qemu-iotests: _cleanup_qemu must be called on exit

For the tests that use the common.qemu functions for running a QEMU
process, _cleanup_qemu must be called in the exit function.

If it is not, if the qemu process aborts, then not all of the droppings
are cleaned up (e.g. pidfile, fifos).

This updates those tests that did not have a cleanup in qemu-iotests.

(I swapped spaces for tabs in test 102 as well)

Reported-by: Eric Blake <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Signed-off-by: Jeff Cody <[email protected]>
Message-id: d59c2f6ad6c1da8b9b3c7f357c94a7122ccfc55a.1492544096[email protected]

block/rbd: Add support for reopen()

This adds support for reopen in rbd, for changing between r/w and r/o.

Note, that this is only a flag change, but we will block a change from
r/o to r/w if we are using an RBD internal snapshot.

Reviewed-by: Stefan Hajnoczi <[email protected]>
Signed-off-by: Jeff Cody <[email protected]>
Reviewed-by: John Snow <[email protected]>
Message-id: d4e87539167ec6527d44c97b164eabcccf96e4f3.1491597120[email protected]

block/rbd - update variable names to more apt names

Update 'clientname' to be 'user', which tracks better with both
the QAPI and rados variable naming.

Update 'name' to be 'image_name', as it indicates the rbd image.
Naming it 'image' would have been ideal, but we are using that for
the rados_image_t value returned by rbd_open().

Reviewed-by: Stefan Hajnoczi <[email protected]>
Signed-off-by: Jeff Cody <[email protected]>
Reviewed-by: John Snow <[email protected]>
Message-id: b7ec1fb2e1cf36f9b6911631447a5b0422590b7d.1491597120[email protected]

block: use bdrv_can_set_read_only() during reopen

Signed-off-by: Jeff Cody <[email protected]>
Reviewed-by: Stefan Hajnoczi <[email protected]>
Reviewed-by: John Snow <[email protected]>
Message-id: 00aed7ffdd7be4b9ed9ce1007d50028a72b34ebe.1491597120[email protected]

block: introduce bdrv_can_set_read_only()

Introduce check function for setting read_only flags.  Will return < 0 on
error, with appropriate Error value set.  Does not alter any flags.

Signed-off-by: Jeff Cody <[email protected]>
Reviewed-by: Stefan Hajnoczi <[email protected]>
Reviewed-by: John Snow <[email protected]>
Message-id: e2bba34ac3bc76a0c42adc390413f358ae0566e8.1491597120[email protected]

block: code movement

Move bdrv_is_read_only() up with its friends.

Reviewed-by: Stefan Hajnoczi <[email protected]>
Reviewed-by: John Snow <[email protected]>
Signed-off-by: Jeff Cody <[email protected]>
Message-id: 73b2399459760c32506f9407efb9dddb3a2789de.1491597120[email protected]

block: honor BDRV_O_ALLOW_RDWR when clearing bs->read_only

The BDRV_O_ALLOW_RDWR flag allows / prohibits the changing of
the BDS 'read_only' state, but there are a few places where it
is ignored.  In the bdrv_set_read_only() helper, make sure to
honor the flag.

Signed-off-by: Jeff Cody <[email protected]>
Reviewed-by: Stefan Hajnoczi <[email protected]>
Reviewed-by: John Snow <[email protected]>
Message-id: be2e5fb2d285cbece2b6d06bed54a6f56520d251.1491597120[email protected]

block: do not set BDS read_only if copy_on_read enabled

A few block drivers will set the BDS read_only flag from their
.bdrv_open() function.  This means the bs->read_only flag could
be set after we enable copy_on_read, as the BDRV_O_COPY_ON_READ
flag check occurs prior to the call to bdrv->bdrv_open().

This adds an error return to bdrv_set_read_only(), and an error will be
return if we try to set the BDS to read_only while copy_on_read is
enabled.

This patch also changes the behavior of vvfat.  Before, vvfat could
override the drive 'readonly' flag with its own, internal 'rw' flag.

For instance, this -drive parameter would result in a writable image:

"-drive format=vvfat,dir=/tmp/vvfat,rw,if=virtio,readonly=on"

This is not correct.  Now, attempting to use the above -drive parameter
will result in an error (i.e., 'rw' is incompatible with 'readonly=on').

