sdk/emulator/qemu.git
9 years agovirtio: Import virtio_vring.h
Fam Zheng [Wed, 10 Sep 2014 06:17:49 +0000 (14:17 +0800)]
virtio: Import virtio_vring.h

This header has no further dependencies. It only has some stable data
types and primitive functions, so we can copy it to include/hw/virtio in
order to allow vring code (and its user virtio-blk dataplane) to be
built unconditionally, even for cross compiling.

Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 1410329871-28885-2-git-send-email-famz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoasync: aio_context_new(): Handle event_notifier_init failure
Chrysostomos Nanakos [Thu, 18 Sep 2014 11:30:49 +0000 (14:30 +0300)]
async: aio_context_new(): Handle event_notifier_init failure

On a system with a low limit of open files the initialization
of the event notifier could fail and QEMU exits without printing any
error information to the user.

The problem can be easily reproduced by enforcing a low limit of open
files and start QEMU with enough I/O threads to hit this limit.

The same problem raises, without the creation of I/O threads, while
QEMU initializes the main event loop by enforcing an even lower limit of
open files.

This commit adds an error message on failure:

 # qemu [...] -object iothread,id=iothread0 -object iothread,id=iothread1
 qemu: Failed to initialize event notifier: Too many open files in system

Signed-off-by: Chrysostomos Nanakos <cnanakos@grnet.gr>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoblock: vhdx - fix reading beyond pointer during image creation
Jeff Cody [Tue, 16 Sep 2014 19:12:06 +0000 (15:12 -0400)]
block: vhdx - fix reading beyond pointer during image creation

In vhdx_create_metadata(), we allocate 40 bytes to entry_buffer for
the various metadata table entries.  However, we write out 64kB from
that buffer into the new file.  Only write out the correct 40 bytes.

Signed-off-by: Jeff Cody <jcody@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoblock: delete cow block driver
Stefan Hajnoczi [Tue, 16 Sep 2014 14:24:24 +0000 (15:24 +0100)]
block: delete cow block driver

This patch removes support for the cow file format.

Normally we do not break backwards compatibility but in this case there
is no impact and it is the most logical option.  Extraordinary claims
require extraordinary evidence so I will show why removing the cow block
driver is the right thing to do.

The cow file format is the disk image format for Usermode Linux, a way
of running a Linux system in userspace.  The performance of UML was
never great and it was hacky, but it enjoyed some popularity before
hardware virtualization support became mainstream.

QEMU's block/cow.c is supposed to read this image file format.
Unfortunately the file format was underspecified:

1. Earlier Linux versions used the MAXPATHLEN constant for the backing
   filename field.  The value of MAXPATHLEN can change, so Linux
   switched to a 4096 literal but QEMU has a 1024 literal.

2. Padding was not used on the header struct (both in the Linux kernel
   and in QEMU) so the struct layout varied across architectures.  In
   particular, i386 and x86_64 were different due to int64_t alignment
   differences.  Linux now uses __attribute__((packed)), QEMU does not.

Therefore:

1. QEMU cow images do not conform to the Linux cow image file format.

2. cow images cannot be shared between different host architectures.

This means QEMU cow images are useless and QEMU has not had bug reports
from users actually hitting these issues.

Let's get rid of this thing, it serves no purpose and no one will be
affected.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-id: 1410877464-20481-1-git-send-email-stefanha@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoblock/archipelago: Fix typo in qemu_archipelago_truncate()
Chrysostomos Nanakos [Tue, 16 Sep 2014 09:17:11 +0000 (12:17 +0300)]
block/archipelago: Fix typo in qemu_archipelago_truncate()

Fix a typo introduced by 94c80a438c85f2c19698547fbb115ea46d80c5f1

Signed-off-by: Chrysostomos Nanakos <cnanakos@grnet.gr>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoahci: Add test_identify case to ahci-test.
John Snow [Thu, 21 Aug 2014 17:44:39 +0000 (13:44 -0400)]
ahci: Add test_identify case to ahci-test.

Utilizing all of the bring-up code in pci_enable and hba_enable,
this test issues a simple IDENTIFY command via the HBA and retrieves
the response via the PIO receive mechanisms of the HBA.

Bugs: The DPS interrupt (Descriptor Processed Status) does not
currently get set. This will need to be adjusted in a future
patch series when the AHCI DMA pathways are reworked to allow
the feature, which may be utilized by OSX guests.

Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 1408643079-30675-9-git-send-email-jsnow@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoahci: Add test_hba_enable to ahci-test.
John Snow [Thu, 21 Aug 2014 17:44:38 +0000 (13:44 -0400)]
ahci: Add test_hba_enable to ahci-test.

This test engages the HBA functionality and initializes
values to sane defaults to allow for minimal HBA functionality.

Buffers are allocated and pointers are updated to allow minimal
I/O commands to complete as expected. Error registers and responses
are sanity checked for specification adherence.

Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 1408643079-30675-8-git-send-email-jsnow@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoahci: Add test_hba_spec to ahci-test.
John Snow [Thu, 21 Aug 2014 17:44:37 +0000 (13:44 -0400)]
ahci: Add test_hba_spec to ahci-test.

Add a test routine that checks the boot-up values of the HBA
configuration memory space against the AHCI 1.3 specification
and Intel ICH9 data sheet (for Q35 machines) for adherence and
sane values.

The HBA is not yet engaged or put into the idle state.

[Replaced g_assert_false(...) with g_assert(!...) for glib <2.38
compatibility, reported by Peter Maydell <peter.maydell@linaro.org>.
--Stefan]

Signed-off-by: John Snow <jsnow@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-id: 1408643079-30675-7-git-send-email-jsnow@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoahci: properly shadow the TFD register
John Snow [Thu, 21 Aug 2014 17:44:36 +0000 (13:44 -0400)]
ahci: properly shadow the TFD register

In a real AHCI device, several S/ATA registers are mirrored or shadowed
within the AHCI register set. These registers are not updated
synchronously for each read access, but are instead updated after a
Device-to-Host Register FIS packet is received. The D2H FIS contains
the values from these registers on the device.

In QEMU, by reaching directly into the device to grab these bits before
they are "sent," we may introduce race conditions where unexpected
values are present "before they are sent" which could cause issues for
some guests, particularly if an attempt is made to read the PxTFD
register prior to enabling the port, where incorrect values will be read.

This patch also addresses the boot-time values for the PxTFD and PxSIG
registers to bring them in line with the AHCI 1.3 specification.

Lastly, several fields (PxTFD, PxSIG and PxSACT) are read-only,
and any attempts to write to them should be ignored.

Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 1408643079-30675-6-git-send-email-jsnow@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoahci: add test_pci_enable to ahci-test.
John Snow [Thu, 21 Aug 2014 17:44:35 +0000 (13:44 -0400)]
ahci: add test_pci_enable to ahci-test.

