sdk/emulator/qemu.git
9 years agospapr: Add pseries-2.3 machine
Alexey Kardashevskiy [Fri, 30 Jan 2015 01:53:18 +0000 (12:53 +1100)]
spapr: Add pseries-2.3 machine

The next patch will make MMIO space bigger and keep the old value for
older pseries machines.

Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alexander Graf <agraf@suse.de>
9 years agopseries: Limit PCI host bridge "index" value
David Gibson [Wed, 14 Jan 2015 02:33:39 +0000 (13:33 +1100)]
pseries: Limit PCI host bridge "index" value

pseries guests can have large numbers of PCI host bridges.  To avoid the
user having to specify a number of different configuration values for every
one, the device supports an "index" property which is a shorthand setting
the various window and configuration addresses from a predefined sensible
set.

There are some problems with the details at present:
  * The "index" propery is signed, but negative values will create PCI
windows below where we expect, potentially colliding with other devices
  * No limit is imposed on the "index" property and large values can
translate to extremely large window addresses.  With PCI passthrough in
particular this can mean we exceed various mapping and physical address
limits causing the guest host bridge to not work in strange ways.

This patch addresses this, by making "index" unsigned, and imposing a
limit.  Currently the limit allows indices from 0..255 which is probably
enough host bridges for the time being.  It's fairly easy to extend if
we discover we need more.

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
9 years agotarget-ppc: Use right page size with hash table lookup
Aneesh Kumar K.V [Mon, 26 Jan 2015 14:21:58 +0000 (19:51 +0530)]
target-ppc: Use right page size with hash table lookup

We look at two sizes specified in ISA (4K, 64K). If not found matching,
we consider it 16MB.

Without this patch we would fail to lookup address above 16MB range.
Below 16MB happened to work before because the kernel have a liner
mapping and we always looked up hash for 0xc000000000000000. The
actual real address was computed by using the 16MB offset
with the real address found with the above hash.

Without Fix:
(gdb) x/16x 0xc000000001000000
0xc000000001000000 <list_entries+453208>:       Cannot access memory at address 0xc000000001000000
(gdb)

With Fix:
(gdb)  x/16x 0xc000000001000000
0xc000000001000000 <list_entries+453208>:       0x00000000      0x00000000      0x00000000      0x00000000
0xc000000001000010 <list_entries+453224>:       0x00000000      0x00000000      0x00000000      0x00000000
0xc000000001000020 <list_entries+453240>:       0x00000000      0x00000000      0x00000000      0x00000000
0xc000000001000030 <list_entries+453256>:       0x00000000      0x00000000      0x00000000      0x00000000

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alexander Graf <agraf@suse.de>
9 years agospapr_vio/spapr_iommu: Move VIO bypass where it belongs
Alexey Kardashevskiy [Thu, 29 Jan 2015 05:04:58 +0000 (16:04 +1100)]
spapr_vio/spapr_iommu: Move VIO bypass where it belongs

Instead of tweaking a TCE table device by adding there a bypass flag,
let's add an alias to RAM and IOMMU memory region, and enable/disable
those according to the selected bypass mode.
This way IOMMU memory region can have size of the actual window rather
than ram_size which is essential for upcoming DDW support.

This moves bypass logic to VIO layer and keeps @bypass flag in TCE table
for migration compatibility only. This replaces spapr_tce_set_bypass()
calls with explicit assignment to avoid confusion as the function could
do something more that just syncing the @bypass flag.

This adds a pointer to VIO device into the sPAPRTCETable struct to provide
the sPAPRTCETable device a way to update bypass mode for the VIO device.

Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alexander Graf <agraf@suse.de>
9 years agoMerge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging
Peter Maydell [Mon, 9 Mar 2015 09:14:28 +0000 (09:14 +0000)]
Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging

pci, pc, virtio fixes and cleanups

A bunch of fixes all over the place.
All of ACPI refactoring has been merged.
Legacy pci commands have been dropped.
virtio header cleanup
initial patches from virtio-1.0 branch

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
* remotes/mst/tags/for_upstream: (130 commits)
  acpi: drop unused code
  aml-build: comment fix
  acpi-build: fix typo in comment
  acpi: update generated files
  vhost user:support vhost user nic for non msi guests
  aml-build: fix build for glib < 2.22
  acpi: update generated files
  Makefile.target: binary depends on config-devices
  acpi-test-data: update after pci rewrite
  acpi, mem-hotplug: use PC_DIMM_SLOT_PROP in acpi_memory_plug_cb().
  pci-hotplug-old: Has been dead for five major releases, bury
  pci: Give a few helpers internal linkage
  acpi: make build_*() routines static to aml-build.c
  pc: acpi: remove not used anymore ssdt-[misc|pcihp].hex.generated blobs
  pc: acpi-build: drop template patching and create PCI bus tree dynamically
  tests: ACPI: update pc/SSDT.bridge due to new alg of PCI tree creation
  pc: acpi-build: simplify PCI bus tree generation
  tests: add ACPI blobs for qemu with bridge cases
  tests: bios-tables-test: add support for testing bridges
  tests: ACPI test blobs update due to PCI0._CRS changes
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Conflicts:
hw/pci/pci-hotplug-old.c

9 years agoMerge remote-tracking branch 'remotes/xtensa/tags/20150307-xtensa' into staging
Peter Maydell [Sun, 8 Mar 2015 14:32:38 +0000 (14:32 +0000)]
Merge remote-tracking branch 'remotes/xtensa/tags/20150307-xtensa' into staging

Xtensa updates:

- implement do_unassigned_access callback;
- fix ML605 xtfpga FLASH size.

# gpg: Signature made Sat Mar  7 12:35:05 2015 GMT using RSA key ID F83FA044
# gpg: Good signature from "Max Filippov <max.filippov@cogentembedded.com>"
# gpg:                 aka "Max Filippov <jcmvbkbc@gmail.com>"

* remotes/xtensa/tags/20150307-xtensa:
  target-xtensa: xtfpga: fix ml605 flash size
  target-xtensa: implement do_unassigned_access callback
  hw/xtensa: allow reads/writes in the system I/O region

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/qmp-unstable/queue/qmp' into staging
Peter Maydell [Sun, 8 Mar 2015 12:47:13 +0000 (12:47 +0000)]
Merge remote-tracking branch 'remotes/qmp-unstable/queue/qmp' into staging

* remotes/qmp-unstable/queue/qmp:
  docs: add memory-hotplug.txt
  qemu-options.hx: improve -m description
  virtio-balloon: Add some trace events
  virtio-balloon: Fix balloon not working correctly when hotplug memory
  pc-dimm: add a function to calculate VM's current RAM size

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoacpi: drop unused code
Michael S. Tsirkin [Sun, 8 Mar 2015 10:49:21 +0000 (11:49 +0100)]
acpi: drop unused code

Recent changes left acpi_get_hex unused,
and clag is unhappy about it:
    error: unused function 'acpi_get_hex'

Drop it, as well as some unused macros.

Signer-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoaml-build: comment fix
Michael S. Tsirkin [Sun, 8 Mar 2015 09:36:52 +0000 (10:36 +0100)]
aml-build: comment fix

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoacpi-build: fix typo in comment
Michael S. Tsirkin [Sun, 8 Mar 2015 09:24:42 +0000 (10:24 +0100)]
acpi-build: fix typo in comment

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoMerge remote-tracking branch 'remotes/spice/tags/pull-spice-20150304-1' into staging
Peter Maydell [Sun, 8 Mar 2015 09:47:55 +0000 (09:47 +0000)]
Merge remote-tracking branch 'remotes/spice/tags/pull-spice-20150304-1' into staging

misc spice/qxl fixes.