Signed-off-by: Jeff Cody <[email protected]>
Reviewed-by: Stefan Hajnoczi <[email protected]>
Reviewed-by: John Snow <[email protected]>
Message-id: 0c5b4c1cc2c651471b131f21376dfd5ea24d2196.1491597120[email protected]

block: add bdrv_set_read_only() helper function

We have a helper wrapper for checking for the BDS read_only flag,
add a helper wrapper to set the read_only flag as well.

Reviewed-by: Stefan Hajnoczi <[email protected]>
Signed-off-by: Jeff Cody <[email protected]>
Reviewed-by: John Snow <[email protected]>
Message-id: 9b18972d05f5fa2ac16c014f0af98d680553048d.1491597120[email protected]

qemu-iotests: exclude vxhs from image creation via protocol

The protocol VXHS does not support image creation.  Some tests expect
to be able to create images through the protocol.  Exclude VXHS from
these tests.

Signed-off-by: Jeff Cody <[email protected]>

block/vxhs.c: Add qemu-iotests for new block device type "vxhs"

These changes use a vxhs test server that is a part of the following
repository:
https://github.com/VeritasHyperScale/libqnio.git

Signed-off-by: Ashish Mittal <[email protected]>
Reviewed-by: Stefan Hajnoczi <[email protected]>
Reviewed-by: Jeff Cody <[email protected]>
Signed-off-by: Jeff Cody <[email protected]>
Message-id: 1491277689[email protected]

block/vxhs.c: Add support for a new block device type called "vxhs"

Source code for the qnio library that this code loads can be downloaded from:
https://github.com/VeritasHyperScale/libqnio.git

Sample command line using JSON syntax:
./x86_64-softmmu/qemu-system-x86_64 -name instance-00000008 -S -vnc 0.0.0.0:0
-k en-us -vga cirrus -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x5
-msg timestamp=on
'json:{"driver":"vxhs","vdisk-id":"c3e9095a-a5ee-4dce-afeb-2a59fb387410",
"server":{"host":"172.172.17.4","port":"9999"}}'

Sample command line using URI syntax:
qemu-img convert -f raw -O raw -n
/var/lib/nova/instances/_base/0c5eacd5ebea5ed914b6a3e7b18f1ce734c386ad
vxhs://192.168.0.1:9999/c6718f6b-0401-441d-a8c3-1f0064d75ee0

Sample command line using TLS credentials (run in secure mode):
./qemu-io --object
tls-creds-x509,id=tls0,dir=/etc/pki/qemu/vxhs,endpoint=client -c 'read
-v 66000 2.5k' 'json:{"server.host": "127.0.0.1", "server.port": "9999",
"vdisk-id": "/test.raw", "driver": "vxhs", "tls-creds":"tls0"}'

[Jeff: Modified trace-events with the correct string formatting]

Signed-off-by: Ashish Mittal <[email protected]>
Reviewed-by: Stefan Hajnoczi <[email protected]>
Reviewed-by: Jeff Cody <[email protected]>
Signed-off-by: Jeff Cody <[email protected]>
Message-id: 1491277689[email protected]

Merge remote-tracking branch 'remotes/armbru/tags/pull-error-2017-04-24' into staging

Error reporting patches for 2017-04-24

# gpg: Signature made Mon 24 Apr 2017 08:16:34 BST
# gpg:                using RSA key 0x3870B400EB918653
# gpg: Good signature from "Markus Armbruster <[email protected]>"
# gpg:                 aka "Markus Armbruster <[email protected]>"
# Primary key fingerprint: 354B C8B3 D7EB 2A6B 6867  4E5F 3870 B400 EB91 8653

* remotes/armbru/tags/pull-error-2017-04-24:
  error: Apply error_propagate_null.cocci again
  qga: Make errp the last parameter of qga_vss_fsfreeze
  migration: Make errp the last parameter of local functions
  scsi: Make errp the last parameter of virtio_scsi_common_realize
  fdc: Make errp the last parameter of fdctrl_connect_drives
  nfs: Make errp the last parameter of nfs_client_open
  block: Make errp the last parameter of commit_active_start
  mirror: Make errp the last parameter of mirror_start_job
  crypto: Make errp the last parameter of functions
  block: Make errp the last parameter of bdrv_img_create
  socket: Make errp the last parameter of vsock_connect_saddr
  socket: Make errp the last parameter of unix_connect_saddr
  socket: Make errp the last parameter of inet_connect_saddr
  socket: Make errp the last parameter of socket_connect
  util/error: Fix leak in error_vprepend()

