When we go into out-of-memory and fail to allocate skbuffs to
refill the receive ring with, rxq_process can end up running into
a receive ring entry that is marked as host-owned but doesn't have
an associated skbuff. If this happens, we must break out of the
rx processing loop instead of trying to process the descriptor.
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
Acked-by: Nicolas Pitre <nico@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
> channel_index loops up to IPW_SCAN_CHANNELS, but is used after being
> incremented. This might be able to access 1 past the end of the array
Reported-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
I noticed oprofile memleaked in linux-2.6 current tree,
and tracked this ring-buffer leak.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
LKML-Reference: <4A7C06B9.2090302@gmail.com>
Cc: stable@kernel.org
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
we should align the GTT after VRAM no matter what, as we can
come back from resume and put in a different place and bad things happen.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Check whether index is within bounds before testing the element.
Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Currently, the machine_flags are stored late in the startup
initialization which results in failing machine type checks
(e.g. for MACHINE_IS_VM).
To allow these checks, store the machine flags in the lowcore
when the machine type has been detected.
Moving the machine_flags to the lowcore has been introduced with
git commit 25097bf153
Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Roland Dreier found that a section that contained only a weak
function in one of the staging drivers and this caused
recordmcount.pl to spit out a warning and fail.
Although it is strange that a driver would have a weak function, and
this function only be used in one place, it should not be something
to make recordmcount.pl fail.
This patch fixes the issue in a simple manner: if only weak
functions exist in a section, then that section will not be
recorded.
Reported-by: Roland Dreier <rdreier@cisco.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
irda_getname() can leak kernel memory to user.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
atalk_getname() can leak 8 bytes of kernel memory to user
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
econet_getname() can leak kernel memory to user.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
rose_getname() can leak kernel memory to user.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Brice Goglin reported this crash with per task precise stats:
> I finally managed to test the threaded perfcounter statistics (thanks a
> lot for implementing it). I am running 2.6.31-rc5 (with the AMD
> magny-cours patches but I don't think they matter here). I am trying to
> measure local/remote memory accesses per thread during the well-known
> stream benchmark. It's compiled with OpenMP using 16 threads on a
> quad-socket quad-core barcelona machine.
>
> Command line is:
> /mnt/scratch/bgoglin/cpunode/linux-2.6.31/tools/perf/perf record -f -s
> -e r1000001e0 -e r1000002e0 -e r1000004e0 -e r1000008e0 ./stream
>
> It seems to work fine with a single -e <counter> on the command line
> while it crashes when there are at least 2 of them.
> It seems to work fine without -s as well.
A silly copy-paste resulted in a messed up iteration which would
cause the OOPS.
Reported-by: Brice Goglin <Brice.Goglin@inria.fr>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Tested-by: Brice Goglin <Brice.Goglin@inria.fr>
LKML-Reference: <1249574786.32113.550.camel@twins>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Adds autodetection for libelf as well, and simplifies the
libbfd code. Furthermore, fail make with an error when libelf
is not found and warn about the lack of libbfd.
Also provide an option to build a 32bit version even though you
might be running a 64bit kernel.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
In some cases distros have binaries and debuginfo in weird places:
[root@doppio tuna]# ls -la /usr/lib64/{xulrunner-1.9.1/xulrunner-stub,firefox-3.5.2/firefox}
-rwxr-xr-x 1 root root 90024 2009-08-03 19:45 /usr/lib64/firefox-3.5.2/firefox
-rwxr-xr-x 1 root root 90024 2009-08-03 18:23 /usr/lib64/xulrunner-1.9.1/xulrunner-stub
[root@doppio tuna]# sha1sum /usr/lib64/{xulrunner-1.9.1/xulrunner-stub,firefox-3.5.2/firefox}
19a858077d263d5de22c9c5da250d3e4396ae739 /usr/lib64/xulrunner-1.9.1/xulrunner-stub
19a858077d263d5de22c9c5da250d3e4396ae739 /usr/lib64/firefox-3.5.2/firefox
[root@doppio tuna]# rpm -qf /usr/lib64/{xulrunner-1.9.1/xulrunner-stub,firefox-3.5.2/firefox}
xulrunner-1.9.1.2-1.fc11.x86_64
firefox-3.5.2-2.fc11.x86_64
[root@doppio tuna]# ls -la /usr/lib/debug/{usr/lib64/xulrunner-1.9.1/xulrunner-stub,usr/lib64/firefox-3.5.2/firefox}.debug
ls: cannot access /usr/lib/debug/usr/lib64/firefox-3.5.2/firefox.debug: No such file or directory
-rwxr-xr-x 1 root root 403608 2009-08-03 18:22 /usr/lib/debug/usr/lib64/xulrunner-1.9.1/xulrunner-stub.debug
Seemingly we don't have a .symtab when we actually can find it
if we use the .note.gnu.build-id ELF section put in place by
some distros. Use it and find the symbols we need.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When calling rb_buffer_peek() from ring_buffer_consume() and a
padding event is returned, the function rb_advance_reader() is
called twice. This may lead to missing samples or under high
workloads to the warning below. This patch fixes this. If a padding
event is returned by rb_buffer_peek() it will be consumed by the
calling function now.
