14 years ago[MIPS] Use generic STABS_DEBUG macro.
Atsushi Nemoto [Wed, 10 May 2006 15:41:26 +0000]
[MIPS] Use generic STABS_DEBUG macro.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

14 years ago[MIPS] Create consistency in "system type" selection.
Martin Michlmayr [Tue, 9 May 2006 21:34:53 +0000]
[MIPS] Create consistency in "system type" selection.

The "system type" Kconfig options on MIPS are not consistent.  For
some platforms, only the name is listed while other entries are
prepended with "Support for".  Remove this as it doesn't make sense
when describing the "system type".

Signed-off-by: Martin Michlmayr <tbm@cyrius.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

14 years ago[MIPS] Use generic DWARF_DEBUG
Atsushi Nemoto [Wed, 10 May 2006 06:36:04 +0000]
[MIPS] Use generic DWARF_DEBUG

When debugging a kernel compiled by gcc 4.1 with gdb 6.4, gdb could
not show filename, linenumber, etc.  It seems fixed if I used generic
DWARF_DEBUG macro.  Although gcc 3.x seems work without this change,
it would be better to use the generic macro unless there were
something MIPS specific.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

14 years ago[MIPS] Fix kgdb exception handler from user mode.
Atsushi Nemoto [Tue, 9 May 2006 11:23:49 +0000]
[MIPS] Fix kgdb exception handler from user mode.

Fix a calculation of saved vector address in trap_low.

(damage done by lmo f4c72cc737561aab0d9c7f877abbc0a853f1c465)

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

14 years ago[MIPS] Update struct sigcontext member names
Daniel Jacobowitz [Mon, 8 May 2006 19:28:22 +0000]
[MIPS] Update struct sigcontext member names

Rename the 64-bit sc_hi and sc_lo arrays to use the same names
as the 32-bit struct sigcontext (sc_mdhi, sc_hi1, et cetera).

Signed-off-by: Daniel Jacobowitz <dan@codesourcery.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

14 years ago[MIPS] Update/fix futex assembly
Ralf Baechle [Wed, 3 May 2006 19:42:39 +0000]
[MIPS] Update/fix futex assembly

 o Implement futex_atomic_op_inuser() operation
 o Don't use the R10000-ll/sc bug workaround version for every processor.
   branch likely is deprecated and some historic ll/sc processors don't
   implement it.  In any case it's slow.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

14 years ago[MIPS] Remove support for sysmips(2) SETNAME and MIPS_RDNVRAM operations.
Ralf Baechle [Wed, 3 May 2006 01:27:40 +0000]
[MIPS] Remove support for sysmips(2) SETNAME and MIPS_RDNVRAM operations.

SETNAME only had a minor defect but probably never had a user and
MIPS_RDNVRAM was unimplemented anyway.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

14 years ago[MIPS] Fix detection and handling of the 74K processor.
Chris Dearman [Tue, 2 May 2006 13:08:46 +0000]
[MIPS] Fix detection and handling of the 74K processor.

Nothing exciting; Linux just didn't know it yet so this is most adding
a value to a case statement.

Signed-off-by: Chris Dearman <chris@mips.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

14 years ago[MIPS] Add missing 34K processor IDs
Nigel Stephens [Thu, 27 Apr 2006 14:50:32 +0000]
[MIPS] Add missing 34K processor IDs

The 34K is very much like a 24K on steroids.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

14 years ago[MIPS] Fix marking buddy of pte global for MIPS32 w/36-bit physical address
Sergei Shtylyov [Sun, 16 Apr 2006 19:27:21 +0000]
[MIPS] Fix marking buddy of pte global for MIPS32 w/36-bit physical address

In case of CONFIG_64BIT_PHYS_ADDR, set_pte() and pte_clear() functions
only set _PAGE_GLOBAL bit in the pte_low field of the buddy PTEs,
forgetting to propagate ito to pte_high. Thus, the both pages might not
really be made global for the CPU (since it AND's the G-bit of the
odd / even PTEs together to decide whether they're global or not). Thus,
if only a single page is allocated via vmalloc() or ioremap(), it's not
really global for CPU (and it must be, since this is kernel mapping),
and thus its ASID is compared against the current process' one -- so,
we'll get into trouble sooner or later...  Also, pte_none() will fail
on global pages because _PAGE_GLOBAL bit is set in both pte_low and
pte_high, and pte_val() will return u64 value consisting of those fields
concateneted.

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

14 years ago[MIPS] AU1xxx mips_timer_interrupt() fixes
Herbert Valerio Riedel [Wed, 12 Apr 2006 07:03:08 +0000]
[MIPS] AU1xxx mips_timer_interrupt() fixes

common/au1000/irq.c was missing a mips_timer_interrupt() prototype,
whereas in common/au1000/time.c the actual mips_timer_interrupt()
implementation was missing an irq_exit() invocation, causing a
preempt_count() leak.

Signed-off-by: Herbert Valerio Riedel <hvr@hvrlab.org>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

14 years ago[MIPS] Fix typo
Ralf Baechle [Thu, 27 Apr 2006 14:44:50 +0000]
[MIPS] Fix typo

Found by Chris Dearman (chris@mips.com).

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

14 years ago[PATCH] ARM: explicitly disable BTB on ixp2350
Deepak Saxena [Tue, 30 May 2006 21:36:49 +0000]
[PATCH] ARM: explicitly disable BTB on ixp2350

We don't enable the BTB on the ixp2350 as that can cause weird
crashes (erratum #42.)  However, some bootloaders enable the BTB,
which means that we have to disable the BTB explicitly.

