sdk/emulator/qemu.git
10 years agoblock/archipelago: Add support for creating images
Chrysostomos Nanakos [Wed, 23 Jul 2014 14:07:31 +0000 (17:07 +0300)]
block/archipelago: Add support for creating images

qemu-img archipelago:<volumename>[/mport=<mapperd_port>[:vport=<vlmcd_port>]
 [:segment=<segment_name>]] [size]

Signed-off-by: Chrysostomos Nanakos <cnanakos@grnet.gr>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agoblock/archipelago: Implement bdrv_parse_filename()
Chrysostomos Nanakos [Wed, 23 Jul 2014 14:07:30 +0000 (17:07 +0300)]
block/archipelago: Implement bdrv_parse_filename()

VM Image on Archipelago volume can also be specified like this:

file=archipelago:<volumename>[/mport=<mapperd_port>[:vport=<vlmcd_port>][:
segment=<segment_name>]]

Examples:

file=archipelago:my_vm_volume
file=archipelago:my_vm_volume/mport=123
file=archipelago:my_vm_volume/mport=123:vport=1234
file=archipelago:my_vm_volume/mport=123:vport=1234:segment=my_segment

Signed-off-by: Chrysostomos Nanakos <cnanakos@grnet.gr>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agoblock: Support Archipelago as a QEMU block backend
Chrysostomos Nanakos [Mon, 4 Aug 2014 14:35:32 +0000 (17:35 +0300)]
block: Support Archipelago as a QEMU block backend

VM Image on Archipelago volume is specified like this:

file.driver=archipelago,file.volume=<volumename>[,file.mport=<mapperd_port>[,
file.vport=<vlmcd_port>][,file.segment=<segment_name>]]

'archipelago' is the protocol.

'mport' is the port number on which mapperd is listening. This is optional
and if not specified, QEMU will make Archipelago to use the default port.

'vport' is the port number on which vlmcd is listening. This is optional
and if not specified, QEMU will make Archipelago to use the default port.

'segment' is the name of the shared memory segment Archipelago stack is using.
This is optional and if not specified, QEMU will make Archipelago to use the
default value, 'archipelago'.

Examples:

file.driver=archipelago,file.volume=my_vm_volume
file.driver=archipelago,file.volume=my_vm_volume,file.mport=123
file.driver=archipelago,file.volume=my_vm_volume,file.mport=123,
file.vport=1234
file.driver=archipelago,file.volume=my_vm_volume,file.mport=123,
file.vport=1234,file.segment=my_segment

Signed-off-by: Chrysostomos Nanakos <cnanakos@grnet.gr>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
10 years agoqemu-img info: show nocow info
Chunyan Liu [Wed, 30 Jul 2014 02:55:06 +0000 (10:55 +0800)]
qemu-img info: show nocow info

Add nocow info in 'qemu-img info' output to show whether the file
currently has NOCOW flag set or not.

Signed-off-by: Chunyan Liu <cyliu@suse.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agovmdk: Optimize cluster allocation
Fam Zheng [Wed, 30 Jul 2014 06:39:10 +0000 (14:39 +0800)]
vmdk: Optimize cluster allocation

This drops the unnecessary bdrv_truncate() from, and also improves,
cluster allocation code path.

Before, when we need a new cluster, get_cluster_offset truncates the
image to bdrv_getlength() + cluster_size, and returns the offset of
added area, i.e. the image length before truncating.

This is not efficient, so it's now rewritten as:

  - Save the extent file length when opening.

  - When allocating cluster, use the saved length as cluster offset.

  - Don't truncate image, because we'll anyway write data there: just
    write any data at the EOF position, in descending priority:

    * New user data (cluster allocation happens in a write request).

    * Filling data in the beginning and/or ending of the new cluster, if
      not covered by user data: either backing file content (COW), or
      zero for standalone images.

One major benifit of this change is, on host mounted NFS images, even
over a fast network, ftruncate is slow (see the example below). This
change significantly speeds up cluster allocation. Comparing by
converting a cirros image (296M) to VMDK on an NFS mount point, over
1Gbe LAN:

    $ time qemu-img convert cirros-0.3.1.img /mnt/a.raw -O vmdk

    Before:
        real    0m21.796s
        user    0m0.130s
        sys     0m0.483s

    After:
        real    0m2.017s
        user    0m0.047s
        sys     0m0.190s

We also get rid of unchecked bdrv_getlength() and bdrv_truncate(), and
get a little more documentation in function comments.

Tested that this passes qemu-iotests for all VMDK subformats.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agoqemu-iotests: Add data pattern in version3 VMDK sample image in 059
Fam Zheng [Wed, 30 Jul 2014 06:39:09 +0000 (14:39 +0800)]
qemu-iotests: Add data pattern in version3 VMDK sample image in 059

It's possible that we diverge from the specification with our
implementation.  Having a reference image in the test cases may detect
such problems when we introduce a bug that can read what it creates, but
can't handle a real VMDK.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agoqdev-monitor: include QOM properties in -device FOO, help output
Stefan Hajnoczi [Wed, 9 Jul 2014 12:01:32 +0000 (14:01 +0200)]
qdev-monitor: include QOM properties in -device FOO, help output

Update -device FOO,help to include QOM properties in addition to qdev
properties.  Devices are gradually adding more QOM properties that are
not reflected as qdev properties.

It is important to report all device properties since management tools
like libvirt use this information (and device-list-properties QMP) to
detect the presence of QEMU features.

This patch reuses the device-list-properties QMP machinery to avoid code
duplication.

Reported-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Tested-by: Cole Robinson <crobinso@redhat.com>
10 years agoqmp: hide "hotplugged" device property from device-list-properties
Stefan Hajnoczi [Wed, 9 Jul 2014 12:01:31 +0000 (14:01 +0200)]
qmp: hide "hotplugged" device property from device-list-properties

The "hotplugged" device property was not reported before commit
f4eb32b590bf58c1c67570775eb78beb09964fad ("qmp: show QOM properties in
device-list-properties").  Fix this difference.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
10 years agodocs/multiple-iothreads.txt: add documentation on IOThread programming
Stefan Hajnoczi [Wed, 23 Jul 2014 11:55:32 +0000 (12:55 +0100)]
docs/multiple-iothreads.txt: add documentation on IOThread programming

This document explains how IOThreads and the main loop are related,
especially how to write code that can run in an IOThread.  Currently
only virtio-blk-data-plane uses these techniques.  The next obvious
target is virtio-scsi; there has also been work on virtio-net.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
10 years agoxen_disk: fix possible null-ptr dereference
Gonglei (Arei) [Mon, 28 Jul 2014 06:03:45 +0000 (06:03 +0000)]
xen_disk:  fix possible null-ptr dereference

Signed-off-by: Gonglei <arei.gonglei@huawei.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agoconfigure: explicitly state version requirements to devel packages
Hu Tao [Thu, 26 Jun 2014 09:34:50 +0000 (17:34 +0800)]
configure: explicitly state version requirements to devel packages

Signed-off-by: Hu Tao <hutao@cn.fujitsu.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agodocs: Make the recommendation for the backing file name position a requirement
Maria Kustova [Mon, 21 Jul 2014 11:16:33 +0000 (15:16 +0400)]
docs: Make the recommendation for the backing file name position a requirement

The current version of the qcow2 specification recommends to save the backing
file name in the end of the first cluster. It follows that the backing file
name can be saved somewhere in the image, but the first cluster, which
contradicts the current QEMU implementation.

