firefly-linux-kernel-4.4.55.git
13 years agopercpu: fix list_head init bug in __percpu_counter_init()
Masanori ITOH [Tue, 26 Oct 2010 21:21:20 +0000 (14:21 -0700)]
percpu: fix list_head init bug in __percpu_counter_init()

commit 8474b591faf3bb0a1e08a60d21d6baac498f15e4 upstream.

WARNING: at lib/list_debug.c:26 __list_add+0x3f/0x81()
Hardware name: Express5800/B120a [N8400-085]
list_add corruption. next->prev should be prev (ffffffff81a7ea00), but was dead000000200200. (next=ffff88080b872d58).
Modules linked in: aoe ipt_MASQUERADE iptable_nat nf_nat autofs4 sunrpc bridge 8021q garp stp llc ipv6 cpufreq_ondemand acpi_cpufreq freq_table dm_round_robin dm_multipath kvm_intel kvm uinput lpfc scsi_transport_fc igb ioatdma scsi_tgt i2c_i801 i2c_core dca iTCO_wdt iTCO_vendor_support pcspkr shpchp megaraid_sas [last unloaded: aoe]
Pid: 54, comm: events/3 Tainted: G        W  2.6.34-vanilla1 #1
Call Trace:
[<ffffffff8104bd77>] warn_slowpath_common+0x7c/0x94
[<ffffffff8104bde6>] warn_slowpath_fmt+0x41/0x43
[<ffffffff8120fd2e>] __list_add+0x3f/0x81
[<ffffffff81212a12>] __percpu_counter_init+0x59/0x6b
[<ffffffff810d8499>] bdi_init+0x118/0x17e
[<ffffffff811f2c50>] blk_alloc_queue_node+0x79/0x143
[<ffffffff811f2d2b>] blk_alloc_queue+0x11/0x13
[<ffffffffa02a931d>] aoeblk_gdalloc+0x8e/0x1c9 [aoe]
[<ffffffffa02aa655>] aoecmd_sleepwork+0x25/0xa8 [aoe]
[<ffffffff8106186c>] worker_thread+0x1a9/0x237
[<ffffffffa02aa630>] ? aoecmd_sleepwork+0x0/0xa8 [aoe]
[<ffffffff81065827>] ? autoremove_wake_function+0x0/0x39
[<ffffffff810616c3>] ? worker_thread+0x0/0x237
[<ffffffff810653ad>] kthread+0x7f/0x87
[<ffffffff8100aa24>] kernel_thread_helper+0x4/0x10
[<ffffffff8106532e>] ? kthread+0x0/0x87
[<ffffffff8100aa20>] ? kernel_thread_helper+0x0/0x10

It's because there is no initialization code for a list_head contained in
the struct backing_dev_info under CONFIG_HOTPLUG_CPU, and the bug comes up
when block device drivers calling blk_alloc_queue() are used.  In case of
me, I got them by using aoe.

Signed-off-by: Masanori Itoh <itoumsn@nttdata.co.jp>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoolpc_battery: Fix endian neutral breakage for s16 values
Richard A. Smith [Sat, 25 Sep 2010 18:19:26 +0000 (19:19 +0100)]
olpc_battery: Fix endian neutral breakage for s16 values

commit 7cfbb29466633e6ecdc14f76a693c8478c2b22af upstream.

When the driver was updated to be endian neutral (8e9c7716c)
the signed part of the s16 values was lost.  This is because be16_to_cpu()
returns an unsigned value.  This patch casts the values back to a s16
number prior to the the implicit cast up to an int.

Signed-off-by: Richard A. Smith <richard@laptop.org>
Signed-off-by: Daniel Drake <dsd@laptop.org>
Signed-off-by: Anton Vorontsov <cbouatmailru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agohpet: unmap unused I/O space
Jiri Slaby [Tue, 26 Oct 2010 21:22:11 +0000 (14:22 -0700)]
hpet: unmap unused I/O space

commit a56d5318716d120e040294bb258901ba89fb9c90 upstream.

When the initialization code in hpet finds a memory resource and does not
find an IRQ, it does not unmap the memory resource previously mapped.

There are buggy BIOSes which report resources exactly like this and what
is worse the memory region bases point to normal RAM.  This normally would
not matter since the space is not touched.  But when PAT is turned on,
ioremap causes the page to be uncached and sets this bit in page->flags.

Then when the page is about to be used by the allocator, it is reported
as:

BUG: Bad page state in process md5sum  pfn:3ed00
page:ffffea0000dbd800 count:0 mapcount:0 mapping:(null) index:0x0
page flags: 0x20000001000000(uncached)
Pid: 7956, comm: md5sum Not tainted 2.6.34-12-desktop #1
Call Trace:
 [<ffffffff810df851>] bad_page+0xb1/0x100
 [<ffffffff810dfa45>] prep_new_page+0x1a5/0x1c0
 [<ffffffff810dfe01>] get_page_from_freelist+0x3a1/0x640
 [<ffffffff810e01af>] __alloc_pages_nodemask+0x10f/0x6b0
...

In this particular case:

1) HPET returns 3ed00000 as memory region base, but it is not in
reserved ranges reported by the BIOS (excerpt):
 BIOS-e820: 0000000000100000 - 00000000af6cf000 (usable)
 BIOS-e820: 00000000af6cf000 - 00000000afdcf000 (reserved)

2) there is no IRQ resource reported by HPET method. On the other
hand, the Intel HPET specs (1.0a) says (3.2.5.1):
_CRS (
  // Report 1K of memory consumed by this Timer Block
  memory range consumed
  // Optional: only used if BIOS allocates Interrupts [1]
  IRQs consumed
)

[1] For case where Timer Block is configured to consume IRQ0/IRQ8 AND
Legacy 8254/Legacy RTC hardware still exists, the device objects
associated with 8254 & RTC devices should not report IRQ0/IRQ8 as
"consumed resources".

So in theory we should check whether if it is the case and use those
interrupts instead.

Anyway the address reported by the BIOS here is bogus, so non-presence
of IRQ doesn't mean the "optional" part in point 2).

Since I got no reply previously, fix this by simply unmapping the space
when IRQ is not found and memory region was mapped previously.  It would
be probably more safe to walk the resources again and unmap appropriately
depending on type.  But as we now use only ioremap for both 2 memory
resource types, it is not necessarily needed right now.

Addresses https://bugzilla.novell.com/show_bug.cgi?id=629908

Reported-by: Olaf Hering <olaf@aepfle.de>
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Acked-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agohpet: fix unwanted interrupt due to stale irq status bit
Clemens Ladisch [Tue, 26 Oct 2010 21:22:13 +0000 (14:22 -0700)]
hpet: fix unwanted interrupt due to stale irq status bit

commit 96e9694df446d1154ec2f4fdba8908588b9cba38 upstream.

Jaswinder Singh Rajput wrote:
> By executing Documentation/timers/hpet_example.c
>
> for polling, I requested for 3 iterations but it seems iteration work
> for only 2 as first expired time is always very small.
>
> # ./hpet_example poll /dev/hpet 10 3
> -hpet: executing poll
> hpet_poll: info.hi_flags 0x0
> hpet_poll: expired time = 0x13
> hpet_poll: revents = 0x1
> hpet_poll: data 0x1
> hpet_poll: expired time = 0x1868c
> hpet_poll: revents = 0x1
> hpet_poll: data 0x1
> hpet_poll: expired time = 0x18645
> hpet_poll: revents = 0x1
> hpet_poll: data 0x1

Clearing the HPET interrupt enable bit disables interrupt generation
but does not disable the timer, so the interrupt status bit will still
be set when the timer elapses.  If another interrupt arrives before
the timer has been correctly programmed (due to some other device on
the same interrupt line, or CONFIG_DEBUG_SHIRQ), this results in an
extra unwanted interrupt event because the status bit is likely to be
set from comparator matches that happened before the device was opened.

Therefore, we have to ensure that the interrupt status bit is and
stays cleared until we actually program the timer.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Reported-by: Jaswinder Singh Rajput <jaswinderlinux@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: john stultz <johnstul@us.ibm.com>
Cc: Bob Picco <bpicco@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoStaging: line6: fix up some sysfs attribute permissions
Greg Kroah-Hartman [Tue, 16 Nov 2010 19:23:33 +0000 (11:23 -0800)]
Staging: line6: fix up some sysfs attribute permissions

commit 2018845b6a169f75341f8e68ad1089cb6697cf24 and
2018845b6a169f75341f8e68ad1089cb6697cf24 upstream merged together as it
had to be backported by hand.

They should not be writable by any user

Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Markus Grabner <grabner@icg.tugraz.at>
Cc: Mariusz Kozlowski <m.kozlowski@tuxland.pl>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoStaging: asus_oled: fix up my fixup for some sysfs attribute permissions
Greg Kroah-Hartman [Thu, 18 Nov 2010 19:21:04 +0000 (11:21 -0800)]
Staging: asus_oled: fix up my fixup for some sysfs attribute permissions

commit 515b4987ccd097cdf5416530b05fdf9e01afe95a upstream.

They should be writable by root, not readable.
Doh, stupid me with the wrong flags.

Reported-by: Jonathan Cameron <jic23@cam.ac.uk>
Cc: Jakub Schmidtke <sjakub@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoStaging: asus_oled: fix up some sysfs attribute permissions
Greg Kroah-Hartman [Tue, 16 Nov 2010 19:17:01 +0000 (11:17 -0800)]
Staging: asus_oled: fix up some sysfs attribute permissions

commit 590b0b9754bd8928926bae7194b6da7ead9bda3b upstream.

They should not be writable by any user

Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Jakub Schmidtke <sjakub@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agonet: clear heap allocation for ETHTOOL_GRXCLSRLALL
Kees Cook [Thu, 7 Oct 2010 10:03:48 +0000 (10:03 +0000)]
net: clear heap allocation for ETHTOOL_GRXCLSRLALL

commit ae6df5f96a51818d6376da5307d773baeece4014 upstream.