Found thanks to Tom Rini.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years agoRevert "[PATCH] i386/x86_64: Force pci=noacpi on HP XW9300"
Linus Torvalds [Wed, 31 May 2006 03:32:15 +0000]
Revert "[PATCH] i386/x86_64: Force pci=noacpi on HP XW9300"

This reverts commit 5491d0f3e206beb95eeb506510d62a1dab462df1.

As per Andi:

  "After some discussion with people who have the affected system it
   seems best to revert for 2.6.17.  It broke a common BIOS workaround
   and PCI-X still doesn't work.  Alternative is for people to change
   the BIOS which seems to be better right now."

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64: Don't do syscall exit tracing twice
Andi Kleen [Tue, 30 May 2006 20:48:03 +0000]
[PATCH] x86_64: Don't do syscall exit tracing twice

int_ret_from_syscall already does syscall exit tracing, so
no need to do it again in the caller.

This caused problems for UML and some other special programs doing
syscall interception.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64: Fix off by one in bad_addr checking in find_e820_area
Robert Hentosh [Tue, 30 May 2006 20:48:00 +0000]
[PATCH] x86_64: Fix off by one in bad_addr checking in find_e820_area

From: Robert Hentosh <robert_hentosh@dell.com>

Actually, we just stumbled on a different bug found in find_e820_area() in
e820.c.  The following code does not handle the edge condition correctly:

   while (bad_addr(&addr, size) && addr+size < ei->addr + ei->size)
       ;
   last = addr + size;
   if ( last > ei->addr + ei->size )
       continue;

The second statement in the while loop needs to be a <= b so that it is the
logical negavite of the if (a > b) outside it. It needs to read:

   while (bad_addr(&addr, size) && addr+size <= ei->addr + ei->size)
       ;

In the case that failed bad_addr was returning an address that is exactly size
bellow the end of the e820 range.

AK: Again together with the earlier avoid edma fix this fixes
boot on a Dell PE6850/16GB

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64: Handle empty node zero
Daniel Yeisley [Tue, 30 May 2006 20:47:57 +0000]
[PATCH] x86_64: Handle empty node zero

From: Daniel Yeisley <dan.yeisley@unisys.com>

It is possible to boot a Unisys ES7000 with CPUs from multiple cells, and not
also include the memory from those cells.  This can create a scenario where
node 0 has cpus, but no associated memory.  The system will boot fine in a
configuration where node 0 has memory, but nodes 2 and 3 do not.

[AK: I rechecked the code and generic code seems to indeed handle that already.
Dan's original patch had a change for mm/slab.c that seems to be already in now.]

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64: fix last_tsc calculation of PM timer
Jan Beulich [Tue, 30 May 2006 20:47:54 +0000]
[PATCH] x86_64: fix last_tsc calculation of PM timer

From: "Jan Beulich" <jbeulich@novell.com>

The PM timer code updates vxtime.last_tsc, but this update was done
incorrectly in two ways:
- offset_delay being in microseconds requires multiplying with cpu_mhz
  rather than cpu_khz
- the multiplication of offset_delay and cpu_khz (both being 32-bit
  values) on most current CPUs would overflow (observed value of the
  delay was approximately 4000us, yielding an overflow for frequencies
  starting a little above 1GHz)

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] i386: apic= command line option should always be
Jan Beulich [Tue, 30 May 2006 20:47:51 +0000]
[PATCH] i386: apic= command line option should always be

From: "Jan Beulich" <jbeulich@novell.com>

When using apic= on the kernel command line, this had no effect for machines
matched by either the ACPI MADT or the MPS OEM table scan. However, when such
option is specified, it should also take effect for this set of systems.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64: Fix no IOMMU warning in PCI-GART driver
Andi Kleen [Tue, 30 May 2006 20:47:48 +0000]
[PATCH] x86_64: Fix no IOMMU warning in PCI-GART driver

Complaining about the IOMMU not compiled in doesn't make sense
here because it is clearly compiled in.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64: Fix stack/mmap randomization for compat tasks
Andi Kleen [Tue, 30 May 2006 20:47:45 +0000]
[PATCH] x86_64: Fix stack/mmap randomization for compat tasks

ia32_setup_arg_pages would ignore the passed in random stack top
and use its own static value.

Now it uses the 8bit of randomness native i386 would use too.

This indirectly fixes mmap randomization for 32bit processes too,
which depends on the stack randomization.

Should also give slightly better virtual cache colouring and
possibly better performance with HyperThreading.

Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] cfq-iosched: fixup locking and ->queue_list list management
Jens Axboe [Tue, 30 May 2006 19:11:04 +0000]
[PATCH] cfq-iosched: fixup locking and ->queue_list list management

- Drop cic from the list when seen as dead.
- Fixup the locking, just use a simple spinlock.

Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/davej/agpgart
Linus Torvalds [Tue, 30 May 2006 18:54:32 +0000]
Merge /pub/scm/linux/kernel/git/davej/agpgart

* master.kernel.org:/pub/scm/linux/kernel/git/davej/agpgart:
  [AGPGART] VIA PT880 Ultra support.
  [AGPGART] Fix Nforce3 suspend on amd64.
  [AGPGART] Enable SIS AGP driver on x86-64 for EM64T systems

14 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6
Linus Torvalds [Tue, 30 May 2006 18:46:45 +0000]
Merge git://git./linux/kernel/git/sfrench/cifs-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6:
  [[CIFS] Pass truncate open flag through on file open in case setattr fails
  [CIFS] Fix typos in previous fix
  [CIFS] endian fix for new POSIX byte range lock support
  [CIFS] fix memory leak in cifs session info struct on reconnect
  [CIFS] ACPI suspend oops
  [CIFS] Do not limit the length of share names (was 100 for whole UNC name)
  [CIFS] Fix new POSIX Locking for setting lock_type correctly on unlock

14 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Tue, 30 May 2006 18:31:10 +0000]
Merge /pub/scm/linux/kernel/git/davem/net-2.6

* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6:
  [NETFILTER]: PPTP helper: fix sstate/cstate typo
  [NETFILTER]: mark H.323 helper experimental
  [NETFILTER]: Fix small information leak in SO_ORIGINAL_DST (CVE-2006-1343)

14 years ago[[CIFS] Pass truncate open flag through on file open in case setattr fails
Steve French [Tue, 30 May 2006 18:09:31 +0000]
[[CIFS] Pass truncate open flag through on file open in case setattr fails

on set size to zero.

