linux-2.6
17 years ago[POWERPC] Fix breakage caused by 72486f1f8f0a2bc828b9d30cf4690cf2dd6807fc
Giuliano Pochini [Tue, 27 Mar 2007 05:40:28 +0000 (21:40 -0800)] 
[POWERPC] Fix breakage caused by 72486f1f8f0a2bc828b9d30cf4690cf2dd6807fc

72486f1f8f0a2bc828b9d30cf4690cf2dd6807fc inverted the sense for enabling
hotplug CPU controls without reference to any other architecture other than
i386, ia64 and PowerPC.  This left everyone else without hotplug CPU control.

Fix powerpc for this brain damage.

(akpm: patch adapted from rmk's ARM fix.  Changelog stolen from rmk)

Signed-off-by: Giuliano Pochini <pochini@shiny.it>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years ago[POWERPC] Remove unused header file: arch/ppc/syslib/cpc710.h.
Robert P. J. Day [Tue, 27 Mar 2007 05:40:27 +0000 (21:40 -0800)] 
[POWERPC] Remove unused header file: arch/ppc/syslib/cpc710.h.

Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years ago[POWERPC] Ocotea board: ntpd complains that the frequency error exceeds the tolerance
Bart Van Assche [Tue, 27 Mar 2007 05:40:29 +0000 (21:40 -0800)] 
[POWERPC] Ocotea board: ntpd complains that the frequency error exceeds the tolerance

Lifted from http://bugzilla.kernel.org/show_bug.cgi?id=8182

Steps to reproduce:
- Boot an Ocotea board with the mainline 2.6.20.1 kernel.
- Create an /etc/ntp.conf file with at least one NTP server and iburst mode set.
- Issue the command "ntpd -g".
- Wait about two minutes.
- Verify ntpd's status via "ntpq -pn" and by looking in /var/log/ntp.

This fixes this problem by adjusting the expected clock frequency.

Cc: Kumar Gala <galak@gate.crashing.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years ago[POWERPC] bootwrapper: Add dt_xlate_reg(), and use it to find serial registers.
Scott Wood [Mon, 26 Mar 2007 20:52:24 +0000 (15:52 -0500)] 
[POWERPC] bootwrapper: Add dt_xlate_reg(), and use it to find serial registers.

dt_xlate_reg() uses the ranges properties of a node's parentage to find
the absolute physical address of the node's registers.

The ns16550 driver uses this when no virtual-reg property is found.

Signed-off-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years ago[POWERPC] spufs: don't yield CPU in spu_yield
Christoph Hellwig [Fri, 23 Mar 2007 14:12:26 +0000 (15:12 +0100)] 
[POWERPC] spufs: don't yield CPU in spu_yield

There is no reason to yield the CPU in spu_yield - if the backing
thread reenters spu_run it gets added to the end of the runqueue for
it's priority.  So the yield is just a slowdown for the case where
we have higher priority contexts waiting.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years ago[POWERPC] CBE thermal support on PS3
Geert Uytterhoeven [Fri, 23 Mar 2007 13:06:43 +0000 (14:06 +0100)] 
[POWERPC] CBE thermal support on PS3

I wanted to enable CBE_THERM on PS3.  So I had to enable CBE_RAS first.

But the resulting kernel doesn't link, as cbe_regs.c isn't compiled for
non-PPC_CELL_NATIVE.

CBE_RAS should depend on PPC_CELL_NATIVE; this makes it so.

Signed-off-by: Geert Uytterhoeven <Geert.Uytterhoeven@sonycom.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years ago[POWERPC] Remove stale comment from head_64.S
Sonny Rao [Fri, 23 Mar 2007 02:44:31 +0000 (22:44 -0400)] 
[POWERPC] Remove stale comment from head_64.S

This is now inaccurate because we may not have entered prom_init() and
r3 is overwritten immediately anyway.

Signed-off-by: Sonny Rao <sonny@burdell.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years ago[POWERPC] Remove unused inclusion of linux/ide.h
Olaf Hering [Fri, 23 Mar 2007 00:11:59 +0000 (01:11 +0100)] 
[POWERPC] Remove unused inclusion of linux/ide.h

Remove unneeded inclusion of linux/ide.h
It does not compile with CONFIG_BLOCK=n.

Remove asm/ide.h from ksyms file, it gets included earlier via
linux/ide.h.

Compile tested with all defconfig files.

Signed-off-by: Olaf Hering <olaf@aepfle.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years ago[POWERPC] PMU LED whitelisting of PowerMac 7,2 and 7,3
Tony Vroon [Thu, 22 Mar 2007 23:31:08 +0000 (23:31 +0000)] 
[POWERPC] PMU LED whitelisting of PowerMac 7,2 and 7,3

This allows the PMU LED on both a PowerMac 7,2 (Dual G5 2.0GHz, June 2003)
and a PowerMac 7,3 (Dual G5 2.0GHz, June 2004) to be controlled.
The physical LED is never off, unlike an iBook/PowerBook LED.
It is rather dim ("off") or very bright ("on").

Signed-off-by: Tony Vroon <chainsaw@gentoo.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years ago[POWERPC] Fix link errors when EEH is disabled
Olaf Hering [Thu, 22 Mar 2007 22:14:07 +0000 (23:14 +0100)] 
[POWERPC] Fix link errors when EEH is disabled

Fix link errors with CONFIG_EEH=n:

arch/powerpc/platforms/built-in.o: In function `.pcibios_fixup_new_pci_devices':
(.text+0x41c8): undefined reference to `.eeh_add_device_tree_late'
arch/powerpc/platforms/built-in.o: In function `.init_phb_dynamic':
(.text+0x4280): undefined reference to `.eeh_add_device_tree_early'
arch/powerpc/platforms/built-in.o: In function `.pcibios_remove_pci_devices':
(.text+0x42fc): undefined reference to `.eeh_remove_bus_device'
arch/powerpc/platforms/built-in.o: In function `.pcibios_add_pci_devices':
(.text+0x43c0): undefined reference to `.eeh_add_device_tree_early'
arch/powerpc/platforms/built-in.o: In function `.pSeries_final_fixup':
(.init.text+0xb4): undefined reference to `.pci_addr_cache_build'
make[1]: *** [.tmp_vmlinux1] Error 1

Signed-off-by: Olaf Hering <olaf@aepfle.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years ago[POWERPC] Clean up zImage handling of the command line
David Gibson [Thu, 22 Mar 2007 06:02:21 +0000 (17:02 +1100)] 
[POWERPC] Clean up zImage handling of the command line

This cleans up how the zImage code manipulates the kernel
command line.  Notable improvements from the old handling:
- Command line manipulation is consolidated into a new
prep_cmdline() function, rather than being scattered across start()
and some helper functions
- Less stack space use: we use just a single global command
line buffer, which can be initialized by an external tool as before,
we no longer need another command line sized buffer on the stack.
- Easier to support platforms whose firmware passes a
commandline, but not a device tree.  Platform code can now point new
loader_info fields to the firmware's command line, rather than having
to do early manipulation of the /chosen bootargs property which may
then be rewritten again by the core.

