platform/kernel/linux-starfive.git
17 years agogit-libata-all-ipr-fix
Andrew Morton [Mon, 26 Mar 2007 09:17:43 +0000 (02:17 -0700)]
git-libata-all-ipr-fix

drivers/scsi/ipr.c: In function '__ipr_eh_dev_reset':
drivers/scsi/ipr.c:3865: warning: passing argument 4 of 'ata_do_eh' from incompatible pointer type

Cc: Tejun Heo <htejun@gmail.com>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agolibata: hardreset on SERR_INTERNAL
Tejun Heo [Tue, 13 Mar 2007 16:20:51 +0000 (01:20 +0900)]
libata: hardreset on SERR_INTERNAL

There was a rare report where SB600 reported SERR_INTERNAL and SRST
couldn't get it out of the failure mode.  Hardreset on SERR_INTERNAL.
As the problem is intermittent, whether this fixes the problem or not
hasn't been verified yet, but hardresetting the channel on internal
error is a good idea anyway.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agosata_promise: decode and report error reasons
Mikael Pettersson [Wed, 14 Mar 2007 08:51:35 +0000 (09:51 +0100)]
sata_promise: decode and report error reasons

This patch adds much needed error reason decoding and
reporting to sata_promise. It's simplistic but should
log all relevant error info the controller provides.

Signed-off-by: Mikael Pettersson <mikpe@it.uu.se>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agosata_promise: separate SATA and PATA ops
Mikael Pettersson [Sun, 11 Mar 2007 20:20:43 +0000 (21:20 +0100)]
sata_promise: separate SATA and PATA ops

This patch changes sata_promise so that the PATA ports
on TX2plus chips are bound to the pdc_pata_ops structure.
This means that operations called from the SATA ops
structures don't need any SATA-vs-PATA tests any more.
Instead, operations that depend on a port being SATA or
PATA are separated into different procedures.

* pdc_cable_type() is split into a PATA version and a
  SATA version
* pdc_error_handler() is split into a PATA version and a
  SATA version, that both call a common version after
  setting up the `hardreset' function pointer
* pdc_old_check_atapi_dma() is now only used for SATAI
  ports, so is renamed to pdc_old_sata_check_atapi_dma()
  and simplified
* pdc_sata_scr_{read,write}() are now only used for SATA
  ports, so their is-not-SATA tests are removed
* pdc_port_start() is split into three procedures: a wrapper
  which performs the ->ops adjustment on TX2plus PATA ports,
  a procedure with the common code, and a procedure with
  the SATA-specific code (this bit might be cleaned up by
  Tejun's new init model)

Tested on 20619, 20575, and 20775 chips.

Signed-off-by: Mikael Pettersson <mikpe@it.uu.se>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agosata_promise: add missing cable_detect hooks
Mikael Pettersson [Sun, 11 Mar 2007 20:19:27 +0000 (21:19 +0100)]
sata_promise: add missing cable_detect hooks

The recent change which moved cable detection from
pdc_pre_reset() to the new ->cable_detect hook only
added the hook for SATAII chips, leaving SATAI chips
and the 20619 without the hook. Fixed by this patch.

Signed-off-by: Mikael Pettersson <mikpe@it.uu.se>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agolibata/IDE: remove combined mode quirk
Jeff Garzik [Fri, 9 Mar 2007 15:54:42 +0000 (10:54 -0500)]
libata/IDE: remove combined mode quirk

Both old-IDE and libata should be able handle all controllers and
devices found using normal resource reservation methods.

This eliminates the awful, low-performing split-driver configuration
where old-IDE drove the PATA portion of a PCI device, in PIO-only mode,
and libata drove the SATA portion of the /same/ PCI device, in DMA mode.
Typically vendors would ship SATA hard drive / PATA optical
configuration, which would lend itself to slow (PIO-only) CD-ROM
performance.

For Intel users running in combined mode, it is now wholly dependent on
your driver choice (potentially link order, if you compile both drivers
in) whether old-IDE or libata will drive your hardware.

In either case, you will get full performance from both SATA and PATA
ports now, without having to pass a kernel command line parameter.

Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years ago[libata] turn on !IORDY filter
Jeff Garzik [Fri, 9 Mar 2007 14:56:46 +0000 (09:56 -0500)]
[libata] turn on !IORDY filter

The previous commit erroneously noted that the !IORDY filter was turned
on.  No true, that change was split out into this commit.

Originally authored and signed-off-by Alan Cox.

Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years ago[libata] sata_mv: support ->cable_detect
Jeff Garzik [Fri, 9 Mar 2007 14:46:47 +0000 (09:46 -0500)]
[libata] sata_mv: support ->cable_detect

Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agolibata: Change prototype of mode_filter to remove ata_port*
Alan Cox [Fri, 9 Mar 2007 14:34:07 +0000 (09:34 -0500)]
libata: Change prototype of mode_filter to remove ata_port*

With Tejun having added adev->ap some time ago we can get rid of the
almost unused port being passed to mode filters. And while we are
doing filters, lets turn on the !IORDY filter as well.