This adds a test wherein we engage the PCI AHCI
device and ensure that the memory region for the
HBA functionality is now accessible.

Under Q35 environments, additional PCI configuration
is performed to ensure that the HBA functionality
will become usable.

Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 1408643079-30675-5-git-send-email-jsnow@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoahci: Add test_pci_spec to ahci-test.
John Snow [Thu, 21 Aug 2014 17:44:34 +0000 (13:44 -0400)]
ahci: Add test_pci_spec to ahci-test.

Adds a specification adherence test for AHCI
where the boot-up values for the PCI configuration space
are compared against the AHCI 1.3 specification.

This test does not itself attempt to engage the device.

Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 1408643079-30675-4-git-send-email-jsnow@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoahci: MSI capability should be at 0x80, not 0x50.
John Snow [Thu, 21 Aug 2014 17:44:33 +0000 (13:44 -0400)]
ahci: MSI capability should be at 0x80, not 0x50.

In the Intel ICH9 data sheet, the MSI capability offset
in the PCI configuration space for ICH9 AHCI devices is
specified to be 0x80.

Further, the PCI capability pointer should always point
to 0x80 in ICH9 devices, despite the fact that AHCI 1.3
specifies that it should be pointing to PMCAP (Which in
this instance would be 0x70) to maintain adherence to
the Intel data sheet specifications and real observed behavior.

Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 1408643079-30675-3-git-send-email-jsnow@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoahci: Adding basic functionality qtest.
John Snow [Thu, 21 Aug 2014 17:44:32 +0000 (13:44 -0400)]
ahci: Adding basic functionality qtest.

Currently, there is no qtest to test the functionality of
the AHCI functionality present within the Q35 machine type.

This patch adds a skeleton for an AHCI test suite,
and adds a simple sanity-check test case where we
identify that the AHCI device is present, then
disengage the virtual machine.

Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 1408643079-30675-2-git-send-email-jsnow@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agolayout: Add generators for refcount table and blocks
Maria Kustova [Tue, 19 Aug 2014 12:25:13 +0000 (16:25 +0400)]
layout: Add generators for refcount table and blocks

Refcount structures are placed in clusters randomly selected from all
unallocated host clusters.

Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Maria Kustova <maria.k@catit.be>
Reviewed-by: Fam Zheng <famz@redhat.com>
Message-id: 7e2f38608db6fba2da53997390b19400d445c45d.1408450493.git.maria.k@catit.be
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agofuzz: Add fuzzing functions for entries of refcount table and blocks
Maria Kustova [Tue, 19 Aug 2014 12:25:12 +0000 (16:25 +0400)]
fuzz: Add fuzzing functions for entries of refcount table and blocks

Reviewed-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Maria Kustova <maria.k@catit.be>
Message-id: c9f4027b6f401c67e9d18f94aed29be445e81d48.1408450493.git.maria.k@catit.be
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agodocs: List all image elements currently supported by the fuzzer
Maria Kustova [Tue, 19 Aug 2014 12:25:11 +0000 (16:25 +0400)]
docs: List all image elements currently supported by the fuzzer

Reviewed-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Maria Kustova <maria.k@catit.be>
Message-id: cb71485d0f55d1d8401eebaead8324eb78673060.1408450493.git.maria.k@catit.be
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoqapi/block-core: Add "new" qcow2 options
Max Reitz [Wed, 20 Aug 2014 17:59:36 +0000 (19:59 +0200)]
qapi/block-core: Add "new" qcow2 options

qcow2 supports more than four options by now, add the new options
(overlap check mode and metadata cache size)

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-id: 1408557576-14574-5-git-send-email-mreitz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoqcow2: Add overlap-check.template option
Max Reitz [Wed, 20 Aug 2014 17:59:35 +0000 (19:59 +0200)]
qcow2: Add overlap-check.template option

Being able to set the overlap-check option to a string and then refine
it via the overlap-check.* options is a nice idea for the command line
but does not work so well for non-flattened dicts. In that case, one can
only specify either but not both, so add a field to overlap-check.*
which does the same as directly specifying overlap-check but can be used
in conjunction with the other fields in non-flattened dicts.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-id: 1408557576-14574-4-git-send-email-mreitz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoqapi: Allow enums in anonymous unions
Max Reitz [Wed, 20 Aug 2014 17:59:34 +0000 (19:59 +0200)]
qapi: Allow enums in anonymous unions

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-id: 1408557576-14574-3-git-send-email-mreitz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoqcow2: Fix leak of QemuOpts in qcow2_open()
Max Reitz [Wed, 20 Aug 2014 17:59:33 +0000 (19:59 +0200)]
qcow2: Fix leak of QemuOpts in qcow2_open()

Currently, the QemuOpts object opts is leaked if anything fails from its
creation up to and including the image repair block. Fix this by freeing
that object in the fail path.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Benoît Canet <benoit.canet@nodalink.com>
Message-id: 1408557576-14574-2-git-send-email-mreitz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agohmp: fix memory leak at hmp_info_block_jobs()
Gonglei [Tue, 16 Sep 2014 13:36:55 +0000 (21:36 +0800)]
hmp: fix memory leak at hmp_info_block_jobs()

Signed-off-by: Gonglei <arei.gonglei@huawei.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-id: 1410874615-14292-1-git-send-email-arei.gonglei@huawei.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoimage-fuzzer: Trivial readability and formatting improvements
Maria Kustova [Sun, 14 Sep 2014 12:07:02 +0000 (16:07 +0400)]
image-fuzzer: Trivial readability and formatting improvements

Signed-off-by: Maria Kustova <maria.k@catit.be>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoiotests: Add more tests for qcow2 corruption
Max Reitz [Fri, 5 Sep 2014 14:07:19 +0000 (16:07 +0200)]
iotests: Add more tests for qcow2 corruption

Add tests for unaligned L1/L2/reftable entries and non-fatal corruption
reports.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-id: 1409926039-29044-6-git-send-email-mreitz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoqcow2: Check L1/L2/reftable entries for alignment
Max Reitz [Fri, 5 Sep 2014 14:07:18 +0000 (16:07 +0200)]
qcow2: Check L1/L2/reftable entries for alignment

Offsets taken from the L1, L2 and refcount tables are generally assumed
to be correctly aligned. However, this cannot be guaranteed if the image
has been written to by something different than qemu, thus check all
offsets taken from these tables for correct cluster alignment.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-id: 1409926039-29044-5-git-send-email-mreitz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoqcow2: Use qcow2_signal_corruption() for overlaps
Max Reitz [Fri, 5 Sep 2014 14:07:17 +0000 (16:07 +0200)]
qcow2: Use qcow2_signal_corruption() for overlaps

Use the new function in case of a failed overlap check.

