8 years agoMerge branch 'pm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/suspe...
Linus Torvalds [Thu, 7 Jul 2011 20:22:41 +0000]
Merge branch 'pm-fixes' of git://git./linux/kernel/git/rafael/suspend-2.6

* 'pm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/suspend-2.6:
  PM / Hibernate: Fix free_unnecessary_pages()

8 years agoMerge branch 'for-linus' of git://git.kernel.dk/linux-block
Linus Torvalds [Thu, 7 Jul 2011 20:22:26 +0000]
Merge branch 'for-linus' of git://git.kernel.dk/linux-block

* 'for-linus' of git://git.kernel.dk/linux-block:
  drbd: we should write meta data updates with FLUSH FUA
  drbd: fix limit define, we support 1 PiByte now
  drbd: when receive times out on meta socket, also check last receive time on data socket
  drbd: account bitmap IO during resync as resync-(related-)-io
  drbd: don't cond_resched_lock with IRQs disabled
  drbd: add missing spinlock to bitmap receive
  drbd: Use the correct max_bio_size when creating resync requests
  cfq-iosched: make code consistent
  cfq-iosched: fix a rcu warning

8 years agoFS-Cache: Add a helper to bulk uncache pages on an inode
David Howells [Thu, 7 Jul 2011 11:19:48 +0000]
FS-Cache: Add a helper to bulk uncache pages on an inode

Add an FS-Cache helper to bulk uncache pages on an inode.  This will
only work for the circumstance where the pages in the cache correspond
1:1 with the pages attached to an inode's page cache.

This is required for CIFS and NFS: When disabling inode cookie, we were
returning the cookie and setting cifsi->fscache to NULL but failed to
invalidate any previously mapped pages.  This resulted in "Bad page
state" errors and manifested in other kind of errors when running
fsstress.  Fix it by uncaching mapped pages when we disable the inode
cookie.

This patch should fix the following oops and "Bad page state" errors
seen during fsstress testing.

  ------------[ cut here ]------------
  kernel BUG at fs/cachefiles/namei.c:201!
  invalid opcode: 0000 [#1] SMP
  Pid: 5, comm: kworker/u:0 Not tainted 2.6.38.7-30.fc15.x86_64 #1 Bochs Bochs
  RIP: 0010: cachefiles_walk_to_object+0x436/0x745 [cachefiles]
  RSP: 0018:ffff88002ce6dd00  EFLAGS: 00010282
  RAX: ffff88002ef165f0 RBX: ffff88001811f500 RCX: 0000000000000000
  RDX: 0000000000000000 RSI: 0000000000000100 RDI: 0000000000000282
  RBP: ffff88002ce6dda0 R08: 0000000000000100 R09: ffffffff81b3a300
  R10: 0000ffff00066c0a R11: 0000000000000003 R12: ffff88002ae54840
  R13: ffff88002ae54840 R14: ffff880029c29c00 R15: ffff88001811f4b0
  FS:  00007f394dd32720(0000) GS:ffff88002ef00000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
  CR2: 00007fffcb62ddf8 CR3: 000000001825f000 CR4: 00000000000006e0
  DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
  DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
  Process kworker/u:0 (pid: 5, threadinfo ffff88002ce6c000, task ffff88002ce55cc0)
  Stack:
   0000000000000246 ffff88002ce55cc0 ffff88002ce6dd58 ffff88001815dc00
   ffff8800185246c0 ffff88001811f618 ffff880029c29d18 ffff88001811f380
   ffff88002ce6dd50 ffffffff814757e4 ffff88002ce6dda0 ffffffff8106ac56
  Call Trace:
   cachefiles_lookup_object+0x78/0xd4 [cachefiles]
   fscache_lookup_object+0x131/0x16d [fscache]
   fscache_object_work_func+0x1bc/0x669 [fscache]
   process_one_work+0x186/0x298
   worker_thread+0xda/0x15d
   kthread+0x84/0x8c
   kernel_thread_helper+0x4/0x10
  RIP  cachefiles_walk_to_object+0x436/0x745 [cachefiles]
  ---[ end trace 1d481c9af1804caa ]---

I tested the uncaching by the following means:

 (1) Create a big file on my NFS server (104857600 bytes).

 (2) Read the file into the cache with md5sum on the NFS client.  Look in
     /proc/fs/fscache/stats:

Pages  : mrk=25601 unc=0

 (3) Open the file for read/write ("bash 5<>/warthog/bigfile").  Look in proc
     again:

Pages  : mrk=25601 unc=25601

Reported-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-and-Tested-by: Suresh Jayaraman <sjayaraman@suse.de>
cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

8 years agoMerge branch 'stable/bug.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Thu, 7 Jul 2011 20:19:04 +0000]
Merge branch 'stable/bug.fixes' of git://git./linux/kernel/git/konrad/xen

* 'stable/bug.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen:
  xen/pci: Move check for acpi_sci_override_gsi to xen_setup_acpi_sci.

8 years agoMerge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Thu, 7 Jul 2011 20:18:13 +0000]
Merge branch 'x86-urgent-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  x86: Don't use the EFI reboot method by default
  x86, suspend: Restore MISC_ENABLE MSR in realmode wakeup
  x86, reboot: Acer Aspire One A110 reboot quirk
  x86-32, NUMA: Fix boot regression caused by NUMA init unification on highmem machines

8 years agoMerge branches 'core-urgent-for-linus', 'perf-urgent-for-linus' and 'sched-urgent...
Linus Torvalds [Thu, 7 Jul 2011 20:17:45 +0000]
Merge branches 'core-urgent-for-linus', 'perf-urgent-for-linus' and 'sched-urgent-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  debugobjects: Fix boot crash when kmemleak and debugobjects enabled

* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  jump_label: Fix jump_label update for modules
  oprofile, x86: Fix race in nmi handler while starting counters

* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  sched: Disable (revert) SCHED_LOAD_SCALE increase
  sched, cgroups: Fix MIN_SHARES on 64-bit boxen

8 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Thu, 7 Jul 2011 20:16:21 +0000]
Merge git://git./linux/kernel/git/davem/net-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (31 commits)
  sctp: fix missing send up SCTP_SENDER_DRY_EVENT when subscribe it
  net: refine {udp|tcp|sctp}_mem limits
  vmxnet3: round down # of queues to power of two
  net: sh_eth: fix the parameter for the ETHER of SH7757
  net: sh_eth: fix cannot work half-duplex mode
  net: vlan: enable soft features regardless of underlying device
  vmxnet3: fix starving rx ring whenoc_skb kb fails
  bridge: Always flood broadcast packets
  greth: greth_set_mac_add would corrupt the MAC address.
  net: bind() fix error return on wrong address family
  natsemi: silence dma-debug warnings
  net: 8139too: Initial necessary vlan_features to support vlan
  Fix call trace when interrupts are disabled while sleeping function kzalloc is called
  qlge:Version change to v1.00.00.29
  qlge: Fix printk priority so chip fatal errors are always reported.
  qlge:Fix crash caused by mailbox execution on wedged chip.
  xfrm4: Don't call icmp_send on local error
  ipv4: Don't use ufo handling on later transformed packets
  xfrm: Remove family arg from xfrm_bundle_ok
  ipv6: Don't put artificial limit on routing table size.
  ...