Signed-off-by: Alan Cox <alan@redhat.com>
With some hand massaging from
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agosata_sil: First step to removing ->post_set_mode
Alan Cox [Thu, 8 Mar 2007 23:09:12 +0000 (23:09 +0000)]
sata_sil: First step to removing ->post_set_mode

Now that we have ata_do_set_mode() available for drivers to use we don't
actually need ->post_set_mode() as the driver can wrap set_mode nicely
and do stuff before or after (eg PCMCIA needs before), so we can kill off
a method in all the structs

While I was at it I added kernel-doc to the function involved.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agolibata-core: Fix the iordy methods
Alan Cox [Thu, 8 Mar 2007 23:22:59 +0000 (23:22 +0000)]
libata-core: Fix the iordy methods

This alone isn't sufficient to save the universe from prehistoric disks
and controllers but it is a first important step. Split off a separate
function to provide a mode filter when controller iordy is not available.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agopata: expose set_mode method so it can be wrapped
Alan [Tue, 6 Mar 2007 10:37:52 +0000 (02:37 -0800)]
pata: expose set_mode method so it can be wrapped

This splits set_mode into do_set_mode and the wrapper so that a driver can
call the standard method inside its own.  This in theory also obsoletes
->post_set_mode().

Signed-off-by: Alan Cox <alan@redhat.com>
Cc: Jeff Garzik <jeff@garzik.org>
Cc: Tejun Heo <htejun@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agopata_hpt37x: Updates from drivers/ide work
Alan Cox [Thu, 8 Mar 2007 23:28:52 +0000 (23:28 +0000)]
pata_hpt37x: Updates from drivers/ide work

Drag pata_hpt37x kicking and screaming in the direction of
drivers/ide/pci/hpt366.c and all the work that Sergei has been doing
there. Plenty left to be done but this is a good snapshot for folks to
work on and to review

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoiomap: implement pcim_iounmap_regions()
Tejun Heo [Fri, 9 Mar 2007 10:45:58 +0000 (19:45 +0900)]
iomap: implement pcim_iounmap_regions()

Implement pcim_iounmap_regions() - the opposite of
pcim_iomap_regions().

Signed-off-by: Tejun heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years ago[libata] More PATA driver ->cable_detect support
Jeff Garzik [Fri, 9 Mar 2007 13:37:46 +0000 (08:37 -0500)]
[libata] More PATA driver ->cable_detect support

Roll-up of ->cable_detect feature addition patches, authored and
signed-off-by Alan Cox.

Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agopata_hpt366: support ->cable_detect
Alan Cox [Thu, 8 Mar 2007 19:34:28 +0000 (19:34 +0000)]
pata_hpt366: support ->cable_detect

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agopata_efar: support ->cable_detect
Alan Cox [Thu, 8 Mar 2007 19:33:29 +0000 (19:33 +0000)]
pata_efar: support ->cable_detect

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agopata_atiixp: support ->cable_detect
Alan Cox [Thu, 8 Mar 2007 19:27:31 +0000 (19:27 +0000)]
pata_atiixp: support ->cable_detect

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agopata_radisys: support cable_detect
Alan Cox [Thu, 8 Mar 2007 22:56:07 +0000 (22:56 +0000)]
pata_radisys: support cable_detect

Signed-off-by: Alan Cox <number6@the-village.bc.nu>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agopata_sc1200: restore cable type
Alan Cox [Thu, 8 Mar 2007 22:58:03 +0000 (22:58 +0000)]
pata_sc1200: restore cable type

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agopata_rz1000: support cable_detect
Alan Cox [Thu, 8 Mar 2007 22:57:03 +0000 (22:57 +0000)]
pata_rz1000: support cable_detect

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agopata_platform: Add cable_detect method
Alan Cox [Thu, 8 Mar 2007 22:54:04 +0000 (22:54 +0000)]
pata_platform: Add cable_detect method

Signed-off-by: Alan Cox <number6@the-village.bc.nu>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agosata_promise: Switch to cable method, clean up some bits as a result
Alan Cox [Thu, 8 Mar 2007 23:06:47 +0000 (23:06 +0000)]
sata_promise: Switch to cable method, clean up some bits as a result

Signed-off-by: Alan Cox <number6@the-village.bc.nu>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agolibata-core: fix comments on cable type
Alan Cox [Thu, 8 Mar 2007 23:19:19 +0000 (23:19 +0000)]
libata-core: fix comments on cable type

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agolibata: Restore Kconfig updated experimental levels and correct
Alan Cox [Thu, 8 Mar 2007 23:24:49 +0000 (23:24 +0000)]
libata: Restore Kconfig updated experimental levels and correct

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agopata_cmd640: Multiple updates
Alan Cox [Wed, 7 Mar 2007 16:43:29 +0000 (16:43 +0000)]
pata_cmd640: Multiple updates

Fix suspend/resume support
Write 0x5B to 0 not 0x5C

The former is important as we must kill the FIFO on a resume

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agopata_via: Use cable_detect method
Alan Cox [Wed, 7 Mar 2007 16:56:54 +0000 (16:56 +0000)]
pata_via: Use cable_detect method

We end up shifting a few bits of logic around in this driver but the
basic change is the same.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agopata_sis: Clean up using cable_detect methods
Alan Cox [Wed, 7 Mar 2007 16:54:24 +0000 (16:54 +0000)]
pata_sis: Clean up using cable_detect methods

This changeset revolves around the fact that all the SiS controllers have
the same enable bits, but differing cable detection methods. Previously
that meant each type had its own error_handler methods. Instead we can
now implement different ->cable_detect methods and share a single
error_handler which does the filtering by enable bits.

