safe/jmp/linux-2.6
18 years agoMerge git://git.infradead.org/jffs2-xattr-2.6
David Woodhouse [Sat, 20 May 2006 16:27:32 +0000 (17:27 +0100)]
Merge git://git.infradead.org/jffs2-xattr-2.6

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[JFFS2] Reduce calls to ref_totlen() in jffs2_mark_node_obsolete()
David Woodhouse [Sat, 20 May 2006 15:20:19 +0000 (16:20 +0100)]
[JFFS2] Reduce calls to ref_totlen() in jffs2_mark_node_obsolete()

We were calling ref_totlen() 18 times. Even before that becomes a real
function rather than just a dereference, apparently some compilers still
suck anyway. It'll _certainly_ suck after ref_totlen() becomes more
complicated, so calculate it once and don't rely on CSE.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[JFFS2] Optimise reading of eraseblock summary nodes
David Woodhouse [Sat, 20 May 2006 15:13:34 +0000 (16:13 +0100)]
[JFFS2] Optimise reading of eraseblock summary nodes

This improves the time to mount 512MiB of NAND flash on my OLPC prototype
by about 4%. We used to read the last page of the eraseblock twice -- once
to find the offset of the summary node, and again to actually _read_ the
summary node. Now we read the last page only once, and read more only if
we need to.

We also don't allocate a new buffer just for the summary code -- we use
the buffer which was already allocated for the scan. Better still, if the
'buffer' for the scan is actually just a pointer directly into NOR flash,
we use that too, avoiding the memcpy() which we used to do.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD] Avoid 64-bit division in mtdconcat
Andrew Morton [Sat, 20 May 2006 09:17:21 +0000 (10:17 +0100)]
[MTD] Avoid 64-bit division in mtdconcat

WARNING: "__moddi3" [drivers/mtd/mtdconcat.ko] undefined!

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD] Use __symbol_get() instead of symbol_get() in NOR chip probe
David Woodhouse [Sat, 20 May 2006 01:41:34 +0000 (02:41 +0100)]
[MTD] Use __symbol_get() instead of symbol_get() in NOR chip probe

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years agogit-mtd: symbol_get() fix
Andrew Morton [Sat, 20 May 2006 01:06:34 +0000 (18:06 -0700)]
git-mtd: symbol_get() fix

drivers/mtd/devices/docprobe.c: In function `DoC_Probe':
drivers/mtd/devices/docprobe.c:338: warning: assignment from incompatible pointer type
drivers/mtd/devices/docprobe.c:341: warning: assignment from incompatible pointer type

Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[JFFS2] Remove forgotten summary code
Ferenc Havasi [Fri, 19 May 2006 20:00:36 +0000 (21:00 +0100)]
[JFFS2] Remove forgotten summary code

Remove forgotten lines from jffs2_scan_eraseblock() which
were unnecessary and may cause problem in some environments.

Thanks to Alexander Belyakov <alexander.belyakov@intel.com>.

Signed-off-by: Ferenc Havasi <havasi@inf.u-szeged.hu>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[JFFS2] Support new device nodes
David Woodhouse [Thu, 18 May 2006 23:28:49 +0000 (00:28 +0100)]
[JFFS2] Support new device nodes

Device node major/minor numbers are just stored in the payload of a single
data node. Just extend that to 4 bytes and use new_encode_dev() for it.

We only use the 4-byte format if we _need_ to, if !old_valid_dev(foo).
This preserves backwards compatibility with older code as much as
possible. If we do make devices with major or minor numbers above 255, and
then mount the file system with the old code, it'll just read the first
two bytes and get the numbers wrong. If it comes to garbage-collect it,
it'll then write back those wrong numbers. But that's about the best we
can expect.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years agoMerge git://git.infradead.org/mtd-2.6
KaiGai Kohei [Thu, 18 May 2006 15:43:53 +0000 (00:43 +0900)]
Merge git://git.infradead.org/mtd-2.6

18 years ago[MTD] Fix printk format error in gen_probe.c
David Woodhouse [Wed, 17 May 2006 21:03:10 +0000 (22:03 +0100)]
[MTD] Fix printk format error in gen_probe.c

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD] Fix mtdconcat build. We didn't introduce mtd->writesize yet.
David Woodhouse [Wed, 17 May 2006 20:53:58 +0000 (21:53 +0100)]
[MTD] Fix mtdconcat build. We didn't introduce mtd->writesize yet.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years agoMTD: mtdconcat NAND/Sibley support (rev.2)
Alexander Belyakov [Wed, 17 May 2006 15:11:16 +0000 (19:11 +0400)]
MTD: mtdconcat NAND/Sibley support (rev.2)

There is a second revision of "mtdconcat NAND/Sibley" patch. I hope
the patch will not get damaged as I'm posting it from gmail account,
thanks to Jorn.

The patch adds previously missing concat_writev(),
concat_writev_ecc(), concat_block_isbad(), concat_block_markbad()
functions to make concatenation layer compatible with Sibley and NAND
chips.

Patch has been cleared from whitespaces, fixed some lines of code as
requested. Also I have added code for alignment check that should
support Jorn's "writesize" patch.

Signed-off-by: Alexander Belyakov <alexander.belyakov@intel.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years agoLASAT depends on MTD_CFI
Martin Michlmayr [Mon, 20 Mar 2006 04:40:46 +0000 (04:40 +0000)]
LASAT depends on MTD_CFI

The following difference was found between the mainline and linux-mips
kernel.  LASAT depends on MTD_CFI.

Signed-off-by: Martin Michlmayr <tbm@cyrius.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years agoRe-add module description for ms02-nv to Kconfig
Martin Michlmayr [Mon, 20 Mar 2006 04:40:26 +0000 (04:40 +0000)]
Re-add module description for ms02-nv to Kconfig

In an unrelated MTD commit, a description about the ms02-nv module
got removed from Kconfig.  While I personally agree with this
removal, the module maintainer (Maciej W. Rozycki) would like to
see it added back.  In the absense of any consistency regarding
Kconfig descriptions his wish should be followed.

Signed-off-by: Martin Michlmayr <tbm@cyrius.com>
Acked-by: Maciej W. Rozycki <macro@linux-mips.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD] RFD FTL: Be noisier, and don't assume block without RFD magic are erased
Sean Young [Wed, 17 May 2006 11:45:34 +0000 (12:45 +0100)]
[MTD] RFD FTL: Be noisier, and don't assume block without RFD magic are erased

Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD] generic: propagate oobavail to MTD partitions
Vitaly Wool [Wed, 17 May 2006 11:34:27 +0000 (12:34 +0100)]
[MTD] generic: propagate oobavail to MTD partitions

'oobavail' parameter of mtd_info structure is now propagated to the MTD
partitions

Signed-off-by: Vitaly Wool <vwool@ru.mvista.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[PCMCIA MTD] Fix leak and crash on reboot
Sean Young [Wed, 17 May 2006 10:47:48 +0000 (11:47 +0100)]
[PCMCIA MTD] Fix leak and crash on reboot

Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years agocfi_cmdset_0001: factorize code to wait for flash status
Nicolas Pitre [Thu, 30 Mar 2006 14:52:41 +0000 (15:52 +0100)]
cfi_cmdset_0001: factorize code to wait for flash status

This allows for much better abstraction and separation of the XIP and
non-XIP cases with their own specific implementations.  This fixes the
case where a timeout was tripped on in the XIP case by the code that
was meant for the non-XIP case only.