This changes output in case of corruption, so adapt iotest 060's
reference output accordingly.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Benoît Canet <benoit.canet@nodalink.com>
Message-id: 1409926039-29044-4-git-send-email-mreitz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoqcow2: Add qcow2_signal_corruption()
Max Reitz [Fri, 5 Sep 2014 14:07:16 +0000 (16:07 +0200)]
qcow2: Add qcow2_signal_corruption()

Add a helper function for easily marking an image corrupt (on fatal
corruptions) while outputting an informative message to stderr and via
QAPI.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Benoît Canet <benoit.canet@nodalink.com>
Message-id: 1409926039-29044-3-git-send-email-mreitz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoqapi/block: Add "fatal" to BLOCK_IMAGE_CORRUPTED
Max Reitz [Fri, 5 Sep 2014 14:07:15 +0000 (16:07 +0200)]
qapi/block: Add "fatal" to BLOCK_IMAGE_CORRUPTED

Not every BLOCK_IMAGE_CORRUPTED event must be fatal; for example, when
reading from an image, they should generally not be. Nonetheless, even
an image only read from may of course be corrupted and this can be
detected during normal operation. In this case, a non-fatal event should
be emitted, but the image should not be marked corrupt (in accordance to
"fatal" set to false).

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-id: 1409926039-29044-2-git-send-email-mreitz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoqapi: Sort items in BlockdevOptions definition
Fam Zheng [Thu, 11 Sep 2014 06:09:58 +0000 (14:09 +0800)]
qapi: Sort items in BlockdevOptions definition

Signed-off-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Benoît Canet <benoit.canet@nodalink.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-id: 1410415798-20673-4-git-send-email-famz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoqapi: Sort BlockdevDriver enum data list
Fam Zheng [Thu, 11 Sep 2014 06:09:57 +0000 (14:09 +0800)]
qapi: Sort BlockdevDriver enum data list

Reviewed-by: Benoît Canet <benoit.canet@nodalink.com>
Signed-off-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-id: 1410415798-20673-3-git-send-email-famz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoblock: Introduce "null" drivers
Fam Zheng [Thu, 11 Sep 2014 06:09:56 +0000 (14:09 +0800)]
block: Introduce "null" drivers

This is an analogue to Linux null_blk. It can be used for testing or
benchmarking block device emulation and general block layer
functionalities such as coroutines and throttling, where disk IO is not
necessary or wanted.

Use null-aio:// for AIO version, and null-co:// for coroutine version.

[Resolved conflict with Fam's async bdrv_aio_cancel() series:
1. Drop .bdrv_aio_cancel() since it is now done by block.c
2. Rename qemu_aio_release() to qemu_aio_unref()
--Stefan]

Signed-off-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Benoît Canet <benoit.canet@nodalink.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-id: 1410415798-20673-2-git-send-email-famz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoaio-win32: avoid out-of-bounds access to the events array
Paolo Bonzini [Mon, 15 Sep 2014 12:52:58 +0000 (14:52 +0200)]
aio-win32: avoid out-of-bounds access to the events array

If ret is WAIT_TIMEOUT and there was an event returned by select(),
we can write to a location after the end of the array.  But in
that case we can retry the WaitForMultipleObjects call with the
same set of events, so just move the event[ret - WAIT_OBJECT_0]
assignment inside the existin conditional.

Reported-by: TeLeMan <geleman@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: TeLeMan <geleman@gmail.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoqdev-monitor: fix segmentation fault on qdev_device_help()
Gonglei [Tue, 16 Sep 2014 02:19:33 +0000 (10:19 +0800)]
qdev-monitor: fix segmentation fault on qdev_device_help()

Normally, qmp_device_list_properties() may return NULL when
a device haven't special properties excpet Object and DeviceState
properties, such as virtio-balloon-device.

We just need check local_err instead of prop_list.

Example:

Segmentation fault (core dumped)

The backtrace as below:

Program received signal SIGSEGV, Segmentation fault.
0x00005555559af1a8 in error_get_pretty (err=0x0) at util/error.c:152
152         return err->msg;
(gdb) bt
    func=0x55555574a6ca <device_help_func>, opaque=0x0, abort_on_failure=0) at util/qemu-option.c:1072

Signed-off-by: Gonglei <arei.gonglei@huawei.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoblock: Rename qemu_aio_release -> qemu_aio_unref
Fam Zheng [Thu, 11 Sep 2014 05:41:28 +0000 (13:41 +0800)]
block: Rename qemu_aio_release -> qemu_aio_unref

Suggested-by: Benoît Canet <benoit.canet@irqsave.net>
Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoblock: Drop AIOCBInfo.cancel
Fam Zheng [Thu, 11 Sep 2014 05:41:27 +0000 (13:41 +0800)]
block: Drop AIOCBInfo.cancel

Now that all the implementations are converted to asynchronous version
and we can emulate synchronous cancellation with it. Let's drop the
unused member.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoide: Convert trim_aiocb_info.cancel to .cancel_async
Fam Zheng [Thu, 11 Sep 2014 05:41:26 +0000 (13:41 +0800)]
ide: Convert trim_aiocb_info.cancel to .cancel_async

We know that either bh is scheduled or ide_issue_trim_cb will be called
again, so we just set i, j and ret to the right values. In both cases,
ide_trim_bh_cb will be called.

Also forward the cancellation to the iocb->aiocb which we get from
bdrv_aio_discard.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agowin32-aio: Drop win32_aiocb_info.cancel
Fam Zheng [Thu, 11 Sep 2014 05:41:25 +0000 (13:41 +0800)]
win32-aio: Drop win32_aiocb_info.cancel

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agosheepdog: Convert sd_aiocb_info.cancel to .cancel_async
Fam Zheng [Thu, 11 Sep 2014 05:41:24 +0000 (13:41 +0800)]
sheepdog: Convert sd_aiocb_info.cancel to .cancel_async

Also drop the now unused SheepdogAIOCB.finished field. Note that this
aio is internal to sheepdog driver and has NULL cb and opaque, and
should be unused at all.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agorbd: Drop rbd_aiocb_info.cancel
Fam Zheng [Thu, 11 Sep 2014 05:41:23 +0000 (13:41 +0800)]
rbd: Drop rbd_aiocb_info.cancel

And also drop the now unused "cancelled" field.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoquorum: Convert quorum_aiocb_info.cancel to .cancel_async
Fam Zheng [Thu, 11 Sep 2014 05:41:22 +0000 (13:41 +0800)]
quorum: Convert quorum_aiocb_info.cancel to .cancel_async

Before, we cancel all the child requests with bdrv_aio_cancel, then free
the acb..

Now we just kick off asynchronous cancellation of child requests and
return, we know quorum_aio_cb will be called later, so in the end
quorum_aio_finalize will take care of calling the caller's cb.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoquorum: fix quorum_aio_cancel()
Liu Yuan [Thu, 11 Sep 2014 05:41:21 +0000 (13:41 +0800)]
quorum: fix quorum_aio_cancel()

For a fifo read pattern, we only have one running aio (possible other cases that
has less number than num_children in the future), so we need to check if
.acb is NULL against bdrv_aio_cancel() to avoid segfault.