Signed-off-by: Peter Maydell <[email protected]>

ppc: Add SM501 device in ppc softmmu targets default configs

This is not used by default on any emulated machine yet but it is
still useful to have it compiled so it can be added from the command
line for clients that can use it (e.g. MorphOS has no driver for any
other emulated video cards but can output via SM501)

Signed-off-by: BALATON Zoltan <[email protected]>
Acked-by: David Gibson <[email protected]>
Tested-by: Aurelien Jarno <[email protected]>
Message-id: bf305f36dcde152668cf12438ef983cd09ed2d3f.1492787889[email protected]
Signed-off-by: Peter Maydell <[email protected]>

sm501: Add vmstate descriptor

Signed-off-by: BALATON Zoltan <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Tested-by: Aurelien Jarno <[email protected]>
Message-id: 86803c6f40cd678b61b3b1a1429683f60f0aa89a.1492787889[email protected]
Signed-off-by: Peter Maydell <[email protected]>

sm501: Add some more missing registers

This is to allow clients to initialise these without failing as long
as no 2D engine function is called that would use the written value.
Saved values are not used yet (may get used when more of 2D engine is
added sometimes) and clients normally only write to most of these
registers, nothing is known to ever read them but they are documented
as read/write so also implement read for these.

Signed-off-by: BALATON Zoltan <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Tested-by: Aurelien Jarno <[email protected]>
Message-id: 80adf8e4d084ec6cc30d149f8e8215debb67314a.1492787889[email protected]
Signed-off-by: Peter Maydell <[email protected]>

sm501: Add support for panel layer

Signed-off-by: BALATON Zoltan <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Tested-by: Aurelien Jarno <[email protected]>
Message-id: 2029a276362c0c3a14c78acb56baa9466848dd51.1492787889[email protected]
Signed-off-by: Peter Maydell <[email protected]>

sm501: Misc clean ups

- Rename a variable
- Move variable declarations out of loop to the beginning in draw_hwc_line

Signed-off-by: BALATON Zoltan <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Tested-by: Aurelien Jarno <[email protected]>
Message-id: 187c9e4e09d9bc2967b2454b36bb088ceef0b8bc.1492787889[email protected]
Signed-off-by: Peter Maydell <[email protected]>

sm501: Fix hardware cursor

Rework HWC handling to simplify it and fix cursor not updating on
screen as needed. Previously cursor was not updated because checking
for changes in a line overrode the update flag set for the cursor but
fixing this is not enough because the cursor should also be updated if
its shape or location changes. Introduce hwc_invalidate() function to
handle that similar to other display controller models.

Signed-off-by: BALATON Zoltan <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Tested-by: Aurelien Jarno <[email protected]>
Message-id: 6970a5e9868b7246656c1d02038dc5d5fa369507.1492787889[email protected]
Signed-off-by: Peter Maydell <[email protected]>

sm501: Fix device endianness

We only emulate the sysbus device in its default LE mode and PCI is LE
as well so specify this for registers and framebuffer memory.

Note that though the Linux kernel driver has code which claims to
handle both big and little endian, it is obviously bogus for 16 bit
and cannot be trusted as a source of information on the framebuffer
pixel format. This is our best guess about device behaviour based on
the specs and testing with MorphOS that is known to work on real HW.

Signed-off-by: BALATON Zoltan <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Tested-by: Aurelien Jarno <[email protected]>
Message-id: 8b9605a569f8bf54074e15903620b18cd9967c89.1492787889[email protected]
Signed-off-by: Peter Maydell <[email protected]>

sm501: Add emulation of chip connected via PCI

Only the display controller part is created automatically on PCI

Signed-off-by: BALATON Zoltan <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Tested-by: Aurelien Jarno <[email protected]>
Message-id: 647d292c6f5abba8b2a614687229949b5dcb864e.1492787889[email protected]
Signed-off-by: Peter Maydell <[email protected]>

sm501: Get rid of base address in draw_hwc_line

Do not use the base address to access data in local memory. This is in
preparation to allow chip connected via PCI where base address depends
on where the BAR is mapped so it will be unknown.

Signed-off-by: BALATON Zoltan <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Tested-by: Aurelien Jarno <[email protected]>
Message-id: 79dab21bc6ec4d563aabf265c3bab40e2e95aae8.1492787889[email protected]
Signed-off-by: Peter Maydell <[email protected]>

sm501: QOMify

Adding vmstate saving is not in this patch because the state structure
will be changed in further patches, then another patch will add
vmstate descriptor after those changes.