This also makes for a nice code reduction.

Signed-off-by: Nicolas Pitre <nico@cam.org>
CC: "Alexey, Korolev" <alexey.korolev@intel.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years agocfi-cmdset-0001: always update the chip status
Nicolas Pitre [Wed, 29 Mar 2006 22:31:42 +0000 (23:31 +0100)]
cfi-cmdset-0001: always update the chip status

... otherwise xip_enable() won't do the right thing.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years agoCHIPS: Fix potential starvation in cfi_cmdset_0001
Josh Boyer [Tue, 6 Dec 2005 17:28:19 +0000 (17:28 +0000)]
CHIPS: Fix potential starvation in cfi_cmdset_0001

The patch below fixes a potential starvation issue that can arise when
there is contention on a chip during a period when a process is
currently writing to it.  The starvation is avoided by conditionally
rescheduling when the chip is left in a state usable by other processes.

Signed-off-by: Josh Boyer <jdub@us.ibm.com>
Signed-off-by: Tom Gall <tom_gall@vnet.ibm.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[JFFS2] Repack some on-medium structures. ARM is weirder than I thought.
David Woodhouse [Tue, 16 May 2006 22:03:08 +0000 (23:03 +0100)]
[JFFS2] Repack some on-medium structures. ARM is weirder than I thought.

We have to pack at least the jint16_t structure, because otherwise it'll
be four bytes in size. Thankfully, we can do that and _not_ pack the
actual node structures, and the compiler still doesn't emit stupid code.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years agoNAND: Fix NAND ECC errors on AMD Au1550
Sergei Shtylyov [Tue, 16 May 2006 16:52:06 +0000 (20:52 +0400)]
NAND: Fix NAND ECC errors on AMD Au1550

    On AMD Au1550 the static bus controller fails to keep -CE asserted during
chip ready delay on read commands and the NAND chip being used requires this.
So, the current driver allows nand_base.c to drive -CE manually during the
entire sector read. When the PCMCIA driver is enabled however, occasionally
the ECC errors occur on NAND reads. This happens because the PCMCIA driver
polls sockets periodically and reads one of the board's control/status regs
(BCSRs) which are on the same static bus as the NAND flash, and just use
another chip select (and the NOR flash also resides on that bus), so as the
NAND driver forces NAND chip select asserted and the -RE signal is shared, a
contention occurs on the static bus when BCSR or NOR flash is read while we're
reading from NAND.
    So, we either can't keep interrupts enabled during the whole NAND sector
read (which is hardly acceptable), or have to implement some interlocking
scheme between multiple drivers (which is painful, and makes me shudder :-).
    There's a third way which has proven to work: to force -CE asserted only
while we're waiting for a NAND chip to become ready after a read command,
disabling interrupts for a maximum of 25 microseconds (according to Toshiba
TC58DVM92A1FT00 datasheet -- this chip is mentioned in the board schematics);
for Samsung NAND chip which seems to be actually used this delay is even less,
12 us.

Signed-off-by: Konstantin Baydarov <kbaidarov@ru.mvista.com>
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD] Fix build warnings in RedBoot MTD partition parser.
Ben Dooks [Tue, 16 May 2006 16:31:15 +0000 (17:31 +0100)]
[MTD] Fix build warnings in RedBoot MTD partition parser.

Fix build warnings from drivers/mtd/redboot.c due to
use of `unsigned long` in `struct fis_image_desc` for
fields being passed to swab32s() which expects __u32 *

Change the entries to uint32_t to make them compatible
with the swab32s() function

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years agoNAND: AMD Au1550 driver reads write-only register
Sergei Shtylyov [Tue, 16 May 2006 16:16:41 +0000 (20:16 +0400)]
NAND: AMD Au1550 driver reads write-only register

     During the last cleanup of the AMD Au1550 NAND driver the old buglet was
reintroduced: as the MEM_STNDCTL register is write-only and seem to always
read as 0x31, read-modify-write to it done in au1xxx_nand_init() will have the
side effect of enabling -RCS0/1 pin override (via bits 4/5 of this reg.), thus
possibly causing a contention on the static bus when the NOR flash (using
-RCS0) or board control status registers (using -RCS2) are read. Luckily, this
goes away with a first NAND access, since au1550_hwcontrol() doesn't try to
read this register before writing anymore.

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[JFFS2] Fix printk format in jffs2_sum_write_data() error message.
David Woodhouse [Tue, 16 May 2006 16:05:33 +0000 (17:05 +0100)]
[JFFS2] Fix printk format in jffs2_sum_write_data() error message.

fs/jffs2/summary.c: In function ‘jffs2_sum_write_data’:
fs/jffs2/summary.c:658: warning: format ‘%zd’ expects type ‘signed size_t’, but argument 4 has type ‘uint32_t’

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[JFFS2] Fix section mismatch warnings in JFFS2.
David Brownell [Tue, 16 May 2006 15:08:10 +0000 (16:08 +0100)]
[JFFS2] Fix section mismatch warnings in JFFS2.

Mark certain functions with __init and __exit appropriately.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD NAND] Make various initfuncs static, remove #ifdef MODULE from exitfuncs
David Woodhouse [Tue, 16 May 2006 12:54:50 +0000 (13:54 +0100)]
[MTD NAND] Make various initfuncs static, remove #ifdef MODULE from exitfuncs

We all inherited the same error from the original NAND board driver which
got copied and changed. Fix them all at once...

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years agodrivers/mtd/devices/docprobe.c: correct #if's
Adrian Bunk [Tue, 16 May 2006 11:46:44 +0000 (13:46 +0200)]
drivers/mtd/devices/docprobe.c: correct #if's

On Mon, May 15, 2006 at 12:56:37AM -0700, Andrew Morton wrote:
>...
> Changes since 2.6.17-rc3-mm1:
>...
>  git-mtd.patch
>...
>  git trees
>...

If we correct the names of the config options, the code might actually
work as intended...

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD] Add help text for MTD_NAND_CS553X option.
David Woodhouse [Tue, 16 May 2006 12:11:47 +0000 (13:11 +0100)]
[MTD] Add help text for MTD_NAND_CS553X option.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6
David Woodhouse [Tue, 16 May 2006 00:19:52 +0000 (01:19 +0100)]
Merge git://git./linux/kernel/git/torvalds/linux-2.6

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[PATCH] jffs2: memory leak in jffs2_scan_medium()
Florin Malita [Mon, 15 May 2006 22:42:31 +0000 (23:42 +0100)]
[PATCH] jffs2: memory leak in jffs2_scan_medium()

If jffs2_scan_eraseblock() fails and the exit path is taken, 's' is not
being deallocated.

Reported by Coverity, CID: 1258.

