qemu.git
13 months agoamd_iommu: Use TYPE_AMD_IOMMU_PCI constant
Eduardo Habkost [Wed, 26 Aug 2020 18:43:29 +0000 (14:43 -0400)] 
amd_iommu: Use TYPE_AMD_IOMMU_PCI constant

This will make future conversion to use OBJECT_DECLARE* easier.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Reviewed-by: Li Qiang <liq3ea@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20200826184334.4120620-4-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agonios2_iic: Use TYPE_ALTERA_IIC constant
Eduardo Habkost [Wed, 26 Aug 2020 18:43:28 +0000 (14:43 -0400)] 
nios2_iic: Use TYPE_ALTERA_IIC constant

This will make future conversion to use OBJECT_DECLARE* easier.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Reviewed-by: Li Qiang <liq3ea@gmail.com>
Message-Id: <20200826184334.4120620-3-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoetsec: Use TYPE_ETSEC_COMMON constant
Eduardo Habkost [Wed, 26 Aug 2020 18:43:27 +0000 (14:43 -0400)] 
etsec: Use TYPE_ETSEC_COMMON constant

This will make future conversion to use OBJECT_DECLARE* easier.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Li Qiang <liq3ea@gmail.com>
Message-Id: <20200826184334.4120620-2-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agomigration: Rename class type checking macros
Eduardo Habkost [Tue, 25 Aug 2020 19:20:46 +0000 (15:20 -0400)] 
migration: Rename class type checking macros

Rename the macros to make them consistent with the MIGRATION_OBJ
macro name.

This will make future conversion to OBJECT_DECLARE* easier.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-51-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoswim: Rename struct SWIM to Swim
Eduardo Habkost [Tue, 25 Aug 2020 19:20:45 +0000 (15:20 -0400)] 
swim: Rename struct SWIM to Swim

Currently we have a SWIM typedef and a SWIM type checking macro,
but OBJECT_DECLARE* would transform the SWIM macro into a
function, and the function name would conflict with the SWIM
typedef name.

Rename the struct and typedef to "Swim". This will make future
conversion to OBJECT_DECLARE* easier.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Acked-by: Laurent Vivier <laurent@vivier.eu>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-50-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agos390-virtio-ccw: Rename S390_MACHINE_CLASS macro
Eduardo Habkost [Tue, 25 Aug 2020 19:20:44 +0000 (15:20 -0400)] 
s390-virtio-ccw: Rename S390_MACHINE_CLASS macro

Rename it to be consistent with S390_CCW_MACHINE and
TYPE_S390_CCW_MACHINE.

This will make future conversion to OBJECT_DECLARE* easier.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-49-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agonubus: Rename class type checking macros
Eduardo Habkost [Tue, 25 Aug 2020 19:20:39 +0000 (15:20 -0400)] 
nubus: Rename class type checking macros

Rename the existing class type checking macros to be consistent
with the type name and instance type checking macro.  Use a
NUBUS_MACFB prefix instead of MACFB_NUBUS.

This will make future conversion to OBJECT_DECLARE* easier.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Acked-by: Laurent Vivier <laurent@vivier.eu>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-44-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agovfio/pci: Move QOM macros to header
Eduardo Habkost [Tue, 25 Aug 2020 19:20:38 +0000 (15:20 -0400)] 
vfio/pci: Move QOM macros to header

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-43-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agokvm: Move QOM macros to kvm.h
Eduardo Habkost [Tue, 25 Aug 2020 19:20:37 +0000 (15:20 -0400)] 
kvm: Move QOM macros to kvm.h

Move QOM macros close to the KVMState typedef.

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-42-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agomptsas: Move QOM macros to header
Eduardo Habkost [Tue, 25 Aug 2020 19:20:36 +0000 (15:20 -0400)] 
mptsas: Move QOM macros to header

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-41-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agopxa2xx: Move QOM macros to header
Eduardo Habkost [Tue, 25 Aug 2020 19:20:35 +0000 (15:20 -0400)] 
pxa2xx: Move QOM macros to header

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-40-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agorocker: Move QOM macros to header
Eduardo Habkost [Tue, 25 Aug 2020 19:20:34 +0000 (15:20 -0400)] 
rocker: Move QOM macros to header

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-39-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoauxbus: Move QOM macros to header
Eduardo Habkost [Tue, 25 Aug 2020 19:20:33 +0000 (15:20 -0400)] 
auxbus: Move QOM macros to header

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-38-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agopiix: Move QOM macros to header
Eduardo Habkost [Tue, 25 Aug 2020 19:20:32 +0000 (15:20 -0400)] 
piix: Move QOM macros to header

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-37-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agovirtio-serial-bus: Move QOM macros to header
Eduardo Habkost [Tue, 25 Aug 2020 19:20:31 +0000 (15:20 -0400)] 
virtio-serial-bus: Move QOM macros to header

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-36-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agovmbus: Move QOM macros to vmbus.h
Eduardo Habkost [Tue, 25 Aug 2020 19:20:30 +0000 (15:20 -0400)] 
vmbus: Move QOM macros to vmbus.h

Move all declarations related to TYPE_VMBUS to the same place in
vmbus.h.

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-35-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agopckbd: Move QOM macro to header
Eduardo Habkost [Tue, 25 Aug 2020 19:20:29 +0000 (15:20 -0400)] 
pckbd: Move QOM macro to header

Move the I8042 macro close to the TYPE_I8042 define.

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-34-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoahci: Move QOM macro to header
Eduardo Habkost [Tue, 25 Aug 2020 19:20:28 +0000 (15:20 -0400)] 
ahci: Move QOM macro to header

Move the ALLWINNER_AHCI macro close to the TYPE_ALLWINNER_AHCI
define.

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-33-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoi8257: Move QOM macro to header
Eduardo Habkost [Tue, 25 Aug 2020 19:20:27 +0000 (15:20 -0400)] 
i8257: Move QOM macro to header

Move the I8257 macro to i8257.h, close to the TYPE_I8257 define.

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-32-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoahci: Move QOM macros to header
Eduardo Habkost [Tue, 25 Aug 2020 19:20:26 +0000 (15:20 -0400)] 
ahci: Move QOM macros to header

The TYPE_* constants and the typedefs are defined in ahci.h, so
we can move the type checking macros there too.

