Hi, Jens,
If you recall, I posted an RFC patch for this back in July of last year:
http://lkml.org/lkml/2010/7/13/279
The basic problem is that a process can issue a never-ending stream of
async direct I/Os to the same sector on a device, thus starving out
other I/O in the system (due to the way the alias handling works in both
cfq and deadline). The solution I proposed back then was to start
dispatching from the fifo after a certain number of aliases had been
dispatched. Vivek asked why we had to treat aliases differently at all,
and I never had a good answer. So, I put together a simple patch which
allows aliases to be added to the rb tree (it adds them to the right,
though that doesn't matter as the order isn't guaranteed anyway). I
think this is the preferred solution, as it doesn't break up time slices
in CFQ or batches in deadline. I've tested it, and it does solve the
starvation issue. Let me know what you think.
Cheers,
Jeff
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
This reverts commit a197b59ae6.
As rmk says:
"Commit a197b59ae6 (mm: fail GFP_DMA allocations when ZONE_DMA is not
configured) is causing regressions on ARM with various drivers which
use GFP_DMA.
The behaviour up until now has been to silently ignore that flag when
CONFIG_ZONE_DMA is not enabled, and to allocate from the normal zone.
However, as a result of the above commit, such allocations now fail
which causes drivers to fail. These are regressions compared to the
previous kernel version."
so just revert it.
Requested-by: Russell King <linux@arm.linux.org.uk>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* git://git.infradead.org/iommu-2.6:
intel-iommu: Fix off-by-one in RMRR setup
intel-iommu: Add domain check in domain_remove_one_dev_info
intel-iommu: Remove Host Bridge devices from identity mapping
intel-iommu: Use coherent DMA mask when requested
intel-iommu: Dont cache iova above 32bit
intel-iommu: Speed up processing of the identity_mapping function
intel-iommu: Check for identity mapping candidate using system dma mask
intel-iommu: Only unlink device domains from iommu
intel-iommu: Enable super page (2MiB, 1GiB, etc.) support
intel-iommu: Flush unmaps at domain_exit
intel-iommu: Remove obsolete comment from detect_intel_iommu
intel-iommu: fix VT-d PMR disable for TXT on S3 resume
Jens' back-merge commit 698567f3fa ("Merge commit 'v2.6.39' into
for-2.6.40/core") was incorrectly done, and re-introduced the
DISK_EVENT_MEDIA_CHANGE lines that had been removed earlier in commits
- 9fd097b149 ("block: unexport DISK_EVENT_MEDIA_CHANGE for
legacy/fringe drivers")
- 7eec77a181 ("ide: unexport DISK_EVENT_MEDIA_CHANGE for ide-gd
and ide-cd")
because of conflicts with the "g->flags" updates near-by by commit
d4dc210f69 ("block: don't block events on excl write for non-optical
devices")
As a result, we re-introduced the hanging behavior due to infinite disk
media change reports.
Tssk, tssk, people! Don't do back-merges at all, and *definitely* don't
do them to hide merge conflicts from me - especially as I'm likely
better at merging them than you are, since I do so many merges.
Reported-by: Steven Rostedt <rostedt@goodmis.org>
Cc: Jens Axboe <jaxboe@fusionio.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We were mapping an extra byte (and hence usually an extra page):
iommu_prepare_identity_map() expects to be given an 'end' argument which
is the last byte to be mapped; not the first byte *not* to be mapped.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
The comment in domain_remove_one_dev_info() states "No need to compare
PCI domain; it has to be the same". But for the si_domain that isn't
going to be true, as it consists of all the PCI devices that are
identity mapped thus multiple PCI domains can be in si_domain. The
code needs to validate the PCI domain too.
Signed-off-by: Mike Habeck <habeck@sgi.com>
Signed-off-by: Mike Travis <travis@sgi.com>
Cc: stable@kernel.org
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
When using the 1:1 (identity) PCI DMA remapping, PCI Host Bridge devices
that do not use the IOMMU causes a kernel panic. Fix that by not
inserting those devices into the si_domain.
Signed-off-by: Mike Travis <travis@sgi.com>
Reviewed-by: Mike Habeck <habeck@sgi.com>
Cc: stable@kernel.org
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
The __intel_map_single function is not honoring the passed in DMA mask.
This results in not using the coherent DMA mask when called from
intel_alloc_coherent().
