GitHub/exynos8895/android_kernel_samsung_universal8895.git
15 years agox86, setup: guard against pre-ACPI 3 e820 code not updating %ecx
H. Peter Anvin [Wed, 1 Apr 2009 18:35:00 +0000 (11:35 -0700)]
x86, setup: guard against pre-ACPI 3 e820 code not updating %ecx

Impact: BIOS bug safety

For pre-ACPI 3 BIOSes, pre-initialize the end of the e820 buffer just
in case the BIOS returns an unchanged %ecx but without actually
touching the ACPI 3 extended flags field.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
15 years agox86, setup: ACPI 3, BIOS workaround for E820-probing code
H. Peter Anvin [Sat, 28 Mar 2009 20:53:26 +0000 (13:53 -0700)]
x86, setup: ACPI 3, BIOS workaround for E820-probing code

Impact: ACPI 3 spec compliance, BIOS bug workaround

The ACPI 3 spec added another field to the E820 buffer -- which is
backwards incompatible, since it contains a validity bit.
Furthermore, there has been at least one report of a BIOS which
assumes that the buffer it is pointed at is the same buffer as for the
previous E820 call.  Therefore, read the data into a temporary buffer
and copy the standard part of it if and only if the valid bit is set.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
15 years agox86, setup: preemptively save/restore edi and ebp around INT 15 E820
H. Peter Anvin [Sat, 28 Mar 2009 20:53:26 +0000 (13:53 -0700)]
x86, setup: preemptively save/restore edi and ebp around INT 15 E820

Impact: BIOS bugproofing

Since there are BIOSes known to clobber %ebx and %esi for INT 15 E820,
assume there is something out there clobbering %edi and/or %ebp too,
and don't wait for it to fail.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
15 years agox86, setup: mark %esi as clobbered in E820 BIOS call
Michael K. Johnson [Fri, 27 Mar 2009 17:14:41 +0000 (13:14 -0400)]
x86, setup: mark %esi as clobbered in E820 BIOS call

Jordan Hargrave diagnosed a BIOS clobbering %esi in the E820 call.
That particular BIOS has been fixed, but there is a possibility that
this is responsible for other occasional reports of early boot
failure, and it does not hurt to add %esi to the clobbers.

-stable candidate patch.

Cc: Justin Forbes <jmforbes@linuxtx.org>
Signed-off-by: Michael K Johnson <johnsonm@rpath.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: stable@kernel.org
15 years agox86/dmi: fix dmi_alloc() section mismatches
Jeremy Fitzhardinge [Sun, 22 Mar 2009 21:48:44 +0000 (14:48 -0700)]
x86/dmi: fix dmi_alloc() section mismatches

Impact: section mismatch fix

Ingo reports these warnings:
> WARNING: vmlinux.o(.text+0x6a288e): Section mismatch in reference from
> the function dmi_alloc() to the function .init.text:extend_brk()
> The function dmi_alloc() references
> the function __init extend_brk().
> This is often because dmi_alloc lacks a __init annotation or the
> annotation of extend_brk is wrong.

dmi_alloc() is a static inline, and so should be immune to this
kind of error.  But force it to be inlined and make it __init
anyway, just to be extra sure.

All of dmi_alloc()'s callers are already __init.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <49C6B23C.2040308@goop.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
15 years agoMerge branches 'x86/cleanups', 'x86/mm', 'x86/setup' and 'linus' into x86/core
Ingo Molnar [Fri, 20 Mar 2009 09:34:22 +0000 (10:34 +0100)]
Merge branches 'x86/cleanups', 'x86/mm', 'x86/setup' and 'linus' into x86/core

15 years agoaio: lookup_ioctx can return the wrong value when looking up a bogus context
Jeff Moyer [Thu, 19 Mar 2009 00:04:21 +0000 (17:04 -0700)]
aio: lookup_ioctx can return the wrong value when looking up a bogus context

The libaio test harness turned up a problem whereby lookup_ioctx on a
bogus io context was returning the 1 valid io context from the list
(harness/cases/3.p).

Because of that, an extra put_iocontext was done, and when the process
exited, it hit a BUG_ON in the put_iocontext macro called from exit_aio
(since we expect a users count of 1 and instead get 0).

The problem was introduced by "aio: make the lookup_ioctx() lockless"
(commit abf137dd7712132ee56d5b3143c2ff61a72a5faa).

Thanks to Zach for pointing out that hlist_for_each_entry_rcu will not
return with a NULL tpos at the end of the loop, even if the entry was
not found.

Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Acked-by: Zach Brown <zach.brown@oracle.com>
Acked-by: Jens Axboe <jens.axboe@oracle.com>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agoeventfd: remove fput() call from possible IRQ context
Davide Libenzi [Thu, 19 Mar 2009 00:04:19 +0000 (17:04 -0700)]
eventfd: remove fput() call from possible IRQ context

Remove a source of fput() call from inside IRQ context.  Myself, like Eric,
wasn't able to reproduce an fput() call from IRQ context, but Jeff said he was
able to, with the attached test program.  Independently from this, the bug is
conceptually there, so we might be better off fixing it.  This patch adds an
optimization similar to the one we already do on ->ki_filp, on ->ki_eventfd.
Playing with ->f_count directly is not pretty in general, but the alternative
here would be to add a brand new delayed fput() infrastructure, that I'm not
sure is worth it.

Signed-off-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Cc: Zach Brown <zach.brown@oracle.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agoMove cc-option to below arch-specific setup
Linus Torvalds [Thu, 19 Mar 2009 22:53:19 +0000 (15:53 -0700)]
Move cc-option to below arch-specific setup

Sam Ravnborg says:
 "We have several architectures that plays strange games with $(CC) and
  $(CROSS_COMPILE).

  So we need to postpone any use of $(call cc-option..) until we have
  included the arch specific Makefile so we try with the correct $(CC)
  version."

Requested-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agoMerge branch 'for-linus' of git://git390.marist.edu/pub/scm/linux-2.6
Linus Torvalds [Thu, 19 Mar 2009 21:56:35 +0000 (14:56 -0700)]
Merge branch 'for-linus' of git://git390.marist.edu/linux-2.6

* 'for-linus' of git://git390.marist.edu/pub/scm/linux-2.6:
  [S390] make page table upgrade work again
  [S390] make page table walking more robust
  [S390] Dont check for pfn_valid() in uaccess_pt.c
  [S390] ftrace/mcount: fix kernel stack backchain
  [S390] topology: define SD_MC_INIT to fix performance regression
  [S390] __div64_31 broken for CONFIG_MARCH_G5

15 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid
Linus Torvalds [Thu, 19 Mar 2009 21:50:15 +0000 (14:50 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/jikos/hid

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid:
  HID: fix waitqueue usage in hiddev
  HID: fix incorrect free in hiddev

15 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable
Linus Torvalds [Thu, 19 Mar 2009 21:49:55 +0000 (14:49 -0700)]
Merge git://git./linux/kernel/git/mason/btrfs-unstable

* git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable:
  Btrfs: Clear space_info full when adding new devices
  Btrfs: Fix locking around adding new space_info

15 years agoFix race in create_empty_buffers() vs __set_page_dirty_buffers()
Linus Torvalds [Thu, 19 Mar 2009 18:32:05 +0000 (11:32 -0700)]
Fix race in create_empty_buffers() vs __set_page_dirty_buffers()

Nick Piggin noticed this (very unlikely) race between setting a page
dirty and creating the buffers for it - we need to hold the mapping
private_lock until we've set the page dirty bit in order to make sure
that create_empty_buffers() might not build up a set of buffers without
the dirty bits set when the page is dirty.

