cascardo/linux.git
14 years agosched: move rq_weight data array out of .percpu
Jiri Kosina [Wed, 28 Oct 2009 15:26:00 +0000 (00:26 +0900)]
sched: move rq_weight data array out of .percpu

Commit 34d76c41 introduced percpu array update_shares_data, size of which
being proportional to NR_CPUS. Unfortunately this blows up ia64 for large
NR_CPUS configuration, as ia64 allows only 64k for .percpu section.

Fix this by allocating this array dynamically and keep only pointer to it
percpu.

The per-cpu handling doesn't impose significant performance penalty on
potentially contented path in tg_shares_up().

...
ffffffff8104337c:       65 48 8b 14 25 20 cd    mov    %gs:0xcd20,%rdx
ffffffff81043383:       00 00
ffffffff81043385:       48 c7 c0 00 e1 00 00    mov    $0xe100,%rax
ffffffff8104338c:       48 c7 45 a0 00 00 00    movq   $0x0,-0x60(%rbp)
ffffffff81043393:       00
ffffffff81043394:       48 c7 45 a8 00 00 00    movq   $0x0,-0x58(%rbp)
ffffffff8104339b:       00
ffffffff8104339c:       48 01 d0                add    %rdx,%rax
ffffffff8104339f:       49 8d 94 24 08 01 00    lea    0x108(%r12),%rdx
ffffffff810433a6:       00
ffffffff810433a7:       b9 ff ff ff ff          mov    $0xffffffff,%ecx
ffffffff810433ac:       48 89 45 b0             mov    %rax,-0x50(%rbp)
ffffffff810433b0:       bb 00 04 00 00          mov    $0x400,%ebx
ffffffff810433b5:       48 89 55 c0             mov    %rdx,-0x40(%rbp)
...

After:

...
ffffffff8104337c:       65 8b 04 25 28 cd 00    mov    %gs:0xcd28,%eax
ffffffff81043383:       00
ffffffff81043384:       48 98                   cltq
ffffffff81043386:       49 8d bc 24 08 01 00    lea    0x108(%r12),%rdi
ffffffff8104338d:       00
ffffffff8104338e:       48 8b 15 d3 7f 76 00    mov    0x767fd3(%rip),%rdx        # ffffffff817ab368 <update_shares_data>
ffffffff81043395:       48 8b 34 c5 00 ee 6d    mov    -0x7e921200(,%rax,8),%rsi
ffffffff8104339c:       81
ffffffff8104339d:       48 c7 45 a0 00 00 00    movq   $0x0,-0x60(%rbp)
ffffffff810433a4:       00
ffffffff810433a5:       b9 ff ff ff ff          mov    $0xffffffff,%ecx
ffffffff810433aa:       48 89 7d c0             mov    %rdi,-0x40(%rbp)
ffffffff810433ae:       48 c7 45 a8 00 00 00    movq   $0x0,-0x58(%rbp)
ffffffff810433b5:       00
ffffffff810433b6:       bb 00 04 00 00          mov    $0x400,%ebx
ffffffff810433bb:       48 01 f2                add    %rsi,%rdx
ffffffff810433be:       48 89 55 b0             mov    %rdx,-0x50(%rbp)
...

Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Tejun Heo <tj@kernel.org>
14 years agopercpu: allow pcpu_alloc() to be called with IRQs off
Jiri Kosina [Wed, 28 Oct 2009 15:25:59 +0000 (00:25 +0900)]
percpu: allow pcpu_alloc() to be called with IRQs off

pcpu_alloc() and pcpu_extend_area_map() perform a series of
spin_lock_irq()/spin_unlock_irq() calls, which make them unsafe
with respect to being called from contexts which have IRQs off.

This patch converts the code to perform save/restore of flags instead,
making pcpu_alloc() (or __alloc_percpu() respectively) to be called
from early kernel startup stage, where IRQs are off.

This is needed for proper initialization of per-cpu rq_weight data from
sched_init().

tj: added comment explaining why irqsave/restore is used in alloc path.

Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Tejun Heo <tj@kernel.org>
14 years agovirtio-net: fix data corruption with OOM
Michael S. Tsirkin [Wed, 28 Oct 2009 11:03:38 +0000 (04:03 -0700)]
virtio-net: fix data corruption with OOM

virtio net used to unlink skbs from send queues on error,
but ever since 48925e372f04f5e35fec6269127c62b2c71ab794
we do not do this. This causes guest data corruption and crashes
with vhost since net core can requeue the skb or free it without
it being taken off the list.

This patch fixes this by queueing the skb after successful
transmit.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosfc: Set ip_summed correctly for page buffers passed to GRO
Ben Hutchings [Wed, 28 Oct 2009 10:43:49 +0000 (03:43 -0700)]
sfc: Set ip_summed correctly for page buffers passed to GRO

Page buffers containing packets with an incorrect checksum or using a
protocol not handled by hardware checksum offload were previously not
passed to LRO.  The conversion to GRO changed this, but did not set
the ip_summed value accordingly.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agocnic: Fix L2CTX_STATUSB_NUM offset in context memory.
Michael Chan [Wed, 28 Oct 2009 10:41:59 +0000 (03:41 -0700)]
cnic: Fix L2CTX_STATUSB_NUM offset in context memory.

The BNX2_L2CTX_STATUSB_NUM definition needs to be changed to match
the recent firmware update:

commit 078b0735881c7969aaf21469f3577831cddd9f8c
bnx2: Update firmware to 5.0.0.j3.

Without the fix, bnx2 can crash intermittently in bnx2_rx_int() when
iSCSI is enabled.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agodrm/kms: fix kms/fbdev colormap support properly.
Dave Airlie [Fri, 23 Oct 2009 08:49:03 +0000 (18:49 +1000)]
drm/kms: fix kms/fbdev colormap support properly.

This sets the fbcon to use TRUECOLOR by default, it then
only modifies the pseudo palette for fbcon, and only touches
the real palette when in 8-bit pseudo color mode.

Signed-off-by: Dave Airlie <airlied@redhat.com>
14 years agodrm: Add the basic check for the detailed timing in EDID
Zhao Yakui [Wed, 14 Oct 2009 01:11:25 +0000 (09:11 +0800)]
drm: Add the basic check for the detailed timing in EDID

Sometimes we will get the incorrect display modeline when parsing the detailed
timing in EDID. For example:
   >hsync/vsync width is zero
   >sync is beyond the blank.

So add the basic check for the detailed timing in EDID to avoid the incorrect
display modeline.

Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
14 years agodrm/radeon/kms: ignore vga arbiter return.
Dave Airlie [Wed, 28 Oct 2009 01:09:58 +0000 (11:09 +1000)]
drm/radeon/kms: ignore vga arbiter return.

Since we register all radeon devices, and the arbiter only cares about
VGA class ones, we will fail to startup on display controller class devices.
We don't gain anything by using the return value here.

this helps kms on sparc64 get started.

Reported-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Dave Airlie <airlied@redhat.com>
14 years agoxen: set up mmu_ops before trying to set any ptes
Jeremy Fitzhardinge [Tue, 27 Oct 2009 23:54:19 +0000 (16:54 -0700)]
xen: set up mmu_ops before trying to set any ptes

xen_setup_stackprotector() ends up trying to set page protections,
so we need to have vm_mmu_ops set up before trying to do so.
Failing to do so causes an early boot crash.

[ Impact: Fix early crash under Xen. ]

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
14 years agoMerge commit 'gcl/merge' into merge
Benjamin Herrenschmidt [Tue, 27 Oct 2009 22:56:18 +0000 (09:56 +1100)]
Merge commit 'gcl/merge' into merge

14 years agoMAINTAINERS: rt2x00 list is moderated
Bartlomiej Zolnierkiewicz [Mon, 26 Oct 2009 19:14:33 +0000 (20:14 +0100)]
MAINTAINERS: rt2x00 list is moderated

Cc: users@rt2x00.serialmonkey.com
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoairo: Reorder tests, check bounds before element
Roel Kluin [Mon, 26 Oct 2009 14:28:11 +0000 (15:28 +0100)]
airo: Reorder tests, check bounds before element

Test whether index is within bounds before reading the element

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agomac80211: fix for incorrect sequence number on hostapd injected frames
Björn Smedman [Sat, 24 Oct 2009 18:55:09 +0000 (20:55 +0200)]
mac80211: fix for incorrect sequence number on hostapd injected frames

When hostapd injects a frame, e.g. an authentication or association
response, mac80211 looks for a suitable access point virtual interface
to associate the frame with based on its source address. This makes it
possible e.g. to correctly assign sequence numbers to the frames.