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years ago[POWERPC] Add device tree utility functions to zImage
David Gibson [Thu, 22 Mar 2007 06:02:21 +0000 (17:02 +1100)] 
[POWERPC] Add device tree utility functions to zImage

This patch adds a library of useful device tree manipulation functions
to the zImage library, for use by platform code.  These functions are
based on the hooks already in dt_ops, so they're not dependent on a
particular device tree implementation.  This patch also slightly
streamlines the code in main.c using these new functions.

This is a consolidation of my work in this area with Scott Wood's
patches to a very similar end.

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years ago[POWERPC] Use mtocrf instruction in asm when CONFIG_POWER4_ONLY=y
Olof Johansson [Thu, 22 Mar 2007 14:34:13 +0000 (09:34 -0500)] 
[POWERPC] Use mtocrf instruction in asm when CONFIG_POWER4_ONLY=y

mtocrf is a faster single-field mtcrf (move to condition register
fields) instruction available in POWER4 and later processors.  It can
make quite a difference in performance on some implementations, so use
it for CONFIG_POWER4_ONLY builds.

Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years ago[POWERPC] DMA 4GB boundary protection
Jake Moilanen [Thu, 29 Mar 2007 13:44:02 +0000 (08:44 -0500)] 
[POWERPC] DMA 4GB boundary protection

There are many adapters which cannot handle DMAing across any 4 GB
boundary.  For instance, the latest Emulex adapters.

This normally is not an issue as firmware gives dma-windows under
4gigs.  However, some of the new System-P boxes have dma-windows above
4gigs, and this present a problem.

During initialization of the IOMMU tables, the last entry at each 4GB
boundary is marked as used.  Thus no mappings can cross the boundary.
If a table ends at a 4GB boundary, the entry is not marked as used.

A boot option to remove this 4GB protection is given w/ protect4gb=off.
This exposes the potential issue for driver and hardware development
purposes.

Signed-off-by: Jake Moilanen <moilanen@austin.ibm.com>
Acked-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years ago[POWERPC] Add uevent handler for of_platform_bus
Sylvain Munaut [Mon, 12 Feb 2007 22:13:27 +0000 (23:13 +0100)] 
[POWERPC] Add uevent handler for of_platform_bus

Adding this handler allow userspace to properly handle the module
autoloading. The generation of the uevent itself is now common to
all bus using of_device, so not much code here.

Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years ago[POWERPC] powermac: Use the new of_device common uevent handler
Sylvain Munaut [Mon, 12 Feb 2007 22:13:26 +0000 (23:13 +0100)] 
[POWERPC] powermac: Use the new of_device common uevent handler

The generation of the uevent is now common to all bus using
of_device.

Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years ago[POWERPC] Add a unified uevent handler for bus based on of_device
Sylvain Munaut [Mon, 12 Feb 2007 22:13:25 +0000 (23:13 +0100)] 
[POWERPC] Add a unified uevent handler for bus based on of_device

This common uevent handler allow the several bus types based on
of_device to generate the uevent properly and avoiding
code duplication.

This handlers take a struct device as argument and can therefore
be used as the uevent call directly if no special treatment is
needed for the bus.

Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years agoMerge branch 'linux-2.6' into for-2.6.22
Paul Mackerras [Thu, 12 Apr 2007 17:50:03 +0000 (03:50 +1000)] 
Merge branch 'linux-2.6' into for-2.6.22

17 years agoMerge branch 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc
Linus Torvalds [Wed, 11 Apr 2007 22:44:45 +0000 (15:44 -0700)] 
Merge branch 'merge' of git://git./linux/kernel/git/paulus/powerpc

* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc:
  [POWERPC] mpc52xx_pic: fix main interrupt masking

17 years agoMerge branch 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik...
Linus Torvalds [Wed, 11 Apr 2007 22:40:45 +0000 (15:40 -0700)] 
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:
  myri10ge: update driver version to 1.3.0-1.233
  myri10ge: more Intel chipsets providing aligned PCIe completions
  myri10ge: fix management of the firmware 4KB boundary crossing restriction
  cxgb3 - missing CPL hanler and register setting.
  cxgb3 - MAC watchdog update
  cxgb3 - avoid deadlock with mac watchdog
  skge: fix wake on lan
  sky2: phy workarounds for Yukon EC-U A1
  sky2: turn on clocks when doing resume
  sky2: turn carrier off when down
  skge: turn carrier off when down
  [PATCH] bcm43xx: Fix PPC machine checks and match loopback gain specs
  [PATCH] bcm43xx: Fix 802.11b/g scan limits to match regulatory reqs
  [PATCH] zd1211rw: Fix E2P_PHY_REG patching
  [PATCH] zd1211rw: Reject AL2230S devices

17 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/gregkh/usb-2.6
Linus Torvalds [Wed, 11 Apr 2007 22:40:03 +0000 (15:40 -0700)] 
Merge /pub/scm/linux/kernel/git/gregkh/usb-2.6

* master.kernel.org:/pub/scm/linux/kernel/git/gregkh/usb-2.6:
  USB: Nikon D80 unusual device patch
  EHCI: fix remote wakeup regression in 2.6.21-rc

17 years ago[PATCH] PM: use kobject_name() to access kobject names
Dmitry Torokhov [Wed, 11 Apr 2007 05:37:18 +0000 (01:37 -0400)] 
[PATCH] PM: use kobject_name() to access kobject names

Noone should use kobj.name directly since it may contain garbage.
Objects with longer names have them stored in separately allocated
memory pointed to by kobj->k_name.

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years agoUSB: Nikon D80 unusual device patch
Emil Larsson [Fri, 30 Mar 2007 05:48:46 +0000 (22:48 -0700)] 
USB: Nikon D80 unusual device patch

the Nikon D80 camera will not work without an UNUSUAL_DEV entry embodied
in the attached patch (made against 2.6.20.3). Hope you find it helpful,
or if not, pass it along to someone who does.

From: Emil Larsson <emil@swip.net>
Signed-off-by: Phil Dibowitz <phil@ipom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
17 years agoEHCI: fix remote wakeup regression in 2.6.21-rc
Alan Stern [Mon, 9 Apr 2007 15:52:31 +0000 (11:52 -0400)] 
EHCI: fix remote wakeup regression in 2.6.21-rc

There is one significant difference between the behavior of root hubs
(as embodied in host controller hardware) and external hubs: When a
remote-wakeup signal is received, an external hub sends an interrupt
message at the _end_ of the resume sequence but a root hub generates
and interrupt at the _beginning_ of the resume sequence.  The host
system must poll for the end of the sequence.