I doubt anybody has ever hit this race (and it didn't solve the issue
Nick was looking at), but as Nick says: "Still, it does appear to solve
a real race, which we should close."

Acked-by: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agoAdd '-fwrapv' to gcc CFLAGS
Linus Torvalds [Thu, 19 Mar 2009 18:10:17 +0000 (11:10 -0700)]
Add '-fwrapv' to gcc CFLAGS

This makes sure that gcc doesn't try to optimize away wrapping
arithmetic, which the kernel occasionally uses for overflow testing, ie
things like

if (ptr + offset < ptr)

which technically is undefined for non-unsigned types. See

http://bugzilla.kernel.org/show_bug.cgi?id=12597

for details.

Not all versions of gcc support it, so we need to make it conditional
(it looks like it was introduced in gcc-3.4).

Reminded-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agox86: with the last user gone, remove set_pte_present
Jeremy Fitzhardinge [Wed, 18 Mar 2009 20:03:33 +0000 (13:03 -0700)]
x86: with the last user gone, remove set_pte_present

Impact: cleanup

set_pte_present() is no longer used, directly or indirectly,
so remove it.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Xen-devel <xen-devel@lists.xensource.com>
Cc: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Alok Kataria <akataria@vmware.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Avi Kivity <avi@redhat.com>
LKML-Reference: <1237406613-2929-2-git-send-email-jeremy@goop.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
15 years agox86/32: no need to use set_pte_present in set_pte_vaddr
Jeremy Fitzhardinge [Wed, 18 Mar 2009 20:03:32 +0000 (13:03 -0700)]
x86/32: no need to use set_pte_present in set_pte_vaddr

Impact: cleanup, remove last user of set_pte_present

set_pte_vaddr() is only used to install ptes in fixmaps, and
should never be used to overwrite a present mapping.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Xen-devel <xen-devel@lists.xensource.com>
LKML-Reference: <1237406613-2929-1-git-send-email-jeremy@goop.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
15 years agox86: mpparse: clean up code by introducing a few helper functions, fix
Ingo Molnar [Thu, 19 Mar 2009 07:50:35 +0000 (08:50 +0100)]
x86: mpparse: clean up code by introducing a few helper functions, fix

Impact: fix boot crash

This fixes commit a6830278568a8bb9758aac152db15187741e0113.

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <1237403503.22438.21.camel@ht.satnam>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
15 years agox86, setup: fix the setting of 480-line VGA modes
H. Peter Anvin [Wed, 18 Mar 2009 23:54:05 +0000 (16:54 -0700)]
x86, setup: fix the setting of 480-line VGA modes

Impact: fix rarely-used feature

The VGA Miscellaneous Output Register is read from address 0x3CC but
written to address 0x3C2.  This was missed when this code was
converted from assembly to C.  While we're at it, clean up the code by
making the overflow bits and the math used to set the bits explicit.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
15 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging-2.6
Linus Torvalds [Wed, 18 Mar 2009 16:34:17 +0000 (09:34 -0700)]
Merge git://git./linux/kernel/git/gregkh/staging-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging-2.6:
  Staging: benet: remove driver now that it is merged in drivers/net/

15 years agoMerge branch 'for-2.6.29' of git://linux-nfs.org/~bfields/linux
Linus Torvalds [Wed, 18 Mar 2009 16:27:20 +0000 (09:27 -0700)]
Merge branch 'for-2.6.29' of git://linux-nfs.org/~bfields/linux

* 'for-2.6.29' of git://linux-nfs.org/~bfields/linux:
  nfsd: nfsd should drop CAP_MKNOD for non-root
  NFSD: provide encode routine for OP_OPENATTR

15 years agoStaging: benet: remove driver now that it is merged in drivers/net/
Greg Kroah-Hartman [Wed, 18 Mar 2009 16:22:17 +0000 (09:22 -0700)]
Staging: benet: remove driver now that it is merged in drivers/net/

The benet driver is now in the proper place in drivers/net/benet, so we
can remove the staging version.

Acked-by: Sathya Perla <sathyap@serverengines.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agox86: mpparse: clean up code by introducing a few helper functions
Jaswinder Singh Rajput [Wed, 18 Mar 2009 15:12:28 +0000 (20:42 +0530)]
x86: mpparse: clean up code by introducing a few helper functions

Impact: cleanup

Refactor the MP-table parsing code via the introduction of the
following helper functions:

  skip_entry()
  smp_reserve_bootmem()
  check_irq_src()
  check_slot()

To simplify the code flow and to reduce the size of the
following oversized functions: smp_read_mpc(), smp_scan_config().

There should be no impact to functionality.

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
15 years agoMerge branch 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc
Linus Torvalds [Wed, 18 Mar 2009 16:05:40 +0000 (09:05 -0700)]
Merge branch 'merge' of git://git./linux/kernel/git/benh/powerpc

* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc:
  powerpc/ps3: ps3_defconfig updates
  powerpc/mm: Respect _PAGE_COHERENT on classic ppc32 SW
  powerpc/5200: Enable CPU_FTR_NEED_COHERENT for MPC52xx
  ps3/block: Replace mtd/ps3vram by block/ps3vram

15 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus
Linus Torvalds [Wed, 18 Mar 2009 16:04:25 +0000 (09:04 -0700)]
Merge git://git./linux/kernel/git/rusty/linux-2.6-for-linus

* git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus:
  module: fix refptr allocation and release order

15 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb-2.6
Linus Torvalds [Wed, 18 Mar 2009 16:03:18 +0000 (09:03 -0700)]
Merge git://git./linux/kernel/git/gregkh/usb-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb-2.6:
  USB: storage: Unusual USB device Prolific 2507 variation added
  USB: Add device id for Option GTM380 to option driver
  USB: Add Vendor/Product ID for new CDMA U727 to option driver
  USB: Updated unusual-devs entry for USB mass storage on Nokia 6233
  USB: Option: let cdc-acm handle Sony Ericsson F3507g / Dell 5530
  USB: EHCI: expedite unlinks when the root hub is suspended
  USB: EHCI: Fix isochronous URB leak
  USB: option.c: add ZTE 622 modem device
  USB: wusbcore/wa-xfer, fix lock imbalance
  USB: misc/vstusb, fix lock imbalance
  USB: misc/adutux, fix lock imbalance
  USB: image/mdc800, fix lock imbalance
  USB: atm/cxacru, fix lock imbalance
  USB: unusual_devs: Add support for GI 0431 SD-Card interface
  USB: serial: new cp2101 device id
  USB: serial: ftdi: enable UART detection on gnICE JTAG adaptors blacklist interface0
  USB: serial: add FTDI USB/Serial converter devices
  USB: usbfs: keep async URBs until the device file is closed
  USB: usbtmc: add protocol 1 support
  USB: usbtmc: fix stupid bug in open()

15 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6
Linus Torvalds [Wed, 18 Mar 2009 14:39:11 +0000 (07:39 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/tiwai/sound-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6:
  ALSA: Fix vunmap and free order in snd_free_sgbuf_pages()
  ALSA: mixart, fix lock imbalance
  ALSA: pcm_oss, fix locking typo
  ALSA: oss-mixer - Fixes recording gain control
  ALSA: hda - Workaround for buggy DMA position on ATI controllers
  ALSA: hda - Fix DMA mask for ATI controllers
  ALSA: opl3sa2 - Fix NULL dereference when suspending snd_opl3sa2

15 years ago[S390] make page table upgrade work again
Martin Schwidefsky [Wed, 18 Mar 2009 12:27:37 +0000 (13:27 +0100)]
[S390] make page table upgrade work again