# gpg: Signature made Wed Mar  4 13:57:42 2015 GMT 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-20150304-1:
  hmp: info spice: take out webdav
  hmp: info spice: Show string channel name
  qxl: drop update_displaychangelistener call for secondary qxl devices
  vga: refactor vram_size clamping and rounding
  qxl: refactor rounding up to a nearest power of 2
  spice: fix invalid memory access to vga.vram
  qxl: document minimal video memory for new modes

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/kraxel/tags/pull-usb-20150303-1' into staging
Peter Maydell [Sun, 8 Mar 2015 07:33:45 +0000 (07:33 +0000)]
Merge remote-tracking branch 'remotes/kraxel/tags/pull-usb-20150303-1' into staging

xhci: generate a Transfer Event for each Transfer TRB with the IOC bit set

# gpg: Signature made Tue Mar  3 07:38:43 2015 GMT 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-20150303-1:
  xhci: generate a Transfer Event for each Transfer TRB with the IOC bit set

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/gonglei/tags/bootdevice-next-20150303' into...
Peter Maydell [Sun, 8 Mar 2015 06:43:32 +0000 (06:43 +0000)]
Merge remote-tracking branch 'remotes/gonglei/tags/bootdevice-next-20150303' into staging

bootdevice: bug fixes

# gpg: Signature made Tue Mar  3 05:18:39 2015 GMT using RSA key ID DDE30FBB
# gpg: Good signature from "Gonglei <arei.gonglei@huawei.com>"
# gpg: WARNING: This key is not certified with a trusted signature!
# gpg:          There is no indication that the signature belongs to the owner.
# Primary key fingerprint: 5178 9C82 617F 2F58 8693  63B1 BA7A 65B0 DDE3 0FBB

* remotes/gonglei/tags/bootdevice-next-20150303:
  bootdevice: add check in restore_boot_order()
  bootdevice: check boot order argument validation before vm running

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/bkoppelmann/tags/pull-tricore-20150303' into...
Peter Maydell [Sun, 8 Mar 2015 00:16:27 +0000 (00:16 +0000)]
Merge remote-tracking branch 'remotes/bkoppelmann/tags/pull-tricore-20150303' into staging

TriCore RRR1, RRR2 instructions and bugfixes

# gpg: Signature made Tue Mar  3 01:12:02 2015 GMT using RSA key ID 6B69CA14
# gpg: Good signature from "Bastian Koppelmann <kbastian@mail.uni-paderborn.de>"

* remotes/bkoppelmann/tags/pull-tricore-20150303:
  target-tricore: Add instructions of RRR1 opcode format, which have 0xc3 as first opcode
  target-tricore: Add instructions of RRR1 opcode format, which have 0x43 as first opcode
  target-tricore: Add instructions of RRR1 opcode format, which have 0x83 as first opcode
  target-tricore: Add instructions of RRR2 opcode format
  target-tricore: fix msub32_suov return wrong results
  target-tricore: Fix RLC_ADDI, RLC_ADDIH using wrong microcode helper

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agotarget-xtensa: xtfpga: fix ml605 flash size
Max Filippov [Mon, 16 Feb 2015 19:30:21 +0000 (22:30 +0300)]
target-xtensa: xtfpga: fix ml605 flash size

ML605 daughterboard has 16MB flash, not 32MB. Change board definition
accordingly.

Cc: qemu-stable@nongnu.org
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
9 years agotarget-xtensa: implement do_unassigned_access callback
Max Filippov [Wed, 12 Feb 2014 10:35:56 +0000 (14:35 +0400)]
target-xtensa: implement do_unassigned_access callback

Depending on access type raise either InstrPIFDataError or
LoadStorePIFDataError exception.

Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
9 years agohw/xtensa: allow reads/writes in the system I/O region
Max Filippov [Mon, 17 Feb 2014 16:57:45 +0000 (20:57 +0400)]
hw/xtensa: allow reads/writes in the system I/O region

Ignore writes to unassigned areas of system I/O regison and return 0 for
reads. This makes drivers for unimportant unimplemented hardware blocks
happy.

Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
9 years agodocs: add memory-hotplug.txt
Luiz Capitulino [Thu, 26 Feb 2015 19:43:07 +0000 (14:43 -0500)]
docs: add memory-hotplug.txt

This document describes how to use memory hotplug in QEMU.

Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Paulo Vital <paulo.vital@profitbricks.com>
9 years agoqemu-options.hx: improve -m description
Luiz Capitulino [Thu, 26 Feb 2015 19:35:45 +0000 (14:35 -0500)]
qemu-options.hx: improve -m description

Add memory hotplug options to the command-line format. Also,
add a complete command-line example and improve description.

Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Paulo Vital <paulo.vital@profitbricks.com>
9 years agovirtio-balloon: Add some trace events
zhanghailiang [Mon, 17 Nov 2014 05:11:10 +0000 (13:11 +0800)]
virtio-balloon: Add some trace events

Add some trace events for easier debugging

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
9 years agovirtio-balloon: Fix balloon not working correctly when hotplug memory
zhanghailiang [Mon, 17 Nov 2014 05:11:09 +0000 (13:11 +0800)]
virtio-balloon: Fix balloon not working correctly when hotplug memory

When do memory balloon, it takes the 'ram_size' as the VM's current ram size,
But 'ram_size' is the startup configured ram size, it does not take into
account the hotplugged memory.

As a result, the balloon result will be confused.
Steps to reproduce:
(1)Start VM: qemu -m size=1024,slots=4,maxmem=8G
(2)In VM: #free -m : 1024M
(3)qmp balloon 512M
(4)In VM: #free -m : 512M
(5)hotplug pc-dimm 1G
(6)In VM: #free -m : 1512M
(7)qmp balloon 256M
(8)In VM: #free -m :1256M
We expect the VM's available ram size to be 256M after 'qmp balloon 256M'
command, but VM's real available ram size is 1256M.

For "qmp balloon" is not performance critical code, we use function
'get_current_ram_size' to get VM's current ram size.

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
9 years agopc-dimm: add a function to calculate VM's current RAM size
zhanghailiang [Mon, 17 Nov 2014 05:11:08 +0000 (13:11 +0800)]
pc-dimm: add a function to calculate VM's current RAM size

The global parameter 'ram_size' does not take into account
the hotplugged memory.

In some codes, we use 'ram_size' as current VM's real RAM size,
which is not correct.

Add function 'get_current_ram_size' to calculate VM's current RAM size,
it will enumerate present memory devices and also plus ram_size.

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
9 years agoMerge remote-tracking branch 'remotes/stefanha/tags/tracing-pull-request' into staging
Peter Maydell [Wed, 4 Mar 2015 15:33:05 +0000 (15:33 +0000)]
Merge remote-tracking branch 'remotes/stefanha/tags/tracing-pull-request' into staging

# gpg: Signature made Mon Mar  2 21:45:18 2015 GMT using RSA key ID 81AB73C8
# gpg: Good signature from "Stefan Hajnoczi <stefanha@redhat.com>"
# gpg:                 aka "Stefan Hajnoczi <stefanha@gmail.com>"

* remotes/stefanha/tags/tracing-pull-request:
  trace: add DTrace reserved words for .d files
  unbreak dtrace tracing due to double _ in rdma names

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoacpi: update generated files
Michael S. Tsirkin [Wed, 4 Mar 2015 15:05:32 +0000 (16:05 +0100)]
acpi: update generated files

Fixes up build on systems without iasl.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoMerge remote-tracking branch 'remotes/awilliam/tags/vfio-update-20150302.0' into...
Peter Maydell [Wed, 4 Mar 2015 14:37:31 +0000 (14:37 +0000)]
Merge remote-tracking branch 'remotes/awilliam/tags/vfio-update-20150302.0' into staging

Updates for QEMU 2.3-rc0:
 - Error reporting and static cleanup (Alexey Kardashevskiy)
 - Runtime mmap disable for tracing (Samuel Pitoiset)
 - Support for host directed device request (Alex Williamson)

# gpg: Signature made Mon Mar  2 18:42:50 2015 GMT using RSA key ID 3BB08B22
# gpg: Good signature from "Alex Williamson <alex.williamson@redhat.com>"
# gpg:                 aka "Alex Williamson <alex@shazbot.org>"
# gpg:                 aka "Alex Williamson <alwillia@redhat.com>"
# gpg:                 aka "Alex Williamson <alex.l.williamson@gmail.com>"

* remotes/awilliam/tags/vfio-update-20150302.0:
  vfio-pci: Enable device request notification support
  vfio: allow to disable MMAP per device with -x-mmap=off option
  vfio: Make type1 listener symbols static
  vfio: Add ioctl number to error report

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agohmp: info spice: take out webdav
Gerd Hoffmann [Tue, 3 Mar 2015 08:27:28 +0000 (09:27 +0100)]
hmp: info spice: take out webdav

Obvious suggestion for the next spice-protocol
release: Add some way to #ifdef new stuff.

