Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux
Pull drm fixes from Dave Airlie:
"Final drm fixes: one core locking imbalance regression, and a bunch of
i915 baytrail s/r fixes"
* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
drm: fix drm_mode_getconnector() locking imbalance regression
drm/i915/vlv: remove wait for previous GFX clk disable request
drm/i915/chv: Remove Wait for a previous gfx force-off
drm/i915/vlv: save/restore the power context base reg
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client
Pull ceph revert from Sage Weil:
"This corrects a recent misadventure with __GFP_MEMALLOC and
PF_MEMALLOC; it turns out it's not a good fit for RBD and we're better
off relying on dirty page throttling"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
Revert "libceph: use memalloc flags for net IO"
Merge misc fixes from Andrew Morton:
"Three fixes"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
mm: numa: disable change protection for vma(VM_HUGETLB)
include/linux/dmapool.h: declare struct device
mm: move zone lock to a different cache line than order-0 free page lists
Copy the kernel module data from user space in chunks
Unlike most (all?) other copies from user space, kernel module loading
is almost unlimited in size. So we do a potentially huge
"copy_from_user()" when we copy the module data from user space to the
kernel buffer, which can be a latency concern when preemption is
disabled (or voluntary).
Also, because 'copy_from_user()' clears the tail of the kernel buffer on
failures, even a *failed* copy can end up wasting a lot of time.
Normally neither of these are concerns in real life, but they do trigger
when doing stress-testing with trinity. Running in a VM seems to add
its own overheadm causing trinity module load testing to even trigger
the watchdog.
The simple fix is to just chunk up the module loading, so that it never
tries to copy insanely big areas in one go. That bounds the latency,
and also the amount of (unnecessarily, in this case) cleared memory for
the failure case.
The rule for 'copy_from_user()' is that it zeroes the remaining kernel
buffer even when the copy fails halfway, just to make sure that we don't
leave uninitialized kernel memory around. Because even if we check for
errors, some kernel buffers stay around after thge copy (think page
cache).
However, the x86-64 logic for user copies uses a copy_user_generic()
function for all the cases, that set the "zerorest" flag for any fault
on the source buffer. Which meant that it didn't just try to clear the
kernel buffer after a failure in copy_from_user(), it also tried to
clear the destination user buffer for the "copy_in_user()" case.
Not only is that pointless, it also means that the clearing code has to
worry about the tail clearing taking page faults for the user buffer
case. Which is just stupid, since that case shouldn't happen in the
first place.
Get rid of the whole "zerorest" thing entirely, and instead just check
if the destination is in kernel space or not. And then just use
memset() to clear the tail of the kernel buffer if necessary.
Dave Airlie [Wed, 8 Apr 2015 20:59:50 +0000 (06:59 +1000)]
Merge tag 'drm-intel-fixes-2015-04-08' of git://anongit.freedesktop.org/drm-intel into drm-fixes
three commits, all cc: stable, to address Baytrail
suspend/resume issues.
* tag 'drm-intel-fixes-2015-04-08' of git://anongit.freedesktop.org/drm-intel:
drm/i915/vlv: remove wait for previous GFX clk disable request
drm/i915/chv: Remove Wait for a previous gfx force-off
drm/i915/vlv: save/restore the power context base reg
const __read_mostly is a senseless combination. If something
is already const it cannot be __read_mostly. Remove the bogus
__read_mostly in the fou driver.
This fixes section conflicts with LTO.
Signed-off-by: Andi Kleen <ak@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Alessio Igor Bogani <alessio.bogani@elettra.eu> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Wed, 8 Apr 2015 19:17:38 +0000 (15:17 -0400)]
Merge branch 'rds'
Sowmini Varadhan says:
====================
RDS: RDS-core fixes
This patch-series updates the RDS core and rds-tcp modules with
some bug fixes that were originally authored by Andy Grover,
Zach Brown, and Chris Mason.
v2: Code review comment by Sergei Shtylov
V3: DaveM comments:
- dropped patches 3, 5 for "heuristic" changes in rds_send_xmit().
Investigation into the root-cause of these IB-triggered changes
produced the feedback: "I don't remember seeing "RDS: Stuck RM"
message in last 1-1.5 years and checking with other folks. It may very
well be some old workaround for stale connection for which long term
fix is already made and this part of code not exercised anymore."
Any such fixes, *if* they are needed, can/should be done in the
IB specific RDS transport modules.
- similarly dropped the LL_SEND_FULL patch (patch 6 in v2 set)
v4: Documentation/networking/rds.txt contains incorrect references
to "missing sysctl values for pf_rds and sol_rds in mainline".
