During my recent work on the rtnl lock deadlock in the IPoIB driver, I
saw that even once I fixed the apparent races for a single device, as
soon as that device had any children, new races popped up. It turns
out that this is because no matter how well we protect against races
on a single device, the fact that all devices use the same workqueue,
and flush_workqueue() flushes *everything* from that workqueue, we can
have one device in the middle of a down and holding the rtnl lock and
another totally unrelated device needing to run mcast_restart_task,
which wants the rtnl lock and will loop trying to take it unless is
sees its own FLAG_ADMIN_UP flag go away. Because the unrelated
interface will never see its own ADMIN_UP flag drop, the interface
going down will deadlock trying to flush the queue. There are several
possible solutions to this problem:
Make carrier_on_task and mcast_restart_task try to take the rtnl for
some set period of time and if they fail, then bail. This runs the
real risk of dropping work on the floor, which can end up being its
own separate kind of deadlock.
Set some global flag in the driver that says some device is in the
middle of going down, letting all tasks know to bail. Again, this can
drop work on the floor. I suppose if our own ADMIN_UP flag doesn't go
away, then maybe after a few tries on the rtnl lock we can queue our
own task back up as a delayed work and return and avoid dropping work
on the floor that way. But I'm not 100% convinced that we won't cause
other problems.
Or the method this patch attempts to use, which is when we bring an
interface up, create a workqueue specifically for that interface, so
that when we take it back down, we are flushing only those tasks
associated with our interface. In addition, keep the global
workqueue, but now limit it to only flush tasks. In this way, the
flush tasks can always flush the device specific work queues without
having deadlock issues.
Signed-off-by: Doug Ledford <dledford@redhat.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
In preparation for using per device work queues, we need to move the
start of the neighbor thread task to after ipoib_ib_dev_init and move
the destruction of the neighbor task to before ipoib_ib_dev_cleanup.
Otherwise we will end up freeing our workqueue with work possibly
still on it.
Signed-off-by: Doug Ledford <dledford@redhat.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Our mcast_dev_flush routine and our mcast_restart_task can race
against each other. In particular, they both hold the priv->lock
while manipulating the rbtree and while removing mcast entries from
the multicast_list and while adding entries to the remove_list, but
they also both drop their locks prior to doing the actual removes.
The mcast_dev_flush routine is run entirely under the rtnl lock and so
has at least some locking. The actual race condition is like this:
Thread 1 Thread 2
ifconfig ib0 up
start multicast join for broadcast
multicast join completes for broadcast
start to add more multicast joins
call mcast_restart_task to add new entries
ifconfig ib0 down
mcast_dev_flush
mcast_leave(mcast A)
mcast_leave(mcast A)
As mcast_leave calls ib_sa_multicast_leave, and as member in
core/multicast.c is ref counted, we run into an unbalanced refcount
issue. To avoid stomping on each others removes, take the rtnl lock
specifically when we are deleting the entries from the remove list.
Signed-off-by: Doug Ledford <dledford@redhat.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Commit a9c8ba5884 ("IPoIB: Fix usage of uninitialized multicast
objects") added a new flag MCAST_JOIN_STARTED, but was not very strict
in how it was used. We didn't always initialize the completion struct
before we set the flag, and we didn't always call complete on the
completion struct from all paths that complete it. This made it less
than totally effective, and certainly made its use confusing. And in
the flush function we would use the presence of this flag to signal
that we should wait on the completion struct, but we never cleared
this flag, ever. This is further muddied by the fact that we overload
the MCAST_FLAG_BUSY flag to mean two different things: we have a join
in flight, and we have succeeded in getting an ib_sa_join_multicast.
In order to make things clearer and aid in resolving the rtnl deadlock
bug I've been chasing, I cleaned this up a bit.
1) Remove the MCAST_JOIN_STARTED flag entirely
2) Un-overload MCAST_FLAG_BUSY so it now only means a join is in-flight
3) Test on mcast->mc directly to see if we have completed
ib_sa_join_multicast (using IS_ERR_OR_NULL)
4) Make sure that before setting MCAST_FLAG_BUSY we always initialize
the mcast->done completion struct
5) Make sure that before calling complete(&mcast->done), we always clear
the MCAST_FLAG_BUSY bit
6) Take the mcast_mutex before we call ib_sa_multicast_join and also
take the mutex in our join callback. This forces
ib_sa_multicast_join to return and set mcast->mc before we process
the callback. This way, our callback can safely clear mcast->mc
if there is an error on the join and we will do the right thing as
a result in mcast_dev_flush.