Signed-off-by: Sebastian Voitzsch <sebastoam/vpotzscj@web.de>
Signed-off-by: Steve French <sfrench@us.ibm.com>

14 years ago[CIFS] Fix typos in previous fix
Steve French [Tue, 30 May 2006 18:08:26 +0000]
[CIFS] Fix typos in previous fix

Signed-off-by: Steve French <sfrench@us.ibm.com>

14 years ago[CIFS] endian fix for new POSIX byte range lock support
Steve French [Tue, 30 May 2006 18:07:17 +0000]
[CIFS] endian fix for new POSIX byte range lock support

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>

14 years ago[CIFS] fix memory leak in cifs session info struct on reconnect
Steve French [Tue, 30 May 2006 18:06:04 +0000]
[CIFS] fix memory leak in cifs session info struct on reconnect

Signed-off-by: Steve French <sfrench@us.ibm.com>

14 years ago[CIFS] ACPI suspend oops
Steve French [Tue, 30 May 2006 18:05:10 +0000]
[CIFS] ACPI suspend oops

Wasn't able to reproduce a hard hang, but was able to get an oops if
suspended the machine during a copy to the cifs mount.  This led to some
things hanging, including a "sync".  Also got I/O errors when trying to
access the mount afterwards (even when didn't see the oops), and had
to unmount and remount in order to access the filesystem.

This patch fixed the oops.

Signed-off-by: Dave Kleikamp <shaggy@austin.ibm.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>

14 years ago[CIFS] Do not limit the length of share names (was 100 for whole UNC name)
Steve French [Tue, 30 May 2006 18:04:19 +0000]
[CIFS] Do not limit the length of share names (was 100 for whole UNC name)
during mount. Especially important for some non-Western languages.

Signed-off-by: Steve French <sfrench@us.ibm.com>

14 years ago[CIFS] Fix new POSIX Locking for setting lock_type correctly on unlock
Steve French [Tue, 30 May 2006 18:03:32 +0000]
[CIFS] Fix new POSIX Locking for setting lock_type correctly on unlock

Signed-off-by: Steve French <sfrench@us.ibm.com>

14 years agoInput: psmouse - DMI updates for lifebook protocol
Kenan Esau [Tue, 30 May 2006 03:31:12 +0000]
Input: psmouse - DMI updates for lifebook protocol

Added different lifebook-versions and the CF-18 to the corresponding
dmi-table.

Signed-off-by: Kenan Esau <kenan.esau@conan.de>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

14 years agoInput: change from numbered to named switches
Richard Purdie [Tue, 30 May 2006 03:31:03 +0000]
Input: change from numbered to named switches

Remove the numbered SW_* entries from the input system and assign names
to the existing users.

Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

14 years agoInput: alps - fix old protocol decoding
Yotam Medini [Tue, 30 May 2006 03:30:36 +0000]
Input: alps - fix old protocol decoding

Correct touchpad left & right keys assignments for ALPS_OLDPROTO
that were swapped. Old protocol is used on UMAX ActionBook-530T
notebook.

Signed-off-by: Yotam Medini <yotam.medini@gmail.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

14 years agoInput: wistron - add support for AOpen Barebook 1559as
masc@theaterzentrum.at [Tue, 30 May 2006 03:29:36 +0000]
Input: wistron - add support for AOpen Barebook 1559as

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

14 years agoInput: psmouse - add detection of Logitech TrackMan Wheel trackball
Zbigniew Luszpinski [Tue, 30 May 2006 03:29:19 +0000]
Input: psmouse - add detection of Logitech TrackMan Wheel trackball

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

14 years agoInput: sidewinder - fix memory leak
Jesper Juhl [Tue, 30 May 2006 03:28:05 +0000]
Input: sidewinder - fix memory leak

In sw_connect we leak 'buf' and 'idbuf' when we do not leave via one of
the fail* labels. This was spotted by the coverity checker.

Patch is compile tested only due to lack of hardware.

Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

14 years agoInput: add KEY_BATTERY keycode
Matthew Garrett [Tue, 30 May 2006 03:27:39 +0000]
Input: add KEY_BATTERY keycode

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

14 years ago[NETFILTER]: PPTP helper: fix sstate/cstate typo
Alexey Dobriyan [Mon, 29 May 2006 05:51:05 +0000]
[NETFILTER]: PPTP helper: fix sstate/cstate typo

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

14 years ago[NETFILTER]: mark H.323 helper experimental
Patrick McHardy [Mon, 29 May 2006 05:50:40 +0000]
[NETFILTER]: mark H.323 helper experimental

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

14 years ago[NETFILTER]: Fix small information leak in SO_ORIGINAL_DST (CVE-2006-1343)
Marcel Holtmann [Mon, 29 May 2006 05:50:18 +0000]
[NETFILTER]: Fix small information leak in SO_ORIGINAL_DST (CVE-2006-1343)

It appears that sockaddr_in.sin_zero is not zeroed during
getsockopt(...SO_ORIGINAL_DST...) operation. This can lead
to an information leak (CVE-2006-1343).

