This will allow TYPE_KVM_ACCEL objects to be created for querying host
capabilities without affecting global QEMU state.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

On many simple error cases in kvm_init(), we can easily call
error_setg*() instead of printing to stderr and using strerr(), so
convert them.

On some other cases, stderr printing is being kept because it includes a
long explanation for the user, which doesn't seem to be appropriate as
an error_setg() message.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

The new open/close methods will allow basic accel operations that don't
affect global state (such as opening files or querying host
capabilities) to be performed on an accelerator object.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

Properties that will call open/close functions when changed to
true/false are a common pattern in QEMU (e.g. the DeviceState "realized"
property and the RNG backend "opened" property). So to reduce the amount
of boilerplate code needed for those properties, introduce a
object_add_flip_property() helper that will register the property and
call the open/close functions when necessary.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

Instead of returning errors using negative return values, use a Error**
parameter to report errors.

To keep the patch small, existing kvm_init() stderr error reporting code
is being kept, except that sterror(ret) is being used as the error
message before returning. It can be changed later to only call
error_setg() instead of printing to stderr.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

Instead of passing icc_bridge from the PC initialization code to
cpu_x86_create(), make the PC initialization code attach the CPU to
icc_bridge.

The only difference here is that icc_bridge attachment will now be done
after x86_cpu_parse_featurestr() is called. But this shouldn't make any
difference, as property setters shouldn't depend on icc_bridge.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

To allow new code to ask the CPU classes for CPU model information and
allow QOM properties to be queried by qmp_device_list_properties(), we
need to be able to safely instantiate a X86CPU object without any
side-effects.

cpu_exec_init() has lots of side-effects on global QEMU state, move it
to realize so it will be called only if the X86CPU instance is realized.

For reference, this is the current cpu_exec_init() code:

> void cpu_exec_init(CPUArchState *env)
> {
>     CPUState *cpu = ENV_GET_CPU(env);
>     CPUClass *cc = CPU_GET_CLASS(cpu);
>     CPUState *some_cpu;
>     int cpu_index;
>
> #ifndef CONFIG_USER_ONLY
>     cpu->as = &address_space_memory;
>     cpu->thread_id = qemu_get_thread_id();
> #endif

Those fields should be used only after actually starting the VCPU and can be
initialized on realize.

>
> #if defined(CONFIG_USER_ONLY)
>     cpu_list_lock();
> #endif
>     cpu_index = 0;
>     CPU_FOREACH(some_cpu) {
>         cpu_index++;
>     }
>     cpu->cpu_index = cpu_index;
>     QTAILQ_INSERT_TAIL(&cpus, cpu, node);
> #if defined(CONFIG_USER_ONLY)
>     cpu_list_unlock();
> #endif

The above initializes cpu_index and add the CPU to the global CPU list.
This affects QEMU global state and must be done only on realize.

>     if (qdev_get_vmsd(DEVICE(cpu)) == NULL) {
>         vmstate_register(NULL, cpu_index, &vmstate_cpu_common, cpu);
>     }
> #if defined(CPU_SAVE_VERSION) && !defined(CONFIG_USER_ONLY)
>     register_savevm(NULL, "cpu", cpu_index, CPU_SAVE_VERSION,
>                     cpu_save, cpu_load, env);
>     assert(cc->vmsd == NULL);
>     assert(qdev_get_vmsd(DEVICE(cpu)) == NULL);
> #endif
>     if (cc->vmsd != NULL) {
>         vmstate_register(NULL, cpu_index, cc->vmsd, cpu);
>     }

vmstate and savevm registration also affects global QEMU state and should be
done only on realize.

> }

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

To ask the CPU classes for CPU model information and allow QOM
properties to be queried by qmp_device_list_properties(), we need to be
able to safely instantiate a X86CPU object without any side-effects.

Move TCG initialization to realize time so it won't be called when just
doing object_new() on a X86CPU subclass.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
---
Changes v1 -> v2:
 * Now the inited/tcg_initialized variable doesn't exist anymore

Instead of requiring cpu.c to check if TCG was already initialized,
simply let the function be called multiple times.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

Rename the function so that the reason for its existence is clearer: it
does x86-specific initialization of TCG structures.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

Instead of initializing cpu->as and cpu->thread_id while holding
cpu_list_lock(), initialize it earlier.

This allows the code handling cpu_index and global CPU list to be
isolated from the rest.

Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

One small step in the simplification of cpu_exec_init().

Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

QOM objects are already zero-filled when instantiated, there's no need
to explicitly set numa_node to 0.

Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

On softmuu, instead of setting APIC ID automatically when creating a
X86CPU, require the property to be set before realizing the object
(which is already done by the CPU creation code on PC).

Keep apic_id = 0 by default on *-user so it can simply create a new CPU
object and realize it without extra steps (so target-i386 will be able
to use cpu_generic_init() eventually).

Reviewed-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

The APIC ID compatibility code is required only for PC, and now that
x86_cpu_initfn() doesn't use x86_cpu_apic_id_from_index() anymore, that
code can be moved to pc.c.

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

The field doesn't need to be inside CPUX86State, and it is not specific
for the CPUID instruction, so move and rename it.

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

The existing apic_id = cpu_index code has no visible effect: the PC code
already initializes the APIC ID according to the topology on
pc_new_cpu(), and linux-user memcpy()s the CPU state (including
cpuid_apic_id) on cpu_copy().

Remove the dead code and simply let APIC ID to to be 0 by default. This
doesn't change behavior of PC because apic-id is already explicitly set,
and doesn't affect linux-user because APIC ID was already always 0.

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

The function was used in only two places. In one of them, the function
made the code less readable by requiring temporary te[bcd]x variables.
In the other one we can simply inline the existing code.

Reviewed-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

listflags() had lots of unnecessary complexity. Instead of printing to a
buffer that will be immediately printed, simply call the printing
function directly. Also, remove the fbits and flags arguments that were
always set to the same value. Also, there's no need to list the flags in
reverse order.

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

This will allow the PC code to use the header, and lets us eliminate the
QEMU_INCLUDES hack inside tests/Makefile.

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

…302.0' 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>

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>

… 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>

…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>

…request' into staging"

This reverts commit b8a173b, reversing
changes made to 5de0904.

(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>

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>

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>

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>

This makes the error report more informative.

Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

… 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>

…eter' 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>

…' 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>

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>

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>

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>