This allows increasing the rx queue size up to 1024: unlike with tx,
guests don't put in huge S/G lists into RX so the risk of running into
the max 1024 limitation due to some off-by-one seems small.

It's helpful for users like OVS-DPDK which don't do any buffering on the
host - 1K roughly matches 500 entries in tun + 256 in the current rx
queue, which seems to work reasonably well. We could probably make do
with ~750 entries but virtio spec limits us to powers of two.
It might be a good idea to specify an s/g size limit in a future
version.

It also might be possible to make the queue size smaller down the road, 64
seems like the minimal value which will still work (as guests seem to
assume a queue full of 1.5K buffers is enough to process the largest
incoming packet, which is ~64K).  No one actually asked for this, and
with virtio 1 guests can reduce ring size without need for host
configuration, so don't bother with this for now.

Cc: Cornelia Huck <cornelia.huck@de.ibm.com>
Cc: Jason Wang <jasowang@redhat.com>
Suggested-by: NPatrik Hermansson <phermansson@gmail.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>
Reviewed-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: NJason Wang <jasowang@redhat.com>

virtio, pc: fixes and features

beginning of guest error handling for virtio devices
amd iommu
pc compat fixes
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

# gpg: Signature made Fri 23 Sep 2016 23:02:09 BST
# gpg:                using RSA key 0x281F0DB8D28D5469
# gpg: Good signature from "Michael S. Tsirkin <mst@kernel.org>"
# gpg:                 aka "Michael S. Tsirkin <mst@redhat.com>"
# Primary key fingerprint: 0270 606B 6F3C DF3D 0B17  0970 C350 3912 AFBE 8E67
#      Subkey fingerprint: 5D09 FD08 71C8 F85B 94CA  8A0D 281F 0DB8 D28D 5469

* remotes/mst/tags/for_upstream:
  hw/i386: AMD IOMMU IVRS table
  hw/i386: Introduce AMD IOMMU
  hw/i386/trace-events: Add AMD IOMMU trace events
  hw/pci: Prepare for AMD IOMMU
  virtio: handle virtqueue_get_head() errors
  virtio: handle virtqueue_num_heads() errors
  virtio: handle virtqueue_read_next_desc() errors
  virtio: use unsigned int for virtqueue_get_avail_bytes() index
  virtio: handle virtqueue_get_avail_bytes() errors
  virtio: handle virtqueue_map_desc() errors
  virtio: migrate vdev->broken flag
  virtio: stop virtqueue processing if device is broken
  virtio: fix stray tab character
  target-i386: turn off CPU.l3-cache only for 2.7 and older machine types
  pc: clean up COMPAT macro chaining
  virtio: add check for descriptor's mapped address
  tests: add /vhost-user/flags-mismatch test
  tests: add a simple /vhost-user/multiqueue test
  tests: add /vhost-user/connect-fail test
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

Add IVRS table for AMD IOMMU. Generate IVRS or DMAR
depending on emulated IOMMU.
Signed-off-by: NDavid Kiarie <davidkiarie4@gmail.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

Add AMD IOMMU emulaton to Qemu in addition to Intel IOMMU.
The IOMMU does basic translation, error checking and has a
minimal IOTLB implementation. This IOMMU bypassed the need
for target aborts by responding with IOMMU_NONE access rights
and exempts the region 0xfee00000-0xfeefffff from translation
as it is the q35 interrupt region.

We advertise features that are not yet implemented to please
the Linux IOMMU driver.

IOTLB aims at implementing commands on real IOMMUs which is
essential for debugging and may not offer any performance
benefits
Signed-off-by: NDavid Kiarie <davidkiarie4@gmail.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

Signed-off-by: NDavid Kiarie <davidkiarie4@gmail.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

Introduce PCI macros from for use by AMD IOMMU
Signed-off-by: NDavid Kiarie <davidkiarie4@gmail.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

Stop processing the vring if virtqueue_get_head() fetches an
out-of-bounds head index.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>
Reviewed-by: NCornelia Huck <cornelia.huck@de.ibm.com>

If the avail ring index is bogus virtqueue_num_heads() must return
-EINVAL.