8 years agoxen/pci: Move check for acpi_sci_override_gsi to xen_setup_acpi_sci.
Konrad Rzeszutek Wilk [Wed, 6 Jul 2011 13:43:16 +0000]
xen/pci: Move check for acpi_sci_override_gsi to xen_setup_acpi_sci.

Previously we would check for acpi_sci_override_gsi == gsi every time
a PCI device was enabled. That works during early bootup, but later
on it could lead to triggering unnecessarily the acpi_gsi_to_irq(..) lookup.
The reason is that acpi_sci_override_gsi was declared in __initdata and
after early bootup could contain bogus values.

This patch moves the check for acpi_sci_override_gsi to the
site where the ACPI SCI is preset.

CC: stable@kernel.org
Reported-by: Raghavendra D Prabhu <rprabhu@wnohang.net>
Tested-by: Raghavendra D Prabhu <rprabhu@wnohang.net>
[http://lists.xensource.com/archives/html/xen-devel/2011-07/msg00154.html]
Suggested-by:  Ian Campbell <ijc@hellion.org.uk>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>

8 years agosctp: fix missing send up SCTP_SENDER_DRY_EVENT when subscribe it
Wei Yongjun [Sat, 2 Jul 2011 09:28:04 +0000]
sctp: fix missing send up SCTP_SENDER_DRY_EVENT when subscribe it

We forgot to send up SCTP_SENDER_DRY_EVENT notification when
user app subscribes to this event, and there is no data to be
sent or retransmit.

This is required by the Socket API and used by the DTLS/SCTP
implementation.

Reported-by: Michael Tüxen <Michael.Tuexen@lurchi.franken.de>
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Tested-by: Robin Seggelmann <seggelmann@fh-muenster.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

8 years agox86: Don't use the EFI reboot method by default
Matthew Garrett [Wed, 6 Jul 2011 20:52:37 +0000]
x86: Don't use the EFI reboot method by default

Testing suggests that at least some Lenovos and some Intels will
fail to reboot via EFI, attempting to jump to an unmapped
physical address. In the long run we could handle this by
providing a page table with a 1:1 mapping of physical addresses,
but for now it's probably just easier to assume that ACPI or
legacy methods will be present and reboot via those.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Alan Cox <alan@linux.intel.com>
Link: http://lkml.kernel.org/r/1309985557-15350-1-git-send-email-mjg@redhat.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>

8 years agonet: refine {udp|tcp|sctp}_mem limits
Eric Dumazet [Thu, 7 Jul 2011 07:27:05 +0000]
net: refine {udp|tcp|sctp}_mem limits

Current tcp/udp/sctp global memory limits are not taking into account
hugepages allocations, and allow 50% of ram to be used by buffers of a
single protocol [ not counting space used by sockets / inodes ...]

Lets use nr_free_buffer_pages() and allow a default of 1/8 of kernel ram
per protocol, and a minimum of 128 pages.
Heavy duty machines sysadmins probably need to tweak limits anyway.

References: https://bugzilla.stlinux.com/show_bug.cgi?id=38032
Reported-by: starlight <starlight@binnacle.cx>
Suggested-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

8 years agovmxnet3: round down # of queues to power of two
Shreyas Bhatewara [Thu, 7 Jul 2011 07:25:52 +0000]
vmxnet3: round down # of queues to power of two

vmxnet3 device supports only power-of-two number of queues. The driver
therefore needs to check this and rounds down the number of queues to the
nearest power of two.

Signed-off-by: Yong Wang <yongwang@vmware.com>
Signed-off-by: Shreyas N Bhatewara <sbhatewara@vmware.com>
Reviewed-by: Dmitry Torokhov <dtor@vmware.com>

8 years agox86, suspend: Restore MISC_ENABLE MSR in realmode wakeup
Kees Cook [Thu, 7 Jul 2011 01:10:34 +0000]
x86, suspend: Restore MISC_ENABLE MSR in realmode wakeup

Some BIOSes will reset the Intel MISC_ENABLE MSR (specifically the
XD_DISABLE bit) when resuming from S3, which can interact poorly with
ebba638ae723d8a8fc2f7abce5ec18b688b791d7. In 32bit PAE mode, this can
lead to a fault when EFER is restored by the kernel wakeup routines,
due to it setting the NX bit for a CPU that (thanks to the BIOS reset)
now incorrectly thinks it lacks the NX feature. (64bit is not affected
because it uses a common CPU bring-up that specifically handles the
XD_DISABLE bit.)

The need for MISC_ENABLE being restored so early is specific to the S3
resume path. Normally, MISC_ENABLE is saved in save_processor_state(),
but this happens after the resume header is created, so just reproduce
the logic here. (acpi_suspend_lowlevel() creates the header, calls
do_suspend_lowlevel, which calls save_processor_state(), so the saved
processor context isn't available during resume header creation.)

[ hpa: Consider for stable if OK in mainline ]

Signed-off-by: Kees Cook <kees.cook@canonical.com>
Link: http://lkml.kernel.org/r/20110707011034.GA8523@outflux.net
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: Rafael J. Wysocki <rjw@sisk.pl>
Cc: <stable@kernel.org> 2.6.38+

8 years agoMerge branch 'gpio/merge' of git://git.secretlab.ca/git/linux-2.6
Linus Torvalds [Thu, 7 Jul 2011 01:36:53 +0000]
Merge branch 'gpio/merge' of git://git.secretlab.ca/git/linux-2.6

* 'gpio/merge' of git://git.secretlab.ca/git/linux-2.6:
  gpio: tps65910: add missing breaks in tps65910_gpio_init

8 years agoDocumentation: fix cgroup blkio throttle filenames
Andrea Righi [Wed, 6 Jul 2011 18:26:26 +0000]
Documentation: fix cgroup blkio throttle filenames

All the blkio.throttle.* file names are incorrectly reported without
".throttle" in the documentation. Fix it.

Signed-off-by: Andrea Righi <andrea@betterlinux.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

8 years agoDocumentation: update CodingStyle memory allocators
Jesper Juhl [Wed, 6 Jul 2011 18:27:17 +0000]
Documentation: update CodingStyle memory allocators

The list of available general purpose memory allocators in
Documentation/CodingStyle chapter 14 is incomplete. This patch adds
the missing vzalloc() to the list.

Signed-off-by: Jesper Juhl <jj@chaosbits.net>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

8 years agoMAINTAINERS: move kernel-doc patches location
Randy Dunlap [Wed, 6 Jul 2011 18:19:56 +0000]
MAINTAINERS: move kernel-doc patches location

Move location of quilt series for kernel-doc patches.

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

8 years agoMerge branch 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab...
Linus Torvalds [Wed, 6 Jul 2011 19:16:49 +0000]
Merge branch 'v4l_for_linus' of git://git./linux/kernel/git/mchehab/linux-2.6

* 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6: (46 commits)
  [media] rc: call input_sync after scancode reports
  [media] imon: allow either proto on unknown 0xffdc
  [media] imon: auto-config ffdc 7e device
  [media] saa7134: fix raw IR timeout value
  [media] rc: fix ghost keypresses with certain hw
  [media] [staging] lirc_serial: allocate irq at init time
  [media] lirc_zilog: fix spinning rx thread
  [media] keymaps: fix table for pinnacle pctv hd devices
  [media] ite-cir: 8709 needs to use pnp resource 2
  [media] V4L: mx1-camera: fix uninitialized variable
  [media] omap_vout: Added check in reqbuf & mmap for buf_size allocation
  [media] OMAP_VOUT: Change hardcoded device node number to -1
  [media] OMAP_VOUTLIB: Fix wrong resizer calculation
  [media] uvcvideo: Disable the queue when failing to start
  [media] uvcvideo: Remove buffers from the queues when freeing
  [media] uvcvideo: Ignore entities for terminals with no supported format
  [media] v4l: Don't access media entity after is has been destroyed
  [media] media: omap3isp: fix a potential NULL deref
  [media] media: vb2: fix allocation failure check
  [media] media: vb2: reset queued_count value during queue reinitialization
  ...