After TASK_SIZE now gives the current size of the address space the
upgrade of a 64 bit process from 3 to 4 levels of page table  needs
to use the arch_mmap_check hook to catch large mmap lengths. The
get_unmapped_area* functions need to check for -ENOMEM from the
arch_get_unmapped_area*, upgrade the page table and retry.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
15 years ago[S390] make page table walking more robust
Martin Schwidefsky [Wed, 18 Mar 2009 12:27:36 +0000 (13:27 +0100)]
[S390] make page table walking more robust

Make page table walking on s390 more robust. The current code requires
that the pgd/pud/pmd/pte loop is only done for address ranges that are
below the end address of the last vma of the address space. But this
is not always true, e.g. the generic page table walker does not guarantee
this. Change TASK_SIZE/TASK_SIZE_OF to reflect the current size of the
address space. This makes the generic page table walker happy but it
breaks the upgrade of a 3 level page table to a 4 level page table.
To make the upgrade work again another fix is required.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
15 years ago[S390] Dont check for pfn_valid() in uaccess_pt.c
Gerald Schaefer [Wed, 18 Mar 2009 12:27:35 +0000 (13:27 +0100)]
[S390] Dont check for pfn_valid() in uaccess_pt.c

pfn_valid() actually checks for a valid struct page and not for a
valid pfn. Using xip mappings w/o struct pages, this will result in
-EFAULT returned by the (page table walk) user copy functions,
even though there is valid memory. Those user copy functions don't
need a struct page, so this patch just removes the pfn_valid() check.

Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
15 years ago[S390] ftrace/mcount: fix kernel stack backchain
Heiko Carstens [Wed, 18 Mar 2009 12:27:34 +0000 (13:27 +0100)]
[S390] ftrace/mcount: fix kernel stack backchain

With packed stack the backchain is at a different location.
Just use __SF_BACKCHAIN as an offset to store the backchain.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
15 years ago[S390] topology: define SD_MC_INIT to fix performance regression
Heiko Carstens [Wed, 18 Mar 2009 12:27:33 +0000 (13:27 +0100)]
[S390] topology: define SD_MC_INIT to fix performance regression

The default values for SD_MC_INIT cause an additional cpu usage of up
to 40% on some network benchmarks compared to the plain SD_CPU_INIT
values. So just define SD_MC_INIT to SD_CPU_INIT.
More tuning needs to be done.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
15 years ago[S390] __div64_31 broken for CONFIG_MARCH_G5
Martin Schwidefsky [Wed, 18 Mar 2009 12:27:32 +0000 (13:27 +0100)]
[S390] __div64_31 broken for CONFIG_MARCH_G5

The implementation of __div64_31 for G5 machines is broken. The comments
in __div64_31 are correct, only the code does not do what the comments
say. The part "If the remainder has overflown subtract base and increase
the quotient" is only partially realized, the base is subtracted correctly
but the quotient is only increased if the dividend had the last bit set.
Using the correct instruction fixes the problem.

Cc: stable@kernel.org
Reported-by: Frans Pop <elendil@planet.nl>
Tested-by: Frans Pop <elendil@planet.nl>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
15 years agox86: kprobes.c fix compilation warning
Jaswinder Singh Rajput [Wed, 18 Mar 2009 12:07:45 +0000 (17:37 +0530)]
x86: kprobes.c fix compilation warning

arch/x86/kernel/kprobes.c:196: warning: passing argument 1 of ‘search_exception_tables’ makes integer from pointer without a cast

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
LKML-Reference:<49BED952.2050809@redhat.com>
LKML-Reference: <1237378065.13488.2.camel@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
15 years agoMerge branches 'x86/cleanups', 'x86/cpu', 'x86/debug', 'x86/mce2', 'x86/mm', 'x86...
Ingo Molnar [Wed, 18 Mar 2009 12:19:49 +0000 (13:19 +0100)]
Merge branches 'x86/cleanups', 'x86/cpu', 'x86/debug', 'x86/mce2', 'x86/mm', 'x86/mtrr', 'x86/setup', 'x86/setup-memory', 'x86/urgent', 'x86/uv', 'x86/x2apic' and 'linus' into x86/core

Conflicts:
arch/parisc/kernel/irq.c

15 years agox86: cpu/mttr/cleanup.c fix compilation warning
Jaswinder Singh Rajput [Wed, 18 Mar 2009 12:06:55 +0000 (17:36 +0530)]
x86: cpu/mttr/cleanup.c fix compilation warning

arch/x86/kernel/cpu/mtrr/cleanup.c:197: warning: format ‘%d’ expects type ‘int’, but argument 2 has type ‘long unsigned int’

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <1237378015.13488.1.camel@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
15 years agox86, uv: fix cpumask iterator in uv_bau_init()
Rusty Russell [Tue, 17 Mar 2009 21:52:30 +0000 (08:22 +1030)]
x86, uv: fix cpumask iterator in uv_bau_init()

Impact: fix boot crash on UV systems

Commit 76ba0ecda0de9accea9a91cb6dbde46782110e1c "cpumask: use
cpumask_var_t in uv_flush_tlb_others" used cur_cpu as an iterator;
it was supposed to be zero for the code below it.

Reported-by: Cliff Wickman <cpw@sgi.com>
Original-From: Cliff Wickman <cpw@sgi.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Mike Travis <travis@sgi.com>
Cc: steiner@sgi.com
Cc: <stable@kernel.org>
LKML-Reference: <200903180822.31196.rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
15 years agox86: add x2apic_wrmsr_fence() to x2apic flush tlb paths
Suresh Siddha [Tue, 17 Mar 2009 18:16:54 +0000 (10:16 -0800)]
x86: add x2apic_wrmsr_fence() to x2apic flush tlb paths

Impact: optimize APIC IPI related barriers

Uncached MMIO accesses for xapic are inherently serializing and hence
we don't need explicit barriers for xapic IPI paths.

x2apic MSR writes/reads don't have serializing semantics and hence need
a serializing instruction or mfence, to make all the previous memory
stores globally visisble before the x2apic msr write for IPI.

Add x2apic_wrmsr_fence() in flush tlb path to x2apic specific paths.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: "steiner@sgi.com" <steiner@sgi.com>
Cc: Nick Piggin <npiggin@suse.de>
LKML-Reference: <1237313814.27006.203.camel@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
15 years agoMerge branch 'fix/opl3sa2-suspend' into for-linus
Takashi Iwai [Wed, 18 Mar 2009 07:04:36 +0000 (08:04 +0100)]
Merge branch 'fix/opl3sa2-suspend' into for-linus

15 years agoMerge branch 'fix/hda' into for-linus
Takashi Iwai [Wed, 18 Mar 2009 07:04:16 +0000 (08:04 +0100)]
Merge branch 'fix/hda' into for-linus

15 years agoALSA: Fix vunmap and free order in snd_free_sgbuf_pages()
Takashi Iwai [Tue, 17 Mar 2009 13:00:06 +0000 (14:00 +0100)]
ALSA: Fix vunmap and free order in snd_free_sgbuf_pages()

In snd_free_sgbuf_pags(), vunmap() is called after releasing the SG
pages, and it causes errors on Xen as Xen manages the pages
differently.  Although no significant errors have been reported on
the actual hardware, this order should be fixed other way round,
first vunmap() then free pages.

