iotests: Fix nonportable use of od --endian

Tests 261 and 272 fail on RHEL 7 with coreutils 8.22, since od
--endian was not added until coreutils 8.23.  Fix this by manually
constructing the final value one byte at a time.

Fixes: fc8ba423
Reported-by: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20200226125424.481840-1-eblake@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>

block/curl: HTTP header field names are case insensitive

RFC 7230 section 3.2 indicates that HTTP header field names are case
insensitive.

Signed-off-by: David Edmondson <david.edmondson@oracle.com>
Message-Id: <20200224101310.101169-3-david.edmondson@oracle.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>

block/curl: HTTP header fields allow whitespace around values

RFC 7230 section 3.2 indicates that whitespace is permitted between
the field name and field value and after the field value.

Signed-off-by: David Edmondson <david.edmondson@oracle.com>
Message-Id: <20200224101310.101169-2-david.edmondson@oracle.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>

iotests: add 288 luks qemu-img measure test

This test exercises the block/crypto.c "luks" block driver
.bdrv_measure() code.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20200221112522.1497712-5-stefanha@redhat.com>
[mreitz: Renamed test from 282 to 288]
Signed-off-by: Max Reitz <mreitz@redhat.com>

qemu-img: allow qemu-img measure --object without a filename

In most qemu-img sub-commands the --object option only makes sense when
there is a filename.  qemu-img measure is an exception because objects
may be referenced from the image creation options instead of an existing
image file.  Allow --object without a filename.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20200221112522.1497712-4-stefanha@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>

luks: implement .bdrv_measure()

Add qemu-img measure support in the "luks" block driver.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20200221112522.1497712-3-stefanha@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>

luks: extract qcrypto_block_calculate_payload_offset()

The qcow2 .bdrv_measure() code calculates the crypto payload offset.
This logic really belongs in crypto/block.c where it can be reused by
other image formats.

The "luks" block driver will need this same logic in order to implement
.bdrv_measure(), so extract the qcrypto_block_calculate_payload_offset()
function now.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20200221112522.1497712-2-stefanha@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>

Merge remote-tracking branch 'remotes/borntraeger/tags/s390x-20200310' into staging

s390x/ipl: Fixes for ipl and bios

- provide a pointer to the loadparm. This fixes crashes in zipl
- do not throw away guest changes of the IPL parameter during reset
- refactor IPLB checks

# gpg: Signature made Tue 10 Mar 2020 14:50:31 GMT
# gpg:                using RSA key 117BBC80B5A61C7C
# gpg: Good signature from "Christian Borntraeger (2nd IBM address) <borntraeger@linux.ibm.com>" [unknown]
# gpg:                 aka "Christian Borntraeger (IBM) <borntraeger@de.ibm.com>" [full]
# gpg:                 aka "Christian Borntraeger (kernel.org email address) <borntraeger@kernel.org>" [unknown]
# Primary key fingerprint: F922 9381 A334 08F9 DBAB  FBCA 117B BC80 B5A6 1C7C

* remotes/borntraeger/tags/s390x-20200310:
  s390x: ipl: Consolidate iplb validity check into one function
  s390/ipl: sync back loadparm
  s390x/bios: rebuild s390-ccw.img
  pc-bios: s390x: Save iplb location in lowcore

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

9p/proxy: Fix export_flags

The common fsdev options are set by qemu_fsdev_add() before it calls
the backend specific option parsing code. In the case of "proxy" this
means "writeout" or "readonly" were simply ignored. This has been
broken from the beginning.

Reported-by: Stéphane Graber <stgraber@ubuntu.com>
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <158349633705.1237488.8895481990204796135.stgit@bahia.lan>

Merge remote-tracking branch 'remotes/kraxel/tags/usb-20200310-pull-request' into staging

usb: bugfixes for ehci & serial.

# gpg: Signature made Tue 10 Mar 2020 08:23:37 GMT
# gpg:                using RSA key 4CB6D8EED3E87138
# gpg: Good signature from "Gerd Hoffmann (work) <kraxel@redhat.com>" [full]
# gpg:                 aka "Gerd Hoffmann <gerd@kraxel.org>" [full]
# gpg:                 aka "Gerd Hoffmann (private) <kraxel@gmail.com>" [full]
# Primary key fingerprint: A032 8CFF B93A 17A7 9901  FE7D 4CB6 D8EE D3E8 7138

* remotes/kraxel/tags/usb-20200310-pull-request:
  usb/hcd-ehci: Remove redundant statements
  usb-serial: wakeup device on input

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

s390x: ipl: Consolidate iplb validity check into one function

It's nicer to just call one function than calling a function for each
possible iplb type.

Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-Id: <20200310090950.61172-1-frankja@linux.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>

s390/ipl: sync back loadparm

We expose loadparm as a r/w machine property, but if loadparm is set by
the guest via DIAG 308, we don't update the property. Having a
disconnect between the guest view and the QEMU property is not nice in
itself, but things get even worse for SCSI, where under certain
circumstances (see 789b5a401b "s390: Ensure IPL from SCSI works as
expected" for details) we call s390_gen_initial_iplb() on resets
effectively overwriting the guest/user supplied loadparm with the stale
value.

Signed-off-by: Halil Pasic <pasic@linux.ibm.com>
Fixes: 7104bae9de ("hw/s390x: provide loadparm property for the machine")
Reported-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Viktor Mihajlovski <mihajlov@linux.ibm.com>
Tested-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-Id: <20200309133223.100491-1-pasic@linux.ibm.com>
[borntraeger@de.ibm.com: use reverse xmas tree]
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>

s390x/bios: rebuild s390-ccw.img

contains 98df96d4edae ("pc-bios: s390x: Save iplb location in lowcore")

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>

pc-bios: s390x: Save iplb location in lowcore

The POP states that for a list directed IPL the IPLB is stored into
memory by the machine loader and its address is stored at offset 0x14
of the lowcore.

ZIPL currently uses the address in offset 0x14 to access the IPLB and
acquire flags about secure boot. If the IPLB address points into
memory which has an unsupported mix of flags set, ZIPL will panic
instead of booting the OS.

As the lowcore can have quite a high entropy for a guest that did drop
out of protected mode (i.e. rebooted) we encountered the ZIPL panic
quite often.

Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Tested-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Message-Id: <20200304114231.23493-19-frankja@linux.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>

Merge remote-tracking branch 'remotes/dgilbert/tags/pull-hmp-20200309' into staging

HMP Pull 2020-03-09

Maxim's hmp block move, Thomas's deprecation in hostfwd.

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
# gpg: Signature made Mon 09 Mar 2020 19:43:33 GMT
# gpg:                using RSA key 45F5C71B4A0CB7FB977A9FA90516331EBC5BFDE7
# gpg: Good signature from "Dr. David Alan Gilbert (RH2) <dgilbert@redhat.com>" [full]
# Primary key fingerprint: 45F5 C71B 4A0C B7FB 977A  9FA9 0516 331E BC5B FDE7

* remotes/dgilbert/tags/pull-hmp-20200309:
  net: Remove deprecated [hub_id name] tuple of 'hostfwd_add' / 'hostfwd_remove'
  monitor/hmp: Move hmp_drive_add_node to block-hmp-cmds.c
  monitor/hmp: move hmp_info_block* to block-hmp-cmds.c
  monitor/hmp: move remaining hmp_block* functions to block-hmp-cmds.c
  monitor/hmp: move hmp_nbd_server* to block-hmp-cmds.c
  monitor/hmp: move hmp_snapshot_* to block-hmp-cmds.c
  monitor/hmp: move hmp_block_job* to block-hmp-cmds.c
  monitor/hmp: move hmp_drive_mirror and hmp_drive_backup to block-hmp-cmds.c
  monitor/hmp: move hmp_drive_del and hmp_commit to block-hmp-cmds.c
  monitor/hmp: rename device-hotplug.c to block/monitor/block-hmp-cmds.c
  monitor/hmp: inline add_init_drive
  usb/dev-storage: remove unused include

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

net: Remove deprecated [hub_id name] tuple of 'hostfwd_add' / 'hostfwd_remove'

It's been deprecated since QEMU v3.1.0. Time to finally remove it now.

Signed-off-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20191205104109.18680-1-thuth@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
  Reworked Thomas's deprecated.texi to the rst

monitor/hmp: Move hmp_drive_add_node to block-hmp-cmds.c

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20200308092440.23564-12-mlevitsk@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

monitor/hmp: move hmp_info_block* to block-hmp-cmds.c

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20200308092440.23564-11-mlevitsk@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

monitor/hmp: move remaining hmp_block* functions to block-hmp-cmds.c

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20200308092440.23564-10-mlevitsk@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

monitor/hmp: move hmp_nbd_server* to block-hmp-cmds.c

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20200308092440.23564-9-mlevitsk@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

monitor/hmp: move hmp_snapshot_* to block-hmp-cmds.c

hmp_snapshot_blkdev is from GPLv2 version of the hmp-cmds.c thus
have to change the licence to GPLv2

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20200308092440.23564-8-mlevitsk@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

monitor/hmp: move hmp_block_job* to block-hmp-cmds.c

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20200308092440.23564-7-mlevitsk@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

monitor/hmp: move hmp_drive_mirror and hmp_drive_backup to block-hmp-cmds.c

Moved code was added after 2012-01-13, thus under GPLv2+

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20200308092440.23564-6-mlevitsk@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
  Fixed commit message

monitor/hmp: move hmp_drive_del and hmp_commit to block-hmp-cmds.c

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20200308092440.23564-5-mlevitsk@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

monitor/hmp: rename device-hotplug.c to block/monitor/block-hmp-cmds.c

These days device-hotplug.c only contains the hmp_drive_add
In the next patch, rest of hmp_drive* functions will be moved
there.

Also add block-hmp-cmds.h to contain prototypes of these
functions

License for block-hmp-cmds.h since it contains the code
moved from sysemu.h which lacks license and thus according
to LICENSE is under GPLv2+

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20200308092440.23564-4-mlevitsk@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

monitor/hmp: inline add_init_drive

This function is only used by hmp_drive_add.
The code is just a bit shorter this way.

No functional changes

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20200308092440.23564-3-mlevitsk@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

usb/dev-storage: remove unused include

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200308092440.23564-2-mlevitsk@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

aio-posix: remove idle poll handlers to improve scalability

When there are many poll handlers it's likely that some of them are idle
most of the time.  Remove handlers that haven't had activity recently so
that the polling loop scales better for guests with a large number of
devices.

This feature only takes effect for the Linux io_uring fd monitoring
implementation because it is capable of combining fd monitoring with
userspace polling.  The other implementations can't do that and risk
starving fds in favor of poll handlers, so don't try this optimization
when they are in use.

IOPS improves from 10k to 105k when the guest has 100
virtio-blk-pci,num-queues=32 devices and 1 virtio-blk-pci,num-queues=1
device for rw=randread,iodepth=1,bs=4k,ioengine=libaio on NVMe.

[Clarified aio_poll_handlers locking discipline explanation in comment
after discussion with Paolo Bonzini <pbonzini@redhat.com>.
--Stefan]

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Link: https://lore.kernel.org/r/20200305170806.1313245-8-stefanha@redhat.com
Message-Id: <20200305170806.1313245-8-stefanha@redhat.com>

qemu.nsi: Install Sphinx documentation

The old qemu-doc.html is no longer built, so update the Windows
installer to install the new Sphinx manual sets.

We install all five of the manuals, even though some of them
(notably the user-mode manual) will not be very useful to Windows
users, because skipping some of them would mean broken links
in the top level 'index.html' page.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20200306134751.2572-1-peter.maydell@linaro.org

aio-posix: support userspace polling of fd monitoring

Unlike ppoll(2) and epoll(7), Linux io_uring completions can be polled
from userspace.  Previously userspace polling was only allowed when all
AioHandler's had an ->io_poll() callback.  This prevented starvation of
fds by userspace pollable handlers.

Add the FDMonOps->need_wait() callback that enables userspace polling
even when some AioHandlers lack ->io_poll().

For example, it's now possible to do userspace polling when a TCP/IP
socket is monitored thanks to Linux io_uring.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Link: https://lore.kernel.org/r/20200305170806.1313245-7-stefanha@redhat.com
Message-Id: <20200305170806.1313245-7-stefanha@redhat.com>

aio-posix: add io_uring fd monitoring implementation

The recent Linux io_uring API has several advantages over ppoll(2) and
epoll(2).  Details are given in the source code.