Fix up trivial conflict in MAINTAINERS as per Mauro

8 years agoFDPIC: Fix memory leak
Davidlohr Bueso [Wed, 6 Jul 2011 11:26:05 +0000]
FDPIC: Fix memory leak

The shdr4extnum variable isn't being freed in the cleanup process of
elf_fdpic_core_dump().

Signed-off-by: Davidlohr Bueso <dave@gnu.org>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

8 years agoPM / Hibernate: Fix free_unnecessary_pages()
Rafael J. Wysocki [Wed, 6 Jul 2011 18:15:23 +0000]
PM / Hibernate: Fix free_unnecessary_pages()

There is a bug in free_unnecessary_pages() that causes it to
attempt to free too many pages in some cases, which triggers the
BUG_ON() in memory_bm_clear_bit() for copy_bm.  Namely, if
count_data_pages() is initially greater than alloc_normal, we get
to_free_normal equal to 0 and "save" greater from 0.  In that case,
if the sum of "save" and count_highmem_pages() is greater than
alloc_highmem, we subtract a positive number from to_free_normal.
Hence, since to_free_normal was 0 before the subtraction and is
an unsigned int, the result is converted to a huge positive number
that is used as the number of pages to free.

Fix this bug by checking if to_free_normal is actually greater
than or equal to the number we're going to subtract from it.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Reported-and-tested-by: Matthew Garrett <mjg@redhat.com>
Cc: stable@kernel.org

8 years agoresource: ability to resize an allocated resource
Ram Pai [Wed, 6 Jul 2011 06:44:30 +0000]
resource: ability to resize an allocated resource

Provides the ability to resize a resource that is already allocated.
This functionality is put in place to support reallocation needs of
pci resources.

Signed-off-by: Ram Pai <linuxram@us.ibm.com>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

8 years agofs: fix lock initialization
Miklos Szeredi [Wed, 6 Jul 2011 10:33:55 +0000]
fs: fix lock initialization

locks_alloc_lock() assumed that the allocated struct file_lock is
already initialized to zero members.  This is only true for the first
allocation of the structure, after reuse some of the members will have
random values.

This will for example result in passing random fl_start values to
userspace in fuse for FL_FLOCK locks, which is an information leak at
best.

Fix by reinitializing those members which may be non-zero after freeing.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
CC: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

8 years agonet: sh_eth: fix the parameter for the ETHER of SH7757
Yoshihiro Shimoda [Tue, 5 Jul 2011 20:33:57 +0000]
net: sh_eth: fix the parameter for the ETHER of SH7757

If the driver didn't set this parameter on the ETHER, the CPU will
encounter the "data address error" exception.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

8 years agonet: sh_eth: fix cannot work half-duplex mode
Yoshihiro Shimoda [Tue, 5 Jul 2011 20:33:51 +0000]
net: sh_eth: fix cannot work half-duplex mode

When link was down, the bit of DM in ECMR was always set.
So, we could not use half-duplex mode on the controller.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

8 years agogpio: tps65910: add missing breaks in tps65910_gpio_init
Axel Lin [Wed, 6 Jul 2011 02:08:27 +0000]
gpio: tps65910: add missing breaks in tps65910_gpio_init

Signed-off-by: Axel Lin <axel.lin@gmail.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>

8 years agoMerge branch 'usb-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Wed, 6 Jul 2011 03:57:45 +0000]
Merge branch 'usb-linus' of git://git./linux/kernel/git/gregkh/usb-2.6

* 'usb-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb-2.6:
  USB: fix regression occurring during device removal
  USB: fsl_udc_core: fix build breakage when building for ARM arch

8 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sameo/mfd-2.6
Linus Torvalds [Wed, 6 Jul 2011 03:57:08 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/sameo/mfd-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sameo/mfd-2.6:
  mfd: Add Makefile and Kconfig Entries for tps65911 comparator
  mfd: Fix build error for tps65911-comparator.c
  Revert "mfd: Add omap-usbhs runtime PM support"
  input: pmic8xxx-pwrkey: Do not use mfd_get_data()
  input: pmic8xxx-keypad: Do not use mfd_get_data()

8 years agonet: vlan: enable soft features regardless of underlying device
Shan Wei [Wed, 6 Jul 2011 03:43:12 +0000]
net: vlan: enable soft features regardless of underlying device

If gso/gro feature of underlying device is turned off,
then new created vlan device never can turn gso/gro on.

Although underlying device don't support TSO, we still
should use software segments for vlan device.

Signed-off-by: Shan Wei <shanwei@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

8 years agox86, reboot: Acer Aspire One A110 reboot quirk
Peter Chubb [Wed, 6 Jul 2011 00:56:30 +0000]
x86, reboot: Acer Aspire One A110 reboot quirk

Since git commit
  660e34cebf0a11d54f2d5dd8838607452355f321 x86: reorder reboot method
  preferences,
my Acer Aspire One hangs on reboot.  It appears that its ACPI method
for rebooting is broken.  The attached patch adds a quirk so that the
machine will reboot via the BIOS.

[ hpa: verified that the ACPI control on this machine is just plain broken. ]

Signed-off-by: Peter Chubb <peter.chubb@nicta.com.au>
Link: http://lkml.kernel.org/r/w439iki5vl.wl%25peter@chubb.wattle.id.au
Signed-off-by: H. Peter Anvin <hpa@zytor.com>

8 years agovmxnet3: fix starving rx ring whenoc_skb kb fails
Shreyas Bhatewara [Tue, 5 Jul 2011 14:34:05 +0000]
vmxnet3: fix starving rx ring whenoc_skb kb fails

If the rx ring is completely empty, then the device may never fire an rx
interrupt. Unfortunately, the rx interrupt is what triggers populating the
rx ring with fresh buffers, so this will cause networking to lock up.

This patch replenishes the skb in recv descriptor as soon as it is
peeled off while processing rx completions. If the skb/buffer
allocation fails, existing one is recycled and the packet in hand is
dropped. This way none of the RX desc is ever left empty, thus avoiding
starvation

Signed-off-by: Scott J. Goldman <scottjg@vmware.com>
Signed-off-by: Shreyas N Bhatewara <sbhatewara@vmware.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

8 years agobridge: Always flood broadcast packets
Herbert Xu [Tue, 5 Jul 2011 13:58:33 +0000]
bridge: Always flood broadcast packets

As is_multicast_ether_addr returns true on broadcast packets as
well, we need to explicitly exclude broadcast packets so that
they're always flooded.  This wasn't an issue before as broadcast
packets were considered to be an unregistered multicast group,
which were always flooded.  However, as we now only flood such
packets to router ports, this is no longer acceptable.