Cc: Jan Beulich <jbeulich@novell.com>
Cc: <stable@kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
15 years agoALSA: mixart, fix lock imbalance
Jiri Slaby [Wed, 11 Mar 2009 19:11:41 +0000 (20:11 +0100)]
ALSA: mixart, fix lock imbalance

There is an omitted unlock in one snd_mixart_hw_params fail path. Fix it.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: <stable@kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
15 years agoALSA: pcm_oss, fix locking typo
Jiri Slaby [Wed, 11 Mar 2009 19:11:40 +0000 (20:11 +0100)]
ALSA: pcm_oss, fix locking typo

s/mutex_lock/mutex_unlock/ on 2 fail paths in snd_pcm_oss_proc_write.
Probably a typo, lock should be unlocked when leaving the function.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: <stable@kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
15 years agoALSA: oss-mixer - Fixes recording gain control
Viral Mehta [Tue, 10 Mar 2009 14:43:18 +0000 (15:43 +0100)]
ALSA: oss-mixer - Fixes recording gain control

At the time of initialization, SNDRV_MIXER_OSS_PRESENT_PVOLUME bit is not
set for MIC (slot 7).
So, the same should not be checked when an application tries to do gain
control for audio recording devices.

Just check slot->present for SNDRV_MIXER_OSS_PRESENT_CVOLUME independently.
Verified with a simple application which opens /dev/dsp for recording and
/dev/mixer for volume control.

Have tested two usb audio mic devices.

Signed-off-by: Viral Mehta <viral.mehta@einfochips.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
15 years agoALSA: hda - Workaround for buggy DMA position on ATI controllers
Takashi Iwai [Tue, 17 Mar 2009 06:49:14 +0000 (07:49 +0100)]
ALSA: hda - Workaround for buggy DMA position on ATI controllers

The position-buffer on ATI controllers are unreliable as well as
on VIA chips, thus the same workaround for DMA position reading as
VIA is useful for ATI.

Cc: <stable@kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
15 years agoALSA: hda - Fix DMA mask for ATI controllers
Takashi Iwai [Tue, 17 Mar 2009 06:47:18 +0000 (07:47 +0100)]
ALSA: hda - Fix DMA mask for ATI controllers

ATI controllers (at least some SB0600 models) appear buggy to handle
64bit DMA.  As a workaround, reset GCAP bit0 and let the driver to
use only 32bit DMA on these controllers.

Cc: <stable@kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
15 years agox86: use smp_call_function_single() in arch/x86/kernel/cpu/mcheck/mce_amd_64.c
Andrew Morton [Wed, 18 Mar 2009 00:10:25 +0000 (10:40 +1030)]
x86: use smp_call_function_single() in arch/x86/kernel/cpu/mcheck/mce_amd_64.c

Attempting to rid us of the problematic work_on_cpu().  Just use
smp_call_function_single() here.

Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
LKML-Reference: <20090318042217.EF3F1DDF39@ozlabs.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
15 years agoMerge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4
Linus Torvalds [Wed, 18 Mar 2009 03:55:40 +0000 (20:55 -0700)]
Merge branch 'for_linus' of git://git./linux/kernel/git/tytso/ext4

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
  ext4: fix bb_prealloc_list corruption due to wrong group locking
  ext4: fix bogus BUG_ONs in in mballoc code
  ext4: Print the find_group_flex() warning only once
  ext4: fix header check in ext4_ext_search_right() for deep extent trees.

15 years agopowerpc/ps3: ps3_defconfig updates
Geoff Levand [Fri, 13 Mar 2009 06:52:22 +0000 (06:52 +0000)]
powerpc/ps3: ps3_defconfig updates

Update ps3_defconfig.

Sets these options:

  CONFIG_PS3_VRAM=m
  CONFIG_BLK_DEV_DM=m
  CONFIG_USB_HIDDEV=y
  CONFIG_EXT4_FS=y

Signed-off-by: Geoff Levand <geoffrey.levand@am.sony.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
15 years agoMerge commit 'gcl/merge' into merge
Benjamin Herrenschmidt [Wed, 18 Mar 2009 02:16:30 +0000 (13:16 +1100)]
Merge commit 'gcl/merge' into merge

15 years agox86, dmar: use atomic allocations for QI and Intr-remapping init
Suresh Siddha [Tue, 17 Mar 2009 00:05:05 +0000 (17:05 -0700)]
x86, dmar: use atomic allocations for QI and Intr-remapping init

Impact: invalid use of GFP_KERNEL in interrupt context

Queued invalidation and interrupt-remapping will get initialized with
interrupts disabled (while enabling interrupt-remapping). So use
GFP_ATOMIC instead of GFP_KERNEL for memory alloacations.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
15 years agox86: fix broken irq migration logic while cleaning up multiple vectors
Suresh Siddha [Tue, 17 Mar 2009 00:05:04 +0000 (17:05 -0700)]
x86: fix broken irq migration logic while cleaning up multiple vectors

Impact: fix spurious IRQs

During irq migration, we send a low priority interrupt to the previous
irq destination. This happens in non interrupt-remapping case after interrupt
starts arriving at new destination and in interrupt-remapping case after
modifying and flushing the interrupt-remapping table entry caches.

This low priority irq cleanup handler can cleanup multiple vectors, as
multiple irq's can be migrated at almost the same time. While
there will be multiple invocations of irq cleanup handler (one cleanup
IPI for each irq migration), first invocation of the cleanup handler
can potentially cleanup more than one vector (as the first invocation can
see the requests for more than vector cleanup). When we cleanup multiple
vectors during the first invocation of the smp_irq_move_cleanup_interrupt(),
other vectors that are to be cleanedup can still be pending in the local
cpu's IRR (as smp_irq_move_cleanup_interrupt() runs with interrupts disabled).

When we are ready to unhook a vector corresponding to an irq, check if that
vector is registered in the local cpu's IRR. If so skip that cleanup and
do a self IPI with the cleanup vector, so that we give a chance to
service the pending vector interrupt and then cleanup that vector
allocation once we execute the lowest priority handler.

This fixes spurious interrupts seen when migrating multiple vectors
at the same time.

[ This is apparently possible even on conventional xapic, although to
  the best of our knowledge it has never been seen.  The stable
  maintainers may wish to consider this one for -stable. ]

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Cc: stable@kernel.org
15 years agomodule: fix refptr allocation and release order
Masami Hiramatsu [Mon, 16 Mar 2009 22:13:36 +0000 (18:13 -0400)]
module: fix refptr allocation and release order

Impact: fix ref-after-free crash on failed module load

Fix refptr bug: Change refptr allocation and release order not to access a module
data structure pointed by 'mod' after freeing mod->module_core.
This bug will cause kernel panic(e.g. failed to find undefined symbols).

This bug was reported on systemtap bugzilla.
http://sources.redhat.com/bugzilla/show_bug.cgi?id=9927

Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
15 years agox86, ioapic: Fix non atomic allocation with interrupts disabled
Suresh Siddha [Tue, 17 Mar 2009 00:05:03 +0000 (17:05 -0700)]
x86, ioapic: Fix non atomic allocation with interrupts disabled

Impact: fix possible race

save_mask_IO_APIC_setup() was using non atomic memory allocation while getting
called with interrupts disabled. Fix this by splitting this into two different
function. Allocation part save_IO_APIC_setup() now happens before
disabling interrupts.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
15 years agox86, x2apic: cleanup ifdef CONFIG_INTR_REMAP in io_apic code
Suresh Siddha [Tue, 17 Mar 2009 00:05:02 +0000 (17:05 -0700)]
x86, x2apic: cleanup ifdef CONFIG_INTR_REMAP in io_apic code

Impact: cleanup

Clean up #ifdefs and replace them with helper functions.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
15 years agox86, x2apic: cleanup the IO-APIC level migration with interrupt-remapping
Suresh Siddha [Tue, 17 Mar 2009 00:05:01 +0000 (17:05 -0700)]
x86, x2apic: cleanup the IO-APIC level migration with interrupt-remapping

Impact: simplification

In the current code, for level triggered migration, we need to modify the
io-apic RTE with the update vector information, along with modifying interrupt
remapping table entry(IRTE) with vector and destination. This is to ensure that
remote IRR bit inthe IOAPIC RTE gets cleared when the cpu does EOI.