Signed-off-by: BALATON Zoltan <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Tested-by: Aurelien Jarno <[email protected]>
Message-id: a32b7fc981a20205f96d530d8e958f12ace1104c.1492787889[email protected]
Signed-off-by: Peter Maydell <[email protected]>

sm501: Add missing arbitration control register

Signed-off-by: BALATON Zoltan <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Tested-by: Aurelien Jarno <[email protected]>
Message-id: d1eaf3b19c40aeb32a343a211f2b56664a67f948.1492787889[email protected]
Signed-off-by: Peter Maydell <[email protected]>

sm501: Use defined constants instead of literal values where available

Signed-off-by: BALATON Zoltan <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Tested-by: Aurelien Jarno <[email protected]>
Message-id: 31205c2df623e7b133ef942ff4f5e95fff800a14.1492787889[email protected]
Signed-off-by: Peter Maydell <[email protected]>

sm501: Fixed code style and a few typos in comments

Signed-off-by: BALATON Zoltan <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Tested-by: Aurelien Jarno <[email protected]>
Message-id: 36288b703e7d56822c818567193ff28cdc47377e.1492787889[email protected]
Signed-off-by: Peter Maydell <[email protected]>

Merge remote-tracking branch 'remotes/mcayland/tags/qemu-openbios-signed' into staging

Update OpenBIOS images

# gpg: Signature made Sat 22 Apr 2017 07:36:01 BST
# gpg:                using RSA key 0x5BC2C56FAE0F321F
# gpg: Good signature from "Mark Cave-Ayland <[email protected]>"
# Primary key fingerprint: CC62 1AB9 8E82 200D 915C  C9C4 5BC2 C56F AE0F 321F

* remotes/mcayland/tags/qemu-openbios-signed:
  Update OpenBIOS images to 04898e8 built from submodule.

Signed-off-by: Peter Maydell <[email protected]>

Merge remote-tracking branch 'remotes/mcayland/tags/qemu-sparc-signed' into staging

qemu-sparc update

# gpg: Signature made Fri 21 Apr 2017 20:09:35 BST
# gpg:                using RSA key 0x5BC2C56FAE0F321F
# gpg: Good signature from "Mark Cave-Ayland <[email protected]>"
# Primary key fingerprint: CC62 1AB9 8E82 200D 915C  C9C4 5BC2 C56F AE0F 321F

* remotes/mcayland/tags/qemu-sparc-signed:
  tcx: switch to load_image_mr() and remove prom_addr hack
  tcx: use tcx_set_dirty() for accelerated ops
  tcx: remove primitives for non-32-bit surfaces
  tcx: remove TARGET_PAGE_SIZE from tcx24_update_display()
  tcx: remove TARGET_PAGE_SIZE from tcx_update_display()
  tcx: remove page24 and cpage from tcx24_update_display()
  tcx: alter tcx24_reset_dirty() to accept address and length parameters
  tcx: alter tcx24_check_dirty() to accept address and length parameters
  tcx: ensure tcx_set_dirty() also invalidates the 24-bit plane and cplane
  tcx: alter tcx_set_dirty() to accept address and length parameters
  cg3: switch to load_image_mr() and remove prom-addr hack
  cg3: fix up size parameter for memory_region_get_dirty()
  cg3: remove TARGET_PAGE_SIZE rounding on dirty page detection

Signed-off-by: Peter Maydell <[email protected]>

error: Apply error_propagate_null.cocci again

Signed-off-by: Fam Zheng <[email protected]>
Message-Id: <20170421122710[email protected]>
Reviewed-by: Markus Armbruster <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Signed-off-by: Markus Armbruster <[email protected]>

qga: Make errp the last parameter of qga_vss_fsfreeze

Signed-off-by: Fam Zheng <[email protected]>
Message-Id: <20170421122710[email protected]>
Reviewed-by: Markus Armbruster <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Signed-off-by: Markus Armbruster <[email protected]>

migration: Make errp the last parameter of local functions

Signed-off-by: Fam Zheng <[email protected]>
Message-Id: <20170421122710[email protected]>
Reviewed-by: Markus Armbruster <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Signed-off-by: Markus Armbruster <[email protected]>

scsi: Make errp the last parameter of virtio_scsi_common_realize

Signed-off-by: Fam Zheng <[email protected]>
Message-Id: <20170421122710[email protected]>
Reviewed-by: Markus Armbruster <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Signed-off-by: Markus Armbruster <[email protected]>