7) Because we need the mutex to synchronize mcast->mc, we can no
longer call mcast_sendonly_join directly from mcast_send and
instead must add sendonly join processing to the mcast_join_task
A number of different races are resolved with these changes. These
races existed with the old MCAST_FLAG_BUSY usage, the
MCAST_JOIN_STARTED flag was an attempt to address them, and while it
helped, a determined effort could still trip things up.
One race looks something like this:
Thread 1 Thread 2
ib_sa_join_multicast (as part of running restart mcast task)
alloc member
call callback
ifconfig ib0 down
wait_for_completion
callback call completes
wait_for_completion in
mcast_dev_flush completes
mcast->mc is PTR_ERR_OR_NULL
so we skip ib_sa_leave_multicast
return from callback
return from ib_sa_join_multicast
set mcast->mc = return from ib_sa_multicast
We now have a permanently unbalanced join/leave issue that trips up the
refcounting in core/multicast.c
Another like this:
Thread 1 Thread 2 Thread 3
ib_sa_multicast_join
ifconfig ib0 down
priv->broadcast = NULL
join_complete
wait_for_completion
mcast->mc is not yet set, so don't clear
return from ib_sa_join_multicast and set mcast->mc
complete
return -EAGAIN (making mcast->mc invalid)
call ib_sa_multicast_leave
on invalid mcast->mc, hang
forever
By holding the mutex around ib_sa_multicast_join and taking the mutex
early in the callback, we force mcast->mc to be valid at the time we
run the callback. This allows us to clear mcast->mc if there is an
error and the join is going to fail. We do this before we complete
the mcast. In this way, mcast_dev_flush always sees consistent state
in regards to mcast->mc membership at the time that the
wait_for_completion() returns.
Signed-off-by: Doug Ledford <dledford@redhat.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
We blindly assume that we can just take the rtnl lock and that will
prevent races with downing this interface. Unfortunately, that's not
the case. In ipoib_mcast_stop_thread() we will call flush_workqueue()
in an attempt to clear out all remaining instances of ipoib_join_task.
But, since this task is put on the same workqueue as the join task,
the flush_workqueue waits on this thread too. But this thread is
deadlocked on the rtnl lock. The better thing here is to use trylock
and loop on that until we either get the lock or we see that
FLAG_ADMIN_UP has been cleared, in which case we don't need to do
anything anyway and we just return.
Signed-off-by: Doug Ledford <dledford@redhat.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Setting the MTU can safely be moved to the carrier_on_task, which keeps
us from needing to take the rtnl lock in the join_finish section.
Signed-off-by: Doug Ledford <dledford@redhat.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
cxgb4_create_server() and cxgb4_create_server6() return NET_XMIT_*
values or a negative errno. iw_cxgb4 need to handle this correctly.
Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
When transitioning into ERROR state, the QP was getting flushed after
waking up any waiters. This can cause applications to miss flushed work
requests which can stall an NFS mount.
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
T4/T5 hardware can't handle MRs >= 8GB due to a hardware bug. So limit
registrations to < 8GB for thse devices.
Based on original work by Steve Wise <swise@opengridcomputing.com>.
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
0B MRs need some tweaks to work correctly with HW. When writing the
TPTE, if the MR length is zero we now:
1) turn off all permissions
2) set the length to -1
While functionality/capabilities of the MR are the same with these
changes, it resolves a dapltest 0B RDMA Read test failure. Based on
original work by Steve Wise <swise@opengridcomputing.com>.
Signed-off-by: Pramod Kumar <pramod@chelsio.com>
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Address resolution always does a context switch to a work-queue to
deliver the address resolution event. When the IP address is already
cached in the system ARP table, we're going through the following:
chain:
rdma_resolve_ip --> addr_resolve (cache hit) -->
which ends up with:
queue_req --> set_timeout (now) --> mod_delayed_work(,, delay=1)
We actually do realize that the timeout should be zero, but the code
forces it to a minimum of one jiffie.
Using one jiffie as the minimum delay value results in sub-optimal
scheduling of executing this work item by the workqueue, which on the
below testbed costs about 3-4ms out of 12ms total time.