A small typo in the ethernet address comparison statement caused a
failure to find a suitable ap interface. Sequence numbers on such
frames where therefore left unassigned causing some clients
(especially windows-based 11b/g clients) to reject them and fail to
authenticate or associate with the access point. This patch fixes the
typo in the address comparison statement.

Signed-off-by: Björn Smedman <bjorn.smedman@venatech.se>
Reviewed-by: Johannes Berg <johannes@sipsolutions.net>
Cc: stable@kernel.org
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agolibertas spi: fix sparse errors
Holger Schurig [Thu, 22 Oct 2009 13:30:44 +0000 (15:30 +0200)]
libertas spi: fix sparse errors

This fixes the following sparse warnings:

$ make modules SUBDIRS=drivers/net/wireless/libertas C=1 CF=-D__CHECK_ENDIAN__
make: Entering directory `/usr/src/linux-wl'
  CHECK   drivers/net/wireless/libertas/if_spi.c
drivers/net/wireless/libertas/if_spi.c:137:16: warning: incorrect type in initializer (different base types)
drivers/net/wireless/libertas/if_spi.c:137:16:    expected unsigned short [unsigned] [usertype] reg_out
drivers/net/wireless/libertas/if_spi.c:137:16:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:191:16: warning: incorrect type in initializer (different base types)
drivers/net/wireless/libertas/if_spi.c:191:16:    expected unsigned short [unsigned] [usertype] reg_out
drivers/net/wireless/libertas/if_spi.c:191:16:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:256:24: warning: incorrect type in argument 1 (different base types)
drivers/net/wireless/libertas/if_spi.c:256:24:    expected restricted __le32 const [usertype] *p
drivers/net/wireless/libertas/if_spi.c:256:24:    got unsigned int *<noident>
drivers/net/wireless/libertas/if_spi.c:243:24: warning: incorrect type in argument 1 (different base types)
drivers/net/wireless/libertas/if_spi.c:243:24:    expected restricted __le16 const [usertype] *p
drivers/net/wireless/libertas/if_spi.c:243:24:    got unsigned short *<noident>
drivers/net/wireless/libertas/if_spi.c:171:7: warning: incorrect type in assignment (different base types)
drivers/net/wireless/libertas/if_spi.c:171:7:    expected unsigned short [unsigned] [usertype] buff
drivers/net/wireless/libertas/if_spi.c:171:7:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:171:7: warning: incorrect type in assignment (different base types)
drivers/net/wireless/libertas/if_spi.c:171:7:    expected unsigned short [unsigned] [usertype] buff
drivers/net/wireless/libertas/if_spi.c:171:7:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:171:7: warning: incorrect type in assignment (different base types)
drivers/net/wireless/libertas/if_spi.c:171:7:    expected unsigned short [unsigned] [usertype] buff
drivers/net/wireless/libertas/if_spi.c:171:7:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:171:7: warning: incorrect type in assignment (different base types)
drivers/net/wireless/libertas/if_spi.c:171:7:    expected unsigned short [unsigned] [usertype] buff
drivers/net/wireless/libertas/if_spi.c:171:7:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:171:7: warning: incorrect type in assignment (different base types)
drivers/net/wireless/libertas/if_spi.c:171:7:    expected unsigned short [unsigned] [usertype] buff
drivers/net/wireless/libertas/if_spi.c:171:7:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:243:24: warning: incorrect type in argument 1 (different base types)
drivers/net/wireless/libertas/if_spi.c:243:24:    expected restricted __le16 const [usertype] *p
drivers/net/wireless/libertas/if_spi.c:243:24:    got unsigned short *<noident>
drivers/net/wireless/libertas/if_spi.c:171:7: warning: incorrect type in assignment (different base types)
drivers/net/wireless/libertas/if_spi.c:171:7:    expected unsigned short [unsigned] [usertype] buff
drivers/net/wireless/libertas/if_spi.c:171:7:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:171:7: warning: incorrect type in assignment (different base types)
drivers/net/wireless/libertas/if_spi.c:171:7:    expected unsigned short [unsigned] [usertype] buff
drivers/net/wireless/libertas/if_spi.c:171:7:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:171:7: warning: incorrect type in assignment (different base types)
drivers/net/wireless/libertas/if_spi.c:171:7:    expected unsigned short [unsigned] [usertype] buff
drivers/net/wireless/libertas/if_spi.c:171:7:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:171:7: warning: incorrect type in assignment (different base types)
drivers/net/wireless/libertas/if_spi.c:171:7:    expected unsigned short [unsigned] [usertype] buff
drivers/net/wireless/libertas/if_spi.c:171:7:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:171:7: warning: incorrect type in assignment (different base types)
drivers/net/wireless/libertas/if_spi.c:171:7:    expected unsigned short [unsigned] [usertype] buff
drivers/net/wireless/libertas/if_spi.c:171:7:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:171:7: warning: incorrect type in assignment (different base types)
drivers/net/wireless/libertas/if_spi.c:171:7:    expected unsigned short [unsigned] [usertype] buff
drivers/net/wireless/libertas/if_spi.c:171:7:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:243:24: warning: incorrect type in argument 1 (different base types)
drivers/net/wireless/libertas/if_spi.c:243:24:    expected restricted __le16 const [usertype] *p
drivers/net/wireless/libertas/if_spi.c:243:24:    got unsigned short *<noident>
drivers/net/wireless/libertas/if_spi.c:171:7: warning: incorrect type in assignment (different base types)
drivers/net/wireless/libertas/if_spi.c:171:7:    expected unsigned short [unsigned] [usertype] buff
drivers/net/wireless/libertas/if_spi.c:171:7:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:171:7: warning: incorrect type in assignment (different base types)
drivers/net/wireless/libertas/if_spi.c:171:7:    expected unsigned short [unsigned] [usertype] buff
drivers/net/wireless/libertas/if_spi.c:171:7:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:243:24: warning: incorrect type in argument 1 (different base types)
drivers/net/wireless/libertas/if_spi.c:243:24:    expected restricted __le16 const [usertype] *p
drivers/net/wireless/libertas/if_spi.c:243:24:    got unsigned short *<noident>
drivers/net/wireless/libertas/if_spi.c:243:24: warning: incorrect type in argument 1 (different base types)
drivers/net/wireless/libertas/if_spi.c:243:24:    expected restricted __le16 const [usertype] *p
drivers/net/wireless/libertas/if_spi.c:243:24:    got unsigned short *<noident>
drivers/net/wireless/libertas/if_spi.c:171:7: warning: incorrect type in assignment (different base types)
drivers/net/wireless/libertas/if_spi.c:171:7:    expected unsigned short [unsigned] [usertype] buff
drivers/net/wireless/libertas/if_spi.c:171:7:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:171:7: warning: incorrect type in assignment (different base types)
drivers/net/wireless/libertas/if_spi.c:171:7:    expected unsigned short [unsigned] [usertype] buff
drivers/net/wireless/libertas/if_spi.c:171:7:    got restricted __le16 [usertype] <noident>
drivers/net/wireless/libertas/if_spi.c:243:24: warning: incorrect type in argument 1 (different base types)
drivers/net/wireless/libertas/if_spi.c:243:24:    expected restricted __le16 const [usertype] *p
drivers/net/wireless/libertas/if_spi.c:243:24:    got unsigned short *<noident>

Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agomac80211: trivial: fix spelling in mesh_hwmp
Andrey Yurovsky [Tue, 20 Oct 2009 19:17:34 +0000 (12:17 -0700)]
mac80211: trivial: fix spelling in mesh_hwmp

Fix a typo in the description of hwmp_route_info_get(), no function
changes.

Signed-off-by: Andrey Yurovsky <andrey@cozybit.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agocfg80211: sme: deauthenticate on assoc failure
Johannes Berg [Tue, 20 Oct 2009 06:08:53 +0000 (15:08 +0900)]
cfg80211: sme: deauthenticate on assoc failure

When the in-kernel SME gets an association failure from
the AP we don't deauthenticate, and thus get into a very
confused state which will lead to warnings later on. Fix
this by actually deauthenticating when the AP indicates
an association failure.