When ehci-hcd was converted to interrupt-driven operation instead of
using polling, the remaining need for this particular poll was
overlooked.  This patch (as894) fixes the problem.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
17 years ago[POWERPC] mpc52xx_pic: fix main interrupt masking
Domen Puncer [Tue, 10 Apr 2007 22:27:49 +0000 (00:27 +0200)] 
[POWERPC] mpc52xx_pic: fix main interrupt masking

The bit setting was off by one.
Tested with RTC and GPIO_WKUP interrupts.

Signed-off-by: Domen Puncer <domen.puncer@telargo.com>
Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
17 years agoMerge branch 'upstream-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/linvil...
Jeff Garzik [Wed, 11 Apr 2007 15:55:45 +0000 (11:55 -0400)] 
Merge branch 'upstream-fixes' of git://git./linux/kernel/git/linville/wireless-2.6 into upstream-fixes

17 years agomyri10ge: update driver version to 1.3.0-1.233
Brice Goglin [Tue, 10 Apr 2007 19:22:19 +0000 (21:22 +0200)] 
myri10ge: update driver version to 1.3.0-1.233

Update the myri10ge driver version number to 1.3.0-1.233.

Signed-off-by: Brice Goglin <brice@myri.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agomyri10ge: more Intel chipsets providing aligned PCIe completions
Brice Goglin [Tue, 10 Apr 2007 19:21:39 +0000 (21:21 +0200)] 
myri10ge: more Intel chipsets providing aligned PCIe completions

Add the Intel 5000 southbridge (aka Intel 6310/6311/6321ESB) PCIe ports
and the Intel E30x0 chipsets to the whitelist of aligned PCIe completion.

Signed-off-by: Brice Goglin <brice@myri.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agomyri10ge: fix management of the firmware 4KB boundary crossing restriction
Brice Goglin [Tue, 10 Apr 2007 19:21:08 +0000 (21:21 +0200)] 
myri10ge: fix management of the firmware 4KB boundary crossing restriction

Simpler way of dealing with the firmware 4KB boundary crossing
restriction for rx buffers.  This fixes a variety of memory
corruption issues when using an "uncommon" MTU with a 16KB
page size.

Signed-off-by: Brice Goglin <brice@myri.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agocxgb3 - missing CPL hanler and register setting.
Divy Le Ray [Tue, 10 Apr 2007 03:10:33 +0000 (20:10 -0700)] 
cxgb3 - missing CPL hanler and register setting.

Remove specific CPL handler.
Add missing CPL handler.
Add missing register setting when the interface is brought up.

Signed-off-by: Divy Le Ray <divy@chelsio.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agocxgb3 - MAC watchdog update
Divy Le Ray [Tue, 10 Apr 2007 03:10:27 +0000 (20:10 -0700)] 
cxgb3 - MAC watchdog update

The MAC watchdog was failing if the peer interface was brought down.

Signed-off-by: Divy Le Ray <divy@chelsio.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agocxgb3 - avoid deadlock with mac watchdog
Divy Le Ray [Tue, 10 Apr 2007 03:10:22 +0000 (20:10 -0700)] 
cxgb3 - avoid deadlock with mac watchdog

Fix a deadlock when the interface s configured down and
the watchdog tack is sleeping on rtnl_lock.

Signed-off-by: Divy Le Ray <divy@chelsio.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoskge: fix wake on lan
Stephen Hemminger [Mon, 9 Apr 2007 22:32:45 +0000 (15:32 -0700)] 
skge: fix wake on lan

Need to rework wake on lan code to setup properly and get activated
on shutdown (and suspend), not when ethtool is run.

This does not need to go to stable queue because wake on lan
was not even included in 2.6.20 (or earlier versions).

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agosky2: phy workarounds for Yukon EC-U A1
Stephen Hemminger [Sat, 7 Apr 2007 23:02:28 +0000 (16:02 -0700)] 
sky2: phy workarounds for Yukon EC-U A1

The workaround Yukon EC-U wasn't comparing with correct
version and wasn't doing correct setup. Without it, 88e8056
throws all sorts of errors.

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agosky2: turn on clocks when doing resume
Stephen Hemminger [Sat, 7 Apr 2007 23:02:27 +0000 (16:02 -0700)] 
sky2: turn on clocks when doing resume

Some of these chips are disabled until clock is enabled.
This fixes:
     http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=404107

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agosky2: turn carrier off when down
Stephen Hemminger [Sat, 7 Apr 2007 23:02:26 +0000 (16:02 -0700)] 
sky2: turn carrier off when down

Driver needs to turn off carrier when down.

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years agoskge: turn carrier off when down
Stephen Hemminger [Sat, 7 Apr 2007 23:02:25 +0000 (16:02 -0700)] 
skge: turn carrier off when down

Driver needs to turn off carrier when down, otherwise it can
confuse bonding and bridging and looks like carrier is on immediately
when it is brought back up.

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
17 years ago[PATCH] FRV: Update the FRV arch MAINTAINER record to get a hit on "grep -i frv"
David Howells [Tue, 10 Apr 2007 14:10:45 +0000 (15:10 +0100)] 
[PATCH] FRV: Update the FRV arch MAINTAINER record to get a hit on "grep -i frv"

Update the FRV arch MAINTAINER record to get a hit on "grep -i frv".  Whilst
FR-V is technically correct, it's normally thought of as FRV.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] Input: ucb1400 - set up driver's name to show in sysfs
Cliff Brake [Tue, 10 Apr 2007 03:50:50 +0000 (23:50 -0400)] 
[PATCH] Input: ucb1400 - set up driver's name to show in sysfs

The UCB1400 is missing a name parameter in the device_driver struct.
This causes missing information in the /sys tree and seems to cause
other problems with the AC97 functionality. This was tested on a PXA270
system.

