Always check it immediately after calling bdrv_acct_done, and
always do a "goto done" in case the "done" label has to free
some memory---as is the case for scsi_unmap_complete in the
previous patch.

This patch could fix problems that happen when a request is
split into multiple parts, and one of them is canceled.  Then
the next part is fired, but the HBA's cancellation callbacks have
fired already.  Whether this happens or not, depends on how the
block/ driver implements AIO cancellation.  It it does a simple
bdrv_drain_all() or similar, then it will not have a problem.
If it only cancels the given AIOCB, this scenario could happen.

Cc: qemu-stable@nongnu.org
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Canceled requests should never be completed, and doing that could cause
accesses to a NULL hba_private field.

Cc: qemu-stable@nongnu.org
Reported-by: NStefan Priebe <s.priebe@profihost.ag>
Tested-by: NStefan Priebe <s.priebe@profihost.ag>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Cc: qemu-stable@nongnu.org
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Due to library conflicts, Fedora will have to put libiscsi in
/usr/lib/iscsi.  Simplify configuration by using a pkg-config
file.  The Fedora package will distribute one, and the patch
to add it has been sent to upstream libiscsi as well.

Cc: qemu-stable@nongnu.org
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

this patch adds iscsi_truncate which effectively allows for
online resizing of iscsi volumes. for this to work you have
to resize the volume on your storage and then call
block_resize command in qemu which will issue a
readcapacity16 to update the capacity.

v4:
  - factor out complete readcapacity logic into a separate function
  - handle capacity change check condition in readcapacity function
    (this happens if the block_resize cmd is the first iscsi task
    executed after a resize on the storage)

v3:
  - remove switch statement in iscsi_open
  - create separate patch for brdv_drain_all() in bdrv_truncate()

v2:
  - add a general bdrv_drain_all() before bdrv_truncate() to avoid
    in-flight AIOs while the device is truncated
  - since no AIOs are in flight we can use a sync libiscsi call
    to re-read the capacity
  - factor out the readcapacity16 logic as it is redundant
    to iscsi_open() and iscsi_truncate().
Signed-off-by: NPeter Lieven <pl@kamp.de>
[allow any type of unit attention check condition in iscsi_readcapacity_sync(),
 as in Message-ID: <51263A2A.6070304@dlhnet.de> - Paolo]
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

the storage might return a check condition status for various reasons.
(e.g. bus reset, capacity change, thin-provisioning info etc.)

currently all these informative status responses lead to an I/O error
which is populated to the guest. this patch introduces a retry mechanism
to avoid this.
Signed-off-by: NPeter Lieven <pl@kamp.de>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

* 'target-arm.next' of git://git.linaro.org/people/pmaydell/qemu-arm:
  MAINTAINERS: add entry for ARM KVM guest cores
  configure: Enable KVM on ARM
  hw/kvm/arm_gic: Implement support for KVM in-kernel ARM GIC
  target-arm: Use MemoryListener to identify GIC base address for KVM
  hw/arm_gic: Convert ARM GIC classes to use init/realize
  hw/arm_gic: Add presave/postload hooks
  ARM KVM: save and load VFP registers from kernel
  ARM: KVM: Add support for KVM on ARM architecture
  target-arm: Drop CPUARMState* argument from bank_number()
  linux-headers: resync from mainline to add ARM KVM headers
  oslib-posix: Align to permit transparent hugepages on ARM Linux
  target-arm: Don't decode RFE or SRS on M profile cores
  target-arm: Factor out handling of SRS instruction

N32 is a 64-bit cpu with a 32-bit address space.  We have
existing cpp defines for this situation, but weren't using them.

