik...
Linus Torvalds [Fri, 2 Feb 2007 17:14:48 +0000 (09:14 -0800)]
Merge branch 'upstream-linus' of /linux/kernel/git/jgarzik/libata-dev

* 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/libata-dev:
  libata: Initialize nbytes for internal sg commands
  libata: Fix ata_busy_wait() kernel docs
  pata_via: Correct missing comments
  pata_atiixp: propogate cable detection hack from drivers/ide to the new driver
  ahci/pata_jmicron: fix JMicron quirk

14 years agolibata: Initialize nbytes for internal sg commands
Brian King [Tue, 30 Jan 2007 17:32:26 +0000 (11:32 -0600)]
libata: Initialize nbytes for internal sg commands

Some LLDDs, like ipr, use nbytes and pad_len to determine
the total data transfer length of a command. Make sure
nbytes gets initialized for internally generated commands.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agolibata: Fix ata_busy_wait() kernel docs
Alan [Wed, 31 Jan 2007 17:47:24 +0000 (17:47 +0000)]
libata: Fix ata_busy_wait() kernel docs

> Looks like you should use ata_busy_wait() here, rather than reproducing
> the same code again.

It waits in 10uS chunks while 1uS chunks were used in the workaround.
Could indeed do that once I know the fix is right. While I'm at it the
ata_busy_wait kerneldoc is borked so here's a fix

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agopata_via: Correct missing comments
Alan [Wed, 31 Jan 2007 17:14:38 +0000 (17:14 +0000)]
pata_via: Correct missing comments

The 8237S was added to the chipsets but not to the comments. Fix this

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agopata_atiixp: propogate cable detection hack from drivers/ide to the new driver
Alan [Wed, 31 Jan 2007 17:10:46 +0000 (17:10 +0000)]
pata_atiixp: propogate cable detection hack from drivers/ide to the new driver

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoahci/pata_jmicron: fix JMicron quirk
Tejun Heo [Fri, 2 Feb 2007 05:51:09 +0000 (14:51 +0900)]
ahci/pata_jmicron: fix JMicron quirk

For all JMicrons except for 361 and 368, AHCI mode enable bits in the
Control(1) should be set.  This used to be done in both ahci and
pata_jmicron but while moving programming to PCI quirk, it was removed
from ahci part while still left in pata_jmicron.

The implemented JMicron PCI quirk was incorrect in that it didn't
program AHCI mode enable bits.  If pata_jmicron is loaded first and
programs those bits, the ahci ports work; otherwise, ahci device
detection fails miserably.

This patch makes JMicron PCI quirk clear SATA IDE mode bits and set
AHCI mode bits and remove the respective part from pata_jmicron.
Tested on JMB361, 363 and 368.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
14 years agoMerge branch 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik...
Linus Torvalds [Fri, 2 Feb 2007 16:13:23 +0000 (08:13 -0800)]
Merge branch 'upstream-linus' of /linux/kernel/git/jgarzik/netdev-2.6

* 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6:
  spidernet : fix memory leak in spider_net_stop
  e100: fix napi ifdefs removing needed code
  netxen patches

14 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/davem/bnx2-2.6
Linus Torvalds [Fri, 2 Feb 2007 16:10:58 +0000 (08:10 -0800)]
Merge /pub/scm/linux/kernel/git/davem/bnx2-2.6

* master.kernel.org:/pub/scm/linux/kernel/git/davem/bnx2-2.6:
  [BNX2]: PHY workaround for 5709 A0.

14 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Fri, 2 Feb 2007 16:10:30 +0000 (08:10 -0800)]
Merge /pub/scm/linux/kernel/git/davem/net-2.6

* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6:
  [NET_SCHED]: act_ipt: fix regression in ipt action

14 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/davem/sparc-2.6
Linus Torvalds [Fri, 2 Feb 2007 16:10:17 +0000 (08:10 -0800)]
Merge /pub/scm/linux/kernel/git/davem/sparc-2.6

* master.kernel.org:/pub/scm/linux/kernel/git/davem/sparc-2.6:
  [SPARC32]: Fix over-optimization by GCC near ip_fast_csum.

14 years ago[PATCH] MAINTAINERS: ufs entry
Evgeniy Dushistov [Fri, 2 Feb 2007 08:36:34 +0000 (11:36 +0300)]
[PATCH] MAINTAINERS: ufs entry

Mark ufs file system as maintainable, and add me as maintainer,
to help people find appropriate person to assign bugs.

Signed-off-by: Evgeniy Dushistov <dushistov@mail.ru>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years agoRevert "[PATCH] fix typo in geode_configre()@cyrix.c"
Linus Torvalds [Fri, 2 Feb 2007 16:07:42 +0000 (08:07 -0800)]
Revert "[PATCH] fix typo in geode_configre()@cyrix.c"

This reverts commit e4f0ae0ea63caceff37a13f281a72652b7ea71ba.

It's not wrong, but it's not right either, and everybody seems to agree
that the right fix is probably to do the ccr3 write after the ccr4 one
(and that we also should clean it up a bit).  And after that we need to
really validate that all the bits that we write to ccr4 actually do
work.

The old 2.6.19 code was insane, and basically didn't change ccr4 at all
(even though it certainly looks like it was the *intent* to do so).  So
let's revert the change that may fix things, just because it's not what
was actually ever tested when the code was written, even if it _was_ the
intent.

There's a discussion on http://lkml.org/lkml/2007/1/9/63 that was
started by the patch that now gets reverted, and that discussion may
well contain the proper long-term fix.

Suggested-by: Adrian Bunk <bunk@stusta.de>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years agospidernet : fix memory leak in spider_net_stop
Jens Osterkamp [Thu, 1 Feb 2007 11:07:47 +0000 (12:07 +0100)]
spidernet : fix memory leak in spider_net_stop

We forget to call spider_net_free_rx_chain_contents which does the
actual dev_kfree_skb. New skbs are allocated from skbuff_head_cache
on each "ifconfig up" letting the cache grow infinitely.

This patch fixes it.

Signed-off-by: Jens Osterkamp <jens@de.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>