The only caller is virtqueue_get_avail_bytes().  Return saying no bytes
are available when virtqueue_num_heads() fails.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>
Reviewed-by: NCornelia Huck <cornelia.huck@de.ibm.com>

Stop processing the vring if an avail ring index is invalid.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>
Reviewed-by: NCornelia Huck <cornelia.huck@de.ibm.com>

The virtio code uses int, unsigned int, and uint16_t for virtqueue
indices.  The uint16_t is used for the low-level descriptor layout in
virtio_ring.h while code that isn't concerned with descriptor layout can
use unsigned int.

Use of int is problematic because it can result in signed/unsigned
comparison and incompatible int*/unsigned int* pointer types.

Make the virtqueue_get_avail_bytes() 'i' variable unsigned int.  This
eliminates the need to introduce casts and modify code further in the
patches that follow.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>
Reviewed-by: NCornelia Huck <cornelia.huck@de.ibm.com>

If the vring is invalid, tell the caller no bytes are available and mark
the device broken.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>
Reviewed-by: NCornelia Huck <cornelia.huck@de.ibm.com>

Errors can occur during virtqueue_pop(), especially in
virtqueue_map_desc().  In order to handle this we must unmap iov[]
before returning NULL.  The caller will consider the virtqueue empty and
the virtio_error() call will have marked the device broken.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

Send a subsection if the vdev->broken flag is set.  This allows live
migration of broken virtio devices.

The subsection is only sent if vdev->broken has been set.  In most cases
the flag will be clear and no subsection will be sent.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>
Reviewed-by: NCornelia Huck <cornelia.huck@de.ibm.com>

QEMU prints an error message and exits when the device enters an invalid
state.  Terminating the process is heavy-handed.  The guest may still be
able to function even if there is a bug in a virtio guest driver.

Moreover, exiting is a bug in nested virtualization where a nested guest
could DoS other nested guests by killing a pass-through virtio device.
I don't think this configuration is possible today but it is likely in
the future.

If the broken flag is set, do not process virtqueues or write back used
descriptors.  The broken flag can be cleared again by resetting the
device.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>
Reviewed-by: NCornelia Huck <cornelia.huck@de.ibm.com>

Fix a single occurrence of a tab character in a file that otherwise uses
spaces for indentation.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: NFam Zheng <famz@redhat.com>
Acked-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>
Acked-by: NCornelia Huck <cornelia.huck@de.ibm.com>

commit (14c985cf target-i386: present virtual L3 cache info for vcpus)
misplaced compat property putting it in new 2.8 machine type
which would effectively to disable feature until 2.9 is released.
Intent of commit probably should be to disable feature for 2.7
and older while allowing not yet released 2.8 to have feature
enabled by default.

Cc: qemu-stable@nongnu.org
Signed-off-by: NIgor Mammedov <imammedo@redhat.com>
Reviewed-by: NMarcel Apfelbaum <marcel@redhat.com>
Reviewed-by: NEduardo Habkost <ehabkost@redhat.com>

Since commit
 bacc344c ("machine: add properties to compat_props incrementaly")
there is no need to chain per machine type compat macro.

Clean up places where it was done anyway so it will be
consistent and won't confuse contributors during addtion
of new machine types.
Signed-off-by: NIgor Mammedov <imammedo@redhat.com>
Reviewed-by: NEduardo Habkost <ehabkost@redhat.com>

virtio back end uses set of buffers to facilitate I/O operations.
If its size is too large, 'cpu_physical_memory_map' could return
a null address. This would result in a null dereference while
un-mapping descriptors. Add check to avoid it.
Reported-by: NQinghao Tang <luodalongde@gmail.com>
Signed-off-by: NPrasad J Pandit <pjp@fedoraproject.org>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>
Reviewed-by: NLaszlo Ersek <lersek@redhat.com>

Check that qemu disconnects the backend that doesn't have the previously
acked features.
Signed-off-by: NMarc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

This test just checks that 2 virtio-net queues can be setup over
vhost-user and waits for them to be started.
Signed-off-by: NMarc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

Check early connection failure and resume.
Signed-off-by: NMarc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

Block layer patches

# gpg: Signature made Fri 23 Sep 2016 12:59:46 BST
# gpg:                using RSA key 0x7F09B272C88F2FD6
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>"
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74  56FE 7F09 B272 C88F 2FD6