In addition we had some auto const arrays that should be static const. I'm
not sure if gcc already treats them intelligently but adding the static
will make sure.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agopata_marvell: Cable and reset fixes
Alan Cox [Wed, 7 Mar 2007 16:48:09 +0000 (16:48 +0000)]
pata_marvell: Cable and reset fixes

There are two changes here. Firstly we switch to a cable detect method,
secondly the old code forgot to call ata_std_prereset() but somehow
managed to work anyway. Fix the missing call.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agopata_it8213: Cable detect
Alan Cox [Wed, 7 Mar 2007 16:46:20 +0000 (16:46 +0000)]
pata_it8213: Cable detect

Another not-quite PIIX, another cable type conversion

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years ago[libata] Update several PATA drivers for new ->cable_detect hook
Jeff Garzik [Fri, 9 Mar 2007 12:24:15 +0000 (07:24 -0500)]
[libata] Update several PATA drivers for new ->cable_detect hook

All patches authored and signed-off-by Alan Cox, sent on Mar 7, 2007.
I merely combined them all into a single patch.

Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoSupport for Marvell 7042 Chip
Morrison, Tom [Tue, 6 Mar 2007 10:38:10 +0000 (02:38 -0800)]
Support for Marvell 7042 Chip

Added Support for Marvell 7042 Chip - 7042 has same capabilities & behavior
as 6042.

Signed-off-by: Thomas A. Morrison <tmorrison@empirix.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agopata_cmd640: CMD640 PCI support
Alan Cox [Tue, 6 Mar 2007 10:38:11 +0000 (02:38 -0800)]
pata_cmd640: CMD640 PCI support

Support for the PCI CMD640 (not VLB)

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agolibata: cable detection fixes
Alan Cox [Tue, 6 Mar 2007 10:37:56 +0000 (02:37 -0800)]
libata: cable detection fixes

2.6.21-rc has horrible problems with libata and PATA cable types (and
thus speeds). This occurs because Tejun fixed a pile of other bugs and
we now do cable detect enforcement for drive side detection properly.

Unfortunately we don't do the process around cable detection right. Tejun
identified the problem and pointed to the right Annex in the spec, this patch
implements the rest of the needed changes.

We add a ->cable_detect() method called after the identify
sequence which allows a host to do host side detection at this point
should it wish, or to modify the results of the drive side identify.

This separate ->cable_detect method also cleans up a lot of code because
many drivers have their own error_handler methods which really just set
the cable type.

If there is no ->cable_detect method the cable type is left alone so a
driver setting it earlier (eg because it has the SATA flags set or
because it uses the old error_handler approach) will still do the right
thing (or at least the same thing) as before.

This patch simply adds the cable_detect method and helpers it doesn't use
them but other follow up patches will (ie Adrian please don't submit
patches to unexport them ;))

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agolibata: handle ata_pci_device_do_resume() failure while resuming
Dmitriy Monakhov [Tue, 6 Mar 2007 10:37:54 +0000 (02:37 -0800)]
libata: handle ata_pci_device_do_resume() failure while resuming

Since commit:553c4aa630af7bc885e056d0436e4eb7f238579b
ata_pci_device_do_resume() can return error code, all callers was updated
except this one.

Signed-off-by: Monakhov Dmitriy <dmonakhov@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agolibata: warn if speed limited due to 40-wire cable
Robert Hancock [Tue, 6 Mar 2007 10:37:51 +0000 (02:37 -0800)]
libata: warn if speed limited due to 40-wire cable

Warn the user if a drive's transfer rate is limited because of a 40-wire
cable detection.

Signed-off-by: Robert Hancock <hancockr@shaw.ca>
Cc: Jeff Garzik <jeff@garzik.org>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoRESEND: libata: check cdb len per dev instead of per host
Mark Lord [Fri, 2 Feb 2007 17:36:25 +0000 (12:36 -0500)]
RESEND: libata: check cdb len per dev instead of per host

Resending, with s/printk/DPRINTK/ as pointed out by Alan.

Fix libata to perform CDB len validation per device
rather than per host.  This way, validation still works
when we have a mix of 12-byte and 16-byte devices on
a common host interface.

Signed-off-by: Mark Lord <mlord@pobox.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agolibata: dev_config does not need ap and adev passing
Alan [Fri, 2 Mar 2007 00:56:15 +0000 (00:56 +0000)]
libata: dev_config does not need ap and adev passing

It used to be impossible to get from ata_device to ata_port but that is
no longer true. Various methods have been cleaned up over time but
dev_config still takes both and most users don't need both anyway. Tidy
this one up

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years ago[libata] sata_mv: clean up DMA boundary issues, turn on 64-bit DMA
Jeff Garzik [Mon, 26 Feb 2007 06:26:06 +0000 (01:26 -0500)]
[libata] sata_mv: clean up DMA boundary issues, turn on 64-bit DMA

The chips covered by sata_mv have a 32-bit DMA boundary they must not
cross, not a 64K boundary.  We are merely limited to a 64K maximum
segment size.  Therefore, the DMA scatter/gather table fill code can be
greatly simplified, and we need not cut in half the S/G table size as
reported to the SCSI layer.

Also, the driver forget to turn on 64-bit DMA at the PCI layer.  All
other data structures (both hardware and software) have been prepped for
64-bit PCI DMA.  It was simply never turned on.  <fingers crossed> let's
see if it still works...

Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years ago[libata] export sata_print_link_status()
Jeff Garzik [Sun, 25 Feb 2007 21:50:52 +0000 (16:50 -0500)]
[libata] export sata_print_link_status()

To be used in sata_mv's exception handling code, and overall is a
generally useful function.

Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years ago[libata] sata_mv: remove extra braces
Jeff Garzik [Sun, 25 Feb 2007 02:26:42 +0000 (21:26 -0500)]
[libata] sata_mv: remove extra braces

Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agosmc911x: fix compilation breakage wjen debug is on
Vitaly Wool [Fri, 27 Apr 2007 10:42:09 +0000 (14:42 +0400)]
smc911x: fix compilation breakage wjen debug is on

the patch below fixes compilation breakage of smc911x driver when
ENABLE_SMC_DEBUG_PKTS equals to 1.

Signed-off-by: Vitaly Wool <vitalywool@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years ago[netdrvr] eexpress: minor corrections
Shani Moideen [Sat, 28 Apr 2007 15:05:43 +0000 (11:05 -0400)]
[netdrvr] eexpress: minor corrections

A few corrections related to time_after and time_before in
drivers/net/eexpress.c as suggested by Marcin slusarz.

Signed-off-by: Shani Moideen <shani.moideen@wipro.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoadd NAPI support to sb1250-mac.c
Mark Mason [Thu, 26 Apr 2007 07:23:22 +0000 (00:23 -0700)]
add NAPI support to sb1250-mac.c

Patch to add NAPI support to sb1250-mac.c (rev 2).  This patch differs from
the last in that the NAPI support isn't marked as experimental, nor is it
configurable (ie.  once applied - NAPI is enabled all the time).  This was
based on feedback from Ralf and others.

Signed-off-by: Mark Mason <mason@broadcom.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoixgb: ROUND_UP macro cleanup in drivers/net/ixgb
Milind Arun Choudhary [Fri, 27 Apr 2007 20:55:31 +0000 (13:55 -0700)]
ixgb: ROUND_UP macro cleanup in drivers/net/ixgb

IXGB_ROUNDUP macro cleanup ,use ALIGN

Signed-off-by: Milind Arun Choudhary <milindchoudhary@gmail.com>
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoe1000: ROUND_UP macro cleanup in drivers/net/e1000
Milind Arun Choudhary [Fri, 27 Apr 2007 20:55:29 +0000 (13:55 -0700)]
e1000: ROUND_UP macro cleanup in drivers/net/e1000

E1000_ROUNDUP macro cleanup, use ALIGN

Signed-off-by: Milind Arun Choudhary <milindchoudhary@gmail.com>
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoGeneric HDLC sparse annotations
Krzysztof Halasa [Fri, 27 Apr 2007 11:13:33 +0000 (13:13 +0200)]
Generic HDLC sparse annotations

Sparse annotations, including two minor bugfixes.

Signed-off-by: Krzysztof Halasa <khc@pm.waw.pl>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoe100: Optionally use I/O mode only to access register space
Jesse Brandeburg [Thu, 26 Apr 2007 16:43:20 +0000 (09:43 -0700)]
e100: Optionally use I/O mode only to access register space

It appears that some systems still like e100 better if it uses
I/O access mode.  Setting the new parameter use_io=1 will cause
all driver instances to use io mapping to access the register
space on the e100 device.

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoe100: allow bad MAC address when running with invalid eeprom csum
Jesse Brandeburg [Thu, 26 Apr 2007 16:43:11 +0000 (09:43 -0700)]
e100: allow bad MAC address when running with invalid eeprom csum

Seved Torstendahl <seved.torstendahl@netinsight.net> suggested to
let the module parameter for invalid eeprom checksum control the valid
mac address test.

If this bypass happens we should print a different message,
or at least one that is correct, maybe something like below

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoehea: fix for dlpar support
Thomas Klein [Thu, 26 Apr 2007 09:56:43 +0000 (11:56 +0200)]
ehea: fix for dlpar support

Certain resources may only be allocated when first logical port is available,
and must be removed when last logical port has been removed.

Signed-off-by: Thomas Klein <tklein@de.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoehea: fix for sysfs entries
Thomas Klein [Thu, 26 Apr 2007 09:56:13 +0000 (11:56 +0200)]
ehea: fix for sysfs entries

Create symbolic link from each logical port to ehea driver

Signed-off-by: Thomas Klein <tklein@de.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years ago3C509: Remove unnecessary include of <linux/pm_legacy.h>
Robert P. J. Day [Thu, 26 Apr 2007 07:23:25 +0000 (00:23 -0700)]
3C509: Remove unnecessary include of <linux/pm_legacy.h>

Remove the apparently redundant include of <linux/pm_legacy.h>.

Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoNetXen: Fix for vmalloc issues
Mithlesh Thukral [Fri, 20 Apr 2007 14:57:26 +0000 (07:57 -0700)]
NetXen: Fix for vmalloc issues

NetXen: Fix vmalloc errors on seen on some X86 high end machines.

Signed-off by: Milan Bag <mbag@netxen.com>
Acked-by: Mithlesh Thukral <mithlesh@netxen.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoNetXen: Fixes for Power PC architecture
Mithlesh Thukral [Fri, 20 Apr 2007 14:56:42 +0000 (07:56 -0700)]
NetXen: Fixes for Power PC architecture

NetXen: Fix PPC architecture specific bugs
Fixes some issues seen on Big endian machines.