The sysctl values were never needed in mainline, thus fix the
documentation.
v5: Clarify comment per http://www.spinics.net/lists/netdev/msg324220.html
v6: Re-added entire version history to cover letter.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
RDS: make sure not to loop forever inside rds_send_xmit
If a determined set of concurrent senders keep the send queue full,
we can loop forever inside rds_send_xmit. This fix has two parts.
First we are dropping out of the while(1) loop after we've processed a
large batch of messages.
Second we add a generation number that gets bumped each time the
xmit bit lock is acquired. If someone else has jumped in and
made progress in the queue, we skip our goto restart.
Original patch by Chris Mason.
Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
RDS: only use passive connections when addresses match
Passive connections were added for the case where one loopback IB
connection between identical addresses needs another connection to store
the second QP. Unfortunately, they were also created in the case where
the addesses differ and we already have both QPs.
This lead to a message reordering bug.
- two different IB interfaces and addresses on a machine: A B
- traffic is sent from A to B
- connection from A-B is created, connect request sent
- listening accepts connect request, B-A is created
- traffic flows, next_rx is incremented
- unacked messages exist on the retrans list
- connection A-B is shut down, new connect request sent
- listen sees existing loopback B-A, creates new passive B-A
- retrans messages are sent and delivered because of 0 next_rx
The problem is that the second connection request saw the previously
existing parent connection. Instead of using it, and using the existing
next_rx_seq state for the traffic between those IPs, it mistakenly
thought that it had to create a passive connection.
We fix this by only using passive connections in the special case where
laddr and faddr match. In this case we'll only ever have one parent
sending connection requests and one passive connection created as the
listening path sees the existing parent connection which initiated the
request.
Original patch by Zach Brown
Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
RDS: Documentation: Document AF_RDS, PF_RDS and SOL_RDS correctly.
AF_RDS, PF_RDS and SOL_RDS are available in header files,
and there is no need to get their values from /proc. Document
this correctly.
Fixes: 0c5f9b8830aa ("RDS: Documentation") Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
ALSA: hda - Fix headphone pin config for Lifebook T731
Some BIOS version of Fujitsu Lifebook T731 seems to set up the
headphone pin (0x21) without the assoc number 0x0f while it's set only
to the output on the docking port (0x1a). With the recent commit
[03ad6a8c93b6: ALSA: hda - Fix "PCM" name being used on one DAC when
there are two DACs], this resulted in the weird mixer element
mapping where the headphone on the laptop is assigned as a shared
volume with the speaker and the docking port is assigned as an
individual headphone.
This patch improves the situation by correcting the headphone pin
config to the more appropriate value.
Reported-and-tested-by: Taylor Smock <smocktaylor@gmail.com> Cc: <stable@vger.kernel.org> Signed-off-by: Takashi Iwai <tiwai@suse.de>
ALSA: bebob: fix to processing in big-endian machine for sending cue
Some M-Audio devices require to receive bootup command just after
powering on, while codes in BeBoB driver doesn't work properly in
big-endian machine because the command should be aligned by
little-endian.
This commit fixes this bug. This fix should go to stable kernel.
David Ahern reported that d63e2e1f3df9 breaks booting on an 8-socket T5
sparc system. He also verified that the system boots with d63e2e1f3df9
reverted. Yinghai has some fixes, but they need a little more polishing
than we can do before v4.0.
Bjorn Helgaas [Tue, 24 Mar 2015 16:12:45 +0000 (11:12 -0500)]
PCI: Don't look for ACPI hotplug parameters if ACPI is disabled
Booting a v3.18 or newer Xen domU kernel with PCI devices passed through
results in an oops (this is a 32-bit 3.13.11 dom0 with a 64-bit 4.4.0
hypervisor and 32-bit domU):
Don't look for ACPI configuration information if ACPI has been disabled.
I don't think this is the best fix, because we can boot plain Linux (no
Xen) with "acpi=off", and we don't need this check in pci_get_hp_params().
There should be a better fix that would make Xen domU work the same way.
The domU kernel has ACPI support but it has no AML. There should be a way
to initialize the ACPI data structures so things fail gracefully rather
than oopsing. This is an interim fix to address the regression.