The patch makes the backing file name required to be placed after the header
extensions in the first image cluster.

Signed-off-by: Maria Kustova <maria.k@catit.be>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
10 years agoblock: Avoid bdrv_get_geometry() where errors should be detected
Markus Armbruster [Thu, 26 Jun 2014 11:23:25 +0000 (13:23 +0200)]
block: Avoid bdrv_get_geometry() where errors should be detected

bdrv_get_geometry() hides errors.  Use bdrv_nb_sectors() or
bdrv_getlength() instead where that's obviously inappropriate.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Benoit Canet <benoit@irqsave.net>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agoqemu-img: Make img_convert() get image size just once per image
Markus Armbruster [Thu, 26 Jun 2014 11:23:24 +0000 (13:23 +0200)]
qemu-img: Make img_convert() get image size just once per image

Chiefly so I don't have to do the error checking in quadruplicate in
the next commit.  Moreover, replacing the frequently updated
bs_sectors by an array assigned just once makes the code easier to
understand.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Benoit Canet <benoit@irqsave.net>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agoblock: Drop superfluous aligning of bdrv_getlength()'s value
Markus Armbruster [Thu, 26 Jun 2014 11:23:23 +0000 (13:23 +0200)]
block: Drop superfluous aligning of bdrv_getlength()'s value

It returns a multiple of the sector size.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Benoit Canet <benoit@irqsave.net>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agoblock: Use bdrv_nb_sectors() where sectors, not bytes are wanted
Markus Armbruster [Thu, 26 Jun 2014 11:23:22 +0000 (13:23 +0200)]
block: Use bdrv_nb_sectors() where sectors, not bytes are wanted

Instead of bdrv_getlength().

Aside: a few of these callers don't handle errors.  I didn't
investigate whether they should.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Benoit Canet <benoit@irqsave.net>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agoblock: Use bdrv_nb_sectors() in img_convert()
Markus Armbruster [Thu, 26 Jun 2014 11:23:21 +0000 (13:23 +0200)]
block: Use bdrv_nb_sectors() in img_convert()

Instead of bdrv_getlength().  Replace variable output_length by
output_sectors.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Benoit Canet <benoit@irqsave.net>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agoblock: Use bdrv_nb_sectors() in bdrv_co_get_block_status()
Markus Armbruster [Thu, 26 Jun 2014 11:23:20 +0000 (13:23 +0200)]
block: Use bdrv_nb_sectors() in bdrv_co_get_block_status()

Instead of bdrv_getlength().

Replace variables length, length2 by total_sectors, nb_sectors2.
Bonus: use total_sectors instead of the slightly unclean
bs->total_sectors.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Benoit Canet <benoit@irqsave.net>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agoblock: Use bdrv_nb_sectors() in bdrv_aligned_preadv()
Markus Armbruster [Thu, 26 Jun 2014 11:23:19 +0000 (13:23 +0200)]
block: Use bdrv_nb_sectors() in bdrv_aligned_preadv()

Instead of bdrv_getlength().  Eliminate variable len.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Benoit Canet <benoit@irqsave.net>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agoblock: Use bdrv_nb_sectors() in bdrv_make_zero()
Markus Armbruster [Thu, 26 Jun 2014 11:23:18 +0000 (13:23 +0200)]
block: Use bdrv_nb_sectors() in bdrv_make_zero()

Instead of bdrv_getlength().

Variable target_size is initially in bytes, then changes meaning to
sectors.  Ugh.  Replace by target_sectors.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Benoit Canet <benoit@irqsave.net>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agoblock: New bdrv_nb_sectors()
Markus Armbruster [Thu, 26 Jun 2014 11:23:17 +0000 (13:23 +0200)]
block: New bdrv_nb_sectors()

A call to retrieve the image size converts between bytes and sectors
several times:

* BlockDriver method bdrv_getlength() returns bytes.

* refresh_total_sectors() converts to sectors, rounding up, and stores
  in total_sectors.

* bdrv_getlength() converts total_sectors back to bytes (now rounded
  up to a multiple of the sector size).

* Callers wanting sectors rather bytes convert it right back.
  Example: bdrv_get_geometry().

bdrv_nb_sectors() provides a way to omit the last two conversions.
It's exactly bdrv_getlength() with the conversion to bytes omitted.
It's functionally like bdrv_get_geometry() without its odd error
handling.

Reimplement bdrv_getlength() and bdrv_get_geometry() on top of
bdrv_nb_sectors().

The next patches will convert some users of bdrv_getlength() to
bdrv_nb_sectors().

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Benoit Canet <benoit@irqsave.net>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agoMerge remote-tracking branch 'remotes/pmaydell/tags/pull-target-arm-20140804' into...
Peter Maydell [Mon, 4 Aug 2014 14:01:38 +0000 (15:01 +0100)]
Merge remote-tracking branch 'remotes/pmaydell/tags/pull-target-arm-20140804' into staging

target-arm queue:
 * Set PC correctly when loading AArch64 ELF files
 * sdhci: Fix ADMA dma_memory_read access
 * some more foundational work for EL2/EL3 support
 * fix bugs which reveal themselves if the TARGET_PAGE_SIZE
   is not set to 1K

# gpg: Signature made Mon 04 Aug 2014 14:51:34 BST using RSA key ID 14360CDE
# gpg: Good signature from "Peter Maydell <peter.maydell@linaro.org>"

* remotes/pmaydell/tags/pull-target-arm-20140804:
  target-arm: A64: fix TLB flush instructions
  target-arm: don't hardcode mask values in arm_cpu_handle_mmu_fault
  target-arm: Fix bit test in sp_el0_access
  target-arm: Add FAR_EL2 and 3
  target-arm: Add ESR_EL2 and 3
  target-arm: Make far_el1 an array
  target-arm: A64: Respect SPSEL when taking exceptions
  target-arm: A64: Respect SPSEL in ERET SP restore
  target-arm: A64: Break out aarch64_save/restore_sp
  sd: sdhci: Fix ADMA dma_memory_read access
  hw/arm/virt: formatting: memory map
  hw/arm/boot: Set PC correctly when loading AArch64 ELF files

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agotarget-arm: A64: fix TLB flush instructions
Alex Bennée [Mon, 4 Aug 2014 13:41:56 +0000 (14:41 +0100)]
target-arm: A64: fix TLB flush instructions

According to the ARM ARM we weren't correctly flushing the TLB entries
where bits 63:56 didn't match bit 55 of the virtual address. This
exposed a problem when we switched QEMU's internal TARGET_PAGE_BITS to
12 for aarch64.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1406733627-24255-3-git-send-email-alex.bennee@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agotarget-arm: don't hardcode mask values in arm_cpu_handle_mmu_fault
Alex Bennée [Mon, 4 Aug 2014 13:41:55 +0000 (14:41 +0100)]
target-arm: don't hardcode mask values in arm_cpu_handle_mmu_fault

Otherwise we break quickly when we change TARGET_PAGE_SIZE.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-id: 1406733627-24255-2-git-send-email-alex.bennee@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agotarget-arm: Fix bit test in sp_el0_access
Stefan Weil [Mon, 4 Aug 2014 13:41:55 +0000 (14:41 +0100)]
target-arm: Fix bit test in sp_el0_access

Static code analyzers complain about a dubious & operation used for a
boolean value. The code does not test the PSTATE_SP bit as it should.