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-31-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agovhost-user-gpu: Move QOM macro to header
Eduardo Habkost [Tue, 25 Aug 2020 19:20:25 +0000 (15:20 -0400)] 
vhost-user-gpu: Move QOM macro to header

Move the VHOST_USER_GPU type checking macro to virtio-gpu.h,
close to the TYPE_VHOST_USER_GPU #define.

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-30-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agos390x: Move typedef SCLPEventFacility to event-facility.h
Eduardo Habkost [Tue, 25 Aug 2020 19:20:24 +0000 (15:20 -0400)] 
s390x: Move typedef SCLPEventFacility to event-facility.h

This will make future conversion to OBJECT_DECLARE* easier.

In sclp.h, use "struct SCLPEventFacility" to avoid introducing
unnecessary header dependencies.

Acked-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-29-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agospapr: Move typedef SpaprMachineState to spapr.h
Eduardo Habkost [Tue, 25 Aug 2020 19:20:23 +0000 (15:20 -0400)] 
spapr: Move typedef SpaprMachineState to spapr.h

Move the typedef from spapr_irq.h to spapr.h, and use "struct
SpaprMachineState" in the spapr_*.h headers (to avoid circular
header dependencies).

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-28-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoxen-legacy-backend: Add missing typedef XenLegacyDevice
Eduardo Habkost [Tue, 25 Aug 2020 19:20:22 +0000 (15:20 -0400)] 
xen-legacy-backend: Add missing typedef XenLegacyDevice

The typedef was used in the XENBACKEND_DEVICE macro, but it was
never defined.  Define the typedef close to the type checking
macro.

Acked-by: Anthony PERARD <anthony.perard@citrix.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-27-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoarmsse: Rename QOM macros to avoid conflicts
Eduardo Habkost [Tue, 25 Aug 2020 19:20:21 +0000 (15:20 -0400)] 
armsse: Rename QOM macros to avoid conflicts

Rename TYPE_ARMSSE to TYPE_ARM_SSE, and ARMSSE*() type checking
macros to ARM_SSE*().

This will avoid a future conflict between an ARM_SSE() type
checking macro and the ARMSSE typedef name.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-26-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoplatform-bus: Delete macros for non-existing typedef
Eduardo Habkost [Tue, 25 Aug 2020 19:20:20 +0000 (15:20 -0400)] 
platform-bus: Delete macros for non-existing typedef

PlatformBusDeviceClass doesn't exist.  This will break when we
automatically convert the code to use OBJECT_DEFINE_TYPE().
Delete the macros that reference the non-existing typedef.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-25-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agonubus: Delete unused NUBUS_BRIDGE macro
Eduardo Habkost [Tue, 25 Aug 2020 19:20:19 +0000 (15:20 -0400)] 
nubus: Delete unused NUBUS_BRIDGE macro

The macro never worked because the NubusBridge typedef doesn't
exist.  Delete it.

Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Acked-by: Laurent Vivier <laurent@vivier.eu>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-24-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agocan_emu: Delete macros for non-existing typedef
Eduardo Habkost [Tue, 25 Aug 2020 19:20:18 +0000 (15:20 -0400)] 
can_emu: Delete macros for non-existing typedef

CanBusClass doesn't exist.  This will break when we automatically
convert the code to use OBJECT_DEFINE_TYPE().  Delete the macros
that reference the non-existing typedef.

Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-23-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agos390_flic: Move KVMS390FLICState typedef to header
Eduardo Habkost [Tue, 25 Aug 2020 19:20:17 +0000 (15:20 -0400)] 
s390_flic: Move KVMS390FLICState typedef to header

Move typedef closer to the type check macros, to make it easier
to convert the code to OBJECT_DEFINE_TYPE() in the future.

Acked-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-22-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agomcf_fec: Move mcf_fec_state typedef to header
Eduardo Habkost [Tue, 25 Aug 2020 19:20:16 +0000 (15:20 -0400)] 
mcf_fec: Move mcf_fec_state typedef to header

Move typedef closer to the type check macros, to make it easier
to convert the code to OBJECT_DEFINE_TYPE() in the future.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-21-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agohvf: Move HVFState typedef to hvf.h
Eduardo Habkost [Tue, 25 Aug 2020 19:20:15 +0000 (15:20 -0400)] 
hvf: Move HVFState typedef to hvf.h

Move typedef closer to the type check macros, to make it easier
to convert the code to OBJECT_DEFINE_TYPE() in the future.

Reviewed-by: Roman Bolshakov <r.bolshakov@yadro.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-20-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoi8254: Move PITCommonState/PITCommonClass typedefs to i8254.h
Eduardo Habkost [Tue, 25 Aug 2020 19:20:14 +0000 (15:20 -0400)] 
i8254: Move PITCommonState/PITCommonClass typedefs to i8254.h

Move typedef closer to the type check macros, to make it easier
to convert the code to OBJECT_DEFINE_TYPE() in the future.

Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-19-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agopci: Move PCIBusClass typedef to pci.h
Eduardo Habkost [Tue, 25 Aug 2020 19:20:13 +0000 (15:20 -0400)] 
pci: Move PCIBusClass typedef to pci.h

Move typedef closer to the type check macros, to make it easier
to convert the code to OBJECT_DEFINE_TYPE() in the future.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-18-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agothrottle-groups: Move ThrottleGroup typedef to header
Eduardo Habkost [Tue, 25 Aug 2020 19:20:12 +0000 (15:20 -0400)] 
throttle-groups: Move ThrottleGroup typedef to header

Move typedef closer to the type check macros, to make it easier
to convert the code to OBJECT_DEFINE_TYPE() in the future.

Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-17-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agotulip: Move TulipState typedef to header
Eduardo Habkost [Tue, 25 Aug 2020 19:20:11 +0000 (15:20 -0400)] 
tulip: Move TulipState typedef to header

Move typedef closer to the type check macros, to make it easier
to convert the code to OBJECT_DEFINE_TYPE() in the future.

Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-16-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agohcd-dwc2: Rename USB_*CLASS macros for consistency
Eduardo Habkost [Tue, 25 Aug 2020 19:20:10 +0000 (15:20 -0400)] 
hcd-dwc2: Rename USB_*CLASS macros for consistency

Rename the DWC2_CLASS to DWC2_USB_CLASS and DWC2_GET_CLASS to
DWC2_USB_GET_CLASS, for consistency with the DWC2_USB macro.

Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Reviewed-by: Gerd Hoffmann <kraxel@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-15-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agohvf: Add missing include
Eduardo Habkost [Tue, 25 Aug 2020 19:20:09 +0000 (15:20 -0400)] 
hvf: Add missing include

The sysemu/accel.h header is needed for the ACCEL_CLASS_NAME
macro.  This will be necessary to allow us to use OBJECT_DEFINE*()
for TYPE_HVF_ACCEL.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Roman Bolshakov <r.bolshakov@yadro.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-14-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agovirtio-ccw: Fix definition of VIRTIO_CCW_BUS_GET_CLASS
Eduardo Habkost [Tue, 25 Aug 2020 19:20:08 +0000 (15:20 -0400)] 
virtio-ccw: Fix definition of VIRTIO_CCW_BUS_GET_CLASS

The macro was incorrectly defined using OBJECT_CHECK.

Acked-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-13-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoversatile: Fix typo in PCI_VPB_HOST definition
Eduardo Habkost [Tue, 25 Aug 2020 19:20:07 +0000 (15:20 -0400)] 
versatile: Fix typo in PCI_VPB_HOST definition

Fixes: cd93dbf375bd ("versatile_pci: Update to realize and instance init functions")
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-12-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoaspeed_timer: Fix ASPEED_TIMER macro definition
Eduardo Habkost [Tue, 25 Aug 2020 19:20:06 +0000 (15:20 -0400)] 
aspeed_timer: Fix ASPEED_TIMER macro definition

The macro definition had an extra semicolon.  This was never
noticed because the macro was only being used where it didn't
make a difference.

Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-11-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoopentitan: Rename memmap enum constants
Eduardo Habkost [Tue, 25 Aug 2020 19:20:03 +0000 (15:20 -0400)] 
opentitan: Rename memmap enum constants

Some of the enum constant names conflict with the QOM type check
macros (IBEX_PLIC, IBEX_UART).  This needs to be addressed to
allow us to transform the QOM type check macros into functions
generated by OBJECT_DECLARE_TYPE().

Rename all the constants to IBEX_DEV_*, to avoid conflicts.

Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-8-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoaspeed_soc: Rename memmap/irqmap enum constants
Eduardo Habkost [Tue, 25 Aug 2020 19:20:02 +0000 (15:20 -0400)] 
aspeed_soc: Rename memmap/irqmap enum constants

Some of the enum constant names conflict with the QOM type check
macros:

ASPEED_GPIO
ASPEED_I2C
ASPEED_RTC
ASPEED_SCU
ASPEED_SDHCI
ASPEED_SDMC
ASPEED_VIC
ASPEED_WDT
ASPEED_XDMA

This needs to be addressed to allow us to transform the QOM type
check macros into functions generated by OBJECT_DECLARE_TYPE().

Rename all the constants to ASPEED_DEV_*, to avoid conflicts.

Reviewed-by: Cédric Le Goater <clg@kaod.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-7-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoallwinner-h3: Rename memmap enum constants
Eduardo Habkost [Tue, 25 Aug 2020 19:20:01 +0000 (15:20 -0400)] 
allwinner-h3: Rename memmap enum constants

Some of the enum constant names conflict with the QOM type check
macros (AW_H3_CCU, AW_H3_SYSCTRL).  This needs to be addressed to
allow us to transform the QOM type check macros into functions
generated by OBJECT_DECLARE_TYPE().

Rename all the constants to AW_H3_DEV_*, to avoid conflicts.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Niek Linnenbank <nieklinnenbank@gmail.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-6-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agopl110: Rename pl110_version enum values
Eduardo Habkost [Tue, 25 Aug 2020 19:20:00 +0000 (15:20 -0400)] 
pl110: Rename pl110_version enum values

The PL110 enum value name will conflict with the PL110 type cast
checker, when we replace the existing macro with an inline
function.  Add a VERSION_ prefix to all pl110_version enum
values, to avoid conflicts.

Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-5-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agovmw_pvscsi: Rename QOM class cast macros
Eduardo Habkost [Tue, 25 Aug 2020 19:19:59 +0000 (15:19 -0400)] 
vmw_pvscsi: Rename QOM class cast macros

Rename the PVSCSI_DEVICE_CLASS() and PVSCSI_DEVICE_GET_CLASS()
macros to be consistent with the PVSCSI() instance cast macro.

This will allow us to register the type cast macros using
OBJECT_DECLARE_TYPE later.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-4-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agomegasas: Rename QOM class cast macros
Eduardo Habkost [Tue, 25 Aug 2020 19:19:58 +0000 (15:19 -0400)] 
megasas: Rename QOM class cast macros

Rename the MEGASAS_DEVICE_CLASS() and MEGASAS_DEVICE_GET_CLASS()
macros to be consistent with the MEGASAS() instance cast macro.

This will allow us to register the type cast macros using
OBJECT_DECLARE_TYPE later.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-3-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoe1000: Rename QOM class cast macros
Eduardo Habkost [Tue, 25 Aug 2020 19:19:57 +0000 (15:19 -0400)] 
e1000: Rename QOM class cast macros

Rename the E1000_DEVICE_CLASS() and E1000_DEVICE_GET_CLASS()
macros to be consistent with the E1000() instance cast macro.

