Some versions of Xen 3.x set their magic number to "xen-3.[12]", so
relax the test to match them.
Signed-off-by: NJeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

* 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus:
  [MIPS] Malta: Enable tickless and highres timers.
  [MIPS] Bigsur: Enable tickless and and highres timers.
  qemu: do not enable IP7 blindly
  [MIPS] Alchemy: Fix Au1x SD controller IRQ
  [MIPS] Don't byteswap writes to display when running bigendian

Ultimately to implement /proc perfectly we need an implementation of
d_revalidate because files and directories can be removed behind the back
of the VFS, and d_revalidate is the only way we can let the VFS know that
this has happened.

Unfortunately the linux VFS can not cope with anything in the path to a
mount point going away.  So a proper d_revalidate method that calls d_drop
also needs to call have_submounts which is moderately expensive, so you
really don't want a d_revalidate method that unconditionally calls it, but
instead only calls it when the backing object has really gone away.

proc generic entries only disappear on module_unload (when not counting the
fledgling network namespace) so it is quite rare that we actually encounter
that case and has not actually caused us real world trouble yet.

So until we get a proper test for keeping dentries in the dcache fix the
current d_revalidate method by completely removing it.  This returns us to
the current status quo.

So with CONFIG_NETNS=n things should look as they have always looked.

For CONFIG_NETNS=y things work most of the time but there are a few rare
corner cases that don't behave properly.  As the network namespace is
barely present in 2.6.24 this should not be a problem.
Signed-off-by: NEric W. Biederman <ebiederm@xmission.com>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: "Denis V. Lunev" <den@openvz.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We have a wifi module connected to the spi bus and got sometimes FIFO
overrun errors on the spi bus.

After some investigation i found that the driver loads the TCR (transmit
count) register before the RCR (receive count).  When the transfer list is
not empty the atmel_spi_next_message is called while tx and rx are enabled.
 As soon as the TCR is loaded, hardware starts transfer and causes a rx
fifo overrun because the RCR is not loaded yet.

Load the RCR before the TCR.  After this patch the fifo overrun disapears
at out setup.
Signed-off-by: NHaavard Skinnemoen <hskinnemoen@atmel.com>
Signed-off-by: NRini van Zetten <rini@arvoo.nl>
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Some bad email addresses are removed:

        linux-tr@linuxtr.net
        ehaase@inf.fu-berlin.de

Some are updated:

        linux@maxim.org.za to andrew@sanpeople.com
        linux-kernel@linux-mips.org to linux-mips@linux-mips.org
        jdike@karaya.com to jdike@addtoit.com

The PCMCIA entry is corrected to a web address

The ZR36120 VIDEO FOR LINUX DRIVER entry is removed
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The esp_reset_cleanup() function is called with the host lock held and
invokes starget_for_each_device() which wants to take it too.  Here is a
fix along the lines of shost_for_each_device()/__shost_for_each_device()
adding a __starget_for_each_device() counterpart which assumes the lock
has already been taken.

Eventually, I think the driver should get modified so that more work is
done as a softirq rather than in the interrupt context, but for now it
fixes a bug that causes the spinlock debugger to fire.

While at it, it fixes a small number of cosmetic problems with
starget_for_each_device() too.
Signed-off-by: NMaciej W. Rozycki <macro@linux-mips.org>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Cc: <stable@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix NULL dereference in umem.c
Signed-off-by: NNeil Brown <neilb@suse.de>
Tested-by: NDave Chinner <dgc@sgi.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

There should be an of_node_put when breaking out of a loop that iterates
using for_each_compatible_node.

This was detected and fixed using the following semantic patch.
(http://www.emn.fr/x-info/coccinelle/)

// <smpl>
@@
identifier d;
type T;
expression e;
iterator for_each_compatible_node;
@@

T *d;
...
for_each_compatible_node(d,...)
  {... when != of_node_put(d)
       when != e = d
(
   return d;
|
+  of_node_put(d);
?  return ...;
)
...}
// </smpl>
Signed-off-by: NJulia Lawall <julia@diku.dk>
Acked-by: NGrant Likely <grant.likely@secretlab.ca>
Acked-by: NPeter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The follow_hugetlb_page() fix I posted (merged as git commit
5b23dbe8) missed one case.  If the pte is
present, but not writable and write access is requested by the caller to
get_user_pages(), the code will do the wrong thing.  Rather than calling
hugetlb_fault to make the pte writable, it notes the presence of the pte
and continues.