To fix that, we let the minimum delay to be zero. Note that the
connect step times change too, as there are address resolution calls
from that flow.
The results were taken from running both client and server on the
same node, over mlx4 RoCE port.
before -->
step total ms max ms min us us / conn
create id : 0.01 0.01 6.00 6.00
resolve addr : 4.02 4.01 4013.00 4016.00
resolve route: 0.18 0.18 182.00 183.00
create qp : 1.15 1.15 1150.00 1150.00
connect : 6.73 6.73 6730.00 6731.00
disconnect : 0.55 0.55 549.00 550.00
destroy : 0.01 0.01 9.00 9.00
after -->
step total ms max ms min us us / conn
create id : 0.01 0.01 6.00 6.00
resolve addr : 0.05 0.05 49.00 52.00
resolve route: 0.21 0.21 207.00 208.00
create qp : 1.10 1.10 1104.00 1104.00
connect : 1.22 1.22 1220.00 1221.00
disconnect : 0.71 0.71 713.00 713.00
destroy : 0.01 0.01 9.00 9.00
Signed-off-by: Or Kehati <ork@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Acked-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Applications can request that the SM assign an MGID by passing a mcast
member request containing MGID = 0. When the SM responds by sending
the allocated MGID, this MGID replaces the 0-MGID in the multicast group.
However, the MGID field in the group is also the key field in the IB
core multicast code rbtree containing the multicast groups for the
port.
Since this is a key field, correct handling requires that the group
entry be deleted from the rbtree and then re-inserted with the new
key, so that the table structure is properly maintained.
The current code does not do this correctly. Correct operation
requires that if the key-field gid has changed at all, it should be
deleted and re-inserted.
Note that when inserting, if the new MGID is zero (not the case here
but the code should handle this correctly), we allow duplicate entries
for 0-MGIDs.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Acked-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
For RoCE, resolution of layer 2 address attributes forces no VLAN if
link-local GIDs are used. This patch allows applications to choose
the VLAN ID for link-local based RoCE GIDs by setting IB_QP_VID in
their QP attribute mask, and prevents the core from overriding this
choice.
Cc: Ursula Braun <ursula.braun@de.ibm.com>
Signed-off-by: Moni Shoua <monis@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Here's the big staging tree pull request for 3.19-rc1.
We continued to delete more lines than were added, always a good thing,
but not at a huge rate this release, only about 70k lines removed
overall mostly from removing the horrid bcm driver.
Lots of normal staging driver cleanups and fixes all over the place,
well over a thousand of them, the shortlog shows all the horrid details.
The "contentious" thing here is the movement of the Android binder code
out of staging into the "real" part of the kernel. This is code that
has been stable for a few years now and is working as-is in the tens of
millions of devices with no issues. Yes, the code is horrid, and the
userspace api leaves a lot to be desired, but it's not going to change
due to legacy issues that we have no control over. Because so many
devices and companies rely on this, and the code is stable, might as
well promote it out of staging.
This was all discussed at the Linux Plumbers conference, and everyone
participating agreed that this was the best way forward.
There is work happening to replace the binder code with something new
that is happening right now, but I don't expect to see the results of
that work for another year at the earliest. If that ever happens, and
Android switches over to it, I'll gladly remove this version.
As for maintainers, I'll be glad to maintain this code, I've been doing
it for the past few years with no problems. I'll send a MAINTAINERS
entry for it before 3.19-final is out, still need to talk to the Google
developers about if they are willing to help with it or not, last I
checked they were, which was good.
All of these patches have been in linux-next for a while with no
reported issues.
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2
iEYEABECAAYFAlSPICkACgkQMUfUDdst+yksdwCfSLE9VUy1o2sAPDRe+J3bQced
EWEAoL3RtnejKbo5tHS2IT69pLrwiIDS
=YXyM
-----END PGP SIGNATURE-----
Merge tag 'staging-3.19-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging
Pull staging driver updates from Greg KH:
"Here's the big staging tree pull request for 3.19-rc1.
We continued to delete more lines than were added, always a good
thing, but not at a huge rate this release, only about 70k lines
removed overall mostly from removing the horrid bcm driver.
Lots of normal staging driver cleanups and fixes all over the place,
well over a thousand of them, the shortlog shows all the horrid
details.