Add an io_uring implementation and make it the default on Linux.
Performance is the same as with epoll(7) but later patches add
optimizations that take advantage of io_uring.

It is necessary to change how aio_set_fd_handler() deals with deleting
AioHandlers since removing monitored file descriptors is asynchronous in
io_uring.  fdmon_io_uring_remove() marks the AioHandler deleted and
aio_set_fd_handler() will let it handle deletion in that case.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Link: https://lore.kernel.org/r/20200305170806.1313245-6-stefanha@redhat.com
Message-Id: <20200305170806.1313245-6-stefanha@redhat.com>

aio-posix: simplify FDMonOps->update() prototype

The AioHandler *node, bool is_new arguments are more complicated to
think about than simply being given AioHandler *old_node, AioHandler
*new_node.

Furthermore, the new Linux io_uring file descriptor monitoring mechanism
added by the new patch requires access to both the old and the new
nodes.  Make this change now in preparation.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Link: https://lore.kernel.org/r/20200305170806.1313245-5-stefanha@redhat.com
Message-Id: <20200305170806.1313245-5-stefanha@redhat.com>

aio-posix: extract ppoll(2) and epoll(7) fd monitoring

The ppoll(2) and epoll(7) file descriptor monitoring implementations are
mixed with the core util/aio-posix.c code.  Before adding another
implementation for Linux io_uring, extract out the existing
ones so there is a clear interface and the core code is simpler.

The new interface is AioContext->fdmon_ops, a pointer to a FDMonOps
struct.  See the patch for details.

Semantic changes:
1. ppoll(2) now reflects events from pollfds[] back into AioHandlers
   while we're still on the clock for adaptive polling.  This was
   already happening for epoll(7), so if it's really an issue then we'll
   need to fix both in the future.
2. epoll(7)'s fallback to ppoll(2) while external events are disabled
   was broken when the number of fds exceeded the epoll(7) upgrade
   threshold.  I guess this code path simply wasn't tested and no one
   noticed the bug.  I didn't go out of my way to fix it but the correct
   code is simpler than preserving the bug.

I also took some liberties in removing the unnecessary
AioContext->epoll_available (just check AioContext->epollfd != -1
instead) and AioContext->epoll_enabled (it's implicit if our
AioContext->fdmon_ops callbacks are being invoked) fields.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Link: https://lore.kernel.org/r/20200305170806.1313245-4-stefanha@redhat.com
Message-Id: <20200305170806.1313245-4-stefanha@redhat.com>

aio-posix: move RCU_READ_LOCK() into run_poll_handlers()

Now that run_poll_handlers_once() is only called by run_poll_handlers()
we can improve the CPU time profile by moving the expensive
RCU_READ_LOCK() out of the polling loop.

This reduces the run_poll_handlers() from 40% CPU to 10% CPU in perf's
sampling profiler output.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Link: https://lore.kernel.org/r/20200305170806.1313245-3-stefanha@redhat.com
Message-Id: <20200305170806.1313245-3-stefanha@redhat.com>

aio-posix: completely stop polling when disabled

One iteration of polling is always performed even when polling is
disabled.  This is done because:
1. Userspace polling is cheaper than making a syscall.  We might get
   lucky.
2. We must poll once more after polling has stopped in case an event
   occurred while stopping polling.

However, there are downsides:
1. Polling becomes a bottleneck when the number of event sources is very
   high.  It's more efficient to monitor fds in that case.
2. A high-frequency polling event source can starve non-polling event
   sources because ppoll(2)/epoll(7) is never invoked.

This patch removes the forced polling iteration so that poll_ns=0 really
means no polling.

IOPS increases from 10k to 60k when the guest has 100
virtio-blk-pci,num-queues=32 devices and 1 virtio-blk-pci,num-queues=1
device because the large number of event sources being polled slows down
the event loop.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Link: https://lore.kernel.org/r/20200305170806.1313245-2-stefanha@redhat.com
Message-Id: <20200305170806.1313245-2-stefanha@redhat.com>

aio-posix: remove confusing QLIST_SAFE_REMOVE()

QLIST_SAFE_REMOVE() is confusing here because the node must be on the
list.  We actually just wanted to clear the linked list pointers when
removing it from the list.  QLIST_REMOVE() now does this, so switch to
it.

Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Link: https://lore.kernel.org/r/20200224103406.1894923-3-stefanha@redhat.com
Message-Id: <20200224103406.1894923-3-stefanha@redhat.com>

qemu/queue.h: clear linked list pointers on remove

Do not leave stale linked list pointers around after removal.  It's
safer to set them to NULL so that use-after-removal results in an
immediate segfault.

The RCU queue removal macros are unchanged since nodes may still be
traversed after removal.

Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Link: https://lore.kernel.org/r/20200224103406.1894923-2-stefanha@redhat.com
Message-Id: <20200224103406.1894923-2-stefanha@redhat.com>

Merge remote-tracking branch 'remotes/vivier2/tags/trivial-branch-pull-request' into staging

- includes cleanup
- reduce .data footprint
- fix warnings reported by Clang static code analyzer
- fix dp8393x part lost in merge
- update git.orderfile and rules.mak

# gpg: Signature made Mon 09 Mar 2020 15:07:47 GMT
# gpg:                using RSA key CD2F75DDC8E3A4DC2E4F5173F30C38BD3F2FBE3C
# gpg:                issuer "laurent@vivier.eu"
# gpg: Good signature from "Laurent Vivier <lvivier@redhat.com>" [full]
# gpg:                 aka "Laurent Vivier <laurent@vivier.eu>" [full]
# gpg:                 aka "Laurent Vivier (Red Hat) <lvivier@redhat.com>" [full]
# Primary key fingerprint: CD2F 75DD C8E3 A4DC 2E4F  5173 F30C 38BD 3F2F BE3C