With this patch, for level triggered, we eliminate the io-apic RTE modification
(with the updated vector information), by using a virtual vector (io-apic pin
number).  Real vector that is used for interrupting cpu will be coming from
the interrupt-remapping table entry. Trigger mode in the IRTE will always be
edge, and the actual level or edge trigger will be setup in the IO-APIC RTE.
So a level triggered interrupt will appear as an edge to the local apic
cpu but still as level to the IO-APIC.

With this change, level irq migration can be done by simply modifying
the interrupt-remapping table entry with out changing the io-apic RTE.
And as the interrupt appears as edge at the cpu, in addition to do the
local apic EOI, we need to do IO-APIC directed EOI to clear the remote
IRR bit in  the IO-APIC RTE.

This simplies the irq migration in the presence of interrupt-remapping.

Idea-by: Rajesh Sankaran <rajesh.sankaran@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
15 years agox86, x2apic: fix clear_local_APIC() in the presence of x2apic
Suresh Siddha [Tue, 17 Mar 2009 00:05:00 +0000 (17:05 -0700)]
x86, x2apic: fix clear_local_APIC() in the presence of x2apic

Impact: cleanup, paranoia

We were not clearing the local APIC in clear_local_APIC() in the
presence of x2apic. Fix it.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
15 years agox86, x2apic: use virtual wire A mode in disable_IO_APIC() with interrupt-remapping
Suresh Siddha [Tue, 17 Mar 2009 00:04:59 +0000 (17:04 -0700)]
x86, x2apic: use virtual wire A mode in disable_IO_APIC() with interrupt-remapping

Impact: make kexec work with x2apic

disable_IO_APIC() gets called during crashdump aswell, which configures the
IO-APIC/LAPIC so that legacy interrupts can be delivered for the kexec'd kernel.

In the presence of interrupt-remapping, we need to change the
interrupt-remapping configuration aswell as modifying IO-APIC for virtual wire
B mode.

To keep things simple during the crash, use virtual wire A mode
(for which we don't need to touch io-apic and interrupt-remapping tables).

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
15 years agox86, intr-remapping: fix free_irte() to clear all the IRTE entries
Suresh Siddha [Tue, 17 Mar 2009 00:04:58 +0000 (17:04 -0700)]
x86, intr-remapping: fix free_irte() to clear all the IRTE entries

Impact: fix interrupt table entry leak

Fix the typo which was not clearing all the interrupt remapping table
entries corresponding to an irq.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
15 years agox86, dmar: start with sane state while enabling dma and interrupt-remapping
Suresh Siddha [Tue, 17 Mar 2009 00:04:57 +0000 (17:04 -0700)]
x86, dmar: start with sane state while enabling dma and interrupt-remapping

Impact: cleanup/sanitization

Start from a sane state while enabling dma and interrupt-remapping, by
clearing the previous recorded faults and disabling previously
enabled queued invalidation and interrupt-remapping.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
15 years agox86, dmar: routines for disabling queued invalidation and intr remapping
Suresh Siddha [Tue, 17 Mar 2009 00:04:56 +0000 (17:04 -0700)]
x86, dmar: routines for disabling queued invalidation and intr remapping

Impact: new interfaces (not yet used)

Routines for disabling queued invalidation and interrupt remapping.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
15 years agox86, x2apic: enable fault handling for intr-remapping
Suresh Siddha [Tue, 17 Mar 2009 00:04:55 +0000 (17:04 -0700)]
x86, x2apic: enable fault handling for intr-remapping

Impact: interface augmentation (not yet used)

Enable fault handling flow for intr-remapping aswell. Fault handling
code now shared by both dma-remapping and intr-remapping.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
15 years agox86, dmar: move page fault handling code to dmar.c
Suresh Siddha [Tue, 17 Mar 2009 00:04:54 +0000 (17:04 -0700)]
x86, dmar: move page fault handling code to dmar.c

Impact: code movement

Move page fault handling code to dmar.c
This will be shared both by DMA-remapping and Intr-remapping code.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
15 years agox86, x2apic: fix lock ordering during IRQ migration
Suresh Siddha [Tue, 17 Mar 2009 00:04:53 +0000 (17:04 -0700)]
x86, x2apic: fix lock ordering during IRQ migration

Impact: fix potential deadlock on x2apic

fix "hard-safe -> hard-unsafe lock order detected" with irq_2_ir_lock

On x2apic enabled system:
   [ INFO: hard-safe -> hard-unsafe lock order detected ]
   2.6.27-03151-g4480f15b #1
   ------------------------------------------------------
   swapper/1 [HC0[0]:SC0[0]:HE0:SE1] is trying to acquire:
    (irq_2_ir_lock){--..}, at: [<ffffffff8038ebc0>] get_irte+0x2f/0x95

   and this task is already holding:
    (&irq_desc_lock_class){+...}, at: [<ffffffff802649ed>] setup_irq+0x67/0x281
   which would create a new lock dependency:
    (&irq_desc_lock_class){+...} -> (irq_2_ir_lock){--..}

   but this new dependency connects a hard-irq-safe lock:
    (&irq_desc_lock_class){+...}
   ... which became hard-irq-safe at:
     [<ffffffffffffffff>] 0xffffffffffffffff

   to a hard-irq-unsafe lock:
    (irq_2_ir_lock){--..}
   ... which became hard-irq-unsafe at:
   ...  [<ffffffff802547b5>] __lock_acquire+0x571/0x706
     [<ffffffff8025499f>] lock_acquire+0x55/0x71
     [<ffffffff8062f2c4>] _spin_lock+0x2c/0x38
     [<ffffffff8038ee50>] alloc_irte+0x8a/0x14b
     [<ffffffff8021f733>] setup_IO_APIC_irq+0x119/0x30e
     [<ffffffff8090860e>] setup_IO_APIC+0x146/0x6e5
     [<ffffffff809058fc>] native_smp_prepare_cpus+0x24e/0x2e9
     [<ffffffff808f982c>] kernel_init+0x5a/0x176
     [<ffffffff8020c289>] child_rip+0xa/0x11
     [<ffffffffffffffff>] 0xffffffffffffffff

Fix this theoretical lock order issue by using spin_lock_irqsave() instead of
spin_lock()

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
15 years agox86, setup: move 32-bit code to .text32
H. Peter Anvin [Tue, 17 Mar 2009 22:26:06 +0000 (15:26 -0700)]
x86, setup: move 32-bit code to .text32

Impact: cleanup

The setup code is mostly 16-bit code, but there is a small stub of
32-bit code at the end.  Move the 32-bit code to a separate segment,
.text32, to avoid scrambling the disassembly.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
15 years agox86-32: move _end to a dummy section
H. Peter Anvin [Tue, 17 Mar 2009 21:14:31 +0000 (14:14 -0700)]
x86-32: move _end to a dummy section

Impact: build fix with CONFIG_RELOCATABLE

Move _end into a dummy section, so that relocs.c will know it is a
relocatable symbol.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Cc: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
15 years agoUSB: storage: Unusual USB device Prolific 2507 variation added
Thomas Bartosik [Mon, 16 Mar 2009 15:04:38 +0000 (16:04 +0100)]
USB: storage: Unusual USB device Prolific 2507 variation added

The "c-enter" USB to Toshiba 1.8" IDE enclosure needs special treatment
to work flawlessly. This patch is absolutely trivial, as the integrated
USB-IDE bridge is already identified to be an "unusual" device, only the
bcdDevice is different (lower) to the bcdDeviceMin already included in
the kernel.
It is a Prolific 2507 bridge.