The "contentious" thing here is the movement of the Android binder
code out of staging into the "real" part of the kernel. This is code
that has been stable for a few years now and is working as-is in the
tens of millions of devices with no issues. Yes, the code is horrid,
and the userspace api leaves a lot to be desired, but it's not going
to change due to legacy issues that we have no control over. Because
so many devices and companies rely on this, and the code is stable,
might as well promote it out of staging.
This was all discussed at the Linux Plumbers conference, and everyone
participating agreed that this was the best way forward.
There is work happening to replace the binder code with something new
that is happening right now, but I don't expect to see the results of
that work for another year at the earliest. If that ever happens, and
Android switches over to it, I'll gladly remove this version.
As for maintainers, I'll be glad to maintain this code, I've been
doing it for the past few years with no problems. I'll send a
MAINTAINERS entry for it before 3.19-final is out, still need to talk
to the Google developers about if they are willing to help with it or
not, last I checked they were, which was good.
All of these patches have been in linux-next for a while with no
reported issues"
* tag 'staging-3.19-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging: (1382 commits)
Staging: slicoss: Fix long line issues in slicoss.c
staging: rtl8712: remove unnecessary else after return
staging: comedi: change some printk calls to pr_err
staging: rtl8723au: hal: Removed the extra semicolon
lustre: Deletion of unnecessary checks before three function calls
staging: lustre: fix sparse warnings: static function declaration
staging: lustre: fixed sparse warnings related to static declarations
staging: unisys: remove duplicate header
staging: unisys: remove unneeded structure
staging: ft1000 : replace __attribute ((__packed__) with __packed
drivers: staging: rtl8192e: Include "asm/unaligned.h" instead of "access_ok.h" in "rtl819x_BAProc.c"
Drivers:staging:rtl8192e: Fixed checkpatch warning
Drivers:staging:clocking-wizard: Added a newline
staging: clocking-wizard: check for a valid clk_name pointer
staging: rtl8723au: Hal_InitPGData() avoid unnecessary typecasts
staging: rtl8723au: _DisableAnalog(): Avoid zero-init variables unnecessarily
staging: rtl8723au: Remove unnecessary wrapper _ResetDigitalProcedure1()
staging: rtl8723au: _ResetDigitalProcedure1_92C() reduce code obfuscation
staging: rtl8723au: Remove unnecessary wrapper _DisableRFAFEAndResetBB()
staging: rtl8723au: _DisableRFAFEAndResetBB8192C(): Reduce code obfuscation
...
* update power/reset drivers to use kernel restart handler
* add power off driver for i.mx6
* add DT support for gpio-charger
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQIcBAABCgAGBQJUjv52AAoJENju1/PIO/qaESIP/3rRE/Gw1fmfX+F1wN+jXQao
WXw2JW/IUQG0TQlpTnEqPQQu+H48dX58MbpNVSLy2O7Z4XiMRDG2HkEduyJYGm+P
37pfgiTH9Vj7MJ0EnNK3lLrQmF1l+yNHIjZUGLPEjbKmXSzJYPHwNhtC5U9/QooY
igRwKC51bBw6of6pPaIYP6nxQeaNVgI3MaOnJ1+glLc0uQuy5mc4gODL6yCEkvuR
WCsyJi226m/dBD+ZgejSet7kJS3JG4dR0GuqZkJ22DS5wKbwKrgXfOiVl+xrgCwR
FqbvnZbscr+jt/xFJsAT5zqtvHNgxx37XVwkV3nHsKX8UugYg3rCB5T/mlgNj+m/
Cg5BePP84O1NMJHYXSj7Wo6BYYoFJO8ucT0uIKpcci3UyvtqWcMXzptNjkAtylw1
P+0EtP7HL4jBLJ5Snd0Je47w2mztmaJmqCc9LU8odznlFbaJ25X3chf3zqAoKdpY
Lqe1XWxi5HtLNmuX/NsNhw0tdxuSiH5e2r7yM5Bwtm4uh4Yx9Ab9TEtloF3u+en5
ALR88cryxbP0zu0K/HFUeoM3v7mqH3x29NqAsSpY1D6p87308FUa1rkvYMlHwcFD
Y2W2iZhmxnE5Bi9OTpfDgGQYNEUiELWmMDi3S3aSwb0QM9UWsi043LGMQfpG5QCn
+xKTKJDNJsHAX5jYDuDY
=v7l/
-----END PGP SIGNATURE-----
Merge tag 'for-v3.19' of git://git.infradead.org/battery-2.6
Pull power supply updates from Sebastian Reichel::
"Power supply and reset changes for the v3.19 series
- update power/reset drivers to use kernel restart handler
- add power off driver for i.mx6
- add DT support for gpio-charger"