* remotes/vivier2/tags/trivial-branch-pull-request: (33 commits)
  monitor/hmp-cmds: Remove redundant statement in hmp_rocker_of_dpa_groups()
  display/exynos4210_fimd: Remove redundant statement in exynos4210_fimd_update()
  display/pxa2xx_lcd: Remove redundant statement in pxa2xx_palette_parse()
  scsi/scsi-disk: Remove redundant statement in scsi_disk_emulate_command()
  dma/xlnx-zdma: Remove redundant statement in zdma_write_dst()
  block/file-posix: Remove redundant statement in raw_handle_perm_lock()
  block/stream: Remove redundant statement in stream_run()
  core/qdev: fix memleak in qdev_get_gpio_out_connector()
  hw/i386/pc: Clean up includes
  hw/pci-host/q35: Remove unused includes
  hw/i386: Include "hw/mem/nvdimm.h"
  hw/acpi: Include "hw/mem/nvdimm.h"
  hw/pci-host/piix: Include "qemu/range.h"
  hw/i2c/smbus_ich9: Include "qemu/range.h"
  hw/pci-host/q35: Include "qemu/range.h"
  hw/timer/hpet: Include "exec/address-spaces.h"
  hw/acpi/cpu_hotplug: Include "hw/pci/pci.h"
  hw/hppa/machine: Include "net/net.h"
  hw/alpha/dp264: Include "net/net.h"
  hw/alpha/alpha_sys: Remove unused "hw/ide.h" header
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

monitor/hmp-cmds: Remove redundant statement in hmp_rocker_of_dpa_groups()

Clang static code analyzer show warning:
monitor/hmp-cmds.c:2867:17: warning: Value stored to 'set' is never read
                set = true;
                ^     ~~~~

Reported-by: Euler Robot <euler.robot@huawei.com>
Signed-off-by: Chen Qun <kuhn.chenqun@huawei.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200302130715.29440-14-kuhn.chenqun@huawei.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

display/exynos4210_fimd: Remove redundant statement in exynos4210_fimd_update()

Clang static code analyzer show warning:
hw/display/exynos4210_fimd.c:1313:17: warning: Value stored to 'is_dirty' is never read
                is_dirty = false;

Reported-by: Euler Robot <euler.robot@huawei.com>
Signed-off-by: Chen Qun <kuhn.chenqun@huawei.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200302130715.29440-9-kuhn.chenqun@huawei.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

display/pxa2xx_lcd: Remove redundant statement in pxa2xx_palette_parse()

Clang static code analyzer show warning:
hw/display/pxa2xx_lcd.c:596:9: warning: Value stored to 'format' is never read
        format = 0;
        ^        ~

Reported-by: Euler Robot <euler.robot@huawei.com>
Signed-off-by: Chen Qun <kuhn.chenqun@huawei.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200302130715.29440-8-kuhn.chenqun@huawei.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

scsi/scsi-disk: Remove redundant statement in scsi_disk_emulate_command()

Clang static code analyzer show warning:
scsi/scsi-disk.c:1918:5: warning: Value stored to 'buflen' is never read
    buflen = req->cmd.xfer;
    ^        ~~~~~~~~~~~~~

Reported-by: Euler Robot <euler.robot@huawei.com>
Signed-off-by: Chen Qun <kuhn.chenqun@huawei.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200302130715.29440-7-kuhn.chenqun@huawei.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

dma/xlnx-zdma: Remove redundant statement in zdma_write_dst()

Clang static code analyzer show warning:
hw/dma/xlnx-zdma.c:399:13: warning: Value stored to 'dst_type' is never read
            dst_type = FIELD_EX32(s->dsc_dst.words[3], ZDMA_CH_DST_DSCR_WORD3,
            ^          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Reported-by: Euler Robot <euler.robot@huawei.com>
Signed-off-by: Chen Qun <kuhn.chenqun@huawei.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Francisco Iglesias <frasse.iglesias@gmail.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Message-Id: <20200302130715.29440-11-kuhn.chenqun@huawei.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

block/file-posix: Remove redundant statement in raw_handle_perm_lock()

Clang static code analyzer show warning:
  block/file-posix.c:891:9: warning: Value stored to 'op' is never read
        op = RAW_PL_ABORT;
        ^    ~~~~~~~~~~~~

Reported-by: Euler Robot <euler.robot@huawei.com>
Signed-off-by: Chen Qun <kuhn.chenqun@huawei.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200302130715.29440-5-kuhn.chenqun@huawei.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

block/stream: Remove redundant statement in stream_run()

Clang static code analyzer show warning:
  block/stream.c:186:9: warning: Value stored to 'ret' is never read
        ret = 0;
        ^     ~
Reported-by: Euler Robot <euler.robot@huawei.com>
Signed-off-by: Chen Qun <kuhn.chenqun@huawei.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200302130715.29440-3-kuhn.chenqun@huawei.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

core/qdev: fix memleak in qdev_get_gpio_out_connector()

Fix a memory leak in qdev_get_gpio_out_connector().

Reported-by: Euler Robot <euler.robot@huawei.com>
Signed-off-by: Pan Nengyuan <pannengyuan@huawei.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200307030756.5913-1-pannengyuan@huawei.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/i386/pc: Clean up includes

Various headers are not required by hw/i386/pc.h:

 - "qemu/range.h"
 - "qemu/bitmap.h"
 - "qemu/module.h"
 - "exec/memory.h"
 - "hw/pci/pci.h"
 - "hw/mem/pc-dimm.h"
 - "hw/mem/nvdimm.h"
 - "net/net.h"

Remove them.

Add 3 headers that were missing:

 - "hw/hotplug.h"

   PCMachineState::acpi_dev is of type HotplugHandler

 - "qemu/notify.h"

   PCMachineState::machine_done is of type Notifier

 - "qapi/qapi-types-common.h"

   PCMachineState::vmport/smm is of type OnOffAuto

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-19-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/pci-host/q35: Remove unused includes

Only q35.c requires declarations from "hw/i386/pc.h", move it there.
Remove all the includes not used by "q35.h".

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-18-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/i386: Include "hw/mem/nvdimm.h"

All this files use methods/definitions declared in the NVDIMM
device header. Include it.