Signed-off-by: Mike Travis <travis@sgi.com>
Acked-by: Chris Wright <chrisw@sous-sol.org>
Reviewed-by: Mike Habeck <habeck@sgi.com>
Cc: stable@kernel.org
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Mike Travis and Mike Habeck reported an issue where iova allocation
would return a range that was larger than a device's dma mask.
https://lkml.org/lkml/2011/3/29/423
The dmar initialization code will reserve all PCI MMIO regions and copy
those reservations into a domain specific iova tree. It is possible for
one of those regions to be above the dma mask of a device. It is typical
to allocate iovas with a 32bit mask (despite device's dma mask possibly
being larger) and cache the result until it exhausts the lower 32bit
address space. Freeing the iova range that is >= the last iova in the
lower 32bit range when there is still an iova above the 32bit range will
corrupt the cached iova by pointing it to a region that is above 32bit.
If that region is also larger than the device's dma mask, a subsequent
allocation will return an unusable iova and cause dma failure.
Simply don't cache an iova that is above the 32bit caching boundary.
Reported-by: Mike Travis <travis@sgi.com>
Reported-by: Mike Habeck <habeck@sgi.com>
Cc: stable@kernel.org
Acked-by: Mike Travis <travis@sgi.com>
Tested-by: Mike Habeck <habeck@sgi.com>
Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
When there are a large count of PCI devices, and the pass through
option for iommu is set, much time is spent in the identity_mapping
function hunting though the iommu domains to check if a specific
device is "identity mapped".
Speed up the function by checking the cached info to see if
it's mapped to the static identity domain.
Signed-off-by: Mike Travis <travis@sgi.com>
Reviewed-by: Mike Habeck <habeck@sgi.com>
Cc: stable@kernel.org
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
The identity mapping code appears to make the assumption that if the
devices dma_mask is greater than 32bits the device can use identity
mapping. But that is not true: take the case where we have a 40bit
device in a 44bit architecture. The device can potentially receive a
physical address that it will truncate and cause incorrect addresses
to be used.
Instead check to see if the device's dma_mask is large enough
to address the system's dma_mask.
Signed-off-by: Mike Travis <travis@sgi.com>
Reviewed-by: Mike Habeck <habeck@sgi.com>
Cc: stable@kernel.org
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Commit a97590e5 added unlinking domains from iommus to reciprocate the
iommu from domains unlinking that was already done. We actually want
to only do this for device domains and never for the static
identity map domain or VM domains. The SI domain is special and
never freed, while VM domain->id lives in their own special address
space, separate from iommu->domain_ids.
In the current code, a VM can get domain->id zero, then mark that
domain unused when unbound from pci-stub. This leads to DMAR
write faults when the device is re-bound to the host driver.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Cc: stable@kernel.org
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
There are no externally-visible changes with this. In the loop in the
internal __domain_mapping() function, we simply detect if we are mapping:
- size >= 2MiB, and
- virtual address aligned to 2MiB, and
- physical address aligned to 2MiB, and
- on hardware that supports superpages.
(and likewise for larger superpages).
We automatically use a superpage for such mappings. We never have to
worry about *breaking* superpages, since we trust that we will always
*unmap* the same range that was mapped. So all we need to do is ensure
that dma_pte_clear_range() will also cope with superpages.
Adjust pfn_to_dma_pte() to take a superpage 'level' as an argument, so
it can return a PTE at the appropriate level rather than always
extending the page tables all the way down to level 1. Again, this is
simplified by the fact that we should never encounter existing small
pages when we're creating a mapping; any old mapping that used the same
virtual range will have been entirely removed and its obsolete page
tables freed.
Provide an 'intel_iommu=sp_off' argument on the command line as a
chicken bit. Not that it should ever be required.
==
The original commit seen in the iommu-2.6.git was Youquan's
implementation (and completion) of my own half-baked code which I'd
typed into an email. Followed by half a dozen subsequent 'fixes'.
I've taken the unusual step of rewriting history and collapsing the
original commits in order to keep the main history simpler, and make
life easier for the people who are going to have to backport this to
older kernels. And also so I can give it a more coherent commit comment
which (hopefully) gives a better explanation of what's going on.