Reported-by: Michael Guntsche <mike@it-loops.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>

8 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph...
Linus Torvalds [Tue, 5 Jul 2011 20:15:57 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/sage/ceph-client

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
  ceph: fix sync and dio writes across stripe boundaries
  libceph: fix page calculation for non-page-aligned io
  ceph: fix page alignment corrections

8 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/hch/hfsplus
Linus Torvalds [Tue, 5 Jul 2011 17:04:27 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/hch/hfsplus

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/hch/hfsplus:
  hfsplus: Fix double iput of the same inode in hfsplus_fill_super()
  hfsplus: add missing call to bio_put()

8 years agosched: Disable (revert) SCHED_LOAD_SCALE increase
Peter Zijlstra [Tue, 5 Jul 2011 08:56:32 +0000]
sched: Disable (revert) SCHED_LOAD_SCALE increase

Alex reported that commit c8b281161df ("sched: Increase
SCHED_LOAD_SCALE resolution") caused a power usage regression
under light load as it increases the number of load-balance
operations and keeps idle cpus from staying idle.

Time has run out to find the root cause for this release so
disable the feature for v3.0 until we can figure out what
causes the problem.

Reported-by: "Alex, Shi" <alex.shi@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Nikhil Rao <ncrao@google.com>
Cc: Ming Lei <tom.leiming@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/n/tip-m4onxn0sxnyn5iz9o88eskc3@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>

8 years agomfd: Add Makefile and Kconfig Entries for tps65911 comparator
Axel Lin [Thu, 23 Jun 2011 02:15:51 +0000]
mfd: Add Makefile and Kconfig Entries for tps65911 comparator

Base on Mark's comment [1], I make the Kconfig entry invisible to users.
[1] https://lkml.org/lkml/2011/5/14/136

Signed-off-by: Axel Lin <axel.lin@gmail.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>

8 years agomfd: Fix build error for tps65911-comparator.c
Axel Lin [Thu, 23 Jun 2011 02:17:43 +0000]
mfd: Fix build error for tps65911-comparator.c

Fix below build error:
CC drivers/mfd/tps65911-comparator.o
drivers/mfd/tps65911-comparator.c: In function 'tps65911_comparator_probe':
drivers/mfd/tps65911-comparator.c:131: error: 'struct tps65910_platform_data' has no member named 'vmbch_threshold'
drivers/mfd/tps65911-comparator.c:137: error: 'struct tps65910_platform_data' has no member named 'vmbch2_threshold'
make[2]: *** [drivers/mfd/tps65911-comparator.o] Error 1
make[1]: *** [drivers/mfd] Error 2
make: *** [drivers] Error 2

Signed-off-by: Axel Lin <axel.lin@gmail.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>

8 years agoRevert "mfd: Add omap-usbhs runtime PM support"
Keshava Munegowda [Mon, 20 Jun 2011 13:22:56 +0000]
Revert "mfd: Add omap-usbhs runtime PM support"

This reverts commit 7e6502d577106fb5b202bbaac64c5f1b065e6daa.

Oops are produced during initialization of ehci and ohci
drivers. This is because the run time pm apis are used by
the driver but the corresponding hwmod structures and
initialization is not merged. hence revering  back the
commit id 7e6502d577106fb5b202bbaac64c5f1b065e6daa

Signed-off-by: Keshava Munegowda <keshava_mgowda@ti.com>
Reported-by: Luciano Coelho <coelho@ti.com>
Acked-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>

8 years agoinput: pmic8xxx-pwrkey: Do not use mfd_get_data()
Samuel Ortiz [Wed, 8 Jun 2011 09:06:30 +0000]
input: pmic8xxx-pwrkey: Do not use mfd_get_data()

mfd_get_data() has been removed from the MFD API.

Cc: Anirudh Ghayal <aghayal@codeaurora.org>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>

8 years agoinput: pmic8xxx-keypad: Do not use mfd_get_data()
Samuel Ortiz [Wed, 8 Jun 2011 09:03:23 +0000]
input: pmic8xxx-keypad: Do not use mfd_get_data()

mfd_get_data() has been removed from the MFD API.

Cc: Anirudh Ghayal <aghayal@codeaurora.org>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>

8 years agogreth: greth_set_mac_add would corrupt the MAC address.
Kristoffer Glembo [Tue, 5 Jul 2011 04:39:10 +0000]
greth: greth_set_mac_add would corrupt the MAC address.

The MAC address was set using the signed char sockaddr->sa_addr
field and thus the address could be corrupted through sign extension.

Signed-off-by: Kristoffer Glembo <kristoffer@gaisler.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

8 years agonet: bind() fix error return on wrong address family
Marcus Meissner [Mon, 4 Jul 2011 01:30:29 +0000]
net: bind() fix error return on wrong address family

Hi,

Reinhard Max also pointed out that the error should EAFNOSUPPORT according
to POSIX.

The Linux manpages have it as EINVAL, some other OSes (Minix, HPUX, perhaps BSD) use
EAFNOSUPPORT. Windows uses WSAEFAULT according to MSDN.

Other protocols error values in their af bind() methods in current mainline git as far
as a brief look shows:
EAFNOSUPPORT: atm, appletalk, l2tp, llc, phonet, rxrpc
EINVAL: ax25, bluetooth, decnet, econet, ieee802154, iucv, netlink, netrom, packet, rds, rose, unix, x25,
No check?: can/raw, ipv6/raw, irda, l2tp/l2tp_ip

Ciao, Marcus

Signed-off-by: Marcus Meissner <meissner@suse.de>
Cc: Reinhard Max <max@suse.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

8 years agoLinux 3.0-rc6
Linus Torvalds [Mon, 4 Jul 2011 22:56:24 +0000]
Linux 3.0-rc6

8 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6
Linus Torvalds [Mon, 4 Jul 2011 22:54:37 +0000]
Merge git://git./linux/kernel/git/jejb/scsi-rc-fixes-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6: (277 commits)
  [SCSI] isci: fix checkpatch errors
  isci: Device reset should request sas_phy_reset(phy, true)
  isci: pare back error messsages
  isci: cleanup silicon revision detection
  isci: merge scu_unsolicited_frame.h into unsolicited_frame_control.h
  isci: merge sata.[ch] into request.c
  isci: kill 'get/set' macros
  isci: retire scic_sds_ and scic_ prefixes
  isci: unify isci_host and scic_sds_controller
  isci: unify isci_remote_device and scic_sds_remote_device
  isci: unify isci_port and scic_sds_port
  isci: fix scic_sds_remote_device_terminate_requests
  isci: unify isci_phy and scic_sds_phy
  isci: unify isci_request and scic_sds_request
  isci: rename / clean up scic_sds_stp_request
  isci: preallocate requests
  isci: combine request flags
  isci: unify can_queue tracking on the tci_pool, uplevel tag assignment
  isci: Terminate dev requests on FIS err bit rx in NCQ
  isci: fix frame received locking
  ...