This fixes (when modifying unrelated headers):

  hw/i386/acpi-build.c:2733:9: error: implicit declaration of function 'nvdimm_build_acpi' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
        nvdimm_build_acpi(table_offsets, tables_blob, tables->linker,
        ^
  hw/i386/pc.c:1996:61: error: use of undeclared identifier 'TYPE_NVDIMM'
    const bool is_nvdimm = object_dynamic_cast(OBJECT(dev), TYPE_NVDIMM);
                                                            ^
  hw/i386/pc.c:2032:55: error: use of undeclared identifier 'TYPE_NVDIMM'
    bool is_nvdimm = object_dynamic_cast(OBJECT(dev), TYPE_NVDIMM);
                                                      ^
  hw/i386/pc.c:2040:9: error: implicit declaration of function 'nvdimm_plug' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
        nvdimm_plug(ms->nvdimms_state);
        ^
  hw/i386/pc.c:2040:9: error: this function declaration is not a prototype [-Werror,-Wstrict-prototypes]
        nvdimm_plug(ms->nvdimms_state);
        ^
  hw/i386/pc.c:2065:42: error: use of undeclared identifier 'TYPE_NVDIMM'
    if (object_dynamic_cast(OBJECT(dev), TYPE_NVDIMM)) {
                                         ^
  hw/i386/pc_i440fx.c:307:9: error: implicit declaration of function 'nvdimm_init_acpi_state' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
        nvdimm_init_acpi_state(machine->nvdimms_state, system_io,
        ^
  hw/i386/pc_q35.c:332:9: error: implicit declaration of function 'nvdimm_init_acpi_state' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
        nvdimm_init_acpi_state(machine->nvdimms_state, system_io,
        ^

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-17-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/acpi: Include "hw/mem/nvdimm.h"

Both ich9.c and piix4.c use methods/definitions declared in the
NVDIMM device header. Include it.

This fixes (when modifying unrelated headers):

  hw/acpi/ich9.c:507:46: error: use of undeclared identifier 'TYPE_NVDIMM'
        if (object_dynamic_cast(OBJECT(dev), TYPE_NVDIMM)) {
                                             ^
  hw/acpi/ich9.c:508:13: error: implicit declaration of function 'nvdimm_acpi_plug_cb' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
            nvdimm_acpi_plug_cb(hotplug_dev, dev);
            ^
  hw/acpi/piix4.c:403:46: error: use of undeclared identifier 'TYPE_NVDIMM'
        if (object_dynamic_cast(OBJECT(dev), TYPE_NVDIMM)) {
                                             ^
  hw/acpi/piix4.c:404:13: error: implicit declaration of function 'nvdimm_acpi_plug_cb' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
            nvdimm_acpi_plug_cb(hotplug_dev, dev);
            ^

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-16-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/pci-host/piix: Include "qemu/range.h"

hw/pci-host/piix.c calls various functions from the Range API.
Include "qemu/range.h" which declares them.

This fixes (when modifying unrelated headers):

  hw/pci-host/i440fx.c:54:11: error: field has incomplete type 'Range' (aka 'struct Range')
      Range pci_hole;
           ^
  include/qemu/typedefs.h:116:16: note: forward declaration of 'struct Range'
  typedef struct Range Range;
                 ^
  hw/pci-host/i440fx.c:126:9: error: implicit declaration of function 'ranges_overlap' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
      if (ranges_overlap(address, len, I440FX_PAM, I440FX_PAM_SIZE) ||
          ^
  hw/pci-host/i440fx.c:126:9: error: this function declaration is not a prototype [-Werror,-Wstrict-prototypes]
  hw/pci-host/i440fx.c:127:9: error: implicit declaration of function 'range_covers_byte' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
          range_covers_byte(address, len, I440FX_SMRAM)) {
          ^
  hw/pci-host/i440fx.c:127:9: error: this function declaration is not a prototype [-Werror,-Wstrict-prototypes]
  hw/pci-host/i440fx.c:189:13: error: implicit declaration of function 'range_is_empty' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
      val64 = range_is_empty(&s->pci_hole) ? 0 : range_lob(&s->pci_hole);
              ^

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-15-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/i2c/smbus_ich9: Include "qemu/range.h"

hw/i2c/smbus_ich9.c calls range_covers_byte(). Include "qemu/range.h"
which declares it.

This fixes (when modifying unrelated headers):

  hw/i2c/smbus_ich9.c:66:9: error: implicit declaration of function 'range_covers_byte' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
      if (range_covers_byte(address, len, ICH9_SMB_HOSTC)) {
          ^

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-14-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/pci-host/q35: Include "qemu/range.h"

The MCHPCIState structure uses the Range type which is declared in
"qemu/range.h". Include it.

This fixes (when modifying unrelated headers):

  In file included from hw/pci-host/q35.c:32:
  include/hw/pci-host/q35.h:57:11: error: field has incomplete type 'Range' (aka 'struct Range')
      Range pci_hole;
            ^
  include/qemu/typedefs.h:116:16: note: forward declaration of 'struct Range'
  typedef struct Range Range;
                 ^

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-13-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/timer/hpet: Include "exec/address-spaces.h"

hw/timer/hpet.c calls address_space_stl_le() declared in
"exec/address-spaces.h". Include it.

This fixes (when modifying unrelated headers):

  hw/timer/hpet.c:210:31: error: use of undeclared identifier 'address_space_memory'
          address_space_stl_le(&address_space_memory, timer->fsb >> 32,
                               ^~~~~~~~~~~~~~~~~~~~

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-12-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/acpi/cpu_hotplug: Include "hw/pci/pci.h"

hw/acpi/cpu_hotplug.c calls pci_address_space_io(). Include
"hw/pci/pci.h" which declares it.

This fixes (when modifying unrelated headers):

  hw/acpi/cpu_hotplug.c:103:28: error: implicit declaration of function 'pci_address_space_io' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
      MemoryRegion *parent = pci_address_space_io(PCI_DEVICE(gpe_cpu->device));
                             ^

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-11-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/hppa/machine: Include "net/net.h"

hw/hppa/machine.c uses NICInfo variables which are declared in
"net/net.h". Include it.

This fixes (when modifying unrelated headers):

  hw/hppa/machine.c:126:21: error: use of undeclared identifier 'nb_nics'
      for (i = 0; i < nb_nics; i++) {
                      ^
  hw/hppa/machine.c:127:30: error: use of undeclared identifier 'nd_table'
          pci_nic_init_nofail(&nd_table[i], pci_bus, "e1000", NULL);
                               ^

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-10-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/alpha/dp264: Include "net/net.h"

hw/alpha/dp264.c uses NICInfo variables which are declared in
"net/net.h". Include it.

This fixes (when modifying unrelated headers):

  hw/alpha/dp264.c:89:21: error: use of undeclared identifier 'nb_nics'
      for (i = 0; i < nb_nics; i++) {
                      ^
  hw/alpha/dp264.c:90:30: error: use of undeclared identifier 'nd_table'
          pci_nic_init_nofail(&nd_table[i], pci_bus, "e1000", NULL);
                               ^

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-9-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/alpha/alpha_sys: Remove unused "hw/ide.h" header

alpha_sys.h does not use anything from the "hw/ide.h" header.
Remove it.

