Skip to content

K
Kernel

openeuler / Kernel
1 年多 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 08 4月, 2015 1 次提交
  3. 03 4月, 2015 1 次提交
  5. 13 3月, 2015 4 次提交
  7. 07 3月, 2015 2 次提交
  9. 06 3月, 2015 1 次提交
  11. 05 3月, 2015 5 次提交
    • T
      workqueue: fix hang involving racing cancel[_delayed]_work_sync()'s for PREEMPT_NONE · 8603e1b3
      由 Tejun Heo 提交于 
      cancel[_delayed]_work_sync() are implemented using
__cancel_work_timer() which grabs the PENDING bit using
try_to_grab_pending() and then flushes the work item with PENDING set
to prevent the on-going execution of the work item from requeueing
itself.

try_to_grab_pending() can always grab PENDING bit without blocking
except when someone else is doing the above flushing during
cancelation.  In that case, try_to_grab_pending() returns -ENOENT.  In
this case, __cancel_work_timer() currently invokes flush_work().  The
assumption is that the completion of the work item is what the other
canceling task would be waiting for too and thus waiting for the same
condition and retrying should allow forward progress without excessive
busy looping

Unfortunately, this doesn't work if preemption is disabled or the
latter task has real time priority.  Let's say task A just got woken
up from flush_work() by the completion of the target work item.  If,
before task A starts executing, task B gets scheduled and invokes
__cancel_work_timer() on the same work item, its try_to_grab_pending()
will return -ENOENT as the work item is still being canceled by task A
and flush_work() will also immediately return false as the work item
is no longer executing.  This puts task B in a busy loop possibly
preventing task A from executing and clearing the canceling state on
the work item leading to a hang.

task A			task B			worker

						executing work
__cancel_work_timer()
  try_to_grab_pending()
  set work CANCELING
  flush_work()
    block for work completion
						completion, wakes up A
			__cancel_work_timer()
			while (forever) {
			  try_to_grab_pending()
			    -ENOENT as work is being canceled
			  flush_work()
			    false as work is no longer executing
			}

This patch removes the possible hang by updating __cancel_work_timer()
to explicitly wait for clearing of CANCELING rather than invoking
flush_work() after try_to_grab_pending() fails with -ENOENT.

Link: http://lkml.kernel.org/g/20150206171156.GA8942@axis.com

v3: bit_waitqueue() can't be used for work items defined in vmalloc
    area.  Switched to custom wake function which matches the target
    work item and exclusive wait and wakeup.

v2: v1 used wake_up() on bit_waitqueue() which leads to NULL deref if
    the target bit waitqueue has wait_bit_queue's on it.  Use
    DEFINE_WAIT_BIT() and __wake_up_bit() instead.  Reported by Tomeu
    Vizoso.
Signed-off-by: NTejun Heo <tj@kernel.org>
Reported-by: NRabin Vincent <rabin.vincent@axis.com>
Cc: Tomeu Vizoso <tomeu.vizoso@gmail.com>
Cc: stable@vger.kernel.org
Tested-by: NJesper Nilsson <jesper.nilsson@axis.com>
Tested-by: NRabin Vincent <rabin.vincent@axis.com>
      8603e1b3
    • A
      drm/ttm: device address space != CPU address space · 54c4cd68
      由 Alex Deucher 提交于 
      We need to store device offsets in 64 bit as the device
address space may be larger than the CPU's.

Fixes GPU init failures on radeons with 4GB or more of
vram on 32 bit kernels.  We put vram at the start of the
GPU's address space so the gart aperture starts at 4 GB
causing all GPU addresses in the gart aperture to get
truncated.

bug:
https://bugs.freedesktop.org/show_bug.cgi?id=89072

[airlied: fix warning on nouveau build]
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: thellstrom@vmware.com
Acked-by: NThomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>
      54c4cd68
    • T
      drm/mm: Support 4 GiB and larger ranges · 440fd528
      由 Thierry Reding 提交于 
      The current implementation is limited by the number of addresses that
fit into an unsigned long. This causes problems on 32-bit Tegra where
unsigned long is 32-bit but drm_mm is used to manage an IOVA space of
4 GiB. Given the 32-bit limitation, the range is limited to 4 GiB - 1
(or 4 GiB - 4 KiB for page granularity).