Signed-off-by: Cliff Brake <cbrake@bec-systems.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years agoMerge branch 'for-2.6.21' of master.kernel.org:/pub/scm/linux/kernel/git/davej/agpgart
Linus Torvalds [Wed, 11 Apr 2007 00:24:56 +0000 (17:24 -0700)] 
Merge branch 'for-2.6.21' of /linux/kernel/git/davej/agpgart

* 'for-2.6.21' of master.kernel.org:/pub/scm/linux/kernel/git/davej/agpgart:
  [AGPGART] intel_agp: fix G965 GTT size detect

17 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394...
Linus Torvalds [Wed, 11 Apr 2007 00:24:36 +0000 (17:24 -0700)] 
Merge branch 'for-linus' of git://git./linux/kernel/git/ieee1394/linux1394-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6:
  ieee1394: change deprecation status of dv1394

17 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland...
Linus Torvalds [Wed, 11 Apr 2007 00:23:52 +0000 (17:23 -0700)] 
Merge branch 'for-linus' of git://git./linux/kernel/git/roland/infiniband

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
  IPoIB/cm: Fix DMA direction typo
  IB/iser: Don't defer connection failure notification to workqueue

17 years agoMerge branch 'upstream-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid
Linus Torvalds [Wed, 11 Apr 2007 00:23:36 +0000 (17:23 -0700)] 
Merge branch 'upstream-fixes' of git://git./linux/kernel/git/jikos/hid

* 'upstream-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid:
  HID: Do not discard truncated input reports

17 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/bart/ide-2.6
Linus Torvalds [Wed, 11 Apr 2007 00:22:31 +0000 (17:22 -0700)] 
Merge /pub/scm/linux/kernel/git/bart/ide-2.6

* master.kernel.org:/pub/scm/linux/kernel/git/bart/ide-2.6:
  ide: add "optical" to sysfs "media" attribute
  ide: ugly messages trying to open CD drive with no media present
  ide: correctly prevent IDE timer expiry function to run if request was already handled

17 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Wed, 11 Apr 2007 00:22:16 +0000 (17:22 -0700)] 
Merge /pub/scm/linux/kernel/git/davem/net-2.6

* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6:
  [NETFILTER]: ipt_CLUSTERIP: fix oops in checkentry function
  [TCP]: slow_start_after_idle should influence cwnd validation too
  [SC92031]: Fix priv->lock context
  [NET_SCHED]: cls_tcindex: fix compatibility breakage

17 years agoMerge branch 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6
Linus Torvalds [Wed, 11 Apr 2007 00:21:57 +0000 (17:21 -0700)] 
Merge branch 'release' of git://git./linux/kernel/git/aegl/linux-2.6

* 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6:
  [IA64] SGI Altix : fix pcibr_dmamap_ate32() bug
  [IA64] Fix CPU freq displayed in /proc/cpuinfo
  [IA64] Fix wrong assumption about irq and vector in msi_ia64.c
  [IA64] BTE error timer fix

17 years agoide: add "optical" to sysfs "media" attribute
Danny Kukawka [Tue, 10 Apr 2007 20:39:14 +0000 (22:39 +0200)] 
ide: add "optical" to sysfs "media" attribute

Add "optical" to sysfs "media" attribute as already in /proc

Signed-off-by: Danny Kukawka <dkukawka@suse.de>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
17 years agoide: ugly messages trying to open CD drive with no media present
Stuart Hayes [Tue, 10 Apr 2007 20:38:43 +0000 (22:38 +0200)] 
ide: ugly messages trying to open CD drive with no media present

I get the following error messages when trying to open a CD device
(specifically, the Teac CD-ROM CD-224E) that has no media present:

hda: packet command error: status=3D0x51 { DriveReady SeekComplete Error }
hda: packet command error: error=3D0x54 { AbortedCommand LastFailedSense=0x05 }
ide: failed opcode was: unknown

This happens when a "start stop unit" command (0x1b 0 0 0 3 0 0 0 0 0)
is sent to the drive to try to close the CD-ROM tray, but this drive
doesn't have that capability (it's a slim portable-type CD-ROM), so it
reports sense key 5 (illegal request) with asc/ascq 24/0.  This is
exactly how SFF8090i says it should respond.

But ide-cd.c (in cdrom_decode_status() ) just sees sense key 5 and spews
out an error.  It then goes on to request sense data, and
cdrom_log_sense() understands this error and doesn't log it.

The patch, for kernel 2.6.20.4, suppresses this error message.

Signed-off-by: Stuart Hayes <stuart_hayes@dell.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
17 years agoide: correctly prevent IDE timer expiry function to run if request was already handled
Suleiman Souhlal [Tue, 10 Apr 2007 20:38:37 +0000 (22:38 +0200)] 
ide: correctly prevent IDE timer expiry function to run if request was already handled

It is possible for the timer expiry function to run even though the
request has already been handled: ide_timer_expiry() only checks that
the handler is not NULL, but it is possible that we have handled a
request (thus clearing the handler) and then started a new request
(thus starting the timer again, and setting a handler).

A simple way to exhibit this is to set the DMA timeout to 1 jiffy and
run dd: The kernel will panic after a few minutes because
ide_timer_expiry() tries to add a timer when it's already active.

To fix this, we simply add a request generation count that gets
incremented at every interrupt, and check in ide_timer_expiry() that
we have not already handled a new interrupt before running the expiry
function.

Signed-off-by: Suleiman Souhlal <suleiman@google.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
17 years ago[NETFILTER]: ipt_CLUSTERIP: fix oops in checkentry function
Jaroslav Kysela [Tue, 10 Apr 2007 20:26:48 +0000 (13:26 -0700)] 
[NETFILTER]: ipt_CLUSTERIP: fix oops in checkentry function

The clusterip_config_find_get() already increases entries reference
counter, so there is no reason to do it twice in checkentry() callback.

This causes the config to be freed before it is removed from the list,
resulting in a crash when adding the next rule.

Signed-off-by: Jaroslav Kysela <perex@suse.cz>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
17 years ago[PATCH] bcm43xx: Fix PPC machine checks and match loopback gain specs
Larry Finger [Sat, 7 Apr 2007 19:11:03 +0000 (14:11 -0500)] 
[PATCH] bcm43xx: Fix PPC machine checks and match loopback gain specs

The specifications for loopback_gain calculation and for G PHY
initialization have been updated. This patch implements them and
fixes a machine check error that occurs for PPC architecture with a
phy->rev of 1.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years ago[PATCH] bcm43xx: Fix 802.11b/g scan limits to match regulatory reqs
Larry Finger [Sat, 7 Apr 2007 18:54:35 +0000 (13:54 -0500)] 
[PATCH] bcm43xx: Fix 802.11b/g scan limits to match regulatory reqs

In 802.11b/g mode, bcm43xx actively scans channels 1-14 no matter what
locale has been set, either in the sprom or by the locale option. This
behaviorviolates regulatory rules everywhere in the world except
Japan. This patch changes the default range to the correct value if the
locale has been set, and to channels 1-13 if no locale has been set.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years ago[PATCH] zd1211rw: Fix E2P_PHY_REG patching
Daniel Drake [Tue, 3 Apr 2007 22:17:37 +0000 (23:17 +0100)] 
[PATCH] zd1211rw: Fix E2P_PHY_REG patching

Due to conflicting/confusing defines in the vendor driver, we were
reading E2P_PHY_REG from the wrong location.

CR157 patching was slightly incorrect in that the vendor driver only
patches in an 8-bit value, whereas we were patching 24 bits.

Additionally, CR157 patching was happening on both zd1211 and zd1211b,
but this should only happen on zd1211.