Cc: Eric Blake <eblake@redhat.com>
Cc: Benoit Canet <benoit@irqsave.net>
Cc: Kevin Wolf <kwolf@redhat.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Liu Yuan <namei.unix@gmail.com>
Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoqed: Drop qed_aiocb_info.cancel
Fam Zheng [Thu, 11 Sep 2014 05:41:20 +0000 (13:41 +0800)]
qed: Drop qed_aiocb_info.cancel

Also drop the now unused ->finished field.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agocurl: Drop curl_aiocb_info.cancel
Fam Zheng [Thu, 11 Sep 2014 05:41:19 +0000 (13:41 +0800)]
curl: Drop curl_aiocb_info.cancel

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoblkverify: Drop blkverify_aiocb_info.cancel
Fam Zheng [Thu, 11 Sep 2014 05:41:18 +0000 (13:41 +0800)]
blkverify: Drop blkverify_aiocb_info.cancel

Also the finished pointer is not used any more.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoblkdebug: Drop blkdebug_aiocb_info.cancel
Fam Zheng [Thu, 11 Sep 2014 05:41:17 +0000 (13:41 +0800)]
blkdebug: Drop blkdebug_aiocb_info.cancel

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoarchipelago: Drop archipelago_aiocb_info.cancel
Fam Zheng [Thu, 11 Sep 2014 05:41:16 +0000 (13:41 +0800)]
archipelago: Drop archipelago_aiocb_info.cancel

The cancelled flag is no longer useful. Later the request will complete
as before, and cb will be called.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoiscsi: Convert iscsi_aiocb_info.cancel to .cancel_async
Fam Zheng [Thu, 11 Sep 2014 05:41:15 +0000 (13:41 +0800)]
iscsi: Convert iscsi_aiocb_info.cancel to .cancel_async

Also drop the unused field "canceled".

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agodma: Convert dma_aiocb_info.cancel to .cancel_async
Fam Zheng [Thu, 11 Sep 2014 05:41:14 +0000 (13:41 +0800)]
dma: Convert dma_aiocb_info.cancel to .cancel_async

Just forward the request to bdrv_aio_cancel_async.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agolinux-aio: Convert laio_aiocb_info.cancel to .cancel_async
Fam Zheng [Thu, 11 Sep 2014 05:41:13 +0000 (13:41 +0800)]
linux-aio: Convert laio_aiocb_info.cancel to .cancel_async

Just call io_cancel (2), if it fails, it means the request is not
canceled, so the event loop will eventually call
qemu_laio_process_completion.

In qemu_laio_process_completion, change to call the cb unconditionally.
It is required by bdrv_aio_cancel_async.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agothread-pool: Convert thread_pool_aiocb_info.cancel to cancel_async
Fam Zheng [Thu, 11 Sep 2014 05:41:12 +0000 (13:41 +0800)]
thread-pool: Convert thread_pool_aiocb_info.cancel to cancel_async

The .cancel_async shares the same the first half with .cancel: try to
steal the request if not submitted yet. In this case set the elem to
THREAD_DONE status and ret to -ECANCELED, which means
thread_pool_completion_bh will call the cb with -ECANCELED.

If the request is already submitted, do nothing, as we know the normal
completion will happen in the future.

Testing code update:

Before, done_cb is only called if the request is already submitted by
thread pool. Now done_cb is always called, even before it is submitted,
because we emulate bdrv_aio_cancel with bdrv_aio_cancel_async. So also
update the test criteria accordingly.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoblock: Drop bdrv_em_aiocb_info.cancel
Fam Zheng [Thu, 11 Sep 2014 05:41:11 +0000 (13:41 +0800)]
block: Drop bdrv_em_aiocb_info.cancel

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoblock: Drop bdrv_em_co_aiocb_info.cancel
Fam Zheng [Thu, 11 Sep 2014 05:41:10 +0000 (13:41 +0800)]
block: Drop bdrv_em_co_aiocb_info.cancel

Also drop the now unused ->done pointer.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoblock: Add bdrv_aio_cancel_async
Fam Zheng [Thu, 11 Sep 2014 05:41:09 +0000 (13:41 +0800)]
block: Add bdrv_aio_cancel_async

This is the async version of bdrv_aio_cancel, which doesn't block the
caller. It guarantees that the cb is called either before returning or
some time later.

bdrv_aio_cancel can base on bdrv_aio_cancel_async, later we can convert
all .io_cancel implementations to .io_cancel_async, and the aio_poll is
the common logic. In the end, .io_cancel can be dropped.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoblock: Add refcnt in BlockDriverAIOCB
Fam Zheng [Thu, 11 Sep 2014 05:41:08 +0000 (13:41 +0800)]
block: Add refcnt in BlockDriverAIOCB

This will be useful in synchronous cancel emulation with
bdrv_aio_cancel_async.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoide/ahci: Check for -ECANCELED in aio callbacks
Fam Zheng [Thu, 11 Sep 2014 05:41:07 +0000 (13:41 +0800)]
ide/ahci: Check for -ECANCELED in aio callbacks

Before, bdrv_aio_cancel will either complete the request (like normal)
and call CB with an actual return code, or skip calling the request (for
example when the IO req is not submitted by thread pool yet).

We will change bdrv_aio_cancel to do it differently: always call CB
before return, with either [1] a normal req completion ret code, or [2]
ret == -ECANCELED. So the callers' callback must accept both cases. The
existing logic works with case [1], but not [2].

The simplest transition of callback code is do nothing in case [2], just
as if the CB is not called by the bdrv_aio_cancel() call.

Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoaio-win32: fix uninitialized use of have_select_revents
Paolo Bonzini [Fri, 12 Sep 2014 10:08:49 +0000 (12:08 +0200)]
aio-win32: fix uninitialized use of have_select_revents

Always initialize it with the return value of aio_prepare.

Reported-by: TeLeMan <geleman@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoide/atapi: Mark non-data commands as complete
John Snow [Sat, 13 Sep 2014 03:51:12 +0000 (23:51 -0400)]
ide/atapi: Mark non-data commands as complete

When the command completion code in IDE and AHCI
was unified to put all command completion inside
of a callback, "cmd_done," we neglected to
ensure that all AHCI/ATAPI command paths would
eventually register as finished. for the PCI
interface to IDE this is not a problem because
cmd_done is a nop, but the AHCI implementation
needs to send a D2H_REG_FIS and interrupt back
to the guest to inform of completion.

This patch adds calls to ide_stop_transfer,
which calls ide_cmd_done, inside of
ide_atapi_cmd_ok and ide_atapi_cmd_error.