Also, I simplified some code in ring_buffer_consume().
------------[ cut here ]------------
WARNING: at /dev/shm/.source/linux/kernel/trace/ring_buffer.c:2289 rb_advance_reader+0x2e/0xc5()
Hardware name: Anaheim
Modules linked in:
Pid: 29, comm: events/2 Tainted: G W 2.6.31-rc3-oprofile-x86_64-standard-00059-g5050dc2 #1
Call Trace:
[<ffffffff8106776f>] ? rb_advance_reader+0x2e/0xc5
[<ffffffff81039ffe>] warn_slowpath_common+0x77/0x8f
[<ffffffff8103a025>] warn_slowpath_null+0xf/0x11
[<ffffffff8106776f>] rb_advance_reader+0x2e/0xc5
[<ffffffff81068bda>] ring_buffer_consume+0xa0/0xd2
[<ffffffff81326933>] op_cpu_buffer_read_entry+0x21/0x9e
[<ffffffff810be3af>] ? __find_get_block+0x4b/0x165
[<ffffffff8132749b>] sync_buffer+0xa5/0x401
[<ffffffff810be3af>] ? __find_get_block+0x4b/0x165
[<ffffffff81326c1b>] ? wq_sync_buffer+0x0/0x78
[<ffffffff81326c76>] wq_sync_buffer+0x5b/0x78
[<ffffffff8104aa30>] worker_thread+0x113/0x1ac
[<ffffffff8104dd95>] ? autoremove_wake_function+0x0/0x38
[<ffffffff8104a91d>] ? worker_thread+0x0/0x1ac
[<ffffffff8104dc9a>] kthread+0x88/0x92
[<ffffffff8100bdba>] child_rip+0xa/0x20
[<ffffffff8104dc12>] ? kthread+0x0/0x92
[<ffffffff8100bdb0>] ? child_rip+0x0/0x20
---[ end trace f561c0a58fcc89bd ]---
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: <stable@kernel.org>
Signed-off-by: Robert Richter <robert.richter@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
If the current CPU doesn't support performance counters,
cur_cpu_spec->oprofile_cpu_type can be NULL. The current
perf_counter modules don't test for that case and would thus
crash at boot time.
Bug reported by David Woodhouse.
Reported-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Paul Mackerras <paulus@samba.org>
LKML-Reference: <19066.48028.446975.501454@cargo.ozlabs.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Two defects work together result in KVM device passthrough randomly can't
work:
1. iommu_snooping is not initialized to zero when vm_iommu_init() called.
So it is possible to get a random value.
2. One line added by commit 2c2e2c38("IOMMU Identity Mapping Support")
change the code path, let it bypass domain_update_iommu_cap(), as well as
missing the increment of domain iommu reference count.
The latter is also likely to cause a leak of domains on repeated VMM
assignment and deassignment.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Otherwise the host can spend too long traversing an rmap chain, which
happens under a spinlock.
Cc: stable@kernel.org
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Provide dummt get/setsockopt implementations to stop these
syscalls from oopsing on our sockets.
Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
fix two errors in ioctl processing:
1) if the ioctl isn't supported one should return -ENOIOCTLCMD
2) don't call ndo_do_ioctl if the device doesn't provide it
Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Remove assumption on the shift and size of rows/columns form
matrix_keypad driver.
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
The Prestigio 157, an old no-name clone laptop uses input keys very
similar to the Wistron 1557/MS2141 with the addition of BIOS-controlled
wireless radio frequency kill switch.
This patch adds support for the RF kill switch control and adds manual
identification of the model.
The Prestigio does not expose any recognisable identity via dmidecode
and so requires manual selection at module init using
force=1 keymap=prestigio
Signed-off-by: TJ <ubuntu@tjworld.net>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
GTT object can either be cached,uncached or wc just let core ttm
pick the best mode according to how the bo driver and GTT memory
type was initialized.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Not all tracepoints are created equal, in specific the ftrace
tracepoints are created with TRACE_EVENT_FORMAT() which does
not generate the needed bits to tie them into perf counters.
For those events, don't create the 'id' file and fail
->profile_enable when their ID is specified through other
means.