Cc: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Stefan Weil <sw@weilnetz.de>
Message-id: 1406359601-25583-1-git-send-email-sw@weilnetz.de
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agotarget-arm: Add FAR_EL2 and 3
Edgar E. Iglesias [Mon, 4 Aug 2014 13:41:55 +0000 (14:41 +0100)]
target-arm: Add FAR_EL2 and 3

Reviewed-by: Greg Bellows <greg.bellows@linaro.org>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Message-id: 1402994746-8328-7-git-send-email-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agotarget-arm: Add ESR_EL2 and 3
Edgar E. Iglesias [Mon, 4 Aug 2014 13:41:55 +0000 (14:41 +0100)]
target-arm: Add ESR_EL2 and 3

Reviewed-by: Greg Bellows <greg.bellows@linaro.org>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Message-id: 1402994746-8328-6-git-send-email-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agotarget-arm: Make far_el1 an array
Edgar E. Iglesias [Mon, 4 Aug 2014 13:41:54 +0000 (14:41 +0100)]
target-arm: Make far_el1 an array

No functional change.
Prepares for future additions of the EL2 and 3 versions of this reg.

Reviewed-by: Greg Bellows <greg.bellows@linaro.org>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Message-id: 1402994746-8328-5-git-send-email-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agotarget-arm: A64: Respect SPSEL when taking exceptions
Edgar E. Iglesias [Mon, 4 Aug 2014 13:41:54 +0000 (14:41 +0100)]
target-arm: A64: Respect SPSEL when taking exceptions

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Greg Bellows <greg.bellows@linaro.org>
Message-id: 1402994746-8328-4-git-send-email-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agotarget-arm: A64: Respect SPSEL in ERET SP restore
Edgar E. Iglesias [Mon, 4 Aug 2014 13:41:54 +0000 (14:41 +0100)]
target-arm: A64: Respect SPSEL in ERET SP restore

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Greg Bellows <greg.bellows@linaro.org>
Message-id: 1402994746-8328-3-git-send-email-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agotarget-arm: A64: Break out aarch64_save/restore_sp
Edgar E. Iglesias [Mon, 4 Aug 2014 13:41:54 +0000 (14:41 +0100)]
target-arm: A64: Break out aarch64_save/restore_sp

Break out code to save/restore AArch64 SP into functions.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Greg Bellows <greg.bellows@linaro.org>
Message-id: 1402994746-8328-2-git-send-email-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agosd: sdhci: Fix ADMA dma_memory_read access
Peter Crosthwaite [Mon, 4 Aug 2014 13:41:54 +0000 (14:41 +0100)]
sd: sdhci: Fix ADMA dma_memory_read access

This dma_memory_read was giving too big a size when begin was non-zero.
This could cause segfaults in some circumstances. Fix.

Signed-off-by: Peter Crosthwaite <peter.crosthwaite@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agohw/arm/virt: formatting: memory map
Andrew Jones [Mon, 4 Aug 2014 13:41:53 +0000 (14:41 +0100)]
hw/arm/virt: formatting: memory map

Add some spacing and zeros to make it easier to read and
modify the map. This patch has no functional changes. The
review looks ugly, but it's actually pretty easy to confirm
all the addresses are as they should be - thanks to the new
formatting ;-)

Signed-off-by: Andrew Jones <drjones@redhat.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agohw/arm/boot: Set PC correctly when loading AArch64 ELF files
Peter Maydell [Mon, 4 Aug 2014 13:41:53 +0000 (14:41 +0100)]
hw/arm/boot: Set PC correctly when loading AArch64 ELF files

The code in do_cpu_reset() correctly handled AArch64 CPUs
when running Linux kernels, but was missing code in the
branch of the if() that deals with loading ELF files.
Correctly jump to the ELF entry point on reset rather than
leaving the reset PC at zero.

Reported-by: Christopher Covington <cov@codeaurora.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Tested-by: Christopher Covington <cov@codeaurora.org>
Cc: qemu-stable@nongnu.org
10 years agoMerge remote-tracking branch 'remotes/amit-migration/for-2.2' into staging
Peter Maydell [Mon, 4 Aug 2014 13:41:19 +0000 (14:41 +0100)]
Merge remote-tracking branch 'remotes/amit-migration/for-2.2' into staging

* remotes/amit-migration/for-2.2:
  checker: ignore fields marked unused
  vmstate static checker: whitelist additions

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agoMerge remote-tracking branch 'remotes/amit-virtio-rng/for-2.2' into staging
Peter Maydell [Mon, 4 Aug 2014 12:07:02 +0000 (13:07 +0100)]
Merge remote-tracking branch 'remotes/amit-virtio-rng/for-2.2' into staging

* remotes/amit-virtio-rng/for-2.2:
  virtio-rng: replace error_set calls with error_setg
  virtio-rng: Move error-checking forward to prevent memory leak

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agoMerge remote-tracking branch 'remotes/sstabellini/xen-20140801' into staging
Peter Maydell [Mon, 4 Aug 2014 10:17:24 +0000 (11:17 +0100)]
Merge remote-tracking branch 'remotes/sstabellini/xen-20140801' into staging

* remotes/sstabellini/xen-20140801:
  qemu: support xen hvm direct kernel boot
  tap-bsd: implement a FreeBSD only version of tap_open
  xen: fix usage of ENODATA

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agochecker: ignore fields marked unused
Amit Shah [Tue, 22 Jul 2014 07:36:25 +0000 (13:06 +0530)]
checker: ignore fields marked unused

While comparing qemu-1.0 json output with qemu-2.1, a few fields got
marked unused.  These need to be skipped over, and not flagged as
mismatches.

For handling unused fields, the exact number of bytes need to be skipped
over as the size of the unused field.

Currently, only the term "unused" is matched.  When more field names
turn up, this will have to be updated based on the whitelist matching
method to match more such terms.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
10 years agovirtio-rng: replace error_set calls with error_setg
John Snow [Tue, 29 Jul 2014 23:28:58 +0000 (19:28 -0400)]
virtio-rng: replace error_set calls with error_setg

Under recommendation from Luiz Capitulino, we are changing
the error_set calls to error_setg while we are fixing up
the error handling pathways of virtio-rng.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
10 years agovirtio-rng: Move error-checking forward to prevent memory leak
John Snow [Tue, 29 Jul 2014 23:28:57 +0000 (19:28 -0400)]
virtio-rng: Move error-checking forward to prevent memory leak

This patch pushes the error-checking forward and the virtio
initialization backward in the device realization function
in order to prevent memory leaks for hot plug scenarios.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
10 years agoOpen 2.2 development tree
Peter Maydell [Fri, 1 Aug 2014 17:30:08 +0000 (18:30 +0100)]
Open 2.2 development tree

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agoqemu: support xen hvm direct kernel boot
Chunyan Liu [Mon, 7 Jul 2014 06:34:35 +0000 (14:34 +0800)]
qemu: support xen hvm direct kernel boot

qemu side patch to support xen HVM direct kernel boot:
if -kernel exists, calls xen_load_linux(), which will read kernel/initrd
and add a linuxboot.bin or multiboot.bin option rom. The
linuxboot.bin/multiboot.bin will load kernel/initrd and jump to execute
kernel directly. It's working when xen uses seabios.