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Cole Robinson <crobinso@redhat.com>
9 years agovhost user:support vhost user nic for non msi guests
Haifeng Gao [Sat, 28 Feb 2015 09:02:02 +0000 (17:02 +0800)]
vhost user:support vhost user nic for non msi guests

Currently, vhost user nic doesn't support non msi guests(like pxe stage) by default.
Vhost user nic can't fall back to qemu like normal vhost net nic does. So we should
enable it for non msi guests.

Signed-off-by: Haifeng Gao <gaohaifeng.gao@huawei.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoaml-build: fix build for glib < 2.22
Michael S. Tsirkin [Tue, 3 Mar 2015 16:40:45 +0000 (17:40 +0100)]
aml-build: fix build for glib < 2.22

g_ptr_array_new_with_free_func is there since glib 2.22,
use the older g_ptr_array_foreach instead.

Reported-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agovl: take iothread lock very early
Paolo Bonzini [Tue, 3 Mar 2015 10:36:09 +0000 (11:36 +0100)]
vl: take iothread lock very early

If the iothread lock isn't taken by the main thread, the RCU callbacks
might run concurrently with the main thread.  QEMU's not ready for that.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Tested-by: Christian Borntraeger <borntraeger@de.ibm.com>
Tested-by: Gonglei <arei.gonglei@huawei.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into staging
Peter Maydell [Tue, 3 Mar 2015 12:07:47 +0000 (12:07 +0000)]
Merge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into staging

- more config options
- bootdevice, iscsi, virtio-scsi fixes
- build system patches for MinGW and config-devices.mak
- qemu_mutex_lock_iothread deadlock fixes
- another tiny patch from the record/replay series

# gpg: Signature made Mon Mar  2 09:59:14 2015 GMT using RSA key ID 78C7AE83
# gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>"
# gpg:                 aka "Paolo Bonzini <pbonzini@redhat.com>"
# gpg: WARNING: This key is not certified with a trusted signature!
# gpg:          There is no indication that the signature belongs to the owner.
# Primary key fingerprint: 46F5 9FBD 57D6 12E7 BFD4  E2F7 7E15 100C CD36 69B1
#      Subkey fingerprint: F133 3857 4B66 2389 866C  7682 BFFB D25F 78C7 AE83

* remotes/bonzini/tags/for-upstream:
  cpus: be more paranoid in avoiding deadlocks
  cpus: fix deadlock and segfault in qemu_mutex_lock_iothread
  virtio-scsi: Allocate op blocker reason before blocking
  Makefile.target: binary depends on config-devices
  Makefile: don't silence mak file test with V=1
  Makefile: fix up parallel building under MSYS+MinGW
  iscsi: Handle write protected case in reopen
  Give ivshmem its own config option
  Create specific config option for "platform-bus"
  Add specific config options for PCI-E bridges
  bootdevice: fix segment fault when booting guest with '-kernel' and '-initrd'
  timer: replace time() with QEMU_CLOCK_HOST
  virtio-scsi-dataplane: Call blk_set_aio_context within BQL
  block: Forbid bdrv_set_aio_context outside BQL
  scsi: give device a parent before setting properties

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoxhci: generate a Transfer Event for each Transfer TRB with the IOC bit set
Laszlo Ersek [Mon, 2 Mar 2015 16:02:53 +0000 (17:02 +0100)]
xhci: generate a Transfer Event for each Transfer TRB with the IOC bit set

At the moment, when the XHCI driver in edk2
(MdeModulePkg/Bus/Pci/XhciDxe/XhciDxe.inf) runs on QEMU, with the options

  -device nec-usb-xhci -device usb-kbd

it crashes with:

  ASSERT MdeModulePkg/Bus/Pci/XhciDxe/XhciSched.c(1759):
  TrsRing != ((void*) 0)

The crash hits in the following edk2 call sequence (all files under
MdeModulePkg/Bus/):

UsbEnumerateNewDev()                         [Usb/UsbBusDxe/UsbEnumer.c]
  UsbBuildDescTable()                        [Usb/UsbBusDxe/UsbDesc.c]
    UsbGetDevDesc()                          [Usb/UsbBusDxe/UsbDesc.c]
      UsbCtrlGetDesc(USB_REQ_GET_DESCRIPTOR) [Usb/UsbBusDxe/UsbDesc.c]
        UsbCtrlRequest()                     [Usb/UsbBusDxe/UsbDesc.c]
          UsbHcControlTransfer()             [Usb/UsbBusDxe/UsbUtility.c]
            XhcControlTransfer()             [Pci/XhciDxe/Xhci.c]
              XhcCreateUrb()                 [Pci/XhciDxe/XhciSched.c]
                XhcCreateTransferTrb()       [Pci/XhciDxe/XhciSched.c]
              XhcExecTransfer()              [Pci/XhciDxe/XhciSched.c]
                XhcCheckUrbResult()          [Pci/XhciDxe/XhciSched.c]
                  //
                  // look for TRB_TYPE_DATA_STAGE event [1]
                  //
              //
              // Store a copy of the device descriptor, as the hub device
              // needs this info to configure endpoint. [2]
              //
  UsbSetConfig()                             [Usb/UsbBusDxe/UsbDesc.c]
    UsbCtrlRequest(USB_REQ_SET_CONFIG)       [Usb/UsbBusDxe/UsbDesc.c]
      UsbHcControlTransfer()                 [Usb/UsbBusDxe/UsbUtility.c]
        XhcControlTransfer()                 [Pci/XhciDxe/Xhci.c]
          XhcSetConfigCmd()                  [Pci/XhciDxe/XhciSched.c]
            XhcInitializeEndpointContext()   [Pci/XhciDxe/XhciSched.c]
              //
              // allocate transfer ring for the endpoint [3]
              //

USBKeyboardDriverBindingStart()              [Usb/UsbKbDxe/EfiKey.c]
  UsbIoAsyncInterruptTransfer()              [Usb/UsbBusDxe/UsbBus.c]
    UsbHcAsyncInterruptTransfer()            [Usb/UsbBusDxe/UsbUtility.c]
      XhcAsyncInterruptTransfer()            [Pci/XhciDxe/Xhci.c]
        XhcCreateUrb()                       [Pci/XhciDxe/Xhci.c]
          XhcCreateTransferTrb()             [Pci/XhciDxe/XhciSched.c]
            XhcSyncTrsRing()                 [Pci/XhciDxe/XhciSched.c]
              ASSERT (TrsRing != NULL) [4]

UsbEnumerateNewDev() in the USB bus driver issues a GET_DESCRIPTOR
request, in order to determine the number of configurations that the
endpoint supports. The requests consists of three stages (three TRBs),
setup, data, and status. The length of the response is determined in [1],
namely from the transfer event that the host controller generates in
response to the request's middle stage (ie. the data stage).

If the length of the answer is correct (a full GET_DESCRIPTOR request
takes 18 bytes), then the XHCI driver that underlies the USB bus driver
"snoops" (caches) the descriptor data for later [2].

Later, the USB bus driver sends a SET_CONFIG request. The underlying XHCI
driver allocates a transfer ring for the endpoint, relying on the data
snooped and cached in step [2].

Finally, the USB keyboard driver submits an asynchronous interrupt
transfer to manage the keyboard. As part of this it asserts [4] that the
ring has been allocated in step [3].

And this ASSERT() fires. The root cause can be found in the way QEMU
handles the initial GET_DESCRIPTOR request.

Again, that request consists of three stages (TRBs, Transfer Request
Blocks), "setup", "data", and "status". The XhcCreateTransferTrb()
function sets the IOC ("Interrupt on Completion") flag in each of these
TRBs.

According to the XHCI specification, the host controller shall generate a
Transfer Event in response to *each* individual TRB of the request that
had the IOC flag set. This means that QEMU should queue three events:
setup, data, and status, for edk2's XHCI driver.

However, QEMU only generates two events:
- one for the setup (ie. 1st) stage,
- another for the status (ie. 3rd) stage.

No event is generated for the middle (ie. data) stage. The loop in QEMU's
xhci_xfer_report() function runs three times, but due to the "reported"
variable, only the first and the last TRBs elicit events, the middle (data
stage) results in no event queued.

As a consequence:
- When handling the GET_DESCRIPTOR request, XhcCheckUrbResult() in [1]
  does not update the response length from zero.

- XhcControlTransfer() thinks that the response is invalid (it has zero
  length payload instead of 18 bytes), hence [2] is not reached; the
  device descriptor is not stashed for later, and the number of possible
  configurations is left at zero.

- When handling the SET_CONFIG request, (NumConfigurations == 0) from
  above prevents the allocation of the endpoint's transfer ring.

