]> git.ipfire.org Git - thirdparty/kernel/stable.git/log
thirdparty/kernel/stable.git
20 years ago[PATCH] Fix signedness problem at socket creation
Marcel Holtmann [Sat, 26 Mar 2005 01:49:05 +0000 (17:49 -0800)] 
[PATCH] Fix signedness problem at socket creation

CAN-2005-0750 is assigned to this issue

ilja <ilja@suresec.org> discovered potential local root exploit in
bluetooth socket creation.

This patch fixes a small signedness problem when creating the
socket.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Chris Wright <chrisw@osdl.org>
20 years ago[PATCH] isofs: more defensive checks against corrupt isofs images
chrisw@osdl.org [Sat, 26 Mar 2005 01:46:03 +0000 (17:46 -0800)] 
[PATCH] isofs: more defensive checks against corrupt isofs images

Michal Zalewski <lcamtuf@dione.ids.pl> discovers range checking flaws in
iso9660 filesystem.

http://marc.theaimsgroup.com/?l=bugtraq&m=111110067304783&w=2

CAN-2005-0815 is assigned to this issue.

Some more defensive checks to keep corrupt isofs images from corrupting
memory or causing Oops.

Signed-off-by: Chris Wright <chrisw@osdl.org>
===== fs/isofs/rock.c 1.23 vs edited =====

20 years ago[PATCH] isofs: more "corrupted iso image" error cases
torvalds@osdl.org [Sat, 26 Mar 2005 01:45:24 +0000 (17:45 -0800)] 
[PATCH] isofs: more "corrupted iso image" error cases

Michal Zalewski <lcamtuf@dione.ids.pl> discovers range checking flaws in
iso9660 filesystem.

http://marc.theaimsgroup.com/?l=bugtraq&m=111110067304783&w=2

CAN-2005-0815 is assigned to this issue.

From: Linus Torvalds <torvalds@osdl.org>

isofs: more "corrupted iso image" error cases

Thanks to Michal Zalewski for testing.

Signed-off-by: Chris Wright <chrisw@osdl.org>
20 years ago[PATCH] isofs: Handle corupted rock-ridge info slightly better
torvalds@osdl.org [Sat, 26 Mar 2005 01:44:34 +0000 (17:44 -0800)] 
[PATCH] isofs: Handle corupted rock-ridge info slightly better

Michal Zalewski <lcamtuf@dione.ids.pl> discovers range checking flaws in
iso9660 filesystem.

http://marc.theaimsgroup.com/?l=bugtraq&m=111110067304783&w=2

CAN-2005-0815 is assigned to this issue.

From: Linus Torvalds <torvalds@osdl.org>

isofs: Handle corupted rock-ridge info slightly better.

Keyword here being 'slightly'. The code is a mess.

Signed-off-by: Chris Wright <chrisw@osdl.org>
20 years ago[PATCH] Linux 2.6.11.5
gregkh@suse.de [Sat, 19 Mar 2005 06:11:36 +0000 (22:11 -0800)] 
[PATCH] Linux 2.6.11.5

20 years ago[PATCH] Fix kernel panic on receive with WAN Hitachi SCA HD6457x
khc@pm.waw.pl [Sat, 19 Mar 2005 05:47:12 +0000 (21:47 -0800)] 
[PATCH] Fix kernel panic on receive with WAN Hitachi SCA HD6457x

Another patch for 2.6.11.x: already in main tree, fixes kernel panic on
receive with WAN cards based on Hitachi SCA/SCA-II: N2, C101, PCI200SYN.
The attached patch fixes NULL pointer dereference on RX.

Signed-off-by: Krzysztof Halasa <khc@pm.waw.pl>
Acked-by: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] Possible VIA-Rhine free irq issue
dilinger@debian.org [Sat, 19 Mar 2005 05:43:33 +0000 (21:43 -0800)] 
[PATCH] Possible VIA-Rhine free irq issue

It seems to me that in the VIA Rhine device driver the requested irq might
not be freed in case the alloc_ring() function fails. alloc_ring()
can fail with a ENOMEM return value because of possible
pci_alloc_consistent() failures.

Updated to CodingStyle.

Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] tasklist left locked
hugh@veritas.com [Sat, 19 Mar 2005 05:41:52 +0000 (21:41 -0800)] 
[PATCH] tasklist left locked

On 4-way SMP, about one reboot in twenty hangs while killing processes:
exit needs exclusive tasklist_lock, but something still holds read_lock.
do_signal_stop race case misses unlock, and fixing it fixes the symptom.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] Fix check for underflow
shemminger@osdl.org [Sat, 19 Mar 2005 05:40:56 +0000 (21:40 -0800)] 
[PATCH] Fix check for underflow

http://bugme.osdl.org/show_bug.cgi?id=4279
Summary: When I try to start vpnc the net/core/skbuff.c:91 crash

This check is wrong, gcc optimizes it away:

                if ((len -= sizeof(pi)) > len)
return -EINVAL;

This could be responsible for the BUG. If len is 2 or 3 and TUN_NO_PI
isn't set it underflows. alloc_skb() allocates len + 2, which is 0 or
1 byte. skb_reserve tries to reserve 2 bytes and things explode in
skb_put.

[TUN]: Fix check for underflow

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] Fix __xfrm_find_acq_byseq()
davem@davemloft.net [Sat, 19 Mar 2005 05:39:03 +0000 (21:39 -0800)] 
[PATCH] Fix __xfrm_find_acq_byseq()

This function, as it's name implies, is supposed to only
return IPSEC objects which are in the XFRM_STATE_ACQ
("acquire") state.  But it returns any object with the
matching sequence number.

This is wrong and confuses IPSEC daemons to no end.

[XFRM]: xfrm_find_acq_byseq should only return XFRM_STATE_ACQ states.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] Fix crash while reading /proc/net/route
kaber@trash.net [Sat, 19 Mar 2005 05:37:15 +0000 (21:37 -0800)] 
[PATCH] Fix crash while reading /proc/net/route

[IPV4]: Fix crash while reading /proc/net/route caused by stale pointers

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] Possible AMD8111e free irq issue
dilinger@debian.org [Sat, 19 Mar 2005 05:32:44 +0000 (21:32 -0800)] 
[PATCH] Possible AMD8111e free irq issue

It seems to me that if in the amd8111e_open() fuction dev->irq isn't
zero and the irq request succeeds it might not get released anymore.

Specifically, on failure of the amd8111e_restart() call the function
returns -ENOMEM without releasing the irq. The amd8111e_restart()
function can fail because of various pci_alloc_consistent() and
dev_alloc_skb() calls in amd8111e_init_ring() which is being
called by amd8111e_restart.

1374     if(dev->irq ==0 || request_irq(dev->irq, amd8111e_interrupt, SA_SHIRQ,
1375                      dev->name, dev))
1376         return -EAGAIN;

Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] Fix stereo mutes on Surround volume control
dsd@gentoo.org [Sat, 19 Mar 2005 05:32:08 +0000 (21:32 -0800)] 
[PATCH] Fix stereo mutes on Surround volume control

As of 2.6.11, I have no output out of the rear right speaker of my 4.1
surround sound setup. I am using snd-intel8x0 based on a Realtek ALC650F chip
on an nvidia motherboard.

A gentoo user with completely different hardware also ran into this:
http://bugs.gentoo.org/84276

2.6.11-mm3 fixes this problem and I've identified the specific fix, which is
already in the ALSA development tree. An ALSA developer asked me to submit the
fix for 2.6.11.x when I'd found it, so here it is :)

--
AC97 Codec
Fix stereo mutes on Surround volume control.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] NetROM locking
ralf@linux-mips.org [Sat, 19 Mar 2005 05:30:08 +0000 (21:30 -0800)] 
[PATCH] NetROM locking

Fix deadlock in NetROM due to double locking.  I was sent the patch by
Alan and have doublechecked it.  This bug hits Net/ROM users really hard.
It's accepted by DaveM - but just too late to make it into 2.6.11.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] Linux 2.6.11.4
gregkh@suse.de [Tue, 15 Mar 2005 23:50:14 +0000 (15:50 -0800)] 
[PATCH] Linux 2.6.11.4

20 years ago[PATCH] use strncpy in get_task_comm
Prasanna Meda [Tue, 15 Mar 2005 23:39:08 +0000 (15:39 -0800)] 
[PATCH] use strncpy in get_task_comm

Set_task_comm uses strlcpy, so get_task_comm must use strncpy.