During this work, found the 'kvmvapic' is in option_rom list, it should
not be there in xen case. Set s->vapic_control = 0 in xen_apic_realize()
to handle that.

Signed-off-by: Chunyan Liu <cyliu@suse.com>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
10 years agotap-bsd: implement a FreeBSD only version of tap_open
Roger Pau Monne [Fri, 23 May 2014 15:57:48 +0000 (17:57 +0200)]
tap-bsd: implement a FreeBSD only version of tap_open

The current behaviour of tap_open for BSD systems differ greatly from
it's Linux counterpart. Since FreeBSD supports interface renaming and
tap device cloning by opening /dev/tap, implement a FreeBSD specific
version of tap_open that behaves like it's Linux counterpart.

This is specially important for toolstacks that use Qemu (like Xen
libxl), in order to have a unified behaviour across suported
platforms.

Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
10 years agoxen: fix usage of ENODATA
Roger Pau Monne [Fri, 23 May 2014 15:57:47 +0000 (17:57 +0200)]
xen: fix usage of ENODATA

ENODATA doesn't exist on FreeBSD, so ENODATA errors returned by the
hypervisor are translated to ENOENT.

Also, the error code is returned in errno if the call returns -1, so
compare the error code with the value in errno instead of the value
returned by the function.

Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Cc: xen-devel@lists.xenproject.org
Cc: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Cc: Anthony Perard <anthony.perard@citrix.com>
10 years agoUpdate version for v2.1.0 release
Peter Maydell [Fri, 1 Aug 2014 12:31:29 +0000 (13:31 +0100)]
Update version for v2.1.0 release

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agoUpdate version for v2.1.0-rc5 release
Peter Maydell [Tue, 29 Jul 2014 17:23:34 +0000 (18:23 +0100)]
Update version for v2.1.0-rc5 release

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agohw/arm/virt: fix pl031 addr typo
Andrew Jones [Tue, 29 Jul 2014 16:32:01 +0000 (18:32 +0200)]
hw/arm/virt: fix pl031 addr typo

pl031's base address should be 0x9010000, not 0x90010000, otherwise
it sits in ram when configuring a guest with greater than 1G.

Signed-off-by: Andrew Jones <drjones@redhat.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agoUpdate version for v2.1.0-rc4 release
Peter Maydell [Tue, 29 Jul 2014 12:45:10 +0000 (13:45 +0100)]
Update version for v2.1.0-rc4 release

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agopo: update Italian translation
Paolo Bonzini [Tue, 29 Jul 2014 06:15:12 +0000 (08:15 +0200)]
po: update Italian translation

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agopo: Update French translation
Aurelien Jarno [Mon, 28 Jul 2014 21:44:46 +0000 (23:44 +0200)]
po: Update French translation

Add new translations for recently added messages.

Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
Reviewed-by: Stefan Weil <sw@weilnetz.de>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agoMerge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging
Peter Maydell [Tue, 29 Jul 2014 11:04:01 +0000 (12:04 +0100)]
Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging

pc migration fixes

Last minute fixes for migration.
It seems that if we don't fix it now, fixing
it in the next version will be even more painful ...

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
# gpg: Signature made Tue 29 Jul 2014 11:45:18 BST using RSA key ID D28D5469
# gpg: Good signature from "Michael S. Tsirkin <mst@kernel.org>"
# gpg:                 aka "Michael S. Tsirkin <mst@redhat.com>"

* remotes/mst/tags/for_upstream:
  piix: set legacy table size for 1.7
  acpi-build: tweak acpi migration limits
  pc: future-proof migration-compatibility of ACPI tables
  acpi-build: minor code cleanup
  pc: acpi: generate AML only for PCI0 devices if PCI bridge hotplug is disabled
  bios-tables-test: fix ASL normalization false positive
  pc: hack for migration compatibility from QEMU 2.0
  acpi-dsdt: procedurally generate _PRT

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agopiix: set legacy table size for 1.7
Michael S. Tsirkin [Mon, 28 Jul 2014 21:00:42 +0000 (23:00 +0200)]
piix: set legacy table size for 1.7

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
10 years agoacpi-build: tweak acpi migration limits
Michael S. Tsirkin [Mon, 28 Jul 2014 21:07:11 +0000 (23:07 +0200)]
acpi-build: tweak acpi migration limits

- Tweak error message for legacy machine type:
  Basically if table size exceeds the limits we set all
  bets are off for migration: e.g. it can start failing even
  within given qemu minor version simply because of a bugfix.
- Increase table size to 128k.
- Make sure we notice it long before we start getting close to the
  128k limit: warn at 64k.
- Don't fail if we exceed the limit: most people don't care about
  migration, even less people care about cross version miration.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
10 years agopc: future-proof migration-compatibility of ACPI tables
Paolo Bonzini [Mon, 28 Jul 2014 15:34:16 +0000 (17:34 +0200)]
pc: future-proof migration-compatibility of ACPI tables

This patch avoids that similar changes break QEMU again in the future.
QEMU will now hard-code 64k as the maximum ACPI table size, which
(despite being an order of magnitude smaller than 640k) should be enough
for everyone.

Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Tested-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
10 years agoacpi-build: minor code cleanup
Michael S. Tsirkin [Mon, 28 Jul 2014 20:56:45 +0000 (22:56 +0200)]
acpi-build: minor code cleanup

Fix up and add  comments to clarify code, plus a trivial
code change for clarity.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
10 years agopc: acpi: generate AML only for PCI0 devices if PCI bridge hotplug is disabled
Igor Mammedov [Mon, 28 Jul 2014 15:34:18 +0000 (17:34 +0200)]
pc: acpi: generate AML only for PCI0 devices if PCI bridge hotplug is disabled

Fixes migration regression from QEMU-1.7 to a newer QEMUs.
SSDT table size in QEMU-1.7 doesn't change regardless of
a number of PCI bridge devices present at startup.

However in QEMU-2.0 since addition of hotplug on PCI bridges,
each PCI bridge adds ~1875 bytes to SSDT table, including
pc-i440fx-1.7 machine type where PCI bridge hotplug disabled
via compat property.
It breaks migration from "QEMU-1.7" to "QEMU-2.[01] -M pc-i440fx-1.7"
since RAMBlock size of ACPI tables on target becomes larger
then on source and migration fails with:

"Length mismatch: /rom@etc/acpi/tables: 2000 in != 3000"

error.

Fix this by generating AML only for PCI0 bus if
hotplug on PCI bridges is disabled and preserves PCI brigde
description in AML as it was done in QEMU-1.7 for pc-i440fx-1.7.