fdc: Make errp the last parameter of fdctrl_connect_drives

Signed-off-by: Fam Zheng <[email protected]>
Message-Id: <20170421122710[email protected]>
Reviewed-by: Markus Armbruster <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Signed-off-by: Markus Armbruster <[email protected]>

nfs: Make errp the last parameter of nfs_client_open

Signed-off-by: Fam Zheng <[email protected]>
Message-Id: <20170421122710[email protected]>
Reviewed-by: Markus Armbruster <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Signed-off-by: Markus Armbruster <[email protected]>

block: Make errp the last parameter of commit_active_start

Signed-off-by: Fam Zheng <[email protected]>
Message-Id: <20170421122710[email protected]>
Reviewed-by: Markus Armbruster <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Signed-off-by: Markus Armbruster <[email protected]>

mirror: Make errp the last parameter of mirror_start_job

Signed-off-by: Fam Zheng <[email protected]>
Message-Id: <20170421122710[email protected]>
Reviewed-by: Markus Armbruster <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Signed-off-by: Markus Armbruster <[email protected]>

crypto: Make errp the last parameter of functions

Move opaque to 2nd instead of the 2nd to last, so that compilers help
check with the conversion.

Signed-off-by: Fam Zheng <[email protected]>
Message-Id: <20170421122710[email protected]>
Reviewed-by: Markus Armbruster <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
[Commit message typo corrected]
Signed-off-by: Markus Armbruster <[email protected]>

block: Make errp the last parameter of bdrv_img_create

Signed-off-by: Fam Zheng <[email protected]>
Message-Id: <20170421122710[email protected]>
Reviewed-by: Markus Armbruster <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Signed-off-by: Markus Armbruster <[email protected]>

socket: Make errp the last parameter of vsock_connect_saddr

Signed-off-by: Fam Zheng <[email protected]>
Message-Id: <20170421122710[email protected]>
Reviewed-by: Markus Armbruster <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Signed-off-by: Markus Armbruster <[email protected]>

socket: Make errp the last parameter of unix_connect_saddr

Signed-off-by: Fam Zheng <[email protected]>
Message-Id: <20170421122710[email protected]>
Reviewed-by: Markus Armbruster <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Signed-off-by: Markus Armbruster <[email protected]>

socket: Make errp the last parameter of inet_connect_saddr

Signed-off-by: Fam Zheng <[email protected]>
Message-Id: <20170421122710[email protected]>
Reviewed-by: Markus Armbruster <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Signed-off-by: Markus Armbruster <[email protected]>

socket: Make errp the last parameter of socket_connect

Signed-off-by: Fam Zheng <[email protected]>
Message-Id: <20170421122710[email protected]>
Reviewed-by: Markus Armbruster <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Signed-off-by: Markus Armbruster <[email protected]>

util/error: Fix leak in error_vprepend()

Signed-off-by: Max Reitz <[email protected]>
Message-Id: <20170413160952[email protected]>
Reviewed-by: Fam Zheng <[email protected]>
Reviewed-by: Jeff Cody <[email protected]>
Signed-off-by: Markus Armbruster <[email protected]>
Reviewed-by: Markus Armbruster <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Signed-off-by: Markus Armbruster <[email protected]>

Update OpenBIOS images to 04898e8 built from submodule.

Signed-off-by: Mark Cave-Ayland <[email protected]>

Merge remote-tracking branch 'remotes/juanquintela/tags/migration/20170421' into staging

migration/next for 20170421

# gpg: Signature made Fri 21 Apr 2017 11:28:13 BST
# gpg:                using RSA key 0xF487EF185872D723
# gpg: Good signature from "Juan Quintela <[email protected]>"
# gpg:                 aka "Juan Quintela <[email protected]>"
# Primary key fingerprint: 1899 FF8E DEBF 58CC EE03  4B82 F487 EF18 5872 D723

* remotes/juanquintela/tags/migration/20170421: (65 commits)
  hmp: info migrate_parameters format tunes
  hmp: info migrate_capability format tunes
  migration: rename max_size to threshold_size
  migration: set current_active_state once
  virtio-rng: stop virtqueue while the CPU is stopped
  migration: don't close a file descriptor while it can be in use
  ram: Remove migration_bitmap_extend()
  migration: Disable hotplug/unplug during migration
  qdev: Move qdev_unplug() to qdev-monitor.c
  qdev: Export qdev_hot_removed
  qdev: qdev_hotplug is really a bool
  migration: Remove MigrationState parameter from migration_is_idle()
  ram: Use RAMBitmap type for coherence
  ram: rename last_ram_offset() last_ram_pages()
  ram: Use ramblock and page offset instead of absolute offset
  ram: Change offset field in PageSearchStatus to page
  ram: Remember last_page instead of last_offset
  ram: Use page number instead of an address for the bitmap operations
  ram: reorganize last_sent_block
  ram: ram_discard_range() don't use the mis parameter
  ...