Signed-Off-by: Prasanna Meda <pmeda@akamai.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] CAN-2005-0384: Remote Linux DoS on ppp servers
paulus@au1.ibm.com [Tue, 15 Mar 2005 23:38:47 +0000 (15:38 -0800)] 
[PATCH] CAN-2005-0384: Remote Linux DoS on ppp servers

Martin Schulze writes:

> Ben Martel and Stephen Blackheath have discovered a denial-of-service attack
> that a client of pppd can make that can hang the server machine.  The bug is
> in the Linux kernel 2.6 (tested on 2.6.9), but it looks like it also exists
> in the 2.4 series.

Yes, this is my bug. :(

I would just do this instead:

Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] Linux 2.6.11.3
gregkh@suse.de [Sun, 13 Mar 2005 06:31:28 +0000 (22:31 -0800)] 
[PATCH] Linux 2.6.11.3

20 years ago[PATCH] older chips oops on shutdown
olof@austin.ibm.com [Sun, 13 Mar 2005 05:47:59 +0000 (21:47 -0800)] 
[PATCH] older chips oops on shutdown

Kernel 2.6.11, hardware is a MSI KT333-based board with an XP1800.

I'm oopsing on shutdown on a machine that has a Via Rhine adapter in it:

Unable to handle kernel paging request at virtual address e0803003
  printing eip:
c01f262c
*pde = 014dc067
*pte = 00000000
Oops: 0000 [#1]
Modules linked in: cpufreq_userspace cpufreq_powersave cpufreq_ondemand
CPU:    0
EIP:    0060:[<c01f262c>]    Not tainted VLI
EFLAGS: 00010292   (2.6.11)
EIP is at ioread8+0x2c/0x40
eax: e0803003   ebx: e0803003   ecx: c026b430   edx: e0803003
esi: dff90260   edi: e0802f80   ebp: dd117e74   esp: dd117e74
ds: 007b   es: 007b   ss: 0068
Process reboot (pid: 5769, threadinfo=dd117000 task=dfafa080)
Stack: dd117e8c c026b490 dff90040 c151ccd4 c044a1a8 b7fdc078 dd117ea4
c0253ad9
        c151ccd4 00000042 fee1dead 00000001 dd117fbc c012461c c04d72a8 00000001
        00000000 00010800 00000000 dd117ed8 c013b40b dffe7380 00030800 00000000
Call Trace:
  [<c0103d5f>] show_stack+0x7f/0xa0
  [<c0103efa>] show_registers+0x15a/0x1c0
  [<c01040ce>] die+0xce/0x150
  [<c0113406>] do_page_fault+0x356/0x692
  [<c01039ff>] error_code+0x2b/0x30
  [<c026b490>] rhine_shutdown+0x60/0x140
  [<c0253ad9>] device_shutdown+0x89/0x8b
  [<c012461c>] sys_reboot+0xac/0x200
  [<c0102f71>] sysenter_past_esp+0x52/0x75
Code: 3d ff ff 03 00 89 c2 89 e5 77 20 66 31 c0 3d 00 00 01 00 75 0c
81 e2 ff ff 00 00 ec 0f b6 c0 c9 c3 0f 0b 37 00 7b 65 3b c0 eb ea <0f>
b6 00 eb ec eb 0d 90 90 90 90 90 90 90 90 90 90 90 90 90 55

Seems like it is the ioread8 in:

         /* Hit power state D3 (sleep) */
         iowrite8(ioread8(ioaddr + StickyHW) | 0x03, ioaddr + StickyHW);

that fails. StickyHW is 0x83. lspci says:

0000:00:07.0 Ethernet controller: VIA Technologies, Inc. VT86C100A
[Rhine] (rev 06)
         Flags: bus master, medium devsel, latency 32, IRQ 18
         I/O ports at ec00 [size=128]
         Memory at dfffff80 (32-bit, non-prefetchable) [size=128]

In other words, it's trying to read outside of the I/O range (0x80),
which matches the fauling address.