This does mean that the linux-user/mipsn32 directory must be
merged with the linux-user/mips64 directory, and differences
must be resolved via ifdefs.
Signed-off-by: NRichard Henderson <rth@twiddle.net>
Signed-off-by: NAurelien Jarno <aurelien@aurel32.net>

Signed-off-by: NRichard Henderson <rth@twiddle.net>
Signed-off-by: NAurelien Jarno <aurelien@aurel32.net>

Signed-off-by: NRichard Henderson <rth@twiddle.net>
Signed-off-by: NAurelien Jarno <aurelien@aurel32.net>

Signed-off-by: NRichard Henderson <rth@twiddle.net>
Signed-off-by: NAurelien Jarno <aurelien@aurel32.net>

At this point we can enable compilation, though things
still don't work.
Signed-off-by: NRichard Henderson <rth@twiddle.net>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NAurelien Jarno <aurelien@aurel32.net>

Signed-off-by: NRichard Henderson <rth@twiddle.net>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NAurelien Jarno <aurelien@aurel32.net>

Deleting these first makes the next patch much easier to read.
This doesn't cause any sort of compilation failure because we
have not yet enabled n32/n64 compilation.  This is dead code.
Signed-off-by: NRichard Henderson <rth@twiddle.net>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NAurelien Jarno <aurelien@aurel32.net>

Add an entry indicating maintainer status for the ARM KVM code.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

Enable KVM on ARM hosts, now that all the necessary components
for it exist.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>

Implement support for using the KVM in-kernel GIC for ARM.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NAndreas Färber <afaerber@suse.de>

When using an in-kernel GIC with KVM, we need to tell the kernel where
the GIC's memory mapped registers live. Do this by registering a
MemoryListener which tracks where the board model maps the A15's
private peripherals, so we can finish the GIC initialisation
when the GIC is actually mapped.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>

Convert the ARM GIC classes to use init/realize rather than
SysBusDevice::init. (We have to do them all in one patch to
avoid unconverted subclasses calling a nonexistent SysBusDevice
init function in the base class and crashing.)
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NAndreas Färber <afaerber@suse.de>

Add presave/postload hooks to the ARM GIC common base class.
These will be used by the KVM in-kernel GIC subclass to sync
state between kernel and userspace when migrating.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NAndreas Färber <afaerber@suse.de>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>

Add support for saving and restoring VFP register state from the
kernel. This includes a check that the KVM-created CPU has full
VFP support (as the TCG Cortex-A15 model always does), since for
the moment ARM QEMU doesn't have any way to tweak optional features
on created CPUs.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>

Add basic support for KVM on ARM architecture.
Signed-off-by: NChristoffer Dall <cdall@cs.columbia.edu>
[PMM: Minor tweaks and code cleanup, switch to ONE_REG]
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>

Drop the CPUARMState* argument from bank_number(), since we only
use it for passing to cpu_abort(). Use hw_error() instead.
This avoids propagating further interfaces using env pointers.

In the long term this function's callers need auditing to fix
problems where badly behaved guests can pass invalid bank numbers.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NAndreas Färber <afaerber@suse.de>

Resync QEMU's copy of the Linux kernel headers from
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
commit 2ef14f4. This adds the ARM KVM headers, since ARM KVM
support has just hit mainline via Russell's ARM tree.

This is not a pure sync -- I have removed by hand some changes
that would have reverted updates for s390x and ppc which have not
yet hit mainline.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

ARM Linux (like x86-64 Linux) can use transparent hugepages for
KVM if memory blocks are 2MiB aligned; set QEMU_VMALLOC_ALIGN
accordingly.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>

M profile cores do not have the RFE or SRS instructions, so
correctly UNDEF these insn patterns on those cores.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

Factor out the handling of the SRS instruction rather than
duplicating it between the Thumb and ARM decoders. This in
passing fixes two bugs in the Thumb decoder's SRS handling
which didn't exist in the ARM decoder:
 * (LP:1079080) storing CPSR rather than SPSR (fixed in the
   ARM decoder in commit c67b6b71 in 2009)
 * failing to free the 'addr' TCG temp in the writeback case
Reported-by: NCesson Vincent <vcesson@stmi.com>
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

Add accumulator arguments to gen_HILO and gen_muldiv, rather than
extracting the accumulator directly from ctx->opcode.  The extraction
was only right for the standard encoding: MIPS16 doesn't have access
to the DSP registers, while microMIPS encodes the accumulator register
in a different field (bits 14 and 15).

