linux-2.6
16 years agofsl_diu_fb: fix build with CONFIG_PM=y, plus fix some warnings
Anton Vorontsov [Fri, 4 Jul 2008 16:59:38 +0000 (09:59 -0700)] 
fsl_diu_fb: fix build with CONFIG_PM=y, plus fix some warnings

This patch fixes following build error when CONFIG_PM is set.

  CC      drivers/video/fsl-diu-fb.o
drivers/video/fsl-diu-fb.c: In function 'fsl_diu_suspend':
drivers/video/fsl-diu-fb.c:1327: error: 'ofdev' undeclared (first use in this function)
drivers/video/fsl-diu-fb.c:1327: error: (Each undeclared identifier is reported only once
drivers/video/fsl-diu-fb.c:1327: error: for each function it appears in.)
drivers/video/fsl-diu-fb.c: In function 'fsl_diu_resume':
drivers/video/fsl-diu-fb.c:1337: error: 'ofdev' undeclared (first use in this function)

While I'm at it, also fix this warning:

drivers/video/fsl-diu-fb.c: In function 'fsl_diu_alloc':
drivers/video/fsl-diu-fb.c:314: warning: format '%lx' expects type 'long unsigned int', but argument 3 has type 'phys_addr_t'

And these section mismatches:

..from the function fsl_diu_remove() to the function .exit.text:uninstall_fb()
..from the function fsl_diu_remove() to the function .exit.text:uninstall_fb()
..from the function install_fb() to the variable .devinit.data:fsl_diu_mode_db
..from the function install_fb() to the variable .devinit.data:fsl_diu_mode_db
..from the function fsl_diu_probe() to the function .exit.text:uninstall_fb()
..from the function fsl_diu_probe() to the function .exit.text:uninstall_fb()

Also, some sparse fixes: make two functions static, and use NULL where
appropriate. There are still a lot of sparse warnings, mainly wrt absence
of __iomem annotations, but some will require ugly __force stuff. I'll leave
them for now, since proper fix would be not that trivial as few one-liners
below.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Cc: Timur Tabi <timur@freescale.com>
Cc: Antonino Daplas <adaplas@gmail.com>
Cc: York Sun <yorksun@freescale.com>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agogpio: pca953x (i2c) handles max7310 too
David Brownell [Fri, 4 Jul 2008 16:59:37 +0000 (09:59 -0700)] 
gpio: pca953x (i2c) handles max7310 too

The pca953x driver can handle another 8-bit I/O expander, the max7310.
This patch adds that chip to the list of supported IDs in that driver, and
expands the Kconfig helptext accordingly.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agolib: taint kernel in common report_bug() WARN path.
Paul Mundt [Fri, 4 Jul 2008 16:59:36 +0000 (09:59 -0700)] 
lib: taint kernel in common report_bug() WARN path.

Commit 95b570c9cef3b12356454c7112571b7e406b4b51 ("Taint kernel after
WARN_ON(condition)") introduced a TAINT_WARN that was implemented for
all architectures using the generic warn_on_slowpath(), which excluded
any architecture that set HAVE_ARCH_WARN_ON.

As all of the architectures that implement their own WARN_ON() all go
through the report_bug() path (specifically handling BUG_TRAP_TYPE_WARN),
taint the kernel there as well for consistency.

Tested on avr32 and sh. Also relevant for s390, parisc, and powerpc.

Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Acked-by: Kyle McMartin <kyle@mcmartin.ca>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoecryptfs: remove unnecessary mux from ecryptfs_init_ecryptfs_miscdev()
Michael Halcrow [Fri, 4 Jul 2008 16:59:35 +0000 (09:59 -0700)] 
ecryptfs: remove unnecessary mux from ecryptfs_init_ecryptfs_miscdev()

The misc_mtx should provide all the protection required to keep the daemon
hash table sane during miscdev registration.  Since this mutex is causing
gratuitous lockdep warnings, this patch removes it.

Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com>
Reported-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoreiserfs: add missing unlock to an error path in reiserfs_quota_write()
Jan Kara [Fri, 4 Jul 2008 16:59:34 +0000 (09:59 -0700)] 
reiserfs: add missing unlock to an error path in reiserfs_quota_write()

When write in reiserfs_quota_write() fails, we have to properly release
i_mutex. One error path has been missing the unlock...

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoext4: add missing unlock to an error path in ext4_quota_write()
Jan Kara [Fri, 4 Jul 2008 16:59:34 +0000 (09:59 -0700)] 
ext4: add missing unlock to an error path in ext4_quota_write()

When write in ext4_quota_write() fails, we have to properly release
i_mutex.  One error path has been missing the unlock...

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoext3: add missing unlock to error path in ext3_quota_write()
Jan Kara [Fri, 4 Jul 2008 16:59:33 +0000 (09:59 -0700)] 
ext3: add missing unlock to error path in ext3_quota_write()

When write in ext3_quota_write() fails, we have to properly release
i_mutex.  One error path has been missing the unlock...

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoMiguel Ojeda has moved
Miguel Ojeda [Fri, 4 Jul 2008 16:59:33 +0000 (09:59 -0700)] 
Miguel Ojeda has moved

Signed-off-by: Miguel Ojeda <miguel.ojeda.sandonis@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agopci: VT3336 can't do MSI either
Tejun Heo [Fri, 4 Jul 2008 16:59:32 +0000 (09:59 -0700)] 
pci: VT3336 can't do MSI either

It seems VT3336 can't do msi either as with its bro 3351.  Disable it.
Reported in the following SUSE bug.

  https://bugzilla.novell.com/show_bug.cgi?id=300001

Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agortc: fix CMOS time error after writing /proc/acpi/alarm
Huacai Chen [Fri, 4 Jul 2008 16:59:31 +0000 (09:59 -0700)] 
rtc: fix CMOS time error after writing /proc/acpi/alarm

When writing /proc/acpi/alarm in adjust mode, e.g.
echo "+0000-00-00 00:00:15" >/proc/acpi/alarm
The "century" field should be read and added to "year" field before
writing, otherwise the CMOS time will go back to 2000 years ago, e.g.
# cat /proc/acpi/alarm
0008-06-21 11:38:46
Then the system time may be reset to the date of manufacture after
rebooting. This patch fixed this issue.

Signed-off-by: Huacai Chen <huacai.chen@intel.com>
Acked-by: Pavel Machek <pavel@suse.cz>
Acked-by: Zhao Yakui <yakui.zhao@intel.com>
Acked-by: Alessandro Zummo <a.zummo@towertech.it>
Acked-by: Len Brown <lenb@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agortc-x1205: Fix alarm set
Michael Hamel [Fri, 4 Jul 2008 16:59:30 +0000 (09:59 -0700)] 
rtc-x1205: Fix alarm set

I have discovered that the current version of rtc-x1205.c does not work
correctly when asked to set the alarm time by the RTC_WKALM_SET ioctl()
call.  This happens because the alarm registers do not behave like the
current-time registers.  They are non-volatile.  Two things go wrong:

- the X1205 requires a 10 msec delay after any attempt to write to the
  non-volatile registers.  The x1205_set_datetime() routine does the write
  as 8 single-byte writes without any delay.  Only the first write
  succeeds.  The second is NAKed because the chip is busy.

- the X1205 resets the RWEL bit after any write to the non-volatile
  registers.  This would lock out any further writes after the first even
  with a 10msec delay.

I fix this by doing a single 8-byte write and then waiting 10msec for the
chip to be ready.  A side effect of this change is that it will speed up
x1205_rtc_set_time() which uses the same code.

I have also implemented the 'enable' bit in the rtc_wkalm structure, which
the existing driver does not attempt to do.  I have modified both
x1205_rtc_set_alarm() to set the AL0E bit, and x1205_rtc_read_alarm() to
return it.

I have tested this patch on a LinkSys NSLU2 under OpenWRT, but on no other
hardware.  On the NSLU2 the X1205 correctly asserts its IRQ pin when the
alarm time matches the current time.