Calling ETHTOOL_GRXCLSRLALL with a large rule_cnt will allocate kernel
heap without clearing it. For the one driver (niu) that implements it,
it will leave the unused portion of heap unchanged and copy the full
contents back to userspace.

Signed-off-by: Kees Cook <kees.cook@canonical.com>
Acked-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agomicroblaze: Fix build with make 3.82
Thomas Backlund [Thu, 21 Oct 2010 10:19:10 +0000 (13:19 +0300)]
microblaze: Fix build with make 3.82

commit b843e4ec01991a386a9e0e9030703524446e03da upstream.

When running make headers_install_all on x86_64 and make 3.82 I hit this:

arch/microblaze/Makefile:80: *** mixed implicit and normal rules.  Stop.
make: *** [headers_install_all] Error 2

So split the rules to satisfy make 3.82.

Signed-off-by: Thomas Backlund <tmb@mandriva.org>
Signed-off-by: Michal Simek <monstr@monstr.eu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoi2c-pca-platform: Change device name of request_irq
Nobuhiro Iwamatsu [Sun, 24 Oct 2010 16:16:57 +0000 (18:16 +0200)]
i2c-pca-platform: Change device name of request_irq

commit 323584436db0cb05286425d4dfd9516fce88487f upstream.

i2c->adap.name shouldn't be used in request_irq.
Instead the driver name "i2c-pca-platform" should be used.

Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Acked-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoirda: Fix heap memory corruption in iriap.c
Samuel Ortiz [Tue, 5 Oct 2010 23:03:12 +0000 (01:03 +0200)]
irda: Fix heap memory corruption in iriap.c

commit 37f9fc452d138dfc4da2ee1ce5ae85094efc3606 upstream.

While parsing the GetValuebyClass command frame, we could potentially write
passed the skb->data pointer.

Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
Signed-off-by: Samuel Ortiz <samuel@sortiz.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoirda: Fix parameter extraction stack overflow
Samuel Ortiz [Sun, 10 Oct 2010 23:17:56 +0000 (01:17 +0200)]
irda: Fix parameter extraction stack overflow

commit efc463eb508798da4243625b08c7396462cabf9f upstream.

Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
Signed-off-by: Samuel Ortiz <samuel@sortiz.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agojme: Fix PHY power-off error
Guo-Fu Tseng [Mon, 18 Oct 2010 14:10:40 +0000 (14:10 +0000)]
jme: Fix PHY power-off error

commit c8a8684d5cfb0f110a962c93586630c0bf91ebc1 upstream.

Adding phy_on in opposition to phy_off.

Signed-off-by: Guo-Fu Tseng <cooldavid@cooldavid.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoblock: check for proper length of iov entries in blk_rq_map_user_iov()
Jens Axboe [Fri, 29 Oct 2010 14:10:18 +0000 (08:10 -0600)]
block: check for proper length of iov entries in blk_rq_map_user_iov()

commit 9284bcf4e335e5f18a8bc7b26461c33ab60d0689 upstream.

Ensure that we pass down properly validated iov segments before
calling into the mapping or copy functions.

Reported-by: Dan Rosenberg <drosenberg@vsecurity.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoblock: take care not to overflow when calculating total iov length
Jens Axboe [Fri, 29 Oct 2010 17:31:42 +0000 (11:31 -0600)]
block: take care not to overflow when calculating total iov length

commit 9f864c80913467312c7b8690e41fb5ebd1b50e92 upstream.

Reported-by: Dan Rosenberg <drosenberg@vsecurity.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoblock: limit vec count in bio_kmalloc() and bio_alloc_map_data()
Jens Axboe [Fri, 29 Oct 2010 17:46:56 +0000 (11:46 -0600)]
block: limit vec count in bio_kmalloc() and bio_alloc_map_data()

commit f3f63c1c28bc861a931fac283b5bc3585efb8967 upstream.

Reported-by: Dan Rosenberg <drosenberg@vsecurity.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoblock: Ensure physical block size is unsigned int
Martin K. Petersen [Wed, 13 Oct 2010 19:18:03 +0000 (21:18 +0200)]
block: Ensure physical block size is unsigned int

commit 892b6f90db81cccb723d5d92f4fddc2d68b206e1 upstream.

Physical block size was declared unsigned int to accomodate the maximum
size reported by READ CAPACITY(16).  Make sure we use the right type in
the related functions.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Acked-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoLinux 2.6.32.26
Greg Kroah-Hartman [Mon, 22 Nov 2010 18:48:20 +0000 (10:48 -0800)]
Linux 2.6.32.26

13 years agosgi-xp: incoming XPC channel messages can come in after the channel's partition struc...
Robin Holt [Tue, 26 Oct 2010 21:21:15 +0000 (14:21 -0700)]
sgi-xp: incoming XPC channel messages can come in after the channel's partition structures have been torn down

commit 09358972bff5ce99de496bbba97c85d417b3c054 upstream.

Under some workloads, some channel messages have been observed being
delayed on the sending side past the point where the receiving side has
been able to tear down its partition structures.

This condition is already detected in xpc_handle_activate_IRQ_uv(), but
that information is not given to xpc_handle_activate_mq_msg_uv().  As a
result, xpc_handle_activate_mq_msg_uv() assumes the structures still exist
and references them, causing a NULL-pointer deref.

Signed-off-by: Robin Holt <holt@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoFix regressions in scsi_internal_device_block
Mike Christie [Wed, 6 Oct 2010 08:10:59 +0000 (03:10 -0500)]
Fix regressions in scsi_internal_device_block

commit 986fe6c7f50974e871b8ab5a800f5310ea25b361 upstream.

Deleting a SCSI device on a blocked fc_remote_port (before
fast_io_fail_tmo fires) results in a hanging thread:

  STACK:
  0 schedule+1108 [0x5cac48]
  1 schedule_timeout+528 [0x5cb7fc]
  2 wait_for_common+266 [0x5ca6be]
  3 blk_execute_rq+160 [0x354054]
  4 scsi_execute+324 [0x3b7ef4]
  5 scsi_execute_req+162 [0x3b80ca]
  6 sd_sync_cache+138 [0x3cf662]
  7 sd_shutdown+138 [0x3cf91a]
  8 sd_remove+112 [0x3cfe4c]
  9 __device_release_driver+124 [0x3a08b8]
10 device_release_driver+60 [0x3a0a5c]
11 bus_remove_device+266 [0x39fa76]
12 device_del+340 [0x39d818]
13 __scsi_remove_device+204 [0x3bcc48]
14 scsi_remove_device+66 [0x3bcc8e]
15 sysfs_schedule_callback_work+50 [0x260d66]
16 worker_thread+622 [0x162326]
17 kthread+160 [0x1680b0]
18 kernel_thread_starter+6 [0x10aaea]

During the delete, the SCSI device is in moved to SDEV_CANCEL.  When
the FC transport class later calls scsi_target_unblock, this has no
effect, since scsi_internal_device_unblock ignores SCSI devics in this
state.

It looks like all these are regressions caused by:
5c10e63c943b4c67561ddc6bf61e01d4141f881f
[SCSI] limit state transitions in scsi_internal_device_unblock

Fix by rejecting offline and cancel in the state transition.

Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
[jejb: Original patch by Christof Schmitt, modified by Mike Christie]
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoFix race when removing SCSI devices
Christof Schmitt [Wed, 6 Oct 2010 11:19:44 +0000 (13:19 +0200)]
Fix race when removing SCSI devices

commit 546ae796bfac6399e30da4b5af2cf7a6d0f8a4ec upstream.

Removing SCSI devices through
echo 1 > /sys/bus/scsi/devices/ ... /delete

while the FC transport class removes the SCSI target can lead to an
oops:

Unable to handle kernel pointer dereference at virtual kernel address 00000000b6815000
Oops: 0011 [#1] PREEMPT SMP DEBUG_PAGEALLOC
Modules linked in: sunrpc qeth_l3 binfmt_misc dm_multipath scsi_dh dm_mod ipv6 qeth ccwgroup [last unloaded: scsi_wait_scan]
CPU: 1 Not tainted 2.6.35.5-45.x.20100924-s390xdefault #1
Process fc_wq_0 (pid: 861, task: 00000000b7331240, ksp: 00000000b735bac0)
Krnl PSW : 0704200180000000 00000000003ff6e4 (__scsi_remove_device+0x24/0xd0)
           R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:0 CC:2 PM:0 EA:3
Krnl GPRS: 0000000000000001 0000000000000000 00000000b6815000 00000000bc24a8c0
           00000000003ff7c8 000000000056dbb8 0000000000000002 0000000000835d80
           ffffffff00000000 0000000000001000 00000000b6815000 00000000bc24a7f0
           00000000b68151a0 00000000b6815000 00000000b735bc20 00000000b735bbf8
Krnl Code: 00000000003ff6d6a7840001            brc 8,3ff6d8
           00000000003ff6daa7fbffd8            aghi %r15,-40
           00000000003ff6dee3e0f0980024        stg %r14,152(%r15)
          >00000000003ff6e4e31021200004        lg %r1,288(%r2)
           00000000003ff6eaa71f0000            cghi    %r1,0
           00000000003ff6eea7a40011            brc 10,3ff710
           00000000003ff6f2a7390003            lghi    %r3,3
           00000000003ff6f6c0e5ffffc8b1        brasl %r14,3f8858
Call Trace:
([<0000000000001000>] 0x1000)
 [<00000000003ff7d2>] scsi_remove_device+0x42/0x54
 [<00000000003ff8ba>] __scsi_remove_target+0xca/0xfc
 [<00000000003ff99a>] __remove_child+0x3a/0x48
 [<00000000003e3246>] device_for_each_child+0x72/0xbc
 [<00000000003ff93a>] scsi_remove_target+0x4e/0x74
 [<0000000000406586>] fc_rport_final_delete+0xb2/0x23c
 [<000000000015d080>] worker_thread+0x200/0x344
 [<000000000016330c>] kthread+0xa0/0xa8
 [<0000000000106c1a>] kernel_thread_starter+0x6/0xc
 [<0000000000106c14>] kernel_thread_starter+0x0/0xc
INFO: lockdep is turned off.
Last Breaking-Event-Address:
 [<00000000003ff7cc>] scsi_remove_device+0x3c/0x54

The function __scsi_remove_target iterates through the SCSI devices on
the host, but it drops the host_lock before calling
scsi_remove_device. When the SCSI device is deleted from another
thread, the pointer to the SCSI device in scsi_remove_device can
become invalid. Fix this by getting a reference to the SCSI device
before dropping the host_lock to keep the SCSI device alive for the
call to scsi_remove_device.

Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agogdth: integer overflow in ioctl
Dan Carpenter [Fri, 8 Oct 2010 07:03:07 +0000 (09:03 +0200)]
gdth: integer overflow in ioctl

commit f63ae56e4e97fb12053590e41a4fa59e7daa74a4 upstream.

gdth_ioctl_alloc() takes the size variable as an int.
copy_from_user() takes the size variable as an unsigned long.
gen.data_len and gen.sense_len are unsigned longs.
On x86_64 longs are 64 bit and ints are 32 bit.

We could pass in a very large number and the allocation would truncate
the size to 32 bits and allocate a small buffer.  Then when we do the
copy_from_user(), it would result in a memory corruption.

Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agolibsas: fix NCQ mixing with non-NCQ
David Milburn [Fri, 3 Sep 2010 22:13:03 +0000 (17:13 -0500)]
libsas: fix NCQ mixing with non-NCQ

commit f0ad30d3d2dc924decc0e10b1ff6dc32525a5d99 upstream.

Some cards (like mvsas) have issue troubles if non-NCQ commands are
mixed with NCQ ones.  Fix this by using the libata default NCQ check
routine which waits until all NCQ commands are complete before issuing
a non-NCQ one.  The impact to cards (like aic94xx) which don't need
this logic should be minimal

Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agosd name space exhaustion causes system hang
Michael Reed [Mon, 20 Sep 2010 16:20:22 +0000 (11:20 -0500)]
sd name space exhaustion causes system hang

commit 1a03ae0f556a931aa3747b70e44b78308f5b0590 upstream.

Following a site power outage which re-enabled all the ports on my FC
switches, my system subsequently booted with far too many luns!  I had
let it run hoping it would make multi-user.  It didn't.  :(  It hung solid
after exhausting the last sd device, sdzzz, and attempting to create sdaaaa
and beyond.  I was unable to get a dump.

Discovered using a 2.6.32.13 based system.

correct this by detecting when the last index is utilized and failing
the sd probe of the device.  Patch applies to scsi-misc-2.6.

Signed-off-by: Michael Reed <mdr@sgi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: accept some invalid ep0-maxpacket values
Alan Stern [Thu, 14 Oct 2010 19:25:21 +0000 (15:25 -0400)]
USB: accept some invalid ep0-maxpacket values

commit 56626a72a47bf3e50875d960d6b5f17b9bee0ab2 upstream.

A few devices (such as the RCA VR5220 voice recorder) are so
non-compliant with the USB spec that they have invalid maxpacket sizes
for endpoint 0.  Nevertheless, as long as we can safely use them, we
may as well do so.

This patch (as1432) softens our acceptance criterion by allowing
high-speed devices to have ep0-maxpacket sizes other than 64.  A
warning is printed in the system log when this happens, and the
existing error message is clarified.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Reported-by: James <bjlockie@lockie.ca>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: opticon: Fix long-standing bugs in opticon driver
Alon Ziv [Sun, 10 Oct 2010 06:32:18 +0000 (08:32 +0200)]
USB: opticon: Fix long-standing bugs in opticon driver

commit 97cd8dc4ca9a1a5efb2cc38758e01492e3b013e2 upstream.

The bulk-read callback had two bugs:
a) The bulk-in packet's leading two zeros were returned (and the two last
   bytes truncated)
b) The wrong URB was transmitted for the second (and later) read requests,
   causing further reads to return the entire packet (including leading
   zeros)

Signed-off-by: Alon Ziv <alon-git@nolaviz.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: disable endpoints after unbinding interfaces, not before
Alan Stern [Thu, 30 Sep 2010 19:16:23 +0000 (15:16 -0400)]
USB: disable endpoints after unbinding interfaces, not before

commit 80f0cf3947889014d3a3dc0ad60fb87cfda4b12a upstream.

This patch (as1430) fixes a bug in usbcore.  When a device
configuration change occurs or a device is removed, the endpoints for
the old config should be completely disabled.  However it turns out
they aren't; this is because usb_unbind_interface() calls
usb_enable_interface() or usb_set_interface() to put interfaces back
in altsetting 0, which re-enables the interfaces' endpoints.

As a result, when a device goes through a config change or is
unconfigured, the ep_in[] and ep_out[] arrays may be left holding old
pointers to usb_host_endpoint structures.  If the device is
deauthorized these structures get freed, and the stale pointers cause
errors when the the device is eventually unplugged.

The solution is to disable the endpoints after unbinding the
interfaces instead of before.  This isn't as large a change as it
sounds, since usb_unbind_interface() disables all the interface's
endpoints anyway before calling the driver's disconnect routine,
unless the driver claims to support "soft" unbind.

This fixes Bugzilla #19192.  Thanks to "Tom" Lei Ming for diagnosing
the underlying cause of the problem.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Tested-by: Carsten Sommer <carsten_sommer@ymail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: atmel_usba_udc: force vbus_pin at -EINVAL when gpio_request failled
Jean-Christophe PLAGNIOL-VILLARD [Mon, 20 Sep 2010 16:31:07 +0000 (18:31 +0200)]
USB: atmel_usba_udc: force vbus_pin at -EINVAL when gpio_request failled

commit 969affff54702785330de553b790372e261e93f9 upstream.

to ensure gpio_is_valid return false

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: cp210x: Add WAGO 750-923 Service Cable device ID
Anders Larsen [Wed, 6 Oct 2010 21:46:25 +0000 (23:46 +0200)]
USB: cp210x: Add WAGO 750-923 Service Cable device ID

commit 93ad03d60b5b18897030038234aa2ebae8234748 upstream.

The WAGO 750-923 USB Service Cable is used for configuration and firmware
updates of several industrial automation products from WAGO Kontakttechnik GmbH.

Bus 004 Device 002: ID 1be3:07a6
Device Descriptor:
  bLength                18
  bDescriptorType         1
  bcdUSB               1.10
  bDeviceClass            0 (Defined at Interface level)
  bDeviceSubClass         0
  bDeviceProtocol         0
  bMaxPacketSize0        64
  idVendor           0x1be3
  idProduct          0x07a6
  bcdDevice            1.00
  iManufacturer           1 Silicon Labs
  iProduct                2 WAGO USB Service Cable
  iSerial                 3 1277796751
  . . .

Signed-off-by: Anders Larsen <al@alarsen.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: cp210x: Add Renesas RX-Stick device ID
DJ Delorie [Fri, 17 Sep 2010 15:09:06 +0000 (11:09 -0400)]
USB: cp210x: Add Renesas RX-Stick device ID

commit 2f1136d1d08a63dcdbcd462621373f30d8dfe590 upstream.

RX610 development board by Renesas

Bus 001 Device 024: ID 045b:0053 Hitachi, Ltd
Device Descriptor:
  bLength                18
  bDescriptorType         1
  bcdUSB               1.10
  bDeviceClass            0 (Defined at Interface level)
  bDeviceSubClass         0
  bDeviceProtocol         0
  bMaxPacketSize0        64
  idVendor           0x045b Hitachi, Ltd
  idProduct          0x0053
  bcdDevice            1.00
  iManufacturer           1 Silicon Labs
  iProduct                2 RX-Stick
  iSerial                 3 0001
  . . .

http://am.renesas.com/rx610stick

Signed-off-by: DJ Delorie <dj@delorie.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: option: Add more ZTE modem USB id's
Mauro Carvalho Chehab [Sun, 12 Sep 2010 14:41:50 +0000 (11:41 -0300)]
USB: option: Add more ZTE modem USB id's

commit ecfa153ef616b901e86d9a051b329fcda7a6ce7b upstream.

There are lots of ZTE USB id's currently not covered by usb/serial. Adds them,
to allow those devices to work properly on Linux.

While here, put the USB ID's for 0x2002/0x2003 at the sorted order.

This patch is based on zte.c file found on MF645.

PS.: The ZTE driver is commenting the USB ID for 0x0053. It also adds, commented,
an USB ID for 0x0026.

Not sure why, but I think that 0053 is used by their devices in storage mode only.
So, I opted to keep the comment on this patch.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb: musb: blackfin: call gpio_free() on error path in musb_platform_init()
Sergei Shtylyov [Wed, 29 Sep 2010 06:54:31 +0000 (09:54 +0300)]
usb: musb: blackfin: call gpio_free() on error path in musb_platform_init()

commit 00be545e49d83485d49a598d3b7e090088934be8 upstream.

Blackfin's musb_platform_init() needs to call gpio_free() for error cleanup iff
otg_get_transceiver() call returns NULL.

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: ftdi_sio: add device ids for ScienceScope
Greg Kroah-Hartman [Tue, 19 Oct 2010 16:05:43 +0000 (09:05 -0700)]
USB: ftdi_sio: add device ids for ScienceScope

commit 0f266abd70cd83571eca019f764b5f1992da7361 upstream.

This adds the requested device ids to the ftdi_sio driver.