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-8-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/i386/intel_iommu: Remove unused includes

intel_iommu.h does not use any of these includes, remove them.

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-7-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/usb/dev-storage: Remove unused "ui/console.h" header

The USB models related to storage don't need anything from
"ui/console.h". Remove it.

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-6-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/timer: Remove unused "ui/console.h" header

The timer models don't need anything from "ui/console.h".
Remove it.

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-5-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/i386/ioapic_internal: Remove unused "hw/i386/ioapic.h" header

The "ioapic_internal.h" does not use anything from
"hw/i386/ioapic.h", remove it.

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-4-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/southbridge/ich9: Removed unused headers

The ICH9 chipset is not X86/PC specific.

These files don't use anything declared by the "hw/i386/pc.h"
or "hw/i386/ioapic.h" headers. Remove them.

Reviewed-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-3-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

vl: Add missing "hw/boards.h" include

vl.c calls machine_usb() declared in "hw/boards.h". Include it.

This fixes (when modifying unrelated headers):

  vl.c:1283:10: error: implicit declaration of function 'machine_usb' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
      if (!machine_usb(current_machine)) {
           ^
  vl.c:1283:10: error: this function declaration is not a prototype [-Werror,-Wstrict-prototypes]
  vl.c:1283:22: error: use of undeclared identifier 'current_machine'
      if (!machine_usb(current_machine)) {
                       ^

Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200228114649.12818-2-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

virtfs-proxy-helper: Make the helper_opts[] array const

Reduce a bit the memory footprint by making the helper_opts[]
array const.

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Acked-by: Greg Kurz <groug@kaod.org>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20200305010446.17029-4-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/net/e1000: Move macreg[] arrays to .rodata to save 1MiB of .data

Each array consumes 256KiB of .data. As we do not reassign entries,
we can move it to the .rodata section, and save a total of 1MiB of
.data (size reported on x86_64 host).

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Dmitry Fleytman <dmitry.fleytman@gmail.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20200305010446.17029-3-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/net/e1000: Add readops/writeops typedefs

Express the macreg[] arrays using typedefs.
No logical changes introduced here.

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Dmitry Fleytman <dmitry.fleytman@gmail.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20200305010446.17029-2-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/audio/fmopl: Fix a typo twice

Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20200305124525.14555-2-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

build-sys: Move the print-variable rule to rules.mak

Currently the print-variable rule can only be used in the
root directory:

  $ make print-vhost-user-json-y
  vhost-user-json-y= contrib/vhost-user-gpu/50-qemu-gpu.json tools/virtiofsd/50-qemu-virtiofsd.json

  $ make -C i386-softmmu print-obj-y
  make: Entering directory 'build/i386-softmmu'
  make: *** No rule to make target 'print-obj-y'.  Stop.
  make: Leaving directory 'build/i386-softmmu'

Move it to rules.mak so we can use it from other directories:

  $ make -C i386-softmmu print-obj-y
  make: Entering directory 'build/i386-softmmu'
  obj-y=qapi-introspect.o qapi-types-machine-target.o qapi-types-misc-target.o qapi-types.o qapi-visit-machine-target.o qapi-visit-misc-target.o qapi-visit.o qapi-events-machine-target.o qapi-events-misc-target.o qapi-events.o qapi-commands-machine-target.o qapi-commands-misc-target.o qapi-commands.o qapi-init-commands.o
  make: Leaving directory 'build/i386-softmmu'

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200306170456.21977-1-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

maint: Include top-level *.rst files early in git diff

We are converting more doc files to *.rst rather than *.texi.  Most
doc files are already listed early in diffs due to our catchall
docs/*, but a few top-level files get missed by that glob.

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20200220162214.3474280-1-eblake@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

dp8393x: Mask EOL bit from descriptor addresses, take 2

A portion of a recent patch got lost due to a merge snafu. That patch is
now commit 88f632fbb1 ("dp8393x: Mask EOL bit from descriptor addresses").
This patch restores the portion that got lost.

Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <alpine.LNX.2.22.394.2003041421280.12@nippy.intranet>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

Merge remote-tracking branch 'remotes/armbru/tags/pull-error-2020-03-09' into staging

Error reporting patches for 2020-03-09

# gpg: Signature made Mon 09 Mar 2020 12:37:04 GMT
# gpg:                using RSA key 354BC8B3D7EB2A6B68674E5F3870B400EB918653
# gpg:                issuer "armbru@redhat.com"
# gpg: Good signature from "Markus Armbruster <armbru@redhat.com>" [full]
# gpg:                 aka "Markus Armbruster <armbru@pond.sub.org>" [full]
# Primary key fingerprint: 354B C8B3 D7EB 2A6B 6867  4E5F 3870 B400 EB91 8653

* remotes/armbru/tags/pull-error-2020-03-09:
  qga: Fix a memory leak
  qga: Improve error report by calling error_setg_win32()
  util/osdep: Improve error report by calling error_setg_win32()
  chardev: Improve error report by calling error_setg_win32()

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

qga: Fix a memory leak

The string returned by g_win32_error_message() has to be
deallocated with g_free().

Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200228100726.8414-5-philmd@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>

qga: Improve error report by calling error_setg_win32()

Use error_setg_win32() which adds a hint similar to strerror(errno)).

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200228100726.8414-4-philmd@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>

util/osdep: Improve error report by calling error_setg_win32()

Use error_setg_win32() which adds a hint similar to strerror(errno)).

Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200228100726.8414-3-philmd@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>

chardev: Improve error report by calling error_setg_win32()

Use error_setg_win32() which adds a hint similar to strerror(errno)).

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200228100726.8414-2-philmd@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>

Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging

virtio, pci, pc: fixes, cleanups, features

Bugfixes, cleanups all over the place.
Ability to disable hotplug for pci express ports.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
# gpg: Signature made Sun 08 Mar 2020 13:27:54 GMT
# gpg:                using RSA key 5D09FD0871C8F85B94CA8A0D281F0DB8D28D5469
# gpg:                issuer "mst@redhat.com"
# gpg: Good signature from "Michael S. Tsirkin <mst@kernel.org>" [full]
# gpg:                 aka "Michael S. Tsirkin <mst@redhat.com>" [full]
# 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/intel_iommu: Simplify vtd_find_as_from_bus_num() logic
  vhost-vsock: fix error message output
  vhost: correctly turn on VIRTIO_F_IOMMU_PLATFORM
  pcie_root_port: Add hotplug disabling option

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

usb/hcd-ehci: Remove redundant statements

The "again" assignment is meaningless before g_assert_not_reached.
In addition, the break statements no longer needs to be after
g_assert_not_reached.