Reported-by: Chris Mason <chris.mason@oracle.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Steven Rostedt <rostedt@goodmis.org>
LKML-Reference: <1249497664.5890.4.camel@laptop>
[ v2: fix build error in the !CONFIG_EVENT_PROFILE case ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
In the event of a lock steal or owner died,
rt_mutex_start_proxy_lock() will give the rt_mutex to the
waiting task, but it fails to release the wait_lock. This leads
to subsequent deadlocks when other tasks try to acquire the
rt_mutex.
I also removed a few extra blank lines that really spaced this
routine out. I must have been high on the \n when I wrote this
originally...
Signed-off-by: Darren Hart <dvhltc@us.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Dinakar Guniguntala <dino@in.ibm.com>
Cc: John Stultz <johnstul@linux.vnet.ibm.com>
LKML-Reference: <4A79D7F1.4000405@us.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
We found this old card which was not supported, and physically
looks similar to the other 3C905B we have (9055).
After adding the IDs it seems to work fine (MII report, dhcp, scp, ...)
Acked-by: Steffen Klassert <klassert@mathematik.tu-chemnitz.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
On Wed, 5 Aug 2009, Ingo Molnar wrote:
> * Dave Airlie <airlied@gmail.com> wrote:
>
> > Hey,
> >
> > So I spent 3-4 hrs today (I'm stupid yes) tracking down a .o
> > breakage by blaming rawhide gcc/binutils as I was using make
> > V=1and seeing only the compiler chain running,
>
> Hm, is this that powerpc related build bug you just reported?
Well we tracked it down and it is powerpc64 specific.
Seems that in drivers/hwmon/lm93.c there's a function called:
LM93_IN_FROM_REG()
But PPC64 has function descriptors and the real function names (the ones
you see in objdump) start with a '.'. Thus this in objdump you have:
Disassembly of section .text:
0000000000000000 <.LM93_IN_FROM_REG>:
0: 7c 08 02 a6 mflr r0
4: fb 81 ff e0 std r28,-32(r1)
The function name used is .LM93_IN_FROM_REG. But gcc considers symbols
that start with ".L" as a special symbol that is used inside the assembly
stage.
The nm passed into recordmcount uses the --synthetic option which shows
the ".L" symbols (my runs outside of the build did not include the
--synthetic option, so my older patch worked). We see the function as a
local.
Now to capture all the locations that use "mcount" we need to have a
reference to link into the object file a list of mcount callers. We need a
reference that will not disappear. We try to use a global function and if
that does not work, we use a local function as a reference. But to relink
the section back into the object, we need to make it global. In this case,
we run objcopy using --globalize-symbol and --localize-symbol to convert
the symbol into a global symbol, link the mcount list, then convert it
back to a local symbol.
This works great except for this case. .L* symbols can not be converted
into a global symbol, and the mcount section referencing it will remain
unresolved.
Reported-by: Dave Airlie <airlied@gmail.com>
LKML-Reference: <alpine.DEB.2.00.0908052011590.5010@gandalf.stny.rr.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The commit:
commit e0fdace10e
Author: David Miller <davem@davemloft.net>
Date: Fri Aug 1 01:11:22 2008 -0700
debug_locks: set oops_in_progress if we will log messages.
Otherwise lock debugging messages on runqueue locks can deadlock the
system due to the wakeups performed by printk().
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Will permanently set oops_in_progress on any lockdep failure.
When this triggers it will cause any read from the ring buffer to
permanently disable the ring buffer (not to mention no locking of
printk).
This patch removes the check. It keeps the print in NMI which makes
sense. This is probably OK, since the ring buffer should not cause
something to set oops_in_progress anyway.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The function ring_buffer_discard_commit inversed the code path
of the result of try_to_discard. It should skip incrementing the
entry counter if try_to_discard succeeded. But instead, it increments
the entry conder if it succeeded to discard, and does not increment
it if it fails.
The result of this bug is that filtering will make the stat counters
incorrect.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
<linux/clk.h> should be included to get the base API prototypes.
This fixes the following sparse warnings:
arch/arm/common/clkdev.c:65:12:
warning: symbol 'clk_get_sys' was not declared. Should it be static?
arch/arm/common/clkdev.c:79:12:
warning: symbol 'clk_get' was not declared. Should it be static?
arch/arm/common/clkdev.c:87:6:
warning: symbol 'clk_put' was not declared. Should it be static?
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
preserve_crunch_context() calls __copy_to_user() which expects the
destination address to be in __user space. setup_sigframe() properly
passes the destination address.
restore_crunch_context() calls __copy_from_user() which expects the
source address to be in __user space. restore_sigframe() properly
passes the source address.