Fixes: 6cd33649fa83 ("PCI: Add pci_configure_device() during enumeration") Link: https://bugzilla.kernel.org/show_bug.cgi?id=96301 Reported-by: Michael D Labriola <mlabriol@gdeb.com> Tested-by: Michael D Labriola <mlabriol@gdeb.com> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> CC: stable@vger.kernel.org # v3.18+
ARM: mvebu: use 0xf1000000 as internal registers on Armada 370 DB
All Marvell EBU SoCs (Kirkwood, Dove, Orion, Armada) have the
capability of changing the location of their internal registers (i.e
the registers for most hardware blocks inside the SoC). When coming
out of reset, the internal registers are mapped at 0xd0000000, but
since years and years, the tradition has been to have the internal
registers remapped at 0xf1000000 by the bootloader, and Linux has
since then assumed that the internal registers for the SoC were
located at 0xf1000000 on Kirkwood, Dove, Orion, etc. Linux has never
been aware that those registers are remappable (and there is no way to
know where they are mapped at runtime, since the register to configure
the address of the registers is itself within the internal registers).
Then came the Armada 370 and Armada XP, in which some of the very
early silicon steppings had an issue, which forced to use 0xd0000000:
the SoC was no longer working properly when the internal registers
were remapped at 0xf1000000. This issue is only affecting very early
silicon steppings and production steppings are not affected: the issue
has been fixed in between.
Since what we (Free Electrons) used to do the initial submission of
the Armada 370 and Armada XP platforms was evaluation boards with
those very early steppings, we submitted Device Tree that assumed the
internal registers were mapped at 0xd0000000. This is the case for
Armada 370 DB, Armada XP DB and Armada XP GP.
However, in practice, since Marvell has been shipping the evaluation
boards with production steppings of the SoC, they are shipping those
boards with bootloaders that remap the registers to 0xf1000000. We
have already changed this internal register address to 0xf1000000 for
the Armada XP DB in commit 82066bdb5a75 and for the Armada XP GP in
commit 91ed32200e6e (both merged in v3.15).
We only recently got our hand on an Armada 370 DB with a production
stepping of the SoC, which uses a bootloader that remaps internal
registers at 0xf1000000. Therefore, this commit aligns the Armada 370
DB to be like the Armada XP DB and Armada XP GP: assume that the
internal registers are mapped at 0xf1000000.
We would like to stress out the fact that the usage of 0xd0000000 as
the internal register base address was a temporary workaround for
early steppings deficiencies, and that the real long-term solution is
the usage of 0xf1000000. Having 0xd0000000 is an *accident* in the
life of the Marvell platform support in the kernel, as is confirmed by
the usage of 0xf1000000 in all previous Marvell platforms (Dove,
Kirkwood, Orion).
There are unfortunately a number of commercial devices that continue
to use 0xd0000000 even though they use production steppings of the
SoC, simply because the vendors of such devices have never bothered
using a more recent bootloader version from Marvell. There is not much
we can do about it, and we plan on keeping 0xd0000000 in the Device
Tree of such devices.
The main reason for remapping the internal registers at 0xf1000000
instead of 0xd0000000 is that it leaves more space in the 0 -> 4 GB
part of the physical address space for RAM. With registers at
0xd0000000, all RAM between 0xd0000000 to 0xffffffff is lost because
it's covered by the I/O registers.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Acked-by: Andrew Lunn <andrew@lunn.ch> Acked-by: Jason Cooper <jason@lakedameon.net> Acked-by: Gregory CLEMENT <gregory.clement@free-electrons.com> Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Abhi Das [Wed, 8 Apr 2015 14:03:56 +0000 (09:03 -0500)]
gfs2: fix quota refresh race in do_glock()
quotad periodically syncs in-memory quotas to the ondisk quota file
and sets the QDF_REFRESH flag so that a subsequent read of a synced
quota is re-read from disk.
gfs2_quota_lock() checks for this flag and sets a 'force' bit to
force re-read from disk if requested. However, there is a race
condition here. It is possible for gfs2_quota_lock() to find the
QDF_REFRESH flag unset (i.e force=0) and quotad comes in immediately
after and syncs the relevant quota and sets the QDF_REFRESH flag.
gfs2_quota_lock() resumes with force=0 and uses the stale in-memory
quota usage values that result in miscalculations.
This patch fixes this race by moving the check for the QDF_REFRESH
flag check further out into the gfs2_quota_lock() process, i.e, in
do_glock(), under the protection of the quota glock.
Signed-off-by: Abhi Das <adas@redhat.com> Signed-off-by: Bob Peterson <rpeterso@redhat.com> Acked-by: Steven Whitehouse <swhiteho@redhat.com>
Kailang Yang [Wed, 8 Apr 2015 08:34:00 +0000 (16:34 +0800)]
ALSA: hda/realtek - Make more stable to get pin sense for ALC283
Pin sense will active when power pin is wake up.