Reported-by: Ewan Bingham <ewan@auc.co.uk>
Cc: Kuba Ober <kuba@mareimbrium.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: ftdi_sio: new VID/PIDs for various Papouch devices
Daniel Suchy [Tue, 12 Oct 2010 13:44:24 +0000 (15:44 +0200)]
USB: ftdi_sio: new VID/PIDs for various Papouch devices

commit 59c6ccd9f9aecfa59c99ceba6d4d34b180547a05 upstream.

This patch for FTDI USB serial driver ads new VID/PIDs used on various
devices manufactured by Papouch (http://www.papouch.com). These devices
have their own VID/PID, although they're using standard FTDI chip. In
ftdi_sio.c, I also made small cleanup to have declarations for all
Papouch devices together.

Signed-off-by: Daniel Suchy <danny@danysek.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: add PID for FTDI based OpenDCC hardware
Rainer Keller [Tue, 28 Sep 2010 10:27:43 +0000 (12:27 +0200)]
USB: add PID for FTDI based OpenDCC hardware

commit 99c1e4f89d1033444ce4d0c064bd2826e81c3775 upstream.

The OpenDCC project is developing a new hardware. This patch adds its
PID to the list of known FTDI devices. The PID can be found at
http://www.opendcc.de/elektronik/usb/opendcc_usb.html

Signed-off-by: Rainer Keller <mail@rainerkeller.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: ftdi_sio: Add PID for accesio products
Rich Mattes [Tue, 14 Sep 2010 04:35:40 +0000 (00:35 -0400)]
USB: ftdi_sio: Add PID for accesio products

commit 3126d8236ca6f68eb8292c6af22c2e59afbeef24 upstream.

Adds support for Accesio USB to Serial adapters, which are built around
FTDI FT232 UARTs.  Tested with the Accesio USB-COM-4SM.

Signed-off-by: Rich Mattes <richmattes@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agodrivers/net/wireless/p54/eeprom.c: Return -ENOMEM on memory allocation failure
Julia Lawall [Fri, 15 Oct 2010 13:00:06 +0000 (15:00 +0200)]
drivers/net/wireless/p54/eeprom.c: Return -ENOMEM on memory allocation failure

commit 0d91f22b75347d9503b17a42b6c74d3f7750acd6 upstream.

In this code, 0 is returned on memory allocation failure, even though other
failures return -ENOMEM or other similar values.

A simplified version of the semantic match that finds this problem is as
follows: (http://coccinelle.lip6.fr/)

// <smpl>
@@
expression ret;
expression x,e1,e2,e3;
@@

ret = 0
... when != ret = e1
*x = \(kmalloc\|kcalloc\|kzalloc\)(...)
... when != ret = e2
if (x == NULL) { ... when != ret = e3
  return ret;
}
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>
Acked-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agop54usb: add five more USBIDs
Christian Lamparter [Fri, 1 Oct 2010 20:01:24 +0000 (22:01 +0200)]
p54usb: add five more USBIDs

commit 1a92795dac419128eb511dce30a6aad672064b88 upstream.

Source:
http://www.wikidevi.com/wiki/Intersil/p54/usb/windows

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agop54usb: fix off-by-one on !CONFIG_PM
Christian Lamparter [Sun, 22 Aug 2010 20:41:33 +0000 (22:41 +0200)]
p54usb: fix off-by-one on !CONFIG_PM

commit 11791a6f7534906b4a01ffb54ba0b02ca39398ef upstream.

The ISL3887 chip needs a USB reset, whenever the
usb-frontend module "p54usb" is reloaded.

This patch fixes an off-by-one bug, if the user
is running a kernel without the CONFIG_PM option
set and for some reason (e.g.: compat-wireless)
wants to switch between different p54usb modules.

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agopipe: fix failure to return error code on ->confirm()
Nicolas Kaiser [Thu, 21 Oct 2010 12:56:00 +0000 (14:56 +0200)]
pipe: fix failure to return error code on ->confirm()

commit e5953cbdff26f7cbae7eff30cd9b18c4e19b7594 upstream.

The arguments were transposed, we want to assign the error code to
'ret', which is being returned.

Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoKVM: Fix fs/gs reload oops with invalid ldt
Avi Kivity [Thu, 28 Oct 2010 18:48:16 +0000 (16:48 -0200)]
KVM: Fix fs/gs reload oops with invalid ldt

commit 9581d442b9058d3699b4be568b6e5eae38a41493 upstream

kvm reloads the host's fs and gs blindly, however the underlying segment
descriptors may be invalid due to the user modifying the ldt after loading
them.

Fix by using the safe accessors (loadsegment() and load_gs_index()) instead
of home grown unsafe versions.

This is CVE-2010-3698.

Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoKVM: x86: Move TSC reset out of vmcb_init
Zachary Amsden [Thu, 28 Oct 2010 18:48:15 +0000 (16:48 -0200)]
KVM: x86: Move TSC reset out of vmcb_init

commit 47008cd887c1836bcadda123ba73e1863de7a6c4 upstream.

The VMCB is reset whenever we receive a startup IPI, so Linux is setting
TSC back to zero happens very late in the boot process and destabilizing
the TSC.  Instead, just set TSC to zero once at VCPU creation time.

Why the separate patch?  So git-bisect is your friend.

Signed-off-by: Zachary Amsden <zamsden@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoKVM: x86: Fix SVM VMCB reset
Zachary Amsden [Thu, 28 Oct 2010 18:48:14 +0000 (16:48 -0200)]
KVM: x86: Fix SVM VMCB reset

commit 58877679fd393d3ef71aa383031ac7817561463d upstream.

On reset, VMCB TSC should be set to zero.  Instead, code was setting
tsc_offset to zero, which passes through the underlying TSC.

Signed-off-by: Zachary Amsden <zamsden@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoKVM: SVM: Adjust tsc_offset only if tsc_unstable
Joerg Roedel [Thu, 28 Oct 2010 18:48:13 +0000 (16:48 -0200)]
KVM: SVM: Adjust tsc_offset only if tsc_unstable

commit 953899b659adce62cbe83d6a7527550ab8797c48 upstream.

The tsc_offset adjustment in svm_vcpu_load is executed
unconditionally even if Linux considers the host tsc as
stable. This causes a Linux guest detecting an unstable tsc
in any case.
This patch removes the tsc_offset adjustment if the host tsc
is stable. The guest will now get the benefit of a stable
tsc too.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoKVM: VMX: Fix host GDT.LIMIT corruption
Avi Kivity [Thu, 28 Oct 2010 18:48:12 +0000 (16:48 -0200)]
KVM: VMX: Fix host GDT.LIMIT corruption

commit 3444d7da1839b851eefedd372978d8a982316c36 upstream.

vmx does not restore GDT.LIMIT to the host value, instead it sets it to 64KB.
This means host userspace can learn a few bits of host memory.

Fix by reloading GDTR when we load other host state.

Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoKVM: MMU: fix conflict access permissions in direct sp
Xiao Guangrong [Thu, 28 Oct 2010 18:48:11 +0000 (16:48 -0200)]
KVM: MMU: fix conflict access permissions in direct sp

commit 5fd5387c89ec99ff6cb82d2477ffeb7211b781c2 upstream.

In no-direct mapping, we mark sp is 'direct' when we mapping the
guest's larger page, but its access is encoded form upper page-struct
entire not include the last mapping, it will cause access conflict.

For example, have this mapping:
        [W]
      / PDE1 -> |---|
  P[W]          |   | LPA
      \ PDE2 -> |---|
        [R]

P have two children, PDE1 and PDE2, both PDE1 and PDE2 mapping the
same lage page(LPA). The P's access is WR, PDE1's access is WR,
PDE2's access is RO(just consider read-write permissions here)

When guest access PDE1, we will create a direct sp for LPA, the sp's
access is from P, is W, then we will mark the ptes is W in this sp.

Then, guest access PDE2, we will find LPA's shadow page, is the same as
PDE's, and mark the ptes is RO.

So, if guest access PDE1, the incorrect #PF is occured.

Fixed by encode the last mapping access into direct shadow page

Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoKVM: MMU: fix direct sps access corrupted
Xiao Guangrong [Thu, 28 Oct 2010 18:48:10 +0000 (16:48 -0200)]
KVM: MMU: fix direct sps access corrupted

commit 9e7b0e7fba45ca3c6357aeb7091ebc281f1de365 upstream.

If the mapping is writable but the dirty flag is not set, we will find
the read-only direct sp and setup the mapping, then if the write #PF
occur, we will mark this mapping writable in the read-only direct sp,
now, other real read-only mapping will happily write it without #PF.

It may hurt guest's COW

Fixed by re-install the mapping when write #PF occur.

Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoKVM: SVM: Fix wrong intercept masks on 32 bit
Joerg Roedel [Thu, 28 Oct 2010 18:48:09 +0000 (16:48 -0200)]
KVM: SVM: Fix wrong intercept masks on 32 bit

commit 061e2fd16863009c8005b4b5fdfb75c7215c0b99 upstream.

This patch makes KVM on 32 bit SVM working again by
correcting the masks used for iret interception. With the
wrong masks the upper 32 bits of the intercepts are masked
out which leaves vmrun unintercepted. This is not legal on
svm and the vmrun fails.
Bug was introduced by commits 95ba827313 and 3cfc3092.

Cc: Jan Kiszka <jan.kiszka@siemens.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agox86, kdump: Change copy_oldmem_page() to use cached addressing
Cliff Wickman [Wed, 8 Sep 2010 15:14:27 +0000 (10:14 -0500)]
x86, kdump: Change copy_oldmem_page() to use cached addressing

commit 37a2f9f30a360fb03522d15c85c78265ccd80287 upstream.

The copy of /proc/vmcore to a user buffer proceeds much faster
if the kernel addresses memory as cached.

With this patch we have seen an increase in transfer rate from
less than 15MB/s to 80-460MB/s, depending on size of the
transfer. This makes a big difference in time needed to save a
system dump.