(Brought to you by the hacking session at Kernel Summit 2009 in Tokyo,
Japan. -- JWL)

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agomac80211: keep auth state when assoc fails
Johannes Berg [Tue, 20 Oct 2009 06:08:12 +0000 (15:08 +0900)]
mac80211: keep auth state when assoc fails

When association fails, we should stay authenticated,
which in mac80211 is represented by the existence of
the mlme work struct, so we cannot free that, instead
we need to just set it to idle.

(Brought to you by the hacking session at Kernel Summit 2009 in Tokyo,
Japan. -- JWL)

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agomac80211: fix ibss joining
Reinette Chatre [Mon, 19 Oct 2009 21:55:37 +0000 (14:55 -0700)]
mac80211: fix ibss joining

Recent commit "mac80211: fix logic error ibss merge bssid check" fixed
joining of ibss cell when static bssid is provided. In this case
ifibss->bssid is set before the cell is joined and comparing that address
to a bss should thus always succeed. Unfortunately this change broke the
other case of joining a ibss cell without providing a static bssid where
the value of ifibss->bssid is not set before the cell is joined.

Since ifibss->bssid may be set before or after joining the cell we do not
learn anything by comparing it to a known bss. Remove this check.

Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agob43: add 'struct b43_wl' missing declaration
Miguel Boton [Fri, 9 Oct 2009 14:13:53 +0000 (16:13 +0200)]
b43: add 'struct b43_wl' missing declaration

'struct b43_wl' declaration is missing at 'leds.h'.
It should be declared to avoid getting some GCC warnings at 'b43_leds_unregister'.

Signed-off-by: Miguel Botón <mboton@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agob43: Fix Bugzilla #14181 and the bug from the previous 'fix'
Larry Finger [Fri, 16 Oct 2009 15:18:09 +0000 (10:18 -0500)]
b43: Fix Bugzilla #14181 and the bug from the previous 'fix'

"b43: Fix PPC crash in rfkill polling on unload" fixed the bug reported
in Bugzilla No. 14181; however, it introduced a new bug. Whenever the
radio switch was turned off, it was necessary to unload and reload
the driver for it to recognize the switch again.

This patch fixes both the original bug in #14181 and the bug introduced by
the previous patch. It must be stated, however, that if there is a BCM4306/3
with an rfkill switch (not yet proven), then the driver will need an
unload/reload cycle to turn the device back on.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agort2x00: Fix crypto in TX frame for rt2800usb
Benoit PAPILLAULT [Thu, 15 Oct 2009 19:17:09 +0000 (21:17 +0200)]
rt2x00: Fix crypto in TX frame for rt2800usb

Signed-off-by: Benoit PAPILLAULT <benoit.papillault@free.fr>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
14 years agoRevert "PCI: get larger bridge ranges when space is available"
Jesse Barnes [Tue, 27 Oct 2009 16:39:18 +0000 (09:39 -0700)]
Revert "PCI: get larger bridge ranges when space is available"

This reverts commit 308cf8e13f42f476dfd6552aeff58fdc0788e566.  This
patch had trouble with transparent bridges, among other things.  A more
readable and correct version should land in 2.6.33.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
14 years agopowerpc/ppc64: Use preempt_schedule_irq instead of preempt_schedule
Benjamin Herrenschmidt [Mon, 26 Oct 2009 19:41:17 +0000 (19:41 +0000)]
powerpc/ppc64: Use preempt_schedule_irq instead of preempt_schedule

Based on an original patch by Valentine Barshak <vbarshak@ru.mvista.com>

Use preempt_schedule_irq to prevent infinite irq-entry and
eventual stack overflow problems with fast-paced IRQ sources.

This kind of problems has been observed on the PASemi Electra IDE
controller. We have to make sure we are soft-disabled before calling
preempt_schedule_irq and hard disable interrupts after that
to avoid unrecoverable exceptions.

This patch also moves the "clrrdi r9,r1,THREAD_SHIFT" out of
the #ifdef CONFIG_PPC_BOOK3E scope, since r9 is clobbered
and has to be restored in both cases.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
14 years agopowerpc: Minor cleanup to lib/Kconfig.debug
Kumar Gala [Fri, 16 Oct 2009 07:21:39 +0000 (07:21 +0000)]
powerpc: Minor cleanup to lib/Kconfig.debug

We don't need an explicit PPC64 in the DEBUG_PREEMPT dependancies as all
PPC platforms now support TRACE_IRQFLAGS_SUPPORT.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
14 years agopowerpc: Minor cleanup to sound/ppc/Kconfig
Kumar Gala [Fri, 16 Oct 2009 07:21:38 +0000 (07:21 +0000)]
powerpc: Minor cleanup to sound/ppc/Kconfig

We can replace PPC32 || PPC64 as a dependancy with just PPC as all
powerpc platforms (32-bit and 64-bit) define PPC now.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
14 years agopowerpc: Minor cleanup to init/Kconfig
Kumar Gala [Fri, 16 Oct 2009 07:21:37 +0000 (07:21 +0000)]
powerpc: Minor cleanup to init/Kconfig

We dont need to depend on PPC64 explicitly as all powerpc platforms
(32-bit and 64-bit) define PPC now.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
14 years agopowerpc: Limit memory hotplug support to PPC64 Book-3S machines
Kumar Gala [Fri, 16 Oct 2009 07:21:36 +0000 (07:21 +0000)]
powerpc: Limit memory hotplug support to PPC64 Book-3S machines

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
14 years agopowerpc: Limit hugetlbfs support to PPC64 Book-3S machines
Kumar Gala [Fri, 16 Oct 2009 07:21:35 +0000 (07:21 +0000)]
powerpc: Limit hugetlbfs support to PPC64 Book-3S machines

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
14 years agopowerpc: Fix compile errors found by new ppc64e_defconfig
Kumar Gala [Fri, 16 Oct 2009 07:05:17 +0000 (07:05 +0000)]
powerpc: Fix compile errors found by new ppc64e_defconfig

Fix the following 3 issues:

arch/powerpc/kernel/process.c: In function 'arch_randomize_brk':
arch/powerpc/kernel/process.c:1183: error: 'mmu_highuser_ssize' undeclared (first use in this function)
arch/powerpc/kernel/process.c:1183: error: (Each undeclared identifier is reported only once
arch/powerpc/kernel/process.c:1183: error: for each function it appears in.)
arch/powerpc/kernel/process.c:1183: error: 'MMU_SEGSIZE_1T' undeclared (first use in this function)

In file included from arch/powerpc/kernel/setup_64.c:60:
arch/powerpc/include/asm/mmu-hash64.h:132: error: redefinition of 'struct mmu_psize_def'
arch/powerpc/include/asm/mmu-hash64.h:159: error: expected identifier or '(' before numeric constant
arch/powerpc/include/asm/mmu-hash64.h:396: error: conflicting types for 'mm_context_t'
arch/powerpc/include/asm/mmu-book3e.h:184: error: previous declaration of 'mm_context_t' was here

cc1: warnings being treated as errors
arch/powerpc/kernel/pci_64.c: In function 'pcibios_unmap_io_space':
arch/powerpc/kernel/pci_64.c:100: error: unused variable 'res'

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
14 years agopowerpc: Add a Book-3E 64-bit defconfig
Kumar Gala [Fri, 16 Oct 2009 07:05:16 +0000 (07:05 +0000)]
powerpc: Add a Book-3E 64-bit defconfig

This defconfig's purpose at this time is to help catch compile errors
between Book-3S and Book-3E support in ppc64.  It is based on the
ppc64_defconfig with some things disabled that we dont support on
Book-3E right now (hugetlbfs, slices, etc.)

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
14 years agopowerpc/booke: Fix xmon single step on PowerPC Book-E
Josh Boyer [Mon, 5 Oct 2009 04:46:05 +0000 (04:46 +0000)]
powerpc/booke: Fix xmon single step on PowerPC Book-E

Prior to the arch/ppc -> arch/powerpc transition, xmon had support for single
stepping on 4xx boards.  The functionality was lost when arch/ppc was removed.
This patch restores single step support for 44x boards, and Book-E in general.