[akpm@linux-foundation.org: clean up over-parenthesisation]
Signed-off-by: Michael Hamel <mhamel@adi.co.nz>
Signed-off-by: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoget_user_pages(): fix possible page leak on oom
Oleg Nesterov [Fri, 4 Jul 2008 16:59:28 +0000 (09:59 -0700)] 
get_user_pages(): fix possible page leak on oom

get_user_pages() must not return the error when i != 0.  When pages !=
NULL we have i get_page()'ed pages.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoserial: fix serial_match_port() for dynamic major tty-device numbers
Guennadi Liakhovetski [Fri, 4 Jul 2008 16:59:28 +0000 (09:59 -0700)] 
serial: fix serial_match_port() for dynamic major tty-device numbers

As reported by Vipul Gandhi, the current serial_match_port() doesn't work
for tty-devices using dynamic major number allocation.  Fix it.

It oopses if you suspend a serial port with _dynamic_ major number.  ATM,
I think, there's only the drivers/serial/jsm/jsm_driver.c driver, that
does it in-tree.

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Tested-by: Vipul Gandhi <vcgandhi1@aol.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agofirmware: fix the request_firmware() dummy
James Bottomley [Fri, 4 Jul 2008 16:59:27 +0000 (09:59 -0700)] 
firmware: fix the request_firmware() dummy

> the build (.config attached) failed, make ends with :
> ...
>   UPD     include/linux/compile.h
>   CC      init/version.o
>   LD      init/built-in.o
>   LD      vmlinux
> drivers/built-in.o: In function `sas_request_addr':
> (.text+0x33bab): undefined reference to `request_firmware'
> drivers/built-in.o: In function `sas_request_addr':
> (.text+0x33c3f): undefined reference to `release_firmware'
> make: *** [vmlinux] Error 1

There's a slight fault in the stub logic.  It fails for FW_LOADER=m and
the user =y.

This should fix it.

This patch fixes the following 2.6.26-rc regression:
  http://bugzilla.kernel.org/show_bug.cgi?id=10730

Reviewed-by: Toralf Foerster <toralf.foerster@gmx.de>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agortc: rtc_read_alarm() handles wraparound
David Brownell [Fri, 4 Jul 2008 16:59:26 +0000 (09:59 -0700)] 
rtc: rtc_read_alarm() handles wraparound

While 0e36a9a4a788e4e92407774df76c545910810d35 ("rtc: fix readback from
/sys/class/rtc/rtc?/wakealarm") made sure that active alarms were never
returned with invalid "wildcard" fields (negative), it can still report
(wrongly) that the alarm triggers in the past.

Example, if it's now 10am, an alarm firing at 5am will be triggered
TOMORROW not today.  (Which may also be next month or next year...)

This updates that alarm handling in three ways:

  * Handle alarm rollover in the common cases of RTCs that don't
    support matching on all date fields.

  * Skip the invalid-field logic when it's not needed.

  * Minor bugfix ... tm_isdst should be ignored, it's one of the
    fields Linux doesn't maintain.

A warning is emitted for some of the unhandled rollover cases, but the
possible combinations are a bit too numerous to handle every bit of
potential hardware and firmware braindamage.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Mark Lord <lkml@rtr.ca>
Acked-by: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agomm: dirty page accounting vs VM_MIXEDMAP
Peter Zijlstra [Fri, 4 Jul 2008 16:59:24 +0000 (09:59 -0700)] 
mm: dirty page accounting vs VM_MIXEDMAP

Dirty page accounting accurately measures the amound of dirty pages in
writable shared mappings by mapping the pages RO (as indicated by
vma_wants_writenotify).  We then trap on first write and call
set_page_dirty() on the page, after which we map the page RW and
continue execution.

When we launder dirty pages, we call clear_page_dirty_for_io() which
clears both the dirty flag, and maps the page RO again before we start
writeout so that the story can repeat itself.

vma_wants_writenotify() excludes VM_PFNMAP on the basis that we cannot
do the regular dirty page stuff on raw PFNs and the memory isn't going
anywhere anyway.

The recently introduced VM_MIXEDMAP mixes both !pfn_valid() and
pfn_valid() pages in a single mapping.

We can't do dirty page accounting on !pfn_valid() pages as stated
above, and mapping them RO causes them to be COW'ed on write, which
breaks VM_SHARED semantics.

Excluding VM_MIXEDMAP in vma_wants_writenotify() would mean we don't do
the regular dirty page accounting for the pfn_valid() pages, which
would bring back all the head-aches from inaccurate dirty page
accounting.

So instead, we let the !pfn_valid() pages get mapped RO, but fix them
up unconditionally in the fault path.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Acked-by: Hugh Dickins <hugh@veritas.com>
Cc: "Jared Hulbert" <jaredeh@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoChristoph has moved
Christoph Lameter [Fri, 4 Jul 2008 16:59:22 +0000 (09:59 -0700)] 
Christoph has moved

Remove all clameter@sgi.com addresses from the kernel tree since they will
become invalid on June 27th.  Change my maintainer email address for the
slab allocators to cl@linux-foundation.org (which will be the new email
address for the future).

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoMerge branch 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzi...
Linus Torvalds [Fri, 4 Jul 2008 16:52:54 +0000 (09:52 -0700)] 
Merge branch 'upstream-linus' of git://git./linux/kernel/git/jgarzik/libata-dev

* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev:
  sata_mv: safer logic for limit_warnings
  libata-sff: improve HSM violation reporting
  ahci: always clear all bits in irq_stat
  sata_sil24: add DID for another adaptec flavor
  sata_uli: hardreset is broken

16 years agoMerge branch 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus
Linus Torvalds [Fri, 4 Jul 2008 16:51:51 +0000 (09:51 -0700)] 
Merge branch 'upstream' of git://ftp.linux-mips.org/upstream-linus

* 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus:
  [MIPS] Fix bug in atomic_sub_if_positive.

16 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/penberg...
Linus Torvalds [Fri, 4 Jul 2008 16:48:21 +0000 (09:48 -0700)] 
Merge branch 'for-linus' of git://git./linux/kernel/git/penberg/slab-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/penberg/slab-2.6:
  slub: Do not use 192 byte sized cache if minimum alignment is 128 byte

16 years agoUpdate maintainers for powerpc
Paul Mackerras [Fri, 4 Jul 2008 11:04:42 +0000 (21:04 +1000)] 
Update maintainers for powerpc

This updates the MAINTAINERS entries for powerpc.  It adds Ben H to
the overall Linux for PowerPC entry and makes it clear this covers
both 32-bit and 64-bit machines.  It removes the separate entry we had
for Linux on 64-bit PowerPC where Anton and I were listed as
maintainers - Anton hasn't been involved in the day-to-day maintenance
of the code for several years.  Finally, it removes the entry for the
Linux for PowerPC boot code where Tom Rini was listed as the
maintainer.  That code got completely rewritten when we merged
32-bit and 64-bit, and I and the various platform maintainers have
been maintaining that code since.

Signed-off-by: Paul Mackerras <paulus@samba.org>
Acked-by: Tom Rini <trini@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agox86: cacheline_align tss_struct
Richard Kennedy [Fri, 4 Jul 2008 12:56:16 +0000 (13:56 +0100)] 
x86: cacheline_align tss_struct

The manual padding to align on cacheline size only worked in 32 bit
In 64 bit the structure was not aligned and contained wasted space.

use the compiler ____cachline_aligned to save space & properly align
this structure.

x86_64_default size goes from 9136 -> 8960
x86_64_AMD     size goes from 9136 -> 8896

built & running on 2.6.26-rc8.

Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agosata_mv: safer logic for limit_warnings
Mark Lord [Thu, 19 Jun 2008 01:57:42 +0000 (21:57 -0400)] 
sata_mv: safer logic for limit_warnings

There is a miniscule chance that two separate host controllers
might be in sata_mv at the same time and manage to decrement
the static limit_warnings variable below zero.