This will allow us to register the type cast macros using
OBJECT_DECLARE_TYPE later.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tested-By: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20200825192110.3528606-2-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agonuma: hmat: fix cache size check
Igor Mammedov [Fri, 21 Aug 2020 10:05:19 +0000 (06:05 -0400)] 
numa: hmat: fix cache size check

when QEMU is started like:

qemu-system-x86_64 -smp 2 -machine hmat=on \
 -m 2G \
 -object memory-backend-ram,size=1G,id=m0 \
 -object memory-backend-ram,size=1G,id=m1 \
 -numa node,nodeid=0,memdev=m0 \
 -numa node,nodeid=1,memdev=m1,initiator=0 \
 -numa cpu,node-id=0,socket-id=0 \
 -numa cpu,node-id=0,socket-id=1 \
 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-latency,latency=5 \
 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-bandwidth,bandwidth=200M \
 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-latency,latency=10 \
 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-bandwidth,bandwidth=100M \
 -numa hmat-cache,node-id=0,size=8K,level=1,associativity=direct,policy=write-back,line=5 \
 -numa hmat-cache,node-id=0,size=16K,level=2,associativity=direct,policy=write-back,line=5

it errors out with:
 -numa hmat-cache,node-id=0,size=16K,level=2,associativity=direct,policy=write-back,line=5:
        Invalid size=16384, the size of level=2 should be less than the size(8192) of level=1

which doesn't look right as one would expect that L1 < L2 < L3 ...
Fix it by sawpping relevant size checks.

Fixes: c412a48d4d91 (numa: Extend CLI to provide memory side cache information)
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <20200821100519.1325691-1-imammedo@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
13 months agoMerge remote-tracking branch 'remotes/cohuck/tags/s390x-20200827' into staging
Peter Maydell [Thu, 27 Aug 2020 12:48:12 +0000 (13:48 +0100)] 
Merge remote-tracking branch 'remotes/cohuck/tags/s390x-20200827' into staging

- document s390x boot devices
- bug fixes

# gpg: Signature made Thu 27 Aug 2020 13:16:41 BST
# gpg:                using RSA key C3D0D66DC3624FF6A8C018CEDECF6B93C6F02FAF
# gpg:                issuer "cohuck@redhat.com"
# gpg: Good signature from "Cornelia Huck <conny@cornelia-huck.de>" [unknown]
# gpg:                 aka "Cornelia Huck <huckc@linux.vnet.ibm.com>" [full]
# gpg:                 aka "Cornelia Huck <cornelia.huck@de.ibm.com>" [full]
# gpg:                 aka "Cornelia Huck <cohuck@kernel.org>" [unknown]
# gpg:                 aka "Cornelia Huck <cohuck@redhat.com>" [unknown]
# Primary key fingerprint: C3D0 D66D C362 4FF6 A8C0  18CE DECF 6B93 C6F0 2FAF

* remotes/cohuck/tags/s390x-20200827:
  hw/vfio/ap: Plug memleak in vfio_ap_get_group()
  docs/system/s390x: Add a chapter about s390x boot devices
  virtio-ccw-input: fix description

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
13 months agohw/vfio/ap: Plug memleak in vfio_ap_get_group()
Pan Nengyuan [Fri, 14 Aug 2020 16:02:31 +0000 (12:02 -0400)] 
hw/vfio/ap: Plug memleak in vfio_ap_get_group()

Missing g_error_free() in vfio_ap_get_group() error path. Fix that.

Reported-by: Euler Robot <euler.robot@huawei.com>
Signed-off-by: Pan Nengyuan <pannengyuan@huawei.com>
Reviewed-by: Pierre Morel <pmorel@linux.ibm.com>
Reviewed-by: Li Qiang <liq3ea@gmail.com>
Message-Id: <20200814160241.7915-3-pannengyuan@huawei.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
13 months agodocs/system/s390x: Add a chapter about s390x boot devices
Thomas Huth [Thu, 6 Aug 2020 15:05:07 +0000 (17:05 +0200)] 
docs/system/s390x: Add a chapter about s390x boot devices

Booting on s390x is a little bit different compared to other architectures.
Let's add some information for people who are not yet used to this.

Signed-off-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20200806150507.12073-1-thuth@redhat.com>
[CH: minor wording tweaks]
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
13 months agovirtio-ccw-input: fix description
Cornelia Huck [Tue, 28 Jul 2020 10:28:20 +0000 (12:28 +0200)] 
virtio-ccw-input: fix description

Fix a copy/paste error.

Signed-off-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20200728102820.273598-1-cohuck@redhat.com>

13 months agoMerge remote-tracking branch 'remotes/hdeller/tags/target-hppa-v3-pull-request' into...
Peter Maydell [Wed, 26 Aug 2020 21:23:53 +0000 (22:23 +0100)] 
Merge remote-tracking branch 'remotes/hdeller/tags/target-hppa-v3-pull-request' into staging

artist out of bounds fixes

# gpg: Signature made Wed 26 Aug 2020 22:09:55 BST
# gpg:                using EDDSA key BCE9123E1AD29F07C049BBDEF712B510A23A0F5F
# gpg: Good signature from "Helge Deller <deller@gmx.de>" [unknown]
# gpg:                 aka "Helge Deller <deller@kernel.org>" [unknown]
# 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: 4544 8228 2CD9 10DB EF3D  25F8 3E5F 3D04 A7A2 4603
#      Subkey fingerprint: BCE9 123E 1AD2 9F07 C049  BBDE F712 B510 A23A 0F5F

* remotes/hdeller/tags/target-hppa-v3-pull-request:
  hw/display/artist: Fix invalidation of lines near screen border
  hw/display/artist: Fix invalidation of lines in artist_draw_line()
  hw/display/artist: Unbreak size mismatch memory accesses
  hw/display/artist: Prevent out of VRAM buffer accesses
  Revert "hw/display/artist: Avoid drawing line when nothing to display"
  hw/display/artist: Refactor artist_rop8() to avoid buffer over-run
  hw/display/artist: Check offset in draw_line to avoid buffer over-run
  hw/hppa/lasi: Don't abort on invalid IMR value
  hw/display/artist.c: fix out of bounds check
  hw/hppa: Implement proper SeaBIOS version check
  seabios-hppa: Update to SeaBIOS hppa version 1
  hw/hppa: Sync hppa_hardware.h file with SeaBIOS sources

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
13 months agohw/display/artist: Fix invalidation of lines near screen border
Sven Schnelle [Sat, 8 Aug 2020 18:51:57 +0000 (20:51 +0200)] 
hw/display/artist: Fix invalidation of lines near screen border

If parts of the invalidated screen lines are outside of the VRAM buffer,
the code skips the whole invalidate. This is incorrect when only parts
of the buffer are invisble - which is the case when the mouse cursor is
located near the screen border.