Signed-off-by: Josh Boyer <jwboyer@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
14 years agopowerpc: Align vDSO base address
Andreas Schwab [Sun, 4 Oct 2009 02:35:41 +0000 (02:35 +0000)]
powerpc: Align vDSO base address

The ABI specifies a 64K alignment, we need to map the vDSO accordingly

Signed-off-by: Andreas Schwab <schwab@linux-m68k.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
14 years agopowerpc: Fix segment mapping in vdso32
Andreas Schwab [Fri, 2 Oct 2009 23:25:31 +0000 (23:25 +0000)]
powerpc: Fix segment mapping in vdso32

Due to missing segment assignments the .text section was put in the NOTES
segment (and marked as NOTE section), and the .got was put in the DYNAMIC
segment.

Signed-off-by: Andreas Schwab <schwab@linux-m68k.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
14 years agopowerpc/iseries: Remove compiler version dependent hack
Stephen Rothwell [Mon, 5 Oct 2009 20:28:47 +0000 (20:28 +0000)]
powerpc/iseries: Remove compiler version dependent hack

The creation of the flattened device tree depended on the compiler
putting the constant strings for an object in a section with a
particular name.  This was changed with recent compilers.  Do this
explicitly instead.

Without this patch, iseries kernels may silently not boot when built with
some compilers.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
14 years agopowerpc/perf_events: Fix priority of MSR HV vs PR bits
Michael Neuling [Wed, 14 Oct 2009 19:32:15 +0000 (19:32 +0000)]
powerpc/perf_events: Fix priority of MSR HV vs PR bits

The architecture defines that if MSR PR is set we are in problem state
irrespective of the HV bit.  This fixes perf events to reflect this.

Also, on bare metal systems, samples taken in Linux will now be reported
as kernel rather than hypervisor.

Signed-off-by: Michael Neuling <mikey@neuling.org>
CC: paulus@samba.org
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
14 years agosh_eth: Add asm/cacheflush.h
Nobuhiro Iwamatsu [Mon, 26 Oct 2009 13:49:50 +0000 (13:49 +0000)]
sh_eth: Add asm/cacheflush.h

Add include asm/cacheflush.h,  because declaration of __flush_purge_region
moved to asm/cacheflush.h.

Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoPPPoE: Fix flush/close races.
Michal Ostrowski [Mon, 26 Oct 2009 23:23:20 +0000 (16:23 -0700)]
PPPoE: Fix flush/close races.

Be more careful about the state of pointers during tear-down.
The "pppoe_dev" field can only be looked at safely while holding socket locks.
This subsequently allows for the flush_lock to be killed.

We depend on the PPPOX_CONNECTED state to tell us that that those fields are
valid, so whoever clears that state (pppox_unbind_sock()) is responsible for
the dev_put() call.

We also have to ensure that we delete_item() on all sockets before they are
cleaned up.

The need for these changes has been exposed by scenarios wherein namespace
bindings of ethernet devices change while there are ongoing PPPoE sessions,
which resulted in oopses due to unusual socket connection termination paths,
exposing these issues.

Signed-off-by: Michal Ostrowski <mostrows@gmail.com>
Reviewed-by: Cyril Gorcunov <gorcunov@gmail.com>
Reported-by: Denys Fedoryschenko <denys@visp.net.lb>
Tested-by: Denys Fedoryschenko <denys@visp.net.lb>
14 years agomfd: Do not dereference null pointer in twl4030 error path
Ilkka Koskinen [Tue, 20 Oct 2009 13:22:52 +0000 (16:22 +0300)]
mfd: Do not dereference null pointer in twl4030 error path

We are mistakenly dereferencing twl->client in the twl->client null checking
path.

Signed-off-by: Ilkka Koskinen <ilkka.koskinen@nokia.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
14 years agomfd: Always initialise WM831x IRQ mutex
Mark Brown [Mon, 19 Oct 2009 10:07:05 +0000 (11:07 +0100)]
mfd: Always initialise WM831x IRQ mutex

This avoids crashes when running without interrupt support.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
14 years agoe1000e: allow for swflag to be held over consecutive PHY accesses
Bruce Allan [Mon, 26 Oct 2009 11:24:02 +0000 (11:24 +0000)]
e1000e: allow for swflag to be held over consecutive PHY accesses

PCH-based parts (82577/82578) and some ICH8-based parts (82566) need to
hold the swflag (sw/fw/hw hardware semaphore) over consecutive PHY accesses
in order to perform sw-driven PHY configuration during initialization to
workaround known hardware issues (see follow-on patch).  This patch
provides new PHY read/write functions (and function pointers) that will
allow accessing the PHY registers assuming the swflag has already been
acquired.  The actual PHY register access code has moved into helper
functions that are called with a flag indicating whether or not the swflag
has already been acquired and acquires/releases it if not.

The functions called from within the updated PHY access functions had to be
updated to assume the swflag was already acquired, and other functions that
called those functions were also updated to acquire/release the swflag.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: separate mutex usage between NVM and PHY/CSR register for ICHx/PCH
Bruce Allan [Mon, 26 Oct 2009 11:23:43 +0000 (11:23 +0000)]
e1000e: separate mutex usage between NVM and PHY/CSR register for ICHx/PCH

Accesses to NVM and PHY/CSR registers on ICHx/PCH-based parts are protected
from concurrent accesses with a mutex that is acquired when the access is
initiated and released when the access has completed.  However, the two
types of accesses should not be protected by the same mutex because the
driver may have to access the NVM while already holding the mutex over
several consecutive PHY/CSR accesses which would result in livelock.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: 82577/82578 requires a different method to configure LPLU
Bruce Allan [Mon, 26 Oct 2009 11:23:25 +0000 (11:23 +0000)]
e1000e: 82577/82578 requires a different method to configure LPLU

Unlike previous ICHx-based parts, the PCH-based parts (82577/82578) require
LPLU (Low Power Link Up, or "reverse auto-negotiation") to be configured in
the PHY rather than the MAC.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: increase swflag acquisition timeout for ICHx/PCH
Bruce Allan [Mon, 26 Oct 2009 11:23:06 +0000 (11:23 +0000)]
e1000e: increase swflag acquisition timeout for ICHx/PCH

In some conditions (e.g. when AMT is enabled on the system), it is possible
to take an extended period of time to for the driver to acquire the sw/fw/hw
hardware semaphore used to protect against concurrent access of a shared
resource (e.g. PHY registers).  This could cause PHY registers to not get
configured properly resulting in link issues.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: clear PHY wakeup bit after LCD reset on 82577/82578
Bruce Allan [Mon, 26 Oct 2009 11:22:47 +0000 (11:22 +0000)]
e1000e: clear PHY wakeup bit after LCD reset on 82577/82578

Performing a dummy read of the PHY Wakeup Control (WUC) register clears the
wakeup enable bit set by an PHY reset.  If this bit remains set, link
problems may occur.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoigbvf: fix memory leak when ring size changed while interface down
Alexander Duyck [Mon, 26 Oct 2009 11:32:25 +0000 (11:32 +0000)]
igbvf: fix memory leak when ring size changed while interface down

This patch resolves a memory leak which occurs while changing the ring size
while the interface is down.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoixgbe: fix memory leak when resizing rings while interface is down
Alexander Duyck [Mon, 26 Oct 2009 11:32:05 +0000 (11:32 +0000)]
ixgbe: fix memory leak when resizing rings while interface is down

This patch resolves a memory leak that occurs when you resize the rings via
the ethtool -G option while the interface is down.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoigb: fix memory leak when setting ring size while interface is down
Alexander Duyck [Mon, 26 Oct 2009 11:31:47 +0000 (11:31 +0000)]
igb: fix memory leak when setting ring size while interface is down

Changing ring sizes while the interface was down was causing a double
allocation of the receive and transmit rings.  This issue is amplified when
there are multiple rings enabled.  To prevent this we need to add an
additional check which will just update the ring counts when the interface
is not up and skip the allocation steps.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosh: Fix hugetlbfs dependencies for SH-3 && MMU configurations.
Paul Mundt [Mon, 26 Oct 2009 22:22:37 +0000 (07:22 +0900)]
sh: Fix hugetlbfs dependencies for SH-3 && MMU configurations.