I'm guessing my chip revision doesn't support WOL, it's a crappy noname
card.

It does seem as if rhine_power_init checks quirks for rqWOL before
touching any registers. Should rhine_shutdown do the same? Proposed
patch below, which resolves the problem on my system.

Check to make sure WOL is supported before setting it up in
rhine_shutdown.

Signed-off-by: Olof Johansson <olof@austin.ibm.com>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] sis900 kernel oops fix
chrisw@osdl.org [Sun, 13 Mar 2005 05:46:45 +0000 (21:46 -0800)] 
[PATCH] sis900 kernel oops fix

Backport of fix described below.

  From: Herbert Xu <herbert@gondor.apana.org.au>

  Fix bug #4223.

  OK, this happened because we got preempted before sis900_mii_probe
  finished setting the sis_priv->mii.  Theoretically this can happen
  with SMP as well but I suppose the number of SMP machines with sis900
  is fairly small.

  Anyway, the fix is to make sure that sis900_mii_probe is done before
  the device can be opened.  This patch does it by moving the setup
  before register_netdevice.

  Since the netdev name is not available before register_netdev, I've
  changed the relevant printk's to use pci_name instead.  Note that
  one of those printk's may be called after register_netdev as well.

Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] r8169: receive descriptor length fix
shemminger@osdl.org [Sun, 13 Mar 2005 05:46:01 +0000 (21:46 -0800)] 
[PATCH] r8169: receive descriptor length fix

The status and received packets indication in the Rx descriptor ring
are not correctly reset when a descriptor is recycled.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] PCI: fix hotplug double free
alexn@dsv.su.se [Sun, 13 Mar 2005 05:43:53 +0000 (21:43 -0800)] 
[PATCH] PCI: fix hotplug double free

[PATCH] PCI: fix hotplug double free

With the brackets missed out func could be freed twice.

Found by Coverity tool

Signed-off-by: Alexander Nyberg <alexn@dsv.su.se>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] ppc32: trivial fix for e500 oprofile build
galak@freescale.com [Sun, 13 Mar 2005 05:42:42 +0000 (21:42 -0800)] 
[PATCH] ppc32: trivial fix for e500 oprofile build

Fix for trivial fix for 2.6.11 oprofile compilation on e500 based ppc.

Signed-off-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] Put back tcp_timer_bug_msg[] symbol export.
davem@davemloft.net [Sun, 13 Mar 2005 05:41:54 +0000 (21:41 -0800)] 
[PATCH] Put back tcp_timer_bug_msg[] symbol export.

This wrecks the ipv6 modular build for a lot of people.
In fact, since I always build ipv6 modular I am surprised
I never hit this.  My best guess is that my compiler is
optimizing the reference away, but that can never be
depended upon and the symbol export really is needed.

[TCP]: Put back tcp_timer_bug_msg[] symbol export.

It is needed for tcp_reset_xmit_timer(), which is invoked by
tcp_prequeue() which is invoked from tcp_ipv6.c

Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] Fix i2c messsage flags in video drivers
khali@linux-fr.org [Sun, 13 Mar 2005 05:40:46 +0000 (21:40 -0800)] 
[PATCH] Fix i2c messsage flags in video drivers

While working on the saa7110 driver I found a problem with the way
various video drivers (found on Zoran-based boards) prepare i2c messages
to be used by i2c_transfer. The drivers improperly copy the i2c client
flags as the message flags, while both sets are mostly unrelated. The
net effect in this case is to trigger an I2C block read instead of the
expected I2C block write. The fix is simply not to pass any flag,
because none are needed.

I think this patch qualifies hands down as a "critical bug fix" to be
included in whatever bug-fix-only trees exist these days. As far as I
can see, all Zoran-based boards are broken in 2.6.11 without this patch.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] ppc32: Compilation fixes for Ebony, Luan and Ocotea
Matt Porter [Sun, 13 Mar 2005 05:39:19 +0000 (21:39 -0800)] 
[PATCH] ppc32: Compilation fixes for Ebony, Luan and Ocotea

this patch fixes the problem, that the current kernel (linux-2.6.11-rc5)
could not be compiled, when "support for early boot texts over serial port"
(CONFIG_SERIAL_TEXT_DEBUG=y) is active.