This simple one-liner makes sure we also fault on the pte for this case.
Please apply.
Signed-off-by: NAdam Litke <agl@us.ibm.com>
Acked-by: NDave Kleikamp <shaggy@us.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NWill Newton <will.newton@gmail.com>
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NWill Newton <will.newton@gmail.com>
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NWill Newton <will.newton@gmail.com>
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Some places where CLOCK_TICK_RATE may be used incorrectly:

arch/arm/mach-mx3/time.c:125:   __raw_writel((v / CLOCK_TICK_RATE) - 1, MXC_GPT_GPTPR);
drivers/watchdog/davinci_wdt.c:103:     timer_margin = (((u64)heartbeat * CLOCK_TICK_RATE) & 0xffffffff);
drivers/watchdog/davinci_wdt.c:105:     timer_margin = (((u64)heartbeat * CLOCK_TICK_RATE) >> 32);
drivers/watchdog/ks8695_wdt.c:64:       unsigned long tval = wdt_time * CLOCK_TICK_RATE;

I'm not sure whether this definition is used there, but adding parentheses
should be good anyway.
Signed-off-by: NRoel Kluin <12o3l@tiscali.nl>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix kernel-doc comments in drivers/pcmcia/:

- ti113x.h does not contain kernel-doc, so don't use /** to begin a doc
  comment
- yenta_socket.c: remove /** on non-kernel-doc comments;
  escape the ':' in an "http:" comment so that it won't be treated as a
  section heading;
- cs.c: remove /** on non-kernel-doc comments & add function parameter info
- ds.c: fix function parameter info
Signed-off-by: NRandy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

revert

    commit 55d9fcf5
    Author: Matthew Wilcox <matthew@wil.cx>
    Date:   Mon Jul 30 15:19:18 2007 -0600

        [SCSI] dpt_i2o: convert to SCSI hotplug model

         - Delete refereces to HOSTS_C
         - Switch to module_init/module_exit instead of detect/release
         - Don't pass around the host template and rename it to adpt_template
         - Switch from scsi_register/scsi_unregister to scsi_host_alloc,
           scsi_add_host, scsi_scan_host and scsi_host_put.

Because it caused (for unknown reasons) Andres' all-data-reads-as-zeroes
problem, reported at
http://groups.google.com/group/fa.linux.kernel/msg/083a9acff0330234

Cc: Matthew Wilcox <matthew@wil.cx>
Cc: "Salyzyn, Mark" <mark_salyzyn@adaptec.com>
Cc: James Bottomley <James.Bottomley@SteelEye.com>
Acked-by: NFUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Anders Henke <anders.henke@1und1.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
Cc: <stable@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

AOE forgot to initialise its queue's backing_dev_info, so kernels crash.
(http://bugzilla.kernel.org/show_bug.cgi?id=9482)

Fix that and consoldate aoeblk_gdalloc()'s error handling.

Thanks be to Jon for reporting and testing.

Cc: "Ed L. Cashin" <ecashin@coraid.com>
Cc: <stable@kernel.org>
Cc: "Jon Nelson" <jnelson@jamponi.net>
Acked-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6:
  firewire: OHCI 1.0 Isochronous Receive support

* 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6:
  [IA64] Fix iosapic interrupt delivery mode for CPE
  [IA64] kprobe: make kreturn probe handler stack unwind correct
  [IA64] operator priority fix in acpi_map_lsapic()
  [IA64] Add missing "space" to concatenated strings
  [IA64] make full use of macro efi_md_size
  [IA64] rename _bss to __bss_start
  [IA64] SGI Altix : fix bug in sn_io_late_init()
  [IA64] iosapic cleanup
  [IA64] signal : fix missing error checkings
  [IA64] export copy_page() to modules
  [IA64] don't assume that unwcheck.py is executable
  [IA64] increase .data.patch offset

Third rendition of FireWire OHCI 1.0 Isochronous Receive support, using a
zer-copy method similar to OHCI 1.1 which puts the IR data payload directly
into the userspace buffer. The zero-copy implementation eliminates the
video artifacts, audio popping, and buffer underrun problems seen with
version 1 of this patch, as well as fixing a regression in OHCI 1.1 support
introduced by version 2 of this patch.