Signed-off-by: Peter Maydell <[email protected]>

Merge remote-tracking branch 'remotes/stefanha/tags/tracing-pull-request' into staging

# gpg: Signature made Fri 21 Apr 2017 10:52:18 BST
# gpg:                using RSA key 0x9CA4ABB381AB73C8
# gpg: Good signature from "Stefan Hajnoczi <[email protected]>"
# gpg:                 aka "Stefan Hajnoczi <[email protected]>"
# Primary key fingerprint: 8695 A8BF D3F9 7CDA AC35  775A 9CA4 ABB3 81AB 73C8

* remotes/stefanha/tags/tracing-pull-request:
  simpletrace: document Analyzer method signatures
  trace: Put all trace.o into libqemuutil.a
  configure: eliminate Python dependency for --help

Signed-off-by: Peter Maydell <[email protected]>

Merge remote-tracking branch 'remotes/stefanha/tags/block-pull-request' into staging

# gpg: Signature made Fri 21 Apr 2017 10:43:04 BST
# gpg:                using RSA key 0x9CA4ABB381AB73C8
# gpg: Good signature from "Stefan Hajnoczi <[email protected]>"
# gpg:                 aka "Stefan Hajnoczi <[email protected]>"
# Primary key fingerprint: 8695 A8BF D3F9 7CDA AC35  775A 9CA4 ABB3 81AB 73C8

* remotes/stefanha/tags/block-pull-request:
  MAINTAINERS: update my email address
  MAINTAINERS: update Wen's email address
  migration/block: use blk_pwrite_zeroes for each zero cluster
  throttle: make throttle_config(throttle_get_config()) symmetric
  throttle: do not use invalid config in test
  qemu-options: explain disk I/O throttling options

Signed-off-by: Peter Maydell <[email protected]>

Merge remote-tracking branch 'remotes/cohuck/tags/s390x-20170421' into staging

The first batch of s390x changes for 2.10:
- the new compat machine
- several cleanups and optimizations
- introspection for css ids

# gpg: Signature made Fri 21 Apr 2017 08:36:25 BST
# gpg:                using RSA key 0xDECF6B93C6F02FAF
# gpg: Good signature from "Cornelia Huck <[email protected]>"
# gpg:                 aka "Cornelia Huck <[email protected]>"
# Primary key fingerprint: C3D0 D66D C362 4FF6 A8C0  18CE DECF 6B93 C6F0 2FAF

* remotes/cohuck/tags/s390x-20170421:
  s390x: Drop useless casts
  s390x: register I/O adapters per ISC during init
  s390x/flic: cache flic in s390_get_flic
  s390x: initialize flic before I/O subsystems
  s390x: use enum for adapter type and standardize its naming
  s390x/css: consolidate the devno property for ccw devices
  s390x/css: provide introspection for virtual subchannel and device busid
  s390x/css: introduce read-only property type for device ids
  s390x/pci: make printf always compile in debug output
  s390x/kvm: make printf always compile in debug output
  s390x: introduce 2.10 compat machine

Signed-off-by: Peter Maydell <[email protected]>

Merge remote-tracking branch 'remotes/armbru/tags/pull-qdev-2017-04-21' into staging

qdev patches for 2017-04-21

# gpg: Signature made Fri 21 Apr 2017 06:37:19 BST
# gpg:                using RSA key 0x3870B400EB918653
# gpg: Good signature from "Markus Armbruster <[email protected]>"
# gpg:                 aka "Markus Armbruster <[email protected]>"
# Primary key fingerprint: 354B C8B3 D7EB 2A6B 6867  4E5F 3870 B400 EB91 8653

* remotes/armbru/tags/pull-qdev-2017-04-21:
  qdev: remove cannot_destroy_with_object_finalize_yet
  versatile: remove cannot_destroy_with_object_finalize_yet
  ppc: remove cannot_destroy_with_object_finalize_yet
  arm: remove remaining cannot_destroy_with_object_finalize_yet

Signed-off-by: Peter Maydell <[email protected]>

hmp: info migrate_parameters format tunes

Do the same (one per line) to the parameter list.