Signed-off-by: Sven Schnelle <svens@stackframe.org>
Signed-off-by: Helge Deller <deller@gmx.de>
13 months agohw/display/artist: Fix invalidation of lines in artist_draw_line()
Sven Schnelle [Sat, 8 Aug 2020 19:56:29 +0000 (21:56 +0200)] 
hw/display/artist: Fix invalidation of lines in artist_draw_line()

The old code didn't invalidate correctly when vertical lines were drawn.
Fix this and move the invalidation out of the loop.

Signed-off-by: Sven Schnelle <svens@stackframe.org>
Signed-off-by: Helge Deller <deller@gmx.de>
13 months agohw/display/artist: Unbreak size mismatch memory accesses
Helge Deller [Sat, 8 Aug 2020 20:29:01 +0000 (22:29 +0200)] 
hw/display/artist: Unbreak size mismatch memory accesses

Commit 5d971f9e6725 ("memory: Revert "memory: accept mismatching sizes
in memory_region_access_valid") broke the artist driver in a way that
the dtwm window manager on HP-UX rendered wrong.

Fixes: 5d971f9e6725 ("memory: Revert "memory: accept mismatching sizes in memory_region_access_valid")
Signed-off-by: Sven Schnelle <svens@stackframe.org>
Signed-off-by: Helge Deller <deller@gmx.de>
13 months agohw/display/artist: Prevent out of VRAM buffer accesses
Helge Deller [Sun, 9 Aug 2020 13:35:38 +0000 (15:35 +0200)] 
hw/display/artist: Prevent out of VRAM buffer accesses

Simplify various bounds checks by changing parameters like row and column
numbers to become unsigned instead of signed.
With that we can check if the calculated offset is bigger than the size of the
VRAM region and bail out if not.

Reported-by: LLVM libFuzzer
Reported-by: Alexander Bulekov <alxndr@bu.edu>
Buglink: https://bugs.launchpad.net/qemu/+bug/1880326
Buglink: https://bugs.launchpad.net/qemu/+bug/1890310
Buglink: https://bugs.launchpad.net/qemu/+bug/1890311
Buglink: https://bugs.launchpad.net/qemu/+bug/1890312
Buglink: https://bugs.launchpad.net/qemu/+bug/1890370
Acked-by: Alexander Bulekov <alxndr@bu.edu>
Signed-off-by: Helge Deller <deller@gmx.de>
13 months agoRevert "hw/display/artist: Avoid drawing line when nothing to display"
Helge Deller [Sat, 8 Aug 2020 20:22:37 +0000 (22:22 +0200)] 
Revert "hw/display/artist: Avoid drawing line when nothing to display"

This reverts commit b0f6455feac97e41045ee394e11c24d92c370f6e.
It's wrong. A line could even be a dot.

Signed-off-by: Helge Deller <deller@gmx.de>
13 months agohw/display/artist: Refactor artist_rop8() to avoid buffer over-run
Philippe Mathieu-Daudé [Sat, 23 May 2020 19:15:16 +0000 (21:15 +0200)] 
hw/display/artist: Refactor artist_rop8() to avoid buffer over-run

Invalid I/O writes can craft an offset out of the vram_buffer range.
Instead of passing an unsafe pointer to artist_rop8(), pass the vram_buffer and
the offset. We can now check if the offset is in range before accessing it.

We avoid:

  Program terminated with signal SIGSEGV, Segmentation fault.
  284             *dst &= ~plane_mask;
  (gdb) bt
  #0  0x000056367b2085c0 in artist_rop8 (s=0x56367d38b510, dst=0x7f9f972fffff <error: Cannot access memory at address 0x7f9f972fffff>, val=0 '\000') at hw/display/artist.c:284
  #1  0x000056367b209325 in draw_line (s=0x56367d38b510, x1=-20480, y1=-1, x2=0, y2=17920, update_start=true, skip_pix=-1, max_pix=-1) at hw/display/artist.c:646

Reported-by: LLVM libFuzzer
Buglink: https://bugs.launchpad.net/qemu/+bug/1880326
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Helge Deller <deller@gmx.de>
13 months agohw/display/artist: Check offset in draw_line to avoid buffer over-run
Philippe Mathieu-Daudé [Sat, 23 May 2020 19:15:15 +0000 (21:15 +0200)] 
hw/display/artist: Check offset in draw_line to avoid buffer over-run

Invalid I/O writes can craft an offset out of the vram_buffer range.

We avoid:

  Program terminated with signal SIGSEGV, Segmentation fault.
  284             *dst &= ~plane_mask;
  (gdb) bt
  #0  0x000055d5dccdc5c0 in artist_rop8 (s=0x55d5defee510, dst=0x7f8e84ed8216 <error: Cannot access memory at address 0x7f8e84ed8216>, val=0 '\000') at hw/display/artist.c:284
  #1  0x000055d5dccdcf83 in fill_window (s=0x55d5defee510, startx=22, starty=5674, width=65, height=5697) at hw/display/artist.c:551
  #2  0x000055d5dccddfb9 in artist_reg_write (opaque=0x55d5defee510, addr=1051140, val=4265537, size=4) at hw/display/artist.c:902
  #3  0x000055d5dcb42a7c in memory_region_write_accessor (mr=0x55d5defeea10, addr=1051140, value=0x7ffe57db08c8, size=4, shift=0, mask=4294967295, attrs=...) at memory.c:483

Reported-by: LLVM libFuzzer
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Helge Deller <deller@gmx.de>
13 months agohw/hppa/lasi: Don't abort on invalid IMR value
Helge Deller [Mon, 3 Aug 2020 05:41:46 +0000 (07:41 +0200)] 
hw/hppa/lasi: Don't abort on invalid IMR value

NetBSD initializes the LASI IMR value with 0xffffffff to disable all LASI
interrupts. This triggered an assert() and stopped the emulation.  By replacing
the check with a warning in the guest log we now allow NetBSD to boot again.