Signed-off-by: Florin Malita <fmalita@gmail.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[PATCH] dl2k needs dma-mapping.h
Andrew Morton [Mon, 15 May 2006 16:44:43 +0000 (09:44 -0700)]
[PATCH] dl2k needs dma-mapping.h

On alpha:

drivers/net/dl2k.c: In function `rio_free_tx':
drivers/net/dl2k.c:768: error: `DMA_48BIT_MASK' undeclared (first use in this function)
drivers/net/dl2k.c:768: error: (Each undeclared identifier is reported only once
drivers/net/dl2k.c:768: error: for each function it appears in.)
drivers/net/dl2k.c: In function `receive_packet':
drivers/net/dl2k.c:896: error: `DMA_48BIT_MASK' undeclared (first use in this function)
drivers/net/dl2k.c: In function `rio_close':
drivers/net/dl2k.c:1803: error: `DMA_48BIT_MASK' undeclared (first use in this function)

Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] jffs2 warning fixes
Andrew Morton [Mon, 15 May 2006 16:44:42 +0000 (09:44 -0700)]
[PATCH] jffs2 warning fixes

fs/jffs2/nodelist.c: In function `check_node_data':
fs/jffs2/nodelist.c:441: warning: unsigned int format, different type arg (arg 4)
fs/jffs2/nodelist.c:464: warning: int format, different type arg (arg 5)

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] devices.txt: remove pktcdvd entry
Peter Osterlund [Mon, 15 May 2006 16:44:40 +0000 (09:44 -0700)]
[PATCH] devices.txt: remove pktcdvd entry

Changing the driver to use dynamic device numbers was one of the many
changes that were made in order to have the driver accepted into the
mainline kernel.  Therefore I would say that the entry in devices.txt is
obsolete.  This patch removes it.

Signed-off-by: Peter Osterlund <petero2@telia.com>
Cc: Torben Mathiasen <device@lanana.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] ide_cs: Add IBM microdrive to known IDs
Thomas Kleffel [Mon, 15 May 2006 16:44:37 +0000 (09:44 -0700)]
[PATCH] ide_cs: Add IBM microdrive to known IDs

Add the IBM microdrive to the known PCMCIA IDs for ide_cs.

Signed-off-by: Thomas Kleffel <tk@maintech.de>
Cc: Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] Fix typos in Documentation/memory-barriers.txt
Aneesh Kumar [Mon, 15 May 2006 16:44:36 +0000 (09:44 -0700)]
[PATCH] Fix typos in Documentation/memory-barriers.txt

Fix some typos in Documentation/memory-barriers.txt

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@gmail.com>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] gigaset: endian fix
Alexey Dobriyan [Mon, 15 May 2006 16:44:35 +0000 (09:44 -0700)]
[PATCH] gigaset: endian fix

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Hansjoerg Lipp <hjlipp@web.de>
Cc: Tilman Schmidt <tilman@imap.cc>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] fix hotplug kconfig help
Pavel Machek [Mon, 15 May 2006 16:44:34 +0000 (09:44 -0700)]
[PATCH] fix hotplug kconfig help

HOTPLUG_CPU entry says "Say Y..." then "Say N.".  Slightly ugly, so I fixed
it up, and added remark about suspend on SMP as a bonus.

Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] smbus unhiding kills thermal management
Carl-Daniel Hailfinger [Mon, 15 May 2006 16:44:33 +0000 (09:44 -0700)]
[PATCH] smbus unhiding kills thermal management

Do not enable the SMBus device on Asus boards if suspend is used.  We do
not reenable the device on resume, leading to all sorts of undesirable
effects, the worst being a total fan failure after resume on Samsung P35
laptop.

Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
Signed-off-by: Pavel Machek <pavel@suse.cz>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] revert "vfs: propagate mnt_flags into do_loopback/vfsmount"
Andrew Morton [Mon, 15 May 2006 16:44:30 +0000 (09:44 -0700)]
[PATCH] revert "vfs: propagate mnt_flags into do_loopback/vfsmount"

Revert commit f6422f17d3a480f21917a3895e2a46b968f56a08, due to

Valdis.Kletnieks@vt.edu wrote:
>
> There seems to have been a bug introduced in this changeset:
>
> Am running 2.6.17-rc3-mm1.  When this changeset is applied, 'mount --bind'
> misbehaves:
>
> > # mkdir /foo
> > # mount -t tmpfs -o rw,nosuid,nodev,noexec,noatime,nodiratime none /foo
> > # mkdir /foo/bar
> > # mount --bind /foo/bar /foo
> > # tail -2 /proc/mounts
> > none /foo tmpfs rw,nosuid,nodev,noexec,noatime,nodiratime 0 0
> > none /foo tmpfs rw 0 0
>
> Reverting this changeset causes both mounts to have the same options.
>
> (Thanks to Stephen Smalley for tracking down the changeset...)
>

Cc: Herbert Poetzl <herbert@13thfloor.at>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: <Valdis.Kletnieks@vt.edu>
Cc: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] root mount failure: emit filesystems attempted
Andy Whitcroft [Mon, 15 May 2006 16:44:29 +0000 (09:44 -0700)]
[PATCH] root mount failure: emit filesystems attempted

When we fail to mount from a valid root device list out the filesystems we
have tried to mount it with.  This gives the user vital diagnostics as to
what is missing from their kernel.

For example in the fragment below the kernel does not have CRAMFS compiled
into the kernel and yet appears to recognise it at the RAMDISK detect
stage.  Later the mount fails as we don't have the filesystem.

  RAMDISK: cramfs filesystem found at block 0
  RAMDISK: Loading 1604KiB [1 disk] into ram disk... done.
  XFS: bad magic number
  XFS: SB validate failed
  No filesystem could mount root, tried: reiserfs ext3 ext2 msdos vfat
    iso9660 jfs xfs
  Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(8,1)

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] fs/compat.c: fix 'if (a |= b )' typo
Alexey Dobriyan [Mon, 15 May 2006 16:44:27 +0000 (09:44 -0700)]
[PATCH] fs/compat.c: fix 'if (a |= b )' typo

Mentioned by Mark Armbrust somewhere on Usenet.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Ulrich Drepper <drepper@redhat.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] tpm_register_hardware gcc 4.1 warning fix
Daniel Walker [Mon, 15 May 2006 16:44:27 +0000 (09:44 -0700)]
[PATCH] tpm_register_hardware gcc 4.1 warning fix

drivers/char/tpm/tpm.c: In function 'tpm_register_hardware':
drivers/char/tpm/tpm.c:1157: warning: assignment from incompatible pointer type

Signed-off-by: Daniel Walker <dwalker@mvista.com>
Acked-by: Kylene Hall <kjhall@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] Final rio polish
Alan Cox [Mon, 15 May 2006 16:44:26 +0000 (09:44 -0700)]
[PATCH] Final rio polish

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] tpm: fix constant
Kylene Jo Hall [Mon, 15 May 2006 16:44:25 +0000 (09:44 -0700)]
[PATCH] tpm: fix constant

Fix the constant used for the base address when it cannot be determined
from ACPI.  It was off by one order of magnitude.

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>
18 years ago[PATCH] Add Core Solo and Core Duo support to oprofile
Benjamin LaHaise [Mon, 15 May 2006 16:44:24 +0000 (09:44 -0700)]
[PATCH] Add Core Solo and Core Duo support to oprofile

Add support to oprofile for the Intel Core Solo and Core Duo processors.
See also the patch to add support to oprofile-0.9.1-8.1.1 at
http://www.kvack.org/~bcrl/patches/oprofile/oprofile-core-0.9.1.diff .