- When the keyboard driver tries to use the endpoint, the ASSERT() blows
  up.

The solution is to correct the emulation in QEMU, and to generate a
transfer event whenever IOC is set in a TRB.

The patch replaces

  !reported && (IOC || foo)    == !reported && IOC ||
                                  !reported && foo

with

  IOC || (!reported && foo)    == IOC ||
                                  !reported && foo

which only changes how

  reported && IOC

is handled. (Namely, it now generates an event.)

Tested with edk2 built for "qemu-system-aarch64 -M virt" (ie.
"ArmVirtualizationQemu.dsc", aka "AAVMF"), and guest Linux.

Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
9 years agohmp: info spice: Show string channel name
Cole Robinson [Sun, 1 Mar 2015 14:29:18 +0000 (09:29 -0500)]
hmp: info spice: Show string channel name

Useful for debugging.

https://bugzilla.redhat.com/show_bug.cgi?id=822418
Signed-off-by: Cole Robinson <crobinso@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
9 years agoqxl: drop update_displaychangelistener call for secondary qxl devices
Gerd Hoffmann [Mon, 2 Mar 2015 16:01:50 +0000 (17:01 +0100)]
qxl: drop update_displaychangelistener call for secondary qxl devices

Commit 3dcadce5076d4b42fa395c39662d65e050b77784 added three
update_displaychangelistener call sites:

Two for primary qxl cards, when entering/leaving vga mode, which are
correct.

One for secondary qxl cards, which is wrong because we don't register
a displaychangelistener in the first place for secondary cards.

Remove it.

Reported-by: Brad Campbell <lists2009@fnarfbargle.com>
Tested-by: Brad Campbell <lists2009@fnarfbargle.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
9 years agovga: refactor vram_size clamping and rounding
Radim Krčmář [Tue, 17 Feb 2015 16:30:53 +0000 (17:30 +0100)]
vga: refactor vram_size clamping and rounding

Make the code a bit more obvious.

We don't have min/max, so a general helper for clamp probably isn't
acceptable either.

Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
9 years agoqxl: refactor rounding up to a nearest power of 2
Radim Krčmář [Tue, 17 Feb 2015 16:30:52 +0000 (17:30 +0100)]
qxl: refactor rounding up to a nearest power of 2

We already have pow2floor, mirror it and use instead of a function with
similar results (same in used domain), to clarify our intent.

Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
9 years agospice: fix invalid memory access to vga.vram
Radim Krčmář [Tue, 17 Feb 2015 16:30:51 +0000 (17:30 +0100)]
spice: fix invalid memory access to vga.vram

vga_common_init() doesn't allow more than 256 MiB vram size and silently
shrinks any larger value.  qxl_dirty_surfaces() used the unshrinked size
via qxl->shadow_rom.surface0_area_size when accessing the memory, which
resulted in segfault.

Add a workaround for this case and an assert if it happens again.

We have to bump the vga memory limit too, because 256 MiB wouldn't have
allowed 8k (it requires more than 128 MiB).
1024 MiB doesn't work, but 512 MiB seems fine.

Proposed-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
9 years agoqxl: document minimal video memory for new modes
Radim Krčmář [Tue, 17 Feb 2015 16:30:50 +0000 (17:30 +0100)]
qxl: document minimal video memory for new modes

The alternative to removing existing comments.

Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
9 years agobootdevice: add check in restore_boot_order()
Gonglei [Thu, 29 Jan 2015 13:13:47 +0000 (13:13 +0000)]
bootdevice: add check in restore_boot_order()

qemu_boot_set() can't fail in restore_boot_order(),
then simply assert it doesn't fail, by passing
&error_abort if boot_set_handler set.

Suggested-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Gonglei <arei.gonglei@huawei.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
9 years agobootdevice: check boot order argument validation before vm running
Gonglei [Tue, 3 Feb 2015 11:31:09 +0000 (11:31 +0000)]
bootdevice: check boot order argument validation before vm running

Either 'once' option or 'order' option can take effect for -boot at
the same time, that is say initial startup processing can check only
one. And pc.c's set_boot_dev() fails when its boot order argument
is invalid. This patch provide a solution fix this problem:

 1. If "once" is given, register reset handler to restore boot order.

 2. Pass the normal boot order to machine creation.  Should fail when
   the normal boot order is invalid.

 3. If "once" is given, set it with qemu_boot_set().  Fails when the
   once boot order is invalid.

 4. Start the machine.

 5. On reset, the reset handler calls qemu_boot_set() to restore boot
   order.  Should never fail.

Suggested-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Gonglei <arei.gonglei@huawei.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
9 years agoMerge remote-tracking branch 'remotes/armbru/tags/pull-error-2015-02-26' into staging
Peter Maydell [Tue, 3 Mar 2015 01:28:54 +0000 (01:28 +0000)]
Merge remote-tracking branch 'remotes/armbru/tags/pull-error-2015-02-26' into staging

QemuOpts: Convert various setters to Error

# gpg: Signature made Thu Feb 26 13:56:43 2015 GMT using RSA key ID EB918653
# gpg: Good signature from "Markus Armbruster <armbru@redhat.com>"
# gpg:                 aka "Markus Armbruster <armbru@pond.sub.org>"

* remotes/armbru/tags/pull-error-2015-02-26:
  qtest: Use qemu_opt_set() instead of qemu_opts_parse()
  pc: Use qemu_opt_set() instead of qemu_opts_parse()
  qemu-sockets: Simplify setting numeric and boolean options
  block: Simplify setting numeric options
  qemu-img: Suppress unhelpful extra errors in convert, amend
  QemuOpts: Propagate errors through opts_parse()
  QemuOpts: Propagate errors through opts_do_parse()
  QemuOpts: Drop qemu_opt_set(), rename qemu_opt_set_err(), fix use
  block: Suppress unhelpful extra errors in bdrv_img_create()
  qemu-img: Suppress unhelpful extra errors in convert, resize
  QemuOpts: Convert qemu_opts_set() to Error, fix its use
  QemuOpts: Convert qemu_opt_set_number() to Error, fix its use
  QemuOpts: Convert qemu_opt_set_bool() to Error, fix its use

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agotarget-tricore: Add instructions of RRR1 opcode format, which have 0xc3 as first...
Bastian Koppelmann [Fri, 6 Feb 2015 14:48:33 +0000 (14:48 +0000)]
target-tricore: Add instructions of RRR1 opcode format, which have 0xc3 as first opcode

Add helpers helper_addsur_h/_ssov which adds one halfword and subtracts one
halfword, rounds / and saturates each half word independently.

Add microcode helper functions:
    * gen_maddsu_h/sus_h: multiply two halfwords left justified and add to the
                          first one word and subtract from the second one word
                          / and saturate each resulting word independetly.
    * gen_maddsum_h/sums_h: multiply two halfwords in q-format left justified
                            and add to the first one word and subtract from
                            the second one word / and saturate each resulting
                            word independetly.
    * gen_maddsur32_h/32s_h: multiply two halfwords in q-format left justified
                             and add to the first one word and subtract from
                             the second one word, round both results / and
                             saturate each resulting word independetly.

Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
Reviewed-by: Richard Henderson <rth@twiddle.net>
9 years agotarget-tricore: Add instructions of RRR1 opcode format, which have 0x43 as first...
Bastian Koppelmann [Tue, 3 Feb 2015 18:36:53 +0000 (18:36 +0000)]
target-tricore: Add instructions of RRR1 opcode format, which have 0x43 as first opcode

Add helpers:
    * madd64_q_ssov: multiply two 32 bit q-format number, add them with a
                     64 bit q-format number and saturate.
    * madd32_q_add_ssov: add two 64 bit q-format numbers and return a 32 bit
                         result.
    * maddr_q_ssov: multiplay two 32 bit q-format numbers, add a 32 bit
                    q-format number and saturate.
    * maddr_q: multiplay two 32 bit q-format numbers and add a 32 bit
               q-format number.

Note: madd instructions in the q format can behave strange, e.g.
0x1 + (0x80000000 * 0x80000000) << 1 for 32 bit signed values does not cause an
overflow on the guest, because all intermediate results should be handled as if
they are indefinitely precise. We handle this by inverting the overflow bit for
all cases: a + (0x80000000 * 0x80000000) << 1.

Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
Reviewed-by: Richard Henderson <rth@twiddle.net>
9 years agotarget-tricore: Add instructions of RRR1 opcode format, which have 0x83 as first...
Bastian Koppelmann [Thu, 29 Jan 2015 15:35:56 +0000 (15:35 +0000)]
target-tricore: Add instructions of RRR1 opcode format, which have 0x83 as first opcode

Add helpers:
    * add64_ssov: adds two 64 bit values and saturates the result.
    * addr_h/_ssov: adds two halfwords with two words in q-format with rounding
                    / and saturates each result independetly.

Add microcode generator:
    * gen_add64_d: adds two 64 bit values.
    * gen_addsub64_h: adds/subtracts one halfwords with a word and adds/
                      subtracts another halftword with another word.
    * gen_madd_h/s_h: multiply four halfwords, add each result left justfied
                      to two word values / and saturate each result.
    * gen_maddm_h/s_h: multiply four halfwords, add each result left justfied
                       to two words values in q-format / and saturate each
                       result.
    * gen_maddr32/64_h/s_h: multiply four halfwords, add each result left
                            justfied to two halftwords/words values in q-format
                            / and saturate each result.

Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
Reviewed-by: Richard Henderson <rth@twiddle.net>
9 years agotarget-tricore: Add instructions of RRR2 opcode format
Bastian Koppelmann [Wed, 21 Jan 2015 14:57:51 +0000 (14:57 +0000)]
target-tricore: Add instructions of RRR2 opcode format

Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
Reviewed-by: Richard Henderson <rth@twiddle.net>
9 years agotarget-tricore: fix msub32_suov return wrong results
Bastian Koppelmann [Wed, 28 Jan 2015 12:15:05 +0000 (12:15 +0000)]
target-tricore: fix msub32_suov return wrong results

If the signed result of the multiplication overflows, we would get a negative
value, which would result in a addition instead of a subtraction.

Now we do the overflow calculation and saturation by hand instead of using
suov32_neg.

Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
Reviewed-by: Richard Henderson <rth@twiddle.net>
9 years agotarget-tricore: Fix RLC_ADDI, RLC_ADDIH using wrong microcode helper
Bastian Koppelmann [Fri, 6 Feb 2015 15:06:05 +0000 (15:06 +0000)]
target-tricore: Fix RLC_ADDI, RLC_ADDIH using wrong microcode helper

Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
Reviewed-by: Richard Henderson <rth@twiddle.net>
9 years agoRevert "Merge remote-tracking branch 'remotes/ehabkost/tags/x86-pull-request' into...
Peter Maydell [Tue, 3 Mar 2015 00:29:17 +0000 (00:29 +0000)]
Revert "Merge remote-tracking branch 'remotes/ehabkost/tags/x86-pull-request' into staging"

This reverts commit b8a173b25c887a606681fc35a46702c164d5b2d0, reversing
changes made to 5de090464f1ec5360c4f30faa01d8a9f8826cd58.

(I applied this pull request when I should not have done so, and
am now immediately reverting it.)

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agotrace: add DTrace reserved words for .d files
Stefan Hajnoczi [Thu, 14 Aug 2014 10:39:23 +0000 (11:39 +0100)]
trace: add DTrace reserved words for .d files

DTrace on Mac OS X fails due to trace events using 'self' as an argument
name:

  GEN   trace/generated-tracers-dtrace.h
dtrace: failed to compile script trace/generated-tracers-dtrace.dtrace: line 1330: syntax error, unexpected DT_KEY_SELF, expecting ) near "self"
make: *** [trace/generated-tracers-dtrace.h] Error 1

Filter argument names according to the list of DTrace .d file reserved
keywords.

Note that DTrace on Mac and Linux still do not work after this patch.
There are additional build issues remaining.

Reported-by: Henk Poley <henkpoley@gmail.com>
Tested-by: Henk Poley <henkpoley@gmail.com>
Cc: Lluís Vilanova <vilanova@ac.upc.edu>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agounbreak dtrace tracing due to double _ in rdma names
Dr. David Alan Gilbert [Mon, 16 Feb 2015 16:58:05 +0000 (16:58 +0000)]
unbreak dtrace tracing due to double _ in rdma names

It looks like the dtrace trace code gets upset if you have trace names
with __ in, which the migration/rdma.c code does.

Rename the functions and the associated traces.

Fixes: 733252deb8b7d37beacda5976c2769e18642b2fa

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reported-by: Andreas Färber <afaerber@suse.de>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Tested-by: Andreas Färber <afaerber@suse.de>
Message-id: 1424105885-12149-1-git-send-email-dgilbert@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agovfio-pci: Enable device request notification support
Alex Williamson [Mon, 2 Mar 2015 18:38:55 +0000 (11:38 -0700)]
vfio-pci: Enable device request notification support

Linux v4.0-rc1 vfio-pci introduced a new virtual interrupt to allow
the kernel to request a device from the user.  When signaled, QEMU
will by default attmempt to hot-unplug the device.  This is a one-
shot attempt with the expectation that the kernel will continue to
poll for the device if it is not returned.  Returning the device when
requested is the expected standard model of cooperative usage, but we
also add an option option to disable this feature.  Initially this
opt-out is set as an experimental option because we really should
honor kernel requests for the device.

Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
9 years agovfio: allow to disable MMAP per device with -x-mmap=off option
Samuel Pitoiset [Mon, 2 Mar 2015 18:38:55 +0000 (11:38 -0700)]
vfio: allow to disable MMAP per device with -x-mmap=off option

Disabling MMAP support uses the slower read/write accesses but allows to
trace all MMIO accesses, which is not good for performance, but very
useful for reverse engineering PCI drivers. This option allows to
disable MMAP per device without a compile-time change.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
9 years agovfio: Make type1 listener symbols static
Alexey Kardashevskiy [Mon, 2 Mar 2015 18:38:55 +0000 (11:38 -0700)]
vfio: Make type1 listener symbols static

They are not used from anywhere but common.c which is where these are
defined so make them static.

Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
9 years agovfio: Add ioctl number to error report
Alexey Kardashevskiy [Mon, 2 Mar 2015 18:38:54 +0000 (11:38 -0700)]
vfio: Add ioctl number to error report

This makes the error report more informative.

Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
9 years agoacpi: update generated files
Michael S. Tsirkin [Mon, 2 Mar 2015 15:14:33 +0000 (16:14 +0100)]
acpi: update generated files

Fixes up build on systems without iasl.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoMerge remote-tracking branch 'remotes/ehabkost/tags/x86-pull-request' into staging
Peter Maydell [Mon, 2 Mar 2015 14:25:48 +0000 (14:25 +0000)]
Merge remote-tracking branch 'remotes/ehabkost/tags/x86-pull-request' into staging

* remotes/ehabkost/tags/x86-pull-request:
  target-i386: Move APIC ID compatibility code to pc.c
  target-i386: Require APIC ID to be explicitly set before CPU realize
  target-i386: Set APIC ID using cpu_index on CONFIG_USER
  linux-user: Check for cpu_init() errors
  target-i386: Move CPUX86State.cpuid_apic_id to X86CPU.apic_id
  target-i386: Simplify error handling on cpu_x86_init_user()
  target-i386: Eliminate cpu_init() function
  target-i386: Rename cpu_x86_init() to cpu_x86_init_user()
  target-i386: Move topology.h to include/hw/i386
  target-i386: Eliminate unnecessary get_cpuid_vendor() function
  target-i386: Simplify listflags() function

Conflicts:
target-i386/cpu.c

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/afaerber/tags/qom-devices-for-peter' into staging
Peter Maydell [Mon, 2 Mar 2015 13:20:43 +0000 (13:20 +0000)]
Merge remote-tracking branch 'remotes/afaerber/tags/qom-devices-for-peter' into staging

QOM infrastructure fixes and device conversions

* Assertion fix for device_add with non-device types
* Documentation fix
* qdev_init() error reporting cleanups

# gpg: Signature made Tue Feb 24 13:56:33 2015 GMT 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-peter:
  parallel: parallel_hds_isa_init() shouldn't fail
  parallel: Factor out common parallel_hds_isa_init()
  serial: serial_hds_isa_init() shouldn't fail
  serial: Factor out common serial_hds_isa_init()
  etsec: Replace qdev_init() by qdev_init_nofail()
  leon3: Replace unchecked qdev_init() by qdev_init_nofail()
  ide/isa: Replace unchecked qdev_init() by qdev_init_nofail()
  qdev: Improve qdev_init_nofail()'s error reporting
  qom: Fix typo, 'my_class_init' -> 'derived_class_init'
  qdev: Avoid type assertion in qdev_build_hotpluggable_device_list()