* remotes/kevin/tags/for-upstream: (33 commits)
  block: Remove BB interface from blockdev-add/del
  qemu-iotests/141: Avoid blockdev-add with id
  block: Avoid printing NULL string in error messages
  qemu-iotests/139: Avoid blockdev-add with id
  qemu-iotests/124: Avoid blockdev-add with id
  qemu-iotests/118: Avoid blockdev-add with id
  qemu-iotests/117: Avoid blockdev-add with id
  qemu-iotests/087: Avoid blockdev-add with id
  qemu-iotests/081: Avoid blockdev-add with id
  qemu-iotests/071: Avoid blockdev-add with id
  qemu-iotests/067: Avoid blockdev-add with id
  qemu-iotests/041: Avoid blockdev-add with id
  qemu-iotests/118: Test media change with qdev name
  block: Accept device model name for block_set_io_throttle
  block: Accept device model name for blockdev-change-medium
  block: Accept device model name for eject
  block: Accept device model name for x-blockdev-remove-medium
  block: Accept device model name for x-blockdev-insert-medium
  block: Accept device model name for blockdev-open/close-tray
  qdev-monitor: Add blk_by_qdev_id()
  ...
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

MIPS patches 2016-09-23

Changes:
* 24KEc CPU definition
* SYNC instructions make use of tcg memory barrier ops
* various MIPS linux-user bug fixes

# gpg: Signature made Fri 23 Sep 2016 08:09:08 BST
# gpg:                using RSA key 0x52118E3C0B29DA6B
# gpg: Good signature from "Leon Alrae <leon.alrae@imgtec.com>"
# Primary key fingerprint: 8DD3 2F98 5495 9D66 35D4  4FC0 5211 8E3C 0B29 DA6B

* remotes/lalrae/tags/mips-20160923:
  linux-user: Add missing Mips syscalls items in strace.list
  linux-user: Add missing TARGET_EDQUOT error code for Mips
  linux-user: Fix certain argument alignment cases for Mips64
  linux-user: Fix structure target_semid64_ds definition for Mips
  linux-user: Fix structure target_flock definition for Mips
  linux-user: Fix TARGET_F_GETOWN definition for Mips
  linux-user: Fix TARGET_SIOCATMARK definition for Mips
  target-mips: generate fences
  target-mips: add 24KEc CPU definition
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

ppc patch queue 2016-09-23

This pull request supersedes ppc-for-2.8-20160922.  There was a clang
build error in that, and I've also added one extra patch in the new pull.

Included in this set of ppc and spapr patches are:
    * TCG implementations for more POWER9 instructions
    * Some preliminary XICS fixes in preparataion for the pnv machine type
    * A significant ADB (Macintosh kbd/mouse) cleanup
    * Some conversions to use trace instead of debug macros
    * Fixes to correctly handle global TLB flush synchronization in
      TCG.  This is already a bug, but it will have much more impact
      when we get MTTCG
    * Add more qtest testcases for Power
    * Some MAINTAINERS updates
    * Assorted bugfixes
    * Add the basics of NUMA associativity to the spapr PCI host bridge

This touches some test files and monitor.c which are technically
outside the ppc code, but coming through this tree because the changes
are primarily of interest to ppc.

# gpg: Signature made Fri 23 Sep 2016 08:14:47 BST
# gpg:                using RSA key 0x6C38CACA20D9B392
# gpg: Good signature from "David Gibson <david@gibson.dropbear.id.au>"
# gpg:                 aka "David Gibson (Red Hat) <dgibson@redhat.com>"
# gpg:                 aka "David Gibson (ozlabs.org) <dgibson@ozlabs.org>"
# gpg:                 aka "David Gibson (kernel.org) <dwg@kernel.org>"
# Primary key fingerprint: 75F4 6586 AE61 A66C C44E  87DC 6C38 CACA 20D9 B392

* remotes/dgibson/tags/ppc-for-2.8-20160923: (45 commits)
  spapr_pci: Add numa node id
  monitor: fix crash for platforms without a CPU 0
  linux-user: ppc64: fix ARCH_206 bit in AT_HWCAP
  ppc/kvm: Mark 64kB page size support as disabled if not available
  ppc/xics: An ICS with offset 0 is assumed to be uninitialized
  ppc/xics: account correct irq status
  Enable H_CLEAR_MOD and H_CLEAR_REF hypercalls on KVM/PPC64.
  target-ppc: tlbie/tlbivax should have global effect
  target-ppc: add flag in check_tlb_flush()
  target-ppc: add TLB_NEED_LOCAL_FLUSH flag
  spapr: Introduce sPAPRCPUCoreClass
  target-ppc: implement darn instruction
  target-ppc: add stxsi[bh]x instruction
  target-ppc: add lxsi[bw]zx instruction
  target-ppc: add xxspltib instruction
  target-ppc: consolidate store conditional
  target-ppc: move out stqcx impementation
  target-ppc: consolidate load with reservation
  target-ppc: convert st[16,32,64]r to use new macro
  target-ppc: convert st64 to use new macro
  ...
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