Signed-off-by: Cliff Wickman <cpw@sgi.com>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: kexec@lists.infradead.org
LKML-Reference: <E1OtMLz-0001yp-Ia@eag09.americas.sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agox86, intr-remap: Set redirection hint in the IRTE
Suresh Siddha [Fri, 27 Aug 2010 18:09:48 +0000 (11:09 -0700)]
x86, intr-remap: Set redirection hint in the IRTE

commit 75e3cfbed6f71a8f151dc6e413b6ce3c390030cb upstream.

Currently the redirection hint in the interrupt-remapping table entry
is set to 0, which means the remapped interrupt is directed to the
processors listed in the destination. So in logical flat mode
in the presence of intr-remapping, this results in a single
interrupt multi-casted to multiple cpu's as specified by the destination
bit mask. But what we really want is to send that interrupt to one of the cpus
based on the lowest priority delivery mode.

Set the redirection hint in the IRTE to '1' to indicate that we want
the remapped interrupt to be directed to only one of the processors
listed in the destination.

This fixes the issue of same interrupt getting delivered to multiple cpu's
in the logical flat mode in the presence of interrupt-remapping. While
there is no functional issue observed with this behavior, this will
impact performance of such configurations (<=8 cpu's using logical flat
mode in the presence of interrupt-remapping)

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <20100827181049.013051492@sbsiddha-MOBL3.sc.intel.com>
Cc: Weidong Han <weidong.han@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agox86, mtrr: Assume SYS_CFG[Tom2ForceMemTypeWB] exists on all future AMD CPUs
Andreas Herrmann [Thu, 30 Sep 2010 12:32:35 +0000 (14:32 +0200)]
x86, mtrr: Assume SYS_CFG[Tom2ForceMemTypeWB] exists on all future AMD CPUs

commit 3fdbf004c1706480a7c7fac3c9d836fa6df20d7d upstream.

Instead of adapting the CPU family check in amd_special_default_mtrr()
for each new CPU family assume that all new AMD CPUs support the
necessary bits in SYS_CFG MSR.

Tom2Enabled is architectural (defined in APM Vol.2).
Tom2ForceMemTypeWB is defined in all BKDGs starting with K8 NPT.
In pre K8-NPT BKDG this bit is reserved (read as zero).

W/o this adaption Linux would unnecessarily complain about bad MTRR
settings on every new AMD CPU family, e.g.

[    0.000000] WARNING: BIOS bug: CPU MTRRs don't cover all of memory, losing 4863MB of RAM.

Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com>
LKML-Reference: <20100930123235.GB20545@loge.amd.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agox86, olpc: Don't retry EC commands forever
Paul Fox [Fri, 1 Oct 2010 17:17:19 +0000 (18:17 +0100)]
x86, olpc: Don't retry EC commands forever

commit 286e5b97eb22baab9d9a41ca76c6b933a484252c upstream.

Avoids a potential infinite loop.

It was observed once, during an EC hacking/debugging
session - not in regular operation.

Signed-off-by: Daniel Drake <dsd@laptop.org>
Cc: dilinger@queued.net
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agox86, kexec: Make sure to stop all CPUs before exiting the kernel
Alok Kataria [Mon, 11 Oct 2010 21:37:08 +0000 (14:37 -0700)]
x86, kexec: Make sure to stop all CPUs before exiting the kernel

commit 76fac077db6b34e2c6383a7b4f3f4f7b7d06d8ce upstream.

x86 smp_ops now has a new op, stop_other_cpus which takes a parameter
"wait" this allows the caller to specify if it wants to stop until all
the cpus have processed the stop IPI.  This is required specifically
for the kexec case where we should wait for all the cpus to be stopped
before starting the new kernel.  We now wait for the cpus to stop in
all cases except for panic/kdump where we expect things to be broken
and we are doing our best to make things work anyway.

This patch fixes a legitimate regression, which was introduced during
2.6.30, by commit id 4ef702c10b5df18ab04921fc252c26421d4d6c75.

Signed-off-by: Alok N Kataria <akataria@vmware.com>
LKML-Reference: <1286833028.1372.20.camel@ank32.eng.vmware.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agox86, cpu: Fix renamed, not-yet-shipping AMD CPUID feature bit
Andre Przywara [Mon, 6 Sep 2010 13:14:17 +0000 (15:14 +0200)]
x86, cpu: Fix renamed, not-yet-shipping AMD CPUID feature bit

commit 7ef8aa72ab176e0288f363d1247079732c5d5792 upstream.

The AMD SSE5 feature set as-it has been replaced by some extensions
to the AVX instruction set. Thus the bit formerly advertised as SSE5
is re-used for one of these extensions (XOP).
Although this changes the /proc/cpuinfo output, it is not user visible, as
there are no CPUs (yet) having this feature.
To avoid confusion this should be added to the stable series, too.

Signed-off-by: Andre Przywara <andre.przywara@amd.com>
LKML-Reference: <1283778860-26843-2-git-send-email-andre.przywara@amd.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agomm, x86: Saving vmcore with non-lazy freeing of vmas
Cliff Wickman [Thu, 16 Sep 2010 16:44:02 +0000 (11:44 -0500)]
mm, x86: Saving vmcore with non-lazy freeing of vmas

commit 3ee48b6af49cf534ca2f481ecc484b156a41451d upstream.

During the reading of /proc/vmcore the kernel is doing
ioremap()/iounmap() repeatedly. And the buildup of un-flushed
vm_area_struct's is causing a great deal of overhead. (rb_next()
is chewing up most of that time).

This solution is to provide function set_iounmap_nonlazy(). It
causes a subsequent call to iounmap() to immediately purge the
vma area (with try_purge_vmap_area_lazy()).

With this patch we have seen the time for writing a 250MB
compressed dump drop from 71 seconds to 44 seconds.

Signed-off-by: Cliff Wickman <cpw@sgi.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: kexec@lists.infradead.org
LKML-Reference: <E1OwHZ4-0005WK-Tw@eag09.americas.sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agofutex: Fix errors in nested key ref-counting
Darren Hart [Sun, 17 Oct 2010 15:35:04 +0000 (08:35 -0700)]
futex: Fix errors in nested key ref-counting

commit 7ada876a8703f23befbb20a7465a702ee39b1704 upstream.

futex_wait() is leaking key references due to futex_wait_setup()
acquiring an additional reference via the queue_lock() routine. The
nested key ref-counting has been masking bugs and complicating code
analysis. queue_lock() is only called with a previously ref-counted
key, so remove the additional ref-counting from the queue_(un)lock()
functions.

Also futex_wait_requeue_pi() drops one key reference too many in
unqueue_me_pi(). Remove the key reference handling from
unqueue_me_pi(). This was paired with a queue_lock() in
futex_lock_pi(), so the count remains unchanged.

Document remaining nested key ref-counting sites.

Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Reported-and-tested-by: Matthieu Fertré<matthieu.fertre@kerlabs.com>
Reported-by: Louis Rilling<louis.rilling@kerlabs.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: John Kacur <jkacur@redhat.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
LKML-Reference: <4CBB17A8.70401@linux.intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agobluetooth: Fix missing NULL check
Alan Cox [Fri, 22 Oct 2010 13:11:26 +0000 (14:11 +0100)]
bluetooth: Fix missing NULL check

commit c19483cc5e56ac5e22dd19cf25ba210ab1537773 upstream.

Fortunately this is only exploitable on very unusual hardware.

[Reported a while ago but nothing happened so just fixing it]

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agosched: Fix string comparison in /proc/sched_features
Mathieu Desnoyers [Mon, 13 Sep 2010 21:47:00 +0000 (17:47 -0400)]
sched: Fix string comparison in /proc/sched_features

commit 7740191cd909b75d75685fb08a5d1f54b8a9d28b upstream.

Fix incorrect handling of the following case:

 INTERACTIVE
 INTERACTIVE_SOMETHING_ELSE

The comparison only checks up to each element's length.

Changelog since v1:
 - Embellish using some Rostedtisms.
  [ mingo:                 ^^ == smaller and cleaner ]

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Reviewed-by: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tony Lindgren <tony@atomide.com>
LKML-Reference: <20100913214700.GB16118@Krystal>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agopcmcia: synclink_cs: fix information leak to userland
Vasiliy Kulikov [Sun, 17 Oct 2010 14:41:24 +0000 (18:41 +0400)]
pcmcia: synclink_cs: fix information leak to userland

commit 5b917a1420d3d1a9c8da49fb0090692dc9aaee86 upstream.

Structure new_line is copied to userland with some padding fields unitialized.
It leads to leaking of stack memory.

Signed-off-by: Vasiliy Kulikov <segooon@gmail.com>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agopowerpc/perf: Fix sampling enable for PPC970
Paul Mackerras [Thu, 9 Sep 2010 19:02:40 +0000 (19:02 +0000)]
powerpc/perf: Fix sampling enable for PPC970

commit 9f5f9ffe50e90ed73040d2100db8bfc341cee352 upstream.

The logic to distinguish marked instruction events from ordinary events
on PPC970 and derivatives was flawed.  The result is that instruction
sampling didn't get enabled in the PMU for some marked instruction
events, so they would never trigger.  This fixes it by adding the
appropriate break statements in the switch statement.

Reported-by: David Binderman <dcb314@hotmail.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agostaging: usbip: Process event flags without delay
Max Vozeler [Tue, 21 Sep 2010 15:43:30 +0000 (17:43 +0200)]
staging: usbip: Process event flags without delay

commit 584c5b7cf06194464240280483ee0376cdddbbae upstream.

The way the event handler works can cause it to delay
events until eventual wakeup for another event.

For example, on device detach (vhci):

 - Write to sysfs detach file
    -> usbip_event_add(VDEV_EVENT_DOWN)
      -> wakeup()

#define VDEV_EVENT_DOWN (USBIP_EH_SHUTDOWN | USBIP_EH_RESET).

 - Event thread wakes up and passes the event to
   event_handler() to process.

 - It processes and clears the USBIP_EH_SHUTDOWN
   flag then returns.