Signed-off-by: Benjamin LaHaise <benjamin.c.lahaise@intel.com>
Cc: Philippe Elie <phil.el@wanadoo.fr>
Cc: John Levon <levon@movementarian.org>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] fix can_share_swap_page() when !CONFIG_SWAP
Hua Zhong [Mon, 15 May 2006 16:44:22 +0000 (09:44 -0700)]
[PATCH] fix can_share_swap_page() when !CONFIG_SWAP

can_share_swap_page() is used to check if the page has the last reference.
This avoids allocating a new page for COW if it's the last page.

However, if CONFIG_SWAP is not set, can_share_swap_page() is defined as 0,
thus always causes a copy for the last COW page.  The below simple patch
fixes it.

Signed-off-by: Hua Zhong <hzhong@gmail.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] v9fs: signal handling fixes
Latchesar Ionkov [Mon, 15 May 2006 16:44:21 +0000 (09:44 -0700)]
[PATCH] v9fs: signal handling fixes

Multiple races can happen when v9fs is interrupted by a signal and Tflush
message is sent to the server.  After v9fs sends Tflush it doesn't wait
until it receives Rflush, and possibly the response of the original
message.  This behavior may confuse v9fs what fids are allocated by the
file server.

This patch fixes the races and the fid allocation.

Signed-off-by: Latchesar Ionkov <lucho@ionkov.net>
Cc: Eric Van Hensbergen <ericvh@hera.kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] v9fs: Twalk memory leak
Latchesar Ionkov [Mon, 15 May 2006 16:44:18 +0000 (09:44 -0700)]
[PATCH] v9fs: Twalk memory leak

v9fs leaks memory if the file server responds with Rerror to a Twalk
message.  The patch fixes the leak.

Signed-off-by: Latchesar Ionkov <lucho@ionkov.net>
Cc: Eric Van Hensbergen <ericvh@hera.kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] LED: Fix sysfs store function error handling
Richard Purdie [Mon, 15 May 2006 16:44:17 +0000 (09:44 -0700)]
[PATCH] LED: Fix sysfs store function error handling

Fix the error handling of some LED _store functions.  This corrects them to
return -EINVAL if the value is not numeric with an optional byte of trailing
whitespace.

Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] LED: Add maintainer entry for the LED subsystem
Richard Purdie [Mon, 15 May 2006 16:44:16 +0000 (09:44 -0700)]
[PATCH] LED: Add maintainer entry for the LED subsystem

Add a MAINTAINERS entry for the LED subsystem.

Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] Backlight/LCD Class: Fix sysfs _store error handling
Richard Purdie [Mon, 15 May 2006 16:44:15 +0000 (09:44 -0700)]
[PATCH] Backlight/LCD Class: Fix sysfs _store error handling

The backlight and LCD class _store functions currently accept values like "34
some random strings" without error.  This corrects them to return -EINVAL if
the value is not numeric with an optional byte of trailing whitespace.

Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] LED: Improve Kconfig information
Richard Purdie [Mon, 15 May 2006 16:44:14 +0000 (09:44 -0700)]
[PATCH] LED: Improve Kconfig information

Improve the NEW_LEDS Kconfig information to say what it does as well as what
it doesn't.

Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] add slab_is_available() routine for boot code
Mike Kravetz [Mon, 15 May 2006 16:44:13 +0000 (09:44 -0700)]
[PATCH] add slab_is_available() routine for boot code

slab_is_available() indicates slab based allocators are available for use.
SPARSEMEM code needs to know this as it can be called at various times
during the boot process.

Signed-off-by: Mike Kravetz <kravetz@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] smbfs: Fix slab corruption in samba error path
Jan Niehusmann [Mon, 15 May 2006 16:44:12 +0000 (09:44 -0700)]
[PATCH] smbfs: Fix slab corruption in samba error path

Yesterday, I got the following error with 2.6.16.13 during a file copy from
a smb filesystem over a wireless link.  I guess there was some error on the
wireless link, which in turn caused an error condition for the smb
filesystem.

In the log, smb_file_read reports error=4294966784 (0xfffffe00), which also
shows up in the slab dumps, and also is -ERESTARTSYS.  Error code 27499
corresponds to 0x6b6b, so the rq_errno field seems to be the only one being
set after freeing the slab.

In smb_add_request (which is the only place in smbfs where I found
ERESTARTSYS), I found the following:

        if (!timeleft || signal_pending(current)) {
                /*
                 * On timeout or on interrupt we want to try and remove the
                 * request from the recvq/xmitq.
                 */
                smb_lock_server(server);
                if (!(req->rq_flags & SMB_REQ_RECEIVED)) {
                        list_del_init(&req->rq_queue);
                        smb_rput(req);
                }
                smb_unlock_server(server);
        }
[...]
        if (signal_pending(current))
                req->rq_errno = -ERESTARTSYS;

I guess that some codepath like smbiod_flush() caused the request to be
removed from the queue, and smb_rput(req) be called, without
SMB_REQ_RECEIVED being set.  This violates an asumption made by the quoted
code.

Then, the above code calls smb_rput(req) again, the req gets freed, and
req->rq_errno = -ERESTARTSYS writes into the already freed slab.  As
list_del_init doesn't cause an error if called multiple times, that does
cause the observed behaviour (freed slab with rq_errno=-ERESTARTSYS).

If this observation is correct, the following patch should fix it.

I wonder why the smb code uses list_del_init everywhere - using list_del
instead would catch such situations by poisoning the next and prev
pointers.