T:  Bus=02 Lev=01 Prnt=01 Port=02 Cnt=01 Dev#=  4 Spd=480 MxCh= 0
D:  Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs=  1
P:  Vendor=067b ProdID=2507 Rev= 0.01
S:  Manufacturer=Prolific Technology Inc.
S:  Product=ATAPI-6 Bridge Controller
S:  SerialNumber=00000272
C:* #Ifs= 1 Cfg#= 1 Atr=c0 MxPwr=100mA
I:* If#= 0 Alt= 0 #EPs= 2 Cls=08(stor.) Sub=06 Prot=50 Driver=usb-storage
E:  Ad=01(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=82(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms

Signed-off-by: Thomas Bartosik <tbartdev@gmx-topmail.de>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: Add device id for Option GTM380 to option driver
Achilleas Kotsis [Mon, 16 Mar 2009 14:35:02 +0000 (16:35 +0200)]
USB: Add device id for Option GTM380 to option driver

Option GTM380 in Modem mode uses Product ID 0x7201. This has been tested and works
on production systems for over 6 months.

Signed-off-by: Achilleas Kotsis <akots@exponent.gr>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: Add Vendor/Product ID for new CDMA U727 to option driver
Dirk Hohndel [Sun, 15 Mar 2009 03:47:39 +0000 (20:47 -0700)]
USB: Add Vendor/Product ID for new CDMA U727 to option driver

* newer versions of the Novatel Wireless U727 CDMA 3G USB stick
   have a different Product ID (0x5010); adding this ID makes them
   work just fine with the option driver

Signed-off-by: Dirk Hohndel <hohndel@infradead.org>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: Updated unusual-devs entry for USB mass storage on Nokia 6233
Moritz Muehlenhoff [Fri, 13 Mar 2009 23:43:21 +0000 (00:43 +0100)]
USB: Updated unusual-devs entry for USB mass storage on Nokia 6233

Current firmware revision 5.60 still behaves the same,
so update the quirk up a (non-existing) 99.99 revision.

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=493415

Signed-off-by: Moritz Muehlenhoff <jmm@debian.org>
Tested-by: Jan Heitkoetter <devnull@heitkoetter.net>
Cc: stable <stable@kernel.org>
Signed-off-by: Phil Dibowitz <phil@ipom.com>
15 years agoUSB: Option: let cdc-acm handle Sony Ericsson F3507g / Dell 5530
Dan Williams [Thu, 12 Mar 2009 10:53:00 +0000 (06:53 -0400)]
USB: Option: let cdc-acm handle Sony Ericsson F3507g / Dell 5530

The generic cdc-acm driver is now the best one to handle Sony Ericsson
F3507g-based devices (which the Dell 5530 is a rebrand of), now that all
the pieces are in place (ie, cac477e8f1038c41b6f29d3161ce351462ef3df7).
Removing the IDs from option allows cdc-acm to handle the device.

Signed-off-by: Dan Williams <dcbw@redhat.com>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: EHCI: expedite unlinks when the root hub is suspended
Alan Stern [Mon, 16 Mar 2009 18:21:56 +0000 (14:21 -0400)]
USB: EHCI: expedite unlinks when the root hub is suspended

This patch (as1225) fixes a bug in ehci-hcd.  The condition for
whether unlinked QHs can become IDLE should not be that the controller
is halted, but rather that the controller isn't running.  In other
words when the root hub is suspended, the hardware doesn't own any
QHs.

This fixes a problem that can show up during hibernation: If a QH is
only partially unlinked when the root hub is frozen, then when the
root hub is thawed the QH won't be in the IDLE state.  As a result it
can't be used properly for new URB submissions.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Reported-by: Brandon Philips <brandon@ifup.org>
Tested-by: Brandon Philips <brandon@ifup.org>
Acked-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: EHCI: Fix isochronous URB leak
Karsten Wiese [Thu, 26 Feb 2009 00:47:48 +0000 (01:47 +0100)]
USB: EHCI: Fix isochronous URB leak

ehci-hcd uses usb_get_urb() and usb_put_urb() in an unbalanced way causing
isochronous URB's kref.counts incrementing once per usb_submit_urb() call.
The culprit is *usb being set to NULL when usb_put_urb() is called after URB
is given back.
Due to other fixes there is no need for ehci-hcd to deal with usb_get_urb()
nor usb_put_urb() anymore, so patch removes their usages in ehci-hcd.
Patch also makes ehci_to_hcd(ehci)->self.bandwidth_allocated adjust, if a
stream finishes.

Signed-off-by: Karsten Wiese <fzu@wemgehoertderstaat.de>
Cc: David Brownell <david-b@pacbell.net>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: option.c: add ZTE 622 modem device
Albert Pauw [Sun, 1 Mar 2009 08:37:52 +0000 (09:37 +0100)]
USB: option.c: add ZTE 622 modem device

Please consider this small patch for the usb option-card driver.
This patch adds the ZTE 622 usb modem device.

Signed-off-by: Albert Pauw <albert.pauw@gmail.com>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: wusbcore/wa-xfer, fix lock imbalance
Jiri Slaby [Wed, 11 Mar 2009 20:47:40 +0000 (21:47 +0100)]
USB: wusbcore/wa-xfer, fix lock imbalance

Fix locking on one wa_urb_enqueue_b's fail path. There was omitted unlock.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: misc/vstusb, fix lock imbalance
Jiri Slaby [Wed, 11 Mar 2009 20:47:39 +0000 (21:47 +0100)]
USB: misc/vstusb, fix lock imbalance

Make sure we don't leak locked vstdev->lock in vstusb_write. Unlock
properly on one fail path.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: misc/adutux, fix lock imbalance
Jiri Slaby [Wed, 11 Mar 2009 20:47:38 +0000 (21:47 +0100)]
USB: misc/adutux, fix lock imbalance

Don't unlock adutux_mutex when not held.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: image/mdc800, fix lock imbalance
Jiri Slaby [Wed, 11 Mar 2009 20:47:37 +0000 (21:47 +0100)]
USB: image/mdc800, fix lock imbalance

There is an omitted unlock in mdc800_usb_probe's fail path. Add it.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: Henning Zabel <henning@uni-paderborn.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: atm/cxacru, fix lock imbalance
Jiri Slaby [Wed, 11 Mar 2009 20:47:36 +0000 (21:47 +0100)]
USB: atm/cxacru, fix lock imbalance

We do not hold mutex in one place in cxacru_cm, but unlock it on fail path.
Fix this.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: Simon Arlott <cxacru@fire.lp0.eu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: unusual_devs: Add support for GI 0431 SD-Card interface
Jan Dumon [Tue, 10 Mar 2009 16:29:47 +0000 (17:29 +0100)]
USB: unusual_devs: Add support for GI 0431 SD-Card interface

Enable the SD-Card interface on the GI 0431 HSUPA stick from Option.

The unusual_devs.h entry is necessary because the device descriptor is
vendor-specific. That prevents usb-storage from binding to it as an
interface driver.