Successfully tested in OHCI 1.1 mode on the following chipsets:

- NEC uPD72847 (rev 01), OHCI 1.1 (PCI)
- Ti XIO2200(A) (rev 01), OHCI 1.1 (PCIe)
- Ti TSB41AB2 (rev 01), OHCI 1.1 (PCI on SB Audigy)
- Apple UniNorth 2 (rev 81), OHCI 1.1 (PowerBook G4 onboard)

Successfully tested in OHCI 1.0 mode on the following chipsets:

- Agere FW323 (rev 06), OHCI 1.0 (Mac Mini onboard)
- Agere FW323 (rev 06), OHCI 1.0 (PCI)
- Via VT6306 (rev 46), OHCI 1.0 (PCI)
- NEC OrangeLink (rev 01), OHCI 1.0 (PCI)
- NEC uPD72847 (rev 01), OHCI 1.1 (PCI)
- Ti XIO2200(A) (rev 01), OHCI 1.1 (PCIe)

The bulk of testing was done in an x86_64 system, but was also successfully
sanity-tested on other systems, including a PPC(32) PowerBook G4 and an i686
EPIA M10k. Crude benchmarking (watching top during capture) puts the cpu
utilization during capture on the EPIA's 1GHz Via C3 processor around 13%,
which is down from 30% with the v1 code.

Some implementation details:

To maintain the same userspace API as dual-buffer mode, we set up two
descriptors for every incoming packet. The first is an INPUT_MORE descriptor,
pointing to a buffer large enough to hold just the packet's iso headers,
immediately followed by an INPUT_LAST descriptor, pointing to a chunk of the
userspace buffer big enough for the packet's data payload. With this setup,
each incoming packet fills in these two descriptors in a manner that very
closely emulates dual-buffer receive, to the point where the bulk of the
handle_ir_* code is now identical between the two (and probably primed for
some restructuring to share code between them).

The only caveat I have at the moment is that neither of my OHCI 1.0 Via
VT6307-based FireWire controllers work particularly well with this code
for reasons I have yet to figure out.
Signed-off-by: NJarod Wilson <jwilson@redhat.com>
Signed-off-by: NStefan Richter <stefanr@s5r6.in-berlin.de>

* 'for-linus' of git://oss.sgi.com:8090/xfs/xfs-2.6:
  [XFS] Fix xfs_ichgtime()s broken usage of I_SYNC
  [XFS] Make xfsbufd threads freezable
  [XFS] revert to double-buffering readdir
  [XFS] Fix broken inode cluster setup.
  [XFS] Clear XBF_READ_AHEAD flag on I/O completion.
  [XFS] Fixed a few bugs in xfs_buf_associate_memory()
  [XFS] 971064 Various fixups for xfs_bulkstat().
  [XFS] Fix dbflush panic in xfs_qm_sync.

* git://git.kernel.org/pub/scm/linux/kernel/git/sam/fix-kbuild:
  kbuild: fix building with O=.. options
  kbuild: fix building with redirected output.

This reverts commit fd6e7321, which
helped up things on MIPS, but was wrong for everything else.  As Ralf
Baechle puts it:

  "It seems the whole MIPS resource managment is complicated enough (out
   of necessity) that only a few people actually grok it.  Ioports being
   actually memory mapped on MIPS only makes the confusion worse, sigh."
Requested-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: NAlan Cox <alan@redhat.com>
Acked-by: NRalf Baechle <ralf@linux-mips.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

PowerMac and CHRP/BriQ platforms have quirks to switch some IDE
controllers from legacy mode to fully native mode. Those quirks
however will not work properly anymore due to a change to the
generic code to better handle legacy IDE resources.