Signed-off-by: Helge Deller <deller@gmx.de>
13 months agoMerge remote-tracking branch 'remotes/maxreitz/tags/pull-block-2020-08-26' into staging
Peter Maydell [Wed, 26 Aug 2020 09:28:36 +0000 (10:28 +0100)] 
Merge remote-tracking branch 'remotes/maxreitz/tags/pull-block-2020-08-26' into staging

Block patches:
- qcow2 subclusters (extended L2 entries)

# gpg: Signature made Wed 26 Aug 2020 08:37:04 BST
# gpg:                using RSA key 91BEB60A30DB3E8857D11829F407DB0061D5CF40
# gpg:                issuer "mreitz@redhat.com"
# gpg: Good signature from "Max Reitz <mreitz@redhat.com>" [full]
# Primary key fingerprint: 91BE B60A 30DB 3E88 57D1  1829 F407 DB00 61D5 CF40

* remotes/maxreitz/tags/pull-block-2020-08-26: (34 commits)
  iotests: Add tests for qcow2 images with extended L2 entries
  qcow2: Assert that expand_zero_clusters_in_l1() does not support subclusters
  qcow2: Allow preallocation and backing files if extended_l2 is set
  qcow2: Add the 'extended_l2' option and the QCOW2_INCOMPAT_EXTL2 bit
  qcow2: Add prealloc field to QCowL2Meta
  qcow2: Add subcluster support to qcow2_measure()
  qcow2: Add subcluster support to qcow2_co_pwrite_zeroes()
  qcow2: Add subcluster support to handle_alloc_space()
  qcow2: Clear the L2 bitmap when allocating a compressed cluster
  qcow2: Update L2 bitmap in qcow2_alloc_cluster_link_l2()
  qcow2: Add subcluster support to check_refcounts_l2()
  qcow2: Add subcluster support to discard_in_l2_slice()
  qcow2: Add subcluster support to zero_in_l2_slice()
  qcow2: Add subcluster support to qcow2_get_host_offset()
  qcow2: Add subcluster support to calculate_l2_meta()
  qcow2: Handle QCOW2_SUBCLUSTER_UNALLOCATED_ALLOC
  qcow2: Replace QCOW2_CLUSTER_* with QCOW2_SUBCLUSTER_*
  qcow2: Add cluster type parameter to qcow2_get_host_offset()
  qcow2: Add qcow2_cluster_is_allocated()
  qcow2: Add qcow2_get_subcluster_range_type()
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
13 months agoiotests: Add tests for qcow2 images with extended L2 entries
Alberto Garcia [Fri, 10 Jul 2020 16:13:16 +0000 (18:13 +0200)] 
iotests: Add tests for qcow2 images with extended L2 entries

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <e6dd0429cafe84ca603179c298a8703bddca2904.1594396418.git.berto@igalia.com>
[mreitz: Use env in shebang line]
Signed-off-by: Max Reitz <mreitz@redhat.com>
13 months agoMerge remote-tracking branch 'remotes/alistair/tags/pull-riscv-to-apply-20200825...
Peter Maydell [Tue, 25 Aug 2020 21:50:42 +0000 (22:50 +0100)] 
Merge remote-tracking branch 'remotes/alistair/tags/pull-riscv-to-apply-20200825' into staging

This pull request first adds support for multi-socket NUMA RISC-V
machines. The Spike and Virt machines both support NUMA sockets.

This PR also updates the current experimental Hypervisor support to the
v0.6.1 spec.

# gpg: Signature made Tue 25 Aug 2020 19:47:41 BST
# gpg:                using RSA key F6C4AC46D4934868D3B8CE8F21E10D29DF977054
# gpg: Good signature from "Alistair Francis <alistair@alistair23.me>" [full]
# Primary key fingerprint: F6C4 AC46 D493 4868 D3B8  CE8F 21E1 0D29 DF97 7054

* remotes/alistair/tags/pull-riscv-to-apply-20200825:
  target/riscv: Support the Virtual Instruction fault
  target/riscv: Return the exception from invalid CSR accesses
  target/riscv: Support the v0.6 Hypervisor extension CRSs
  target/riscv: Only support little endian guests
  target/riscv: Only support a single VSXL length
  target/riscv: Update the CSRs to the v0.6 Hyp extension
  target/riscv: Update the Hypervisor trap return/entry
  target/riscv: Fix the interrupt cause code
  target/riscv: Convert MSTATUS MTL to GVA
  target/riscv: Don't allow guest to write to htinst
  target/riscv: Do two-stage lookups on hlv/hlvx/hsv instructions
  target/riscv: Allow generating hlv/hlvx/hsv instructions
  target/riscv: Allow setting a two-stage lookup in the virt status
  hw/riscv: virt: Allow creating multiple NUMA sockets
  hw/riscv: spike: Allow creating multiple NUMA sockets
  hw/riscv: Add helpers for RISC-V multi-socket NUMA machines
  hw/riscv: Allow creating multiple instances of PLIC
  hw/riscv: Allow creating multiple instances of CLINT

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
13 months agotarget/riscv: Support the Virtual Instruction fault
Alistair Francis [Wed, 12 Aug 2020 19:13:49 +0000 (12:13 -0700)] 
target/riscv: Support the Virtual Instruction fault

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 4c744dce9b0b057cbb5cc0f4d4ac75cda682a8af.1597259519.git.alistair.francis@wdc.com
Message-Id: <4c744dce9b0b057cbb5cc0f4d4ac75cda682a8af.1597259519.git.alistair.francis@wdc.com>

13 months agotarget/riscv: Return the exception from invalid CSR accesses
Alistair Francis [Wed, 12 Aug 2020 19:13:46 +0000 (12:13 -0700)] 
target/riscv: Return the exception from invalid CSR accesses

When performing a CSR access let's return a negative exception value on
an error instead of -1. This will allow us to specify the exception in
future patches.