Fix the comparison to deal with it.

Signed-off-by: Mark Lord <mlord@pobox.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agolibata-sff: improve HSM violation reporting
Tejun Heo [Fri, 27 Jun 2008 16:39:43 +0000 (01:39 +0900)] 
libata-sff: improve HSM violation reporting

Improve SFF HSM violation reporting such that each HSM violation can
be distinguished using ehi_desc.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoahci: always clear all bits in irq_stat
Tejun Heo [Fri, 27 Jun 2008 16:49:02 +0000 (01:49 +0900)] 
ahci: always clear all bits in irq_stat

Some AHCI controllers (ICH7 was reported) set pending bit in
HOST_IRQ_STAT for non-existent ports and when it's not cleared falls
into IRQ storm.  Always clear full irq_stat instead of only the bits
that are handled.  As nothing changes for recognized ports, the risk
of breaking things is pretty low.

Reported and verified by Philipp Thomas in the following suse
bugzilla.

https://bugzilla.novell.com/attachment.cgi?id=215692

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Philipp Thomas <pth@novell.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agosata_sil24: add DID for another adaptec flavor
Tejun Heo [Wed, 2 Jul 2008 08:50:23 +0000 (17:50 +0900)] 
sata_sil24: add DID for another adaptec flavor

There's another DID used for Adaptec card.  Add it.

Reported by Travis Read.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Travis Read <ics@dark.net.au>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
16 years agoxen: fix address truncation in pte mfn<->pfn conversion
Jeremy Fitzhardinge [Fri, 4 Jul 2008 05:10:18 +0000 (22:10 -0700)] 
xen: fix address truncation in pte mfn<->pfn conversion

When converting the page number in a pte/pmd/pud/pgd between
machine and pseudo-physical addresses, the converted result was
being truncated at 32-bits.  This caused failures on machines
with more than 4G of physical memory.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: "Christopher S. Aker" <caker@theshore.net>
Cc: Ian Campbell <Ian.Campbell@eu.citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years ago[MIPS] Fix bug in atomic_sub_if_positive.
Ralf Baechle [Thu, 3 Jul 2008 22:28:35 +0000 (23:28 +0100)] 
[MIPS] Fix bug in atomic_sub_if_positive.

The branch optimization fixes in 2.6.21 introduced a bug in
atomic_sub_if_positive that causes it to return even when the sc
instruction fails. The result is that e.g. down_trylock becomes unreliable
as the semaphore counter is not always decremented.

Original MUA-shredded patch from Morten Larsen <mlarsen@broadcom.com>.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb-2.6
Linus Torvalds [Fri, 4 Jul 2008 04:29:08 +0000 (21:29 -0700)] 
Merge git://git./linux/kernel/git/gregkh/usb-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb-2.6:
  USB: adding comment for ipaq forcing number of ports
  USB: fix Oops on loading ipaq module since 2.6.26
  USB: add a pl2303 device id
  USB: another option device id
  USB: don't lose disconnections during suspend
  USB: fix interrupt disabling for HCDs with shared interrupt handlers
  USB: New device ID for ftdi_sio driver
  sisusbvga: Fix oops on disconnect.
  USB: mass storage: new id for US_SC_CYP_ATACB
  USB: ohci - record data toggle after unlink
  USB: ehci - fix timer regression
  USB: fix cdc-acm resume()
  OHCI: Fix problem if SM501 and another platform driver is selected

16 years agoUSB: adding comment for ipaq forcing number of ports
Oliver Neukum [Thu, 3 Jul 2008 15:14:16 +0000 (17:14 +0200)] 
USB: adding comment for ipaq forcing number of ports

The reason for forcing a number of ports should be documented.

Signed-off-by: Oliver Neukum <oneukum@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
16 years agoUSB: fix Oops on loading ipaq module since 2.6.26
Oliver Neukum [Thu, 3 Jul 2008 08:05:57 +0000 (10:05 +0200)] 
USB: fix Oops on loading ipaq module since 2.6.26

Fixes bugzilla.kernel.org #10868

Signed-off-by: Oliver Neukum <oneukum@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
16 years agoUSB: add a pl2303 device id
Greg Kroah-Hartman [Wed, 2 Jul 2008 20:25:41 +0000 (15:25 -0500)] 
USB: add a pl2303 device id

As reported by Ken A Scott <kscott9@sent.com>

Cc: Ken A Scott <kscott9@sent.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
16 years agoUSB: another option device id
Greg Kroah-Hartman [Tue, 1 Jul 2008 07:41:56 +0000 (13:11 +0530)] 
USB: another option device id

Thanks to umesh b <umesh.kollam@gmail.com> for the information here.

Cc: umesh b <umesh.kollam@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
16 years agoUSB: don't lose disconnections during suspend
Alan Stern [Tue, 1 Jul 2008 14:45:51 +0000 (10:45 -0400)] 
USB: don't lose disconnections during suspend

This patch (as1111) fixes a bug in the hub driver.  When a hub
resumes, disconnections that occurred while the hub was suspended are
lost.

A completely different fix for this problem has already been accepted
for 2.6.27; however the problem still needs to be handled in 2.6.26.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Tested-by: Lukas Hejtmanek <xhejtman@ics.muni.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
16 years agoUSB: fix interrupt disabling for HCDs with shared interrupt handlers
Stefan Becker [Tue, 1 Jul 2008 16:19:22 +0000 (19:19 +0300)] 
USB: fix interrupt disabling for HCDs with shared interrupt handlers

USB: fix interrupt disabling for HCDs with shared interrupt handlers

As has been discussed several times on LKML, IRQF_SHARED | IRQF_DISABLED
doesn't work reliably, i.e. a shared interrupt handler CAN'T be certain to
be called with interrupts disabled. Most USB HCD handlers use IRQF_DISABLED
and therefore havoc can break out if they share their interrupt with a
handler that doesn't use it.

On my test machine the yenta_socket interrupt handler (no IRQF_DISABLED)
was registered before ehci_hcd and one uhci_hcd instance. Therefore all
usb_hcd_irq() invocations for ehci_hcd and for one uhci_hcd instance
happened with interrupts enabled. That led to random lockups as USB core
HCD functions that acquire the same spinlock could be called twice
from interrupt handlers.

This patch updates usb_hcd_irq() to always disable/restore interrupts.
usb_add_hcd() will silently remove any IRQF_DISABLED requested from HCD code.

Signed-off-by: Stefan Becker <stefan.becker@nokia.com>
Cc: stable <stable@kernel.org>
Acked-by: David Brownell <david-b@pacbell.net>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
16 years agoUSB: New device ID for ftdi_sio driver
Jon K Hellan [Tue, 24 Jun 2008 09:43:13 +0000 (11:43 +0200)] 
USB: New device ID for ftdi_sio driver

Here's a new device ID for the ftdio_sio driver.
The diff is with linus's tree as of this morning.

The device is the RigExpert Tiny USB Soundcard Transceiver Interface for ham
radio.

(I didn't actually test this. A fellow ham couldn't get the device to work, and
I suggested binding the device ID using sysfs - see
"http://jk.ufisa.uninett.no/usb/". However, he had had moved on to other things
by then. I guess adding the device ID to the kernel "on spec" won't hurt.
The relevant part of cat /proc/bus/usb/devices shows:

T:  Bus=02 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#=  2 Spd=12  MxCh= 0
D:  Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS= 8 #Cfgs=  1
P:  Vendor=0403 ProdID=ed22 Rev= 5.00
S:  Manufacturer=FTDI
S:  Product=MixW RigExpert Tiny
S:  SerialNumber=00000000
C:* #Ifs= 2 Cfg#= 1 Atr=80 MxPwr=100mA
I:* If#= 0 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
E:  Ad=81(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=02(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
E:  Ad=83(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=04(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
)

From: Jon K Hellan <hellan@acm.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
16 years agosisusbvga: Fix oops on disconnect.
Will Newton [Fri, 27 Jun 2008 12:08:08 +0000 (13:08 +0100)] 
sisusbvga: Fix oops on disconnect.