* tag 'for-v3.19' of git://git.infradead.org/battery-2.6:
power: reset: adjust priority of simple syscon reboot driver
power: ds2782_battery: Simplify the PM hooks
power/reset: brcmstb: Register with kernel restart handler
power/reset: hisi: Register with kernel restart handler
power/reset: keystone: Register with kernel restart handler
power/reset: axxia: Register with kernel restart handler
power/reset: xgene: Register with kernel restart handler
power/reset: xgene: Use mdelay instead of jiffies based timeout
power/reset: xgene: Use local variable dev instead of pdev->dev
power/reset: xgene: Drop devm_kfree
power/reset: xgene: Return -ENOMEM if out of memory
power/reset: vexpress: Register with kernel restart handler
power: reset: imx-snvs-poweroff: add power off driver for i.mx6
power: gpio-charger: add device tree support
dt-bindings: document gpio-charger bindings
Pull irq domain ARM updates from Thomas Gleixner:
"This set of changes make use of hierarchical irqdomains to provide:
- MSI/ITS support for GICv3
- MSI support for GICv2m
- Interrupt polarity extender for GICv1
Marc has come more cleanups for the existing extension hooks of GIC in
the pipeline, but they are going to be 3.20 material"
* 'irq-irqdomain-arm-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (22 commits)
irqchip: gicv3-its: Fix ITT allocation
irqchip: gicv3-its: Move some alloc/free code to activate/deactivate
irqchip: gicv3-its: Fix domain free in multi-MSI case
irqchip: gic: Remove warning by including linux/irqdomain.h
irqchip: gic-v2m: Add DT bindings for GICv2m
irqchip: gic-v2m: Add support for ARM GICv2m MSI(-X) doorbell
irqchip: mtk-sysirq: dt-bindings: Add bindings for mediatek sysirq
irqchip: mtk-sysirq: Add sysirq interrupt polarity support
irqchip: gic: Support hierarchy irq domain.
irqchip: GICv3: Binding updates for ITS
irqchip: GICv3: ITS: enable compilation of the ITS driver
irqchip: GICv3: ITS: plug ITS init into main GICv3 code
irqchip: GICv3: ITS: DT probing and initialization
irqchip: GICv3: ITS: MSI support
irqchip: GICv3: ITS: device allocation and configuration
irqchip: GICv3: ITS: tables allocators
irqchip: GICv3: ITS: LPI allocator
irqchip: GICv3: ITS: irqchip implementation
irqchip: GICv3: ITS command queue
irqchip: GICv3: rework redistributor structure
...
Pull core block fix from Jens Axboe:
"Jan reported a problem this morning with a crash in blk-mq, and after
looking over the recent changes, it's obvious that the blk-mq-tag
waitqueue handling change is buggy. We could end up _not_ doing
finish_wait() before switching to a new waitqueue, thus corrupting the
wait task list"
* 'for-linus' of git://git.kernel.dk/linux-block:
Revert "blk-mq: Micro-optimize bt_get()"
use less buffers when small vrings are being used.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=Qf82
-----END PGP SIGNATURE-----
Merge tag 'rpmsg-3.19-next' of git://git.kernel.org/pub/scm/linux/kernel/git/ohad/rpmsg
Pull rpmsg update from Ohad Ben-Cohen:
"A single patch from Suman Anna which makes rpmsg use less buffers when
small vrings are being used"
* tag 'rpmsg-3.19-next' of git://git.kernel.org/pub/scm/linux/kernel/git/ohad/rpmsg:
rpmsg: use less buffers when vrings are small
Pull drm updates from Dave Airlie:
"Highlights:
- AMD KFD driver merge
This is the AMD HSA interface for exposing a lowlevel interface for
GPGPU use. They have an open source userspace built on top of this
interface, and the code looks as good as it was going to get out of
tree.
- Initial atomic modesetting work
The need for an atomic modesetting interface to allow userspace to
try and send a complete set of modesetting state to the driver has
arisen, and been suffering from neglect this past year. No more,
the start of the common code and changes for msm driver to use it
are in this tree. Ongoing work to get the userspace ioctl finished
and the code clean will probably wait until next kernel.