# gpg: Signature made Fri 23 Sep 2016 05:58:28 BST
# gpg:                using RSA key 0xCA35624C6A9171C6
# gpg: Good signature from "Fam Zheng <famz@redhat.com>"
# gpg: WARNING: This key is not certified with a trusted signature!
# gpg:          There is no indication that the signature belongs to the owner.
# Primary key fingerprint: 5003 7CB7 9706 0F76 F021  AD56 CA35 624C 6A91 71C6

* remotes/famz/tags/various-pull-request: (23 commits)
  docker: exec $CMD
  docker: Terminate instances at SIGTERM and SIGHUP
  docker: Support showing environment information
  docker: Print used options before doing configure
  docker: Flatten default target list in test-quick
  docker: Update fedora image to latest
  docker: Generate /packages.txt in ubuntu image
  docker: Generate /packages.txt in fedora image
  docker: Generate /packages.txt in centos6 image
  tests: Ignore test-uuid
  Add UUID files to MAINTAINERS
  tests: Add uuid tests
  uuid: Tighten uuid parse
  vl: Switch qemu_uuid to QemuUUID
  configure: Remove detection code for UUID
  tests: No longer dependent on CONFIG_UUID
  crypto: Switch to QEMU UUID API
  vpc: Use QEMU UUID API
  vdi: Use QEMU UUID API
  vhdx: Use QEMU UUID API
  ...
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

# Conflicts:
#	tests/Makefile.include

With this patch, blockdev-add always works on a node level, i.e. it
creates a BDS, but no BB. Consequently, x-blockdev-del doesn't need the
'device' option any more, but 'node-name' becomes mandatory.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

We want to remove the 'id' option for blockdev-add. This removes one
user of the option and makes it use only node names.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

Even for nodes that have a BlockBackend attached, bdrv_get_parent_name()
can return NULL if the BB is anonymous (e.g. it belongs to a block job
or a device that was created with a drive=<node-name> option).

Remove the information from the error message. The user probably knows
already why the node is still in use.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

We want to remove the 'id' option for blockdev-add. This removes one
user of the option and makes it use only node names.

Some test cases that used to work with an unattached BlockBackend are
removed, either because they don't make sense with an attached device or
because the equivalent test case with an attached device already exists.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

We want to remove the 'id' option for blockdev-add. This removes one
user of the option and makes it use only node names.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

We want to remove the 'id' option for blockdev-add. This removes one
user of the option and makes it use only node names.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

We want to remove the 'id' option for blockdev-add. This removes one
user of the option and makes it use only node names.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

We want to remove the 'id' option for blockdev-add. This removes one
user of the option and makes it use only node names.

The test cases that test conflicts between the 'id' option to
blockdev-add and existing block devices or the 'node-name' of the same
command can be removed because it won't be possible to specify this at
the end of the series.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

We want to remove the 'id' option for blockdev-add. This removes one
user of the option and makes it use only node names.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

We want to remove the 'id' option for blockdev-add. This removes one
user of the option and makes it use only node names.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

We want to remove the 'id' option for blockdev-add. This removes one
user of the option and makes it use only node names.

In order to keep the test meaningful, some instances of query-block that
want to check whether the node still exists and would now turn up empty
must be converted to query-named-block-nodes (which also return the
protocol level node, but that shouldn't hurt).
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

We want to remove the 'id' option for blockdev-add. This removes one
user of the option and makes it use only node names.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

We just added the option to use qdev device names in all device related
block QMP commands. This patch converts some of the test cases in 118 to
use qdev device names instead of BlockBackend names to cover the new
way. It converts cases for each of the media change commands, but only
for CD-ROM and not everywhere, so that the old way is still tested, too.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

In order to remove the need for BlockBackend names in the external API,
we want to allow qdev device names in all device related commands.

This converts block_set_io_throttle to accept a qdev device name.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

In order to remove the need for BlockBackend names in the external API,
we want to allow qdev device names in all device related commands.

This converts blockdev-change-medium to accept a qdev device name.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>