Remove dev_info call on disconnect. The sisusb_dev pointer may have been
set to zero by sisusb_delete at this point causing an oops.

The message does not provide any extra information over the standard USB
subsystem output so removing it does not affect functionality.

Signed-off-by: Will Newton <will.newton@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
16 years agoUSB: mass storage: new id for US_SC_CYP_ATACB
matthieu castet [Mon, 16 Jun 2008 17:49:06 +0000 (19:49 +0200)] 
USB: mass storage: new id for US_SC_CYP_ATACB

CY7C68310 chip also support cypress atacb "ATA command" pass_thru.

Signed-off-by: Matthieu CASTET <castet.matthieu@free.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
16 years agoUSB: ohci - record data toggle after unlink
David Brownell [Sat, 14 Jun 2008 06:59:54 +0000 (23:59 -0700)] 
USB: ohci - record data toggle after unlink

This patch fixes a problem with OHCI where canceling bulk or
interrupt URBs may lose track of the right data toggle.  This
seems to be a longstanding bug, possibly dating back to the
Linux 2.4 kernel, which stayed hidden because

 (a) about half the time the data toggle bit was correct;
 (b) canceling such URBs is unusual; and
 (c) the few drivers which cancel these URBs either
      [1] do it only as part of shutting down, or
      [2] have fault recovery logic, which recovers.

For those transfer types, the toggle is normally written back
into the ED when each TD is retired.  But canceling bypasses
the mechanism used to retire TDs ... so on average, half the
time the toggle bit will be invalid after cancelation.

The fix is simple:  the toggle state of any canceled TDs are
propagated back to the ED in the finish_unlinks function.

(Issue found by leonidv11@gmail.com ...)

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Leonid <leonidv11@gmail.com>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
16 years agoUSB: ehci - fix timer regression
David Brownell [Sat, 14 Jun 2008 06:56:48 +0000 (23:56 -0700)] 
USB: ehci - fix timer regression

This patch fixes a regression in the EHCI driver's TIMER_IO_WATCHDOG
behavior.  The patch "USB: EHCI: add separate IAA watchdog timer" changed
how that timer is handled, so that short timeouts on the remaining
timer (unfortunately, overloaded) would never be used.

This takes a more direct approach, reorganizing the code slightly to
be explicit about only the I/O watchdog role now being overridable.
It also replaces a now-obsolete comment describing older timer behavior.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Leonid <leonidv11@gmail.com>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
16 years agoUSB: fix cdc-acm resume()
Oliver Neukum [Tue, 17 Jun 2008 20:30:48 +0000 (22:30 +0200)] 
USB: fix cdc-acm resume()

cdc-acm has
- a memory leak in resume()
- will fail to reactivate the read code path if this is needed.
his corrects it by deleting the useless relict code.

Signed-off-by: Oliver Neukum <oneukum@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
16 years agoOHCI: Fix problem if SM501 and another platform driver is selected
Ben Dooks [Sun, 8 Jun 2008 16:20:11 +0000 (17:20 +0100)] 
OHCI: Fix problem if SM501 and another platform driver is selected

If the SM501 and another platform driver, such as the SM501
then we end up defining PLATFORM_DRIVER twice. This patch
seperated the SM501 onto a seperate define of SM501_OHCI_DRIVER
so that it can be selected without overwriting the original
definition.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Acked-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
16 years agosvcrpc: fix handling of garbage args
J. Bruce Fields [Thu, 3 Jul 2008 19:26:35 +0000 (15:26 -0400)] 
svcrpc: fix handling of garbage args

To return garbage_args, the accept_stat must be 0, and we must have a
verifier.  So we shouldn't be resetting the write pointer as we reject
the call.

Also, we must add the two placeholder words here regardless of success
of the unwrap, to ensure the output buffer is left in a consistent state
for svcauth_gss_release().

This fixes a BUG() in svcauth_gss.c:svcauth_gss_release().

Thanks to Aime Le Rouzic for bug report, debugging help, and testing.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Tested-by: Aime Le Rouzic <aime.le-rouzic@bull.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoMerge branch 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus
Linus Torvalds [Thu, 3 Jul 2008 18:37:19 +0000 (11:37 -0700)] 
Merge branch 'upstream' of git://ftp.linux-mips.org/upstream-linus

* 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus:
  [MIPS] cevt-txx9: Reset timer counter on initialization
  [MIPS] IP22: Fix crashes due to wrong L1_CACHE_BYTES
  [MIPS] IP32: Fix unexpected irq 71

16 years agohrtimer: prevent migration for raising softirq
Steven Rostedt [Thu, 3 Jul 2008 18:31:26 +0000 (14:31 -0400)] 
hrtimer: prevent migration for raising softirq

Due to a possible deadlock, the waking of the softirq was pushed outside
of the hrtimer base locks. See commit 0c96c5979a522c3323c30a078a70120e29b5bdbc

Unfortunately this allows the task to migrate after setting up the softirq
and raising it. Since softirqs run a queue that is per-cpu we may raise the
softirq on the wrong CPU and this will keep the queued softirq task from
running.

To solve this issue, this patch disables preemption around the releasing
of the hrtimer lock and raising of the softirq.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years ago[MIPS] cevt-txx9: Reset timer counter on initialization
Atsushi Nemoto [Tue, 24 Jun 2008 14:26:38 +0000 (23:26 +0900)] 
[MIPS] cevt-txx9: Reset timer counter on initialization

The txx9_tmr_init() will not clear a timer counter register in a certain
case.  The counter register is cleared on 1->0 transition of TCE bit if
CRE=1.  So just clearing the TCE bit is not enough.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
16 years ago[MIPS] IP22: Fix crashes due to wrong L1_CACHE_BYTES
Thomas Bogendoerfer [Fri, 27 Jun 2008 21:52:26 +0000 (23:52 +0200)] 
[MIPS] IP22: Fix crashes due to wrong L1_CACHE_BYTES

The introduction of a real dma cache invalidate makes it important
to have a correct cache line size, otherwise the kernel will gives
out two memory segment, which might share one cache line. The R4400
Indy/Indigo2 CPU modules are using a second level cache line size
of 128 bytes, so MIPS_L1_CACHE_SHIFT needs to be bumped up to 7 for
IP22.

Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
16 years ago[MIPS] IP32: Fix unexpected irq 71
Thomas Bogendoerfer [Mon, 23 Jun 2008 22:48:05 +0000 (00:48 +0200)] 
[MIPS] IP32: Fix unexpected irq 71

It's possible that the crime interrupt handler is called without
pending interrupts (probably a hardware issue). To avoid irritating
"unexpected irq 71" messages, we now just ignore the spurious crime
interrupts.

Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
16 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ericvh...
Linus Torvalds [Thu, 3 Jul 2008 16:28:44 +0000 (09:28 -0700)] 
Merge branch 'for-linus' of git://git./linux/kernel/git/ericvh/v9fs

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ericvh/v9fs:
  9p: fix O_APPEND in legacy mode

16 years agoDo not overwrite nr_zones on !NUMA when initialising zlcache_ptr
Mel Gorman [Thu, 3 Jul 2008 04:27:51 +0000 (05:27 +0100)] 
Do not overwrite nr_zones on !NUMA when initialising zlcache_ptr

The non-NUMA case of build_zonelist_cache() would initialize the
zlcache_ptr for both node_zonelists[] to NULL.

Which is problematic, since non-NUMA only has a single node_zonelists[]
entry, and trying to zero the non-existent second one just overwrote the
nr_zones field instead.