- DisplayID 1.3 and tiled monitor exposed to userspace.
Tiled monitor property is now exposed for userspace to make use of.
- Rockchip drm driver merged.
- imx gpu driver moved out of staging
Other stuff:
- core:
panel - MIPI DSI + new panels.
expose suggested x/y properties for virtual GPUs
- i915:
Initial Skylake (SKL) support
gen3/4 reset work
start of dri1/ums removal
infoframe tracking
fixes for lots of things.
- nouveau:
tegra k1 voltage support
GM204 modesetting support
GT21x memory reclocking work
- radeon:
CI dpm fixes
GPUVM improvements
Initial DPM fan control
- rcar-du:
HDMI support added
removed some support for old boards
slave encoder driver for Analog Devices adv7511
- exynos:
Exynos4415 SoC support
- msm:
a4xx gpu support
atomic helper conversion
- tegra:
iommu support
universal plane support
ganged-mode DSI support
- sti:
HDMI i2c improvements
- vmwgfx:
some late fixes.
- qxl:
use suggested x/y properties"
* 'drm-next' of git://people.freedesktop.org/~airlied/linux: (969 commits)
drm: sti: fix module compilation issue
drm/i915: save/restore GMBUS freq across suspend/resume on gen4
drm: sti: correctly cleanup CRTC and planes
drm: sti: add HQVDP plane
drm: sti: add cursor plane
drm: sti: enable auxiliary CRTC
drm: sti: fix delay in VTG programming
drm: sti: prepare sti_tvout to support auxiliary crtc
drm: sti: use drm_crtc_vblank_{on/off} instead of drm_vblank_{on/off}
drm: sti: fix hdmi avi infoframe
drm: sti: remove event lock while disabling vblank
drm: sti: simplify gdp code
drm: sti: clear all mixer control
drm: sti: remove gpio for HDMI hot plug detection
drm: sti: allow to change hdmi ddc i2c adapter
drm/doc: Document drm_add_modes_noedid() usage
drm/i915: Remove '& 0xffff' from the mask given to WA_REG()
drm/i915: Invert the mask and val arguments in wa_add() and WA_REG()
drm: Zero out DRM object memory upon cleanup
drm/i915/bdw: Fix the write setting up the WIZ hashing mode
...
The VM_FAULT_RETRY handling was confusing and incorrect for the case of
returning to kernel mode. We need to handle the exception table fixup
if we return to kernel mode due to a fatal signal - it will basically
look to the kernel user mode access like the access failed due to the VM
going away from udner it. Which is correct - the process is dying - and
avoids the whole "repeat endless kernel page faults" case.
Handling the VM_FAULT_RETRY early and in just one place also simplifies
the mmap_sem handling, since once we've taken care of VM_FAULT_RETRY we
know that we can just drop the lock. The remaining accounting and
possible error handling is thread-local and does not need the mmap_sem.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This replaces four copies in various stages of mm_fault_error() handling
with just a single one. It will also allow for more natural placement
of the unlocking after some further cleanup.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
into -rc1:
- Update SoC revision detection for am43x es1.2
- Fix regression with GPMC timings on 2430sdp for some versions
of u-boot
- Fix dra7 watchdog compatible property
- Fix am437x-sk-evm LCD timings
- Fix dra7 DSS clock muxing
- Fix dra7-evm voltages
- Remove a unused function prototype for am33xx_clk_init
- Enable AHCI in the omap2plus_defconfig
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQIcBAABAgAGBQJUiJUMAAoJEBvUPslcq6VzGzEP/Ayn/OoI3NXw7MZhru2Yp2iA
3oBrjGK4Sy0ak/osf+UB/KMsmxGiQnStRdW0s+QrqXXpBYvDZKy8uZUJ+9ftG0t5
cdpvn/oAI9wXy0WbS5cEkfSXrT8ggoCu7h13NZMSHbD4d6/L9tqJes66/7njwt9d
BdQlXPtG++A7XlV0v1KN1lSmkNmz3VbZfxL4j0ihKmiYeqI+TQI5PIBMFEB8AnFW
FTaUMSrZD2yrLzfh2BxVTeoEmycTdmPC7lgz3QiJLMijleDRpZRIl7oKK2zF3QxC
8fldyEJCDW6hiMP6y8bF2nE8jkuto0x3hKb0Di34ccgZeR2Ez3pPU77ecAIEV6eG