The hugetlb dependencies presently depend on SUPERH && MMU while the
hugetlb page size definitions depend on CPU_SH4 or CPU_SH5. This
unfortunately allows SH-3 + MMU configurations to enable hugetlbfs
without a corresponding HPAGE_SHIFT definition, resulting in the build
blowing up.

As SH-3 doesn't support variable page sizes, we tighten up the
dependenies a bit to prevent hugetlbfs from being enabled. These days
we also have a shiny new SYS_SUPPORTS_HUGETLBFS, so switch to using
that rather than adding to the list of corner cases in fs/Kconfig.

Reported-by: Kristoffer Ericson <kristoffer.ericson@gmail.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
14 years agosh: Document uImage.bin target in archhelp.
Paul Mundt [Mon, 26 Oct 2009 22:17:12 +0000 (07:17 +0900)]
sh: Document uImage.bin target in archhelp.

This was missing from the previous patch.

Reported-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
14 years agoblock: use after free bug in __blkdev_get
Neil Brown [Mon, 26 Oct 2009 07:59:17 +0000 (08:59 +0100)]
block: use after free bug in __blkdev_get

commit 0762b8bde9729f10f8e6249809660ff2ec3ad735
(from 14 months ago) introduced a use-after-free bug which has just
recently started manifesting in my md testing.
I tried git bisect to find out what caused the bug to start
manifesting, and it could have been the recent change to
blk_unregister_queue (48c0d4d4c04) but the results were inconclusive.

This patch certainly fixes my symptoms and looks correct as the two
calls are now in the same order as elsewhere in that function.

Signed-off-by: NeilBrown <neilb@suse.de>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
14 years agoNFSv4: The link() operation should return any delegation on the file
Trond Myklebust [Mon, 26 Oct 2009 12:09:46 +0000 (08:09 -0400)]
NFSv4: The link() operation should return any delegation on the file

Otherwise, we have to wait for the server to recall it.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
14 years agoNFSv4: Fix two unbalanced put_rpccred() issues.
Trond Myklebust [Mon, 26 Oct 2009 12:09:46 +0000 (08:09 -0400)]
NFSv4: Fix two unbalanced put_rpccred() issues.

Commits 29fba38b (nfs41: lease renewal) and fc01cea9 (nfs41: sequence
operation) introduce a couple of put_rpccred() calls on credentials for
which there is no corresponding get_rpccred().

See http://bugzilla.kernel.org/show_bug.cgi?id=14249

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
14 years agosh: add uImage.bin target
Magnus Damm [Mon, 26 Oct 2009 10:12:39 +0000 (10:12 +0000)]
sh: add uImage.bin target

Add an uImage.bin target to allow uncompressed uImages.
Useful for boards with busted u-boot decompression like
the rsk7203 on my desk.

Signed-off-by: Magnus Damm <damm@opensource.se>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
14 years agosh: rsk7203 CONFIG_MTD=n fix
Magnus Damm [Mon, 26 Oct 2009 10:08:55 +0000 (10:08 +0000)]
sh: rsk7203 CONFIG_MTD=n fix

Fix the rsk7203 board code to build with CONFIG_MTD=n.

Signed-off-by: Magnus Damm <damm@opensource.se>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
14 years agosh: Check for return_to_handler when unwinding the stack
Matt Fleming [Sat, 24 Oct 2009 18:56:57 +0000 (18:56 +0000)]
sh: Check for return_to_handler when unwinding the stack

When CONFIG_FUNCTION_GRAPH_TRACER is enabled the function graph tracer
may patch return addresses on the stack with the address of
return_to_handler(). This really confuses the DWARF unwinder because it
will try find the caller of return_to_handler(), not the caller of the
real return address.

So teach the DWARF unwinder how to find the real return address whenever
it encounters return_to_handler().

This patch does not cope very well when multiple return addresses on the
stack have been patched. To make it work properly it would require state
to track how many return_to_handler()'s have been seen so that we'd know
where to look in current->curr_ret_stack[]. So for now, instead of
trying to handle this, just moan if more than one return address on the
stack has been patched.

Signed-off-by: Matt Fleming <matt@console-pimps.org>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
14 years agosh: Build fix: define more __movmem* symbols
Lubomir Rintel [Thu, 22 Oct 2009 11:58:37 +0000 (11:58 +0000)]
sh: Build fix: define more __movmem* symbols

ERROR: "__movmemSI12" [net/unix/unix.ko] undefined!
ERROR: "__movmemSI52" [net/ipv6/sit.ko] undefined!
ERROR: "__movmemSI24" [net/ipv6/ipv6.ko] undefined!
ERROR: "__movmemSI60" [net/ipv6/ipv6.ko] undefined!
ERROR: "__movmemSI16" [net/ipv6/ipv6.ko] undefined!
ERROR: "__movmemSI20" [net/ipv6/ipv6.ko] undefined!
ERROR: "__movmemSI32" [net/ipv6/ipv6.ko] undefined!

Signed-off-by: Lubomir Rintel <lkundrak@v3.sk>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
14 years agosh: __irq_entry annotate do_IRQ().
Paul Mundt [Mon, 26 Oct 2009 00:58:31 +0000 (09:58 +0900)]
sh: __irq_entry annotate do_IRQ().

This adds an __irq_entry annotation for do_IRQ() so that the IRQ
annotation in the function graph tracer works as advertized. We already
have the IRQENTRY section wired up, so this is just a trivial addition
to actually make use of it.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
14 years agopcmcia: do not load the pd6729 driver if io_base is NULL
Komuro [Fri, 23 Oct 2009 23:07:39 +0000 (08:07 +0900)]
pcmcia: do not load the pd6729 driver if io_base is NULL

The CL-PD6729 chip in some docking station is not initialized properly
under Linux. In that case, do not load the pd6729 driver.

[Dominik Brodowski <linux@dominikbrodowski.net>: spelling fixes, check for NULL not 0]
Signed-off-by: Komuro <komurojun-mbn@nifty.com>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
14 years agobonding: Modify hash transmit policies to use the packet's source MAC address
Jasper Spaans [Fri, 23 Oct 2009 04:08:46 +0000 (04:08 +0000)]
bonding: Modify hash transmit policies to use the packet's source MAC address

Modify bonding hash transmit policies to use the psource MAC address of
the packet instead of the MAC address configured for the bonding device.

The old sitation conflicts with the documentation.

Signed-off-by: Jasper Spaans <spaans@fox-it.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agopktgen: Dont leak kernel memory
Eric Dumazet [Sat, 24 Oct 2009 13:55:20 +0000 (06:55 -0700)]
pktgen: Dont leak kernel memory

While playing with pktgen, I realized IP ID was not filled and a
random value was taken, possibly leaking 2 bytes of kernel memory.

We can use an increasing ID, this can help diagnostics anyway.

Also clear packet payload, instead of leaking kernel memory.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoDM9000: Fix revision ID for DM9000B
Ben Dooks [Sat, 24 Oct 2009 13:53:07 +0000 (06:53 -0700)]
DM9000: Fix revision ID for DM9000B

The DM9000B revision ID is 0x1A, not 0x1B as set in the curernt
dm9000.h header.

Fix bug reported by Paolo Zebelloni.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
Signed-off-by: Simtec Linux Team <linux@simtec.co.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agor8169: fix Ethernet Hangup for RTL8110SC rev d
Simon Wunderlich [Sat, 24 Oct 2009 13:47:33 +0000 (06:47 -0700)]
r8169: fix Ethernet Hangup for RTL8110SC rev d

The 8110SC rev d chip on our board shows a regression which the 8110SB chip
did not have. When inbound traffic is overflowing the receive descriptor queue,
"holes" in the ring buffer may occur which lead to a hangup until the buffer
is filled again. The packets are than completely processed, but the ring
remains porous and no packets are processed until the next overflow. Setting
the interface down and up can fix the problem temporary from userspace.

For some reason we don't know, this behaviour is not occuring if the RxVlan
bit for hardware VLAN untagging is set. There is another "Work around for
AMD plateform" in the current code which checks the VLAN status
word in receive descriptors, but does never come to effect when hardware
VLAN support is enabled. We assume that this is a bug in the chip.