This fixes regressions observed by trying to boot QEMU
with a Fedora 20 live CD under Q35/AHCI, which uses
ATAPI command 0x00, which is a status check that may
cause a hang because we never complete, and ATAPI
command 0x56, which is unsupported by our current
implementation and results in an error that we never
report back to the guest.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoblock/vhdx.c: Mark parent_vhdx_guid variable as unused
Peter Maydell [Sun, 14 Sep 2014 19:29:59 +0000 (20:29 +0100)]
block/vhdx.c: Mark parent_vhdx_guid variable as unused

The parent_vhdx_guid variable is defined but never used, which provokes
complaints from newer versions of clang. Since the variable definition
is here acting as documentation of the image format, mark it with the
'unused' attribute to keep the compiler happy rather than simply
deleting it.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Jeff Cody <jcody@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoexec.c: fix setting 1-byte-long watchpoints
Max Filippov [Thu, 18 Sep 2014 05:03:36 +0000 (22:03 -0700)]
exec.c: fix setting 1-byte-long watchpoints

With commit 05068c0dfb5b 'exec.c: Relax restrictions on watchpoint length
and alignment' it's no longer possible to set 1-byte-long watchpoint
because of incorrect address range check.
Fix that by changing condition that checks for address wraparound.

Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1411016616-29879-1-git-send-email-jcmvbkbc@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoFix cross compilation (nm command)
Stefan Weil [Thu, 18 Sep 2014 19:55:08 +0000 (21:55 +0200)]
Fix cross compilation (nm command)

Commit c261d774fb9093d00e0938a19f502fb220f62718 added one more binutils
tool: nm also needs a cross prefix.

Signed-off-by: Stefan Weil <sw@weilnetz.de>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1411070108-8954-1-git-send-email-sw@weilnetz.de
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging
Peter Maydell [Thu, 18 Sep 2014 19:02:00 +0000 (20:02 +0100)]
Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging

pci, pc, virtio, misc bugfixes

A bunch of bugfixes - some of these will make sense for 2.1.2
I put Cc: qemu-stable included where appropriate.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
# gpg: Signature made Thu 18 Sep 2014 19:52:18 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:
  pc: leave more space for BIOS allocations
  virtio-pci: fix migration for pci bus master
  vhost-user: fix VIRTIO_NET_F_MRG_RXBUF negotiation
  virtio-pci: enable bus master for old guests
  Revert "virtio: don't call device on !vm_running"
  virtio-net: drop assert on vm stop
  Revert "rng-egd: remove redundant free"
  qdev: Move global validation to a single function
  qdev: Rename qdev_prop_check_global() to qdev_prop_check_globals()
  test-qdev-global-props: Test handling of hotpluggable and non-device types
  test-qdev-global-props: Initialize not_used=true for all props
  test-qdev-global-props: Run tests on subprocess
  tests: disable global props test for old glib
  test-qdev-global-props: Trivial comment fix
  hw/machine: Free old values of string properties

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agopc: leave more space for BIOS allocations
Michael S. Tsirkin [Thu, 18 Sep 2014 13:32:07 +0000 (16:32 +0300)]
pc: leave more space for BIOS allocations

Since QEMU 2.1, we are allocating more space for ACPI tables, so no
space is left after initrd for the BIOS to allocate memory.

Besides ACPI tables, there are a few other uses of high memory in
SeaBIOS: SMBIOS tables and USB drivers use it in particular.  These uses
allocate a very small amount of memory.  Malloc metadata also lives
there.  So we need _some_ extra padding there to avoid initrd breakage,
but not much.

John Snow found a case where RHEL5 was broken by the recent change to
ACPI_TABLE_SIZE; in his case 4KB of extra padding are fine, but just to
be safe I am adding 32KB, which is roughly the same amount of padding
that was left by QEMU 2.0 and earlier.

Move initrd to leave some space for the BIOS.

Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reported-by: John Snow <jsnow@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agovirtio-pci: fix migration for pci bus master
Michael S. Tsirkin [Thu, 11 Sep 2014 15:34:29 +0000 (18:34 +0300)]
virtio-pci: fix migration for pci bus master

Current support for bus master (clearing OK bit)
together with the need to support guests which do not
enable PCI bus mastering, leads to extra state in
VIRTIO_PCI_FLAG_BUS_MASTER_BUG bit, which isn't robust
in case of cross-version migration for the case when
guests use the device before setting DRIVER_OK.

Rip out VIRTIO_PCI_FLAG_BUS_MASTER_BUG and implement a simpler
work-around: treat clearing of PCI_COMMAND as a virtio reset.  Old
guests never touch this bit so they will work.

As reset clears device status, DRIVER and MASTER bits are
now in sync, so we can fix up cross-version migration simply
by synchronising them, without need to detect a buggy guest
explicitly.

Drop tracking VIRTIO_PCI_FLAG_BUS_MASTER_BUG completely.

As reset makes the device quiescent, in the future we'll be able to drop
checking OK bit in a bunch of places.

Cc: Jason Wang <jasowang@redhat.com>
Cc: Greg Kurz <gkurz@linux.vnet.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agovhost-user: fix VIRTIO_NET_F_MRG_RXBUF negotiation
Damjan Marion [Thu, 11 Sep 2014 21:55:48 +0000 (14:55 -0700)]
vhost-user: fix VIRTIO_NET_F_MRG_RXBUF negotiation

Header length check should happen only if backend is kernel. For user
backend there is no reason to reset this bit.

vhost-user code does not define .has_vnet_hdr_len so
VIRTIO_NET_F_MRG_RXBUF cannot be negotiated even if both sides
support it.

Signed-off-by: Damjan Marion <damarion@cisco.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agovirtio-pci: enable bus master for old guests
Michael S. Tsirkin [Thu, 11 Sep 2014 16:45:33 +0000 (18:45 +0200)]
virtio-pci: enable bus master for old guests

commit cc943c36faa192cd4b32af8fe5edb31894017d35
    pci: Use bus master address space for delivering MSI/MSI-X messages
breaks virtio-net for rhel6.[56] x86 guests because they don't
enable bus mastering for virtio PCI devices. For the same reason,
rhel6.[56] ppc64 guests cannot boot on a virtio-blk disk anymore.

Old guests forgot to enable bus mastering, enable it automatically on
DRIVER (guests use some devices before DRIVER_OK).

Reported-by: Greg Kurz <gkurz@linux.vnet.ibm.com>
Reviewed-by: Greg Kurz <gkurz@linux.vnet.ibm.com>
Tested-by: Greg Kurz <gkurz@linux.vnet.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoRevert "virtio: don't call device on !vm_running"
Michael S. Tsirkin [Thu, 11 Sep 2014 15:42:02 +0000 (18:42 +0300)]
Revert "virtio: don't call device on !vm_running"

This reverts commit a1bc7b827e422e1ff065640d8ec5347c4aadfcd8.
    virtio: don't call device on !vm_running
It turns out that virtio net assumes that vm_running
is updated before device status callback in many places,
so this change leads to asserts.
Previous commit fixes the root issue that motivated
a1bc7b827e422e1ff065640d8ec5347c4aadfcd8 differently,
so there's no longer a need for this change.