This fixes {preserve/restore}_crunch_context() to accept the
address as __user space and resolves the following sparse warnings:
arch/arm/kernel/signal.c:146:31:
warning: incorrect type in argument 1 (different address spaces)
expected void [noderef] <asn:1>*to
got struct crunch_sigframe *frame
arch/arm/kernel/signal.c:156:38:
warning: incorrect type in argument 2 (different address spaces)
expected void const [noderef] <asn:1>*from
got struct crunch_sigframe *frame
arch/arm/kernel/signal.c:250:48:
warning: incorrect type in argument 1 (different address spaces)
expected struct crunch_sigframe *frame
got struct crunch_sigframe [noderef] <asn:1>*<noident>
arch/arm/kernel/signal.c:365:49:
warning: incorrect type in argument 1 (different address spaces)
expected struct crunch_sigframe *frame
got struct crunch_sigframe [noderef] <asn:1>*<noident>
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Stop referencing CLOCK_TICK_RATE in the KS8695 drivers, rather refer
to a KS8695_CLOCK_RATE.
Issue pointed out by Russell King on arm-linux-kernel mailing list.
Signed-off-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
net/phonet/pn_dev.c: In function `phonet_device_get':
net/phonet/pn_dev.c:99: warning: 'dev' might be used uninitialized in this function
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
dev_priv->saveSWF1 is a 16 element array, but this reads up to index 22,
and restored values from the wrong registers.
Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
percpu counter dccp_orphan_count is init in dccp_init() by
percpu_counter_init() while dccp module is loaded, but the
destroy of it is missing while dccp module is unloaded. We
can get the kernel WARNING about this. Reproduct by the
following commands:
$ modprobe dccp
$ rmmod dccp
$ modprobe dccp
WARNING: at lib/list_debug.c:26 __list_add+0x27/0x5c()
Hardware name: VMware Virtual Platform
list_add corruption. next->prev should be prev (c080c0c4), but was (null). (next
=ca7188cc).
Modules linked in: dccp(+) nfsd lockd nfs_acl auth_rpcgss exportfs sunrpc
Pid: 1956, comm: modprobe Not tainted 2.6.31-rc5 #55
Call Trace:
[<c042f8fa>] warn_slowpath_common+0x6a/0x81
[<c053a6cb>] ? __list_add+0x27/0x5c
[<c042f94f>] warn_slowpath_fmt+0x29/0x2c
[<c053a6cb>] __list_add+0x27/0x5c
[<c053c9b3>] __percpu_counter_init+0x4d/0x5d
[<ca9c90c7>] dccp_init+0x19/0x2ed [dccp]
[<c0401141>] do_one_initcall+0x4f/0x111
[<ca9c90ae>] ? dccp_init+0x0/0x2ed [dccp]
[<c06971b5>] ? notifier_call_chain+0x26/0x48
[<c0444943>] ? __blocking_notifier_call_chain+0x45/0x51
[<c04516f7>] sys_init_module+0xac/0x1bd
[<c04028e4>] sysenter_do_call+0x12/0x22
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix dma mask calculation that caps at 63-bit addressing even
when firmware advertises full 64-bit support.
Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since the C++ demangling isn't needed for everybody and
bfd/iberty aren't widely/easily available on all machines, make
it optional.
It also allows you to forcefully disable demangling by using
NO_DEMANGLE=1 and otherwise tries to detect libbfd/libiberty
combinations that result in a compiling demangler.
Reported-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
LKML-Reference: <20090801082048.GX12579@kernel.dk>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Eliminates this compiler warning:
arch/powerpc/kvm/../../../virt/kvm/kvm_main.c:1178: error: integer overflow in expression
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Avi Kivity <avi@redhat.com>
If you're doing performance testing, you're interested in the
symbols anyway so lets make "--sort comm,dso,symbol" the
default sort option.
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: acme@redhat.com
LKML-Reference: <1249467921-10450-1-git-send-email-penberg@cs.helsinki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
kvm_notify_acked_irq does not check irq type, so that it sometimes
interprets msi vector as irq. As a result, ack notifiers are not
called, which typially hangs the guest. The fix is to track and
check irq type.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
There are two waitqueues in kvm for wait handling:
vcpu->wq for virt/kvm/kvm_main.c and
vpcu->arch.local_int.wq for the s390 specific wait code.
the wait handling in kvm_s390_handle_wait was broken by using different
wait_queues for add_wait queue and remove_wait_queue.
There are two options to fix the problem:
o move all the s390 specific code to vcpu->wq and remove
vcpu->arch.local_int.wq
o move all the s390 specific code to vcpu->arch.local_int.wq
This patch chooses the 2nd variant for two reasons:
o s390 does not use kvm_vcpu_block but implements its own enabled wait
handling.
Having a separate wait_queue make it clear, that our wait mechanism is
different
o the patch is much smaller
Report-by: Julia Lawall <julia@diku.dk>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
We have to disable preemption and IRQs on every exit from
handle_invalid_guest_state, otherwise we generate at least a
preempt_disable imbalance.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