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: a487dad60c9b8fe7a2b992c5e0dcc2504a9000a7.1597259519.git.alistair.francis@wdc.com
Message-Id: <a487dad60c9b8fe7a2b992c5e0dcc2504a9000a7.1597259519.git.alistair.francis@wdc.com>

13 months agotarget/riscv: Support the v0.6 Hypervisor extension CRSs
Alistair Francis [Wed, 12 Aug 2020 19:13:44 +0000 (12:13 -0700)] 
target/riscv: Support the v0.6 Hypervisor extension CRSs

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 644b6c114b1a81adbee0ab8c9c66a8672059ec96.1597259519.git.alistair.francis@wdc.com
Message-Id: <644b6c114b1a81adbee0ab8c9c66a8672059ec96.1597259519.git.alistair.francis@wdc.com>

13 months agotarget/riscv: Only support little endian guests
Alistair Francis [Wed, 12 Aug 2020 19:13:41 +0000 (12:13 -0700)] 
target/riscv: Only support little endian guests

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 93e5d4f13eca0d2a588e407187f33c6437aeaaf9.1597259519.git.alistair.francis@wdc.com
Message-Id: <93e5d4f13eca0d2a588e407187f33c6437aeaaf9.1597259519.git.alistair.francis@wdc.com>

13 months agotarget/riscv: Only support a single VSXL length
Alistair Francis [Wed, 12 Aug 2020 19:13:38 +0000 (12:13 -0700)] 
target/riscv: Only support a single VSXL length

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: f3f4fd2ec22a07cc1d750e96895d6813f131de4d.1597259519.git.alistair.francis@wdc.com
Message-Id: <f3f4fd2ec22a07cc1d750e96895d6813f131de4d.1597259519.git.alistair.francis@wdc.com>

13 months agotarget/riscv: Update the CSRs to the v0.6 Hyp extension
Alistair Francis [Wed, 12 Aug 2020 19:13:36 +0000 (12:13 -0700)] 
target/riscv: Update the CSRs to the v0.6 Hyp extension

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 4f227b30cb1816795296c0994f1123fab143666a.1597259519.git.alistair.francis@wdc.com
Message-Id: <4f227b30cb1816795296c0994f1123fab143666a.1597259519.git.alistair.francis@wdc.com>

13 months agotarget/riscv: Update the Hypervisor trap return/entry
Alistair Francis [Wed, 12 Aug 2020 19:13:33 +0000 (12:13 -0700)] 
target/riscv: Update the Hypervisor trap return/entry

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: e7e4e801234f2934306e734f65860f601a5745bd.1597259519.git.alistair.francis@wdc.com
Message-Id: <e7e4e801234f2934306e734f65860f601a5745bd.1597259519.git.alistair.francis@wdc.com>

13 months agotarget/riscv: Fix the interrupt cause code
Alistair Francis [Wed, 12 Aug 2020 19:13:30 +0000 (12:13 -0700)] 
target/riscv: Fix the interrupt cause code

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 85b7fdba8abd87adb83275cdc3043ce35a1ed5c3.1597259519.git.alistair.francis@wdc.com
Message-Id: <85b7fdba8abd87adb83275cdc3043ce35a1ed5c3.1597259519.git.alistair.francis@wdc.com>

13 months agotarget/riscv: Convert MSTATUS MTL to GVA
Alistair Francis [Wed, 12 Aug 2020 19:13:27 +0000 (12:13 -0700)] 
target/riscv: Convert MSTATUS MTL to GVA

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 9308432988946de550a68524ed76e4b8683f10e2.1597259519.git.alistair.francis@wdc.com
Message-Id: <9308432988946de550a68524ed76e4b8683f10e2.1597259519.git.alistair.francis@wdc.com>

13 months agotarget/riscv: Don't allow guest to write to htinst
Alistair Francis [Wed, 12 Aug 2020 19:13:25 +0000 (12:13 -0700)] 
target/riscv: Don't allow guest to write to htinst

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: ca5359fec6b2aff851eef3b3bc4b53cb5d4ad194.1597259519.git.alistair.francis@wdc.com
Message-Id: <ca5359fec6b2aff851eef3b3bc4b53cb5d4ad194.1597259519.git.alistair.francis@wdc.com>

13 months agotarget/riscv: Do two-stage lookups on hlv/hlvx/hsv instructions
Alistair Francis [Wed, 12 Aug 2020 19:13:22 +0000 (12:13 -0700)] 
target/riscv: Do two-stage lookups on hlv/hlvx/hsv instructions

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 024ad8a594fb2feaf0950fbfad1508cfa82ce7f0.1597259519.git.alistair.francis@wdc.com
Message-Id: <024ad8a594fb2feaf0950fbfad1508cfa82ce7f0.1597259519.git.alistair.francis@wdc.com>

13 months agotarget/riscv: Allow generating hlv/hlvx/hsv instructions
Alistair Francis [Wed, 12 Aug 2020 19:13:19 +0000 (12:13 -0700)] 
target/riscv: Allow generating hlv/hlvx/hsv instructions

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 477c864312280ea55a98dc84cb01d826751b6c14.1597259519.git.alistair.francis@wdc.com
Message-Id: <477c864312280ea55a98dc84cb01d826751b6c14.1597259519.git.alistair.francis@wdc.com>

13 months agotarget/riscv: Allow setting a two-stage lookup in the virt status
Alistair Francis [Wed, 12 Aug 2020 19:13:16 +0000 (12:13 -0700)] 
target/riscv: Allow setting a two-stage lookup in the virt status

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 08cdefb171b1bdb0c9e3151c509aaadefc3dcd3e.1597259519.git.alistair.francis@wdc.com
Message-Id: <08cdefb171b1bdb0c9e3151c509aaadefc3dcd3e.1597259519.git.alistair.francis@wdc.com>

13 months agohw/riscv: virt: Allow creating multiple NUMA sockets
Anup Patel [Fri, 15 May 2020 09:28:50 +0000 (14:58 +0530)] 
hw/riscv: virt: Allow creating multiple NUMA sockets

We extend RISC-V virt machine to allow creating a multi-socket
machine. Each RISC-V virt machine socket is a NUMA node having
a set of HARTs, a memory instance, a CLINT instance, and a PLIC
instance. Other devices are shared between all sockets. We also
update the generated device tree accordingly.

By default, NUMA multi-socket support is disabled for RISC-V virt
machine. To enable it, users can use "-numa" command-line options
of QEMU.