In the future, we might be able to drop checking vm_running
completely, and check vm state directly.

Reported-by: Dietmar Maurer <dietmar@proxmox.com>
Cc: qemu-stable@nongnu.org
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agovirtio-net: drop assert on vm stop
Michael S. Tsirkin [Thu, 11 Sep 2014 15:32:51 +0000 (18:32 +0300)]
virtio-net: drop assert on vm stop

On vm stop, vm_running state set to stopped
before device is notified, so callbacks can get envoked with
vm_running = false; and this is not an error.

Cc: qemu-stable@nongnu.org
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoRevert "rng-egd: remove redundant free"
Eduardo Habkost [Thu, 4 Sep 2014 16:10:47 +0000 (19:10 +0300)]
Revert "rng-egd: remove redundant free"

This reverts commit 5e490b6a504912225dff0e520e1c6af68295d238.

Cc: qemu-stable@nongnu.org
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoqdev: Move global validation to a single function
Eduardo Habkost [Fri, 8 Aug 2014 19:03:31 +0000 (16:03 -0300)]
qdev: Move global validation to a single function

Currently GlobalProperty.not_used=false has multiple meanings:

* It may be a property for a hotpluggable device, which may or may not
  have been used by a device;
* It may be a machine-type-provided property, which may or may not have
  been used by a device.
* It may be a user-provided property that was actually not used by
  any device.

Simplify the logic by having two separate fields: 'user_provided' and
'used'. This allows the entire global property validation logic to be
contained in a single function, and allows more specific error messages.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoqdev: Rename qdev_prop_check_global() to qdev_prop_check_globals()
Eduardo Habkost [Fri, 8 Aug 2014 19:03:30 +0000 (16:03 -0300)]
qdev: Rename qdev_prop_check_global() to qdev_prop_check_globals()

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agotest-qdev-global-props: Test handling of hotpluggable and non-device types
Eduardo Habkost [Fri, 8 Aug 2014 19:03:29 +0000 (16:03 -0300)]
test-qdev-global-props: Test handling of hotpluggable and non-device types

Ensure no warning will be printed for hotpluggable types, and warnings
will be printed for non-device types.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agotest-qdev-global-props: Initialize not_used=true for all props
Eduardo Habkost [Fri, 8 Aug 2014 19:03:28 +0000 (16:03 -0300)]
test-qdev-global-props: Initialize not_used=true for all props

This will ensure we are actually testing the code which sets
not_used=false when the property is used.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agotest-qdev-global-props: Run tests on subprocess
Eduardo Habkost [Fri, 8 Aug 2014 19:03:27 +0000 (16:03 -0300)]
test-qdev-global-props: Run tests on subprocess

There are multiple reasons for running the global property tests on a
subprocess:

* We need the global_props lists to be empty for each test case, so
  global properties from the previous test won't affect the next one;
* We don't want the qdev_prop_check_global() warnings to pollute test
  output;
* With a subprocess, we can ensure qdev_prop_check_global() is printing
  the warning messages it should.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agotests: disable global props test for old glib
Michael S. Tsirkin [Thu, 18 Sep 2014 17:46:45 +0000 (20:46 +0300)]
tests: disable global props test for old glib

follow-up patch moves global property tests to subprocesses.
Unfortunately with old glib this causes:

tests/test-qdev-global-props.c: In function
‘test_static_prop’:
tests/test-qdev-global-props.c:80:5: error: implicit
declaration of function ‘g_test_trap_subprocess’
[-Werror=implicit-function-declaration]
tests/test-qdev-global-props.c:80:5: error: nested extern
declaration of ‘g_test_trap_subprocess’ [-Werror=nested-externs]

This function was only added in glib 2.38, and our
minimum version is 2.12.

To fix, disable the test for glib < 2.38.

Apply before that patch to avoid breaking bisect.

Reported-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoMerge remote-tracking branch 'remotes/kraxel/tags/pull-vnc-20140918-1' into staging
Peter Maydell [Thu, 18 Sep 2014 16:00:38 +0000 (17:00 +0100)]
Merge remote-tracking branch 'remotes/kraxel/tags/pull-vnc-20140918-1' into staging

vnc: set TCP_NODELAY, cleanup in tlc code

# gpg: Signature made Thu 18 Sep 2014 07:02:37 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-20140918-1:
  vnc-tls: Clean up dead store in vnc_set_x509_credential()
  ui/vnc: set TCP_NODELAY

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agovnc-tls: Clean up dead store in vnc_set_x509_credential()
Markus Armbruster [Wed, 17 Sep 2014 07:33:15 +0000 (09:33 +0200)]
vnc-tls: Clean up dead store in vnc_set_x509_credential()

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
9 years agoui/vnc: set TCP_NODELAY
Peter Lieven [Fri, 5 Sep 2014 20:07:41 +0000 (22:07 +0200)]
ui/vnc: set TCP_NODELAY

we currently have the Nagle algorithm enabled for all outgoing VNC updates.
This may delay sensitive updates as mouse movements or typing in the console.
As we currently prepare all data in a buffer and then send as much as we can
disabling the Nagle algorithm should not cause big trouble. Well established
VNC servers like TightVNC set TCP_NODELAY as well.
A regular framebuffer update request generates exactly one framebuffer update
which should be pushed out as fast as possible.

Signed-off-by: Peter Lieven <pl@kamp.de>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
9 years agoqemu-char: Rename register_char_driver_qapi() to register_char_driver()
Peter Maydell [Tue, 2 Sep 2014 10:24:17 +0000 (11:24 +0100)]
qemu-char: Rename register_char_driver_qapi() to register_char_driver()

Now we have removed the legacy register_char_driver() we can
rename register_char_driver_qapi() to the more obvious and
shorter name.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1409653457-27863-6-git-send-email-peter.maydell@linaro.org

9 years agoqemu-char: Remove register_char_driver() machinery
Peter Maydell [Tue, 2 Sep 2014 10:24:16 +0000 (11:24 +0100)]
qemu-char: Remove register_char_driver() machinery

Now that all the char backends have been converted to the QAPI
framework we can remove the machinery for handling old style
backends.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1409653457-27863-5-git-send-email-peter.maydell@linaro.org

9 years agoqemu-char: Convert udp backend to QAPI
Peter Maydell [Tue, 2 Sep 2014 10:24:15 +0000 (11:24 +0100)]
qemu-char: Convert udp backend to QAPI

Convert the udp char backend to the new style QAPI framework.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1409653457-27863-4-git-send-email-peter.maydell@linaro.org

9 years agoutil/qemu-sockets.c: Support specifying IPv4 or IPv6 in socket_dgram()
Peter Maydell [Tue, 2 Sep 2014 10:24:14 +0000 (11:24 +0100)]
util/qemu-sockets.c: Support specifying IPv4 or IPv6 in socket_dgram()

Currently you can specify whether you want a UDP chardev backend
to be IPv4 or IPv6 using the ipv4 or ipv6 options if you use the
QemuOpts parsing code in inet_dgram_opts(). However the QMP struct
parsing code in socket_dgram() doesn't provide this flexibility
(which in turn prevents us from converting the UDP backend handling
to the new style QAPI framework).