This commit changes the start and size of the range to be an unsigned
64-bit integer, thus allowing much larger ranges to be supported.

[airlied: fix i915 warnings and coloring callback]
Signed-off-by: NThierry Reding <treding@nvidia.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDave Airlie <airlied@redhat.com>

fixupo
      440fd528
    • R
      genirq / PM: Add flag for shared NO_SUSPEND interrupt lines · 17f48034
      由 Rafael J. Wysocki 提交于 
      It currently is required that all users of NO_SUSPEND interrupt
lines pass the IRQF_NO_SUSPEND flag when requesting the IRQ or the
WARN_ON_ONCE() in irq_pm_install_action() will trigger.  That is
done to warn about situations in which unprepared interrupt handlers
may be run unnecessarily for suspended devices and may attempt to
access those devices by mistake.  However, it may cause drivers
that have no technical reasons for using IRQF_NO_SUSPEND to set
that flag just because they happen to share the interrupt line
with something like a timer.

Moreover, the generic handling of wakeup interrupts introduced by
commit 9ce7a258 (genirq: Simplify wakeup mechanism) only works
for IRQs without any NO_SUSPEND users, so the drivers of wakeup
devices needing to use shared NO_SUSPEND interrupt lines for
signaling system wakeup generally have to detect wakeup in their
interrupt handlers.  Thus if they happen to share an interrupt line
with a NO_SUSPEND user, they also need to request that their
interrupt handlers be run after suspend_device_irqs().

In both cases the reason for using IRQF_NO_SUSPEND is not because
the driver in question has a genuine need to run its interrupt
handler after suspend_device_irqs(), but because it happens to
share the line with some other NO_SUSPEND user.  Otherwise, the
driver would do without IRQF_NO_SUSPEND just fine.

To make it possible to specify that condition explicitly, introduce
a new IRQ action handler flag for shared IRQs, IRQF_COND_SUSPEND,
that, when set, will indicate to the IRQ core that the interrupt
user is generally fine with suspending the IRQ, but it also can
tolerate handler invocations after suspend_device_irqs() and, in
particular, it is capable of detecting system wakeup and triggering
it as appropriate from its interrupt handler.

That will allow us to work around a problem with a shared timer
interrupt line on at91 platforms.

Link: http://marc.info/?l=linux-kernel&m=142252777602084&w=2
Link: http://marc.info/?t=142252775300011&r=1&w=2
Link: https://lkml.org/lkml/2014/12/15/552Reported-by: NBoris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: NMark Rutland <mark.rutland@arm.com>
      17f48034
    • P
      netfilter: nf_tables: fix userdata length overflow · 86f1ec32
      由 Patrick McHardy 提交于 
      The NFT_USERDATA_MAXLEN is defined to 256, however we only have a u8
to store its size. Introduce a struct nft_userdata which contains a
length field and indicate its presence using a single bit in the rule.

The length field of struct nft_userdata is also a u8, however we don't
store zero sized data, so the actual length is udata->len + 1.
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>
      86f1ec32
  13. 04 3月, 2015 1 次提交
    • T
      NFS: Fix a regression in the read() syscall · 874f9463
      由 Trond Myklebust 提交于 
      When invalidating the page cache for a regular file, we want to first
sync all dirty data to disk and then call invalidate_inode_pages2().
The latter relies on nfs_launder_page() and nfs_release_page() to deal
respectively with dirty pages, and unstable written pages.

When commit 95905446 ("NFS: avoid deadlocks with loop-back mounted
NFS filesystems.") changed the behaviour of nfs_release_page(), then it
made it possible for invalidate_inode_pages2() to fail with an EBUSY.
Unfortunately, that error is then propagated back to read().