May  4 23:29:21 knautsch kernel: [17180085.456000] ipw2200: Firmware error detected.  Restarting.
May  4 23:29:21 knautsch kernel: [17180085.456000] ipw2200: Sysfs 'error' log captured.
May  4 23:33:02 knautsch kernel: [17180306.316000] ipw2200: Firmware error detected.  Restarting.
May  4 23:33:02 knautsch kernel: [17180306.316000] ipw2200: Sysfs 'error' log already exists.
May  4 23:33:02 knautsch kernel: [17180306.968000] smb_file_read: //some_file validation failed, error=4294966784
May  4 23:34:18 knautsch kernel: [17180383.256000] smb_file_read: //some_file validation failed, error=4294966784
May  4 23:34:18 knautsch kernel: [17180383.284000] SMB connection re-established (-5)
May  4 23:37:19 knautsch kernel: [17180563.956000] smb_file_read: //some_file validation failed, error=4294966784
May  4 23:40:09 knautsch kernel: [17180733.636000] smb_file_read: //some_file validation failed, error=4294966784
May  4 23:40:26 knautsch kernel: [17180750.700000] smb_file_read: //some_file validation failed, error=4294966784
May  4 23:43:02 knautsch kernel: [17180907.304000] smb_file_read: //some_file validation failed, error=4294966784
May  4 23:43:08 knautsch kernel: [17180912.324000] smb_file_read: //some_file validation failed, error=4294966784
May  4 23:43:34 knautsch kernel: [17180938.416000] smb_errno: class Unknown, code 27499 from command 0x6b
May  4 23:43:34 knautsch kernel: [17180938.416000] Slab corruption: start=c4ebe09c, len=244
May  4 23:43:34 knautsch kernel: [17180938.416000] Redzone: 0x5a2cf071/0x5a2cf071.
May  4 23:43:34 knautsch kernel: [17180938.416000] Last user: [<e087b903>](smb_rput+0x53/0x90 [smbfs])
May  4 23:43:34 knautsch kernel: [17180938.416000] 000: 6b 6b 6b 6b 6b 6b 6b 6b 6a 6b 6b 6b 6b 6b 6b 6b
May  4 23:43:34 knautsch kernel: [17180938.416000] 0f0: 00 fe ff ff
May  4 23:43:34 knautsch kernel: [17180938.416000] Next obj: start=c4ebe19c, len=244
May  4 23:43:34 knautsch kernel: [17180938.416000] Redzone: 0x5a2cf071/0x5a2cf071.
May  4 23:43:34 knautsch kernel: [17180938.416000] Last user: [<00000000>](_stext+0x3feffde0/0x30)
May  4 23:43:34 knautsch kernel: [17180938.416000] 000: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
May  4 23:43:34 knautsch kernel: [17180938.416000] 010: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
May  4 23:43:34 knautsch kernel: [17180938.460000] SMB connection re-established (-5)
May  4 23:43:42 knautsch kernel: [17180946.292000] ipw2200: Firmware error detected.  Restarting.
May  4 23:43:42 knautsch kernel: [17180946.292000] ipw2200: Sysfs 'error' log already exists.
May  4 23:45:04 knautsch kernel: [17181028.752000] ipw2200: Firmware error detected.  Restarting.
May  4 23:45:04 knautsch kernel: [17181028.752000] ipw2200: Sysfs 'error' log already exists.
May  4 23:45:05 knautsch kernel: [17181029.868000] smb_file_read: //some_file validation failed, error=4294966784
May  4 23:45:36 knautsch kernel: [17181060.984000] smb_errno: class Unknown, code 27499 from command 0x6b
May  4 23:45:36 knautsch kernel: [17181060.984000] Slab corruption: start=c4ebe09c, len=244
May  4 23:45:36 knautsch kernel: [17181060.984000] Redzone: 0x5a2cf071/0x5a2cf071.
May  4 23:45:36 knautsch kernel: [17181060.984000] Last user: [<e087b903>](smb_rput+0x53/0x90 [smbfs])
May  4 23:45:36 knautsch kernel: [17181060.984000] 000: 6b 6b 6b 6b 6b 6b 6b 6b 6a 6b 6b 6b 6b 6b 6b 6b
May  4 23:45:36 knautsch kernel: [17181060.984000] 0f0: 00 fe ff ff
May  4 23:45:36 knautsch kernel: [17181060.984000] Next obj: start=c4ebe19c, len=244
May  4 23:45:36 knautsch kernel: [17181060.984000] Redzone: 0x5a2cf071/0x5a2cf071.
May  4 23:45:36 knautsch kernel: [17181060.984000] Last user: [<00000000>](_stext+0x3feffde0/0x30)
May  4 23:45:36 knautsch kernel: [17181060.984000] 000: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
May  4 23:45:36 knautsch kernel: [17181060.984000] 010: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
May  4 23:45:36 knautsch kernel: [17181061.024000] SMB connection re-established (-5)
May  4 23:46:17 knautsch kernel: [17181102.132000] smb_file_read: //some_file validation failed, error=4294966784
May  4 23:47:46 knautsch kernel: [17181190.468000] smb_errno: class Unknown, code 27499 from command 0x6b
May  4 23:47:46 knautsch kernel: [17181190.468000] Slab corruption: start=c4ebe09c, len=244
May  4 23:47:46 knautsch kernel: [17181190.468000] Redzone: 0x5a2cf071/0x5a2cf071.
May  4 23:47:46 knautsch kernel: [17181190.468000] Last user: [<e087b903>](smb_rput+0x53/0x90 [smbfs])
May  4 23:47:46 knautsch kernel: [17181190.468000] 000: 6b 6b 6b 6b 6b 6b 6b 6b 6a 6b 6b 6b 6b 6b 6b 6b
May  4 23:47:46 knautsch kernel: [17181190.468000] 0f0: 00 fe ff ff
May  4 23:47:46 knautsch kernel: [17181190.468000] Next obj: start=c4ebe19c, len=244
May  4 23:47:46 knautsch kernel: [17181190.468000] Redzone: 0x5a2cf071/0x5a2cf071.
May  4 23:47:46 knautsch kernel: [17181190.468000] Last user: [<00000000>](_stext+0x3feffde0/0x30)
May  4 23:47:46 knautsch kernel: [17181190.468000] 000: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
May  4 23:47:46 knautsch kernel: [17181190.468000] 010: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
May  4 23:47:46 knautsch kernel: [17181190.492000] SMB connection re-established (-5)
May  4 23:49:20 knautsch kernel: [17181284.828000] smb_file_read: //some_file validation failed, error=4294966784
May  4 23:49:39 knautsch kernel: [17181303.896000] smb_file_read: //some_file validation failed, error=4294966784

Signed-off-by: Jan Niehusmann <jan@gondor.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] Marcelo has moved
Marcelo Tosatti [Mon, 15 May 2006 16:44:08 +0000 (09:44 -0700)]
[PATCH] Marcelo has moved

Signed-off-by: Marcelo Tosatti <marcelo@kvack.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] symbol_put_addr() locks kernel
Trent Piepho [Mon, 15 May 2006 16:44:06 +0000 (09:44 -0700)]
[PATCH] symbol_put_addr() locks kernel

Even since a previous patch:

Fix race between CONFIG_DEBUG_SLABALLOC and modules
Sun, 27 Jun 2004 17:55:19 +0000 (17:55 +0000)
http://www.kernel.org/git/?p=linux/kernel/git/torvalds/old-2.6-bkcvs.git;a=commit;h=92b3db26d31cf21b70e3c1eadc56c179506d8fbe

The function symbol_put_addr() will deadlock the kernel.

symbol_put_addr() would acquire modlist_lock, then while holding the lock call
two functions kernel_text_address() and module_text_address() which also try
to acquire the same lock.  This deadlocks the kernel of course.

This patch changes symbol_put_addr() to not acquire the modlist_lock, it
doesn't need it since it never looks at the module list directly.  Also, it
now uses core_kernel_text() instead of kernel_text_address().  The latter has
an additional check for addr inside a module, but we don't need to do that
since we call module_text_address() (the same function kernel_text_address
uses) ourselves.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Cc: Zwane Mwaikambo <zwane@fsmlabs.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Johannes Stezenbach <js@linuxtv.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] s390: add vmsplice system call
Martin Schwidefsky [Mon, 15 May 2006 16:44:05 +0000 (09:44 -0700)]
[PATCH] s390: add vmsplice system call

Add new vmsplice system call and add missing __NR_xxx defines for
sys_set_robust_list, sys_get_robust_list, sys_splice, sys_sync_file_range
and sys_tee.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] initramfs: fix CPIO hardlink check
Mark Huang [Mon, 15 May 2006 16:44:03 +0000 (09:44 -0700)]
[PATCH] initramfs: fix CPIO hardlink check

Copy the filenames of hardlinks when inserting them into the hash, since
the "name" pointer may point to scratch space (name_buf).  Not doing so
results in corruption if the scratch space is later overwritten: the wrong
file may be hardlinked, or, if the scratch space contains garbage, the link
will fail and a 0-byte file will be created instead.

Signed-off-by: Mark Huang <mlhuang@cs.princeton.edu>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] s390: lcs incorrect test
Greg Smith [Mon, 15 May 2006 16:44:02 +0000 (09:44 -0700)]
[PATCH] s390: lcs incorrect test

While debugging why our LCS emulator is having some problems I noticed the
following weirdness in drivers/s390/net/lcs.c routine lcs_irq.  The `if'
statement is always true since SCHN_STAT_PCI is defined as 0x80.

Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] setup_per_zone_pages_min() overflow fix
Andrew Morton [Mon, 15 May 2006 16:43:59 +0000 (09:43 -0700)]
[PATCH] setup_per_zone_pages_min() overflow fix

As pointed out in http://bugzilla.kernel.org/show_bug.cgi?id=6490, this
function can experience overflows on 32-bit machines, causing our response to
changed values of min_free_kbytes to go whacky.

Fixing it efficiently is all too hard, so fix it with 64-bit math instead.

Cc: Ake Sandgren <ake.sandgren@hpc2n.umu.se>
Cc: Martin Bligh <mbligh@google.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] s390: exploit rcu_needs_cpu() interface
Heiko Carstens [Mon, 15 May 2006 16:43:59 +0000 (09:43 -0700)]
[PATCH] s390: exploit rcu_needs_cpu() interface

Exploit rcu_needs_cpu() interface to keep the cpu 'ticking' if necessary.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] RCU: introduce rcu_needs_cpu() interface
Heiko Carstens [Mon, 15 May 2006 16:43:58 +0000 (09:43 -0700)]
[PATCH] RCU: introduce rcu_needs_cpu() interface

With "Paul E. McKenney" <paulmck@us.ibm.com>

Introduce rcu_needs_cpu() interface.  This can be used to tell if there
will be a new rcu batch on a cpu soon by looking at the curlist pointer.
This can be used to avoid to enter a tickless idle state where the cpu
would miss that a new batch is ready when rcu_start_batch would be called
on a different cpu.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] smbfs chroot issue (CVE-2006-1864)
Olaf Kirch [Mon, 15 May 2006 16:43:57 +0000 (09:43 -0700)]
[PATCH] smbfs chroot issue (CVE-2006-1864)

Mark Moseley reported that a chroot environment on a SMB share can be left
via "cd ..\\".  Similar to CVE-2006-1863 issue with cifs, this fix is for
smbfs.

Steven French <sfrench@us.ibm.com> wrote:

Looks fine to me.  This should catch the slash on lookup or equivalent,
which will be all obvious paths of interest.

Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] VIA quirk fixup, additional PCI IDs
Chris Wedgwood [Mon, 15 May 2006 16:43:55 +0000 (09:43 -0700)]
[PATCH] VIA quirk fixup, additional PCI IDs

An earlier commit (75cf7456dd87335f574dcd53c4ae616a2ad71a11) changed an
overly-zealous PCI quirk to only poke those VIA devices that need it.
However, some PCI devices were not included in what I hope is now the full
list.  Consequently we're failing to run the quirk on all machines which need
it, causing IRQ routing failures.

This should I hope correct this.

Thanks to Masoud Sharbiani <masouds@masoud.ir> for pointing this out
and testing the fix.

Signed-off-by: Chris Wedgwood <cw@f00f.org>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] pcmcia Oopses fixes
Benjamin Herrenschmidt [Mon, 15 May 2006 16:43:53 +0000 (09:43 -0700)]
[PATCH] pcmcia Oopses fixes

Fix some NULL dereferences in the pcmcia code when using old userland
tools.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] tpm: update module dependencies
Kylene Jo Hall [Mon, 15 May 2006 16:43:53 +0000 (09:43 -0700)]
[PATCH] tpm: update module dependencies

The TIS driver is dependent upon information from the ACPI table for device
discovery thus it compiles but does no actual work without this dependency.

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>
18 years ago[PATCH] Fix capi reload by unregistering the correct major
Stefan Schweizer [Mon, 15 May 2006 16:43:52 +0000 (09:43 -0700)]
[PATCH] Fix capi reload by unregistering the correct major

I am having the bug FATAL: Error inserting capi ([..]/capi.ko): Device or
resource busy when I try to reload capi after loading it.  in dmesg:
capi20: unable to get major 68

Fix the issue which is caused by setting the major to zero when registering
the chrdev succeeded.

(akpm: this means that we can again not use `major=0' (dynamic major
allocation) for this driver).

Cc: Karsten Keil <kkeil@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] autofs4: NFY_NONE wait race fix
Ian Kent [Mon, 15 May 2006 16:43:51 +0000 (09:43 -0700)]
[PATCH] autofs4: NFY_NONE wait race fix

This patch fixes two problems.

First, the comparison of entries in the waitq.c was incorrect.

Second, the NFY_NONE check was incorrect. The test of whether the dentry
is mounted if ineffective, for example, if an expire fails then we could
wait forever on a non existant expire. The bug was identified by Jeff
Moyer.

The patch changes autofs4 to wait on expires only as this is all that's
needed.  If there is no existing wait when autofs4_wait is call with a type
of NFY_NONE it delays until either a wait appears or the the expire flag is
cleared.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] fs/open.c: unexport sys_openat
Adrian Bunk [Mon, 15 May 2006 16:43:50 +0000 (09:43 -0700)]
[PATCH] fs/open.c: unexport sys_openat

Remove the unused EXPORT_SYMBOL_GPL(sys_openat).

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] selinux: check for failed kmalloc in security_sid_to_context()
Serge E. Hallyn [Mon, 15 May 2006 16:43:48 +0000 (09:43 -0700)]
[PATCH] selinux: check for failed kmalloc in security_sid_to_context()

Check for NULL kmalloc return value before writing to it.

Signed-off-by: Serge E. Hallyn <serue@us.ibm.com>
Acked-by: James Morris <jmorris@namei.org>
Cc: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[JFFS2] Fix printk format in some error messages.
Andrew Morton [Mon, 15 May 2006 12:45:58 +0000 (13:45 +0100)]
[JFFS2] Fix printk format in some error messages.

fs/jffs2/nodelist.c: In function `check_node_data':
fs/jffs2/nodelist.c:441: warning: unsigned int format, different type arg (arg 4)
fs/jffs2/nodelist.c:464: warning: int format, different type arg (arg 5)

Modified from Andrew's original fix because while his terminal may indeed
only have eighty columns, mine only has _TWENTYFOUR_ lines. So the
cosmetic fluff is perfectly OK out past column 80 where it was -- the
casual reader doesn't _care_ about anything more than the fact that it
goes 'if (foo) JFFS2_WARNING...', and there's no point wasting a whole
line to display the tail end of the printk which nobody actually cares
about.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[JFFS2] Don't pack on-medium structures, because GCC emits crappy code
David Woodhouse [Sun, 14 May 2006 23:49:43 +0000 (00:49 +0100)]
[JFFS2] Don't pack on-medium structures, because GCC emits crappy code

If we use __attribute__((packed)), GCC will _also_ assume that the
structures aren't sensibly aligned, and it'll emit code to cope with
that instead of straight word load/save. This can be _very_ suboptimal
on architectures like ARM.