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

14 years agoMerge branch 'upstream-fixes' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik...
Linus Torvalds [Sun, 28 May 2006 23:35:52 +0000]
Merge branch 'upstream-fixes' of /linux/kernel/git/jgarzik/libata-dev

* 'upstream-fixes' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/libata-dev:
  [PATCH] the latest consensus libata resume fix

14 years agoppc: Fix typo in TI_LOCAL_FLAGS definition
Paul Mackerras [Sun, 28 May 2006 22:42:34 +0000]
ppc: Fix typo in TI_LOCAL_FLAGS definition

A typo crept in with commit ea1e847cc202e805769c3c46ba5e5c53714068a1
which defined TI_LOCAL_FLAGS to be the offset of the `flags' field
of struct thread_info, rather than the `local_flags' field.  This
fixes it.  The typo was pointed out by Guennadi Liakhovetski.

Signed-off-by: Paul Mackerras <paulus@samba.org>

14 years ago[PATCH] the latest consensus libata resume fix
Mark Lord [Sun, 28 May 2006 15:28:00 +0000]
[PATCH] the latest consensus libata resume fix

Okay, just to sum things up.

This forces libata to wait for up to 2 seconds for BUSY|DRQ to clear
on resume before continuing.

[jgarzik adds...]  During testing we never saw DRQ asserted, but
nonetheless (a) this works and (b) testing for DRQ won't hurt.

Signed-off-by: Mark Lord <liml@rtr.ca>
Acked-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

14 years agoMerge branch 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc
Linus Torvalds [Sat, 27 May 2006 16:40:40 +0000]
Merge branch 'merge' of git://git./linux/kernel/git/paulus/powerpc

* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc:
  [PATCH] powerpc: fix RTC/NVRAM accesses on Maple
  [PATCH] ppc32 CPM_UART: various fixes for pq2 uart users
  [PATCH] powerpc: linuxppc64.org no more

14 years ago[netdrvr s/390] trim trailing whitespace
Jeff Garzik [Sat, 27 May 2006 01:58:38 +0000]
[netdrvr s/390] trim trailing whitespace

Previous fix patches added a bunch of trailing whitespace,
which git-applymbox complained loudly about.

14 years ago[PATCH] s390: lcs driver bug fixes and improvements [2/2]
Klaus Wacker [Wed, 24 May 2006 07:51:21 +0000]
[PATCH] s390: lcs driver bug fixes and improvements [2/2]

This is the second lcs driver patch containing the rest of lcs fixes.

Signed-off-by: Frank Pavlic <fpavlic@de.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

14 years ago[PATCH] s390: lcs driver bug fixes and improvements [1/2]
Klaus Wacker [Wed, 24 May 2006 07:51:17 +0000]
[PATCH] s390: lcs driver bug fixes and improvements [1/2]

Several problems occured with lcs device driver:
 - device not operational anymore after cable pull/plug-in.
         - unpredictable results occured, e.g. kernel panic
   using cards of type QD8F.
 - STOPLAN and delete multicast address command
           were not proper recognized by OSA card under heavy network workload.
         - channel/device error checks missing in interrupt handler.
To fix all problems at once recovery of lcs devices has been improved.
missing error checks in lcs interrupt handler has been added.
Once a hardware problem occurs lcs will recover the device now properly.

Signed-off-by: Frank Pavlic <fpavlic@de.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

14 years ago[PATCH] s390: qeth driver fixes
Ursula Braun [Wed, 24 May 2006 07:51:13 +0000]
[PATCH] s390: qeth driver fixes

From: Frank Blaschka <Frank.Blaschka@de.ibm.com>
From: Frank Pavlic <fpavlic@de.ibm.com>

        - fix fake_ll during initial device bringup. fake_ll was
  not active after first start of the device.
  Problem only occured when qeth was built without IPV6 support.
        - avoid skb usage after invocation of qeth_flush_buffers,
  because skb might already be freed.
        - remove yet another useless netif_wake_queue in
  qeth_softsetup_ipv6 since this function is only called
  when device is going online. In this case card->state will
  never be in state UP. So let the net_device queue down .

Signed-off-by: Frank Pavlic <fpavlic@de.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

14 years ago[PATCH] s390: qeth driver fixes
Ursula Braun [Wed, 24 May 2006 07:51:11 +0000]
[PATCH] s390: qeth driver fixes

From: Frank Pavlic <fpavlic@de.ibm.com>

- correct checking of sscanf-%n value in qeth_string_to_ipaddr().
- don't use netif_stop_queue outside the hard_start_xmit routine.
  Rather use netif_tx_disable.
- don't call qeth_netdev_init on a recovery.

Signed-off-by: Frank Pavlic <fpavlic@de.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

14 years ago[PATCH] s390: minor fix in cu3088
Cornelia Huck [Wed, 24 May 2006 07:51:05 +0000]
[PATCH] s390: minor fix in cu3088

In case of a parse error for the cu3088 group attribute,
return -EINVAL instead of count.

Signed-off-by: Frank Pavlic <fpavlic@de.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

14 years ago[PATCH] pcnet32: remove incorrect pcnet32_free_ring
Don Fry [Thu, 25 May 2006 23:22:40 +0000]
[PATCH] pcnet32: remove incorrect pcnet32_free_ring

During a code scan for another change I discovered that this call to
pcnet32_free_ring must be removed.  If the open fails due to a lack of
memory all the ring structures are removed via the call to free_ring
and a subsequent call to open will dereference a null pointer in
pcnet32_init_ring.