Use the existing inet_addr_to_opts() function to convert the
remote->inet address to option strings; this handles ipv4 and
ipv6 flags as well as host and port. (It will also convert any
'to' specification, which is harmless as it is ignored in this
context.)

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1409653457-27863-3-git-send-email-peter.maydell@linaro.org

9 years agoqemu-char: Convert socket backend to QAPI
Peter Maydell [Tue, 2 Sep 2014 10:24:13 +0000 (11:24 +0100)]
qemu-char: Convert socket backend to QAPI

Convert the socket char backend to the new style QAPI framework;
this allows it to return an Error ** to callers who might not
want it to print directly about socket failures.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1409653457-27863-2-git-send-email-peter.maydell@linaro.org

9 years agoMerge remote-tracking branch 'remotes/kraxel/tags/pull-sdl-20140916-1' into staging
Peter Maydell [Tue, 16 Sep 2014 17:29:40 +0000 (18:29 +0100)]
Merge remote-tracking branch 'remotes/kraxel/tags/pull-sdl-20140916-1' into staging

Two minor sdl2 fixes.

# gpg: Signature made Tue 16 Sep 2014 07:20:37 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-sdl-20140916-1:
  sdl2: keymap fixups
  sdl2: drop sdl_zoom.h

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/spice/tags/pull-spice-20140916-2' into staging
Peter Maydell [Tue, 16 Sep 2014 17:28:31 +0000 (18:28 +0100)]
Merge remote-tracking branch 'remotes/spice/tags/pull-spice-20140916-2' into staging

spice: call qemu_spice_set_passwd() during init

# gpg: Signature made Tue 16 Sep 2014 07:11:22 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/spice/tags/pull-spice-20140916-2:
  spice: call qemu_spice_set_passwd() during init

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agospice: call qemu_spice_set_passwd() during init
Marc-André Lureau [Fri, 5 Sep 2014 12:08:48 +0000 (14:08 +0200)]
spice: call qemu_spice_set_passwd() during init

Don't call SPICE API directly to set password given in command line, but
use the internal API, saving password for later calls.

This solves losing password when changing expiration in qemu monitor.

https://bugzilla.redhat.com/show_bug.cgi?id=1138639

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
9 years agosdl2: keymap fixups
Gerd Hoffmann [Tue, 29 Jul 2014 11:52:27 +0000 (13:52 +0200)]
sdl2: keymap fixups

Make a few keys works correctly in SDL2.

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
9 years agosdl2: drop sdl_zoom.h
Gerd Hoffmann [Thu, 24 Jul 2014 11:45:38 +0000 (13:45 +0200)]
sdl2: drop sdl_zoom.h

It isn't used.

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
9 years agoMerge remote-tracking branch 'remotes/qmp-unstable/queue/qmp' into staging
Peter Maydell [Mon, 15 Sep 2014 18:44:34 +0000 (19:44 +0100)]
Merge remote-tracking branch 'remotes/qmp-unstable/queue/qmp' into staging

* remotes/qmp-unstable/queue/qmp:
  exec: file_ram_alloc(): print error when prealloc fails
  monitor: fix debug print compiling error

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging
Peter Maydell [Mon, 15 Sep 2014 16:35:21 +0000 (17:35 +0100)]
Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging

Block patches

# gpg: Signature made Fri 12 Sep 2014 16:09:43 BST using RSA key ID C88F2FD6
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>"

* remotes/kevin/tags/for-upstream: (22 commits)
  qcow2: Add falloc and full preallocation option
  raw-posix: Add falloc and full preallocation option
  qapi: introduce PreallocMode and new PreallocModes full and falloc.
  block: don't convert file size to sector size
  block: round up file size to nearest sector
  iotests: Send the correct fd in socket_scm_helper
  blockdev: Refuse to drive_del something added with blockdev-add
  block: extend BLOCK_IO_ERROR with reason string
  dataplane: fix virtio_blk_data_plane_create() op blocker error path
  qemu-iotests: Run 025 for Archipelago block driver
  block/archipelago: Implement bdrv_truncate()
  block: Make the block accounting functions operate on BlockAcctStats
  block: rename BlockAcctType members to start with BLOCK_ instead of BDRV_
  block: Extract the block accounting code
  block: Extract the BlockAcctStats structure
  IDE: MMIO IDE device control should be little endian
  thread-pool: Drop unnecessary includes
  xen: Drop redundant bdrv_close() from pci_piix3_xen_ide_unplug()
  xen_disk: Plug memory leak on error path
  qemu-io: Clean up openfile() after commit 2e40134
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/kraxel/tags/pull-console-20140915-1' into staging
Peter Maydell [Mon, 15 Sep 2014 15:42:28 +0000 (16:42 +0100)]
Merge remote-tracking branch 'remotes/kraxel/tags/pull-console-20140915-1' into staging

Fix pixman build failure.

# gpg: Signature made Mon 15 Sep 2014 07:15:11 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-console-20140915-1:
  configure: check for pixman-1 version
  pixman: update internal copy to pixman-0.32.6

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoconfigure: check for pixman-1 version
Hu Tao [Mon, 15 Sep 2014 03:41:10 +0000 (11:41 +0800)]
configure: check for pixman-1 version

commit a93a3af9 introduces use of PIXMAN_TYPE_RGBA, but it's only available
in pixman >= 0.21.8. If pixman doesn't meet the version requirement, qemu
will fail to build with following message:

qemu/ui/qemu-pixman.c: In function ‘qemu_pixelformat_from_pixman’:
qemu/ui/qemu-pixman.c:42: error: ‘PIXMAN_TYPE_RGBA’ undeclared (first use in this function)
qemu/ui/qemu-pixman.c:42: error: (Each undeclared identifier is reported only once
qemu/ui/qemu-pixman.c:42: error: for each function it appears in.)

This patch fixes the problem by checking the pixman version.

Signed-off-by: Hu Tao <hutao@cn.fujitsu.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
9 years agopixman: update internal copy to pixman-0.32.6
Hu Tao [Mon, 15 Sep 2014 03:41:09 +0000 (11:41 +0800)]
pixman: update internal copy to pixman-0.32.6

commit a93a3af9 introduces use of PIXMAN_TYPE_RGBA, but it's only available
in pixman >= 0.21.8. Although commit f27b2e1d bumped pixman to pixman-0.28.2,
but the change was reverted later by 7b1b5d19.

This patch updates internal copy of pixman to pixman-0.32.6 to fix the
problem.