As kswapd uses this value to determine what reclaim work is necessary,
the result is that kswapd never reclaims.  This causes processes to
stall frequently in low-memory situations as they always direct reclaim.
This patch initialises zlcache_ptr correctly.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Tested-by: Dan Williams <dan.j.williams@intel.com>
[ Simplified patch a bit ]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoslub: Do not use 192 byte sized cache if minimum alignment is 128 byte
Christoph Lameter [Thu, 3 Jul 2008 14:14:26 +0000 (09:14 -0500)] 
slub: Do not use 192 byte sized cache if minimum alignment is 128 byte

The 192 byte cache is not necessary if we have a basic alignment of 128
byte. If it would be used then the 192 would be aligned to the next 128 byte
boundary which would result in another 256 byte cache. Two 256 kmalloc caches
cause sysfs to complain about a duplicate entry.

MIPS needs 128 byte aligned kmalloc caches and spits out warnings on boot without
this patch.

Signed-off-by: Christoph Lameter <cl@linux-foundation.org>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
16 years ago9p: fix O_APPEND in legacy mode
Eric Van Hensbergen [Tue, 24 Jun 2008 22:39:39 +0000 (17:39 -0500)] 
9p: fix O_APPEND in legacy mode

The legacy protocol's open operation doesn't handle an append operation
(it is expected that the client take care of it).  We were incorrectly
passing the extended protocol's flag through even in legacy mode.  This
was reported in bugzilla report #10689.  This patch fixes the problem
by disallowing extended protocol open modes from being passed in legacy
mode and implemented append functionality on the client side by adding
a seek after the open.

Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
16 years agox86: remove unnecessary #ifdef CONFIG_X86_32...#else
Gustavo Fernando Padovan [Wed, 25 Jun 2008 07:03:19 +0000 (04:03 -0300)] 
x86: remove unnecessary #ifdef CONFIG_X86_32...#else

Remove the #ifdef conditional because this comparison is already done in
user_mode_vm().

Signed-off-by: Gustavo F. Padovan <gustavo@las.ic.unicamp.br>
Cc: akpm@osdl.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agox86, mce_64.c: mce_cpu_quirks being ignored
Venki Pallipadi [Wed, 25 Jun 2008 00:12:56 +0000 (17:12 -0700)] 
x86, mce_64.c: mce_cpu_quirks being ignored

Quirks getting ignored was a bug. Below patch fixes the bug, until
we have the dynamic banks support.

Sysfs choice configuration should not have any issues with the earlier patch
as we look for NR_SYSFS_BANKS in do_machine_check().

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Max Asbock <masbock@us.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agoMerge commit 'v2.6.26-rc8' into x86/mce
Ingo Molnar [Thu, 3 Jul 2008 13:03:02 +0000 (15:03 +0200)] 
Merge commit 'v2.6.26-rc8' into x86/mce

16 years agox86: microcode: cosmetic changes
Ben Castricum [Tue, 10 Jun 2008 11:15:12 +0000 (13:15 +0200)] 
x86: microcode: cosmetic changes

First announce ourself, then start working. Currently this module reports
itself when all is completed which is not most modules do. Plus some
cosmetic/whitespace cleanups.

Signed-off-by: Ben Castricum <lk0806@bencastricum.nl>
Cc: trivial@kernel.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agoarch/x86/mm/init_64.c: early_memtest(): fix types
Andrew Morton [Thu, 3 Jul 2008 08:14:10 +0000 (10:14 +0200)] 
arch/x86/mm/init_64.c: early_memtest(): fix types

fix this warning:

arch/x86/mm/init_64.c: In function 'early_memtest':
arch/x86/mm/init_64.c:524: warning: passing argument 2 of 'find_e820_area_size' from incompatible pointer type

Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agox86: fix Intel Mac booting with EFI
Hugh Dickins [Wed, 2 Jul 2008 21:48:03 +0000 (22:48 +0100)] 
x86: fix Intel Mac booting with EFI

Fedora reports that mem_init()'s zap_low_mappings(), extended to SMP in
61165d7a035f6571c7576e7f51e7230157724c8d x86: fix app crashes after SMP
resume causes 32-bit Intel Mac machines to reboot very early when
booting with EFI.

The EFI code appears to manage low mappings for itself when needed; but
like many before it, confuses PSE with PAE.  So it has only been mapping
half the space it needed when PSE but not PAE.  This remained unnoticed
until we moved the SMP zap_low_mappings() before
efi_enter_virtual_mode().  Presumably could have been noticed years ago
if anyone ran a UP kernel on such machines?

Reported-by: Peter Jones <pjones@redhat.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: Peter Jones <pjones@redhat.com>
Cc: Glauber Costa <gcosta@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Tested-by: Peter Jones <pjones@redhat.com>
16 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes...
Linus Torvalds [Thu, 3 Jul 2008 02:29:16 +0000 (19:29 -0700)] 
Merge branch 'for-linus' of git://git./linux/kernel/git/jbarnes/pci-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6:
  PCI: acpiphp: cleanup notify handler on all root bridges
  PCI: Limit VPD read/write lengths for Broadcom 5706, 5708, 5709 rev.
  PCI: Restrict VPD read permission to root

16 years agoMerge branch 'i2c-fix' of git://aeryn.fluff.org.uk/bjdooks/linux
Linus Torvalds [Thu, 3 Jul 2008 02:26:38 +0000 (19:26 -0700)] 
Merge branch 'i2c-fix' of git://aeryn.fluff.org.uk/bjdooks/linux

* 'i2c-fix' of git://aeryn.fluff.org.uk/bjdooks/linux:
  I2C: S3C2410: Add MODULE_ALIAS() for s3c2440 device.
  I2C: S3C2410: Fixup error codes returned rom a transfer.
  I2C: S3C2410: Check ACK on byte transmission

16 years agoMerge branch 'for-2.6.26' of git://git.kernel.dk/linux-2.6-block
Linus Torvalds [Thu, 3 Jul 2008 02:25:36 +0000 (19:25 -0700)] 
Merge branch 'for-2.6.26' of git://git.kernel.dk/linux-2.6-block

* 'for-2.6.26' of git://git.kernel.dk/linux-2.6-block:
  Properly notify block layer of sync writes
  block: Fix the starving writes bug in the anticipatory IO scheduler

16 years agoMerge branch 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6
Linus Torvalds [Thu, 3 Jul 2008 02:24:48 +0000 (19:24 -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] export account_system_vtime
  [IA64] Bugfix for system with 32 cpus

16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/v4l-dvb
Linus Torvalds [Thu, 3 Jul 2008 02:23:52 +0000 (19:23 -0700)] 
Merge git://git./linux/kernel/git/mchehab/v4l-dvb

* git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/v4l-dvb:
  V4L/DVB (8178): uvc: Fix compilation breakage for the other drivers, if uvc is selected
  V4L/DVB (8145a): USB Video Class driver

16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/bart/ide-2.6
Linus Torvalds [Thu, 3 Jul 2008 02:22:54 +0000 (19:22 -0700)] 
Merge git://git./linux/kernel/git/bart/ide-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/bart/ide-2.6:
  ide: fix /proc/ide/ide?/mate reporting
  Revert "BAST: Remove old IDE driver"

16 years agoMerge master.kernel.org:/home/rmk/linux-2.6-arm
Linus Torvalds [Thu, 3 Jul 2008 02:22:25 +0000 (19:22 -0700)] 
Merge master.kernel.org:/home/rmk/linux-2.6-arm

* master.kernel.org:/home/rmk/linux-2.6-arm:
  [ARM] 5131/1: Annotate platform_secondary_init with trace_hardirqs_off
  [ARM] 5117/1: pxafb: fix __devinit/exit annotations
  [ARM] Export dma_sync_sg_for_device()
  [ARM] 5109/1: Mark rtc sa1100 driver as wakeup source before registering it
  [ARM] 5116/1: pxafb: cleanup and fix order of failure handling
  [ARM] 5115/1: pxafb: fix ifdef for command line option handling
  ARM: OMAP: Correcting the gpmc prefetch control register address
  ARM: OMAP: DMA: Don't mark channel active in omap_enable_channel_irq