8 years agoMerge branch 'at91/fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/linux...
Linus Torvalds [Mon, 4 Jul 2011 22:54:18 +0000]
Merge branch 'at91/fixes' of git://git./linux/kernel/git/arm/linux-2.6-arm-soc

* 'at91/fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/linux-2.6-arm-soc:
  AT91: Change nand buswidth logic to match hardware default configuration
  at91: Use "pclk" as con_id on at91cap9 and at91rm9200
  at91: fix udc, ehci and mmc clock device name for cap9/9g45/9rl
  atmel_serial: fix internal port num
  at91: fix at91_set_serial_console: use platform device id

8 years agoMerge branch 'fbdev-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Mon, 4 Jul 2011 22:53:53 +0000]
Merge branch 'fbdev-fixes-for-linus' of git://git./linux/kernel/git/lethal/fbdev-3.x

* 'fbdev-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/lethal/fbdev-3.x:
  vesafb: fix memory leak
  fbdev: amba: Link fb device to its parent
  fsl-diu-fb: remove check for pixel clock ranges
  udlfb: Correct sub-optimal resolution selection.
  hecubafb: add module_put on error path in hecubafb_probe()
  sm501fb: fix section mismatch warning
  gx1fb: Fix section mismatch warnings
  fbdev: sh_mobile_meram: Correct pointer check for YCbCr chroma plane

8 years agoRDMA: Check for NULL mode in .devnode methods
Goldwyn Rodrigues [Mon, 4 Jul 2011 16:26:57 +0000]
RDMA: Check for NULL mode in .devnode methods

Commits 71c29bd5c235 ("IB/uverbs: Add devnode method to set path/mode")
and c3af0980ce01 ("IB: Add devnode methods to cm_class and umad_class")
added devnode methods that set the mode.

However, these methods don't check for a NULL mode, and so we get a
crash when unloading modules because devtmpfs_delete_node() calls
device_get_devnode() with mode == NULL.

Add the missing checks.

Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.de>
[ Also fix cm.c.  - Roland ]
Signed-off-by: Roland Dreier <roland@purestorage.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

8 years agoAT91: Change nand buswidth logic to match hardware default configuration
Nicolas Ferre [Fri, 1 Jul 2011 10:25:24 +0000]
AT91: Change nand buswidth logic to match hardware default configuration

The recently modified nand buswitth configuration is not aligned with
board reality: the double footprint on boards is always populated with 8bits
buswidth nand flashes.
So we have to consider that without particular configuration the 8bits
buswidth is selected by default.
Moreover, the previous logic was always using !board_have_nand_8bit(), we
change it to a simpler: board_have_nand_16bit().

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Tested-by: Ludovic Desroches <ludovic.desroches@atmel.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>

8 years agovesafb: fix memory leak
Daniel J Blueman [Mon, 27 Jun 2011 23:08:53 +0000]
vesafb: fix memory leak

When releasing framebuffer, free colourmap allocations.

Signed-off-by: Daniel J Blueman <daniel.blueman@gmail.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

8 years agonatsemi: silence dma-debug warnings
FUJITA Tomonori [Mon, 4 Jul 2011 05:34:29 +0000]
natsemi: silence dma-debug warnings

This silences dma-debug warnings:

https://lkml.org/lkml/2011/6/30/341

------------[ cut here ]------------
WARNING: at /home/jimc/projects/lx/linux-2.6/lib/dma-debug.c:820
check_unmap+0x1fe/0x56a()
natsemi 0000:00:06.0: DMA-API: device driver frees DMA memory with
different size [device address=0x0000000006ef0040] [map size=1538
bytes] [unmap size=1522 bytes]
Modules linked in: pc8736x_gpio pc87360 hwmon_vid scx200_gpio nsc_gpio
scx200_hrt scx200_acb i2c_core arc4 rtl8180 mac80211 eeprom_93cx6
cfg80211 pcspkr rfkill scx200 ide_gd_mod ide_pci_generic ohci_hcd
usbcore sc1200 ide_core
Pid: 870, comm: collector Not tainted 3.0.0-rc5-sk-00080-gca56a95 #1
Call Trace:
 [<c011a556>] warn_slowpath_common+0x4a/0x5f
 [<c02565cb>] ? check_unmap+0x1fe/0x56a
 [<c011a5cf>] warn_slowpath_fmt+0x26/0x2a
 [<c02565cb>] check_unmap+0x1fe/0x56a
 [<c0256aaa>] debug_dma_unmap_page+0x53/0x5b
 [<c029d6cd>] pci_unmap_single+0x4d/0x57
 [<c029ea0a>] natsemi_poll+0x343/0x5ca
 [<c0116f41>] ? try_to_wake_up+0xea/0xfc
 [<c0122416>] ? spin_unlock_irq.clone.28+0x18/0x23
 [<c02d4667>] net_rx_action+0x3f/0xe5
 [<c011e35e>] __do_softirq+0x5b/0xd1
 [<c011e303>] ? local_bh_enable+0xa/0xa
 <IRQ>  [<c011e54b>] ? irq_exit+0x34/0x75
 [<c01034b9>] ? do_IRQ+0x66/0x79
 [<c034e869>] ? common_interrupt+0x29/0x30
 [<c0115ed0>] ? finish_task_switch.clone.118+0x31/0x72
 [<c034cb92>] ? schedule+0x3b2/0x3f1
 [<c012f4b0>] ? hrtimer_start_range_ns+0x10/0x12
 [<c012f4ce>] ? hrtimer_start_expires+0x1c/0x24
 [<c034d5aa>] ? schedule_hrtimeout_range_clock+0x8e/0xb4
 [<c012ed27>] ? update_rmtp+0x68/0x68
 [<c034d5da>] ? schedule_hrtimeout_range+0xa/0xc
 [<c017a913>] ? poll_schedule_timeout+0x27/0x3e
 [<c017b051>] ? do_select+0x488/0x4cd
 [<c0115ee2>] ? finish_task_switch.clone.118+0x43/0x72
 [<c01157ad>] ? need_resched+0x14/0x1e
 [<c017a99e>] ? poll_freewait+0x74/0x74
 [<c01157ad>] ? need_resched+0x14/0x1e
 [<c034cbc1>] ? schedule+0x3e1/0x3f1
 [<c011e55e>] ? irq_exit+0x47/0x75
 [<c01157ad>] ? need_resched+0x14/0x1e
 [<c034cf8a>] ? preempt_schedule_irq+0x44/0x4a
 [<c034dd1e>] ? need_resched+0x17/0x19
 [<c024bc12>] ? put_dec_full+0x7b/0xaa
 [<c0240060>] ? blkdev_ioctl+0x434/0x618
 [<c024bc70>] ? put_dec+0x2f/0x6d
 [<c024c6a5>] ? number.clone.1+0x10b/0x1d0
 [<c034cf8a>] ? preempt_schedule_irq+0x44/0x4a
 [<c034dd1e>] ? need_resched+0x17/0x19
 [<c024d046>] ? vsnprintf+0x225/0x264
 [<c024cea0>] ? vsnprintf+0x7f/0x264
 [<c018346f>] ? seq_printf+0x22/0x40
 [<c01a2fcc>] ? do_task_stat+0x582/0x5a3
 [<c017a913>] ? poll_schedule_timeout+0x27/0x3e
 [<c017b1b5>] ? core_sys_select+0x11f/0x1a3
 [<c017a913>] ? poll_schedule_timeout+0x27/0x3e
 [<c01a34a1>] ? proc_tgid_stat+0xd/0xf
 [<c012357c>] ? recalc_sigpending+0x32/0x35
 [<c0123b9c>] ? __set_task_blocked+0x64/0x6a
 [<c011dfb0>] ? timespec_add_safe+0x24/0x48
 [<c0123449>] ? spin_unlock_irq.clone.16+0x18/0x23
 [<c017b3a1>] ? sys_pselect6+0xe5/0x13e
 [<c034dd65>] ? syscall_call+0x7/0xb
 [<c0340000>] ? rpc_clntdir_depopulate+0x26/0x30
---[ end trace 180dcac41a50938b ]---

Reported-by: Jim Cromie <jim.cromie@gmail.com>
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Tested-by: Jim Cromie <jim.cromie@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

8 years agonet: 8139too: Initial necessary vlan_features to support vlan
Shan Wei [Fri, 1 Jul 2011 22:06:44 +0000]
net: 8139too: Initial necessary vlan_features to support vlan

Offload setting of vlan device requires
vlan_features to be initialized.