Signed-off-by: Gerhard Jaeger <gjaeger@sysgo.com>
Signed-off-by: Matt Porter <mporter@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] drm missing memset can crash X server..
airlied@linux.ie [Sun, 13 Mar 2005 05:38:27 +0000 (21:38 -0800)] 
[PATCH] drm missing memset can crash X server..

Egbert Eich reported a bug 2673 on bugs.freedesktop.org and tracked it
down to a missing memset in the setversion ioctl, this causes X server
crashes...

From: Egbert Eich <eich@pdx.freedesktop.org>
Signed-off-by: Dave Airlie <airlied@linux.ie>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] cramfs: small stat(2) fix
Eric Lammerts [Sun, 13 Mar 2005 05:35:18 +0000 (21:35 -0800)] 
[PATCH] cramfs: small stat(2) fix

When I stat(2) a device node on a cramfs, the st_blocks field is bogus
(it's derived from the size field which in this case holds the major/minor
numbers).  This makes du(1) output completely wrong.

Signed-off-by: Eric Lammerts <eric@lammerts.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] fix amd64 2.6.11 oops on modprobe (saa7110)
khali@linux-fr.org [Sun, 13 Mar 2005 05:33:36 +0000 (21:33 -0800)] 
[PATCH] fix amd64 2.6.11 oops on modprobe (saa7110)

This is a rewrite of the saa7110_write_block function, which was plain
broken in the case where the underlying adapter supports I2C_FUNC_I2C.
It also includes related fixes which ensure that different parts of the
driver agree on the number of registers the chip has.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] Linux 2.6.11.2
gregkh@suse.de [Wed, 9 Mar 2005 08:05:27 +0000 (00:05 -0800)] 
[PATCH] Linux 2.6.11.2

20 years ago[PATCH] epoll: return proper error on overflow condition
torvalds@ppc970.osdl.org [Wed, 9 Mar 2005 08:01:53 +0000 (00:01 -0800)] 
[PATCH] epoll: return proper error on overflow condition

Noted by Georgi Guninski.

20 years ago[PATCH] Linux 2.6.11.1
gregkh@suse.de [Fri, 4 Mar 2005 17:15:33 +0000 (09:15 -0800)] 
[PATCH] Linux 2.6.11.1

20 years ago[PATCH] Fix keyboards for Dell machines
dtor_core@ameritech.net [Fri, 4 Mar 2005 16:12:02 +0000 (08:12 -0800)] 
[PATCH] Fix keyboards for Dell machines

Some ACPI-related changes were recently made to i8042 discovery for ia64.
Unfortunately this broke a significant number of Dell laptops due to their
having incorrect BIOS tables.

So, for now, arrange for the new code to be ia64-only.

From: Dmitry Torokhov <dtor_core@ameritech.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] Fix for trivial fix for 2.6.11 raid6 compilation on ppc w/ Altivec
olof@austin.ibm.com [Thu, 3 Mar 2005 23:14:35 +0000 (15:14 -0800)] 
[PATCH] Fix for trivial fix for 2.6.11 raid6 compilation on ppc w/ Altivec

Here's a patch that will work for both PPC and PPC64. The proper way to
fix this in mainline is to merge -mm's cpu_has_feature patch, but for
the stable 2.6.11-series, this much less intrusive (i.e. just the pure
bugfix, not the cleanup part).

Signed-off-by: Olof Johansson <olof@austin.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago[PATCH] trivial fix for 2.6.11 raid6 compilation on ppc w/ Altivec
Rene Rebe [Thu, 3 Mar 2005 18:58:16 +0000 (10:58 -0800)] 
[PATCH] trivial fix for 2.6.11 raid6 compilation on ppc w/ Altivec

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
20 years ago2.6.11 import
Greg KH [Thu, 12 May 2005 16:59:06 +0000 (09:59 -0700)] 
2.6.11 import

20 years agoInitial commit
Greg KH [Thu, 12 May 2005 16:55:53 +0000 (09:55 -0700)] 
Initial commit