T:  Bus=07 Lev=01 Prnt=01 Port=01 Cnt=02 Dev#= 15 Spd=480 MxCh= 0
D:  Ver= 2.00 Cls=ff(vend.) Sub=ff Prot=ff MxPS=64 #Cfgs=  1
P:  Vendor=0af0 ProdID=7501 Rev= 0.00
S:  Manufacturer=Option N.V.
S:  Product=Globetrotter HSUPA Modem
C:* #Ifs=11 Cfg#= 1 Atr=a0 MxPwr=500mA
I:* If#= 0 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=hso
E:  Ad=81(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=01(O) Atr=02(Bulk) MxPS= 512 Ivl=4ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
E:  Ad=82(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=4ms
I:* If#= 2 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
E:  Ad=83(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=03(O) Atr=02(Bulk) MxPS= 512 Ivl=4ms
I:* If#= 3 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=hso
E:  Ad=84(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=04(O) Atr=02(Bulk) MxPS= 512 Ivl=4ms
I:* If#= 4 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
E:  Ad=85(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=05(O) Atr=02(Bulk) MxPS= 512 Ivl=4ms
I:* If#= 5 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
E:  Ad=86(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=06(O) Atr=02(Bulk) MxPS= 512 Ivl=4ms
I:* If#= 6 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=hso
E:  Ad=87(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=07(O) Atr=02(Bulk) MxPS= 512 Ivl=4ms
I:* If#= 7 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=hso
E:  Ad=88(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=08(O) Atr=02(Bulk) MxPS= 512 Ivl=4ms
I:* If#= 8 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
E:  Ad=89(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=09(O) Atr=02(Bulk) MxPS= 512 Ivl=4ms
I:* If#= 9 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=ff Driver=hso
E:  Ad=8a(I) Atr=03(Int.) MxPS=  64 Ivl=2ms
E:  Ad=8b(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=0a(O) Atr=02(Bulk) MxPS= 512 Ivl=4ms
I:* If#=10 Alt= 0 #EPs= 2 Cls=08(stor.) Sub=06 Prot=50 Driver=usb-storage
E:  Ad=0b(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=8c(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms

Signed-off-by: Jan Dumon <j.dumon@option.com>
Signed-off-by: Phil Dibowitz <phil@ipom.com>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: serial: new cp2101 device id
Robert M. Kenney [Thu, 26 Feb 2009 19:58:39 +0000 (14:58 -0500)]
USB: serial: new cp2101 device id

From: Robert M. Kenney <rmk@unh.edu>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: serial: ftdi: enable UART detection on gnICE JTAG adaptors blacklist interface0
Michael Hennerich [Fri, 6 Mar 2009 22:07:43 +0000 (14:07 -0800)]
USB: serial: ftdi: enable UART detection on gnICE JTAG adaptors blacklist interface0

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: serial: add FTDI USB/Serial converter devices
Axel Wachtler [Thu, 5 Mar 2009 22:09:22 +0000 (14:09 -0800)]
USB: serial: add FTDI USB/Serial converter devices

Add the following devices to the USB FTDI SIO device table:

 Bus 001 Device 009: ID 03eb:2109 Atmel Corp.
 http://www.atmel.com/dyn/products/tools_card.asp?tool_id=4187

 Bus 001 Device 008: ID 1cf1:0001
 http://www.dresden-elektronik.de/shop/prod75.html

 Bus 001 Device 007: ID 1c1f:0004
 http://www.dresden-elektronik.de/shop/prod64.html

Signed-off-by: Axel Wachtler <axel.wachtler@atmel.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: usbfs: keep async URBs until the device file is closed
Alan Stern [Mon, 9 Mar 2009 17:44:02 +0000 (13:44 -0400)]
USB: usbfs: keep async URBs until the device file is closed

The usbfs driver manages a list of completed asynchronous URBs.  But
it is too eager to free the entries on this list: destroy_async() gets
called whenever an interface is unbound or a device is removed, and it
deallocates the outstanding struct async entries for all URBs on that
interface or device.  This is wrong; the user program should be able
to reap an URB any time after it has completed, regardless of whether
or not the interface is still bound or the device is still present.

This patch (as1222) moves the code for deallocating the completed list
entries from destroy_async() to usbdev_release().  The outstanding
entries won't be freed until the user program has closed the device
file, thereby eliminating any possibility that the remaining URBs
might still be reaped.

This fixes a bug in which a program can hang in the USBDEVFS_REAPURB
ioctl when the device is unplugged.

Reported-and-tested-by: Martin Poupe <martin.poupe@upek.com>
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: usbtmc: add protocol 1 support
Greg Kroah-Hartman [Wed, 11 Mar 2009 20:51:42 +0000 (13:51 -0700)]
USB: usbtmc: add protocol 1 support

The driver already supports the 1 protocol support, so just add it to
the MODULE_DEVICE_TABLE entry so it properly picks up these devices.

Thanks to Jouni Rynö for pointing this out.

Reported-by: Jouni Ryno <Jouni.Ryno@fmi.fi>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agoUSB: usbtmc: fix stupid bug in open()
Greg Kroah-Hartman [Wed, 11 Mar 2009 03:42:55 +0000 (20:42 -0700)]
USB: usbtmc: fix stupid bug in open()

open() will never succeed, as we always return -ENODEV.  Fix this
obvious bug.

Thanks to Jouni Ryno for reporting it.

Reported-by: Jouni Ryno <Jouni.Ryno@fmi.fi>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
15 years agox86/brk: put the brk reservations in their own section
Jeremy Fitzhardinge [Sun, 15 Mar 2009 06:20:47 +0000 (23:20 -0700)]
x86/brk: put the brk reservations in their own section

Impact: disambiguate real .bss variables from .brk storage

Add a .brk section after the .bss section.  This has no effect
on the final vmlinux, but it more clearly distinguishes the space
taken by actual .bss symbols, and the variable space reserved
by .brk users.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
15 years agox86/brk: make the brk reservation symbols inaccessible from C
Jeremy Fitzhardinge [Sun, 15 Mar 2009 06:19:38 +0000 (23:19 -0700)]
x86/brk: make the brk reservation symbols inaccessible from C

Impact: bulletproofing, clarification

The brk reservation symbols are just there to document the amount
of space reserved by brk users in the final vmlinux file.  Their
addresses are irrelevent, and using their addresses will cause
certain havok.  Name them ".brk.NAME", which is a valid asm symbol
but C can't reference it; it also highlights their special
role in the symbol table.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
15 years agonfsd: nfsd should drop CAP_MKNOD for non-root
J. Bruce Fields [Mon, 16 Mar 2009 22:34:20 +0000 (18:34 -0400)]
nfsd: nfsd should drop CAP_MKNOD for non-root

Since creating a device node is normally an operation requiring special
privilege, Igor Zhbanov points out that it is surprising (to say the
least) that a client can, for example, create a device node on a
filesystem exported with root_squash.

So, make sure CAP_MKNOD is among the capabilities dropped when an nfsd
thread handles a request from a non-root user.

Reported-by: Igor Zhbanov <izh1979@gmail.com>
Cc: stable@kernel.org
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
15 years agoNFSD: provide encode routine for OP_OPENATTR
Benny Halevy [Wed, 4 Mar 2009 21:05:35 +0000 (23:05 +0200)]
NFSD: provide encode routine for OP_OPENATTR

Although this operation is unsupported by our implementation
we still need to provide an encode routine for it to
merely encode its (error) status back in the compound reply.

Thanks for Bill Baker at sun.com for testing with the Sun
OpenSolaris' client, finding, and reporting this bug at
Connectathon 2009.