Let's therefore work around the problem for now by protecting the call
to sync the data and invalidate_inode_pages2() so that they are atomic
w.r.t. the addition of new writes.
Later on, we can revisit whether or not we still need nfs_launder_page()
and nfs_release_page().
Signed-off-by: NTrond Myklebust <trond.myklebust@primarydata.com>
      874f9463
  15. 03 3月, 2015 2 次提交
  17. 02 3月, 2015 3 次提交
  19. 28 2月, 2015 1 次提交
  21. 27 2月, 2015 1 次提交
  23. 26 2月, 2015 2 次提交
    • T
      OMAPDSS: fix regression with display sysfs files · a38bb793
      由 Tomi Valkeinen 提交于 
      omapdss's sysfs directories for displays used to have 'name' file,
giving the name for the display. This file was later renamed to
'display_name' to avoid conflicts with i2c sysfs 'name' file. Looks like
at least xserver-xorg-video-omap3 requires the 'name' file to be
present.

To fix the regression, this patch creates new kobjects for each display,
allowing us to create sysfs directories for the displays. This way we
have the whole directory for omapdss, and there will be no sysfs file
clashes with the underlying display device's sysfs files.

We can thus add the 'name' sysfs file back.
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>
Tested-by: NNeilBrown <neilb@suse.de>
      a38bb793
    • M
      genirq / PM: better describe IRQF_NO_SUSPEND semantics · 737eb030
      由 Mark Rutland 提交于 
      The IRQF_NO_SUSPEND flag is intended to be used for interrupts required
to be enabled during the suspend-resume cycle. This mostly consists of
IPIs and timer interrupts, potentially including chained irqchip
interrupts if these are necessary to handle timers or IPIs. If an
interrupt does not fall into one of the aforementioned categories,
requesting it with IRQF_NO_SUSPEND is likely incorrect.

Using IRQF_NO_SUSPEND does not guarantee that the interrupt can wake the
system from a suspended state. For an interrupt to be able to trigger a
wakeup, it may be necessary to program various components of the system.
In these cases it is necessary to use {enable,disabled}_irq_wake.

Unfortunately, several drivers assume that IRQF_NO_SUSPEND ensures that
an IRQ can wake up the system, and the documentation can be read
ambiguously w.r.t. this property.

This patch updates the documentation regarding IRQF_NO_SUSPEND to make
this caveat explicit, hopefully making future misuse rarer. Cleanup of
existing misuse will occur as part of later patch series.
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Acked-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
      737eb030
  25. 25 2月, 2015 1 次提交
  27. 24 2月, 2015 2 次提交
    • J
      net: sched: export tc_connmark.h so it is uapi accessible · 30ff5476
      由 Jamal Hadi Salim 提交于 
      Signed-off-by: NJamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      30ff5476
    • D
      x86/xen: allow privcmd hypercalls to be preempted · fdfd811d
      由 David Vrabel 提交于 
      Hypercalls submitted by user space tools via the privcmd driver can
take a long time (potentially many 10s of seconds) if the hypercall
has many sub-operations.

A fully preemptible kernel may deschedule such as task in any upcall
called from a hypercall continuation.

However, in a kernel with voluntary or no preemption, hypercall
continuations in Xen allow event handlers to be run but the task
issuing the hypercall will not be descheduled until the hypercall is
complete and the ioctl returns to user space.  These long running
tasks may also trigger the kernel's soft lockup detection.

Add xen_preemptible_hcall_begin() and xen_preemptible_hcall_end() to
bracket hypercalls that may be preempted.  Use these in the privcmd
driver.

When returning from an upcall, call xen_maybe_preempt_hcall() which
adds a schedule point if if the current task was within a preemptible
hypercall.

Since _cond_resched() can move the task to a different CPU, clear and
set xen_in_preemptible_hcall around the call.
Signed-off-by: NDavid Vrabel <david.vrabel@citrix.com>
Reviewed-by: NBoris Ostrovsky <boris.ostrovsky@oracle.com>
      fdfd811d
  29. 23 2月, 2015 5 次提交
  31. 22 2月, 2015 2 次提交
  33. 21 2月, 2015 2 次提交
    • D
      caif: fix a signedness bug in cfpkt_iterate() · 278f7b4f
      由 Dan Carpenter 提交于 
      The cfpkt_iterate() function can return -EPROTO on error, but the
function is a u16 so the negative value gets truncated to a positive
unsigned short.  This causes a static checker warning.