It will help to maintain size of SSDT static regardless of
number of PCI bridges on startup for pc-i440fx-1.7 machine type.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
10 years agobios-tables-test: fix ASL normalization false positive
Paolo Bonzini [Mon, 28 Jul 2014 15:34:17 +0000 (17:34 +0200)]
bios-tables-test: fix ASL normalization false positive

My version of IASL (from RHEL7) puts two newlines between the head comment
and the DefinitionBlock property.  Kill all newlines after the comment,
so that normalize_asl works properly.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
10 years agopo: Update German translation
Stefan Weil [Fri, 18 Jul 2014 14:44:21 +0000 (16:44 +0200)]
po: Update German translation

Line numbers changed, and some translations were missing after commit
3d914488aee3dc1bf495e461aedf8fb4e5bb2270.

Update also "Show Tabs" to a more common translation, and remove some
old unused lines at the end.

Signed-off-by: Stefan Weil <sw@weilnetz.de>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
10 years agotarget-mips/translate.c: Free TCG in OPC_DINSV
Dongxue Zhang [Mon, 28 Jul 2014 15:58:21 +0000 (23:58 +0800)]
target-mips/translate.c: Free TCG in OPC_DINSV

Free t0 and t1 in opcode OPC_DINSV.

Signed-off-by: Dongxue Zhang <elta.era@gmail.com>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
10 years agopc: hack for migration compatibility from QEMU 2.0
Paolo Bonzini [Mon, 28 Jul 2014 15:34:15 +0000 (17:34 +0200)]
pc: hack for migration compatibility from QEMU 2.0

Changing the ACPI table size causes migration to break, and the memory
hotplug work opened our eyes on how horribly we were breaking things in
2.0 already.

The ACPI table size is rounded to the next 4k, which one would think
gives some headroom.  In practice this is not the case, because the user
can control the ACPI table size (each CPU adds 97 bytes to the SSDT and
8 to the MADT) and so some "-smp" values will break the 4k boundary and
fail to migrate.  Similarly, PCI bridges add ~1870 bytes to the SSDT.

This patch concerns itself with fixing migration from QEMU 2.0.  It
computes the payload size of QEMU 2.0 and always uses that one.
The previous patch shrunk the ACPI tables enough that the QEMU 2.0 size
should always be enough; non-AML tables can change depending on the
configuration (especially MADT, SRAT, HPET) but they remain the same
between QEMU 2.0 and 2.1, so we only compute our padding based on the
sizes of the SSDT and DSDT.

Migration from QEMU 1.7 should work for guests that have a number of CPUs
other than 12, 13, 14, 54, 55, 56, 97, 98, 139, 140.  It was already
broken from QEMU 1.7 to QEMU 2.0 in the same way, though.

Even with this patch, QEMU 1.7 and 2.0 have two different ideas of
"-M pc-i440fx-2.0" when there are PCI bridges.  Igor sent a patch to
adopt the QEMU 1.7 definition.  I think distributions should apply
it if they move directly from QEMU 1.7 to 2.1+ without ever packaging
version 2.0.

Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Tested-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
10 years agoacpi-dsdt: procedurally generate _PRT
Paolo Bonzini [Mon, 28 Jul 2014 15:34:14 +0000 (17:34 +0200)]
acpi-dsdt: procedurally generate _PRT

This replaces the _PRT constant with a method that computes it.

The problem is that the DSDT+SSDT have grown from 2.0 to 2.1,
enough to cross the 8k barrier (we align the ACPI tables to 4k
before putting them in fw_cfg).  This causes problems with
migration and the pc-i440fx-2.0 machine type.

The solution to the problem is to hardcode 64k as the limit,
but this doesn't solve the bug with pc-i440fx-2.0.  The fix will be
for QEMU 2.1 to use exactly the same size as QEMU 2.0 for the
ACPI tables.  First, however, we must make the actual AML
equal or smaller; to do this, rewrite _PRT in a way that saves
over 1k of bytecode.

Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Tested-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
10 years agoMerge remote-tracking branch 'remotes/mjt/tags/trivial-patches-2014-07-26' into staging
Peter Maydell [Mon, 28 Jul 2014 10:05:14 +0000 (11:05 +0100)]
Merge remote-tracking branch 'remotes/mjt/tags/trivial-patches-2014-07-26' into staging

trivial patches for 2014-07-26

# gpg: Signature made Sat 26 Jul 2014 08:16:55 BST using RSA key ID A4C3D7DB
# gpg: Good signature from "Michael Tokarev <mjt@tls.msk.ru>"
# gpg:                 aka "Michael Tokarev <mjt@corpit.ru>"
# gpg:                 aka "Michael Tokarev <mjt@debian.org>"
# gpg: WARNING: This key is not certified with a trusted signature!
# gpg:          There is no indication that the signature belongs to the owner.
# Primary key fingerprint: 6EE1 95D1 886E 8FFB 810D  4324 457C E0A0 8044 65C5
#      Subkey fingerprint: 6F67 E18E 7C91 C5B1 5514  66A7 BEE5 9D74 A4C3 D7DB

* remotes/mjt/tags/trivial-patches-2014-07-26:
  qemu-options: fix another allows-to for -net l2tpv3

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agoqemu-options: fix another allows-to for -net l2tpv3
Michael Tokarev [Thu, 24 Jul 2014 16:10:17 +0000 (20:10 +0400)]
qemu-options: fix another allows-to for -net l2tpv3

Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
10 years agoMerge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into staging
Peter Maydell [Fri, 25 Jul 2014 15:58:41 +0000 (16:58 +0100)]
Merge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into staging

Here is the serial fix for 2.1.

# gpg: Signature made Fri 25 Jul 2014 13:36:23 BST using RSA key ID 9B4D86F2
# gpg: Good signature from "Paolo Bonzini <pbonzini@redhat.com>"
# gpg:                 aka "Paolo Bonzini <bonzini@gnu.org>"

* remotes/bonzini/tags/for-upstream:
  qemu-char: ignore flow control if a PTY's slave is not connected

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agoqemu-char: ignore flow control if a PTY's slave is not connected
Paolo Bonzini [Thu, 24 Jul 2014 14:08:04 +0000 (16:08 +0200)]
qemu-char: ignore flow control if a PTY's slave is not connected

After commit f702e62 (serial: change retry logic to avoid concurrency,
2014-07-11), guest boot hangs if the backend is an unconnected PTY.

The reason is that PTYs do not support G_IO_HUP, and serial_xmit is
never called.  To fix this, simply invoke serial_xmit immediately
(via g_idle_source_new) when this happens.

Tested-by: Pavel Hrdina <phrdina@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
10 years agoMerge remote-tracking branch 'remotes/kraxel/tags/pull-vnc-20140725-1' into staging
Peter Maydell [Fri, 25 Jul 2014 09:32:13 +0000 (10:32 +0100)]
Merge remote-tracking branch 'remotes/kraxel/tags/pull-vnc-20140725-1' into staging

vnc: fix two vnc update issues.

# gpg: Signature made Fri 25 Jul 2014 08:44:23 BST using RSA key ID D3E87138
# gpg: Good signature from "Gerd Hoffmann (work) <kraxel@redhat.com>"
# gpg:                 aka "Gerd Hoffmann <gerd@kraxel.org>"
# gpg:                 aka "Gerd Hoffmann (private) <kraxel@gmail.com>"

* remotes/kraxel/tags/pull-vnc-20140725-1:
  vnc update fix
  fix full frame updates for VNC clients

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agovnc update fix
Gerd Hoffmann [Wed, 23 Jul 2014 09:52:02 +0000 (11:52 +0200)]
vnc update fix

We need to remember has_updates for each vnc client.  Otherwise it might
happen that vnc_update_client(has_dirty=1) takes the first exit due to
output buffers not being flushed yet and subsequent calls with
has_dirty=0 take the second exit, wrongly assuming there is nothing to
do because the work defered in the first call is ignored.