8cD6aeV+5XawFCSwMtovV3tLPUZtH93swrDroSY0FIfj0X66z/Wo4erAw7ncppVl
mAG/tyIdEA9c+BY59ADBD45+NwvpoBcecWHZYmtsxaKxTo9yzo3P4qL9eUYG6aNH
ugdcVFvy8P3zcZGSgRD1qsuwrjv6+RPtfhNz+a299IYw35glJI+qMWSlc5YqpdFq
Hb7ZGDKtGCndy6BF1NrEw8xyBaQJX5V6x3QAA5x8JTAe7aEHOAQAzk/9Xvjo1NoD
IB8c2uCzpB/aIm75f9qNmEMO92jHbd6zizF8ejTiTU99h5ONXTOYz1k6c4w/QeiS
fzA1a0rNHQgVn+VcC5qN
=rzAf
-----END PGP SIGNATURE-----
Merge tag 'omap-for-v3.19/fixes-for-merge-window' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap into fixes
From: Tony Lindgren <tony@atomide.com>
Subject: [GIT PULL] few fixes for the v3.19 merge window
Fixes for a few issues found that would be good to get
into -rc1:
- Update SoC revision detection for am43x es1.2
- Fix regression with GPMC timings on 2430sdp for some versions
of u-boot
- Fix dra7 watchdog compatible property
- Fix am437x-sk-evm LCD timings
- Fix dra7 DSS clock muxing
- Fix dra7-evm voltages
- Remove a unused function prototype for am33xx_clk_init
- Enable AHCI in the omap2plus_defconfig
* tag 'omap-for-v3.19/fixes-for-merge-window' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap: (1601 commits)
ARM: omap2plus_defconfig: Enable AHCI_PLATFORM driver
ARM: dts: am437x-sk-evm.dts: fix LCD timings
ARM: dts: dra7-evm: Update SMPS7 (VDD_CORE) max voltage to match DM
ARM: dts: dra7-evm: Fix typo in SMPS6 (VDD_GPU) max voltage
ARM: OMAP2+: AM43x: Add ID for ES1.2
ARM: dts: am437x-sk: fix lcd enable pin mux data
ARM: dts: Fix gpmc regression for omap 2430sdp smc91x
hwmon: (tmp401) Detect TMP435 on all addresses it supports
mfd: rtsx: Add func to split u32 into register
mmc: sdhci-msm: Convert to mmc_send_tuning()
mmc: sdhci-esdhc-imx: Convert to mmc_send_tuning()
mmc: core: Let mmc_send_tuning() to take struct mmc_host* as parameter
nios2: Make NIOS2_CMDLINE_IGNORE_DTB depend on CMDLINE_BOOL
nios2: Add missing NR_CPUS to Kconfig
nios2: asm-offsets: Remove unused definition TI_TASK
nios2: Remove write-only struct member from nios2_timer
nios2: Remove unused extern declaration of shm_align_mask
nios2: include linux/type.h in io.h
nios2: move include asm-generic/io.h to end of file
nios2: remove include asm-generic/iomap.h from io.h
...
This just makes code sparse-clean by using
new memory access APIs in one file I missed.
The new feature bit is not yet negotiated.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
When switching everything over to virtio 1.0 memory access APIs,
I missed converting vringh.
Fortunately, it's straight-forward.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
CONFIG_GENERIC_CPUFREQ_CPU0 disappeared with commit bbcf071969
("cpufreq: cpu0: rename driver and internals to 'cpufreq_dt'") and some
defconfigs are still using it instead of the new one.
Use the renamed CONFIG_CPUFREQ_DT generic driver.
Cc: <stable@vger.kernel.org> # 3.18
Reported-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Kevin Hilman <khilman@linaro.org>
The setting of this flag was missed in previous modifications.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Cc: Stable <stable@vger.kernel.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Debug code prints the fifo name via custom dev_warn() wrappers. The
fifo_names array is only non-zero when debugging is manually enabled,
which is all well and good. However, it's *not* good that this array
uses zero-length arrays in the non-debug case, and so it doesn't
actually have any memory allocated to it. This means that as far as we
know, fifo_names[i] actually points to garbage memory.
I've seen this in my log:
[ 4601.205511] brcmsmac bcma0:1: wl0: brcms_c_d11hdrs_mac80211: �GeL txop exceeded phylen 137/256 dur 1602/1504
So let's give this array space enough to fill it with a NULL byte.