Signed-off by: Milan Bag <mbag@netxen.com>
Signed-off by: Mithlesh Thukral <mithlesh@netxen.com>

Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoNetXen: Port swap feature for multi port cards
Mithlesh Thukral [Fri, 20 Apr 2007 14:55:26 +0000 (07:55 -0700)]
NetXen: Port swap feature for multi port cards

NetXen: Port Swap feature
This patch will allow a port numbers on the card to be swapped in
host driver. This feature is applicable to cards having more than
1 port.

Signed-off by: Milan Bag <mbag@netxen.com>
Signed-off by: Mithlesh Thukral <mithlesh@netxen.com>

Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoNetXen: Removal of redundant macros
Mithlesh Thukral [Fri, 20 Apr 2007 14:54:28 +0000 (07:54 -0700)]
NetXen: Removal of redundant macros

NetXen: Remove 2 redundant macro definitions from header file.

Signed-off by: Mithlesh Thukral <mithlesh@netxen.com>

Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoNetXen: Multi PCI support for Quad cards
Mithlesh Thukral [Fri, 20 Apr 2007 14:53:52 +0000 (07:53 -0700)]
NetXen: Multi PCI support for Quad cards

NetXen: Fix the multi PCI function for cards with more than 2 ports.
This patch fixes the working of multi PCI capable driver on cards with
more than 2 ports by adding the addresses for their rings and sizes.

Signed-off by: Mithlesh Thukral <mithlesh@netxen.com>

Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoNetXen: Removal of redundant argument passing
Mithlesh Thukral [Fri, 20 Apr 2007 14:53:05 +0000 (07:53 -0700)]
NetXen: Removal of redundant argument passing

NetXen: Removal of redundant function call parameters and bug fixes.
This patch will remove the redundant paramters which were being passed to
many functions since now adapter->portnum can be used.

Signed-off-by: Mithlesh Thukral <mithlesh@netxen.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoNetXen: Use multiple PCI functions
Mithlesh Thukral [Fri, 20 Apr 2007 14:52:37 +0000 (07:52 -0700)]
NetXen: Use multiple PCI functions

NetXen: Make driver use multiple PCI functions.
This patch will make NetXen driver work with multiple PCI functions. This will
make the usage of memory resources as well as interrupts more independent
among different functions which results in better throughput. This change has
been done after the multiport support is added in firmware.

Signed-off by: Mithlesh Thukral <mithlesh@netxen.com>

Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years ago[netdrvr e100] experiment with doing RX in a similar manner to eepro100
Scott Feldman [Wed, 9 Nov 2005 07:18:52 +0000 (02:18 -0500)]
[netdrvr e100] experiment with doing RX in a similar manner to eepro100

I was going to say that eepro100's speedo_rx_link() does the same DMA
abuse as e100, but then I noticed one little detail: eepro100 sets  both
EL (end of list) and S (suspend) bits in the RFD as it chains it  to the
RFD list.  e100 was only setting the EL bit.  Hmmm, that's  interesting.
That means that if HW reads a RFD with the S-bit set,  it'll process
that RFD and then suspend the receive unit.  The  receive unit will
resume when SW clears the S-bit.  There is no need  for SW to restart
the receive unit.  Which means a lot of the receive  unit state tracking
code in the driver goes away.

So here's a patch against 2.6.14.  (Sorry for inlining it; the mailer
I'm using now will mess with the word wrap).  I can't test this on
XScale (unless someone has an e100 module for Gumstix :) .  It should
be doing exactly what eepro100 does with RFDs.  I don't believe this
change will introduce a performance hit because the S-bit and EL-bit  go
hand-in-hand meaning if we're going to suspend because of the S- bit,
we're on the last resource anyway, so we'll have to wait for SW  to
replenish.
(cherry picked from 29e79da9495261119e3b2e4e7c72507348e75976 commit)

17 years ago[PATCH] ieee80211: add missing global needed by IEEE80211_DEBUG_XXXX
Larry Finger [Tue, 17 Apr 2007 15:28:47 +0000 (10:28 -0500)]
[PATCH] ieee80211: add missing global needed by IEEE80211_DEBUG_XXXX

Using any of the IEEE80211_DEBUG_XXXX macros in any ieee80211_crypt
routine built as a module results in a missing global for
'ieee80211_debug_level'.  The fix is to export the symbol in ieee80211.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years agomii: add kernel-doc notation
Randy Dunlap [Fri, 6 Apr 2007 18:08:24 +0000 (11:08 -0700)]
mii: add kernel-doc notation

Add kernel-doc notation to drivers/net/mii.c functions.

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoFix sparse errors in drivers/net/ibmveth.c
Michael Ellerman [Tue, 17 Apr 2007 03:12:55 +0000 (13:12 +1000)]
Fix sparse errors in drivers/net/ibmveth.c

drivers/net/ibmveth.c:96:46: error: marked inline, but without a definition
drivers/net/ibmveth.c:96: warning: 'ibmveth_rxq_harvest_buffer' declared inline after being called
drivers/net/ibmveth.c:96: warning: previous declaration of 'ibmveth_rxq_harvest_buffer' was here

Just let the compiler decide, as it happens gcc 4.~ inlines it anyway.

drivers/net/ibmveth.c:957:71: warning: Using plain integer as NULL pointer
drivers/net/ibmveth.c:964:85: warning: Using plain integer as NULL pointer

Split the long lines as well, ugly, but < 80 columns.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoucc_geth: version 1.1
Kim Phillips [Fri, 13 Apr 2007 06:26:36 +0000 (01:26 -0500)]
ucc_geth: version 1.1

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoucc_geth: fixes for ucc_geth_memclean
Nicu Ioan Petru [Fri, 13 Apr 2007 06:26:29 +0000 (01:26 -0500)]
ucc_geth: fixes for ucc_geth_memclean

The ucc_geth_memclean function can be called before the Tx BD rings, Rx
BD rings and associated socket buffers are allocated (for example if
ucc_fast_init fails). The current code doesn't check if p_tx_bd_ring[i]
is null, generating a kernel panic when trying to free the associated
socket buffers.