Passing the accumulator register is probably an over-generalisation
for division and 64-bit multiplication, which never access anything
other than HI and LO, and which always pass 0 as the new argument.
Separating them felt a bit fussy though.
Signed-off-by: NRichard Sandiford <rdsandiford@googlemail.com>
Signed-off-by: NAurelien Jarno <aurelien@aurel32.net>

GCC and GAS are capable of generating traps or breaks to check for
division by zero.  Additionally, GAS is capable of generating traps
or breaks to check for overflow on certain division and multiplication
operations.  The Linux kernel translates these traps and breaks into
signals.  This patch implements the corresponding feature in QEMU.
Signed-off-by: NMeador Inge <meadori@codesourcery.com>
Signed-off-by: NAurelien Jarno <aurelien@aurel32.net>

The previous implementation incorrectly used same macro to detect overflow
for addition and subtraction. This patch makes distinction between these
two, and creates separate macros. The affected routines are changed
accordingly.

This change also includes additions to the existing tests for SUBQ_S_PH and
SUBQ_S_W that would trigger the fixed issue, and it removes dead code from
the test file. The last test case in subq_s_w.c is a bug found/reported/
isolated by Klaus Peichl from Dolby.
Signed-off-by: NPetar Jovanovic <petar.jovanovic@imgtec.com>
Signed-off-by: NAurelien Jarno <aurelien@aurel32.net>

# By MORITA Kazutaka (5) and others
# Via Stefan Hajnoczi
* stefanha/block:
  block: for HMP commit() operations on 'all', skip non-COW drives
  sheepdog: add support for connecting to unix domain socket
  sheepdog: use inet_connect to simplify connect code
  sheepdog: accept URIs
  move socket_set_nodelay to osdep.c
  slirp/tcp_subr.c: fix coding style in tcp_connect
  dataplane: remove EventPoll in favor of AioContext
  virtio-blk: fix unplug + virsh reboot
  ide/macio: Fix macio DMA initialisation.

virtio,vhost,pci,e1000

Mostly bugfixes, but also some ICH work by Laszlo.
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

# gpg: Signature made Thu 28 Feb 2013 07:13:56 AM CST using RSA key ID D28D5469
# gpg: Can't check signature: public key not found

# By Michael S. Tsirkin (2) and others
# Via Michael S. Tsirkin
* mst/tags/for_anthony:
  Set virtio-serial device to have a default of 2 MSI vectors.
  ICH9 LPC: Reset Control Register, basic implementation
  Fix guest OS hang when 64bit PCI bar present
  e1000: unbreak the guest network migration to 1.3
  vhost: memory sync fixes

# By Gerd Hoffmann
# Via Gerd Hoffmann
* kraxel/seabios-1.7.2.1:
  update seabios to 1.7.2.1

# By Jason Wang (2) and others
# Via Stefan Hajnoczi
* stefanha/net:
  qmp: netdev_add is like -netdev, not -net, fix documentation
  doc: document -netdev hubport
  net: reduce the unnecessary memory allocation of multiqueue
  tap: set IFF_ONE_QUEUE per default
  tap: forbid creating multiqueue tap when hub is used
  net: fix unbounded NetQueue
  net: fix qemu_flush_queued_packets() in presence of a hub

During a commit of 'all' using the HMP non-live commit, the operation
is aborted and returns error on the first error enountered.  When
non-COW drives are in use (e.g. ejected floppy, cdrom, or drives without
a backing parent), that means a commit all will return an error of either
-ENOMEDIUM or -ENOTSUP.  This is not desirable, so for the 'all' commit
case, only attempt the commit if both bs->drv and bs->backing_hd are
present.

More succinctly: 'commit all' now means a commit on all COW drives.

This means an individual commit to a specific non-COW drive will still
return the appropriate error (-ENOMEDIUM if eject / not present, -ENOTSUP
if no backing file).
Reported-by: NJan Kiszka <jan.kiszka@web.de>
Signed-off-by: NJeff Cody <jcody@redhat.com>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NKevin Wolf <kwolf@redhat.com>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

This patch adds support for a unix domain socket for a connection
between qemu and local sheepdog server.  You can use the unix domain
socket with the following syntax:

 $ qemu sheepdog+unix:///<vdiname>?socket=<socket path>[#snapid]
Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

This uses the form "<host>:<port>" for the representation of the
sheepdog server to use inet_connect.
Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

The URI syntax is consistent with the NBD and Gluster syntax.  The
syntax is

  sheepdog[+tcp]://[host:port]/vdiname[#snapid|#tag]
Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

Fix coding style in tcp_connect before the next patch.
Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>