This fixes it by moving those quirk to "early" quirks (so they
run before resources are probed for the devices) and clearing
all BARs after the conversion to force a reallocation of sane
values.
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The recent I_LOCK->I_SYNC changes mistakenly changed xfs_ichgtime to look
at I_SYNC instead of I_LOCK. This was incorrect and prevents newly created
inodes from moving to the dirty list. Change this to the correct check
which is for I_NEW, not I_LOCK or I_SYNC so that behaviour is correct.

SGI-PV: 974225
SGI-Modid: xfs-linux-melb:xfs-kern:30204a
Signed-off-by: NDavid Chinner <dgc@sgi.com>
Signed-off-by: NLachlan McIlroy <lachlan@sgi.com>

Fix breakage caused by commit 83144186
that did not introduce the necessary call to set_freezable() in
xfs/linux-2.6/xfs_buf.c .

SGI-PV: 974224
SGI-Modid: xfs-linux-melb:xfs-kern:30203a
Signed-off-by: NRafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: NDavid Chinner <dgc@sgi.com>
Signed-off-by: NLachlan McIlroy <lachlan@sgi.com>

The current readdir implementation deadlocks on a btree buffers locks
because nfsd calls back into ->lookup from the filldir callback. The only
short-term fix for this is to revert to the old inefficient
double-buffering scheme.

SGI-PV: 973377
SGI-Modid: xfs-linux-melb:xfs-kern:30201a
Signed-off-by: NChristoph Hellwig <hch@infradead.org>
Signed-off-by: NDavid Chinner <dgc@sgi.com>
Signed-off-by: NLachlan McIlroy <lachlan@sgi.com>

The radix tree based inode caches did away with the inode cluster hashes,
replacing them with a bunch of masking and gang lookups on the radix tree.

This masking got broken when moving the code to per-ag radix trees and
indexing by agino # rather than straight inode number. The result is
clustered inode writeback does not cluster and things can go extremely
slowly when there are lots of inodes to write.

Fix it up by comparing the agino # of the inode we just looked up to the
index of the cluster we are looking for.
Tested-by: NTorsten Kaiser <just.for.lkml@googlemail.com>

SGI-PV: 972915
SGI-Modid: xfs-linux-melb:xfs-kern:30033a
Signed-off-by: NDavid Chinner <dgc@sgi.com>
Signed-off-by: NLachlan McIlroy <lachlan@sgi.com>

SGI-PV: 972554
SGI-Modid: xfs-linux-melb:xfs-kern:30128a
Signed-off-by: NLachlan McIlroy <lachlan@sgi.com>
Signed-off-by: NChristoph Hellwig <hch@infradead.org>

- calculation of 'page_count' was incorrect as it did not
  consider the offset of 'mem' into the first page. The
  logic to bump 'page_count' didn't work if 'len' was <=
  PAGE_CACHE_SIZE (ie offset = 3k, len = 2k).
- setting b_buffer_length to 'len' is incorrect if 'offset'
  is > 0. Set it to the total length of the buffer.
- I suspect that passing a non-aligned address into
  mem_to_page() for the first page may have been causing
  issues - don't know but just tidy up that code anyway.

SGI-PV: 971596
SGI-Modid: xfs-linux-melb:xfs-kern:30143a
Signed-off-by: NLachlan McIlroy <lachlan@sgi.com>
Signed-off-by: NChristoph Hellwig <hch@infradead.org>