16 years agotty: Fix inverted logic in send_break
Alan Cox [Mon, 30 Jun 2008 16:40:08 +0000 (17:40 +0100)] 
tty: Fix inverted logic in send_break

Not sure how this came to get inverted but it appears to have been my
mess up.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoMerge branch 'sched-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Thu, 3 Jul 2008 02:12:53 +0000 (19:12 -0700)] 
Merge branch 'sched-fixes-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'sched-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  sched: fix divide error when trying to configure rt_period to zero

16 years agoMerge branch 'i2c-for-linus' of git://jdelvare.pck.nerim.net/jdelvare-2.6
Linus Torvalds [Thu, 3 Jul 2008 02:00:29 +0000 (19:00 -0700)] 
Merge branch 'i2c-for-linus' of git://jdelvare.pck.nerim.net/jdelvare-2.6

* 'i2c-for-linus' of git://jdelvare.pck.nerim.net/jdelvare-2.6:
  i2c: Fix bad hint about irqs in i2c.h
  i2c: Documentation: fix device matching description

16 years agoMerge branch 'core-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Thu, 3 Jul 2008 01:59:45 +0000 (18:59 -0700)] 
Merge branch 'core-fixes-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'core-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  rcu: fix hotplug vs rcu race

16 years agoMerge branch 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Thu, 3 Jul 2008 01:58:56 +0000 (18:58 -0700)] 
Merge branch 'x86-fixes-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  x86: fix NODES_SHIFT Kconfig range

16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6
Linus Torvalds [Thu, 3 Jul 2008 01:57:33 +0000 (18:57 -0700)] 
Merge git://git./linux/kernel/git/jejb/scsi-rc-fixes-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6:
  [SCSI] esp: tidy up target reference counting
  [SCSI] esp: Fix OOPS in esp_reset_cleanup().
  [SCSI] ses: Fix timeout

16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/agk/linux-2.6-dm
Linus Torvalds [Thu, 3 Jul 2008 01:55:17 +0000 (18:55 -0700)] 
Merge git://git./linux/kernel/git/agk/linux-2.6-dm

* git://git.kernel.org/pub/scm/linux/kernel/git/agk/linux-2.6-dm:
  dm crypt: use cond_resched

16 years agoMerge branch 'for-2.6.26' of git://neil.brown.name/md
Linus Torvalds [Thu, 3 Jul 2008 01:47:10 +0000 (18:47 -0700)] 
Merge branch 'for-2.6.26' of git://neil.brown.name/md

* 'for-2.6.26' of git://neil.brown.name/md:
  Fix error paths if md_probe fails.
  Don't acknowlege that stripe-expand is complete until it really is.
  Ensure interrupted recovery completed properly (v1 metadata plus bitmap)

16 years agoMerge branch 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc
Linus Torvalds [Thu, 3 Jul 2008 01:45:29 +0000 (18:45 -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/mpc5200: Fix lite5200b suspend/resume
  powerpc/legacy_serial: Bail if reg-offset/shift properties are present
  powerpc/bootwrapper: update for initrd with simpleImage

16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Thu, 3 Jul 2008 01:43:16 +0000 (18:43 -0700)] 
Merge git://git./linux/kernel/git/davem/net-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (55 commits)
  net: fib_rules: fix error code for unsupported families
  netdevice: Fix wrong string handle in kernel command line parsing
  net: Tyop of sk_filter() comment
  netlink: Unneeded local variable
  net-sched: fix filter destruction in atm/hfsc qdisc destruction
  net-sched: change tcf_destroy_chain() to clear start of filter list
  ipv4: fix sysctl documentation of time related values
  mac80211: don't accept WEP keys other than WEP40 and WEP104
  hostap: fix sparse warnings
  hostap: don't report useless WDS frames by default
  textsearch: fix Boyer-Moore text search bug
  netfilter: nf_conntrack_tcp: fixing to check the lower bound of valid ACK
  ipv6 route: Convert rt6_device_match() to use RT6_LOOKUP_F_xxx flags.
  netlabel: Fix a problem when dumping the default IPv6 static labels
  net/inet_lro: remove setting skb->ip_summed when not LRO-able
  inet fragments: fix race between inet_frag_find and inet_frag_secret_rebuild
  CONNECTOR: add a proc entry to list connectors
  netlink: Fix some doc comments in net/netlink/attr.c
  tcp: /proc/net/tcp rto,ato values not scaled properly (v2)
  include/linux/netdevice.h: don't export MAX_HEADER to userspace
  ...

16 years agoDRM/i915: only use tiled blits on 965+
Jesse Barnes [Tue, 1 Jul 2008 19:32:52 +0000 (12:32 -0700)] 
DRM/i915: only use tiled blits on 965+

When scheduled swaps occur, we need to blit between front & back
buffers.  If the buffers are tiled, we need to set the appropriate
XY_SRC_COPY tile bit, but only on 965 chips, since it will cause
corruption on pre-965 (e.g. 945).

Bug reported by and fix tested by Tomas Janousek <tomi@nomi.cz>.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Acked-by: Dave Airlie <airlied@linux.ie>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agodrivers/input/ff-core.c needs <linux/sched.h>
Geert Uytterhoeven [Tue, 1 Jul 2008 11:55:25 +0000 (13:55 +0200)] 
drivers/input/ff-core.c needs <linux/sched.h>

Commit 656acd2bbc4ce7f224de499ee255698701396c48 ("Input: fix locking in
force-feedback core") causes the following regression on m68k:

| linux/drivers/input/ff-core.c: In function 'input_ff_upload':
| linux/drivers/input/ff-core.c:172: error: dereferencing pointer to incomplete type
| linux/drivers/input/ff-core.c: In function 'erase_effect':
| linux/drivers/input/ff-core.c:197: error: dereferencing pointer to incomplete type
| linux/drivers/input/ff-core.c:204: error: dereferencing pointer to incomplete type
| make[4]: *** [drivers/input/ff-core.o] Error 1

As the incomplete type is `struct task_struct', including <linux/sched.h> fixes
it.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Acked-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoMerge branch 'for-2.6.26' of git://git.secretlab.ca/git/linux-2.6-mpc52xx into merge
Paul Mackerras [Thu, 3 Jul 2008 00:05:59 +0000 (10:05 +1000)] 
Merge branch 'for-2.6.26' of git://git.secretlab.ca/git/linux-2.6-mpc52xx into merge

16 years agoPCI: acpiphp: cleanup notify handler on all root bridges
Alex Chiang [Wed, 2 Jul 2008 02:02:23 +0000 (20:02 -0600)] 
PCI: acpiphp: cleanup notify handler on all root bridges

During the development of the physical PCI slot patch series, Gary Hade
kept on reporting strange oopses due to interactions between pci_slot
and acpiphp.

http://lkml.org/lkml/2007/11/28/319

find_root_bridges() unconditionally installs
handle_hotplug_event_bridge() as an ACPI_SYSTEM_NOTIFY handler for all
root bridges.

However, during module cleanup, remove_bridge() will only remove the
notify handler iff the root bridge had a hot-pluggable slot directly
underneath. That is:

root bridge -> hotplug slot

But, if the topology looks like either of the following:

root bridge -> non-hotplug slot
root bridge -> p2p bridge -> hotplug slot

Then we currently do not remove the notify handler from that root
bridge.

This can cause a kernel oops if we modprobe acpiphp later and it gets
loaded somewhere else in memory. If the root bridge then receives a
hotplug event, it will then attempt to call a stale, non-existent notify
handler and we blow up.

Much thanks goes to Gary Hade for his persistent debugging efforts.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Gary Hade <garyhade@us.ibm.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
16 years agoPCI: Limit VPD read/write lengths for Broadcom 5706, 5708, 5709 rev.
Benjamin Li [Wed, 2 Jul 2008 17:59:04 +0000 (10:59 -0700)] 
PCI: Limit VPD read/write lengths for Broadcom 5706, 5708, 5709 rev.