The original sequence of commits leading to identical code was:
Youquan Song (3):
intel-iommu: super page support
intel-iommu: Fix superpage alignment calculation error
intel-iommu: Fix superpage level calculation error in dma_pfn_level_pte()
David Woodhouse (4):
intel-iommu: Precalculate superpage support for dmar_domain
intel-iommu: Fix hardware_largepage_caps()
intel-iommu: Fix inappropriate use of superpages in __domain_mapping()
intel-iommu: Fix phys_pfn in __domain_mapping for sglist pages
Signed-off-by: Youquan Song <youquan.song@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Fix build warnings in physmap.h:
include/linux/mtd/physmap.h:25: warning: 'struct platform_device' declared inside parameter list
include/linux/mtd/physmap.h:25: warning: its scope is only this definition or declaration, which is probably not what you want
include/linux/mtd/physmap.h:26: warning: 'struct platform_device' declared inside parameter list
include/linux/mtd/physmap.h:27: warning: 'struct platform_device' declared inside parameter list
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
The new instruction_pointer_set helper is defined for people who have
converted to asm-generic/ptrace.h, so don't use it generally unless
the arch needs it (in which case it has been converted). This should
fix building of kgdb tests for arches not yet converted.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Jason Wessel <jason.wessel@windriver.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When invalid parameters are passed to apparmor_setprocattr a NULL deref
oops occurs when it tries to record an audit message. This is because
it is passing NULL for the profile parameter for aa_audit. But aa_audit
now requires that the profile passed is not NULL.
Fix this by passing the current profile on the task that is trying to
setprocattr.
Signed-off-by: Kees Cook <kees@ubuntu.com>
Signed-off-by: John Johansen <john.johansen@canonical.com>
Cc: stable@kernel.org
Signed-off-by: James Morris <jmorris@namei.org>
* git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus:
virtio_net: delay TX callbacks
virtio: add api for delayed callbacks
virtio_test: support event index
vhost: support event index
virtio_ring: support event idx feature
virtio ring: inline function to check for events
virtio: event index interface
virtio: add full three-clause BSD text to headers.
virtio balloon: kill tell-host-first logic
virtio console: don't manually set or finalize VIRTIO_CONSOLE_F_MULTIPORT.
drivers, block: virtio_blk: Replace cryptic number with the macro
virtio_blk: allow re-reading config space at runtime
lguest: remove support for VIRTIO_F_NOTIFY_ON_EMPTY.
lguest: fix up compilation after move
lguest: fix timer interrupt setup
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86: Fix mwait_play_dead() faulting on mwait-incapable cpus
x86 idle: Fix mwait deprecation warning message
Evil merge to remove extra quote noticed by Joe Perches
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86: Put back -pg to tsc.o and add no GCOV to vread_tsc_64.o
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6:
autofs4: bogus dentry_unhash() added in ->unlink()
vfs: shrink_dcache_parent before rmdir, dir rename
The Apple custom PIC only exist in some earlier machine models,
anything with an MPIC will crash on suspend if we register those
syscore ops unconditionally.
This is a regression caused by commit f5a592f7d7 ("PM / PowerPC: Use
struct syscore_ops instead of sysdevs for PM")
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit cc3ce5176d (rcu: Start RCU kthreads in TASK_INTERRUPTIBLE
state) fudges a sleeping task' state, resulting in the scheduler seeing
a TASK_UNINTERRUPTIBLE task going to sleep, but a TASK_INTERRUPTIBLE
task waking up. The result is unbalanced load calculation.
The problem that patch tried to address is that the RCU threads could
stay in UNINTERRUPTIBLE state for quite a while and triggering the hung
task detector due to on-demand wake-ups.
Cure the problem differently by always giving the tasks at least one
wake-up once the CPU is fully up and running, this will kick them out of
the initial UNINTERRUPTIBLE state and into the regular INTERRUPTIBLE
wait state.
[ The alternative would be teaching kthread_create() to start threads as
INTERRUPTIBLE but that needs a tad more thought. ]
Reported-by: Damien Wyart <damien.wyart@free.fr>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Paul E. McKenney <paul.mckenney@linaro.org>
Link: http://lkml.kernel.org/r/1306755291.1200.2872.camel@twins
Signed-off-by: Ingo Molnar <mingo@elte.hu>
A logic error in mwait_play_dead() causes the kernel to use
mwait even on cpus which don't support it, such as KVM virtual
cpus.
Introduced by:
349c004e3d: x86: A fast way to check capabilities of the current cpu
Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=36222
Reported-by: Török Edwin <edwintorok@gmail.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Tejun Heo <tj@kernel.org>
Link: http://lkml.kernel.org/r/1306758237-9327-1-git-send-email-avi@redhat.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The dentry_unhash push-down series missed that shink_dcache_parent needs to
be called prior to rmdir or dir rename to clear DCACHE_REFERENCED and
allow efficient dentry reclaim.