Signed-off-by: Shan Wei <shanwei@cn.fujitsu.com>
Acked-by: Francois Romieu <romieu@fr.zoreil.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

8 years agoMerge branch 'pm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/suspe...
Linus Torvalds [Sun, 3 Jul 2011 20:33:16 +0000]
Merge branch 'pm-fixes' of git://git./linux/kernel/git/rafael/suspend-2.6

* 'pm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/suspend-2.6:
  PM / Runtime: Update documentation regarding driver removal
  PM: Documentation: fix typo: pm_runtime_idle_sync() doesn't exist.

8 years ago[SCSI] isci: fix checkpatch errors
James Bottomley [Sun, 3 Jul 2011 19:14:45 +0000]
[SCSI] isci: fix checkpatch errors

Signed-off-by: James Bottomley <JBottomley@Parallels.com>

8 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/djbw/isci
James Bottomley [Sun, 3 Jul 2011 19:19:29 +0000]
Merge git://git./linux/kernel/git/djbw/isci

8 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6
Linus Torvalds [Sun, 3 Jul 2011 18:12:21 +0000]
Merge git://git./linux/kernel/git/sfrench/cifs-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6:
  cifs: set socket send and receive timeouts before attempting connect

8 years agoMerge branch 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jdelv...
Linus Torvalds [Sun, 3 Jul 2011 18:12:06 +0000]
Merge branch 'hwmon-for-linus' of git://git./linux/kernel/git/jdelvare/staging

* 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jdelvare/staging:
  hwmon: (k10temp) Update documentation for Fam12h
  hwmon-vid: Fix typo in VIA CPU name
  hwmon: (f71882fg) Add support for the F71869A
  hwmon: Use <> rather than () around my e-mail address
  hwmon: (emc6w201) Properly handle all errors

8 years agohwmon: (k10temp) Update documentation for Fam12h
Clemens Ladisch [Sun, 3 Jul 2011 11:32:54 +0000]
hwmon: (k10temp) Update documentation for Fam12h

Add some CPU series IDs and links to the Fam12h datasheets.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Jean Delvare <khali@linux-fr.org>

8 years agohwmon-vid: Fix typo in VIA CPU name
Jean Delvare [Sun, 3 Jul 2011 11:32:54 +0000]
hwmon-vid: Fix typo in VIA CPU name

It's Nehemiah, not Nemiah.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Guenter Roeck <guenter.roeck@ericsson.com>

8 years agohwmon: (f71882fg) Add support for the F71869A
Hans de Goede [Sun, 3 Jul 2011 11:32:53 +0000]
hwmon: (f71882fg) Add support for the F71869A

The F71869A is almost the same as the F71869F/E, except that it has
the normal number of temp and pwm zones for a F71882FG derived chip,
rather then the limited number of the F71869F/E.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Tested-by: Max Baldwin <archerseven@gmail.com>
Acked-by: Guenter Roeck <guenter.roeck@ericsson.com>
Signed-off-by: Jean Delvare <khali@linux-fr.org>

8 years agohwmon: Use <> rather than () around my e-mail address
Hans de Goede [Sun, 3 Jul 2011 11:32:53 +0000]
hwmon: Use <> rather than () around my e-mail address

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jean Delvare <khali@linux-fr.org>

8 years agohwmon: (emc6w201) Properly handle all errors
Jean Delvare [Sun, 3 Jul 2011 11:32:53 +0000]
hwmon: (emc6w201) Properly handle all errors

Handle errors on 8-bit register reads and writes too. Also use likely
and unlikely to make the functions faster on success.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Guenter Roeck <guenter.roeck@ericsson.com>

8 years agoisci: Device reset should request sas_phy_reset(phy, true)
Jeff Skirvin [Fri, 1 Jul 2011 20:03:44 +0000]
isci: Device reset should request sas_phy_reset(phy, true)

The hard_reset parameter passed to the LLDD in the direct-attached
phy control case allows the LLDD to filter link failure events
while the direct-attached device reset is executing.

Signed-off-by: Jeff Skirvin <jeffrey.d.skirvin@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: pare back error messsages
Dan Williams [Fri, 1 Jul 2011 19:07:25 +0000]
isci: pare back error messsages

The messages emitted from task.c and some from request.c likely
duplicate (in a less undertandable way) what is reported by the
midlayer.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: cleanup silicon revision detection
Dan Williams [Fri, 1 Jul 2011 18:41:21 +0000]
isci: cleanup silicon revision detection

Perform checking per-pci device (even though all systems will only have
1 pci device in this generation), and delete support for silicon that
does not report a proper revision (i.e. A0).

Reported-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: merge scu_unsolicited_frame.h into unsolicited_frame_control.h
Dan Williams [Fri, 1 Jul 2011 18:15:12 +0000]
isci: merge scu_unsolicited_frame.h into unsolicited_frame_control.h

Does not need its own file.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: merge sata.[ch] into request.c
Dan Williams [Fri, 1 Jul 2011 17:52:55 +0000]
isci: merge sata.[ch] into request.c

Undo some needless separation.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: kill 'get/set' macros
Dan Williams [Fri, 1 Jul 2011 09:25:15 +0000]
isci: kill 'get/set' macros

Most of these simple dereference macros are longer than their open coded
equivalent.  Deleting enum sci_controller_mode is thrown in for good
measure.

Reported-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: retire scic_sds_ and scic_ prefixes
Dan Williams [Fri, 1 Jul 2011 02:14:33 +0000]
isci: retire scic_sds_ and scic_ prefixes

The distinction between scic_sds_ scic_ and sci_ are no longer relevant
so just unify the prefixes on sci_.  The distinction between isci_ and
sci_ is historically significant, and useful for comparing the old
'core' to the current Linux driver. 'sci_' represents the former core as
well as the routines that are closer to the hardware and protocol than
their 'isci_' brethren. sci == sas controller interface.

Also unwind the 'sds1' out of the parameter structs.

Reported-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: unify isci_host and scic_sds_controller
Dan Williams [Fri, 1 Jul 2011 00:38:32 +0000]
isci: unify isci_host and scic_sds_controller

Remove the distinction between these two implementations and unify on
isci_host (local instances named ihost).  Hmmm, we had two
'oem_parameters' instances, one was unused... nice.

Reported-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: unify isci_remote_device and scic_sds_remote_device
Dan Williams [Thu, 30 Jun 2011 23:31:37 +0000]
isci: unify isci_remote_device and scic_sds_remote_device

Remove the distinction between these two implementations and unify on
isci_remote_device (local instances named idev).