Signed-off-by: Hu Tao <hutao@cn.fujitsu.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
9 years agotest-qdev-global-props: Trivial comment fix
Eduardo Habkost [Fri, 8 Aug 2014 19:03:26 +0000 (16:03 -0300)]
test-qdev-global-props: Trivial comment fix

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agohw/machine: Free old values of string properties
Eduardo Habkost [Wed, 6 Aug 2014 18:18:21 +0000 (15:18 -0300)]
hw/machine: Free old values of string properties

Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel.a@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Amos Kong <akong@redhat.com>
Cc: qemu-stable@nongnu.org
9 years agoMerge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into staging
Peter Maydell [Fri, 12 Sep 2014 15:55:48 +0000 (16:55 +0100)]
Merge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into staging

- Memory: improve error reporting and avoid crashes on hotplug
- Build: fixing block/iscsi.so and ranlib warnings on Mac OS X
- Migration fixes for x86
- The odd KVM patch.

# gpg: Signature made Thu 11 Sep 2014 11:21:10 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: (21 commits)
  gdbstub: init mon_chr through qemu_chr_alloc
  pckbd: adding new fields to vmstate
  mc146818rtc: add missed field to vmstate
  piix: do not set irq while loading vmstate
  serial: fixing vmstate for save/restore
  parallel: adding vmstate for save/restore
  fdc: adding vmstate for save/restore
  cpu: init vmstate for ticks and clock offset
  apic_common: vapic_paddr synchronization fix
  vl: use QLIST_FOREACH_SAFE to visit change state handlers
  exec: add parameter errp to gethugepagesize
  exec: report error when memory < hpagesize
  hostmem-ram: don't exit qemu if size of memory-backend-ram is way too big
  memory: add parameter errp to memory_region_init_rom_device
  memory: add parameter errp to memory_region_init_ram
  exec: add parameter errp to qemu_ram_alloc and qemu_ram_alloc_from_ptr
  rules.mak: Fix DSO build by pulling in archive symbols
  util: Don't link host-utils.o if it's empty
  util: Move general qemu_getauxval to util/getauxval.c
  trace: Only link generated-tracers.o with "simple" backend
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoexec: file_ram_alloc(): print error when prealloc fails
Luiz Capitulino [Mon, 8 Sep 2014 17:50:05 +0000 (13:50 -0400)]
exec: file_ram_alloc(): print error when prealloc fails

If memory allocation fails when using the -mem-prealloc command-line
option, QEMU exits without printing any error information to
the user:

 # qemu [...] -m 1G -mem-prealloc -mem-path /dev/hugepages
 # echo $?
 1

This commit adds an error message, so that we print instead:

 # qemu [...] -m 1G -mem-prealloc -mem-path /dev/hugepages
 qemu: unable to map backing store for hugepages: Cannot allocate memory

Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
9 years agomonitor: fix debug print compiling error
Gonglei [Thu, 21 Aug 2014 13:03:09 +0000 (21:03 +0800)]
monitor: fix debug print compiling error

error: 'i' undeclared (first use in this function)

Signed-off-by: Gonglei <arei.gonglei@huawei.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
9 years agoMerge remote-tracking branch 'remotes/pmaydell/tags/pull-target-arm-20140912' into...
Peter Maydell [Fri, 12 Sep 2014 14:12:26 +0000 (15:12 +0100)]
Merge remote-tracking branch 'remotes/pmaydell/tags/pull-target-arm-20140912' into staging

target-arm:
 * add "linux,stdout-path" to the virt DTB
 * fix a long standing bug with IRQ disabling on Cortex-M CPUs
 * implement input interrupt logic in the PL061
 * fix failure to load correct SP/PC on reset of Cortex-M CPUs
   if the vector table is not in a ROM-blob-in-RAM
 * provide flash devices for boot ROMs in the virt board
 * implement architectural watchpoints
 * fix misimplementation of Inner Shareable TLB operations that
   caused instability of guests in TCG SMP configurations
 * configure PL011 and PL031 in the virt board correctly with
   level-triggered interrupts rather than edge-triggered
 * support providing a device tree blob to ROM (firmware)
   images as well as to kernels

# gpg: Signature made Fri 12 Sep 2014 14:19:08 BST using RSA key ID 14360CDE
# gpg: Good signature from "Peter Maydell <peter.maydell@linaro.org>"

* remotes/pmaydell/tags/pull-target-arm-20140912: (23 commits)
  hw/arm/boot: enable DTB support when booting ELF images
  hw/arm/boot: load device tree to base of DRAM if no -kernel option was passed
  hw/arm/boot: pass an address limit to and return size from load_dtb()
  hw/arm/boot: load DTB as a ROM image
  hw/arm/virt: fix pl011 and pl031 irq flags
  target-arm: Make *IS TLB maintenance ops affect all CPUs
  target-arm: Push legacy wildcard TLB ops back into v6
  target-arm: Implement minimal DBGVCR, OSDLR_EL1, MDCCSR_EL0
  target-arm: Remove comment about MDSCR_EL1 being dummy implementation
  target-arm: Set DBGDSCR.MOE for debug exceptions taken to AArch32
  target-arm: Implement handling of fired watchpoints
  target-arm: Move extended_addresses_enabled() to internals.h
  target-arm: Implement setting of watchpoints
  cpu-exec: Make debug_excp_handler a QOM CPU method
  exec.c: Record watchpoint fault address and direction
  exec.c: Provide full set of dummy wp remove functions in user-mode
  exec.c: Relax restrictions on watchpoint length and alignment
  hw/arm/virt: Provide flash devices for boot ROMs
  target-arm: Fix broken indentation in arm_cpu_reest()
  target-arm: Fix resetting issues on ARMv7-M CPUs
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoqcow2: Add falloc and full preallocation option
Hu Tao [Wed, 10 Sep 2014 09:05:49 +0000 (17:05 +0800)]
qcow2: Add falloc and full preallocation option

preallocation=falloc allocates disk space by posix_fallocate(),
preallocation=full allocates disk space by writing zeros to disk.
Both modes imply preallocation=metadata.

Signed-off-by: Hu Tao <hutao@cn.fujitsu.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
9 years agoraw-posix: Add falloc and full preallocation option
Hu Tao [Wed, 10 Sep 2014 09:05:48 +0000 (17:05 +0800)]
raw-posix: Add falloc and full preallocation option

This patch adds a new option preallocation for raw format, and implements
falloc and full preallocation.

Signed-off-by: Hu Tao <hutao@cn.fujitsu.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
9 years agoqapi: introduce PreallocMode and new PreallocModes full and falloc.
Hu Tao [Wed, 10 Sep 2014 09:05:47 +0000 (17:05 +0800)]
qapi: introduce PreallocMode and new PreallocModes full and falloc.

This patch prepares for the subsequent patches.

Signed-off-by: Hu Tao <hutao@cn.fujitsu.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>