Please apply to 2.6.17.

Signed-off-by: Don Fry <brazilnut@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

14 years agoe1000: add shutdown handler back to fix WOL
Auke Kok [Tue, 23 May 2006 20:35:57 +0000]
e1000: add shutdown handler back to fix WOL

Someone was waaay too aggressive and removed e1000's reboot notifier
instead of porting it to the new way of the shutdown handler.  This change
broke wake on lan.  Add the shutdown handler back in using the same method
as e100 uses.

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>

(cherry picked from c653e6351e371b33b29871e5eedf610ffb3be037 commit)

14 years agoMerge branch 'upstream-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/linvil...
Jeff Garzik [Sat, 27 May 2006 01:26:22 +0000]
Merge branch 'upstream-fixes' of git://git./linux/kernel/git/linville/wireless-2.6 into upstream-fixes

14 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Fri, 26 May 2006 22:13:33 +0000]
Merge /pub/scm/linux/kernel/git/davem/net-2.6

* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6:
  [NET]: dev.c comment fixes
  [IPV6] ROUTE: Don't try less preferred routes for on-link routes.
  [IRDA]: *_DONGLE should depend on IRTTY_SIR
  [MAINTAINERS]: Add entry for netem

14 years ago[NET]: dev.c comment fixes
Stephen Hemminger [Fri, 26 May 2006 20:25:24 +0000]
[NET]: dev.c comment fixes

Noticed that dev_alloc_name() comment was incorrect, and more spellung
errors.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

14 years ago[IPV6] ROUTE: Don't try less preferred routes for on-link routes.
YOSHIFUJI Hideaki [Fri, 26 May 2006 20:23:41 +0000]
[IPV6] ROUTE: Don't try less preferred routes for on-link routes.

In addition to the real on-link routes, NONEXTHOP routes
should be considered on-link.

Problem reported by Meelis Roos <mroos@linux.ee>.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Acked-by: Meelis Roos <mroos@linux.ee>
Signed-off-by: David S. Miller <davem@davemloft.net>

14 years ago[PATCH] arlan: fix section mismatch warnings
Randy Dunlap [Thu, 25 May 2006 18:10:08 +0000]
[PATCH] arlan: fix section mismatch warnings

Fix section mismatch warnings:
WARNING: drivers/net/wireless/arlan.o - Section mismatch: reference to
.init.text:arlan_probe from .text between 'init_module' (at offset
0x3526) and 'cleanup_module'
WARNING: drivers/net/wireless/arlan.o - Section mismatch: reference to
.init.text:init_arlan_proc from .text between 'init_module' (at offset
0x3539) and 'cleanup_module'
WARNING: drivers/net/wireless/arlan.o - Section mismatch: reference to
.exit.text:cleanup_arlan_proc from .text between 'cleanup_module' (at
offset 0x356c) and 'arlan_diagnostic_info_string'

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

14 years ago[PATCH] wavelan: fix section mismatch
Randy Dunlap [Thu, 25 May 2006 18:09:21 +0000]
[PATCH] wavelan: fix section mismatch

Fix section mismatch warning:
WARNING: drivers/net/wireless/wavelan.o - Section mismatch: reference to
.init.text: from .text between 'init_module' (at offset 0x371e) and
'cleanup_module'

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

14 years ago[PATCH] tpm: fix bug for TPM on ThinkPad T60 and Z60
Kylene Jo Hall [Fri, 26 May 2006 01:44:27 +0000]
[PATCH] tpm: fix bug for TPM on ThinkPad T60 and Z60

The TPM chip on the ThinkPad T60 and Z60 machines is returning 0xFFFF for
the vendor ID which is a check the driver made to double check it was
actually talking to the memory mapped space of a TPM.  This patch removes
the check since it isn't absolutely necessary and was causing device
discovery to fail on these machines.

Signed-off-by: Kylene Hall <kjhall@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] scx200_acb: fix section mismatch warning
Randy Dunlap [Fri, 26 May 2006 01:44:26 +0000]
[PATCH] scx200_acb: fix section mismatch warning

WARNING: drivers/i2c/busses/scx200_acb.o - Section mismatch: reference to
.init.text: from .text after 'scx200_add_cs553x' (at offset 0x528)

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] s3c24xx: fix spi driver with CONFIG_PM
Ben Dooks [Fri, 26 May 2006 01:44:25 +0000]
[PATCH] s3c24xx: fix spi driver with CONFIG_PM

Fix compile bug with the S3C24XX SPI driver when CONFIG_PM is set.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] PowerMac: force only suspend-to-disk to be valid
Johannes Berg [Fri, 26 May 2006 01:44:24 +0000]
[PATCH] PowerMac: force only suspend-to-disk to be valid

For a very long time, echoing 'standby' or 'mem' into /sys/power/state has
killed the machine on powerpc.  This patch fixes that.

This patch adds the .valid callback to pm_ops on PowerMac so that only the
suspend to disk state can be entered.  Note that just returning 0 would
suffice since the upper layers don't pass PM_SUSPEND_DISK down, but we
handle it there regardless just in case that changes.

Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] affs: possible null pointer dereference in affs_rename()
Florin Malita [Fri, 26 May 2006 01:44:23 +0000]
[PATCH] affs: possible null pointer dereference in affs_rename()

If affs_bread() fails, the exit path calls mark_buffer_dirty_inode() with a
NULL argument.

Coverity CID: 312.