- sanity check for NULL user buffer in xfs_ioc_bulkstat[_compat]()
- remove the special case for XFS_IOC_FSBULKSTAT with count == 1. This
  special case causes bulkstat to fail because the special case uses
  xfs_bulkstat_single() instead of xfs_bulkstat() and the two functions
  have different semantics.  xfs_bulkstat() will return the next inode
  after the one supplied while skipping internal inodes (ie quota inodes).
  xfs_bulkstate_single() will only lookup the inode supplied and return
  an error if it is an internal inode.
- in xfs_bulkstat(), need to initialise 'lastino' to the inode supplied
  so in cases were we return without examining any inodes the scan wont
  restart back at zero.
- sanity check for valid *ubcountp values. Cannot sanity check for valid
  ubuffer here because some users of xfs_bulkstat() don't supply a buffer.
- checks against 'ubleft' (the space left in the user's buffer) should be
  against 'statstruct_size' which is the supplied minimum object size.
  The mixture of checks against statstruct_size and 0 was one of the
  reasons we were skipping inodes.
- if the formatter function returns BULKSTAT_RV_NOTHING and an error and
  the error is not ENOENT or EINVAL then we need to abort the scan. ENOENT
  is for inodes that are no longer valid and we just skip them. EINVAL is
  returned if we try to lookup an internal inode so we skip them too. For
  a DMF scan if the inode and DMF attribute cannot fit into the space left
  in the user's buffer it would return ERANGE. We didn't handle this error
  and skipped the inode. We would continue to skip inodes until one fitted
  into the user's buffer or we completed the scan.
- put back the recalculation of agino (that got removed with the last fix)
  at the end of the while loop. This is because the code at the start of
  the loop expects agino to be the last inode examined if it is non-zero.
- if we found some inodes but then encountered an error, return success
  this time and the error next time. If the formatter aborted with ENOMEM
  we will now return this error but only if we couldn't read any inodes.
  Previously if we encountered ENOMEM without reading any inodes we
  returned a zero count and no error which falsely indicated the scan was
  complete.

SGI-PV: 973431
SGI-Modid: xfs-linux-melb:xfs-kern:30089a
Signed-off-by: NLachlan McIlroy <lachlan@sgi.com>
Signed-off-by: NDavid Chinner <dgc@sgi.com>

The recent behaviour layer removal dropped the check for quotas that have
been requested at mount time but have subsequently been turned off. This
results in a panic when accessing m_quotainfo which has been freed.

This patch adds the check originally made by xfs_qm_syncall() to
xfs_qm_sync().

SGI-PV: 969769
SGI-Modid: xfs-linux-melb:xfs-kern:29908a
Signed-off-by: NDonald Douwsma <donaldd@sgi.com>
Signed-off-by: NDavid Chinner <dgc@sgi.com>
Signed-off-by: NLachlan McIlroy <lachlan@sgi.com>

Both slob and slub react to __GFP_ZERO by clearing the allocation, which
means that passing the GFP_ZERO bit down to the page allocator is just
wasteful and pointless.
Acked-by: NMatt Mackall <mpm@selenic.com>
Reviewed-by: NPekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The check introduced in commit:
4f1127e2 "kbuild: fix
infinite make recursion"

caused certain external modules not to build and
also caused 'make targz-pkg' to fail.
This is a minimal fix so we revert to previous
behaviour - but we do not overwrite the Makefile
in the top-level directory.
Signed-off-by: NSam Ravnborg <sam@ravnborg.org>
Tested-by: NJay Cliburn <jacliburn@bellsouth.net>
Cc: Jay Cliburn <jacliburn@bellsouth.net>

Jan Altenberg <jan.altenberg@linutronix.de> reported that
building with redirected input like this failed:
make O=dir oldconfig bzImage < /dev/null

The problem were caused by a make silentoldconfig being
run before oldconfig and with a non-recent .config the build
failed because silentoldconfig requires non-redirected stdin.

Silentoldconfig was run as a side-effect of having the
top-level Makefile re-made by make.
Introducing an empty rule for the top-level Makefile
(and Kbuild.include) fixed the issue.
Signed-off-by: NSam Ravnborg <sam@ravnborg.org>

Most Malta use an FPGA CPU card which rarely is good for more than 40MHz.
So the performance penalta of the regular timer interrupt, especially
for the VSMP kernel model is significant, even at a mere 100Hz.
Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

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

IP7 will be enabled automatically in mips_clockevent_init(), if available.
Signed-off-by: NAtsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

With the introduction of MIPS_CPU_IRQ_BASE, the hardcoded IRQ number of
the au1100/au1200 SD controller(s) is no longer valid.
Signed-off-by: NManuel Lauss <mano@roarinelk.homelinux.net>
Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

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