Ideally, we want an attribute which just tells GCC not to do any
padding, without the alignment side-effects. In the absense of that,
we'll just drop the 'packed' attribute and hope that everything stays as
it was (which to be fair is fairly much what we expect). And add some
paranoia checks in the initialisation code, which should be optimised
away completely in the normal case.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD NAND] Modify check for modules registering NAND devices without ->owner
David Woodhouse [Sun, 14 May 2006 15:54:39 +0000 (16:54 +0100)]
[MTD NAND] Modify check for modules registering NAND devices without ->owner

Make it work even with compilers which lack the wit to notice that
THIS_MODULE is always non-NULL. Use #ifdef MODULE instead. It's only
a temporary debugging check anyway.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD] Clean up <linux/mtd/physmap.h> to fix modular build
David Woodhouse [Sun, 14 May 2006 11:25:19 +0000 (12:25 +0100)]
[MTD] Clean up <linux/mtd/physmap.h> to fix modular build

... and also fix the multiple inclusion guard so it actually _works_

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD] Remove gratuitous inclusion of ARM-only header from physmap.c
David Woodhouse [Sun, 14 May 2006 10:59:20 +0000 (11:59 +0100)]
[MTD] Remove gratuitous inclusion of ARM-only header from physmap.c

The physmap platform driver conversion added to physmap.c an include
of asm/mach/flash.h which is 1) ARM-specific; and 2) isn't actually
necessary.  Remove it.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[JFFS2] Reduce excessive node count for syslog files.
David Woodhouse [Sun, 14 May 2006 03:06:24 +0000 (04:06 +0100)]
[JFFS2] Reduce excessive node count for syslog files.

We currently get fairly poor behaviour with files which get many short
writes, such as system logs. This is because we end up with many tiny
data nodes, and the rbtree gets massive. None of these nodes are
actually obsolete, so they are counted as 'clean' space. Eraseblocks can
be entirely full of these nodes (which are REF_NORMAL instead of
REF_PRISTINE), and still they count entirely towards 'used_size' and the
eraseblocks can sit on the clean_list for a long time without being
picked for GC.

One way to alleviate this in the long term is to account REF_NORMAL
space separately from REF_PRISTINE space, rather than counting them both
towards used_size. Then these eraseblocks can be picked for GC and the
offending nodes will be garbage collected.

The short-term fix, though -- which probably makes sense even if we do
eventually implement the above -- is to merge these nodes as they're
written. When we write the last byte in a page, write the _whole_ page.
This obsoletes the earlier nodes in the page _immediately_ and we don't
even need to wait for the garbage collection to do it.

Original implementation from Ferenc Havasi <havasi@inf.u-szeged.hu>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD] Fix legacy character sets throughout drivers/mtd, include/linux/mtd
David Woodhouse [Sun, 14 May 2006 00:51:54 +0000 (01:51 +0100)]
[MTD] Fix legacy character sets throughout drivers/mtd, include/linux/mtd

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD] Deal correctly with NOR chips which are smaller than the map window
David Woodhouse [Sun, 14 May 2006 00:40:50 +0000 (01:40 +0100)]
[MTD] Deal correctly with NOR chips which are smaller than the map window

We used to calculate the number of chips to be zero, allocate an array
of that size, then nasty things would happen when we attempt to access
the first object in that zero-sized array.

Now, if the number of _full_ chips that would fit into the map is zero,
we allocate an array of one anyway, and then artificially reduce the
total size of the resulting MTD device to fit in the map.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years agomtd: fix memory leak in block2mtd_setup()
Jesper Juhl [Sat, 13 May 2006 23:42:25 +0000 (01:42 +0200)]
mtd: fix memory leak in block2mtd_setup()

There's a mem leak in drivers/mtd/devices/block2mtd.c::block2mtd_setup()

We can leak 'name' allocated with kmalloc in 'parse_name' if leave via
the 'parse_err' macro since it contains a return but doesn't do any
freeing.

Spotted by coverity checker as bug 615.

Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD] Fix module refcounting in NAND board drivers.
David Woodhouse [Sun, 14 May 2006 00:20:46 +0000 (01:20 +0100)]
[MTD] Fix module refcounting in NAND board drivers.

The _board_ driver needs to be mtd->owner, and it in turn pins the
nand.ko module. Fix them all to actually do that, and fix nand.ko not to
overwrite it -- and also to check that the caller sets it, if the caller
is a module.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years agomtd: fix memory leaks in phram_setup
Jesper Juhl [Sat, 13 May 2006 23:07:18 +0000 (01:07 +0200)]
mtd: fix memory leaks in phram_setup

There are two code paths in drivers/mtd/devices/phram.c::phram_setup() that
will leak memory.
Memory is allocated to the variable 'name' with kmalloc() by the
parse_name() function, but if we leave by way of the parse_err() macro,
then that memory is never kfree()'d, nor is it ever used with
register_device() so it won't be freed later either - leak.

Found by the Coverity checker as #593 - simple fix below.

Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD NAND] Indent all of drivers/mtd/nand/*.c.
David Woodhouse [Sat, 13 May 2006 17:07:53 +0000 (18:07 +0100)]
[MTD NAND] Indent all of drivers/mtd/nand/*.c.

It was just too painful to deal with.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD NAND] Reduce paranoia level when scanning for bad blocks on virgin chips
David Woodhouse [Sat, 13 May 2006 15:14:26 +0000 (16:14 +0100)]
[MTD NAND] Reduce paranoia level when scanning for bad blocks on virgin chips

We were scanning for 0xFF through the entire chip -- which takes a while
when it's a 512MiB device as I have on my current toy. The specs only say
we need to check certain bytes -- so do only that.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years agoRevert "[PATCH] i386: export: memory more than 4G through /proc/iomem"
Linus Torvalds [Sat, 13 May 2006 15:01:23 +0000 (08:01 -0700)]
Revert "[PATCH] i386: export: memory more than 4G through /proc/iomem"

This reverts commit 10dbe196a8da6b3196881269c6639c0ec11c36cb.

The resource struct is still 32-bit, so trying to save a 64-bit memory
size there obviously won't work.

When we merge the 64-bit resource series, we can re-enable this.

Thanks to Sachin Sant and Maneesh Soni for debugging

Cc: Maneesh Soni <maneesh@in.ibm.com>
Cc: Sachin Sant <sachinp@in.ibm.com>
Cc: Russell King <rmk+lkml@arm.linux.org.uk>
Cc: Sharyathi Nagesh <sharyath@in.ibm.com>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[JFFS2][XATTR] Fix obvious typo
KaiGai Kohei [Sat, 13 May 2006 06:22:29 +0000 (15:22 +0900)]
[JFFS2][XATTR] Fix obvious typo

[2/2] jffs2-xattr-v5.2-02-fix_obvious_typo.patch

Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
18 years ago[JFFS2][XATTR] Handling the duplicate JFFS2_NODETYPE_XATTR node cases.
KaiGai Kohei [Sat, 13 May 2006 06:21:38 +0000 (15:21 +0900)]
[JFFS2][XATTR] Handling the duplicate JFFS2_NODETYPE_XATTR node cases.

When jffs2_sum_process_sum_data() found a JFFS2_NODETYPE_XATTR
which has duplicate xid and older version, an error was returned
without appropriate process.
In the result, mounting filesystem is failed.

This patch fix this problem. If jffs2_setup_xattr_datum() returned
-EEXIST, the caller marks this node as DIRTY_SPACE().

[1/2] jffs2-xattr-v5.2-01-fix-duplicate-xdatum.patch

Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
18 years ago[JFFS2][XATTR] remove redundant pointer cast in acl.c
KaiGai Kohei [Sat, 13 May 2006 06:20:24 +0000 (15:20 +0900)]
[JFFS2][XATTR] remove redundant pointer cast in acl.c

remove redundant pointer cast in acl.c.