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/ehabkost/tags/numa-pull-request' into staging
Peter Maydell [Mon, 2 Mar 2015 12:13:45 +0000 (12:13 +0000)]
Merge remote-tracking branch 'remotes/ehabkost/tags/numa-pull-request' into staging

NUMA fixes queue

# gpg: Signature made Mon Feb 23 19:28:42 2015 GMT using RSA key ID 984DC5A6
# gpg: Can't check signature: public key not found

* remotes/ehabkost/tags/numa-pull-request:
  numa: Rename set_numa_modes() to numa_post_machine_init()
  numa: Rename option parsing functions
  numa: Move QemuOpts parsing to set_numa_nodes()
  numa: Make max_numa_nodeid static
  numa: Move NUMA globals to numa.c
  vl.c: Remove unnecessary zero-initialization of NUMA globals
  numa: Move NUMA declarations from sysemu.h to numa.h

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agocpus: be more paranoid in avoiding deadlocks
Paolo Bonzini [Fri, 27 Feb 2015 19:01:03 +0000 (20:01 +0100)]
cpus: be more paranoid in avoiding deadlocks

For good measure, ensure that the following sequence:

   thread 1 calls qemu_mutex_lock_iothread
   thread 2 calls qemu_mutex_lock_iothread
   VCPU thread are created
   VCPU thread enters execution loop

results in the VCPU threads letting the other two threads run
and obeying iothread_requesting_mutex even if the VCPUs are
not halted.  To do this, check iothread_requesting_mutex
before execution starts.

Tested-by: Leon Alrae <leon.alrae@imgtec.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agocpus: fix deadlock and segfault in qemu_mutex_lock_iothread
Paolo Bonzini [Fri, 27 Feb 2015 18:58:23 +0000 (19:58 +0100)]
cpus: fix deadlock and segfault in qemu_mutex_lock_iothread

When two threads (other than the low-priority TCG VCPU thread)
are competing for the iothread lock, a deadlock can happen.  This
is because iothread_requesting_mutex is set to false by the first
thread that gets the mutex, and then the VCPU thread might never
yield from the execution loop.  If iothread_requesting_mutex is
changed from a bool to a counter, the deadlock is fixed.

However, there is another bug in qemu_mutex_lock_iothread that
can be triggered by the new call_rcu thread.  The bug happens
if qemu_mutex_lock_iothread is called before the CPUs are
created.  In that case, first_cpu is NULL and the caller
segfaults in qemu_mutex_lock_iothread.  To fix this, just
do not do the kick if first_cpu is NULL.

Reported-by: Leon Alrae <leon.alrae@imgtec.com>
Reported-by: Andreas Gustafsson <gson@gson.org>
Tested-by: Leon Alrae <leon.alrae@imgtec.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agovirtio-scsi: Allocate op blocker reason before blocking
Max Reitz [Fri, 27 Feb 2015 17:11:53 +0000 (12:11 -0500)]
virtio-scsi: Allocate op blocker reason before blocking

s->blocker is really only used in hw/scsi/virtio-scsi.c; the only places
where it is used in hw/scsi/virtio-scsi-dataplane.c is when it is
allocated and when it is freed. That does not make a whole lot of sense
(and is actually wrong because this leads to s->blocker potentially
being NULL when blk_op_block_all() is called in virtio-scsi.c), so move
the allocation and destruction of s->blocker to the device realization
and unrealization in virtio-scsi.c, respectively.

Case in point:

$ echo -e 'eject drv\nquit' | \
    x86_64-softmmu/qemu-system-x86_64 \
        -monitor stdio -machine accel=qtest -display none \
        -object iothread,id=thr -device virtio-scsi-pci,iothread=thr \
        -drive if=none,file=test.qcow2,format=qcow2,id=drv \
        -device scsi-cd,drive=drv

Without this patch:

(qemu) eject drv
[1]    10102 done
       10103 segmentation fault (core dumped)

With this patch:

(qemu) eject drv
Device 'drv' is busy: block device is in use by data plane
(qemu) quit

Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <1425057113-26940-1-git-send-email-mreitz@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoMakefile.target: binary depends on config-devices
Michael S. Tsirkin [Thu, 19 Feb 2015 07:48:52 +0000 (08:48 +0100)]
Makefile.target: binary depends on config-devices

relink binary whenever config-devices.mak changes:
this makes sense as we are adding/removing devices,
so binary has to be relinked to be up to date.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoacpi-test-data: update after pci rewrite
Michael S. Tsirkin [Sun, 1 Mar 2015 18:39:36 +0000 (19:39 +0100)]
acpi-test-data: update after pci rewrite

more trivial changes as more code has been rewritten in C.
we also got rid of extra Scope operators.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoacpi, mem-hotplug: use PC_DIMM_SLOT_PROP in acpi_memory_plug_cb().
Tang Chen [Thu, 26 Feb 2015 01:16:43 +0000 (09:16 +0800)]
acpi, mem-hotplug: use PC_DIMM_SLOT_PROP in acpi_memory_plug_cb().

Replace string "slot" in acpi_memory_plug_cb() with macro PC_DIMM_SLOT_PROP.

Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Tang Chen <tangchen@cn.fujitsu.com>
Signed-off-by: Zhu Guihua <zhugh.fnst@cn.fujitsu.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agopci-hotplug-old: Has been dead for five major releases, bury
Markus Armbruster [Thu, 26 Feb 2015 16:21:13 +0000 (17:21 +0100)]
pci-hotplug-old: Has been dead for five major releases, bury

Commit 79ca616 (v1.6.0) accidentally disabled legacy x86-only HMP
commands pci_add, pci_del: it defined CONFIG_PCI_HOTPLUG only as make
variable, not as preprocessor macro, killing the code conditional on
defined(CONFIG_PCI_HOTPLUG_OLD).

In all this time, nobody reported the loss.  I only noticed it when I
tried to test some error reporting change that forced me to touch this
old crap again.

Fun: git-log hw/pci/pci-hotplug-old.c shows our faith in the backward
compatibility god has been strong enough to sacrifice at its altar
about a dozen times, but not strong enough to even once verify the
legacy feature's still there, let alone works.

Remove the commands along with the code backing them.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agopci: Give a few helpers internal linkage
Markus Armbruster [Thu, 26 Feb 2015 16:21:14 +0000 (17:21 +0100)]
pci: Give a few helpers internal linkage

None of them should be used in new code.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoacpi: make build_*() routines static to aml-build.c
Igor Mammedov [Fri, 20 Feb 2015 18:22:20 +0000 (18:22 +0000)]
acpi: make build_*() routines static to aml-build.c

build_*() routines were used for composing AML
structures manually in acpi-build.c but after
conversion to AML API they are not used outside
of aml-build.c anymore, so hide them from external
users.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agopc: acpi: remove not used anymore ssdt-[misc|pcihp].hex.generated blobs
Igor Mammedov [Fri, 20 Feb 2015 18:22:19 +0000 (18:22 +0000)]
pc: acpi: remove not used anymore ssdt-[misc|pcihp].hex.generated blobs

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agopc: acpi-build: drop template patching and create PCI bus tree dynamically
Igor Mammedov [Fri, 20 Feb 2015 18:22:18 +0000 (18:22 +0000)]
pc: acpi-build: drop template patching and create PCI bus tree dynamically

Replace AML template patching with direct composing
of PCI device entries in C. It allows to simplify
PCI tree generation further and saves us about 400LOC
scattered through different files, confining tree
generation to one C function which is much easier
to deal with.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agotests: ACPI: update pc/SSDT.bridge due to new alg of PCI tree creation
Igor Mammedov [Fri, 20 Feb 2015 18:22:17 +0000 (18:22 +0000)]
tests: ACPI: update pc/SSDT.bridge due to new alg of PCI tree creation

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agopc: acpi-build: simplify PCI bus tree generation
Igor Mammedov [Fri, 20 Feb 2015 18:22:16 +0000 (18:22 +0000)]
pc: acpi-build: simplify PCI bus tree generation

it basicaly does the same as original approach,
* just without bus/notify tables tracking (less obscure)
  which is easier to follow.
* drops unnecessary loops and bitmaps,
  creating devices and notification method in the same loop.
* saves us ~100LOC

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agotests: add ACPI blobs for qemu with bridge cases
Igor Mammedov [Fri, 20 Feb 2015 18:22:15 +0000 (18:22 +0000)]
tests: add ACPI blobs for qemu with bridge cases

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agotests: bios-tables-test: add support for testing bridges
Igor Mammedov [Fri, 20 Feb 2015 18:22:14 +0000 (18:22 +0000)]
tests: bios-tables-test: add support for testing bridges

Adds alternative ACPI table blob selection for testing
non default QEMU configurations. If blob file for test
variant is not present, fallback to default blob.