Reported-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: unify isci_port and scic_sds_port
Dan Williams [Wed, 29 Jun 2011 20:09:25 +0000]
isci: unify isci_port and scic_sds_port

Remove the distinction between these two implementations and unify on
isci_port (local instances named iport).  The duplicate '->owning_port' and
'->isci_port' in both isci_phy and isci_remote_device will be fixed in a later
patch... this is just the straightforward rename/unification.

Reported-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: fix scic_sds_remote_device_terminate_requests
Dan Williams [Wed, 29 Jun 2011 16:45:48 +0000]
isci: fix scic_sds_remote_device_terminate_requests

Commit 0815632 "isci: unify remote_device stop_handlers" introduced the
possibility that not all requests get terminated if we reach the
request_count.  Now that we properly reference count devices we don't
need this self-defense and can do the straightforward scan of all active
requests.

Reported-by: Jeff Skirvin <jeffrey.d.skirvin@intel.com>
Acked-by: Jeff Skirvin <jeffrey.d.skirvin@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: unify isci_phy and scic_sds_phy
Dan Williams [Tue, 28 Jun 2011 22:05:53 +0000]
isci: unify isci_phy and scic_sds_phy

They are one in the same object so remove the distinction.  The near
duplicate fields (owning_port, and isci_port) will be cleaned up
after the scic_sds_port isci_port unification.

Reported-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: unify isci_request and scic_sds_request
Dan Williams [Mon, 27 Jun 2011 21:57:03 +0000]
isci: unify isci_request and scic_sds_request

They are one in the same object so remove the distinction.  The near
duplicate fields (owning_controller, and isci_host) will be cleaned up
after the scic_sds_contoller isci_host unification.

Reported-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: rename / clean up scic_sds_stp_request
Dan Williams [Mon, 27 Jun 2011 18:56:41 +0000]
isci: rename / clean up scic_sds_stp_request

* Rename scic_sds_stp_request to isci_stp_request
* Remove the unused fields and union indirection

Reported-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: preallocate requests
Dan Williams [Fri, 17 Jun 2011 21:18:39 +0000]
isci: preallocate requests

the dma_pool interface is optimized for object_size << page_size which
is not the case with isci_request objects and the dma_pool routines show
up in the top of the profile.

The old io_request_table which tracked whether tci slots were in-flight
or not is replaced with an IREQ_ACTIVE flag per request.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: combine request flags
Dan Williams [Thu, 23 Jun 2011 21:33:48 +0000]
isci: combine request flags

Combine three bools into one unsigned long 'flags'.  Doesn't increase the
request size due to packing. (to do: optimize the structure layout).

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: unify can_queue tracking on the tci_pool, uplevel tag assignment
Dan Williams [Tue, 28 Jun 2011 20:47:09 +0000]
isci: unify can_queue tracking on the tci_pool, uplevel tag assignment

The tci_pool tracks our outstanding command slots which are also the 'index'
portion of our tags.  Grabbing the tag early in ->lldd_execute_task let's us
drop the isci_host_can_queue() and ->was_tag_assigned_by_user infrastructure.
->was_tag_assigned_by_user required the task context to be duplicated in
request-local buffer.  With the tci established early we can build the
task_context directly into its final location and skip a memcpy.

With the task context buffer at a known address at request construction we
have the opportunity/obligation to also fix sgl handling.  This rework feels
like it belongs in another patch but the sgl handling and task_context are too
intertwined.
1/ fix the 'ab' pair embedded in the task context to point to the 'cd' pair in
   the task context (previously we were prematurely linking to the staging
   buffer).
2/ fix the broken iteration of pio sgls that assumes all sgls are relative to
   the request, and does a dangerous looking reverse lookup of physical
   address to virtual address.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: Terminate dev requests on FIS err bit rx in NCQ
Jeff Skirvin [Fri, 24 Jun 2011 00:09:02 +0000]
isci: Terminate dev requests on FIS err bit rx in NCQ

When the remote device transitions to a not-ready state because of
an NCQ error condition, all outstanding requests to that device
are terminated and completed to libsas on the normal path.  The
device then waits for a READ LOG EXT command to issue on the task
management path.

Signed-off-by: Jeff Skirvin <jeffrey.d.skirvin@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: fix frame received locking
Dan Williams [Fri, 24 Jun 2011 06:44:52 +0000]
isci: fix frame received locking

Updates to the frame_rcvd before need to be atomic with respect to when
they are evaluated by libsas.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: possible buffer overflow in isci_parse_oem_parameters fixed
Maciej Patelczyk [Tue, 21 Jun 2011 22:03:13 +0000]
isci: possible buffer overflow in isci_parse_oem_parameters fixed

scu_index is a parameter of isci_parse_eom_parameters and is an index
in controller table. There is a check: scu_index > SCI_MAX_CONTROLLERS
which is insufficient and should be: scu_index >= SCI_MAX_CONTROLLERS.
scu_index is used as an index in the table which size is
SCI_MAX_CONTROLLERS.

Signed-off-by: Maciej Patelczyk <maciej.patelczyk@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: fix isci_task_execute_tmf completion
Dan Williams [Tue, 21 Jun 2011 23:23:03 +0000]
isci: fix isci_task_execute_tmf completion

1/ fix the timeout for wait_for_completion_timeout
2/ In the tmf timeout case we need to wait for our termination callback
3/ Once the request is successfully started it will be freed according to the
   normal lifetime for requests.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: fix support for arbitrarily large smp requests
Dan Williams [Thu, 16 Jun 2011 23:59:56 +0000]
isci: fix support for arbitrarily large smp requests

Instead of duplicating the smp request buffer reuse the one provided by
libsas.  This future proofs the driver to support arbitrarily large smp
requests, and shrinks the request structure size by ~700 bytes.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: fix dma_unmap_sg usage
Dan Williams [Fri, 17 Jun 2011 17:40:43 +0000]
isci: fix dma_unmap_sg usage

One bug and a cleanup:
1/ Fix cases where we were unmapping invalid addresses (smp requests were
   being unmapped)

[  604.662770] ------------[ cut here ]------------
[  604.668026] WARNING: at lib/dma-debug.c:800 check_unmap+0x418/0x740()
[  604.675315] Hardware name: SandyBridge Platform
[  604.680465] isci 0000:03:00.0: DMA-API: device driver tries to free an invalid DMA memory address

2/ The unmap routine is too large to be an inline function, and
   isci_request_io_request_get_next_sge is unused.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: fix smp response frame overrun
Dan Williams [Fri, 17 Jun 2011 00:20:35 +0000]
isci: fix smp response frame overrun

Due to a typo we currently copy way too much when copying over the
response data, but since a request is likely backed by a full page
allocation we don't corrupt live data.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: kill device_sequence
Dan Williams [Fri, 17 Jun 2011 20:34:43 +0000]
isci: kill device_sequence

Now that we have upleveled device reassignment protection to the
isci_remote_device reference count we no longer need this level of
self-defense.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: kill isci_remote_device_change_state()
Dan Williams [Thu, 16 Jun 2011 18:26:12 +0000]
isci: kill isci_remote_device_change_state()

Now that "stopping/stopped" are one in the same and signalled by a NULL device
pointer the rest of the device status infrastructure can be removed (->status
and ->state_lock).  The "not ready for i/o state" is replaced with a state
flag, and is evaluated under scic_lock so that we don't see transients from
taking the device reference to submitting the i/o.