 - The outer event loop (event_handler_loop()) calls
   wait_event_interruptible().

The processing of the second flag which is part of
VDEV_EVENT_DOWN (USBIP_EH_RESET) did not happen yet.
It is delayed until the next event.

This means the ->reset callback may not happen for
a long time (if ever), leaving the usbip port in a
weird state which prevents its reuse.

This patch changes the handler to process all flags
before waiting for another wakeup.

I have verified this change to fix a problem which
prevented reattach of a usbip device. It also helps
for socket errors which missed the RESET as well.

The delayed event processing also affects the stub
side of usbip and the error handling there.

Signed-off-by: Max Vozeler <mvz@vozeler.com>
Reported-by: Marco Lancione <marco@optikam.com>
Tested-by: Luc Jalbert <ljalbert@optikam.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agostaging: usbip: Notify usb core of port status changes
Max Vozeler [Tue, 21 Sep 2010 15:31:40 +0000 (17:31 +0200)]
staging: usbip: Notify usb core of port status changes

commit 0c9a32f0192e656daa2ff3c9149f6d71b4a1b873 upstream.

This patch changes vhci to behave like dummy and
other hcds when disconnecting a device.

Previously detaching a device from the root hub
did not notify the usb core of the disconnect and
left the device visible.

Signed-off-by: Max Vozeler <mvz@vozeler.com>
Reported-by: Marco Lancione <marco@optikam.com>
Tested-by: Luc Jalbert <ljalbert@optikam.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoLinux 2.6.32.25
Greg Kroah-Hartman [Fri, 29 Oct 2010 04:47:22 +0000 (21:47 -0700)]
Linux 2.6.32.25

13 years agomm: Move vma_stack_continue into mm.h
Stefan Bader [Tue, 31 Aug 2010 13:52:27 +0000 (15:52 +0200)]
mm: Move vma_stack_continue into mm.h

commit 39aa3cb3e8250db9188a6f1e3fb62ffa1a717678 upstream.

So it can be used by all that need to check for that.

Signed-off-by: Stefan Bader <stefan.bader@canonical.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoPhonet: disable network namespace support
Rémi Denis-Courmont [Fri, 17 Sep 2010 22:36:46 +0000 (01:36 +0300)]
Phonet: disable network namespace support

[Solved differently upstream]

Network namespace in the Phonet socket stack causes an OOPS when a
namespace is destroyed. This occurs as the loopback exit_net handler is
called after the Phonet exit_net handler, and re-enters the Phonet
stack. I cannot think of any nice way to fix this in kernel <= 2.6.32.

For lack of a better solution, disable namespace support completely.
If you need that, upgrade to a newer kernel.

Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Ben Hutchings <ben@decadent.org.uk>
Acked-by: David S. Miller <davem@davemloft.net>
13 years agoexecve: make responsive to SIGKILL with large arguments
Roland McGrath [Wed, 8 Sep 2010 02:37:06 +0000 (19:37 -0700)]
execve: make responsive to SIGKILL with large arguments

commit 9aea5a65aa7a1af9a4236dfaeb0088f1624f9919 upstream.

An execve with a very large total of argument/environment strings
can take a really long time in the execve system call.  It runs
uninterruptibly to count and copy all the strings.  This change
makes it abort the exec quickly if sent a SIGKILL.

Note that this is the conservative change, to interrupt only for
SIGKILL, by using fatal_signal_pending().  It would be perfectly
correct semantics to let any signal interrupt the string-copying in
execve, i.e. use signal_pending() instead of fatal_signal_pending().
We'll save that change for later, since it could have user-visible
consequences, such as having a timer set too quickly make it so that
an execve can never complete, though it always happened to work before.

Signed-off-by: Roland McGrath <roland@redhat.com>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Chuck Ebbert <cebbert@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoexecve: improve interactivity with large arguments
Roland McGrath [Wed, 8 Sep 2010 02:36:28 +0000 (19:36 -0700)]
execve: improve interactivity with large arguments

commit 7993bc1f4663c0db67bb8f0d98e6678145b387cd upstream.

This adds a preemption point during the copying of the argument and
environment strings for execve, in copy_strings().  There is already
a preemption point in the count() loop, so this doesn't add any new
points in the abstract sense.

When the total argument+environment strings are very large, the time
spent copying them can be much more than a normal user time slice.
So this change improves the interactivity of the rest of the system
when one process is doing an execve with very large arguments.

Signed-off-by: Roland McGrath <roland@redhat.com>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Chuck Ebbert <cebbert@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agosetup_arg_pages: diagnose excessive argument size
Roland McGrath [Wed, 8 Sep 2010 02:35:49 +0000 (19:35 -0700)]
setup_arg_pages: diagnose excessive argument size

commit 1b528181b2ffa14721fb28ad1bd539fe1732c583 upstream.

The CONFIG_STACK_GROWSDOWN variant of setup_arg_pages() does not
check the size of the argument/environment area on the stack.
When it is unworkably large, shift_arg_pages() hits its BUG_ON.
This is exploitable with a very large RLIMIT_STACK limit, to
create a crash pretty easily.

Check that the initial stack is not too large to make it possible
to map in any executable.  We're not checking that the actual
executable (or intepreter, for binfmt_elf) will fit.  So those
mappings might clobber part of the initial stack mapping.  But
that is just userland lossage that userland made happen, not a
kernel problem.

Signed-off-by: Roland McGrath <roland@redhat.com>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Chuck Ebbert <cebbert@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agofix 2.6.32.23 suspend regression caused by commit 6f6198a
Mike Galbraith [Fri, 1 Oct 2010 04:27:16 +0000 (06:27 +0200)]
fix 2.6.32.23 suspend regression caused by commit 6f6198a

[Not upstream in the same way, as it was fixed differently there]

6f6198a sched: kill migration thread in CPU_POST_DEAD instead of CPU_DEAD
leaves migration threads lying about.  Mask out CPU_TASKS_FROZEN.

Signed-off-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agox86: detect scattered cpuid features earlier
Jacob Pan [Wed, 19 May 2010 19:01:23 +0000 (12:01 -0700)]
x86: detect scattered cpuid features earlier

commit 1dedefd1a066a795a87afca9c0236e1a94de9bf6 upstream.

Some extra CPU features such as ARAT is needed in early boot so
that x86_init function pointers can be set up properly.
http://lkml.org/lkml/2010/5/18/519
At start_kernel() level, this patch moves init_scattered_cpuid_features()
from check_bugs() to setup_arch() -> early_cpu_init() which is earlier than
platform specific x86_init layer setup. Suggested by HPA.

Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
LKML-Reference: <1274295685-6774-2-git-send-email-jacob.jun.pan@linux.intel.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoACPI: Disable Windows Vista compatibility for Toshiba P305D
Zhang Rui [Wed, 29 Sep 2010 02:48:55 +0000 (22:48 -0400)]
ACPI: Disable Windows Vista compatibility for Toshiba P305D

commit 337279ce3aa85d81d34c0f837d1c204df105103b upstream.

Disable the Windows Vista (SP1) compatibility for Toshiba P305D.

http://bugzilla.kernel.org/show_bug.cgi?id=14736

Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoACPI: delete ZEPTO idle=nomwait DMI quirk
Len Brown [Tue, 28 Sep 2010 21:20:20 +0000 (17:20 -0400)]
ACPI: delete ZEPTO idle=nomwait DMI quirk

commit 64a32307b710c100beb101e9c78f8022f0e8ba61 upstream.

per comments in the bug report, this entry
seems to hurt at much as it helps.

https://bugzilla.kernel.org/show_bug.cgi?id=10807

Signed-off-by: Len Brown <len.brown@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoACPI: EC: add Vista incompatibility DMI entry for Toshiba Satellite L355
Len Brown [Tue, 28 Sep 2010 21:51:51 +0000 (17:51 -0400)]
ACPI: EC: add Vista incompatibility DMI entry for Toshiba Satellite L355

commit 7a1d602f5fc35d14907b7da98d5627acb69589d1 upstream.

https://bugzilla.kernel.org/show_bug.cgi?id=12641

Signed-off-by: Len Brown <len.brown@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agointel_idle: PCI quirk to prevent Lenovo Ideapad s10-3 boot hang
Len Brown [Sat, 25 Sep 2010 01:02:27 +0000 (21:02 -0400)]
intel_idle: PCI quirk to prevent Lenovo Ideapad s10-3 boot hang

commit 4731fdcf6f7bdab3e369a3f844d4ea4d4017284d upstream.

When the Lenovo Ideapad S10-3 is booted with HT enabled,
it hits a boot hang in the intel_idle driver.

This occurs when entering ATM-C4 for the first time,
unless BM_STS is first cleared.

acpi_idle doesn't see this because it first checks
and clears BM_STS, but it would hit the same hang
if that check were disabled.

http://bugs.meego.com/show_bug.cgi?id=7093
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/634702

Signed-off-by: Len Brown <len.brown@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoACPI: enable repeated PCIEXP wakeup by clearing PCIEXP_WAKE_STS on resume
Colin Ian King [Mon, 2 Aug 2010 15:14:43 +0000 (15:14 +0000)]
ACPI: enable repeated PCIEXP wakeup by clearing PCIEXP_WAKE_STS on resume

commit 573b638158029898caf9470c8214b7ddd29751e3 upstream.

Section 4.7.3.1.1 (PM1 Status Registers) of version 4.0 of
the ACPI spec concerning PCIEXP_WAKE_STS points out in
in the final note field in table 4-11 that if this bit is
set to 1 and the system is put into a sleeping state then
the system will not automatically wake.

This bit gets set by hardware to indicate that the system
woke up due to a PCI Express wakeup event, so clear it during
acpi_hw_clear_acpi_status() calls to enable subsequent
resumes to work.

BugLink: http://bugs.launchpad.net/bugs/613381
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agob44: fix carrier detection on bind
Paul Fertser [Mon, 11 Oct 2010 22:45:35 +0000 (15:45 -0700)]
b44: fix carrier detection on bind

commit bcf64aa379fcadd074449cbf0c049da70071b06f upstream.