With this change implement testing with a coldplugged
bridge.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agotests: ACPI test blobs update due to PCI0._CRS changes
Igor Mammedov [Fri, 20 Feb 2015 18:22:13 +0000 (18:22 +0000)]
tests: ACPI test blobs update due to PCI0._CRS changes

PCI0._CRS was moved into SSDT and became the same for
PIIX4/Q35 machines.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agopc: acpi-build: drop template patching and create Device(SMC) dynamically
Igor Mammedov [Fri, 20 Feb 2015 18:22:12 +0000 (18:22 +0000)]
pc: acpi-build: drop template patching and create Device(SMC) dynamically

patch moves SMC device into SSDT and creates it only
when device is present, which makes ACPI tables smaller
in default case when device is not present.

Also it fixes wrong IO range in CRS if "iobase"
property is set to a non default value.

PS:
Testing with XP shows that current default "iobase"
used SMC device conflicts with floppy controller IO,
but it's topic for another patch and I'd leave it
to SMC device author for resolving conflict.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
CC: agraf@suse.de
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agopc: export applesmc IO port/len
Igor Mammedov [Fri, 20 Feb 2015 18:22:11 +0000 (18:22 +0000)]
pc: export applesmc IO port/len

IO port and length will be used in following patch
to correctly generate SMC ACPI device in SSDT.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoacpi: add acpi_irq_no_flags() term
Igor Mammedov [Fri, 20 Feb 2015 18:22:10 +0000 (18:22 +0000)]
acpi: add acpi_irq_no_flags() term

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agopc: acpi-build: drop remaining ssdt_misc template
Igor Mammedov [Fri, 20 Feb 2015 18:22:09 +0000 (18:22 +0000)]
pc: acpi-build: drop remaining ssdt_misc template

It drops empty ssdt_misc templete. It also hides
from user almost all pointer arithmetic when building
SSDT which makes resulting code a bit cleaner
and concentrating only on composing ASL construct
/i.e. a task build_ssdt() should be doing/.

Also it makes one binary blob less stored in QEMU
source tree by removing need to keep and update
hw/i386/ssdt-misc.hex.generated file here in total
saving us ~430LOC.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agopc: acpi: drop manual hole punching for GPE0 resources
Igor Mammedov [Fri, 20 Feb 2015 18:22:08 +0000 (18:22 +0000)]
pc: acpi: drop manual hole punching for GPE0 resources

Drops manual hole punching in PCI0._CRS on PIIX4 machine type
for GPE0 resources. Resources will be consumed by Device(GPE0)
that is attached to PCI namespace.
There is GPE device with HID ACPI0006 since ACPI2.0
that should be used for this purpose but none of Windows
versions support it and show it as "unknown device",
so reserve resource in old fashioned way with PNP0A06
device to make windows happy and actually reserve resources.

Along with last hole _CRS layout of PIIX4 machine becomes
the same as Q35 one, so merge them together and use the same
_CRS for both machine types.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agopc: acpi: drop manual hole punching for CPU hotplug resources
Igor Mammedov [Fri, 20 Feb 2015 18:22:07 +0000 (18:22 +0000)]
pc: acpi: drop manual hole punching for CPU hotplug resources

Drops manual hole punching in PCI0._CRS on PIIX4 machine type
for CPU hotplug resources.
Resources will be consumed by Device(PRES) that is attached
to PCI bus. The same way how it currently works for mem hotlpug.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agopc: acpi: drop manual hole punching for PCI hotplug resources
Igor Mammedov [Fri, 20 Feb 2015 18:22:06 +0000 (18:22 +0000)]
pc: acpi: drop manual hole punching for PCI hotplug resources

Drops manual hole punching in PCI0._CRS for PIIX4 machine type.
Resources will be consumed by Device(PHPR) that cwis attached
to PCI bus. The same way how it currently works for mem hotlpug.

Manual hole in PIIX4 _CRS wasn't correct anyway since it was
legacy size 0xF while current PCIHP MMIO region is of size 0x14.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agopc: acpi-build: create PCI0._CRS dynamically
Igor Mammedov [Fri, 20 Feb 2015 18:22:05 +0000 (18:22 +0000)]
pc: acpi-build: create PCI0._CRS dynamically

Replace template patching and runtime calculation
in _CRS() method with static _CRS defined in SSDT.
No functional change except of as mentined above
and _CRS being moved from DSDT to SSDT.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agotpm: Support for capability flags of TIS 1.3
Stefan Berger [Mon, 23 Feb 2015 14:27:20 +0000 (09:27 -0500)]
tpm: Support for capability flags of TIS 1.3

Provide the TIS 1.3 capability flags.
The interface now looks like a TIS 1.3 interface. It's fully
compatible with previous TIS 1.2 and drivers written for
TIS 1.2 continue to work.

Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agotpm: Support for TIS selftest done flag
Stefan Berger [Mon, 23 Feb 2015 14:27:19 +0000 (09:27 -0500)]
tpm: Support for TIS selftest done flag

Extend the backend to check whether the TPM_ContinueSelfTest
finished successfully and provide a flag to the TIS front-end
if it successfully finished. The TIS then sets a flag in
all localities in the STS register and keeps it until the next
reset.

Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agotpm: Support for XFIFO register
Stefan Berger [Mon, 23 Feb 2015 14:27:18 +0000 (09:27 -0500)]
tpm: Support for XFIFO register

Support for the XFIFO register (range) of the TIS 1.3 specification.
We support a range of 64 bytes.

Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agotpm: Allow 32 & 16 bit accesses to the registers
Stefan Berger [Mon, 23 Feb 2015 14:27:17 +0000 (09:27 -0500)]
tpm: Allow 32 & 16 bit accesses to the registers

Improve the access to the registers with 32 and 16 bit reads and writes.
Also enable access to a non-base register address, such as reads of the
2nd byte of a register. Map the FIFO byte access to any byte within
its 4 byte register (following specs).

Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agotpm: Extend sts register to 32 bit
Stefan Berger [Mon, 23 Feb 2015 14:27:16 +0000 (09:27 -0500)]
tpm: Extend sts register to 32 bit

More recent TIS specs extend the STS register to 32 bit. While
we don't store the TIS interface state, yet, we can extend it
without sideeffects.

Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoballoon: call qdev_alias_all_properties for proxy dev in balloon class init
Denis V. Lunev [Thu, 29 Jan 2015 14:24:41 +0000 (17:24 +0300)]
balloon: call qdev_alias_all_properties for proxy dev in balloon class init

The idea is that all other virtio devices are calling this helper
to merge properties of the proxy device. This is the only difference
in between this helper and code in inside virtio_instance_init_common.
The patch should not cause any harm as property list in generic balloon
code is empty.

This also allows to avoid some dummy errors like fixed by this
    commit 91ba21208839643603e7f7fa5864723c3f371ebe
    Author: Gonglei <arei.gonglei@huawei.com>
    Date:   Tue Sep 30 14:10:35 2014 +0800
    virtio-balloon: fix virtio-balloon child refcount in transports

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: Raushaniya Maksudova <rmaksudova@parallels.com>
Revieved-by: Cornelia Huck <cornelia.huck@de.ibm.com>
CC: Christian Borntraeger <borntraeger@de.ibm.com>
CC: Anthony Liguori <aliguori@amazon.com>
CC: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoMakefile.target: binary depends on config-devices
Michael S. Tsirkin [Thu, 19 Feb 2015 07:48:52 +0000 (08:48 +0100)]
Makefile.target: binary depends on config-devices

relink binary whenever config-devices.mak changes:
this makes sense as we are adding/removing devices,
so binary has to be relinked to be up to date.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <1424332114-13440-2-git-send-email-mst@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoMakefile: don't silence mak file test with V=1
Michael S. Tsirkin [Thu, 19 Feb 2015 07:48:46 +0000 (08:48 +0100)]
Makefile: don't silence mak file test with V=1

V=1 should show what's going on, it's not nice
to silence things unconditionally.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <1424332114-13440-1-git-send-email-mst@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoMakefile: fix up parallel building under MSYS+MinGW
Vasily Efimov [Wed, 18 Feb 2015 12:59:37 +0000 (15:59 +0300)]
Makefile: fix up parallel building under MSYS+MinGW

This patch enables parallel building of QEMU in MSYS+MinGW environment.
Currently an attempt to build QEMU in parallel fails on generation of
version.lo (and version.o too).