This also fixes a potential leakage of can_queue slots in the rare case that
SAS_TASK_ABORTED is set at submission.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: atomic device lookup and reference counting
Dan Williams [Tue, 14 Jun 2011 00:39:44 +0000]
isci: atomic device lookup and reference counting

We have unsafe references to remote devices that are notified to
disappear at lldd_dev_gone.  In order to clean this up we need a single
canonical source for device lookups and stable references once a lookup
succeeds.  Towards that end guarantee that domain_device.lldd_dev is
NULL as soon as we start the process of stopping a device.  Any code
path that wants to safely lookup a remote device must do so through
task->dev->lldd_dev (isci_lookup_device()).

For in-flight references outside of scic_lock we need reference counting
to ensure that the device is not recycled before we are done with it.
Simplify device back references to just scic_sds_request.target_device
which is now the only permissible internal reference that is maintained
relative to the reference count.

There were two occasions where we wanted new i/o's to be treated as
SAS_TASK_UNDELIVERED but where the domain_dev->lldd_dev link is still
intact.  Introduce a 'gone' flag to prevent i/o while waiting for libsas
to take action on the port down event.

One 'core' leftover is that we currently call
scic_remote_device_destruct() from isci_remote_device_deconstruct()
which is called when the 'core' says the device is stopped.  It would be
more natural for the final put to trigger
isci_remote_device_deconstruct() but this implementation is deferred as
it requires other changes.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: fix ssp response iu buffer size in isci_tmf
Dan Williams [Wed, 15 Jun 2011 18:11:03 +0000]
isci: fix ssp response iu buffer size in isci_tmf

In isci_task_request_complete() we save the response/sense data from the
command.  Make sure isci_tmf has enough space to hold the full response.

[ it does not look like we actually use this data, and
  response_data_len/sense_data_len should be specifying the byte count,
  in any event do the simple fix first so we don't corrupt memory ]

Reported-by: Adam Gruchala <adam.gruchala@intel.com>
Tested-by: Edmund Nadolski <edmund.nadolski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: cleanup request allocation
Dan Williams [Mon, 13 Jun 2011 07:51:30 +0000]
isci: cleanup request allocation

Rather than return an error code and update a pointer that was passed by
reference just return the request object directly (or null if allocation
failed).

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: cleanup/optimize queue increment macros
Dan Williams [Thu, 9 Jun 2011 23:04:28 +0000]
isci: cleanup/optimize queue increment macros

Every single i/o or event completion incurs a test and branch to see if
the cycle bit changed.  For power-of-2 queue sizes the cycle bit can be
read directly from the rollover of the queue pointer.

Likely premature optimization, but the hidden if() and hidden
assignments / side-effects in the macros were already asking to be
cleaned up.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: cleanup tag macros
Dan Williams [Thu, 9 Jun 2011 18:06:58 +0000]
isci: cleanup tag macros

A tag is a 16 bit number where the upper four bits is a sequence number
and the remainder is the task context index (tci).  Sanitize the macro
names and shave 256-bytes out of scic_sds_controller by reducing the size of
io_request_sequence.

scic_sds_io_tag_construct --> ISCI_TAG
scic_sds_io_tag_get_sequence --> ISCI_TAG_SEQ
scic_sds_io_tag_get_index() --> ISCI_TAG_TCI
scic_sds_io_sequence_increment() [delete / open code]

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: cleanup/optimize pool implementation
Dan Williams [Wed, 8 Jun 2011 01:50:55 +0000]
isci: cleanup/optimize pool implementation

The circ_buf macros are ~6% faster, as measured by perf, because they take
advantage of power-of-two math assumptions i.e. no test and branch for
rollover. Their semantics are clearer than the hidden side effects in pool.h
(like sci_pool_get() which hides an assignment).

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: Disable link layer hang detection
Jeff Skirvin [Mon, 20 Jun 2011 21:09:31 +0000]
isci: Disable link layer hang detection

Some targets exceed the hang detect timer.  Use the OS timeout to
catch hung tasks.

Signed-off-by: Jeff Skirvin <jeffrey.d.skirvin@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: Hard reset failure will link reset all phys in the port
Jeff Skirvin [Mon, 20 Jun 2011 21:09:26 +0000]
isci: Hard reset failure will link reset all phys in the port

In the case where the hard reset process fails, each link in
the port is put through a link reset sequence.

Signed-off-by: Jeff Skirvin <jeffrey.d.skirvin@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: Explicitly decode remote node ready and suspended states
Jeff Skirvin [Mon, 20 Jun 2011 21:09:22 +0000]
isci: Explicitly decode remote node ready and suspended states

The remote node context should only signal a device reset condition
in a suspended state.

Signed-off-by: Jeff Skirvin <jeffrey.d.skirvin@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: fix isci_terminate_pending() list management
Dan Williams [Mon, 20 Jun 2011 22:11:22 +0000]
isci: fix isci_terminate_pending() list management

Walk through the list of pending requests being careful to consider that
multiple requests can be terminated when the lock is dropped (i.e.
invalidating the 'next' reference established by
list_for_each_entry_safe).

Also noticed that all callers to isci_terminate_pending_requests()
specifying terminating, so just drop the parameter.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: Handle timed-out request terminations correctly
Jeff Skirvin [Mon, 20 Jun 2011 21:09:16 +0000]
isci: Handle timed-out request terminations correctly

In the situation where a termination of an I/O times-out,
make sure that the linkage from the request to the task
is severed completely.  Also make sure that the selection
of tasks to terminate occurs under scic_lock.

Signed-off-by: Jeff Skirvin <jeffrey.d.skirvin@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: Requests that do not start must be set to "complete"
Jeff Skirvin [Mon, 20 Jun 2011 21:09:11 +0000]
isci: Requests that do not start must be set to "complete"

Requests that fail at start because of a reset pending condition
must be set to complete in order to allow for later cleanup.

Signed-off-by: Jeff Skirvin <jeffrey.d.skirvin@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: Add decode for SMP request retry error condition
Jeff Skirvin [Mon, 20 Jun 2011 21:09:06 +0000]
isci: Add decode for SMP request retry error condition

There are situations with slow expanders in which a first attempt
to execute an SMP request will fail with a timeout.  Immediate
subsequent retries will generally succeed.  This change makes sure
SMP I/O failures are immediately failed to libsas so that retries
happen with no discovery process timeout delay.

Signed-off-by: Jeff Skirvin <jeffrey.d.skirvin@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

8 years agoisci: filter broadcast change notifications during SMP phy resets
Jeff Skirvin [Tue, 21 Jun 2011 19:16:33 +0000]
isci: filter broadcast change notifications during SMP phy resets

When resetting a sata device in the domain we have seen occasions where
libsas prematurely marks a device gone in the time it takes for the
device to re-establish the link.  This plays badly with software raid
arrays.  Other libsas drivers have non-uniform delays in their reset
handlers to try to cover this condition, but not sufficient to close the
hole.  Given that a sata device can take many seconds to recover we
filter bcns and poll for the device reattach state before notifying
libsas that the port needs the domain to be rediscovered.  Once this has
been proven out at the lldd level we can think about uplevelling this
feature to a common implementation in libsas.

Signed-off-by: Jeff Skirvin <jeffrey.d.skirvin@intel.com>
[ use kzalloc instead of kmem_cache ]
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
[ use eventq and time macros ]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>