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Peter Lieven <pl@kamp.de>
10 years agofix full frame updates for VNC clients
Stephan Kulow [Wed, 23 Jul 2014 14:03:14 +0000 (16:03 +0200)]
fix full frame updates for VNC clients

If the client asks for !incremental frame updates, it has lost its content
so dirty doesn't matter - it has to see the full frame, so setting force_update

Signed-off-by: Stephan Kulow <coolo@suse.de>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Peter Lieven <pl@kamp.de>
10 years agoMerge remote-tracking branch 'remotes/qmp-unstable/queue/qmp' into staging
Peter Maydell [Thu, 24 Jul 2014 14:23:43 +0000 (15:23 +0100)]
Merge remote-tracking branch 'remotes/qmp-unstable/queue/qmp' into staging

* remotes/qmp-unstable/queue/qmp:
  docs: document missing VSERPORT_CHANGE event
  docs: document missing POWERDOWN event
  docs: document missing SPICE_MIGRATE_COMPLETED event
  docs: split SPICE_* event docs
  docs: grammar fixes to qmp-events

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agodocs: document missing VSERPORT_CHANGE event
Eric Blake [Wed, 23 Jul 2014 12:26:18 +0000 (06:26 -0600)]
docs: document missing VSERPORT_CHANGE event

The VSERPORT_CHANGE event was added in e2ae6159.  The patch for
this event was prepared at a time when this file was gone, even
though it got applied immediately after dfab4892 restored this
file.  Duplicate the documentation into this file, so that
anyone using this file instead of qapi will not miss out on this
new event.

* docs/qmp/qmp-events.txt (VSERPORT_CHANGE): Add.

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Wenchao Xia <wenchaoqemu@gmail.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
10 years agodocs: document missing POWERDOWN event
Eric Blake [Wed, 23 Jul 2014 12:26:17 +0000 (06:26 -0600)]
docs: document missing POWERDOWN event

The POWERDOWN event was first documented in 0aab9ec3.  But since
dfab4892 later restored this file to the state prior to qmp events,
and we never documented it in the past, anyone using this file
instead of qapi will miss out on this event.  Tweak the existing
wording of SHUTDOWN to match 84321831, and make the difference
between the two events apparent.

* docs/qmp/qmp-events.txt (POWERDOWN): Add.
(SHUTDOWN): Tweak.

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Wenchao Xia <wenchaoqemu@gmail.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
10 years agodocs: document missing SPICE_MIGRATE_COMPLETED event
Eric Blake [Wed, 23 Jul 2014 12:26:16 +0000 (06:26 -0600)]
docs: document missing SPICE_MIGRATE_COMPLETED event

The SPICE_MIGRATE_COMPLETED event was first documented in
7cfadb6b.  But since dfab4892 later restored this file to the
state prior to qmp events, and we never documented it in the
past, anyone using this file instead of qapi will miss out on
this event.

* docs/qmp/qmp-events.txt (SPICE_MIGRATE_COMPLETED): Add.

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Wenchao Xia <wenchaoqemu@gmail.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
10 years agodocs: split SPICE_* event docs
Eric Blake [Wed, 23 Jul 2014 12:26:15 +0000 (06:26 -0600)]
docs: split SPICE_* event docs

For consistency with the rest of this file, every event should be
listed in isolation.  Compare how commit 7cfadb6b split
SPICE_CONNECTED and SPICE_DISCONNECTED into separate qmp events.

* docs/qmp/qmp-events.txt (SPICE_CONNECTED, SPICE_DISCONNECTED):
Split.

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Wenchao Xia <wenchaoqemu@gmail.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
10 years agodocs: grammar fixes to qmp-events
Eric Blake [Wed, 23 Jul 2014 12:26:14 +0000 (06:26 -0600)]
docs: grammar fixes to qmp-events

When converting to qmp events, commits 7cfadb6b and a6330785
fixed some grammar as part of moving text between files.  But
since dfab4892 later restored this file to the state prior to
qmp events, we have to do it again.

* docs/qmp/qmp-events.txt (RESET, SPICE_INITIALIZED): Tweak.

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Wenchao Xia <wenchaoqemu@gmail.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
10 years agoMerge remote-tracking branch 'remotes/kraxel/tags/pull-usb-20140723-1' into staging
Peter Maydell [Thu, 24 Jul 2014 11:49:54 +0000 (12:49 +0100)]
Merge remote-tracking branch 'remotes/kraxel/tags/pull-usb-20140723-1' into staging

usb: mtp: tag root property as experimental

# gpg: Signature made Wed 23 Jul 2014 07:56:21 BST using RSA key ID D3E87138
# gpg: Good signature from "Gerd Hoffmann (work) <kraxel@redhat.com>"
# gpg:                 aka "Gerd Hoffmann <gerd@kraxel.org>"
# gpg:                 aka "Gerd Hoffmann (private) <kraxel@gmail.com>"

* remotes/kraxel/tags/pull-usb-20140723-1:
  usb: mtp: tag root property as experimental

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agousb: mtp: tag root property as experimental
Gerd Hoffmann [Tue, 22 Jul 2014 07:30:12 +0000 (09:30 +0200)]
usb: mtp: tag root property as experimental

Reason: we don't want commit to that interface yet.  Possibly
the implementation will be switched over to use fsdev.

Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
10 years agoUpdate version for v2.1.0-rc3 release
Peter Maydell [Tue, 22 Jul 2014 17:17:03 +0000 (18:17 +0100)]
Update version for v2.1.0-rc3 release

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agohw/misc/imx_ccm.c: Add missing VMState list terminator
Peter Maydell [Tue, 22 Jul 2014 16:10:01 +0000 (17:10 +0100)]
hw/misc/imx_ccm.c: Add missing VMState list terminator

The VMStateDescription for the imx_ccm device was missing its
terminator. Found by static search of the codebase using
a regex based on one suggested by Ian Jackson:
  pcregrep -rMi '(?s)VMStateField(?:(?!END_OF_LIST).)*?;' $(git grep -l 'VMStateField\[\]')

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Cc: qemu-stable@nongnu.org
10 years agovmstate_xhci_event: fix unterminated field list
Laszlo Ersek [Tue, 22 Jul 2014 15:26:41 +0000 (17:26 +0200)]
vmstate_xhci_event: fix unterminated field list

"vmstate_xhci_event" was introduced in commit 37352df3 ("xhci: add live
migration support"), and first released in v1.6.0. The field list in this
VMSD is not terminated with the VMSTATE_END_OF_LIST() macro.

During normal use (ie. migration), the issue is practically invisible,
because the "vmstate_xhci_event" object (with the unterminated field list)
is only ever referenced -- via "vmstate_xhci_intr" -- if xhci_er_full()
returns true, for the "ev_buffer" test. Since that field_exists() check
(apparently) almost always returns false, we almost never traverse
"vmstate_xhci_event" during migration, which hides the bug.