For Broadcom 5706, 5708, 5709 rev. A nics, any read beyond the
VPD end tag will hang the device.  This problem was initially
observed when a vpd entry was created in sysfs
('/sys/bus/pci/devices/<id>/vpd').   A read to this sysfs entry
will dump 32k of data.  Reading a full 32k will cause an access
beyond the VPD end tag causing the device to hang.  Once the device
is hung, the bnx2 driver will not be able to reset the device.
We believe that it is legal to read beyond the end tag and
therefore the solution is to limit the read/write length.

A majority of this patch is from Matthew Wilcox who gave code for
reworking the PCI vpd size information.  A PCI quirk added for the
Broadcom NIC's to limit the read/write's.

Signed-off-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
16 years agoV4L/DVB (8178): uvc: Fix compilation breakage for the other drivers, if uvc is selected
Mauro Carvalho Chehab [Wed, 2 Jul 2008 14:03:33 +0000 (11:03 -0300)] 
V4L/DVB (8178): uvc: Fix compilation breakage for the other drivers, if uvc is selected

UVC makefile defines obj as:
obj-$(CONFIG_USB_VIDEO_CLASS) := uvcvideo.o
Instead of:
obj-$(CONFIG_USB_VIDEO_CLASS) += uvcvideo.o

Due to that, if uvc is selected, all obj-y or obj-m that were added to
compilation were forget. This breaks a proper kernel build.

Acked-by: Laurent Pinchart <laurent.pinchart@skynet.be>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
16 years agodm crypt: use cond_resched
Milan Broz [Wed, 2 Jul 2008 08:34:28 +0000 (09:34 +0100)] 
dm crypt: use cond_resched

Add cond_resched() to prevent monopolising CPU when processing large bios.

dm-crypt processes encryption of bios in sector units.  If the bio request
is big it can spend a long time in the encryption call.

Signed-off-by: Milan Broz <mbroz@redhat.com>
Tested-by: Yan Li <elliot.li.tech@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
16 years agonet: fib_rules: fix error code for unsupported families
Patrick McHardy [Wed, 2 Jul 2008 02:59:37 +0000 (19:59 -0700)] 
net: fib_rules: fix error code for unsupported families

The errno code returned must be negative.

Fixes "RTNETLINK answers: Unknown error 18446744073709551519".

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agonetdevice: Fix wrong string handle in kernel command line parsing
Wang Chen [Wed, 2 Jul 2008 02:57:19 +0000 (19:57 -0700)] 
netdevice: Fix wrong string handle in kernel command line parsing

v1->v2: Use strlcpy() to ensure s[i].name be null-termination.

1. In netdev_boot_setup_add(), a long name will leak.
   ex. : dev=21,0x1234,0x1234,0x2345,eth123456789verylongname.........
2. In netdev_boot_setup_check(), mismatch will happen if s[i].name
   is a substring of dev->name.
   ex. : dev=...eth1 dev=...eth11

[ With feedback from Ben Hutchings. ]

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agonet: Tyop of sk_filter() comment
Wang Chen [Wed, 2 Jul 2008 02:55:40 +0000 (19:55 -0700)] 
net: Tyop of sk_filter() comment

Parameter "needlock" no long exists.

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agonetlink: Unneeded local variable
Wang Chen [Wed, 2 Jul 2008 02:55:09 +0000 (19:55 -0700)] 
netlink: Unneeded local variable

We already have a variable, which has the same capability.

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agonet-sched: fix filter destruction in atm/hfsc qdisc destruction
Patrick McHardy [Wed, 2 Jul 2008 02:53:09 +0000 (19:53 -0700)] 
net-sched: fix filter destruction in atm/hfsc qdisc destruction

Filters need to be destroyed before beginning to destroy classes
since the destination class needs to still be alive to unbind the
filter.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agonet-sched: change tcf_destroy_chain() to clear start of filter list
Patrick McHardy [Wed, 2 Jul 2008 02:52:38 +0000 (19:52 -0700)] 
net-sched: change tcf_destroy_chain() to clear start of filter list

Pass double tcf_proto pointers to tcf_destroy_chain() to make it
clear the start of the filter list for more consistency.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoipv4: fix sysctl documentation of time related values
Stephen Hemminger [Wed, 2 Jul 2008 00:22:48 +0000 (17:22 -0700)] 
ipv4: fix sysctl documentation of time related values

These sysctl values are time related and all use the same routine
(proc_dointvec_jiffies) that internally converts from seconds to jiffies.
The code is fine, the documentation is just wrong.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agopowerpc/mpc5200: Fix lite5200b suspend/resume
Tim Yamin [Tue, 17 Jun 2008 08:33:14 +0000 (09:33 +0100)] 
powerpc/mpc5200: Fix lite5200b suspend/resume

Suspend/resume ("echo mem > /sys/power/state") does not work with
vanilla kernels -- the system does not suspend correctly and just
hangs. This patch fixes this so suspend/resume works:

1) of_iomap does not map the whole 0xC000 of the MPC5200 immr so
saving registers does not work.
2) PCI registers need to be saved and restored.

Signed-off-by: Tim Yamin <plasm@roo.me.uk>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
16 years agopowerpc/legacy_serial: Bail if reg-offset/shift properties are present
John Linn [Tue, 1 Jul 2008 17:52:41 +0000 (10:52 -0700)] 
powerpc/legacy_serial: Bail if reg-offset/shift properties are present

The legacy serial driver does not work with an 8250 type UART that is
described in the device tree with the reg-offset and reg-shift
properties.  This change makes legacy_serial ignore these devices.

Signed-off-by: John Linn <john.linn@xilinx.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
16 years agoi2c: Fix bad hint about irqs in i2c.h
Wolfram Sang [Tue, 1 Jul 2008 20:38:18 +0000 (22:38 +0200)] 
i2c: Fix bad hint about irqs in i2c.h

i2c.h mentions -1 as a not-issued irq. This false hint was taken by
of_i2c and caused crashes. Don't give any advice as 'no irq' is not
consistent across all architectures yet and it is not needed internally
by the i2c-core.

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
16 years agoi2c: Documentation: fix device matching description
Ben Dooks [Tue, 1 Jul 2008 20:38:18 +0000 (22:38 +0200)] 
i2c: Documentation: fix device matching description

The matching process described for new style clients in
Documentation/i2c/writing-clients is classed as out-of-date
as it requires the presence of an .id_table entry in the
driver's i2c_driver entry.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
16 years agopowerpc/bootwrapper: update for initrd with simpleImage
John Linn [Tue, 1 Jul 2008 20:17:18 +0000 (14:17 -0600)] 
powerpc/bootwrapper: update for initrd with simpleImage

This change to the makefile corrects the build of a simpleImage with initrd.

Signed-off-by: John Linn <john.linn@xilinx>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
16 years agoPCI: Restrict VPD read permission to root
Ben Hutchings [Tue, 1 Jul 2008 16:18:17 +0000 (17:18 +0100)] 
PCI: Restrict VPD read permission to root

Some PCI devices will lock up if we attempt to read from VPD addresses
beyond some device-dependent limit.  Until we can identify these
devices and adjust the file size accordingly, only let root read VPD
through sysfs to prevent a DoS by normal users.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
16 years agoI2C: S3C2410: Add MODULE_ALIAS() for s3c2440 device.
Ben Dooks [Tue, 1 Jul 2008 10:59:43 +0000 (11:59 +0100)] 
I2C: S3C2410: Add MODULE_ALIAS() for s3c2440 device.

Add a MODULE_ALIAS() statement for the i2c-s3c2410 controller
to ensure that it can be autoloaded on the S3C2440 systems that
we support.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
16 years agoI2C: S3C2410: Fixup error codes returned rom a transfer.
Ben Dooks [Tue, 1 Jul 2008 10:59:42 +0000 (11:59 +0100)] 
I2C: S3C2410: Fixup error codes returned rom a transfer.

