fixes crashes under high nfs load

SGI-PV: 941429
SGI-Modid: xfs-linux:xfs-kern:197929a
Signed-off-by: NChristoph Hellwig <hch@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

which can cause an extent hole to be filled and a free extent to be
processed. In this case, we make a few mistakes: forget to pass back the
transaction, forget to put a hold on the buffer and forget to add the buf
to the new transaction.

SGI-PV: 940366
SGI-Modid: xfs-linux:xfs-kern:23594a
Signed-off-by: NTim Shimmin <tes@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 934766
SGI-Modid: xfs-linux:xfs-kern:197783a
Signed-off-by: NChristoph Hellwig <hch@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 940531
SGI-Modid: xfs-linux:xfs-kern:197782a
Signed-off-by: NChristoph Hellwig <hch@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 904196
SGI-Modid: xfs-linux:xfs-kern:23563a
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 938899
SGI-Modid: xfs-linux:xfs-kern:197403a
Signed-off-by: NEric Sandeen <sandeen@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

linvfs_clear_inode(). The behavior may go away in VOP_INACTIVE. 

SGI-PV: 941000
SGI-Modid: xfs-linux:xfs-kern:197355a
Signed-off-by: NFelix Blyakher <felixb@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

having previously mounted with quotas.

SGI-PV: 940491
SGI-Modid: xfs-linux:xfs-kern:23388a
Signed-off-by: NTim Shimmin <tes@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 940531
SGI-Modid: xfs-linux:xfs-kern:196888a
Signed-off-by: NChristoph Hellwig <hch@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 936584
SGI-Modid: xfs-linux:xfs-kern:196886a
Signed-off-by: NChristoph Hellwig <hch@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

because aio+dio completions may happen from irq context but we need
process context for converting unwritten extents.  We also queue regular
direct I/O completions to workqueue for regularity, there's only one
queue_work call per syscall.

SGI-PV: 934766
SGI-Modid: xfs-linux:xfs-kern:196857a
Signed-off-by: NChristoph Hellwig <hch@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 934766
SGI-Modid: xfs-linux:xfs-kern:196857a
Signed-off-by: NChristoph Hellwig <hch@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 934766
SGI-Modid: xfs-linux:xfs-kern:196856a
Signed-off-by: NChristoph Hellwig <hch@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

time

SGI-PV: 934766
SGI-Modid: xfs-linux:xfs-kern:196854a
Signed-off-by: NChristoph Hellwig <hch@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 934766
SGI-Modid: xfs-linux:xfs-kern:196852a
Signed-off-by: NChristoph Hellwig <hch@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

<victor@cetuc.puc-rio.br>

SGI-PV: 940376
SGI-Modid: xfs-linux:xfs-kern:196705a
Signed-off-by: NChristoph Hellwig <hch@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

when it goes to force out the log, and get the tail lsn, it will want to
get the AIL lock.

SGI-PV: 940076
SGI-Modid: xfs-linux:xfs-kern:23260a
Signed-off-by: NTim Shimmin <tes@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 936236
SGI-Modid: xfs-linux:xfs-kern:195878a
Signed-off-by: NChristoph Hellwig <hch@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 907752
SGI-Modid: xfs-linux:xfs-kern:23163a
Signed-off-by: NNathan Scott <nathans@sgi.com>

there..

SGI-PV: 939444
SGI-Modid: xfs-linux:xfs-kern:23162a
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 931456
SGI-Modid: xfs-linux:xfs-kern:23155a
Signed-off-by: NTim Shimmin <tes@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

include all the components which make up the transaction in the ondisk
log. Having this incomplete has shown up as problems on IRIX when some v2
log changes went in. The symptom was the msg of "xfs_log_write:
reservation ran out. Need to up reservation" and was seen on synchronous
writes on files with lots of holes (and therefore lots of extents).

SGI-PV: 931457
SGI-Modid: xfs-linux:xfs-kern:23095a
Signed-off-by: NTim Shimmin <tes@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

writes.

SGI-PV: 938145
SGI-Modid: xfs-linux:xfs-kern:23088a
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 938905
SGI-Modid: xfs-linux:xfs-kern:195240a
Signed-off-by: NEric Sandeen <sandeen@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

are getting ENOSPC errors on writes. When we fail to allocate space for
indirect blocks in xfs_bmapi() make sure we release the direct block
allocation before returning.