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: Ulrich Kunitz <kune@deine-taler.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years agoIPoIB/cm: Fix DMA direction typo
Michael S. Tsirkin [Tue, 10 Apr 2007 15:32:42 +0000 (18:32 +0300)] 
IPoIB/cm: Fix DMA direction typo

Receive buffers need to be mapped with DMA_FROM_DEVICE.  Incorrectly
mapping with DMA_TO_DEVICE causes a hard lock on ppc64 machines with
an IOMMU.

This fixes <https://bugs.openfabrics.org/show_bug.cgi?id=431>

Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
17 years ago[PATCH] zd1211rw: Reject AL2230S devices
Daniel Drake [Sun, 11 Mar 2007 19:54:11 +0000 (19:54 +0000)] 
[PATCH] zd1211rw: Reject AL2230S devices

zd1211rw currently detects AL2230S-based devices as AL2230, and hence
programs the RF incorrectly. Transmit silently fails on this
misconfiguration.

After this patch, AL2230S devices are rejected with an error message, to
avoid any confusion with an apparent driver bug.

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
17 years ago[AGPGART] intel_agp: fix G965 GTT size detect
Wang Zhenyu [Tue, 10 Apr 2007 01:42:48 +0000 (09:42 +0800)] 
[AGPGART] intel_agp: fix G965 GTT size detect

On G965, I810_PGETBL_CTL is a mmio offset, but we wrongly take it
as pci config space offset in detecting GTT size. This one line patch
fixs this.

Signed-off-by: Wang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: Dave Jones <davej@redhat.com>
17 years ago[TCP]: slow_start_after_idle should influence cwnd validation too
David S. Miller [Mon, 9 Apr 2007 20:23:14 +0000 (13:23 -0700)] 
[TCP]: slow_start_after_idle should influence cwnd validation too

For the cases that slow_start_after_idle are meant to deal
with, it is almost a certainty that the congestion window
tests will think the connection is application limited and
we'll thus decrease the cwnd there too.  This defeats the
whole point of setting slow_start_after_idle to zero.

So test it there too.

We do not cancel out the entire tcp_cwnd_validate() function
so that if the sysctl is changed we still have the validation
state maintained.

Signed-off-by: David S. Miller <davem@davemloft.net>
17 years ago[SC92031]: Fix priv->lock context
Herbert Xu [Mon, 9 Apr 2007 18:51:15 +0000 (11:51 -0700)] 
[SC92031]: Fix priv->lock context

The spin_lock calls made in dev->open and dev->close must disable
BH since open/close are made in process context.  Conversely, the
call in dev->hard_start_xmit does not need to disable BH since it
is already executing with BH disabled.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
17 years ago[NET_SCHED]: cls_tcindex: fix compatibility breakage
Patrick McHardy [Mon, 9 Apr 2007 18:42:25 +0000 (11:42 -0700)] 
[NET_SCHED]: cls_tcindex: fix compatibility breakage

Userspace uses an integer for TCA_TCINDEX_SHIFT, the kernel was changed
to expect and use a u16 value in 2.6.11, which broke compatibility on
big endian machines. Change back to use int.

Reported by Ole Reinartz <ole.reinartz@gmx.de>

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
17 years agoMerge branch 'for-2.6.21' of master.kernel.org:/pub/scm/linux/kernel/git/davej/agpgart
Linus Torvalds [Mon, 9 Apr 2007 18:45:18 +0000 (11:45 -0700)] 
Merge branch 'for-2.6.21' of /linux/kernel/git/davej/agpgart

* 'for-2.6.21' of master.kernel.org:/pub/scm/linux/kernel/git/davej/agpgart:
  [AGPGART] intel_agp: PCI id update for Intel 965GM

17 years agoieee1394: change deprecation status of dv1394
Stefan Richter [Wed, 21 Mar 2007 23:40:06 +0000 (00:40 +0100)] 
ieee1394: change deprecation status of dv1394

Nobody ported ffmpeg from dv1394 to rawiso yet, and there is no
justification to remove dv1394 right now.

Nevertheless, a strong deprecation of this ABI makes a lot of sense,
especially as Kristian H's drivers shape up to be an attractive
alternative to the existing ones.  But we don't have a schedule at the
moment.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
17 years ago[PATCH] i386: irqbalance_disable() section fix
Andrew Morton [Sun, 8 Apr 2007 23:04:04 +0000 (16:04 -0700)] 
[PATCH] i386: irqbalance_disable() section fix

WARNING: arch/i386/kernel/built-in.o - Section mismatch: reference to .init.text:irqbalance_disable from .text between 'quirk_intel_irqbalance' (at offset 0x80a5) and 'i8237A_suspend'

Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] x86_64 early quirks: fix early_qrk[] section tag
Andrew Morton [Sun, 8 Apr 2007 23:04:03 +0000 (16:04 -0700)] 
[PATCH] x86_64 early quirks: fix early_qrk[] section tag

WARNING: arch/x86_64/kernel/built-in.o - Section mismatch: reference to .init.text:nvidia_bugs from .data between 'early_qrk' (at offset 0x8428) and 'enable_cpu_hotplug'
WARNING: arch/x86_64/kernel/built-in.o - Section mismatch: reference to .init.text:via_bugs from .data between 'early_qrk' (at offset 0x8438) and 'enable_cpu_hotplug'
WARNING: arch/x86_64/kernel/built-in.o - Section mismatch: reference to .init.text:ati_bugs from .data between 'early_qrk' (at offset 0x8448) and 'enable_cpu_hotplug'

The compiler is putting it into .data because the __initdata is in the wrong
place.

Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] omap_cf: oops-on-suspend fix
David Brownell [Sun, 8 Apr 2007 23:04:02 +0000 (16:04 -0700)] 
[PATCH] omap_cf: oops-on-suspend fix

Convert omap_cf into a platform_driver ...  this resolves oopsing during
suspend/resume.

Evidently folk haven't tried suspend/resume on an OSK (the main platform
for this driver) since September or so, which is when platform_device
learned about suspend_late()/resume_early() and stopped being able to
suspend/resume without a platform_driver.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] Proper fix for highmem kmap_atomic functions for VMI for 2.6.21
Zachary Amsden [Sun, 8 Apr 2007 23:04:01 +0000 (16:04 -0700)] 
[PATCH] Proper fix for highmem kmap_atomic functions for VMI for 2.6.21

Since lazy MMU batching mode still allows interrupts to enter, it is
possible for interrupt handlers to try to use kmap_atomic, which fails when
lazy mode is active, since the PTE update to highmem will be delayed.  The
best workaround is to issue an explicit flush in kmap_atomic_functions
case; this is the only way nested PTE updates can happen in the interrupt
handler.

Thanks to Jeremy Fitzhardinge for noting the bug and suggestions on a fix.