Signed-off-by: Florin Malita <fmalita@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ads7846 conversion accuracy
David Brownell [Fri, 26 May 2006 01:44:20 +0000]
[PATCH] ads7846 conversion accuracy

This improves accuracy of the touchscreen and hwmon sensor readings,
addressing an issue noted by Imre Deak: there's an extra bit written before
the sample (12 bits) gets written out.

It also catches up to various comments, and makes the /proc/interrupts
entry sensible again.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Imre Deak <imre.deak@nokia.com>
Cc: Jean Delvare <khali@linux-fr.org>
Cc: Dmitry Torokhov <dtor_core@ameritech.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86: wire up vmsplice syscall
Jens Axboe [Fri, 26 May 2006 01:44:19 +0000]
[PATCH] x86: wire up vmsplice syscall

Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Add CMSPAR to termbits.h for powerpc and alpha
Paul Mackerras [Fri, 26 May 2006 01:44:17 +0000]
[PATCH] Add CMSPAR to termbits.h for powerpc and alpha

Some driver wants to use CMSPAR, but it was missing on alpha and powerpc.
This adds it, with the same value as every other architecture uses.

(akpm: fixes the build of an upcoming gregkh USB patch)

Signed-off-by: Paul Mackerras <paulus@samba.org>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Unlock md devices when stopping them on reboot.
Neil Brown [Fri, 26 May 2006 00:39:25 +0000]
[PATCH] Unlock md devices when stopping them on reboot.

otherwise we get nasty messages about locks not being released.

Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years agoMerge branch 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband
Linus Torvalds [Fri, 26 May 2006 18:51:08 +0000]
Merge branch 'for-linus' of /linux/kernel/git/roland/infiniband

* 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband:
  IB/mthca: Fix posting lists of 256 receive requests to SRQ for Tavor

14 years ago[PATCH] powerpc: fix RTC/NVRAM accesses on Maple
Hollis Blanchard [Thu, 25 May 2006 21:36:53 +0000]
[PATCH] powerpc: fix RTC/NVRAM accesses on Maple

Due to a firmware device tree bug, RTC and NVRAM accesses (including
halt/reboot) on Maple have been broken since January, when an untested
build fix went in. This code patches the device tree in Linux.

Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Segher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>

14 years ago[PATCH] ppc32 CPM_UART: various fixes for pq2 uart users
Vitaly Bordug [Wed, 24 May 2006 17:40:18 +0000]
[PATCH] ppc32 CPM_UART: various fixes for pq2 uart users

This fixes various odd things that missed update together with cpm_uart
platform_device move. Unified resources names, restructurisation, etc.
Also, addressed issue with recent phys/virt translation rework. Being
cache-coherent, CPM2's do alloc_bootmem() for the console stuff, and it was
used to treat console buffer descriptor mapping 1:1 (as in CPM1 case),
which is definitely wrong.

Signed-off-by: Vitaly Bordug <vbordug@ru.mvista.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>

14 years ago[PATCH] powerpc: linuxppc64.org no more
Arthur Othieno [Fri, 19 May 2006 10:22:23 +0000]
[PATCH] powerpc: linuxppc64.org no more

http://linuxppc64.org has long been a redirect to the canonical
http://penguinppc.org/ppc64/ -- update all instances accordingly,
as ACKed by Hollis:

On Wed, Jan 18, 2006 at 09:48:08AM -0600, Hollis Blanchard wrote:
> On Wed, 2006-01-18 at 13:07 +0100, Olaf Hering wrote:
> > On Wed, Jan 18, Arthur Othieno wrote:
> > >
> > > What about the s/linuxppc64\.org/penguinppc\.org/g case? Or is
> > > penguinppc64.org preferable? Or am I just taking it too far? ;)
> >
> > They are redirected on DNS or HTTP level.
>
> HTTP level, but that doesn't answer his question.
>
> As the maintainer of that site, I would prefer to remove the
> linuxppc64.org reference.

Signed-off-by: Arthur Othieno <apgo@patchbomb.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>

14 years ago[IRDA]: *_DONGLE should depend on IRTTY_SIR
Samuel Ortiz [Thu, 25 May 2006 23:17:53 +0000]
[IRDA]: *_DONGLE should depend on IRTTY_SIR

If a SIR dongle is built in the kernel while IRTTY_SIR is built
as a module, kernel compilation will fail.
Thus, the SIR dongle config should depend on the IRTTY_SIR.

Closes kernel bug# 6512
(http://bugzilla.kernel.org/show_bug.cgi?id=6512)

Signed-off-by: Samuel Ortiz <samuel@sortiz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

14 years ago[MAINTAINERS]: Add entry for netem
Stephen Hemminger [Thu, 25 May 2006 23:14:43 +0000]
[MAINTAINERS]: Add entry for netem

It gets enough reports, that there ought to be a MAINTAINER entry.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

14 years ago[PATCH] s390: fix typo in stop_hz_timer.
Martin Schwidefsky [Tue, 23 May 2006 07:22:42 +0000]
[PATCH] s390: fix typo in stop_hz_timer.

Add missing parentheses for type cast to u64.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Dave Jones <davej@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years agoLinux 2.6.17-rc5
Linus Torvalds [Thu, 25 May 2006 01:50:17 +0000]
Linux 2.6.17-rc5

14 years ago[PATCH] V4L/DVB (4045): Fixes recursive dependency for I2C
Jean Delvare [Wed, 24 May 2006 18:13:14 +0000]
[PATCH] V4L/DVB (4045): Fixes recursive dependency for I2C

Mixing "depends on I2C" and "select I2C" within the media subsystem
leads to the following problem:
Warning! Found recursive dependency: I2C DVB_BUDGET DVB_BUDGET_PATCH
DVB_AV7110 VIDEO_SAA7146_VV VIDEO_SAA7146 I2C

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Manu Abraham <manu@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years agoIB/mthca: Fix posting lists of 256 receive requests to SRQ for Tavor
Michael S. Tsirkin [Wed, 24 May 2006 15:27:07 +0000]
IB/mthca: Fix posting lists of 256 receive requests to SRQ for Tavor

If we post a list of length exactly a multiple of 256, nreq in
doorbell gets set to 256 which is wrong: it should be encoded by 0.
This is because we only zero it out on the next WR, which may not be
there.  The solution is to ring the doorbell after posting a WQE, not
before posting the next one.