The cause of the failure is that when listing prerequisites "Makefile"
references "config-host.h" by absolute path in some rules and by relative
path in others. Make cannot figure out that these references points to the
same file which leads to the race: the generation of "version.*" which
requires "$(BUILD_DIR)/config-host.h" is launched in parallel with the
generation of "config-host.h" needed by other "Makefile" targets.

This patch removes "$(BUILD_DIR)/" prefix from corresponding prerequisite
of "version.*". There is no other prerequisites "$(BUILD_DIR)/config-host.h"
found.

Also note that not every version of MSYS is able to build QEMU in parallel,
see: "http://sourceforge.net/p/mingw/bugs/1950/". The suggested version is
1.0.17.

Signed-off-by: Vasily Efimov <real@ispras.ru>
Message-Id: <1424264377-5992-1-git-send-email-real@ispras.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoiscsi: Handle write protected case in reopen
Fam Zheng [Wed, 25 Feb 2015 04:40:08 +0000 (12:40 +0800)]
iscsi: Handle write protected case in reopen

Save the write protected flag and check before reopen.

Signed-off-by: Fam Zheng <famz@redhat.com>
Message-Id: <1424839208-5195-1-git-send-email-famz@redhat.com>
[Fixed typo in the name of the new field. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoGive ivshmem its own config option
David Gibson [Fri, 27 Feb 2015 06:04:37 +0000 (17:04 +1100)]
Give ivshmem its own config option

Currently the ivshmem device is built whenever both PCI and KVM support are
included.  This patch gives it its own config option to allow easier
customization of whether to include it.  It's enabled by default in the
same circumstances as now - when both PCI and KVM are available.

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Peter Crosthwaite <peter.crosthwaite@xilinx.com>
Message-Id: <1425017077-18487-4-git-send-email-david@gibson.dropbear.id.au>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoCreate specific config option for "platform-bus"
David Gibson [Fri, 27 Feb 2015 06:04:36 +0000 (17:04 +1100)]
Create specific config option for "platform-bus"

Currently the "platform-bus" device is included for all softmmu builds.
This bridge is intended for use on any platforms that require dynamic
creation of sysbus devices.  However, at present it is used only for the
PPC E500 target, with plans for the ARM "virt" target in the immediate
future.

To avoid a not-very-useful entry appearing in "qemu -device ?" output on
other targets, this patch makes a specific config option for platform-bus
and enables it (for now) only on ppc configurations which include E500
and on ARM (which always includes the "virt" target).

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Message-Id: <1425017077-18487-3-git-send-email-david@gibson.dropbear.id.au>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoAdd specific config options for PCI-E bridges
David Gibson [Fri, 27 Feb 2015 06:04:35 +0000 (17:04 +1100)]
Add specific config options for PCI-E bridges

The i82801b11, ioh3420 and xio3130 PCI Express devices are currently
included in the build unconditionally.

While they could theoretically appear on any target platform with PCI-E,
they're pretty unlikely to appear on platforms that aren't Intel derived.

Therefore, to avoid presenting unlikely-to-be-relevant devices to the user,
add config options to enable these components, and enable them by default
only on x86 and arm platforms.

(Note that this patch does include these for aarch64, via its inclusion of
arm-softmmu.mak).

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Peter Crosthwaite <peter.crosthwaite@xilinx.com>
Message-Id: <1425017077-18487-2-git-send-email-david@gibson.dropbear.id.au>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agobootdevice: fix segment fault when booting guest with '-kernel' and '-initrd'
Gonglei [Fri, 27 Feb 2015 01:49:44 +0000 (09:49 +0800)]
bootdevice: fix segment fault when booting guest with '-kernel' and '-initrd'

Reproducer:

 $./qemu-system-x86_64 --enable-kvm -kernel /home/vmlinuz-2.6.32.12-0.7-default \
  -initrd /home/initrd-2.6.32.12-0.7-default -append \
 "root=/dev/ram rw console=ttyS0,115200" -dtb guest.dtb -vnc :10 --monitor stdio -smp 2
QEMU 2.2.50 monitor - type 'help' for more information
(qemu) Segmentation fault (core dumped)

Reported-by: Edivaldo de Araujo Pereira <edivaldoapereira@yahoo.com.br>
Signed-off-by: Gonglei <arei.gonglei@huawei.com>
Message-Id: <1425001784-6752-1-git-send-email-arei.gonglei@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agotimer: replace time() with QEMU_CLOCK_HOST
Pavel Dovgalyuk [Fri, 27 Feb 2015 13:11:02 +0000 (16:11 +0300)]
timer: replace time() with QEMU_CLOCK_HOST

This patch replaces time() function calls with calls to
qemu_clock_get_ns(QEMU_CLOCK_HOST). It makes such requests deterministic
in record/replay mode of icount.

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Pavel Dovgalyuk <Pavel.Dovgaluk@ispras.ru>
Message-Id: <20150227131102.11912.89850.stgit@PASHA-ISP>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agovirtio-scsi-dataplane: Call blk_set_aio_context within BQL
Fam Zheng [Sun, 15 Feb 2015 03:06:31 +0000 (11:06 +0800)]
virtio-scsi-dataplane: Call blk_set_aio_context within BQL

It's not safe to call blk_set_aio_context from outside BQL because of
the bdrv_drain_all there. Let's put it in the hotplug callback which
will be called by qdev device realization for each scsi device attached
to the bus.

Signed-off-by: Fam Zheng <famz@redhat.com>
Message-Id: <1423969591-23646-3-git-send-email-famz@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoblock: Forbid bdrv_set_aio_context outside BQL
Fam Zheng [Sun, 15 Feb 2015 03:06:30 +0000 (11:06 +0800)]
block: Forbid bdrv_set_aio_context outside BQL

Even if the caller has both the old and the new AioContext's, there can
be a deadlock, due to the leading bdrv_drain_all.

Suppose there are four io threads (A, B, A0, B0) with A and B owning a
BDS for each (bs_a, bs_b); Now A wants to move bs_a to iothread A0, and
B wants to move bs_b to B0, at the same time:

  iothread A                           iothread B
--------------------------------------------------------------------------
  aio_context_acquire(A0) /* OK */     aio_context_acquire(B0) /* OK */
  bdrv_set_aio_context(bs_a, A0)       bdrv_set_aio_context(bs_b, B0)
  -> bdrv_drain_all()                  -> bdrv_drain_all()
     -> acquire A /* OK */               -> acquire A /* blocked */
     -> acquire B /* blocked */          -> acquire B
     ...                                 ...

Deadlock happens because A is waiting for B, and B is waiting for A.

Signed-off-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <1423969591-23646-2-git-send-email-famz@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoscsi: give device a parent before setting properties
Paolo Bonzini [Thu, 19 Feb 2015 16:05:46 +0000 (17:05 +0100)]
scsi: give device a parent before setting properties

This mimics what is done in qdev_device_add, and lets the device be
freed in case something goes wrong.  Otherwise, object_unparent returns
immediately without freeing the device, which is on the other hand left
in the parent bus's list of children.

scsi_bus_legacy_handle_cmdline then returns an error, and the HBA is
destroyed as well with object_unparent.  But the lingering device that
was not removed in scsi_bus_legacy_add_drive cannot be removed now either,
and bus_unparent gets stuck in an infinite loop trying to empty the list
of children.

The right fix of course would be to assert in bus_add_child that the
device already has a bus, and remove the "safety net" that adds the
drive to the QOM tree in device_set_realized.  I am not yet sure whether
that would entail changing all callers to qdev_create (as well as
isa_create and usb_create and the corresponding _try_create versions).

Reported-by: Markus Armbruster <armbru@redhat.com>
Tested-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoqtest: Use qemu_opt_set() instead of qemu_opts_parse()
Markus Armbruster [Fri, 13 Feb 2015 14:48:19 +0000 (15:48 +0100)]
qtest: Use qemu_opt_set() instead of qemu_opts_parse()

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>