This patch gets reverted again when we start 2.6.22 and the bug gets fixed
differently.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Cc: Andi Kleen <ak@muc.de>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] fuse: validate rootmode mount option
Timo Savola [Sun, 8 Apr 2007 23:04:00 +0000 (16:04 -0700)] 
[PATCH] fuse: validate rootmode mount option

If rootmode isn't valid, we hit the BUG() in fuse_init_inode.  Now
EINVAL is returned.

Signed-off-by: Timo Savola <tsavola@movial.fi>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[AGPGART] intel_agp: PCI id update for Intel 965GM
Wang Zhenyu [Mon, 9 Apr 2007 00:51:36 +0000 (08:51 +0800)] 
[AGPGART] intel_agp: PCI id update for Intel 965GM

Update PCI id info for Intel 965GM chipset.

Signed-off-by: Wang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: Dave Jones <davej@redhat.com>
17 years agosched.c: Remove unused variable 'relative'
Linus Torvalds [Sat, 7 Apr 2007 17:18:33 +0000 (10:18 -0700)] 
sched.c: Remove unused variable 'relative'

Getting rid of the p->children printout in show_task() left behind an
unused variable.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] sched: get rid of p->children use in show_task()
Ingo Molnar [Fri, 6 Apr 2007 19:18:06 +0000 (21:18 +0200)] 
[PATCH] sched: get rid of p->children use in show_task()

the p->parent PID printout gives us all the information about the
task tree that we need - the eldest_child()/older_sibling()/
younger_sibling() printouts are mostly historic and i do not
remember ever having used those fields. (IMO in fact they confuse
the SysRq-T output.) So remove them.

This code has sentimental value though, those fields and
printouts are one of the oldest ones still surviving from
Linux v0.95's kernel/sched.c:

        if (p->p_ysptr || p->p_osptr)
                printk("   Younger sib=%d, older sib=%d\n\r",
                        p->p_ysptr ? p->p_ysptr->pid : -1,
                        p->p_osptr ? p->p_osptr->pid : -1);
        else
                printk("\n\r");

written 15 years ago, in early 1992.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus 'snif' Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] irq-devres: fix failure path of devm_request_irq()
Tejun Heo [Sat, 7 Apr 2007 05:59:41 +0000 (14:59 +0900)] 
[PATCH] irq-devres: fix failure path of devm_request_irq()

devres should be deallocated with devres_free() not kfree().  This bug
corrupts slab on IRQ request failure.  Fix it.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Greg KH <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] high-res timers: resume fix
Ingo Molnar [Sat, 7 Apr 2007 10:05:00 +0000 (12:05 +0200)] 
[PATCH] high-res timers: resume fix

Soeren Sonnenburg reported that upon resume he is getting
this backtrace:

 [<c0119637>] smp_apic_timer_interrupt+0x57/0x90
 [<c0142d30>] retrigger_next_event+0x0/0xb0
 [<c0104d30>] apic_timer_interrupt+0x28/0x30
 [<c0142d30>] retrigger_next_event+0x0/0xb0
 [<c0140068>] __kfifo_put+0x8/0x90
 [<c0130fe5>] on_each_cpu+0x35/0x60
 [<c0143538>] clock_was_set+0x18/0x20
 [<c0135cdc>] timekeeping_resume+0x7c/0xa0
 [<c02aabe1>] __sysdev_resume+0x11/0x80
 [<c02ab0c7>] sysdev_resume+0x47/0x80
 [<c02b0b05>] device_power_up+0x5/0x10

it turns out that on resume we mistakenly re-enable interrupts too
early.  Do the timer retrigger only on the current CPU.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Soeren Sonnenburg <kernel@nn7.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Sat, 7 Apr 2007 01:15:30 +0000 (18:15 -0700)] 
Merge /pub/scm/linux/kernel/git/davem/net-2.6

* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6:
  [TG3]: Fix crash during tg3_init_one().
  [IPV6]: Revert recent change to rt6_check_dev().
  [XFRM]: beet: fix IP option decapsulation
  [XFRM]: beet: fix beet mode decapsulation
  [XFRM]: beet: use IPOPT_NOP for option padding
  [XFRM]: beet: fix IP option encapsulation

17 years ago[TG3]: Fix crash during tg3_init_one().
Michael Chan [Sat, 7 Apr 2007 00:35:37 +0000 (17:35 -0700)] 
[TG3]: Fix crash during tg3_init_one().

The driver will crash when the chip has been initialized by EFI before
tg3_init_one().  In this case, the driver will call tg3_chip_reset()
before allocating consistent memory.

The bug is fixed by checking for tp->hw_status before accessing it
during tg3_chip_reset().

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
17 years ago[IA64] SGI Altix : fix pcibr_dmamap_ate32() bug
Mike Habeck [Fri, 6 Apr 2007 17:04:39 +0000 (12:04 -0500)] 
[IA64] SGI Altix : fix pcibr_dmamap_ate32() bug

On a SGI Altix TIOCP based PCI bus we need to include the ATE_PIO attribute
bit if we're mapping a 32bit MSI address.

Signed-off-by: Mike Habeck <habeck@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
17 years ago[IA64] Fix CPU freq displayed in /proc/cpuinfo
Venki Pallipadi [Fri, 6 Apr 2007 17:04:49 +0000 (10:04 -0700)] 
[IA64] Fix CPU freq displayed in /proc/cpuinfo

My patch: git commit=95235ca2c20ac0b31a8eb39e2d599bcc3e9c9a10 introduced a bug
in IA64 cpuinfo output.

Patch changed the proc_freq from 1HZ resolution to 1KHz resolution, but left
format string unchanged at " %lu.%06lu". Below is the fix.

Thanks to Bjorn for catching this.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
17 years ago[IA64] Fix wrong assumption about irq and vector in msi_ia64.c
Ishimatsu Yasuaki [Fri, 6 Apr 2007 07:51:12 +0000 (16:51 +0900)] 
[IA64] Fix wrong assumption about irq and vector in msi_ia64.c

This patch fixes a wrong assumption in ia64 MSI code that IRQ equals
vector.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
17 years ago[IA64] BTE error timer fix
Russ Anderson [Fri, 30 Mar 2007 22:13:18 +0000 (17:13 -0500)] 
[IA64] BTE error timer fix

The bte recovery_timer was not being set correctly.

Signed-off-by: Russ Anderson <rja@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
17 years ago[IPV6]: Revert recent change to rt6_check_dev().
David S. Miller [Fri, 6 Apr 2007 18:42:27 +0000 (11:42 -0700)] 
[IPV6]: Revert recent change to rt6_check_dev().

This reverts a0d78ebf3a0e33a1aeacf2fc518ad9273d6a1c2f

It causes pings to link-local addresses to fail.