The following patch fixes the problem. Without the patch we could reproduce
the hang within minutes (given other devices also generating lots of
interrupts), without we couldn't reproduce within a few days of long term
testing.

This version contains minor style adjustments and is sent with mutt which
will hopefully not destroy the formatting again.

Signed-off-by: Bernhard Schmidt <bernhard.schmidt@saxnet.de>
Signed-off-by: Simon Wunderlich <simon.wunderlich@saxnet.de>
Acked-by: Francois Romieu <romieu@zoreil.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoblock: silently error unsupported empty barriers too
Mark McLoughlin [Sat, 24 Oct 2009 12:14:31 +0000 (14:14 +0200)]
block: silently error unsupported empty barriers too

With 2.6.32-rc5 in a KVM guest using dm and virtio_blk, we see the
following errors:

  end_request: I/O error, dev vda, sector 0
  end_request: I/O error, dev vda, sector 0

The errors go away if dm stops submitting empty barriers, by reverting:

  commit 52b1fd5a27c625c78373e024bf570af3c9d44a79
  Author: Mikulas Patocka <mpatocka@redhat.com>
    dm: send empty barriers to targets in dm_flush

We should silently error all barriers, even empty barriers, on devices
like virtio_blk which don't support them.

See also:

  https://bugzilla.redhat.com/514901

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Acked-by: Alasdair G Kergon <agk@redhat.com>
Acked-by: Mikulas Patocka <mpatocka@redhat.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Neil Brown <neilb@suse.de>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
14 years agohwmon: (it87) Fix VID reading on IT8718F/IT8720F
Jean Delvare [Sat, 24 Oct 2009 11:28:47 +0000 (13:28 +0200)]
hwmon: (it87) Fix VID reading on IT8718F/IT8720F

Comparing apples to bananas doesn't seem right. Consistently use the
chips enum for chip type comparisons, to avoid such bugs in the
future.

The bug has been there since support for the IT8718F was added, so
VID never worked for this chip nor for the similar IT8720F.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
14 years agohwmon: (dme1737) No vid attributes for SCH311x
Jean Delvare [Sat, 24 Oct 2009 11:28:46 +0000 (13:28 +0200)]
hwmon: (dme1737) No vid attributes for SCH311x

The SCH311x chips do not have VID inputs, so the cpu0_vid and vrm
attributes shouldn't be created for them.

This fixes lm-sensors ticket #2353:
http://www.lm-sensors.org/ticket/2353

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Tested-by: Udo van den Heuvel <udovdh@xs4all.nl>
Cc: Juerg Haefliger <juergh@gmail.com>
14 years agohwmon: (fschmd) Fix check on unsigned in watchdog_write()
Roel Kluin [Sat, 24 Oct 2009 11:28:45 +0000 (13:28 +0200)]
hwmon: (fschmd) Fix check on unsigned in watchdog_write()

If unsigned the watchdog_trigger() return value will not be
checked correctly.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
14 years agohwmon: (coretemp) Maintainer update
Huaxu Wan [Sat, 24 Oct 2009 11:28:45 +0000 (13:28 +0200)]
hwmon: (coretemp) Maintainer update

Intel will help maintaining the coretemp driver.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
14 years agoNFSv4: Fix a bug when the server returns NFS4ERR_RESOURCE
Trond Myklebust [Fri, 23 Oct 2009 18:46:42 +0000 (14:46 -0400)]
NFSv4: Fix a bug when the server returns NFS4ERR_RESOURCE

RFC 3530 states that when we recieve the error NFS4ERR_RESOURCE, we are not
supposed to bump the sequence number on OPEN, LOCK, LOCKU, CLOSE, etc
operations. The problem is that we map that error into EREMOTEIO in the XDR
layer, and so the NFSv4 middle-layer routines like seqid_mutating_err(),
and nfs_increment_seqid() don't recognise it.

The fix is to defer the mapping until after the middle layers have
processed the error.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
14 years agonfs: Panic when commit fails
Terry Loftin [Fri, 23 Oct 2009 01:36:01 +0000 (21:36 -0400)]
nfs: Panic when commit fails

Actually pass the NFS_FILE_SYNC option to the server to avoid a
Panic in nfs_direct_write_complete() when a commit fails.

At the end of an nfs write, if the nfs commit fails, all the writes
will be rescheduled.  They are supposed to be rescheduled as NFS_FILE_SYNC
writes, but the rpc_task structure is not completely intialized and so
the option is not passed.  When the rescheduled writes complete, the
return indicates that they are NFS_UNSTABLE and we try to do another
commit.  This leads to a Panic because the commit data structure pointer
was set to null in the initial (failed) commit attempt.

Signed-off-by: Terry Loftin <terry.loftin@hp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
14 years agoifb: should not use __dev_get_by_index() without locks
Eric Dumazet [Tue, 20 Oct 2009 02:35:50 +0000 (02:35 +0000)]
ifb: should not use __dev_get_by_index() without locks

At this point (ri_tasklet()), RTNL or dev_base_lock are not held,
we must use dev_get_by_index() instead of __dev_get_by_index()

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: au1000_eth: add missing capability.h
Manuel Lauss [Sat, 17 Oct 2009 02:00:07 +0000 (02:00 +0000)]
net: au1000_eth: add missing capability.h

fixes the following build failure:
  CC      drivers/net/au1000_eth.o
/drivers/net/au1000_eth.c: In function 'au1000_set_settings':
/drivers/net/au1000_eth.c:623: error: implicit declaration of function 'capable'
/drivers/net/au1000_eth.c:623: error: 'CAP_NET_ADMIN' undeclared (first use in this function)
/drivers/net/au1000_eth.c:623: error: (Each undeclared identifier is reported only once
/drivers/net/au1000_eth.c:623: error: for each function it appears in.

Signed-off-by: Manuel Lauss <manuel.lauss@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agomyri10ge: improve port type reporting in ethtool
Brice Goglin [Fri, 23 Oct 2009 04:43:43 +0000 (21:43 -0700)]
myri10ge: improve port type reporting in ethtool

Improve the reporting of myri10ge port type in ethtool,
and update for new boards.

Signed-off-by: Brice Goglin <brice@myri.com>
Signed-off-by: Andrew Gallatin <gallatin@myri.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: use WARN() for the WARN_ON in commit b6b39e8f3fbbb
Arjan van de Ven [Fri, 23 Oct 2009 04:37:56 +0000 (21:37 -0700)]
net: use WARN() for the WARN_ON in commit b6b39e8f3fbbb

Commit b6b39e8f3fbbb (tcp: Try to catch MSG_PEEK bug) added a printk()
to the WARN_ON() that's in tcp.c. This patch changes this combination
to WARN(); the advantage of WARN() is that the printk message shows up
inside the message, so that kerneloops.org will collect the message.

In addition, this gets rid of an extra if() statement.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoe1000e: reset the PHY on 82577/82578 when going to Sx
Bruce Allan [Fri, 23 Oct 2009 04:22:18 +0000 (21:22 -0700)]
e1000e: reset the PHY on 82577/82578 when going to Sx

The PHY on 82577/82578 parts needs a soft reset when transitioning to Sx
state in order for the PHY write which disables gigabit speed to take
effect.  Gigabit speed must be disabled in order for the PHY writes to
registers on page 800 (the wakeup control registers) to work as expected
otherwise the system might not wake via WoL.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoisdn: fix possible circular locking dependency
Xiaotian Feng [Wed, 21 Oct 2009 23:07:04 +0000 (23:07 +0000)]
isdn: fix possible circular locking dependency

There's a circular locking dependency:

---> isdn_net_get_locked_lp
    --->lock &nd->queue_lock
    --->lock &nd->queue->xmit_lock
    .....................
    ---->unlock &nd->queue_lock

---> isdn_net_writebuf_skb (called with &nd->queue->xmit_lock locked)
    ---->isdn_net_inc_frame_cnt
         ---->isdn_net_device_busy
              ----> lock &nd->queue_lock

This will trigger lockdep warnings:

 =======================================================
 [ INFO: possible circular locking dependency detected ]
 2.6.32-rc4-testing #7
 -------------------------------------------------------
 ipppd/28379 is trying to acquire lock:
 (&netdev->queue_lock){......}, at: [<e62ad0fd>] isdn_net_device_busy+0x2c/0x74 [isdn]

 but task is already holding lock:
 (&netdev->local->xmit_lock){+.....}, at: [<e62aefc2>] isdn_net_write_super+0x3f/0x6e [isdn]

 which lock already depends on the new lock.
 .......