This is the same bug that we just fixed for QPs with non-shared RQ.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

14 years agoMerge branch 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband
Linus Torvalds [Wed, 24 May 2006 15:55:12 +0000]
Merge branch 'for-linus' of /linux/kernel/git/roland/infiniband

* 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband:
  IB/ipath: deref correct pointer when using kernel SMA
  IB/ipath: fix null deref during rdma ops
  IB/ipath: register as IB device owner
  IB/ipath: enable PE800 receive interrupts on user ports
  IB/ipath: enable GPIO interrupt on HT-460
  IB/ipath: fix NULL dereference during cleanup
  IB/ipath: replace uses of LIST_POISON
  IB/ipath: fix reporting of driver version to userspace
  IB/ipath: don't modify QP if changes fail
  IB/ipath: fix spinlock recursion bug

14 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/shaggy...
Linus Torvalds [Wed, 24 May 2006 15:37:03 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/shaggy/jfs-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/shaggy/jfs-2.6:
  JFS: Fix multiple errors in metapage_releasepage

14 years agoMerge branch 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik...
Linus Torvalds [Wed, 24 May 2006 15:36:31 +0000]
Merge branch 'upstream-linus' of /linux/kernel/git/jgarzik/libata-dev

* 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/libata-dev:
  [PATCH] libata: add pio flush for via atapi (was: Re: TR: ASUS A8V Deluxe, x86_64)

14 years agoMerge branch 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik...
Linus Torvalds [Wed, 24 May 2006 15:36:03 +0000]
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:
  [PATCH] sky2: fix jumbo packet support

14 years ago[MMC] Fix premature use of md->disk
Pierre Ossman [Wed, 24 May 2006 08:20:45 +0000]
[MMC] Fix premature use of md->disk

md->disk was being used in a debug message before it was allocated.

Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years agoJFS: Fix multiple errors in metapage_releasepage
Dave Kleikamp [Wed, 24 May 2006 12:43:38 +0000]
JFS: Fix multiple errors in metapage_releasepage

It looks like metapage_releasepage was making in invalid assumption that
the releasepage method would not be called on a dirty page.  Instead of
issuing a warning and releasing the metapage, it should return 0, indicating
that the private data for the page cannot be released.

I also realized that metapage_releasepage had the return code all wrong.  If
it is successful in releasing the private data, it should return 1, otherwise
it needs to return 0.

Lastly, there is no need to call wait_on_page_writeback, since
try_to_release_page will not call us with a page in writback state.

Signed-off-by: Dave Kleikamp <shaggy@austin.ibm.com>

14 years ago[PATCH] libata: add pio flush for via atapi (was: Re: TR: ASUS A8V Deluxe, x86_64)
Albert Lee [Mon, 22 May 2006 03:43:46 +0000]
[PATCH] libata: add pio flush for via atapi (was: Re: TR: ASUS A8V Deluxe, x86_64)

Backport the "pio flush" from the libata major update to 2.6.17 for via atapi.

Signed-off-by: Albert Lee <albertcc@tw.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

14 years ago[PATCH] sky2: fix jumbo packet support
Stephen Hemminger [Mon, 22 May 2006 19:03:42 +0000]
[PATCH] sky2: fix jumbo packet support

The truncate threshold calculation to prevent receiver from getting stuck
was incorrect, and it didn't take into account the upper limit on bits
in the register so the jumbo packet support was broken.

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

14 years ago[BRIDGE]: need to ref count the LLC sap
Stephen Hemminger [Tue, 23 May 2006 22:20:25 +0000]
[BRIDGE]: need to ref count the LLC sap

Bridge will OOPS on removal if other application has the SAP open.
The bridge SAP might be shared with other usages, so need
to do reference counting on module removal rather than explicit
close/delete.

Since packet might arrive after or during removal, need to clear
the receive function handle, so LLC only hands it to user (if any).

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

14 years ago[NETFILTER]: SNMP NAT: fix memleak in snmp_object_decode
Chris Wright [Tue, 23 May 2006 22:08:13 +0000]
[NETFILTER]: SNMP NAT: fix memleak in snmp_object_decode

If kmalloc fails, error path leaks data allocated from asn1_oid_decode().

Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

14 years ago[NETFILTER]: H.323 helper: fix sequence extension parsing
Patrick McHardy [Tue, 23 May 2006 22:07:46 +0000]
[NETFILTER]: H.323 helper: fix sequence extension parsing

When parsing unknown sequence extensions the "son"-pointer points behind
the last known extension for this type, don't try to interpret it.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

14 years ago[NETFILTER]: H.323 helper: fix parser error propagation
Patrick McHardy [Tue, 23 May 2006 22:07:07 +0000]
[NETFILTER]: H.323 helper: fix parser error propagation

The condition "> H323_ERROR_STOP" can never be true since H323_ERROR_STOP
is positive and is the highest possible return code, while real errors are
negative, fix the checks. Also only abort on real errors in some spots
that were just interpreting any return value != 0 as error.