The driver should be returning -ENXIO for transfers that do not
pass the initial address byte stage.

Note, also small tidyups to the driver comments in the area.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
16 years agoI2C: S3C2410: Check ACK on byte transmission
Ben Dooks [Tue, 1 Jul 2008 10:59:41 +0000 (11:59 +0100)] 
I2C: S3C2410: Check ACK on byte transmission

We should check for the reception of an ACK after transmitting each
data byte. The address send has been correctly checking this, but the
data write byte state should have also been checking for these failures.

As part of the same fix, we remove the ACK checking from the receive
path where it should not have been checking for an ACK which our hardware
was sending.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
16 years agorcu: fix hotplug vs rcu race
Gautham R Shenoy [Fri, 27 Jun 2008 04:47:38 +0000 (10:17 +0530)] 
rcu: fix hotplug vs rcu race

Dhaval Giani reported this warning during cpu hotplug stress-tests:

| On running kernel compiles in parallel with cpu hotplug:
|
| WARNING: at arch/x86/kernel/smp.c:118
| native_smp_send_reschedule+0x21/0x36()
| Modules linked in:
| Pid: 27483, comm: cc1 Not tainted 2.6.26-rc7 #1
| [...]
|  [<c0110355>] native_smp_send_reschedule+0x21/0x36
|  [<c014fe8f>] force_quiescent_state+0x47/0x57
|  [<c014fef0>] call_rcu+0x51/0x6d
|  [<c01713b3>] __fput+0x130/0x158
|  [<c0171231>] fput+0x17/0x19
|  [<c016fd99>] filp_close+0x4d/0x57
|  [<c016fdff>] sys_close+0x5c/0x97

IMHO the warning is a spurious one.

cpu_online_map is updated by the _cpu_down() using stop_machine_run().
Since force_quiescent_state is invoked from irqs disabled section,
stop_machine_run() won't be executing while a cpu is executing
force_quiescent_state(). Hence the cpu_online_map is stable while we're
in the irq disabled section.

However, a cpu might have been offlined _just_ before we disabled irqs
while entering force_quiescent_state(). And rcu subsystem might not yet
have handled the CPU_DEAD notification, leading to the offlined cpu's
bit being set in the rcp->cpumask.

Hence cpumask = (rcp->cpumask & cpu_online_map) to prevent sending
smp_reschedule() to an offlined CPU.

Here's the timeline:

CPU_A  CPU_B
--------------------------------------------------------------
cpu_down(): .
.     .
. .
stop_machine(): /* disables preemption, .
 * and irqs */ .
. .
. .
take_cpu_down(); .
. .
. .
. .
cpu_disable(); /*this removes cpu  .
*from cpu_online_map  .
*/ .
. .
. .
restart_machine(); /* enables irqs */ .
------WINDOW DURING WHICH rcp->cpumask is stale ---------------
. call_rcu();
. /* disables irqs here */
. .force_quiescent_state();
.CPU_DEAD: .for_each_cpu(rcp->cpumask)
. .   smp_send_reschedule();
. .
. .   WARN_ON() for offlined CPU!
.
.
.
rcu_cpu_notify:
.
-------- WINDOW ENDS ------------------------------------------
rcu_offline_cpu() /* Which calls cpu_quiet()
   * which removes
   * cpu from rcp->cpumask.
   */

If a new batch was started just before calling stop_machine_run(), the
"tobe-offlined" cpu is still present in rcp-cpumask.

During a cpu-offline, from take_cpu_down(), we queue an rt-prio idle
task as the next task to be picked by the scheduler. We also call
cpu_disable() which will disable any further interrupts and remove the
cpu's bit from the cpu_online_map.

Once the stop_machine_run() successfully calls take_cpu_down(), it calls
schedule(). That's the last time a schedule is called on the offlined
cpu, and hence the last time when rdp->passed_quiesc will be set to 1
through rcu_qsctr_inc().

But the cpu_quiet() will be on this cpu will be called only when the
next RCU_SOFTIRQ occurs on this CPU. So at this time, the offlined CPU
is still set in rcp->cpumask.

Now coming back to the idle_task which truely offlines the CPU, it does
check for a pending RCU and raises the softirq, since it will find
rdp->passed_quiesc to be 0 in this case. However, since the cpu is
offline I am not sure if the softirq will trigger on the CPU.

Even if it doesn't the rcu_offline_cpu() will find that rcp->completed
is not the same as rcp->cur, which means that our cpu could be holding
up the grace period progression. Hence we call cpu_quiet() and move
ahead.

But because of the window explained in the timeline, we could still have
a call_rcu() before the RCU subsystem executes it's CPU_DEAD
notification, and we send smp_send_reschedule() to offlined cpu while
trying to force the quiescent states. The appended patch adds comments
and prevents checking for offlined cpu everytime.

cpu_online_map is updated by the _cpu_down() using stop_machine_run().
Since force_quiescent_state is invoked from irqs disabled section,
stop_machine_run() won't be executing while a cpu is executing
force_quiescent_state(). Hence the cpu_online_map is stable while we're
in the irq disabled section.

Reported-by: Dhaval Giani <dhaval@linux.vnet.ibm.com>
Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Acked-by: Dhaval Giani <dhaval@linux.vnet.ibm.com>
Cc: Dipankar Sarma <dipankar@in.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Rusty Russel <rusty@rustcorp.com.au>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agoProperly notify block layer of sync writes
Jens Axboe [Tue, 1 Jul 2008 07:07:34 +0000 (09:07 +0200)] 
Properly notify block layer of sync writes

fsync_buffers_list() and sync_dirty_buffer() both issue async writes and
then immediately wait on them. Conceptually, that makes them sync writes
and we should treat them as such so that the IO schedulers can handle
them appropriately.

This patch fixes a write starvation issue that Lin Ming reported, where
xx is stuck for more than 2 minutes because of a large number of
synchronous IO in the system:

INFO: task kjournald:20558 blocked for more than 120 seconds.
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this
message.
kjournald     D ffff810010820978  6712 20558      2
ffff81022ddb1d10 0000000000000046 ffff81022e7baa10 ffffffff803ba6f2
ffff81022ecd0000 ffff8101e6dc9160 ffff81022ecd0348 000000008048b6cb
0000000000000086 ffff81022c4e8d30 0000000000000000 ffffffff80247537
Call Trace:
[<ffffffff803ba6f2>] kobject_get+0x12/0x17
[<ffffffff80247537>] getnstimeofday+0x2f/0x83
[<ffffffff8029c1ac>] sync_buffer+0x0/0x3f
[<ffffffff8066d195>] io_schedule+0x5d/0x9f
[<ffffffff8029c1e7>] sync_buffer+0x3b/0x3f
[<ffffffff8066d3f0>] __wait_on_bit+0x40/0x6f
[<ffffffff8029c1ac>] sync_buffer+0x0/0x3f
[<ffffffff8066d48b>] out_of_line_wait_on_bit+0x6c/0x78
[<ffffffff80243909>] wake_bit_function+0x0/0x23
[<ffffffff8029e3ad>] sync_dirty_buffer+0x98/0xcb
[<ffffffff8030056b>] journal_commit_transaction+0x97d/0xcb6
[<ffffffff8023a676>] lock_timer_base+0x26/0x4b
[<ffffffff8030300a>] kjournald+0xc1/0x1fb
[<ffffffff802438db>] autoremove_wake_function+0x0/0x2e
[<ffffffff80302f49>] kjournald+0x0/0x1fb
[<ffffffff802437bb>] kthread+0x47/0x74
[<ffffffff8022de51>] schedule_tail+0x28/0x5d
[<ffffffff8020cac8>] child_rip+0xa/0x12
[<ffffffff80243774>] kthread+0x0/0x74
[<ffffffff8020cabe>] child_rip+0x0/0x12

Lin Ming confirms that this patch fixes the issue. I've run tests with
it for the past week and no ill effects have been observed, so I'm
proposing it for inclusion into 2.6.26.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>