This bug was introduced in 2.6.27

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Cc: stable@kernel.org
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
15 years agox86-32: tighten the bound on additional memory to map
H. Peter Anvin [Tue, 17 Mar 2009 18:38:23 +0000 (11:38 -0700)]
x86-32: tighten the bound on additional memory to map

Impact: Tighten bound to avoid masking errors

The definition of MAPPING_BEYOND_END was excessive; this has a nasty
tendency to mask bugs.  We have learned over time that this kind of
bug hiding can cause some very strange errors.  Therefore, tighten the
bound to only need to map the actual kernel area.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Cc: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Yinghai Lu <yinghai@kernel.org>
15 years agox86-32: remove ALLOCATOR_SLOP from head_32.S
Jeremy Fitzhardinge [Mon, 16 Mar 2009 19:10:07 +0000 (12:10 -0700)]
x86-32: remove ALLOCATOR_SLOP from head_32.S

Impact: cleanup

ALLOCATOR_SLOP is a vestigial remain from when we used the
bootmem allocator to allocate the kernel's linear memory mapping.
Now we directly reserve pages from the e820 mapping, and no
longer require secondary structures to keep track of allocated
pages.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
15 years agox86-32: make sure we map enough to fit linear map pagetables
Jeremy Fitzhardinge [Mon, 16 Mar 2009 19:07:54 +0000 (12:07 -0700)]
x86-32: make sure we map enough to fit linear map pagetables

Impact: crash fix

head_32.S needs to map the kernel itself, and enough space so
that mm/init.c can allocate space from the e820 allocator
for the linear map of low memory.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
15 years agoAvoid 64-bit "switch()" statements on 32-bit architectures
Linus Torvalds [Tue, 17 Mar 2009 17:02:35 +0000 (10:02 -0700)]
Avoid 64-bit "switch()" statements on 32-bit architectures

Commit ee6f779b9e0851e2f7da292a9f58e0095edf615a ("filp->f_pos not
correctly updated in proc_task_readdir") changed the proc code to use
filp->f_pos directly, rather than through a temporary variable.  In the
process, that caused the operations to be done on the full 64 bits, even
though the offset is never that big.

That's all fine and dandy per se, but for some unfathomable reason gcc
generates absolutely horrid code when using 64-bit values in switch()
statements.  To the point of actually calling out to gcc helper
functions like __cmpdi2 rather than just doing the trivial comparisons
directly the way gcc does for normal compares.  At which point we get
link failures, because we really don't want to support that kind of
crazy code.

Fix this by just casting the f_pos value to "unsigned long", which
is plenty big enough for /proc, and avoids the gcc code generation issue.

Reported-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Zhang Le <r0bertz@gentoo.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agoprevent boosting kprobes on exception address
Masami Hiramatsu [Mon, 16 Mar 2009 22:57:22 +0000 (18:57 -0400)]
prevent boosting kprobes on exception address

Don't boost at the addresses which are listed on exception tables,
because major page fault will occur on those addresses.  In that case,
kprobes can not ensure that when instruction buffer can be freed since
some processes will sleep on the buffer.

kprobes-ia64 already has same check.

Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/agk/linux-2.6-dm
Linus Torvalds [Tue, 17 Mar 2009 15:59:33 +0000 (08:59 -0700)]
Merge git://git./linux/kernel/git/agk/linux-2.6-dm

* git://git.kernel.org/pub/scm/linux/kernel/git/agk/linux-2.6-dm:
  dm crypt: wait for endio to complete before destruction
  dm crypt: fix kcryptd_async_done parameter
  dm io: respect BIO_MAX_PAGES limit
  dm table: rework reference counting fix
  dm ioctl: validate name length when renaming

15 years agopowerpc/mm: Respect _PAGE_COHERENT on classic ppc32 SW
Kumar Gala [Tue, 17 Mar 2009 15:17:50 +0000 (09:17 -0600)]
powerpc/mm: Respect _PAGE_COHERENT on classic ppc32 SW

Since we now set _PAGE_COHERENT in the Linux PTE we shouldn't be clearing
it out before we setup the SW TLB.  Today all the SW TLB machines
(603/e300) that we support are non-SMP, however there are some errata on
some devices that cause us to set _PAGE_COHERENT via CPU_FTR_NEED_COHERENT.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
15 years agopowerpc/5200: Enable CPU_FTR_NEED_COHERENT for MPC52xx
Piotr Ziecik [Tue, 17 Mar 2009 15:17:50 +0000 (09:17 -0600)]
powerpc/5200: Enable CPU_FTR_NEED_COHERENT for MPC52xx

BestComm, a DMA engine in MPC52xx SoC, requires snooping when
CPU caches are enabled to work properly.

Adding CPU_FTR_NEED_COHERENT fixes NFS problems on MPC52xx machines
introduced by 'powerpc/mm: Fix handling of _PAGE_COHERENT in BAT setup
code' (sha1: 4c456a67f501b8b15542c7c21c28812bf88f484b).

Signed-off-by: Piotr Ziecik <kosmo@semihalf.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
15 years agoFast TSC calibration: calculate proper frequency error bounds
Linus Torvalds [Tue, 17 Mar 2009 15:13:17 +0000 (08:13 -0700)]
Fast TSC calibration: calculate proper frequency error bounds

In order for ntpd to correctly synchronize the clocks, the frequency of
the system clock must not be off by more than 500 ppm (or, put another
way, 1:2000), or ntpd will end up giving up on trying to synchronize
properly, and ends up reseting the clock in jumps instead.

The fast TSC PIT calibration sometimes failed this test - it was
assuming that the PIT reads always took about one microsecond each (2us
for the two reads to get a 16-bit timer), and that calibrating TSC to
the PIT over 15ms should thus be sufficient to get much closer than
500ppm (max 2us error on both sides giving 4us over 15ms: a 270 ppm
error value).

However, that assumption does not always hold: apparently some hardware
is either very much slower at reading the PIT registers, or there was
other noise causing at least one machine to get 700+ ppm errors.

So instead of using a fixed 15ms timing loop, this changes the fast PIT
calibration to read the TSC delta over the individual PIT timer reads,
and use the result to calculate the error bars on the PIT read timing
properly.  We then successfully calibrate the TSC only if the maximum
error bars fall below 500ppm.

In the process, we also relax the timing to allow up to 25ms for the
calibration, although it can happen much faster depending on hardware.

Reported-and-tested-by: Jesper Krogh <jesper@krogh.cc>
Cc: john stultz <johnstul@us.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agoFix potential fast PIT TSC calibration startup glitch
Linus Torvalds [Tue, 17 Mar 2009 14:58:26 +0000 (07:58 -0700)]
Fix potential fast PIT TSC calibration startup glitch

During bootup, when we reprogram the PIT (programmable interval timer)
to start counting down from 0xffff in order to use it for the fast TSC
calibration, we should also make sure to delay a bit afterwards to allow
the PIT hardware to actually start counting with the new value.

That will happens at the next CLK pulse (1.193182 MHz), so the easiest
way to do that is to just wait at least one microsecond after
programming the new PIT counter value.  We do that by just reading the
counter value back once - which will take about 2us on PC hardware.

Reported-and-tested-by: john stultz <johnstul@us.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agox86: MTRR workaround for system with stange var MTRRs
Yinghai Lu [Mon, 16 Mar 2009 23:33:59 +0000 (16:33 -0700)]
x86: MTRR workaround for system with stange var MTRRs

Impact: don't trim e820 according to wrong mtrr

Ozan reports that his server emits strange warning.
it turns out the BIOS sets the MTRRs incorrectly.

Ignore those strange ranges, and don't trim e820,
just emit one warning about BIOS

Reported-by: Ozan Çağlayan <ozan@pardus.org.tr>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <49BEE1E7.7020706@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>