However, Amit's vmstate checker forces recursion into this VMSD as well,
and the lack of VMSTATE_END_OF_LIST() breaks the field list terminator
check (field->name != NULL) in dump_vmstate_vmsd(). The result is
undefined behavior, which in my case translates to infinite recursion
(because the loop happens to overflow into "vmstate_xhci_intr", which then
links back to "vmstate_xhci_event").

Add the missing terminator.

Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Amit Shah <amit.shah@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agoMerge remote-tracking branch 'remotes/agraf/tags/signed-ppc-for-upstream' into staging
Peter Maydell [Tue, 22 Jul 2014 15:40:34 +0000 (16:40 +0100)]
Merge remote-tracking branch 'remotes/agraf/tags/signed-ppc-for-upstream' into staging

Patch queue for ppc - 2014-07-22

Only a single bug fix to make -mem-path only affect RAM regions.

# gpg: Signature made Tue 22 Jul 2014 16:38:04 BST using RSA key ID 03FEDC60
# gpg: Can't check signature: public key not found

* remotes/agraf/tags/signed-ppc-for-upstream:
  ppc: fix -mem-path failure

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agoppc: fix -mem-path failure
Hu Tao [Mon, 21 Jul 2014 09:30:17 +0000 (17:30 +0800)]
ppc: fix -mem-path failure

commit e938ba0c tried to enable -mem-path for ppc but breaked some ppc
boards.

The problems are:

1. it fails when allocating memory for rom, sram whose sizes are less
   than huge page size:

   ./ppc-softmmu/qemu-system-ppc  -m 512 -mem-path /hugepages/ \
   -kernel /home/hutao/Downloads/vmlinux-ppc -initrd \
   /home/hutao/Downloads/initrd-ppc.gz
   qemu-system-ppc: /mnt/data/projects/qemu/exec.c:1184: qemu_ram_set_idstr: Assertion `new_block' failed.

2. if there is a numa node backed by memory backend object, qemu fails
   with message:

   ./ppc-softmmu/qemu-system-ppc  -m 512 \
   -object memory-backend-file,size=512M,mem-path=/hugepages,id=f0 \
   -numa node,nodeid=0,memdev=f0 \
   -kernel /home/hutao/Downloads/vmlinux-ppc \
   -initrd /home/hutao/Downloads/initrd-ppc.gz
   qemu-system-ppc: memory backend f0 is used multiple times. Each -numa option must use a different memdev value.

This patch does following:

1. replaces memory_region_allocate_system_memory() with
   memory_region_init_ram() for rom, sram. Then only system memory
   is backed by hugepages when specifying mem-path.

2. for memory banks, allocates all ram with
   one memory_region_allocate_system_memory(), and use
   memory_region_init_alias() to initialize memory banks.

Tested machines: default(g3beige), mac99, taihu, bamboo, ref405ep.

Signed-off-by: Hu Tao <hutao@cn.fujitsu.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
10 years agoMerge remote-tracking branch 'remotes/amit-virtio-rng/for-2.1' into staging
Peter Maydell [Tue, 22 Jul 2014 12:16:04 +0000 (13:16 +0100)]
Merge remote-tracking branch 'remotes/amit-virtio-rng/for-2.1' into staging

* remotes/amit-virtio-rng/for-2.1:
  virtio-rng: Add human-readable error message for negative max-bytes parameter

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agovirtio-rng: Add human-readable error message for negative max-bytes parameter
John Snow [Mon, 21 Jul 2014 21:44:37 +0000 (17:44 -0400)]
virtio-rng: Add human-readable error message for negative max-bytes parameter

If a negative integer is used for the max_bytes parameter, QEMU currently
calls abort() and leaves behind a core dump. This patch replaces the
abort with a simple error message to make the reason for the termination
clearer. This also ensures device-hotplug with invalid input doesn't
cause qemu to quit.

There is an underlying insufficiency in the parameter parsing code of QEMU
that renders it unable to reject negative values for unsigned properties,
thus the error message "a non-negative integer below 2^63" is the most
user-friendly and correct message we can give until the underlying
insufficiency is corrected.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
10 years agovmstate static checker: whitelist additions
Amit Shah [Tue, 22 Jul 2014 07:36:08 +0000 (13:06 +0530)]
vmstate static checker: whitelist additions

Comparing json outputs from qemu-1.0 with qemu-2.1 turned up a few
description name changes; whitelist them here.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
10 years agoMerge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into staging
Peter Maydell [Tue, 22 Jul 2014 11:03:44 +0000 (12:03 +0100)]
Merge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into staging

One of the two pending migration fix, and a small KVM patch.

# gpg: Signature made Tue 22 Jul 2014 11:49:30 BST using RSA key ID 9B4D86F2
# gpg: Can't check signature: public key not found

* remotes/bonzini/tags/for-upstream:
  kvm-all: Use 'tmpcpu' instead of 'cpu' in sub-looping to avoid 'cpu' be NULL
  exec: fix migration with devices that use address_space_rw

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agokvm-all: Use 'tmpcpu' instead of 'cpu' in sub-looping to avoid 'cpu' be NULL
Chen Gang [Sat, 19 Jul 2014 01:21:46 +0000 (09:21 +0800)]
kvm-all: Use 'tmpcpu' instead of 'cpu' in sub-looping to avoid 'cpu' be NULL

If kvm_arch_remove_sw_breakpoint() in CPU_FOREACH() always be fail, it
will let 'cpu' NULL. And the next kvm_arch_remove_sw_breakpoint() in
QTAILQ_FOREACH_SAFE() will get NULL parameter for 'cpu'.

And kvm_arch_remove_sw_breakpoint() can assumes 'cpu' must never be NULL,
so need define additional temporary variable for 'cpu' to avoid the case.

Cc: qemu-stable@nongnu.org
Signed-off-by: Chen Gang <gang.chen.5i5j@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
10 years agoexec: fix migration with devices that use address_space_rw
Paolo Bonzini [Mon, 21 Jul 2014 14:45:18 +0000 (16:45 +0200)]
exec: fix migration with devices that use address_space_rw

Devices that use address_space_rw to write large areas to memory
(as opposed to address_space_map/unmap) were broken with respect
to migration since fe680d0 (exec: Limit translation limiting in
address_space_translate to xen, 2014-05-07).  Such devices include
IDE CD-ROMs.

The reason is that invalidate_and_set_dirty (called by address_space_rw
but not address_space_map/unmap) was only setting the dirty bit for
the first page in the translation.

To fix this, introduce cpu_physical_memory_set_dirty_range_nocode that
is the same as cpu_physical_memory_set_dirty_range except it does not
muck with the DIRTY_MEMORY_CODE bitmap.  This function can be used if
the caller invalidates translations with tb_invalidate_phys_page_range.

There is another difference between cpu_physical_memory_set_dirty_range
and cpu_physical_memory_set_dirty_flag; the former includes a call
to xen_modified_memory.  This is handled separately in
invalidate_and_set_dirty, and is not needed in other callers of
cpu_physical_memory_set_dirty_range_nocode, so leave it alone.

Just one nit: now that invalidate_and_set_dirty takes care of handling
multiple pages, there is no need for address_space_unmap to wrap it
in a loop.  In fact that loop would now be O(n^2).