Reported-by: Dave Chinner <david@fromorbit.com>
Signed-off-by: Sage Weil <sage@newdream.net>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Ask for delayed callbacks on TX ring full, to give the
other side more of a chance to make progress.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Add an API that tells the other side that callbacks
should be delayed until a lot of work has been done.
Implement using the new event_idx feature.
Note: it might seem advantageous to let the drivers
ask for a callback after a specific capacity has
been reached. However, as a single head can
free many entries in the descriptor table,
we don't really have a clue about capacity
until get_buf is called. The API is the simplest
to implement at the moment, we'll see what kind of
hints drivers can pass when there's more than one
user of the feature.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Support the new event index feature. When acked,
utilize it to reduce the # of interrupts sent to the guest.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Support for the new event idx feature:
1. When enabling interrupts, publish the current avail index
value to the host to get interrupts on the next update.
2. Use the new avail_event feature to reduce the number
of exits from the guest.
Simple test with the simulator:
[virtio]# time ./virtio_test
spurious wakeus: 0x7
real 0m0.169s
user 0m0.140s
sys 0m0.019s
[virtio]# time ./virtio_test --no-event-idx
spurious wakeus: 0x11
real 0m0.649s
user 0m0.295s
sys 0m0.335s
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
With the new used_event and avail_event and features, both
host and guest need similar logic to check whether events are
enabled, so it helps to put the common code in the header.
Note that Xen has similar logic for notification hold-off
in include/xen/interface/io/ring.h with req_event and req_prod
corresponding to event_idx + 1 and new_idx respectively.
+1 comes from the fact that req_event and req_prod in Xen start at 1,
while event index in virtio starts at 0.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Define a new feature bit for the guest and host to utilize
an event index (like Xen) instead if a flag bit to enable/disable
interrupts and kicks.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
It's unclear to me if it's important, but it's obviously causing my
technical colleages some headaches and I'd hate such imprecision to
slow virtio adoption.
I've emailed this to all non-trivial contributors for approval, too.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Acked-by: Ryan Harper <ryanh@us.ibm.com>
Acked-by: Anthony Liguori <aliguori@us.ibm.com>
Acked-by: Eric Van Hensbergen <ericvh@gmail.com>
Acked-by: john cooper <john.cooper@redhat.com>
Acked-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Acked-by: Fernando Luis Vazquez Cao <fernando@oss.ntt.co.jp>
The virtio balloon driver has a VIRTIO_BALLOON_F_MUST_TELL_HOST
feature bit. Whenever the bit is set, the guest kernel must
always tell the host before we free pages back to the allocator.
Without this feature, we might free a page (and have another
user touch it) while the hypervisor is unprepared for it.
But, if the bit is _not_ set, we are under no obligation to
reverse the order; we're under no obligation to do _anything_.
As of now, qemu-kvm defines the bit, but doesn't set it.
This patch makes the "tell host first" logic the only case. This
should make everybody happy, and reduce the amount of untested or
untestable code in the kernel.
This _also_ means that we don't have to preserve a pfn list
after the pages are freed, which should let us get rid of some
temporary storage (vb->pfns) eventually.
Signed-off-by: Dave Hansen <dave@linux.vnet.ibm.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
That's already been done by the virtio infrastructure before the probe
function is called.
Reported-by: alexey.kardashevskiy@au1.ibm.com
Acked-by: Amit Shah <amit.shah@redhat.com>
Tested-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
It is easier to figure out the context by reading SCSI_SENSE_BUFFERSIZE
instead of plain '96'.
Signed-off-by: Liu Yuan <tailai.ly@taobao.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Wire up the virtio_driver config_changed method to get notified about
config changes raised by the host. For now we just re-read the device
size to support online resizing of devices, but once we add more
attributes that might be changeable they could be added as well.
Note that the config_changed method is called from irq context, so
we'll have to use the workqueue infrastructure to provide us a proper
user context for our changes.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Without an IRQ chip set, we now get a WARN_ON and no timer interrupt. This
prevents booting.
Fortunately, the fix is a one-liner: set up the timer IRQ like everything
else.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: stable@kernel.org # .39.x
.. except there are various scripts that really know that there are
three numbers, so it calls itself "3.0.0-rc1".
Hopefully by the time the final 3.0 is out, we'll have that extra zero
all figured out.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We had a few drivers move from arch/arm into drivers/gpio, but they
don't actually compile without the ARM platform headers etc. As a
result they were messing up allyesconfig on x86.
Make them depend on ARM.
Reported-by: Ingo Molnar <mingo@elte.hu>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