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Cc: Brett Rudley <brudley@broadcom.com>
Cc: Arend van Spriel <arend@broadcom.com>
Cc: "Franky (Zhenhui) Lin" <frankyl@broadcom.com>
Cc: Hante Meuleman <meuleman@broadcom.com>
Cc: "John W. Linville" <linville@tuxdriver.com>
Cc: linux-wireless@vger.kernel.org
Cc: brcm80211-dev-list@broadcom.com
Cc: netdev@vger.kernel.org
Acked-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Fixes the following sparse warning:
drivers/net/wireless/rtlwifi/rtl8192cu/hw.c:1595:6: warning:
symbol 'usb_cmd_send_packet' was not declared. Should it be static?
Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Acked-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Replace a misspelled function name by %s and then __func__.
8821 was written as 8812.
This was done using Coccinelle, including the use of Levenshtein distance,
as proposed by Rasmus Villemoes.
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Replace a misspelled function name by %s and then __func__.
This was done using Coccinelle, including the use of Levenshtein distance,
as proposed by Rasmus Villemoes.
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Replace a misspelled function name by %s and then __func__.
This was done using Coccinelle, including the use of Levenshtein distance,
as proposed by Rasmus Villemoes.
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
acerhdf uses thermal interfaces so it should depend on THERMAL.
It also should not select a thermal driver without checking that
THERMAL is enabled.
This fixes the following build errors when THERMAL=m and
ACERHDF=y.
drivers/built-in.o: In function `acerhdf_set_mode':
acerhdf.c:(.text+0x3e02e1): undefined reference to `thermal_zone_device_update'
drivers/built-in.o: In function `acerhdf_unbind':
acerhdf.c:(.text+0x3e052d): undefined reference to `thermal_zone_unbind_cooling_device'
drivers/built-in.o: In function `acerhdf_bind':
acerhdf.c:(.text+0x3e0593): undefined reference to `thermal_zone_bind_cooling_device'
drivers/built-in.o: In function `acerhdf_init':
acerhdf.c:(.init.text+0x1c2f5): undefined reference to `thermal_cooling_device_register'
acerhdf.c:(.init.text+0x1c360): undefined reference to `thermal_zone_device_register'
drivers/built-in.o: In function `acerhdf_unregister_thermal':
acerhdf.c:(.text.unlikely+0x3c67): undefined reference to `thermal_cooling_device_unregister'
acerhdf.c:(.text.unlikely+0x3c91): undefined reference to `thermal_zone_device_unregister'
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Acked-by: Peter Feuerer <peter@piie.net>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Cyrille Pitchen says:
====================
net/macb: fix multiqueue support patch up
This series of patches is a fixup for the multiqueue support patch.
The first patch fixes a bug introduced by the multiqueue support patch.
The second one doesn't fix a bug but simplify the source code by removing
useless calls to devm_free_irq() since we use managed device resources for
IRQs.
They were applied on the net-next tree and tested with a sama5d36ek board.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Inside macb_probe(), when devm_request_irq() fails on queue q, there is no need
to call devm_free_irq() on queues 0..q-1 because the managed device resources
are released later when calling free_netdev().
Also removing devm_free_irq() call from macb_remove() for the same reason.
Signed-off-by: Cyrille Pitchen <cyrille.pitchen@atmel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
fix a bug introduced by the multiqueue support patch:
"net/macb: add TX multiqueue support for gem"
the "bp" pointer to the netdev private data was dereferenced and used after the
associated memory had been freed by calling free_netdev().
Signed-off-by: Cyrille Pitchen <cyrille.pitchen@atmel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/rds/message.c: In function ‘rds_message_inc_copy_to_user’:
net/rds/message.c:328: warning: comparison of distinct pointer types lacks a cast
Use min_t(unsigned long, ...) like is done in
rds_message_copy_from_user().
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
If CONFIG_OF is not set:
drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c: In function ‘sti_dwmac_parse_data’:
drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c:318: warning: ‘rs’ is used uninitialized in this function
of_property_read_string() will return -ENOSYS in this case, and rs will
be an uninitialized pointer.
While the fallback clock selection is already selected correctly in this
case, the string comparisons should be skipped too, else the system will
crash while dereferencing the uninitialized pointer.
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