SGI-PV: 938502
SGI-Modid: xfs-linux:xfs-kern:22986a
Signed-off-by: NDavid Chinner <dgc@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 938410
SGI-Modid: xfs-linux:xfs-kern:22966a
Signed-off-by: NNathan Scott <nathans@sgi.com>

handlers.

SGI-PV: 938409
SGI-Modid: xfs-linux:xfs-kern:22965a
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 938306
SGI-Modid: xfs-linux:xfs-kern:194627a
Signed-off-by: NChristoph Hellwig <hch@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 935317
SGI-Modid: xfs-linux:xfs-kern:192007a
Signed-off-by: NDean Roehrich <roehrich@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 933765
SGI-Modid: xfs-linux:xfs-kern:190760a
Signed-off-by: NDean Roehrich <roehrich@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

SGI-PV: 933551
SGI-Modid: xfs-linux:xfs-kern:190625a
Signed-off-by: NDean Roehrich <roehrich@sgi.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>

We run into problems if we blindly enable L2 prefetching without
checking that the L2 cache is actually enabled.  Additionaly, if we
disable the L2 cache we need to ensure that we disable L2 prefetching.
Signed-off-by: NKumar Gala <kumar.gala@freescale.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This fixes a problem with pci_map_rom() which doesn't properly
update the ROM BAR value with the address thas allocated for it by the
PCI code. This problem, among other, breaks boot on Mac laptops.

It'ss a new version based on Linus latest one with better error
checking.
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

In adjusting the logic for SLB miss for the dynamic hugepage stuff, I
messed up the !CONFIG_HUGETLB_PAGE case, failing to set the SLB flags
properly.

This fixes it.  It also streamlines the logic for the HUGETLB_PAGE case
(removing a couple of branches) while we're at it.

Booted, and roughly tested on POWER5 (with and without HUGETLB_PAGE),
iSeries/RS64 (no hugepage available), and G5 (with and without
HUGETLB_PAGE).
Signed-off-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This fixes the following compile error:

	...
	  LD      .tmp_vmlinux1
	drivers/built-in.o: In function `frontend_init':
	budget-av.c:(.text+0xb9448): undefined reference to `tda10046_attach'
	budget-av.c:(.text+0xb9518): undefined reference to `tda10021_attach'
	drivers/built-in.o: In function `philips_tu1216_request_firmware':
	budget-av.c:(.text+0xb937b): undefined reference to `request_firmware'
	make: *** [.tmp_vmlinux1] Error 1
Signed-off-by: NAdrian Bunk <bunk@stusta.de>
Acked-by: NJohannes Stezenbach <js@linuxtv.org>
Acked-by: NMichael Krufky <mkrufky@linuxtv.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The following patch fixes two warnings in arch/ppc/syslib/m8xx_setup.c
Signed-off-by: NMarcelo Tosatti <marcelo.tosatti@cyclades.com>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

I had some time to think about PCI assign issues in 2.6.13-rc series.

The major problem here is that we call pci_assign_unassigned_resources()
way too early - at subsys_initcall level. Therefore we give no chances
to ACPI and PnP routines (called at fs_initcall level) to reserve their
respective resources properly, as the comments in drivers/pnp/system.c
and drivers/acpi/motherboard.c suggest:

 /**
  * Reserve motherboard resources after PCI claim BARs,
  * but before PCI assign resources for uninitialized PCI devices
  */

So I moved the pci_assign_unassigned_resources() call to
pcibios_assign_resources() (fs_initcall), which should hopefully fix a
lot of problems and make PCIBIOS_MIN_IO tweaks unnecessary.

Other changes:
- remove resource assignment code from pcibios_assign_resources(), since
  it duplicates pci_assign_unassigned_resources() functionality and
  actually does nothing in 2.6.13;
- modify ROM assignment code as per Ben's suggestion: try to use firmware
  settings by default (if PCI_ASSIGN_ROMS is not set);
- set CARDBUS_IO_SIZE back to 4K as it's a wonderful stress test for
  various setups.

Confirmed by Tero Roponen <teanropo@cc.jyu.fi> (who had problems with
the 4kB CardBus IO size previously).
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>