Clang static code analyzer show warning:
hw/usb/hcd-ehci.c:2108:13: warning: Value stored to 'again' is never read
            again = -1;
            ^       ~~

Reported-by: Euler Robot <euler.robot@huawei.com>
Signed-off-by: Chen Qun <kuhn.chenqun@huawei.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200226084647.20636-13-kuhn.chenqun@huawei.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

usb-serial: wakeup device on input

Currently usb-serial devices are unable to send data into guests with
the xhci controller.  Data is copied into the usb-serial's buffer, but
it is not sent into the guest.  Data coming out of the guest works
properly.  usb-serial devices work properly with ehci.

Have usb-serial call usb_wakeup() when receiving data from the chardev.
This seems to notify the xhci controller and fix inbound data flow.

Also add USB_CFG_ATT_WAKEUP to the device's bmAttributes.  This matches
a real FTDI serial adapter's bmAttributes.

Signed-off-by: Jason Andryuk <jandryuk@gmail.com>
Message-id: 20200306140917.26726-1-jandryuk@gmail.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

hw/i386/intel_iommu: Simplify vtd_find_as_from_bus_num() logic

The vtd_find_as_from_bus_num() function was introduced (in commit
dbaabb25f) in a code format that could return an incorrect pointer,
which was later fixed by commit a2e1cd41ccf.
We could have avoided this by writing the if() statement differently.
Do it now, in case this function is re-used. The code is easier to
review (harder to miss bugs).

Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200305102702.31512-1-philmd@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

vhost-vsock: fix error message output

error_setg_errno takes a positive error number, so we should not invert
errno's sign.

Signed-off-by: Nick Erdmann <n@nirf.de>
Message-Id: <04df3f47-c93b-1d02-d250-f9bda8dbc0fa@nirf.de>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Fixes: fc0b9b0e1cbb ("vhost-vsock: add virtio sockets device")
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

vhost: correctly turn on VIRTIO_F_IOMMU_PLATFORM

We turn on device IOTLB via VIRTIO_F_IOMMU_PLATFORM unconditionally on
platform without IOMMU support. This can lead unnecessary IOTLB
transactions which will damage the performance.

Fixing this by check whether the device is backed by IOMMU and disable
device IOTLB.

Reported-by: Halil Pasic <pasic@linux.ibm.com>
Tested-by: Halil Pasic <pasic@linux.ibm.com>
Reviewed-by: Halil Pasic <pasic@linux.ibm.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Message-Id: <20200302042454.24814-1-jasowang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

pcie_root_port: Add hotplug disabling option

Make hot-plug/hot-unplug on PCIe Root Ports optional to allow libvirt
manage it and restrict unplug for the whole machine. This is going to
prevent user-initiated unplug in guests (Windows mostly).
Hotplug is enabled by default.
Usage:
    -device pcie-root-port,hotplug=off,...

If you want to disable hot-unplug on some downstream ports of one
switch, disable hot-unplug on PCIe Root Port connected to the upstream
port as well as on the selected downstream ports.

Discussion related:
    https://lists.gnu.org/archive/html/qemu-devel/2020-02/msg00530.html

Signed-off-by: Julia Suvorova <jusual@redhat.com>
Message-Id: <20200226174607.205941-1-jusual@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>

Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging

Block layer patches:

- Add qemu-storage-daemon (still experimental)
- rbd: Add support for ceph namespaces
- Fix bdrv_reopen() with backing file in different AioContext
- qcow2: Fix read-write reopen with persistent dirty bitmaps
- qcow2: Fix alloc_cluster_abort() for pre-existing clusters

# gpg: Signature made Fri 06 Mar 2020 17:12:31 GMT
# gpg:                using RSA key 7F09B272C88F2FD6
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>" [full]
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74  56FE 7F09 B272 C88F 2FD6

* remotes/kevin/tags/for-upstream: (29 commits)
  block: bdrv_reopen() with backing file in different AioContext
  iotests: Refactor blockdev-reopen test for iothreads
  block/rbd: Add support for ceph namespaces
  qemu-storage-daemon: Add --monitor option
  monitor: Add allow_hmp parameter to monitor_init()
  hmp: Fail gracefully if chardev is already in use
  qmp: Fail gracefully if chardev is already in use
  monitor: Create QAPIfied monitor_init()
  qapi: Create 'pragma' module
  stubs: Update monitor stubs for qemu-storage-daemon
  qemu-storage-daemon: Add --chardev option
  qemu-storage-daemon: Add main loop
  qemu-storage-daemon: Add --export option
  blockdev-nbd: Boxed argument type for nbd-server-add
  qemu-storage-daemon: Add --nbd-server option
  qemu-storage-daemon: Add --object option
  qapi: Flatten object-add
  qemu-storage-daemon: Add --blockdev option
  block: Move sysemu QMP commands to QAPI block module
  block: Move common QMP commands to block-core QAPI module
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

block: bdrv_reopen() with backing file in different AioContext

This patch allows bdrv_reopen() (and therefore the x-blockdev-reopen QMP
command) to attach a node as the new backing file even if the node is in
a different AioContext than the parent if one of both nodes can be moved
to the AioContext of the other node.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Tested-by: Peter Krempa <pkrempa@redhat.com>
Message-Id: <20200306141413.30705-3-kwolf@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

iotests: Refactor blockdev-reopen test for iothreads

We'll want to test more than one successful case in the future, so
prepare the test for that by a refactoring that runs each scenario in a
separate VM.

test_iothreads_switch_{backing,overlay} currently produce errors, but
these are cases that should actually work, by switching either the
backing file node or the overlay node to the AioContext of the other
node.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Tested-by: Peter Krempa <pkrempa@redhat.com>
Message-Id: <20200306141413.30705-2-kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

block/rbd: Add support for ceph namespaces

Starting from ceph Nautilus, RBD has support for namespaces, allowing
for finer grain ACLs on images inside a pool, and tenant isolation.