CC: "Dr. David Alan Gilbert" <[email protected]>
Signed-off-by: Peter Xu <[email protected]>
Reviewed-by: "Dr. David Alan Gilbert" <[email protected]>
Reviewed-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Juan Quintela <[email protected]>
Signed-off-by: Juan Quintela <[email protected]>

hmp: info migrate_capability format tunes

Dump the info in a single line is hard to read. Do it one per line.
Also, the first "capabilities:" didn't help much. Let's remove it.

CC: "Dr. David Alan Gilbert" <[email protected]>
Signed-off-by: Peter Xu <[email protected]>
Reviewed-by: "Dr. David Alan Gilbert" <[email protected]>
Reviewed-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Juan Quintela <[email protected]>
Signed-off-by: Juan Quintela <[email protected]>

migration: rename max_size to threshold_size

In migration codes (especially in migration_thread()), max_size is used
in many place for the threshold value that we will start to do the final
flush and jump to the next stage to dump the whole rest things to
destination. However its name is confusing to first readers. Let's
rename it to "threshold_size" when proper and add a comment for it. No
functional change is made.

CC: Juan Quintela <[email protected]>
CC: "Dr. David Alan Gilbert" <[email protected]>
Signed-off-by: Peter Xu <[email protected]>
Reviewed-by: "Dr. David Alan Gilbert" <[email protected]>
Reviewed-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Juan Quintela <[email protected]>
Signed-off-by: Juan Quintela <[email protected]>

migration: set current_active_state once

We set it right above this one. No need to set it twice.

CC: Juan Quintela <[email protected]>
CC: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Signed-off-by: Peter Xu <[email protected]>
Reviewed-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Juan Quintela <[email protected]>
Signed-off-by: Juan Quintela <[email protected]>

virtio-rng: stop virtqueue while the CPU is stopped

If we modify the virtio-rng virqueue while the
vmstate is already migrated we can have some
inconsistencies between the virtqueue state and
the memory content.

To avoid this, stop the virtqueue while the CPU
is stopped.

Signed-off-by: Laurent Vivier <[email protected]>
Reviewed-by: Amit Shah <[email protected]>
Reviewed-by: Stefan Hajnoczi <[email protected]>
Reviewed-by: Juan Quintela <[email protected]>
Signed-off-by: Juan Quintela <[email protected]>

migration: don't close a file descriptor while it can be in use

If we close the QEMUFile descriptor in process_incoming_migration_co()
while it has been stopped by an error, the postcopy_ram_listen_thread()
can try to continue to use it. And as the memory has been freed
it is working with an invalid pointer and crashes.

Fix this by releasing the memory after having managed the error
case (which, in fact, calls exit())

Signed-off-by: Laurent Vivier <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Juan Quintela <[email protected]>
Reviewed-by: Amit Shah <[email protected]>
Reviewed-by: Stefan Hajnoczi <[email protected]>
Signed-off-by: Juan Quintela <[email protected]>

ram: Remove migration_bitmap_extend()

We have disabled memory hotplug, so we don't need to handle
migration_bitamp there.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Reviewed-by: zhanghailiang <[email protected]>

migration: Disable hotplug/unplug during migration

Until we have reviewed what can/can't be hotplugged during migration,
disable it.  We can enable it later for the things that we know that
work.  For instance, memory hotplug during postcopy doesn't work
currently.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: zhanghailiang <[email protected]>
--

- Fix typo.  Thanks Thomas.
- Delay migration check after we have checked that we can hotplug that
  device.
- more typos

qdev: Move qdev_unplug() to qdev-monitor.c

It is not used by linux-user, otherwise I need to to create one stub
for migration_is_idle() on following patch.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Reviewed-by: zhanghailiang <[email protected]>
Reviewed-by: Markus Armbruster <[email protected]>

qdev: Export qdev_hot_removed

I need to move qdev_unplug to qdev-monitor in the following patch, and
it needs access to this variable.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Reviewed-by: zhanghailiang <[email protected]>

qdev: qdev_hotplug is really a bool

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: zhanghailiang <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Reviewed-by: Philippe Mathieu-Daudé <[email protected]>

migration: Remove MigrationState parameter from migration_is_idle()