Signed-off-by: David S. Miller <davem@davemloft.net>
17 years agoLinux 2.6.21-rc6 v2.6.21-rc6
Linus Torvalds [Fri, 6 Apr 2007 02:36:56 +0000 (19:36 -0700)] 
Linux 2.6.21-rc6

.. perfect? Ahh, sure.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[XFRM]: beet: fix IP option decapsulation
Patrick McHardy [Thu, 5 Apr 2007 23:03:33 +0000 (16:03 -0700)] 
[XFRM]: beet: fix IP option decapsulation

Beet mode looks for the beet pseudo header after the outer IP header,
which is wrong since that is followed by the ESP header. Additionally
it needs to adjust the packet length after removing the pseudo header
and point the data pointer to the real data location.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
17 years ago[XFRM]: beet: fix beet mode decapsulation
Patrick McHardy [Thu, 5 Apr 2007 22:59:41 +0000 (15:59 -0700)] 
[XFRM]: beet: fix beet mode decapsulation

Beet mode decapsulation fails to properly set up the skb pointers, which
only works by accident in combination with CONFIG_NETFILTER, since in that
case the skb is fixed up in xfrm4_input before passing it to the netfilter
hooks.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
17 years ago[XFRM]: beet: use IPOPT_NOP for option padding
Patrick McHardy [Thu, 5 Apr 2007 22:54:39 +0000 (15:54 -0700)] 
[XFRM]: beet: use IPOPT_NOP for option padding

draft-nikander-esp-beet-mode-07.txt states "The padding MUST be filled
with NOP options as defined in Internet Protocol [1] section 3.1
Internet header format.", so do that.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
17 years ago[XFRM]: beet: fix IP option encapsulation
Patrick McHardy [Thu, 5 Apr 2007 22:54:02 +0000 (15:54 -0700)] 
[XFRM]: beet: fix IP option encapsulation

Beet mode calculates an incorrect value for the transport header location
when IP options are present, resulting in encapsulation errors.

The correct location is 4 or 8 bytes before the end of the original IP
header, depending on whether the pseudo header is padded.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
17 years agoIB/iser: Don't defer connection failure notification to workqueue
Erez Zilber [Sun, 1 Apr 2007 10:53:43 +0000 (12:53 +0200)] 
IB/iser: Don't defer connection failure notification to workqueue

When a connection is terminated asynchronously from the iSCSI layer's
perspective, iSER needs to notify the iSCSI layer that the connection
has failed.  This is done using a workqueue (switched to from the iSER
tasklet context).  Meanwhile, the connection object (that holds the
work struct) is released.  If the workqueue function wasn't called
yet, it will be called later with a NULL pointer, which will crash the
kernel.

The context switch (tasklet to workqueue) is not required, and
everything can be done from the iSER tasklet. This eliminates the NULL
work struct bug (and simplifies the code).

Signed-off-by: Erez Zilber <erezz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
17 years agoHID: Do not discard truncated input reports
Adam Kropelin [Thu, 5 Apr 2007 14:06:30 +0000 (16:06 +0200)] 
HID: Do not discard truncated input reports

Truncated reports should not be discarded since it prevents buggy
devices from communicating with userspace.

Prior to the regession introduced in 2.6.20, a shorter-than-expected
report in hid_input_report() was passed thru after having the missing
bytes cleared. This behavior was established over a few patches in the
2.6.early-teens days, including commit
cd6104572bca9e4afe0dcdb8ecd65ef90b01297b.

This patch restores the previous behavior and fixes the regression.

Signed-off-by: Adam Kropelin <akropel1@rochester.rr.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
17 years ago[IPSEC]: Reject packets within replay window but outside the bit mask
Herbert Xu [Thu, 5 Apr 2007 07:07:39 +0000 (00:07 -0700)] 
[IPSEC]: Reject packets within replay window but outside the bit mask

Up until this point we've accepted replay window settings greater than
32 but our bit mask can only accomodate 32 packets.  Thus any packet
with a sequence number within the window but outside the bit mask would
be accepted.

This patch causes those packets to be rejected instead.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
17 years ago[IPv6]: Exclude truncated packets from InHdrErrors statistics
Mitsuru Chinen [Thu, 5 Apr 2007 06:54:59 +0000 (23:54 -0700)] 
[IPv6]: Exclude truncated packets from InHdrErrors statistics

Incoming trancated packets are counted as not only InTruncatedPkts but
also InHdrErrors. They should be counted as InTruncatedPkts only.

Signed-off-by: Mitsuru Chinen <mitch@linux.vnet.ibm.com>
Acked-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
17 years ago[APPLETALK]: Fix a remotely triggerable crash
Jean Delvare [Thu, 5 Apr 2007 06:52:46 +0000 (23:52 -0700)] 
[APPLETALK]: Fix a remotely triggerable crash

When we receive an AppleTalk frame shorter than what its header says,
we still attempt to verify its checksum, and trip on the BUG_ON() at
the end of function atalk_sum_skb() because of the length mismatch.

This has security implications because this can be triggered by simply
sending a specially crafted ethernet frame to a target victim,
effectively crashing that host. Thus this qualifies, I think, as a
remote DoS. Here is the frame I used to trigger the crash, in npg
format:

<Appletalk Killer>
{
# Ethernet header -----

  XX XX XX XX XX XX  # Destination MAC
  00 00 00 00 00 00  # Source MAC
  00 1D              # Length

# LLC header -----

  AA AA 03
  08 00 07 80 9B  # Appletalk

# Appletalk header -----

  00 1B        # Packet length (invalid)
  00 01        # Fake checksum
  00 00 00 00  # Destination and source networks
  00 00 00 00  # Destination and source nodes and ports

# Payload -----

  0C 0D 0E 0F 10 11 12 13
  14
}

The destination MAC address must be set to those of the victim.

The severity is mitigated by two requirements:
* The target host must have the appletalk kernel module loaded. I
  suspect this isn't so frequent.
* AppleTalk frames are non-IP, thus I guess they can only travel on
  local networks. I am no network expert though, maybe it is possible
  to somehow encapsulate AppleTalk packets over IP.

The bug has been reported back in June 2004:
  http://bugzilla.kernel.org/show_bug.cgi?id=2979
But it wasn't investigated, and was closed in July 2006 as both
reporters had vanished meanwhile.

This code was new in kernel 2.6.0-test5:
  http://git.kernel.org/?p=linux/kernel/git/tglx/history.git;a=commitdiff;h=7ab442d7e0a76402c12553ee256f756097cae2d2
And not modified since then, so we can assume that vanilla kernels
2.6.0-test5 and later, and distribution kernels based thereon, are
affected.

Note that I still do not know for sure what triggered the bug in the
real-world cases. The frame could have been corrupted by the kernel if
we have a bug hiding somewhere. But more likely, we are receiving the
faulty frame from the network.

Signed-off-by: Jean Delvare <jdelvare@suse.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
17 years ago[PATCH] ia64: desc_empty thinko/typo fix
Maciej Zenczykowski [Thu, 5 Apr 2007 02:08:27 +0000 (19:08 -0700)] 
[PATCH] ia64: desc_empty thinko/typo fix

Just a one-byter for an ia64 thinko/typo - already fixed for i386 and x86_64.