Power pin will not wake up immediately during resume state.
Add some delay to wait for power pin activated.
Signed-off-by: Kailang Yang <kailang@realtek.com> Cc: <stable@vger.kernel.org> Signed-off-by: Takashi Iwai <tiwai@suse.de>
drm: Fix deadlock due to getconnector locking changes
If the drm_connector_find() call returns NULL, we should no longer
call drm_modeset_unlock() to avoid locking imbalance.
Signed-off-by: Tommi Rantala <tt.rantala@gmail.com> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Dave Airlie <airlied@redhat.com>
Michael Halcrow [Wed, 8 Apr 2015 04:20:21 +0000 (00:20 -0400)]
ext4 crypto: filename encryption modifications
Modifies htree_dirblock_to_tree, dx_make_map, ext4_match search_dir,
and ext4_find_dest_de to support fname crypto. Filename encryption
feature is not yet enabled at this patch.
ext4 crypto: teach ext4_htree_store_dirent() to store decrypted filenames
For encrypted directories, we need to pass in a separate parameter for
the decrypted filename, since the directory entry contains the
encrypted filename.
Michael Halcrow [Wed, 8 Apr 2015 04:19:28 +0000 (00:19 -0400)]
ext4 crypto: add ext4 encryption facilities
On encrypt, we will re-assign the buffer_heads to point to a bounce
page rather than the control_page (which is the original page to write
that contains the plaintext). The block I/O occurs against the bounce
page. On write completion, we re-assign the buffer_heads to the
original plaintext page.
On decrypt, we will attach a read completion callback to the bio
struct. This read completion will decrypt the read contents in-place
prior to setting the page up-to-date.
The current encryption mode, AES-256-XTS, lacks cryptographic
integrity. AES-256-GCM is in-plan, but we will need to devise a
mechanism for handling the integrity data.
This takes code from fs/mpage.c and optimizes it for ext4. Its
primary reason is to allow us to more easily add encryption to ext4's
read path in an efficient manner.
mm: numa: disable change protection for vma(VM_HUGETLB)
Currently when a process accesses a hugetlb range protected with
PROTNONE, unexpected COWs are triggered, which finally puts the hugetlb
subsystem into a broken/uncontrollable state, where for example
h->resv_huge_pages is subtracted too much and wraps around to a very
large number, and the free hugepage pool is no longer maintainable.
This patch simply stops changing protection for vma(VM_HUGETLB) to fix
the problem. And this also allows us to avoid useless overhead of minor
faults.
Signed-off-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Suggested-by: Mel Gorman <mgorman@suse.de> Cc: Hugh Dickins <hughd@google.com> Cc: "Kirill A. Shutemov" <kirill@shutemov.name> Cc: David Rientjes <rientjes@google.com> Cc: Rik van Riel <riel@redhat.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Mark Brown [Tue, 7 Apr 2015 21:26:44 +0000 (14:26 -0700)]
include/linux/dmapool.h: declare struct device
dmapool uses struct device in function arguments but relies on an
implicit inclusion to declare struct device causing warnings in some
configurations:
include/linux/dmapool.h:31:7: warning: 'struct device' declared inside parameter list
Fix this by adding a struct device declaration to the file.
Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm: move zone lock to a different cache line than order-0 free page lists
Huang Ying reported the following problem due to commit 3484b2de9499 ("mm:
rearrange zone fields into read-only, page alloc, statistics and page
reclaim lines") from the Intel performance tests
The problem is specific to very large machines under stress. It was not
reproducible with the machines I had used to justify the original patch
because large numbers of CPUs are required. When pressure is high enough,
the cache line is bouncing between CPUs trying to acquire the lock and the
holder of the lock adjusting free lists. The intention was that the
acquirer of the lock would automatically have the cache line holding the
free lists but according to Huang, this is not a universal win.
One possibility is to move the zone lock to its own cache line but it
increases the size of the zone. This patch moves the lock to the other
end of the free lists where they do not contend under high pressure. It
does mean the page allocator paths now require more cache lines but Huang
reports that it restores performance to previous levels on large machines
Broadcom BCM63xx DSL SoCs utilize BMIPS CPUs, and as such are required
to perform a read-ahead cache flush after a DMA transfer. Utilize
asm/bmips.h to provide a plat_post_dma_flush_hook, and
mach-generic/dma-coherence.h for everything else.
MIPS: DMA: Allow platforms to override only the post DMA hook
Instead of having platforms to copy the entirety of
mach-generic/dma-coherence.h, check whether these platforms have already
defined a plat_post_dma_flush hook, and if not, provide an inline stub.