The function can also fail when accessing the uninitialized list_head
structures ugeth->group_hash_q and ugeth->ind_hash_q. In the current
implementation the list heads are initialized only when
maxGroupAddrInHash and maxIndAddrInHash are positive values, although I
think it's better to always initialize them.

Signed-off-by: Ionut Nicu <ionut.nicu@freescale.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoucc_geth: Implement Transmit on Demand support
Michael Reiss [Fri, 13 Apr 2007 06:26:19 +0000 (01:26 -0500)]
ucc_geth: Implement Transmit on Demand support

Transmit on Demand: Fix spelling in config option, and make it actually enable TOD.

Signed-off-by: Michael Reiss <michael.f.reiss@freescale.com>
Signed-off-by: Michael Barkowski <michael.barkowski@freescale.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoucc_geth: Fix interrupt coalescing size and alignment
Michael Barkowski [Fri, 13 Apr 2007 06:26:15 +0000 (01:26 -0500)]
ucc_geth: Fix interrupt coalescing size and alignment

The rx interrupt coalescing table alignment was "guessed" to be 4,
but should be 64. The size should be 8 * number of queues + 4.
Verified in the MPC8323E manual.

Signed-off-by: Michael Barkowski <Michael.Barkowski@freescale.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoucc_geth: NAPI-related bug fixes
Michael Reiss [Fri, 13 Apr 2007 06:26:11 +0000 (01:26 -0500)]
ucc_geth: NAPI-related bug fixes

Based partly on the gianfar driver, this patch fixes several
bugs which were causing NAPI to be completely unusable.
* An IRQ is still needed in NAPI, to kick off NAPI task,
  and for Tx processing.  Request the IRQ.
* If rx_work_limit = 0 we are not complete.
* While running Rx NAPI processing we must mask Rx events,
  including Rx busy.
* ucc_geth_rx function does not need a lock.
  Could lead to deadlock in NAPI case.
* There's no need to loop reading ucce multiple times in the ISR,
  so while adding the call to schedule NAPI which was not there,
  simplify the event processing into if-else format.
* Rx Busy now kicks off NAPI processing, while still
  being counted as an error.

Signed-off-by: Michael Reiss <michael.f.reiss@freescale.com>
Signed-off-by: Michael Barkowski <michael.barkowski@freescale.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoucc_geth: migrate ucc_geth to phylib
Kim Phillips [Fri, 13 Apr 2007 06:26:03 +0000 (01:26 -0500)]
ucc_geth: migrate ucc_geth to phylib

migrate ucc_geth to use the common phylib code.

There are several side effects from doing this:

o deprecate 'interface' property specification present
  in some old device tree source files in
  favour of a split 'max-speed' and 'interface-type'
  description to appropriately match definitions
  in include/linux/phy.h.  Note that 'interface' property
  is still honoured if max-speed or interface-type
  are not present (backward compatible).
o compile-time CONFIG_UGETH_HAS_GIGA is eliminated
  in favour of probe time speed derivation logic.
o adjust_link streamlined to only operate on maccfg2
  and upsmr.r10m, instead of reapplying static initial
  values related to the interface-type.
o Addition of UEC MDIO of_platform driver requires
  platform code add 'mdio' type to id list
  prior to calling of_platform_bus_probe (separate patch).
o ucc_struct_init introduced to reduce ucc_geth_startup
  complexity.

Signed-off-by: Li Yang <leoli@freescale.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agophylib: add RGMII-ID interface mode definition
Kim Phillips [Fri, 13 Apr 2007 06:25:57 +0000 (01:25 -0500)]
phylib: add RGMII-ID interface mode definition

The RGMII spec allows compliance for devices that implement an internal
delay on TXC or RXC inside the transmitter.  This patch adds an RGMII_ID
definition to support RGMII-ID devices in the phylib.

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years ago[PATCH] ieee80211-crypt: Make some TKIP and CCMP error logging conditional
Larry Finger [Mon, 9 Apr 2007 16:24:41 +0000 (11:24 -0500)]
[PATCH] ieee80211-crypt: Make some TKIP and CCMP error logging conditional

There are messages arising from ieee80211_crypt that spam the logs
of casual users. These are changed to be logged only if the user
specifically requests the IEEE80211_DEBUG_DROP messages. In either
case, the error/drop count is incremented.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years ago[PATCH] zd1211rw: Add ID for ZyXEL AG-220
Daniel Drake [Sat, 7 Apr 2007 15:00:43 +0000 (16:00 +0100)]
[PATCH] zd1211rw: Add ID for ZyXEL AG-220

Tested by Christoph Sager and Tomas Klas

zd1211b chip 0586:3412 v4810 high 00-13-49 AL7230B_RF pa0 g----

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years ago[PATCH] zd1211rw: Add AL7230B RF support for ZD1211B
Daniel Drake [Sat, 7 Apr 2007 15:00:30 +0000 (16:00 +0100)]
[PATCH] zd1211rw: Add AL7230B RF support for ZD1211B

This patch adds support for some new ZD1211B devices which ship with
the AL7230B RF.