Acked-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] Fix build error on zs serial driver
Yoichi Yuasa [Thu, 5 Apr 2007 02:08:25 +0000 (19:08 -0700)] 
[PATCH] Fix build error on zs serial driver

drivers/tc/zs.c:73:24: error: asm/dec/tc.h: No such file or directory

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Acked-by: Maciej W. Rozycki <macro@linux-mips.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] fix jiffies clocksource inittime
john stultz [Thu, 5 Apr 2007 02:08:24 +0000 (19:08 -0700)] 
[PATCH] fix jiffies clocksource inittime

In debugging a problem w/ the -rt tree, I noticed that on systems that mark
the tsc as unstable before it is registered, the TSC would still be
selected and used for a short period of time.  Digging in it looks to be a
result of the mix of the clocksource list changes and my clocksource
initialization changes.

With the -rt tree, using a bad TSC, even for a short period of time can
results in a hang at boot.  I was not able to reproduce this hang w/
mainline, but I'm not completely certain that someone won't trip on it.

This patch resolves the issue by initializing the jiffies clocksource
earlier so a bad TSC won't get selected just because nothing else is yet
registered.

Signed-off-by: John Stultz <johnstul@us.ibm.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] cciss: add init of drv->cylinders back to cciss_geometry_inquiry
Mike Miller (OS Dev) [Thu, 5 Apr 2007 02:08:23 +0000 (19:08 -0700)] 
[PATCH] cciss: add init of drv->cylinders back to cciss_geometry_inquiry

This patch adds initialization of drv->cylinders back into the failing case in
cciss_geometry_inquiry. I inadvertently removed it in one my 2TB updates.

Signed-off-by: Mike Miller <mike.miller@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] remove protection of LANANA-reserved majors
Andrew Morton [Thu, 5 Apr 2007 02:08:22 +0000 (19:08 -0700)] 
[PATCH] remove protection of LANANA-reserved majors

Revert all this.  It can cause device-mapper to receive a different major from
earlier kernels and it turns out that the Amanda backup program (via GNU tar,
apparently) checks major numbers on files when performing incremental backups.

Which is a bit broken of Amanda (or tar), but this feature isn't important
enough to justify the churn.

Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] swsusp: fix memory shrinker
Rafael J. Wysocki [Thu, 5 Apr 2007 02:08:21 +0000 (19:08 -0700)] 
[PATCH] swsusp: fix memory shrinker

Fix a bug in the swsusp's memory shrinker that causes some systems using
highmem to refuse to suspend to disk if image_size is set above 1/2 of
available RAM.

Special thanks to Jiri Slaby for reporting the problem and assistance in
debugging it.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Jiri Slaby <jirislaby@gmail.com>
Cc: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] net/sunrpc/svcsock.c: fix a check
Adrian Bunk [Thu, 5 Apr 2007 02:08:21 +0000 (19:08 -0700)] 
[PATCH] net/sunrpc/svcsock.c: fix a check

The return value of kernel_recvmsg() should be assigned to "err", not
compared with the random value of a never initialized "err" (and the "< 0"
check wrongly always returned false since == comparisons never have a
result < 0).

Spotted by the Coverity checker.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] md: avoid a deadlock when removing a device from an md array via sysfs
NeilBrown [Thu, 5 Apr 2007 02:08:18 +0000 (19:08 -0700)] 
[PATCH] md: avoid a deadlock when removing a device from an md array via sysfs

A device can be removed from an md array via e.g.
  echo remove > /sys/block/md3/md/dev-sde/state

This will try to remove the 'dev-sde' subtree which will deadlock
since
  commit e7b0d26a86943370c04d6833c6edba2a72a6e240

With this patch we run the kobject_del via schedule_work so as to
avoid the deadlock.

Cc: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] kernel/time.c: add missing symbol exports
Thomas Bittermann [Wed, 4 Apr 2007 20:20:54 +0000 (22:20 +0200)] 
[PATCH] kernel/time.c: add missing symbol exports

This patch adds 2 missing symbol exports: jiffies_to_timeval() and
timeval_to_jiffies().  The (not yet merged) dm-raid4-5 module will need
them, and they used to be indirectly exported by virtue of being inline
functions.

Commit 8b9365d753d9870bb6451504c13570b81923228f ("[PATCH] Uninline
jiffies.h functions") uninlined them, and thus modules now need them
explicitly exported to use them.

Signed-off-by: Thomas Bittermann <t.bittermann@online.de>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: john stultz <johnstul@us.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years agoMerge branch 'for-linus' of git://git390.osdl.marist.edu/pub/scm/linux-2.6
Linus Torvalds [Wed, 4 Apr 2007 17:11:16 +0000 (10:11 -0700)] 
Merge branch 'for-linus' of git://git390.osdl.marist.edu/linux-2.6

* 'for-linus' of git://git390.osdl.marist.edu/pub/scm/linux-2.6:
  [S390] cio: Fix handling of interrupt for csch().
  [S390] page_mkclean data corruption.

17 years ago[PATCH] net: Ignore sysfs network device rename bugs.
Eric W. Biederman [Tue, 3 Apr 2007 06:07:30 +0000 (00:07 -0600)] 
[PATCH] net: Ignore sysfs network device rename bugs.

The generic networking code ensures that no two networking devices
have the same name, so  there is no time except when sysfs has
implementation bugs that device_rename when called from
dev_change_name will fail.

The current error handling for errors from device_rename in
dev_change_name is wrong and results in an unusable and unrecoverable
network device if device_rename is happens to return an error.

This patch removes the buggy error handling.  Which confines the mess
when device_rename hits a problem to sysfs, instead of propagating it
the rest of the network stack.  Making linux a little more robust.

Without this patch you can observe what happens when sysfs has a bug
when CONFIG_SYSFS_DEPRECATED is not set and you attempt to rename
a real network device to a name like (broken_parity_status, device,
modalias, power, resource2, subsystem_vendor, class,  driver, irq,
msi_bus, resource, subsystem, uevent, config, enable, local_cpus,
numa_node, resource0, subsystem_device, vendor)

Greg has a patch that fixes the sysfs bugs but he doesn't trust it
for a 2.6.21 timeframe.  This patch which just ignores errors should
be safe and it keeps the system from going completely wacky.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
17 years ago[PATCH] SLAB: Mention slab name when listing corrupt objects
David Howells [Mon, 2 Apr 2007 22:46:28 +0000 (23:46 +0100)] 
[PATCH] SLAB: Mention slab name when listing corrupt objects

Mention the slab name when listing corrupt objects.  Although the function
that released the memory is mentioned, that is frequently ambiguous as such
functions often release several pieces of memory.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>