In the rbd cli tool documentation, the new image-spec and snap-spec are :
 - [pool-name/[namespace-name/]]image-name
 - [pool-name/[namespace-name/]]image-name@snap-name

When using an non namespace's enabled qemu, it complains about not
finding the image called namespace-name/image-name, thus we only need to
parse the image once again to find if there is a '/' in its name, and if
there is, use what is before it as the name of the namespace to later
pass it to rados_ioctx_set_namespace.
rados_ioctx_set_namespace if called with en empty string or a null
pointer as the namespace parameters pretty much does nothing, as it then
defaults to the default namespace.

The namespace is extracted inside qemu_rbd_parse_filename, stored in the
qdict, and used in qemu_rbd_connect to make it work with both qemu-img,
and qemu itself.

Signed-off-by: Florian Florensa <fflorensa@online.net>
Message-Id: <20200110111513.321728-2-fflorensa@online.net>
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

qemu-storage-daemon: Add --monitor option

This adds and parses the --monitor option, so that a QMP monitor can be
used in the storage daemon. The monitor offers commands defined in the
QAPI schema at storage-daemon/qapi/qapi-schema.json.

The --monitor options currently allows to create multiple monitors with
the same ID. This part of the interface is considered unstable. We will
reject such configurations as soon as we have a design for the monitor
subsystem to perform these checks. (In the system emulator, we depend on
QemuOpts rejecting duplicate IDs.)

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200224143008.13362-21-kwolf@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

monitor: Add allow_hmp parameter to monitor_init()

Add a new parameter allow_hmp to monitor_init() so that the storage
daemon can disable HMP.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200224143008.13362-20-kwolf@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

hmp: Fail gracefully if chardev is already in use

Trying to attach a HMP monitor to a chardev that is already in use
results in a crash because monitor_init_hmp() passes &error_abort to
qemu_chr_fe_init():

$ ./x86_64-softmmu/qemu-system-x86_64 --chardev stdio,id=foo --mon foo --mon foo
QEMU 4.2.50 monitor - type 'help' for more information
(qemu) Unexpected error in qemu_chr_fe_init() at chardev/char-fe.c:220:
qemu-system-x86_64: --mon foo: Device 'foo' is in use
Abgebrochen (Speicherabzug geschrieben)

Fix this by allowing monitor_init_hmp() to return an error and passing
any error in qemu_chr_fe_init() to its caller instead of aborting.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200224143008.13362-19-kwolf@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

qmp: Fail gracefully if chardev is already in use

Trying to attach a QMP monitor to a chardev that is already in use
results in a crash because monitor_init_qmp() passes &error_abort to
qemu_chr_fe_init():

$ ./x86_64-softmmu/qemu-system-x86_64 --chardev stdio,id=foo --mon foo,mode=control --mon foo,mode=control
Unexpected error in qemu_chr_fe_init() at chardev/char-fe.c:220:
qemu-system-x86_64: --mon foo,mode=control: Device 'foo' is in use
Abgebrochen (Speicherabzug geschrieben)

Fix this by allowing monitor_init_qmp() to return an error and passing
any error in qemu_chr_fe_init() to its caller instead of aborting.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200224143008.13362-18-kwolf@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

monitor: Create QAPIfied monitor_init()

This adds a new QAPI-based monitor_init() function. The existing
monitor_init_opts() is rewritten to simply put its QemuOpts parameter
into a visitor and pass the resulting QAPI object to monitor_init().

This will cause some change in those error messages for the monitor
options in the system emulator that are now generated by the visitor
rather than explicitly checked in monitor_init_opts().

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200224143008.13362-17-kwolf@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

qapi: Create 'pragma' module

We want to share the whitelists between the system emulator schema and
the storage daemon schema, so move all the pragmas from the main schema
file into a separate file that can be included from both.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200224143008.13362-16-kwolf@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

stubs: Update monitor stubs for qemu-storage-daemon

Before we can add the monitor to qemu-storage-daemon, we need to add a
stubs for monitor_fdsets_cleanup().

We also need to make sure that stubs that are actually implemented in
the monitor core aren't linked to qemu-storage-daemon so that we don't
get linker errors because of duplicate symbols. This is achieved by
moving the stubs in question to a new file stubs/monitor-core.c.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200224143008.13362-15-kwolf@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

qemu-storage-daemon: Add --chardev option

This adds a --chardev option to the storage daemon that works the same
as the -chardev option of the system emulator.

The syntax of the --chardev option is still considered unstable. We want
to QAPIfy it and will potentially make changes to its syntax while
converting it. However, we haven't decided yet on a design for the
QAPIfication, so QemuOpts will have to do for now.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200224143008.13362-14-kwolf@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

qemu-storage-daemon: Add main loop

Instead of exiting after processing all command line options, start a
main loop and keep processing events until exit is requested with a
signal (e.g. SIGINT).

Now qemu-storage-daemon can be used as an alternative for qemu-nbd that
provides a few features that were previously only available from QMP,
such as access to options only available with -blockdev and the socket
types 'vsock' and 'fd'.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200224143008.13362-13-kwolf@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

qemu-storage-daemon: Add --export option

Add a --export option to qemu-storage-daemon to export a block node. For
now, only NBD exports are implemented. Apart from the 'type' option
(which is the implied key), it maps the arguments for nbd-server-add to
the command line. Example:

    --export nbd,device=disk,name=test-export,writable=on

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200224143008.13362-12-kwolf@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

blockdev-nbd: Boxed argument type for nbd-server-add

Move the arguments of nbd-server-add to a new struct BlockExportNbd and
convert the command to 'boxed': true. This makes it easier to share code
with the storage daemon.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200224143008.13362-11-kwolf@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

qemu-storage-daemon: Add --nbd-server option

Add a --nbd-server option to qemu-storage-daemon to start the built-in
NBD server right away. It maps the arguments for nbd-server-start to the
command line, with the exception that it uses SocketAddress instead of
SocketAddressLegacy: New interfaces shouldn't use legacy types, and the
additional nesting would be nasty on the command line.

Example (only with required options):

    --nbd-server addr.type=inet,addr.host=localhost,addr.port=10809

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200224143008.13362-10-kwolf@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

qemu-storage-daemon: Add --object option

Add a command line option to create user-creatable QOM objects.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200224143008.13362-9-kwolf@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>