[10/10] jffs2-xattr-v5.1-10-remove_pointer_cast.patch

Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
18 years ago[JFFS2][XATTR] remove '__KERNEL__' from acl.h
KaiGai Kohei [Sat, 13 May 2006 06:19:36 +0000 (15:19 +0900)]
[JFFS2][XATTR] remove '__KERNEL__' from acl.h

[9/10] jffs2-xattr-v5.1-09-remove__KERNEL__.patch

Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
18 years ago[JFFS2][XATTR] remove senseless comment
KaiGai Kohei [Sat, 13 May 2006 06:19:03 +0000 (15:19 +0900)]
[JFFS2][XATTR] remove senseless comment

remove senseless comment.

[8/10] jffs2-xattr-v5.1-08-remove_senseless_comment.patch

Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
18 years ago[JFFS2][XATTR] Unify each file header part with any jffs2 file.
KaiGai Kohei [Sat, 13 May 2006 06:18:27 +0000 (15:18 +0900)]
[JFFS2][XATTR] Unify each file header part with any jffs2 file.

Unify each file header part with any jffs2 file.

[7/10] jffs2-xattr-v5.1-07-unify_file_header.patch

Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
18 years ago[JFFS2][XATTR] '#include <linux/list.h>' was added into xattr.h.
KaiGai Kohei [Sat, 13 May 2006 06:17:11 +0000 (15:17 +0900)]
[JFFS2][XATTR] '#include <linux/list.h>' was added into xattr.h.

'#include <linux/list.h>' was added into xattr.h.
because 'struct list_head' is used in this header file.

[6/10] jffs2-xattr-v5.1-06-add_list.h.patch

Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
18 years ago[JFFS2][XATTR] Remove jffs2_garbage_collect_xattr(c, ic)
KaiGai Kohei [Sat, 13 May 2006 06:16:13 +0000 (15:16 +0900)]
[JFFS2][XATTR] Remove jffs2_garbage_collect_xattr(c, ic)

Remove jffs2_garbage_collect_xattr(c, ic).
jffs2_garbage_collect_xattr_datum/ref() are called from gc.c directly.

In original implementation, jffs2_garbage_collect_xattr(c, ic) returns
with holding a spinlock if 'ic' is inode_cache. But it returns after
releasing a spinlock if 'ic' is xattr_datum/ref.
It looks so confusable behavior. Thus, this patch makes caller manage
locking/unlocking.

[5/10] jffs2-xattr-v5.1-05-update_xattr_gc.patch

Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
18 years ago[JFFS2][XATTR] Remove 'struct list_head ilist' from jffs2_inode_cache.
KaiGai Kohei [Sat, 13 May 2006 06:15:07 +0000 (15:15 +0900)]
[JFFS2][XATTR] Remove 'struct list_head ilist' from jffs2_inode_cache.

This patch can reduce 4-byte of memory usage per inode_cache.

[4/10] jffs2-xattr-v5.1-04-remove_ilist_from_ic.patch

Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
18 years ago[JFFS2][XATTR] Add a description about c->xattr_sem
KaiGai Kohei [Sat, 13 May 2006 06:14:14 +0000 (15:14 +0900)]
[JFFS2][XATTR] Add a description about c->xattr_sem

Add a description about the c->xattr_sem read/write semaphore
into README.Locking.

[3/10] jffs2-xattr-v5.1-03-append_README.Locking.patch

Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
18 years ago[JFFS2][XATTR] remove typedef from posix_acl related definition.
KaiGai Kohei [Sat, 13 May 2006 06:13:27 +0000 (15:13 +0900)]
[JFFS2][XATTR] remove typedef from posix_acl related definition.

jffs2_acl_header, jffs2_acl_entry and jffs2_acl_entry_short were redefined
with using 'struct' instead of 'typedef' in kernel implementation.

[1/10] jffs2-xattr-v5.1-01-remove_typedef_kernel.patch

Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
18 years ago[JFFS2][XATTR] XATTR support on JFFS2 (version. 5)
KaiGai Kohei [Sat, 13 May 2006 06:09:47 +0000 (15:09 +0900)]
[JFFS2][XATTR] XATTR support on JFFS2 (version. 5)

This attached patches provide xattr support including POSIX-ACL and
SELinux support on JFFS2 (version.5).

There are some significant differences from previous version posted
at last December.
The biggest change is addition of EBS(Erase Block Summary) support.
Currently, both kernel and usermode utility (sumtool) can recognize
xattr nodes which have JFFS2_NODETYPE_XATTR/_XREF nodetype.

In addition, some bugs are fixed.
- A potential race condition was fixed.
- Unexpected fail when updating a xattr by same name/value pair was fixed.
- A bug when removing xattr name/value pair was fixed.

The fundamental structures (such as using two new nodetypes and exclusion
mechanism by rwsem) are unchanged. But most of implementation were reviewed
and updated if necessary.
Espacially, we had to change several internal implementations related to
load_xattr_datum() to avoid a potential race condition.

[1/2] xattr_on_jffs2.kernel.version-5.patch
[2/2] xattr_on_jffs2.utils.version-5.patch

Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD NAND] Update CS553x NAND driver: Hardware ECC support, optimisations.
David Woodhouse [Sat, 13 May 2006 03:12:40 +0000 (04:12 +0100)]
[MTD NAND] Update CS553x NAND driver: Hardware ECC support, optimisations.

- Implement HW ECC support,
- Provide read_buf() and write_buf() routines using memcpy
- Use on-flash bad block table
- Fix module refcounting
- Avoid read/modify/write in hwcontrol()
- Minor cosmetic fixes

Partly based on code and ideas from Tom Sylla <tom.sylla@amd.com>

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years ago[MTD NAND] Use vmalloc for buffer when scanning for bad blocks.
David Woodhouse [Sat, 13 May 2006 03:03:42 +0000 (04:03 +0100)]
[MTD NAND] Use vmalloc for buffer when scanning for bad blocks.

These new chips have 128KiB blocks. Don't try to kmalloc that.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
18 years agoAlternative fix for MMC oops on unmount after removal
Linus Torvalds [Sat, 13 May 2006 01:42:09 +0000 (18:42 -0700)]
Alternative fix for MMC oops on unmount after removal

Make sure to clear the driverfs_dev pointer when we do del_gendisk() (on
disk removal), so that other users that may still have a ref to the disk
won't try to use the stale pointer.

Also move the KOBJ_REMOVE uevent handler up, so that the uevent still
has access to the driverfs_dev data.

This all should hopefully fix the problems with MMC umounts after device
removals that caused commit 56cf6504fc1c0c221b82cebc16a444b684140fb7 and
its reversal (1a2acc9e9214699a99389e323e6686e9e0e2ca67).

Original problem reported by Todd Blumer and others.

Acked-by: Greg KH <gregkh@suse.de>
Cc: Russell King <rmk+lkml@arm.linux.org.uk>
Cc: James Bottomley <James.Bottomley@SteelEye.com>
Cc: Erik Mouw <erik@harddisk-recovery.com>
Cc: Andrew Vasquez <andrew.vasquez@qlogic.com>
Cc: Todd Blumer <todd@sdgsystems.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>