For carrier detection to work properly when binding the driver with a cable
unplugged, netif_carrier_off() should be called after register_netdev(),
not before.

Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agopowerpc: Don't use kernel stack with translation off
Michael Neuling [Wed, 25 Aug 2010 21:04:25 +0000 (21:04 +0000)]
powerpc: Don't use kernel stack with translation off

commit 54a834043314c257210db2a9d59f8cc605571639 upstream.

In f761622e59433130bc33ad086ce219feee9eb961 we changed
early_setup_secondary so it's called using the proper kernel stack
rather than the emergency one.

Unfortunately, this stack pointer can't be used when translation is off
on PHYP as this stack pointer might be outside the RMO.  This results in
the following on all non zero cpus:
  cpu 0x1: Vector: 300 (Data Access) at [c00000001639fd10]
      pc: 000000000001c50c
      lr: 000000000000821c
      sp: c00000001639ff90
     msr: 8000000000001000
     dar: c00000001639ffa0
   dsisr: 42000000
    current = 0xc000000016393540
    paca    = 0xc000000006e00200
      pid   = 0, comm = swapper

The original patch was only tested on bare metal system, so it never
caught this problem.

This changes __secondary_start so that we calculate the new stack
pointer but only start using it after we've called early_setup_secondary.

With this patch, the above problem goes away.

Signed-off-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agopowerpc: Initialise paca->kstack before early_setup_secondary
Matt Evans [Thu, 12 Aug 2010 20:58:28 +0000 (20:58 +0000)]
powerpc: Initialise paca->kstack before early_setup_secondary

commit f761622e59433130bc33ad086ce219feee9eb961 upstream.

As early setup calls down to slb_initialize(), we must have kstack
initialised before checking "should we add a bolted SLB entry for our kstack?"

Failing to do so means stack access requires an SLB miss exception to refill
an entry dynamically, if the stack isn't accessible via SLB(0) (kernel text
& static data).  It's not always allowable to take such a miss, and
intermittent crashes will result.

Primary CPUs don't have this issue; an SLB entry is not bolted for their
stack anyway (as that lives within SLB(0)).  This patch therefore only
affects the init of secondaries.

Signed-off-by: Matt Evans <matt@ozlabs.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agor6040: Fix multicast list iteration when hash filter is used
Ben Hutchings [Fri, 15 Oct 2010 03:36:53 +0000 (04:36 +0100)]
r6040: Fix multicast list iteration when hash filter is used

This was fixed in mainline by the interface change made in commit
f9dcbcc9e338d08c0f7de7eba4eaafbbb7f81249.

After walking the multicast list to set up the hash filter, this
function will walk off the end of the list when filling the
exact-match entries.  This was fixed in mainline by the interface
change made in commit f9dcbcc9e338d08c0f7de7eba4eaafbbb7f81249.

Reported-by: spamalot@hispeed.ch
Reference: https://bugzilla.kernel.org/show_bug.cgi?id=15355
Reported-by: Jason Heeris <jason.heeris@gmail.com>
Reference: http://bugs.debian.org/600155
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agor6040: fix r6040_multicast_list
Florian Fainelli [Wed, 7 Apr 2010 23:50:58 +0000 (16:50 -0700)]
r6040: fix r6040_multicast_list

commit 3bcf8229a8c49769e48d3e0bd1e20d8e003f8106 upstream.

As reported in <https://bugzilla.kernel.org/show_bug.cgi?id=15355>, r6040_
multicast_list currently crashes. This is due a wrong maximum of multicast
entries. This patch fixes the following issues with multicast:

- number of maximum entries if off-by-one (4 instead of 3)

- the writing of the hash table index is not necessary and leads to invalid
values being written into the MCR1 register, so the MAC is simply put in a non
coherent state

- when we exceed the maximum number of mutlticast address, writing the
broadcast address should be done in registers MID_1{L,M,H} instead of
MID_O{L,M,H}, otherwise we would loose the adapter's MAC address

[bwh: Adjust for 2.6.32; should also apply to 2.6.27]

Signed-off-by: Florian Fainelli <florian@openwrt.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Cc: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agobsg: fix incorrect device_status value
FUJITA Tomonori [Thu, 16 Sep 2010 15:46:42 +0000 (00:46 +0900)]
bsg: fix incorrect device_status value

commit 478971600e47cb83ff2d3c63c5c24f2b04b0d6a1 upstream.

bsg incorrectly returns sg's masked_status value for device_status.

[jejb: fix up expression logic]
Reported-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agor8169: allocate with GFP_KERNEL flag when able to sleep
Stanislaw Gruszka [Fri, 8 Oct 2010 04:25:00 +0000 (04:25 +0000)]
r8169: allocate with GFP_KERNEL flag when able to sleep

commit aeb19f6052b5e5c8a24aa444fbff73b84341beac upstream.

We have fedora bug report where driver fail to initialize after
suspend/resume because of memory allocation errors:
https://bugzilla.redhat.com/show_bug.cgi?id=629158

To fix use GFP_KERNEL allocation where possible.

Tested-by: Neal Becker <ndbecker2@gmail.com>
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoskge: add quirk to limit DMA
Stanislaw Gruszka [Tue, 5 Oct 2010 22:11:40 +0000 (15:11 -0700)]
skge: add quirk to limit DMA

commit 392bd0cb000d4aac9e88e4f50823db85e7220688 upstream.

Skge devices installed on some Gigabyte motherboards are not able to
perform 64 dma correctly due to board PCI implementation, so limit
DMA to 32bit if such boards are detected.

Bug was reported here:
https://bugzilla.redhat.com/show_bug.cgi?id=447489

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Tested-by: Luya Tshimbalanga <luya@fedoraproject.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agonet: blackhole route should always be recalculated
Jianzhao Wang [Wed, 8 Sep 2010 21:35:43 +0000 (14:35 -0700)]
net: blackhole route should always be recalculated

[ Upstream commit ae2688d59b5f861dc70a091d003773975d2ae7fb ]

Blackhole routes are used when xfrm_lookup() returns -EREMOTE (error
triggered by IKE for example), hence this kind of route is always
temporary and so we should check if a better route exists for next
packets.
Bug has been introduced by commit d11a4dc18bf41719c9f0d7ed494d295dd2973b92.

Signed-off-by: Jianzhao Wang <jianzhao.wang@6wind.com>
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agorose: Fix signedness issues wrt. digi count.
David S. Miller [Mon, 20 Sep 2010 22:40:35 +0000 (15:40 -0700)]
rose: Fix signedness issues wrt. digi count.

[ Upstream commit 9828e6e6e3f19efcb476c567b9999891d051f52f ]

Just use explicit casts, since we really can't change the
types of structures exported to userspace which have been
around for 15 years or so.

Reported-by: Dan Rosenberg <dan.j.rosenberg@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agonetxen: dont set skb->truesize
Eric Dumazet [Tue, 21 Sep 2010 20:04:04 +0000 (13:04 -0700)]
netxen: dont set skb->truesize

[ Upstream commit 7e96dc7045bff8758804b047c0dfb6868f182500 ]

skb->truesize is set in core network.

Dont change it unless dealing with fragments.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agotcp: Fix race in tcp_poll
Tom Marshall [Mon, 20 Sep 2010 22:42:05 +0000 (15:42 -0700)]
tcp: Fix race in tcp_poll

[ Upstream commit a4d258036ed9b2a1811c3670c6099203a0f284a0 ]

If a RST comes in immediately after checking sk->sk_err, tcp_poll will
return POLLIN but not POLLOUT.  Fix this by checking sk->sk_err at the end
of tcp_poll.  Additionally, ensure the correct order of operations on SMP
machines with memory barriers.

Signed-off-by: Tom Marshall <tdm.code@gmail.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agonet: clear heap allocations for privileged ethtool actions
Kees Cook [Mon, 11 Oct 2010 19:23:25 +0000 (12:23 -0700)]
net: clear heap allocations for privileged ethtool actions

[ Upstream commit b00916b189d13a615ff05c9242201135992fcda3 ]

Several other ethtool functions leave heap uncleared (potentially) by
drivers. Some interfaces appear safe (eeprom, etc), in that the sizes
are well controlled. In some situations (e.g. unchecked error conditions),
the heap will remain unchanged in areas before copying back to userspace.
Note that these are less of an issue since these all require CAP_NET_ADMIN.

Cc: stable@kernel.org
Signed-off-by: Kees Cook <kees.cook@canonical.com>
Acked-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoip: fix truesize mismatch in ip fragmentation
Eric Dumazet [Tue, 21 Sep 2010 08:47:45 +0000 (08:47 +0000)]
ip: fix truesize mismatch in ip fragmentation

[ Upstream commit 3d13008e7345fa7a79d8f6438150dc15d6ba6e9d ]

Special care should be taken when slow path is hit in ip_fragment() :

When walking through frags, we transfert truesize ownership from skb to
frags. Then if we hit a slow_path condition, we must undo this or risk
uncharging frags->truesize twice, and in the end, having negative socket
sk_wmem_alloc counter, or even freeing socket sooner than expected.

Many thanks to Nick Bowler, who provided a very clean bug report and
test program.

Thanks to Jarek for reviewing my first patch and providing a V2

While Nick bisection pointed to commit 2b85a34e911 (net: No more
expensive sock_hold()/sock_put() on each tx), underlying bug is older
(2.6.12-rc5)

A side effect is to extend work done in commit b2722b1c3a893e
(ip_fragment: also adjust skb->truesize for packets not owned by a
socket) to ipv6 as well.