 We don't need to lock nd->queue->xmit_lock to protect single
isdn_net_lp_busy(). This can fix above lockdep warnings.

Reported-and-tested-by: Tilman Schmidt <tilman@imap.cc>
Signed-off-by: Xiaotian Feng <xtfeng@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetxen: avoid undue board config check
Dhananjay Phadke [Wed, 21 Oct 2009 19:39:03 +0000 (19:39 +0000)]
netxen: avoid undue board config check

Old code assumed board config version in the flash to be 1.
When this will get changed by tools, driver just refuses to
attach. This is unnecessary since driver does not have to
parse board config structure directly (maintained by firmware).

Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetxen: fix tx timeout handling on firmware hang
Amit Kumar Salecha [Wed, 21 Oct 2009 19:39:02 +0000 (19:39 +0000)]
netxen: fix tx timeout handling on firmware hang

Clear NX_RESETING bit in netxen_tx_timeout_task() so that
the firmware watchdog task can catch need_reset request
from tx timeout.

Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonetxen: fix i2c init
Dhananjay Phadke [Wed, 21 Oct 2009 19:39:01 +0000 (19:39 +0000)]
netxen: fix i2c init

Avoid resetting subsys ID in i2c block. Also remove duplicate
check for address tranlsation error.

Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoomap4: Fix UART4 platform data on omap4
Santosh Shilimkar [Thu, 22 Oct 2009 21:48:14 +0000 (14:48 -0700)]
omap4: Fix UART4 platform data on omap4

This patch removes the unnecessary UART4 platform which is under
data is wrong because of this

There is a separate platform structure for UART4

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Reviewed-By: Kevin Hilman <khilman@deeprootsystems.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
14 years agoomap4: Allow omap_serial_early_init() for OMAP4430 board
Santosh Shilimkar [Thu, 22 Oct 2009 21:48:14 +0000 (14:48 -0700)]
omap4: Allow omap_serial_early_init() for OMAP4430 board

This patch enables omap_serial_early_init() function for OMAP4430
SDP. Without this the bootup would throw oops in omap_serial_init().

Note that the ifndef CONFIG_ARCH_OMAP4 is split into two sections
to enable omap_serial_early_init(). This ifndef cannot be removed
until omap4 clock framework is implemented.

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Reviewed-By: Tony Lindgren <tony@atomide.com>
Reviewed-By: Kevin Hilman <khilman@deeprootsystems.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
14 years agoomap3: PM: enable UART3 module wakeups
Kevin Hilman [Thu, 22 Oct 2009 21:48:13 +0000 (14:48 -0700)]
omap3: PM: enable UART3 module wakeups

UART3 is in the PER powerdomain.  If PER goes idle/inactive
independently of CORE, for UART3 to wakeup it must have its wakeup
enable bits setup in PM_WKEN_PER.  This patch enables these bits.

The reason it works when PER and CORE work together is because when
CORE goes inactive/retention, the IOPAD wakeups are enabled and
trigger UART3 wakeup.

Without this patch, when the UART inactivity timer fires for UART3,
its clocks are disabled and it's unable to wakeup so will be unusable
until PER is awoken by another source.

Another way of testing is by keeping CORE on during suspend but
allowing PER to hit retention

  # echo 3 > /debug/pm_debug/core_pwrdm/suspend

then enter suspend

  # echo mem > /sys/power/state

Without this patch, UART3 will be unable to wakeup the system.

Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
14 years agoomap2: Fix console serial port number for n8x0
Tony Lindgren [Thu, 22 Oct 2009 21:48:13 +0000 (14:48 -0700)]
omap2: Fix console serial port number for n8x0

With the recent changes omap serial ports match the physical
numbering like they should. Fix the kernel CMDLINE accordingly
so console works.

Signed-off-by: Tony Lindgren <tony@atomide.com>
14 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus
Linus Torvalds [Thu, 22 Oct 2009 22:35:16 +0000 (07:35 +0900)]
Merge git://git./linux/kernel/git/rusty/linux-2.6-for-linus

* git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus:
  move virtrng_remove to .devexit.text
  move virtballoon_remove to .devexit.text
  virtio_blk: Revert serial number support
  virtio: let header files include virtio_ids.h
  virtio_blk: revert QUEUE_FLAG_VIRT addition

14 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Thu, 22 Oct 2009 22:34:23 +0000 (07:34 +0900)]
Merge git://git./linux/kernel/git/davem/net-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (21 commits)
  niu: VLAN_ETH_HLEN should be used to make sure that the whole MAC header was copied to the head buffer in the Vlan packets case
  KS8851: Fix ks8851_set_rx_mode() for IFF_MULTICAST
  KS8851: Fix MAC address write order
  KS8851: Add soft reset at probe time
  net: fix section mismatch in fec.c
  net: Fix struct inet_timewait_sock bitfield annotation
  tcp: Try to catch MSG_PEEK bug
  net: Fix IP_MULTICAST_IF
  bluetooth: static lock key fix
  bluetooth: scheduling while atomic bug fix
  tcp: fix TCP_DEFER_ACCEPT retrans calculation
  tcp: reduce SYN-ACK retrans for TCP_DEFER_ACCEPT
  tcp: accept socket after TCP_DEFER_ACCEPT period
  Revert "tcp: fix tcp_defer_accept to consider the timeout"
  AF_UNIX: Fix deadlock on connecting to shutdown socket
  ethoc: clear only pending irqs
  ethoc: inline regs access
  vmxnet3: use dev_dbg, fix build for CONFIG_BLOCK=n
  virtio_net: use dev_kfree_skb_any() in free_old_xmit_skbs()
  be2net: fix support for PCI hot plug
  ...

14 years agoomap2: Fix detection of n8x0
Tony Lindgren [Thu, 22 Oct 2009 21:48:13 +0000 (14:48 -0700)]
omap2: Fix detection of n8x0

Otherwise the machine_is_nokia_n8*() does not work.

Signed-off-by: Tony Lindgren <tony@atomide.com>
14 years agoomap1: Fix DSP public peripherals support for ams-delta
Janusz Krzysztofik [Thu, 22 Oct 2009 21:47:42 +0000 (14:47 -0700)]
omap1: Fix DSP public peripherals support for ams-delta

DSP public peripherals used to work on OMAP1510 based (or all OMAP1 class?)
machines as long as old dspgateway code were present in the l-o tree. For
several months it is no longer included, breaking support for McBSP1 based
audio on Amstrad Delta, for example.

This patch, derived from the old dspgateway code, corrects the problem for the
board by simply taking the DSP out of reset state, I guess. That way, things
should not break when a new dsp code is added to the tree, and the change can
be reverted then.

If there are any reports on McBSP1 or other DSP public peripherals not working
for other OMAP1 machines (I've not heard of any for now), I can prepare a more
general patch providing an extra include file with a helper function defined.

Created and tested against linux-2.6.32-rc5

Signed-off-by: Janusz Krzysztofik <jkrzyszt@tis.icnet.pl>
Signed-off-by: Tony Lindgren <tony@atomide.com>
14 years agoomap1: Fix redundant UARTs pin muxing that can break other hardware support
Janusz Krzysztofik [Thu, 22 Oct 2009 21:47:42 +0000 (14:47 -0700)]
omap1: Fix redundant UARTs pin muxing that can break other hardware support

Commit 15ac408ee5a509053a765b816e9179515329369f removed enabled_uart
and OMAP_TAG_UART. This works for mach-omap2, but causes issues on
mach-omap1 for some boards as the mach-omap1 serial.c was muxing
pins based on the enabled_uart flag for 15xx.

Fix this by muxing pins in board-*.c files for the 15xx boards for
the uart ports that had enabled_uart flag set before the commit
above.

Tested on Amsdtrad Delta only.

Note that in the future we should add support for powering down
the uarts with a timer like mach-omap2/serial.c does. Otherwise
the enabled uarts will be blocking retention-while-idle.