Reported-by: Dave Gilbert <dgilbert@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Tested-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
10 years agoMerge remote-tracking branch 'remotes/afaerber/tags/qom-devices-for-2.1' into staging
Peter Maydell [Mon, 21 Jul 2014 17:06:12 +0000 (18:06 +0100)]
Merge remote-tracking branch 'remotes/afaerber/tags/qom-devices-for-2.1' into staging

QOM and device refactorings

* Machine: Property name fixups for 2.1 ABI

# gpg: Signature made Mon 21 Jul 2014 18:00:23 BST using RSA key ID 3E7E013F
# gpg: Good signature from "Andreas Färber <afaerber@suse.de>"
# gpg:                 aka "Andreas Färber <afaerber@suse.com>"

* remotes/afaerber/tags/qom-devices-for-2.1:
  machine: Replace underscores in machine's property names

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agomachine: Replace underscores in machine's property names
Marcel Apfelbaum [Fri, 18 Jul 2014 16:32:37 +0000 (19:32 +0300)]
machine: Replace underscores in machine's property names

Replaced '_' with '-' to comply with QOM guidelines.
Made the conversion from command line to QMP in vl.c.

Signed-off-by: Marcel Apfelbaum <marcel.a@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Andreas Färber <afaerber@suse.de>
10 years agoMerge remote-tracking branch 'remotes/mjt/tags/trivial-patches-2014-07-18' into staging
Peter Maydell [Fri, 18 Jul 2014 15:59:29 +0000 (16:59 +0100)]
Merge remote-tracking branch 'remotes/mjt/tags/trivial-patches-2014-07-18' into staging

trivial patches for 2014-07-18

# gpg: Signature made Fri 18 Jul 2014 15:04:43 BST using RSA key ID A4C3D7DB
# gpg: Good signature from "Michael Tokarev <mjt@tls.msk.ru>"
# gpg:                 aka "Michael Tokarev <mjt@corpit.ru>"
# gpg:                 aka "Michael Tokarev <mjt@debian.org>"
# gpg: WARNING: This key is not certified with a trusted signature!
# gpg:          There is no indication that the signature belongs to the owner.
# Primary key fingerprint: 6EE1 95D1 886E 8FFB 810D  4324 457C E0A0 8044 65C5
#      Subkey fingerprint: 6F67 E18E 7C91 C5B1 5514  66A7 BEE5 9D74 A4C3 D7DB

* remotes/mjt/tags/trivial-patches-2014-07-18:
  tests: Add missing 'static' attributes (fix warnings from smatch)
  migration: Add missing 'static' attribute
  qga: Add missing 'static' attribute
  hw/usb: Add missing 'static' attribute
  doc: slirp supports ICMP echo if enabled in Linux
  qemu-img: Remove redundancy "ret = -1"
  Fix new typos in comments (found by codespell)
  slirp: Give error message if hostfwd_add/remove for unrecognized vlan/stack

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agoMerge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into staging
Peter Maydell [Fri, 18 Jul 2014 13:46:53 +0000 (14:46 +0100)]
Merge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into staging

Andreas's fixes to --enable-modules, two 2.1 regression fixes, and a
new qtest.  Michael sent a pull request of his own, so I dropped
the vhost changes.

# gpg: Signature made Fri 18 Jul 2014 14:30:34 BST using RSA key ID 9B4D86F2
# gpg: Can't check signature: public key not found

* remotes/bonzini/tags/for-upstream:
  Revert "kvmclock: Ensure time in migration never goes backward"
  Revert "kvmclock: Ensure proper env->tsc value for kvmclock_current_nsec calculation"
  module: Don't complain when a module is absent
  module: Simplify module_load()
  qtest: new test for wdt_ib700
  target-i386: Allow execute from user mode when SMEP is enabled.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
10 years agotests: Add missing 'static' attributes (fix warnings from smatch)
Stefan Weil [Mon, 7 Jul 2014 19:03:38 +0000 (21:03 +0200)]
tests: Add missing 'static' attributes (fix warnings from smatch)

Smatch also complains about 0 used for pointers, so replace those by
NULL in test-visitor-serialization.c, too.

Signed-off-by: Stefan Weil <sw@weilnetz.de>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
10 years agomigration: Add missing 'static' attribute
Stefan Weil [Mon, 7 Jul 2014 19:09:30 +0000 (21:09 +0200)]
migration: Add missing 'static' attribute

Signed-off-by: Stefan Weil <sw@weilnetz.de>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
10 years agoqga: Add missing 'static' attribute
Stefan Weil [Mon, 7 Jul 2014 19:07:29 +0000 (21:07 +0200)]
qga: Add missing 'static' attribute

This fixes a warning from the static code analysis (smatch).

Signed-off-by: Stefan Weil <sw@weilnetz.de>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
10 years agohw/usb: Add missing 'static' attribute
Stefan Weil [Mon, 7 Jul 2014 19:05:30 +0000 (21:05 +0200)]
hw/usb: Add missing 'static' attribute

Signed-off-by: Stefan Weil <sw@weilnetz.de>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
10 years agodoc: slirp supports ICMP echo if enabled in Linux
Gernot Hillier [Thu, 10 Jul 2014 14:01:25 +0000 (16:01 +0200)]
doc: slirp supports ICMP echo if enabled in Linux

Since QEMU 0.15, slirp (user mode networking) supports ping to the
Internet, see e6d43cfb1f9

Signed-off-by: Gernot Hillier <gernot.hillier@siemens.com>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
10 years agoqemu-img: Remove redundancy "ret = -1"
Chen Gang [Thu, 3 Jul 2014 15:57:15 +0000 (23:57 +0800)]
qemu-img: Remove redundancy "ret = -1"

In this case, 'ret' is already '-1', so need not do it again.

Signed-off-by: Chen Gang <gang.chen.5i5j@gmail.com>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
10 years agoFix new typos in comments (found by codespell)
Stefan Weil [Mon, 7 Jul 2014 19:00:41 +0000 (21:00 +0200)]
Fix new typos in comments (found by codespell)

arbitary -> arbitrary
basicly -> basically

Signed-off-by: Stefan Weil <sw@weilnetz.de>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
10 years agoslirp: Give error message if hostfwd_add/remove for unrecognized vlan/stack
Peter Maydell [Mon, 16 Jun 2014 15:47:11 +0000 (16:47 +0100)]
slirp: Give error message if hostfwd_add/remove for unrecognized vlan/stack

If the user specified a (vlan ID, slirp stack name) tuple in a monitor
hostfwd_add/remove command and we can't find it, give the user an
error message rather than silently doing nothing.

This brings this error case in slirp_lookup() into line with the
other two.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Michael Tokarev <mjt@tls.msk.ru>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
10 years agoRevert "kvmclock: Ensure time in migration never goes backward"
Paolo Bonzini [Tue, 15 Jul 2014 19:46:02 +0000 (21:46 +0200)]
Revert "kvmclock: Ensure time in migration never goes backward"

This reverts commit a096b3a6732f846ec57dc28b47ee9435aa0609bf.

This patch caused a hang that was fixed by commit 9b17868 (kvmclock:
Ensure proper env->tsc value for kvmclock_current_nsec calculation,
2014-06-03), and we just had to revert that commit.  Drop this one
too.

Cc: agraf@suse.de
Cc: mtosatti@redhat.com
Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>