Fixes crashes caused by use of stale data after a parsing error occured:

BUG: unable to handle kernel paging request at virtual address bfffffff
 printing eip:
c01aa0f8
*pde = 1a801067
*pte = 00000000
Oops: 0000 [#1]
PREEMPT
Modules linked in: ip_nat_h323 ip_conntrack_h323 nfsd exportfs sch_sfq sch_red cls_fw sch_hfsc  xt_length ipt_owner xt_MARK iptable_mangle nfs lockd sunrpc pppoe pppoxx
CPU:    0
EIP:    0060:[<c01aa0f8>]    Not tainted VLI
EFLAGS: 00210646   (2.6.17-rc4 #8)
EIP is at memmove+0x19/0x22
eax: d77264e9   ebx: d77264e9   ecx: e88d9b17   edx: d77264e9
esi: bfffffff   edi: bfffffff   ebp: de6a7680   esp: c0349db8
ds: 007b   es: 007b   ss: 0068
Process asterisk (pid: 3765, threadinfo=c0349000 task=da068540)
Stack: <0>00000006 c0349e5e d77264e3 e09a2b4e e09a38a0 d7726052 d7726124 00000491
       00000006 00000006 00000006 00000491 de6a7680 d772601e d7726032 c0349f74
       e09a2dc2 00000006 c0349e5e 00000006 00000000 d76dda28 00000491 c0349f74
Call Trace:
 [<e09a2b4e>] mangle_contents+0x62/0xfe [ip_nat]
 [<e09a2dc2>] ip_nat_mangle_tcp_packet+0xa1/0x191 [ip_nat]
 [<e0a2712d>] set_addr+0x74/0x14c [ip_nat_h323]
 [<e0ad531e>] process_setup+0x11b/0x29e [ip_conntrack_h323]
 [<e0ad534f>] process_setup+0x14c/0x29e [ip_conntrack_h323]
 [<e0ad57bd>] process_q931+0x3c/0x142 [ip_conntrack_h323]
 [<e0ad5dff>] q931_help+0xe0/0x144 [ip_conntrack_h323]
...

Found by the PROTOS c07-h2250v4 testsuite.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

14 years agoIB/ipath: deref correct pointer when using kernel SMA
Bryan O'Sullivan [Tue, 23 May 2006 18:32:38 +0000]
IB/ipath: deref correct pointer when using kernel SMA

At this point, the core QP structure hasn't been initialized, so what's
in there isn't valid.  Get the same information elsewhere.

Signed-off-by: Bryan O'Sullivan <bos@pathscale.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

14 years agoIB/ipath: fix null deref during rdma ops
Bryan O'Sullivan [Tue, 23 May 2006 18:32:37 +0000]
IB/ipath: fix null deref during rdma ops

The problem was that node A's sending thread, which handles sending RDMA
read response data, would write the trigger word, the last packet would
be sent, node B would send a new RDMA read request, node A's interrupt
handler would initialize s_rdma_sge, then node A's sending thread would
update s_rdma_sge.  This didn't happen very often naturally but was more
frequent with 1 byte RDMA reads.  Rather than adding more locking or
increasing the QP structure size and copying sge data, I modified the
copy routine to update the pointers before writing the trigger word to
avoid the update race.

Signed-off-by: Ralph Campbell <ralphc@pathscale.com>
Signed-off-by: Bryan O'Sullivan <bos@pathscale.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

14 years agoIB/ipath: register as IB device owner
Bryan O'Sullivan [Tue, 23 May 2006 18:32:36 +0000]
IB/ipath: register as IB device owner

This fixes an oops.

Signed-off-by: Bryan O'Sullivan <bos@pathscale.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

14 years agoIB/ipath: enable PE800 receive interrupts on user ports
Bryan O'Sullivan [Tue, 23 May 2006 18:32:35 +0000]
IB/ipath: enable PE800 receive interrupts on user ports

Fixed so it works on the PE-800.  It had not previously been updated to
match PE-800 receive interrupt differences from HT-400.

Signed-off-by: Bryan O'Sullivan <bos@pathscale.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

14 years agoIB/ipath: enable GPIO interrupt on HT-460
Bryan O'Sullivan [Tue, 23 May 2006 18:32:34 +0000]
IB/ipath: enable GPIO interrupt on HT-460

This is required for even semi-decent performance on OpenIB.

Signed-off-by: Bryan O'Sullivan <bos@pathscale.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

14 years agoIB/ipath: fix NULL dereference during cleanup
Bryan O'Sullivan [Tue, 23 May 2006 18:32:33 +0000]
IB/ipath: fix NULL dereference during cleanup

Fix NULL deref due to pcidev being clobbered before dd->ipath_f_cleanup()
was called.

Signed-off-by: Bryan O'Sullivan <bos@pathscale.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

14 years agoIB/ipath: replace uses of LIST_POISON
Bryan O'Sullivan [Tue, 23 May 2006 18:32:32 +0000]
IB/ipath: replace uses of LIST_POISON

Per Andrew's request.

Signed-off-by: Bryan O'Sullivan <bos@pathscale.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

14 years agoIB/ipath: fix reporting of driver version to userspace
Bryan O'Sullivan [Tue, 23 May 2006 18:32:31 +0000]
IB/ipath: fix reporting of driver version to userspace

Fix the interface version that gets exported to userspace.

Signed-off-by: Bryan O'Sullivan <bos@pathscale.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

14 years agoIB/ipath: don't modify QP if changes fail
Bryan O'Sullivan [Tue, 23 May 2006 18:32:30 +0000]
IB/ipath: don't modify QP if changes fail

Make sure modify_qp won't modify the QP if any of the changes failed.

Signed-off-by: Bryan O'Sullivan <bos@pathscale.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>