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years ago[PATCH] zd1211rw: rework band edge patching
Daniel Drake [Sat, 7 Apr 2007 15:00:15 +0000 (16:00 +0100)]
[PATCH] zd1211rw: rework band edge patching

This change allows RF drivers to provide their own 6M band edge patching
implementation, while providing a generic implementation shared by most
currently supported RF's.

The upcoming ZD1211B/AL7230B code will use this to define its own
patching function, which is different from the other RF configurations.

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years ago[PATCH] zd1211rw: Remove invalid CR write during ZD1211 phy reset
Daniel Drake [Tue, 3 Apr 2007 22:17:10 +0000 (23:17 +0100)]
[PATCH] zd1211rw: Remove invalid CR write during ZD1211 phy reset

The vendor driver only does the CR123 write for non-USB devices (which
don't exist on the consumer market)

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years ago[PATCH] hostap_cs: support ADLINK 345 CF
Marcin Juszkiewicz [Fri, 30 Mar 2007 13:34:14 +0000 (15:34 +0200)]
[PATCH] hostap_cs: support ADLINK 345 CF

Add ADLINK 345 CF card into hostap_cs

product info: "ADLINK 345 CF"
 manfid 0xd601, 0x0005

Signed-off-by: Marcin Juszkiewicz <openembedded@hrw.one.pl>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years ago[PATCH] Remove comment about IEEE80211_RADIOTAP_FCS
Pavel Roskin [Wed, 28 Mar 2007 21:08:51 +0000 (17:08 -0400)]
[PATCH] Remove comment about IEEE80211_RADIOTAP_FCS

IEEE80211_RADIOTAP_FCS is obsolete and should not be used.  It's no
longer defined.  Remove it from the comment too.

Signed-off-by: Pavel Roskin <proski@gnu.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years ago[PATCH] bcm43xx: Change initialization for 2050 radios
Larry Finger [Wed, 28 Mar 2007 16:07:49 +0000 (11:07 -0500)]
[PATCH] bcm43xx: Change initialization for 2050 radios

This patch implements the changes in the specifications for
2050radio_init that were recently posted.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years agomake drivers/net/qla3xxx.c:PHY_DEVICES[] static
Adrian Bunk [Sat, 31 Mar 2007 20:55:40 +0000 (22:55 +0200)]
make drivers/net/qla3xxx.c:PHY_DEVICES[] static

On Fri, Mar 30, 2007 at 01:05:59AM -0700, Andrew Morton wrote:
>...
> Changes since 2.6.21-rc5-mm2:
>...
>  git-netdev-all.patch
>...
>  git trees
>...

This patch makes the needlessly global PHY_DEVICES[] static.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoS2io: Change of driver maintainers
Ramkrishna Vepa [Sat, 31 Mar 2007 01:42:27 +0000 (18:42 -0700)]
S2io: Change of driver maintainers

- Changed the maintainers for the S2io driver.

Signed-off-by: Ramkrishna Vepa <ram.vepa@neterion.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoPatch:replace with time_after in drivers/net/eexpress.c
Shani [Wed, 28 Mar 2007 05:14:31 +0000 (10:44 +0530)]
Patch:replace with time_after in drivers/net/eexpress.c

Replacing with time_after in drivers/net/eexpress.c
Applies and compiles clean on latest tree.Not tested.

Signed-off-by: Shani Moideen <shani.moideen@wipro.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoehea: removing unused functionality
Jan-Bernd Themann [Fri, 23 Mar 2007 16:18:53 +0000 (17:18 +0100)]
ehea: removing unused functionality

This patch includes:
- removal of unused fields in structs
- ethtool statistics cleanup
- removes unsed functionality from send path

Signed-off-by: Jan-Bernd Themann <themann@de.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agomv643xx_eth: make eth_port_uc_addr_{get,set}() calls symmetric
Gabriel Paubert [Fri, 23 Mar 2007 19:07:26 +0000 (12:07 -0700)]
mv643xx_eth: make eth_port_uc_addr_{get,set}() calls symmetric

There is no good reason for the asymmetry in the parameters of
eth_port_uc_addr_get() and eth_port_uc_addr_set().  Make them
symmetric.  Remove some gratuitous block comments while we're here.

Signed-off-by: Gabriel Paubert <paubert@iram.es>
Signed-off-by: Dale Farnsworth <dale@farnsworth.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoqla3xxx: Bumping driver version number to v2.03.00-k4
Ron Mercer [Mon, 26 Mar 2007 20:43:03 +0000 (13:43 -0700)]
qla3xxx: Bumping driver version number to v2.03.00-k4

Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoqla3xxx: Adding support for the Agere PHY (ET1011C)
Ron Mercer [Mon, 26 Mar 2007 19:43:52 +0000 (12:43 -0700)]
qla3xxx: Adding support for the Agere PHY (ET1011C)

This PHY support patch was written by Benjamin Li.
Signed-off-by: Benjamin Li <benjamin.li@qlogic.com>
Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoqla3xxx: Add ethtool get_pauseparam for improved bonding support.
Ron Mercer [Mon, 26 Mar 2007 20:43:01 +0000 (13:43 -0700)]
qla3xxx: Add ethtool get_pauseparam for improved bonding support.

Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoremove unused header file: drivers/net/skfp/h/lnkstat.h
Robert P. J. Day [Tue, 27 Mar 2007 05:47:24 +0000 (21:47 -0800)]
remove unused header file: drivers/net/skfp/h/lnkstat.h

Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agobaycom_ser_fdx: also allow i/o ports >= 0x1000 and enhanced failure logging
Folkert van Heusden [Tue, 27 Mar 2007 05:47:23 +0000 (21:47 -0800)]
baycom_ser_fdx: also allow i/o ports >= 0x1000 and enhanced failure logging

The baycom_ser_fdx driver did not allow i/o ports >= 0x1000.  Now that
there are pci cards (with rs232 ports) which use for example 0xb800 this
limit should not exists.

Also, for non kernel coders find the cause of problems was challenging so I
added extra logging.

Signed-off-by: Folkert van Heusden <folkert@vanheusden.com>
Cc: Jeff Garzik <jeff@garzik.org>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agostrlcpy is smart enough
Jean Delvare [Tue, 27 Mar 2007 05:47:22 +0000 (21:47 -0800)]
strlcpy is smart enough

strlcpy already accounts for the trailing zero in its length
computation, so there is no need to substract one to the buffer size.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Jan-Bernd Themann <themann@de.ibm.com>
Cc: Thomas Klein <osstklei@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years ago[PATCH] zd1211rw: Add another ID for Linksys WUSBF54G
Daniel Drake [Mon, 26 Mar 2007 00:00:12 +0000 (01:00 +0100)]
[PATCH] zd1211rw: Add another ID for Linksys WUSBF54G

Tested by TiCPU on irc

zd1211 chip 13b1:001e v4802 high 00-14-bf AL2230_RF pa0 g---

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years ago[PATCH] zd1211rw: Don't handle broken frames in monitor mode
Daniel Drake [Sun, 25 Mar 2007 23:59:47 +0000 (00:59 +0100)]
[PATCH] zd1211rw: Don't handle broken frames in monitor mode

Using monitor mode, Johannes Berg observed out that lots of corrupted
and otherwise invalid frames were being passed to the host.

When in monitor mode we were disabling the hardware filtering here, but
this is not how monitor mode should work.

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years ago[PATCH] zd1211rw: Mark some data const
Daniel Drake [Sun, 25 Mar 2007 23:59:18 +0000 (00:59 +0100)]
[PATCH] zd1211rw: Mark some data const

This is a backport of Helge Deller's recent patch for wireless-dev.git

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years ago[PATCH] Update my email address from jkmaline@cc.hut.fi to j@w1.fi
Jouni Malinen [Sun, 25 Mar 2007 00:15:30 +0000 (17:15 -0700)]
[PATCH] Update my email address from jkmaline@cc.hut.fi to j@w1.fi

After 13 years of use, it looks like my email address is finally going
to disappear. While this is likely to drop the amount of incoming spam
greatly ;-), it may also affect more appropriate messages, so let's
update my email address in various places. In addition, Host AP mailing
list is subscribers-only and linux-wireless can also be used for
discussing issues related to this driver which is now shown in
MAINTAINERS.

Signed-off-by: Jouni Malinen <j@w1.fi>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years agoehea: code cleanup
Jan-Bernd Themann [Thu, 22 Mar 2007 16:50:24 +0000 (17:50 +0100)]
ehea: code cleanup

This patch includes:

- code cleanup related to resource management
- extended error data gathering for resource management
- removing trailing whitespaces

Signed-off-by: Jan-Bernd Themann <themann@de.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoehea: fix for dynamic lpar support
Jan-Bernd Themann [Thu, 22 Mar 2007 16:49:42 +0000 (17:49 +0100)]
ehea: fix for dynamic lpar support

The patch fixes bugs related to the probe / remove adapter
functionality (handling of OFDT nodes)

Signed-off-by: Jan-Bernd Themann <themann@de.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoNetpoll support for Sibyte MAC
Deepak Saxena [Mon, 19 Mar 2007 22:43:11 +0000 (15:43 -0700)]
Netpoll support for Sibyte MAC

NETPOLL support for Sibyte MAC

Signed-off-by: Manish Lachwani <mlachwani@mvista.com>
Signed-off-by: Deepak Saxena <dsaxena@mvista.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoS2io: Remove unused variables
Ramkrishna Vepa [Sat, 10 Mar 2007 02:28:32 +0000 (18:28 -0800)]
S2io: Remove unused variables

- Remove unused variables from s2io_nic structure

- Changed the memory failure printk messages to print only in debug mode

- Updated the copyright messages

(Resending; due to patch being corrupted)

Signed-off-by: Santosh Rastapur <santosh.rastapur@neterion.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoMIPSnet: Modernize use platform_device API.
Ralf Baechle [Sun, 18 Mar 2007 23:21:22 +0000 (23:21 +0000)]
MIPSnet: Modernize use platform_device API.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoskge: version 1.11
Stephen Hemminger [Fri, 16 Mar 2007 21:01:32 +0000 (14:01 -0700)]
skge: version 1.11

New version to track changes.

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoskge: rearrange fields
Stephen Hemminger [Fri, 16 Mar 2007 21:01:31 +0000 (14:01 -0700)]
skge: rearrange fields

Do some minor rearrangement of data structures to try and optimize
cache usage.

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>