Only user don't have a MigrationState handly.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>

ram: Use RAMBitmap type for coherence

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>

ram: rename last_ram_offset() last_ram_pages()

We always use it as pages anyways.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>

ram: Use ramblock and page offset instead of absolute offset

This removes the needto pass also the absolute offset.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>

ram: Change offset field in PageSearchStatus to page

We are moving everything to work on pages, not addresses.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>

ram: Remember last_page instead of last_offset

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Philippe Mathieu-Daudé <[email protected]>
--

Improve comment
Fix typo

ram: Use page number instead of an address for the bitmap operations

We use an unsigned long for the page number.  Notice that our bitmaps
already got that for the index, so we have that limit.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
--

rename page to page_abs everywhere.
fix trace types for pages

ram: reorganize last_sent_block

We were setting it far away of when we changed it.  Now everything is
done inside save_page_header.  Once there, reorganize code to pass
RAMState.  We also set CONTINUE flag in a single place.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: ram_discard_range() don't use the mis parameter

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Pass RAMBlock to bitmap_sync

We change the meaning of start to be the offset from the beggining of
the block.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>

ram: Add page-size to output in 'info migrate'

The number of dirty pages is output in 'pages' in the command
'info migrate', so add page-size to calculate the number of dirty
pages in bytes.

Signed-off-by: Chao Fan <[email protected]>
Signed-off-by: Li Zhijian <[email protected]>
Reviewed-by: Eric Blake <[email protected]>
Signed-off-by: Juan Quintela <[email protected]>

ram: Rename qemu_target_page_bits() to qemu_target_page_size()

It was used as a size in all cases except one.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>

ram: We don't need MigrationState parameter anymore

Remove it from callers and callees.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

migration: Remove MigrationState from migration_in_postcopy

We need to call for the migrate_get_current() in more that half of the
uses, so call that inside.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Remove compression_switch and inline its logic

We can calculate its value, so we don't create a variable for it.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
--

After Peter and Dave review, I dropped the variable and just inlined
the condition.

Fix typo

ram: Move QEMUFile into RAMState

We receive the file from save_live operations and we don't use it
until 3 or 4 levels of calls down.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>

ram: Add QEMUFile to RAMState

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>

ram: Move postcopy_requests into RAMState

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Move dirty_pages_rate to RAMState

Treat it like the rest of ram stats counters.  Export its value the
same way.  As an added bonus, no more MigrationState used in
migration_bitmap_sync();

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>
--

Again, dave was the one reviewing it

ram: Remove dirty_bytes_rate

It can be recalculated from dirty_pages_rate.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>
--

Dave was the one that reviewed it O:-)

ram: Create ram_dirty_sync_count()

This is a ram field that was inside MigrationState.  Move it to
RAMState and make it the same that the other ram stats.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Move src_page_req* to RAMState

This are the last postcopy fields still at MigrationState.  Once there
Move MigrationSrcPageRequest to ram.c and remove MigrationState
parameters where appropiate.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Move last_req_rb to RAMState

It was on MigrationState when it is only used inside ram.c for
postcopy.  Problem is that we need to access it without being able to
pass it RAMState directly.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Remove ram_save_remaining

Just unfold it.  Move ram_bytes_remaining() with the rest of exported
functions.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Use the RAMState bytes_transferred parameter

Somewhere it was passed by reference, just use it from RAMState.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Move bytes_transferred into RAMState

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Move migration_bitmap_rcu into RAMState

Once there, rename the type to be shorter.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Move migration_bitmap_mutex into RAMState

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Everything was init to zero, so use memset

And then init only things that are not zero by default.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Move migration_dirty_pages to RAMState

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Move xbzrle_overflows into RAMState

Once there, remove the now unused AccountingInfo struct and var.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Move xbzrle_cache_miss_rate into RAMState

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Move xbzrle_cache_miss into RAMState

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Move xbzrle_pages into RAMState

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>
--

Comment why we need bytes and pages

ram: Move xbzrle_bytes into RAMState

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>

ram: Move iterations into RAMState

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Remove norm_mig_bytes_transferred

Its value can be calculated by other exported.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>

ram: Move norm_pages to RAMState

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Remove unused pages_skipped variable

For compatibility, we need to still send a value, but just specify it
and comment the fact.

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>

ram: Remove unused dup_mig_bytes_transferred()

Signed-off-by: Juan Quintela <[email protected]>
Reviewed-by: Dr. David Alan Gilbert <[email protected]>
Reviewed-by: Peter Xu <[email protected]>