Reported-and-bisected-by: Nick Bowler <nbowler@elliptictech.com>
Tested-by: Nick Bowler <nbowler@elliptictech.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Jarek Poplawski <jarkao2@gmail.com>
CC: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agonet: Fix IPv6 PMTU disc. w/ asymmetric routes
Maciej Żenczykowski [Sun, 3 Oct 2010 21:49:00 +0000 (14:49 -0700)]
net: Fix IPv6 PMTU disc. w/ asymmetric routes

[ Upstream commit ae878ae280bea286ff2b1e1cb6e609dd8cb4501d ]

Signed-off-by: Maciej Żenczykowski <maze@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoPhonet: Correct header retrieval after pskb_may_pull
Kumar Sanghvi [Mon, 27 Sep 2010 23:10:42 +0000 (23:10 +0000)]
Phonet: Correct header retrieval after pskb_may_pull

[ Upstream commit a91e7d471e2e384035b9746ea707ccdcd353f5dd ]

Retrieve the header after doing pskb_may_pull since, pskb_may_pull
could change the buffer structure.

This is based on the comment given by Eric Dumazet on Phonet
Pipe controller patch for a similar problem.

Signed-off-by: Kumar Sanghvi <kumar.sanghvi@stericsson.com>
Acked-by: Linus Walleij <linus.walleij@stericsson.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agonet: Fix the condition passed to sk_wait_event()
Nagendra Tomar [Sat, 2 Oct 2010 23:45:06 +0000 (23:45 +0000)]
net: Fix the condition passed to sk_wait_event()

[ Upstream commit 482964e56e1320cb7952faa1932d8ecf59c4bf75 ]

This patch fixes the condition (3rd arg) passed to sk_wait_event() in
sk_stream_wait_memory(). The incorrect check in sk_stream_wait_memory()
causes the following soft lockup in tcp_sendmsg() when the global tcp
memory pool has exhausted.

>>> snip <<<

localhost kernel: BUG: soft lockup - CPU#3 stuck for 11s! [sshd:6429]
localhost kernel: CPU 3:
localhost kernel: RIP: 0010:[sk_stream_wait_memory+0xcd/0x200]  [sk_stream_wait_memory+0xcd/0x200] sk_stream_wait_memory+0xcd/0x200
localhost kernel:
localhost kernel: Call Trace:
localhost kernel:  [sk_stream_wait_memory+0x1b1/0x200] sk_stream_wait_memory+0x1b1/0x200
localhost kernel:  [<ffffffff802557c0>] autoremove_wake_function+0x0/0x40
localhost kernel:  [ipv6:tcp_sendmsg+0x6e6/0xe90] tcp_sendmsg+0x6e6/0xce0
localhost kernel:  [sock_aio_write+0x126/0x140] sock_aio_write+0x126/0x140
localhost kernel:  [xfs:do_sync_write+0xf1/0x130] do_sync_write+0xf1/0x130
localhost kernel:  [<ffffffff802557c0>] autoremove_wake_function+0x0/0x40
localhost kernel:  [hrtimer_start+0xe3/0x170] hrtimer_start+0xe3/0x170
localhost kernel:  [vfs_write+0x185/0x190] vfs_write+0x185/0x190
localhost kernel:  [sys_write+0x50/0x90] sys_write+0x50/0x90
localhost kernel:  [system_call+0x7e/0x83] system_call+0x7e/0x83

>>> snip <<<

What is happening is, that the sk_wait_event() condition passed from
sk_stream_wait_memory() evaluates to true for the case of tcp global memory
exhaustion. This is because both sk_stream_memory_free() and vm_wait are true
which causes sk_wait_event() to *not* call schedule_timeout().
Hence sk_stream_wait_memory() returns immediately to the caller w/o sleeping.
This causes the caller to again try allocation, which again fails and again
calls sk_stream_wait_memory(), and so on.

[ Bug introduced by commit c1cbe4b7ad0bc4b1d98ea708a3fecb7362aa4088
  ("[NET]: Avoid atomic xchg() for non-error case") -DaveM ]

Signed-off-by: Nagendra Singh Tomar <tomer_iisc@yahoo.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agotcp: Fix >4GB writes on 64-bit.
David S. Miller [Tue, 28 Sep 2010 03:24:54 +0000 (20:24 -0700)]
tcp: Fix >4GB writes on 64-bit.

[ Upstream commit 01db403cf99f739f86903314a489fb420e0e254f ]

Fixes kernel bugzilla #16603

tcp_sendmsg() truncates iov_len to an 'int' which a 4GB write to write
zero bytes, for example.

There is also the problem higher up of how verify_iovec() works.  It
wants to prevent the total length from looking like an error return
value.

However it does this using 'int', but syscalls return 'long' (and
thus signed 64-bit on 64-bit machines).  So it could trigger
false-positives on 64-bit as written.  So fix it to use 'long'.

Reported-by: Olaf Bonorden <bono@onlinehome.de>
Reported-by: Daniel Büse <dbuese@gmx.de>
Reported-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoxfrm4: strip ECN and IP Precedence bits in policy lookup
Ulrich Weber [Tue, 5 Oct 2010 11:46:19 +0000 (13:46 +0200)]
xfrm4: strip ECN and IP Precedence bits in policy lookup

[ Upstream commit 94e2238969e89f5112297ad2a00103089dde7e8f ]

dont compare ECN and IP Precedence bits in find_bundle
and use ECN bit stripped TOS value in xfrm_lookup

Signed-off-by: Ulrich Weber <uweber@astaro.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agodrm/radeon: fix PCI ID 5657 to be an RV410
Dave Airlie [Sat, 25 Sep 2010 07:45:50 +0000 (17:45 +1000)]
drm/radeon: fix PCI ID 5657 to be an RV410

commit f459ffbdfd04edb4a8ce6eea33170eb057a5e695 upstream.

fixes https://bugzilla.kernel.org/show_bug.cgi?id=19012

Signed-off-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoDe-pessimize rds_page_copy_user
Linus Torvalds [Fri, 15 Oct 2010 18:09:28 +0000 (11:09 -0700)]
De-pessimize rds_page_copy_user

commit 799c10559d60f159ab2232203f222f18fa3c4a5f upstream.

Don't try to "optimize" rds_page_copy_user() by using kmap_atomic() and
the unsafe atomic user mode accessor functions.  It's actually slower
than the straightforward code on any reasonable modern CPU.

Back when the code was written (although probably not by the time it was
actually merged, though), 32-bit x86 may have been the dominant
architecture.  And there kmap_atomic() can be a lot faster than kmap()
(unless you have very good locality, in which case the virtual address
caching by kmap() can overcome all the downsides).

But these days, x86-64 may not be more populous, but it's getting there
(and if you care about performance, it's definitely already there -
you'd have upgraded your CPU's already in the last few years).  And on
x86-64, the non-kmap_atomic() version is faster, simply because the code
is simpler and doesn't have the "re-try page fault" case.

People with old hardware are not likely to care about RDS anyway, and
the optimization for the 32-bit case is simply buggy, since it doesn't
verify the user addresses properly.

Reported-by: Dan Rosenberg <drosenberg@vsecurity.com>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agox86, AMD, MCE thresholding: Fix the MCi_MISCj iteration order
Borislav Petkov [Fri, 8 Oct 2010 10:08:34 +0000 (12:08 +0200)]
x86, AMD, MCE thresholding: Fix the MCi_MISCj iteration order

commit 6dcbfe4f0b4e17e289d56fa534b7ce5a6b7f63a3 upstream.

This fixes possible cases of not collecting valid error info in
the MCE error thresholding groups on F10h hardware.

The current code contains a subtle problem of checking only the
Valid bit of MSR0000_0413 (which is MC4_MISC0 - DRAM
thresholding group) in its first iteration and breaking out if
the bit is cleared.

But (!), this MSR contains an offset value, BlkPtr[31:24], which
points to the remaining MSRs in this thresholding group which
might contain valid information too. But if we bail out only
after we checked the valid bit in the first MSR and not the
block pointer too, we miss that other information.

The thing is, MC4_MISC0[BlkPtr] is not predicated on
MCi_STATUS[MiscV] or MC4_MISC0[Valid] and should be checked
prior to iterating over the MCI_MISCj thresholding group,
irrespective of the MC4_MISC0[Valid] setting.

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoatl1: fix resume
Luca Tettamanti [Wed, 22 Sep 2010 10:41:58 +0000 (10:41 +0000)]
atl1: fix resume

commit ec5a32f67c603b11d68eb283d94eb89a4f6cfce1 upstream.

adapter->cmb.cmb is initialized when the device is opened and freed when
it's closed. Accessing it unconditionally during resume results either
in a crash (NULL pointer dereference, when the interface has not been
opened yet) or data corruption (when the interface has been used and
brought down adapter->cmb.cmb points to a deallocated memory area).

Signed-off-by: Luca Tettamanti <kronos.it@gmail.com>
Acked-by: Chris Snook <chris.snook@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agowext: fix potential private ioctl memory content leak
Johannes Berg [Thu, 16 Sep 2010 22:38:25 +0000 (00:38 +0200)]
wext: fix potential private ioctl memory content leak

commit df6d02300f7c2fbd0fbe626d819c8e5237d72c62 upstream.

When a driver doesn't fill the entire buffer, old
heap contents may remain, and if it also doesn't
update the length properly, this old heap content
will be copied back to userspace.

It is very unlikely that this happens in any of
the drivers using private ioctls since it would
show up as junk being reported by iwpriv, but it
seems better to be safe here, so use kzalloc.

Reported-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoocfs2: Don't walk off the end of fast symlinks.
Joel Becker [Thu, 30 Sep 2010 00:33:05 +0000 (17:33 -0700)]
ocfs2: Don't walk off the end of fast symlinks.

commit 1fc8a117865b54590acd773a55fbac9221b018f0 upstream.

ocfs2 fast symlinks are NUL terminated strings stored inline in the
inode data area.  However, disk corruption or a local attacker could, in
theory, remove that NUL.  Because we're using strlen() (my fault,
introduced in a731d1 when removing vfs_follow_link()), we could walk off
the end of that string.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>