Signed-off-by: Janusz Krzysztofik <jkrzyszt@tis.icnet.pl>
Signed-off-by: Tony Lindgren <tony@atomide.com>
14 years agoomap: iommu: fix wrong condition check for SUPERSECTION
Hiroshi DOYU [Thu, 22 Oct 2009 21:46:32 +0000 (14:46 -0700)]
omap: iommu: fix wrong condition check for SUPERSECTION

A bit (2 << 0) is set both on SECTION and SUPERSECTION. To identify
SUPERSECTION correctly, other bits should be compared too.

Reported-by: "Srinivas Pulukuru" <srinivas.pulukuru@ti.com>
Signed-off-by: Hiroshi DOYU <Hiroshi.DOYU@nokia.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
14 years agoomap: SDMA: Fix omap_stop_dma() API for channel linking
Santosh Shilimkar [Thu, 22 Oct 2009 21:46:31 +0000 (14:46 -0700)]
omap: SDMA: Fix omap_stop_dma() API for channel linking

OMAP sDMA driver API omap_stop_dma() doesn't really stop the dma when used
in linking scenario.

The DMA channel needs to be disabled before resetting the chain.
Also fix clearing of the OMAP_DMA_ACTIVE status in the linked case.

Cc: Hari n <hari.zoom@gmail.com>
Cc: Jarkko Nikula <jhnikula@gmail.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Signed-off-by: Venkatraman S <svenkatr@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
14 years agoomap: Fix omap-keypad by restoring old keypad.h without breaking omap2 boards that...
Janusz Krzysztofik [Thu, 22 Oct 2009 21:43:17 +0000 (14:43 -0700)]
omap: Fix omap-keypad by restoring old keypad.h without breaking omap2 boards that use matrix_keypad

Only mach-omap2 boards are currently using matrix_keypad. Allow
mach-omap1 boards to use the old style keypad.h without breaking.

Created against linux-2.6.32-rc5.
Compile tested with omap_3430sdp_defconfig and rx51_defconfig.

Signed-off-by: Janusz Krzysztofik <jkrzyszt@tis.icnet.pl>
Signed-off-by: Tony Lindgren <tony@atomide.com>
14 years agomove virtrng_remove to .devexit.text
Uwe Kleine-König [Thu, 1 Oct 2009 08:28:35 +0000 (10:28 +0200)]
move virtrng_remove to .devexit.text

The function virtrng_remove is used only wrapped by __devexit_p so define
it using __devexit.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agomove virtballoon_remove to .devexit.text
Uwe Kleine-König [Thu, 1 Oct 2009 08:28:33 +0000 (10:28 +0200)]
move virtballoon_remove to .devexit.text

The function virtballoon_remove is used only wrapped by __devexit_p so
define it using __devexit.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio_blk: Revert serial number support
Rusty Russell [Thu, 22 Oct 2009 22:39:28 +0000 (16:39 -0600)]
virtio_blk: Revert serial number support

This reverts "Add serial number support for virtio_blk, V4a".

Turns out that virtio_pci, lguest and s/390 all have an 8 bit limit
on virtio config space, so noone could ever use this.

This is coming back later in a cleaner form.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: john cooper <john.cooper@redhat.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
14 years agovirtio: let header files include virtio_ids.h
Christian Borntraeger [Wed, 30 Sep 2009 09:17:21 +0000 (11:17 +0200)]
virtio: let header files include virtio_ids.h

Rusty,

commit 3ca4f5ca73057a617f9444a91022d7127041970a
    virtio: add virtio IDs file
moved all device IDs into a single file. While the change itself is
a very good one, it can break userspace applications. For example
if a userspace tool wanted to get the ID of virtio_net it used to
include virtio_net.h. This does no longer work, since virtio_net.h
does not include virtio_ids.h.
This patch moves all "#include <linux/virtio_ids.h>" from the C
files into the header files, making the header files compatible with
the old ones.

In addition, this patch exports virtio_ids.h to userspace.

CC: Fernando Luis Vazquez Cao <fernando@oss.ntt.co.jp>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio_blk: revert QUEUE_FLAG_VIRT addition
Christoph Hellwig [Fri, 4 Sep 2009 20:44:42 +0000 (22:44 +0200)]
virtio_blk: revert QUEUE_FLAG_VIRT addition

It seems like the addition of QUEUE_FLAG_VIRT caueses major performance
regressions for Fedora users:

https://bugzilla.redhat.com/show_bug.cgi?id=509383
https://bugzilla.redhat.com/show_bug.cgi?id=505695

while I can't reproduce those extreme regressions myself I think the flag
is wrong.

Rationale:

  QUEUE_FLAG_VIRT expands to QUEUE_FLAG_NONROT which casus the queue
  unplugged immediately.  This is not a good behaviour for at least
  qemu and kvm where we do have significant overhead for every
  I/O operations.  Even with all the latested speeups (native AIO,
  MSI support, zero copy) we can only get native speed for up to 128kb
  I/O requests we already are down to 66% of native performance for 4kb
  requests even on my laptop running the Intel X25-M SSD for which the
  QUEUE_FLAG_NONROT was designed.
  If we ever get virtio-blk overhead low enough that this flag makes
  sense it should only be set based on a feature flag set by the host.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years ago[SCSI] zfcp: Flush SCSI registration work when adding unit
Christof Schmitt [Tue, 13 Oct 2009 08:44:11 +0000 (10:44 +0200)]
[SCSI] zfcp: Flush SCSI registration work when adding unit

When configuring a LUN for use in zfcp, flush the SCSI work to ensure
the SCSI device has been created before returning. This means that a
configuration procedure can run these commands in a script and the
SCSI device is available immediately after the unit_add:

echo 1 > /sys/bus/ccw/drivers/zfcp/0.0.181d/online
echo 0x401040C300000000 > \
        /sys/bus/ccw/drivers/zfcp/0.0.181d/0x500507630313c562/unit_add
lsscsi

Reviewed-by: Swen Schillig <swen@vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] zfcp: Fix timer initialization for ct and els requests
Christof Schmitt [Tue, 13 Oct 2009 08:44:10 +0000 (10:44 +0200)]
[SCSI] zfcp: Fix timer initialization for ct and els requests

Add HZ since the start_timer function expects jiffies, not seconds.

Reviewed-by: Swen Schillig <swen@vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] zfcp: Warn about storage devices with broken PLOGI data
Christof Schmitt [Tue, 13 Oct 2009 08:44:09 +0000 (10:44 +0200)]
[SCSI] zfcp: Warn about storage devices with broken PLOGI data

After opening a remote port zfcp checks if the WWPN returned in the
PLOGI maches the WWPN of the port that should have been opened. On a
mismatch zfcp assumes that the DID just changed, queries the FC
nameserver and tries again. If the situation persists the erp will
give up.

With this strategy, if the remote port always returns the wrong PLOGI
data, the remote port will not be opened. Introduce a warning, so that
the system administrator knows why the remote port is not being opened
and to have a pointer to investigate the problem on the storage
system.

Reviewed-by: Swen Schillig <swen@vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] zfcp: Handle WWPN mismatch in PLOGI payload
Christof Schmitt [Wed, 14 Oct 2009 09:00:43 +0000 (11:00 +0200)]
[SCSI] zfcp: Handle WWPN mismatch in PLOGI payload

For ports, zfcp gets the DID from the FC nameserver and tries to open
the port. If the open succeeds, zfcp compares the WWPN from the
nameserver with the WWPN in the PLOGI payload. In case of a mismatch,
zfcp assumes that the DID of the port just changed and we opened the
wrong port. This means that zfcp has to forget the DID, lookup the DID
again and retry.

This error case had a problem that zfcp forgets the DID, but never
looks up a new one, stalling the ERP in this case. Fix this by
triggering the DID lookup and properly exit from the ERP. The DID
lookup will trigger a new ERP action.

Also ensure when trying to open the port again with the new DID, first
close the open port, even in the NOESC case.

Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years ago[SCSI] zfcp: fix kfree handling in zfcp_init_device_setup
Heiko Carstens [Tue, 13 Oct 2009 08:44:07 +0000 (10:44 +0200)]
[SCSI] zfcp: fix kfree handling in zfcp_init_device_setup

The pointer that is allocated with kmalloc() is passed to strsep()
which modifies it. Later on the modified pointer value will be passed
to kfree. Save the original pointer and pass that one to kfree
instead.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>