The only caller which might care is cffrml_receive(), when it's checking
the frame checksum.  I modified cffrml_receive() so that it never says
-EPROTO is a valid checksum.

Also this isn't ever going to be inlined so I removed the "inline".
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      278f7b4f
    • G
      net: Initialize all members in skb_gro_remcsum_init() · 846cd667
      由 Geert Uytterhoeven 提交于 
      skb_gro_remcsum_init() initializes the gro_remcsum.delta member only,
leading to compiler warnings about a possibly uninitialized
gro_remcsum.offset member:

drivers/net/vxlan.c: In function ‘vxlan_gro_receive’:
drivers/net/vxlan.c:602: warning: ‘grc.offset’ may be used uninitialized in this function
net/ipv4/fou.c: In function ‘gue_gro_receive’:
net/ipv4/fou.c:262: warning: ‘grc.offset’ may be used uninitialized in this function

While these are harmless for now:
  - skb_gro_remcsum_process() sets offset before changing delta,
  - skb_gro_remcsum_cleanup() checks if delta is non-zero before
    accessing offset,
it's safer to let the initialization function initialize all members.
Signed-off-by: NGeert Uytterhoeven <geert@linux-m68k.org>
Acked-by: NTom Herbert <therbert@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      846cd667
  35. 20 2月, 2015 4 次提交
    • K
      NVMe: Fix potential corruption during shutdown · 07836e65
      由 Keith Busch 提交于 
      The driver has to end unreturned commands at some point even if the
controller has not provided a completion. The driver tried to be safe by
deleting IO queues prior to ending all unreturned commands. That should
cause the controller to internally abort inflight commands, but IO queue
deletion request does not have to be successful, so all bets are off. We
still have to make progress, so to be extra safe, this patch doesn't
clear a queue to release the dma mapping for a command until after the
pci device has been disabled.

This patch removes the special handling during device initialization
so controller recovery can be done all the time. This is possible since
initialization is not inlined with pci probe anymore.
Reported-by: NNilish Choudhury <nilesh.choudhury@oracle.com>
Signed-off-by: NKeith Busch <keith.busch@intel.com>
      07836e65
    • K
      NVMe: Asynchronous controller probe · 2e1d8448
      由 Keith Busch 提交于 
      This performs the longest parts of nvme device probe in scheduled work.
This speeds up probe significantly when multiple devices are in use.
Signed-off-by: NKeith Busch <keith.busch@intel.com>
      2e1d8448
    • K
      NVMe: Register management handle under nvme class · b3fffdef
      由 Keith Busch 提交于 
      This creates a new class type for nvme devices to register their
management character devices with. This is so we do not rely on miscdev
to provide enough minors for as many nvme devices some people plan to
use. The previous limit was approximately 60 NVMe controllers, depending
on the platform and kernel. Now the limit is 1M, which ought to be enough
for anybody.

Since we have a new device class, it makes sense to attach the block
devices under this as well, so part of this patch moves the management
handle initialization prior to the namespaces discovery.
Signed-off-by: NKeith Busch <keith.busch@intel.com>
      b3fffdef
    • K
      NVMe: Update SCSI Inquiry VPD 83h translation · 4f1982b4
      由 Keith Busch 提交于 
      The original translation created collisions on Inquiry VPD 83 for many
existing devices. Newer specifications provide other ways to translate
based on the device's version can be used to create unique identifiers.

Version 1.1 provides an EUI64 field that uniquely identifies each
namespace, and 1.2 added the longer NGUID field for the same reason.
Both follow the IEEE EUI format and readily translate to the SCSI device
identification EUI designator type 2h. For devices implementing either,
the translation will use this type, defaulting to the EUI64 8-byte type if
implemented then NGUID's 16 byte version if not. If neither are provided,
the 1.0 translation is used, and is updated to use the SCSI String format
to guarantee a unique identifier.

Knowing when to use the new fields depends on the nvme controller's
revision. The NVME_VS macro was not decoding this correctly, so that is
fixed in this patch and moved to a more appropriate place.

Since the Identify Namespace structure required an update for the NGUID
field, this patch adds the remaining new 1.2 fields to the structure.
Signed-off-by: NKeith Busch <keith.busch@intel.com>
      4f1982b4