Example1: For two NUMA nodes with 2 CPUs each, append following
to command-line options: "-smp 4 -numa node -numa node"

Example2: For two NUMA nodes with 1 and 3 CPUs, append following
to command-line options:
"-smp 4 -numa node -numa node -numa cpu,node-id=0,core-id=0 \
-numa cpu,node-id=1,core-id=1 -numa cpu,node-id=1,core-id=2 \
-numa cpu,node-id=1,core-id=3"

The maximum number of sockets in a RISC-V virt machine is 8
but this limit can be changed in future.

Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Atish Patra <atish.patra@wdc.com>
Message-Id: <20200616032229.766089-6-anup.patel@wdc.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
13 months agohw/riscv: spike: Allow creating multiple NUMA sockets
Anup Patel [Thu, 14 May 2020 11:09:43 +0000 (16:39 +0530)] 
hw/riscv: spike: Allow creating multiple NUMA sockets

We extend RISC-V spike machine to allow creating a multi-socket
machine. Each RISC-V spike machine socket is a NUMA node having
a set of HARTs, a memory instance, and a CLINT instance. Other
devices are shared between all sockets. We also update the
generated device tree accordingly.

By default, NUMA multi-socket support is disabled for RISC-V spike
machine. To enable it, users can use "-numa" command-line options
of QEMU.

Example1: For two NUMA nodes with 2 CPUs each, append following
to command-line options: "-smp 4 -numa node -numa node"

Example2: For two NUMA nodes with 1 and 3 CPUs, append following
to command-line options:
"-smp 4 -numa node -numa node -numa cpu,node-id=0,core-id=0 \
-numa cpu,node-id=1,core-id=1 -numa cpu,node-id=1,core-id=2 \
-numa cpu,node-id=1,core-id=3"

The maximum number of sockets in a RISC-V spike machine is 8
but this limit can be changed in future.

Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Atish Patra <atish.patra@wdc.com>
Message-Id: <20200616032229.766089-5-anup.patel@wdc.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
13 months agohw/riscv: Add helpers for RISC-V multi-socket NUMA machines
Anup Patel [Fri, 29 May 2020 07:56:22 +0000 (13:26 +0530)] 
hw/riscv: Add helpers for RISC-V multi-socket NUMA machines

We add common helper routines which can be shared by RISC-V
multi-socket NUMA machines.

We have two types of helpers:
1. riscv_socket_xyz() - These helper assist managing multiple
   sockets irrespective whether QEMU NUMA is enabled/disabled
2. riscv_numa_xyz() - These helpers assist in providing
   necessary QEMU machine callbacks for QEMU NUMA emulation

Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Atish Patra <atish.patra@wdc.com>
Message-Id: <20200616032229.766089-4-anup.patel@wdc.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
13 months agohw/riscv: Allow creating multiple instances of PLIC
Anup Patel [Fri, 15 May 2020 04:55:33 +0000 (10:25 +0530)] 
hw/riscv: Allow creating multiple instances of PLIC

We extend PLIC emulation to allow multiple instances of PLIC in
a QEMU RISC-V machine. To achieve this, we remove first HART id
zero assumption from PLIC emulation.

Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Palmer Dabbelt <palmerdabbelt@google.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20200616032229.766089-3-anup.patel@wdc.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
13 months agohw/riscv: Allow creating multiple instances of CLINT
Anup Patel [Thu, 14 May 2020 10:21:31 +0000 (15:51 +0530)] 
hw/riscv: Allow creating multiple instances of CLINT

We extend CLINT emulation to allow multiple instances of CLINT in
a QEMU RISC-V machine. To achieve this, we remove first HART id
zero assumption from CLINT emulation.

Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Palmer Dabbelt <palmerdabbelt@google.com>
Message-Id: <20200616032229.766089-2-anup.patel@wdc.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
14 months agoMerge remote-tracking branch 'remotes/berrange/tags/socket-next-pull-request' into...
Peter Maydell [Tue, 25 Aug 2020 14:01:43 +0000 (15:01 +0100)] 
Merge remote-tracking branch 'remotes/berrange/tags/socket-next-pull-request' into staging

Add support for UNIX sockets in the abstract namespace

# gpg: Signature made Tue 25 Aug 2020 11:52:22 BST
# gpg:                using RSA key DAF3A6FDB26B62912D0E8E3FBE86EBB415104FDF
# gpg: Good signature from "Daniel P. Berrange <dan@berrange.com>" [full]
# gpg:                 aka "Daniel P. Berrange <berrange@redhat.com>" [full]
# Primary key fingerprint: DAF3 A6FD B26B 6291 2D0E  8E3F BE86 EBB4 1510 4FDF

* remotes/berrange/tags/socket-next-pull-request:
  tests: fix a memory in test_socket_unix_abstract_good

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
14 months agotests: fix a memory in test_socket_unix_abstract_good
Li Qiang [Wed, 3 Jun 2020 16:14:09 +0000 (09:14 -0700)] 
tests: fix a memory in test_socket_unix_abstract_good

After build qemu with '-fsanitize=address' extra-cflags,
'make check' show following leak:

=================================================================
==44580==ERROR: LeakSanitizer: detected memory leaks

Direct leak of 2500 byte(s) in 1 object(s) allocated from:
    #0 0x7f1b5a8b8d28 in __interceptor_calloc (/usr/lib/x86_64-linux-gnu/libasan.so.4+0xded28)
    #1 0x7f1b5a514b10 in g_malloc0 (/usr/lib/x86_64-linux-gnu/libglib-2.0.so.0+0x51b10)
    #2 0xd79ea4e4c0ad31c3  (<unknown module>)

SUMMARY: AddressSanitizer: 2500 byte(s) leaked in 1 allocation(s).

Call 'g_rand_free' in the end of function to avoid this.

Fixes: 4d3a329af59("tests/util-sockets: add abstract unix socket cases")
Signed-off-by: Li Qiang <liq3ea@163.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by:  xiaoqiang zhao <zxq_yx_007@163.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
14 months agoMerge remote-tracking branch 'remotes/kraxel/tags/fixes-20200825-pull-request' into...
Peter Maydell [Tue, 25 Aug 2020 09:54:50 +0000 (10:54 +0100)] 
Merge remote-tracking branch 'remotes/kraxel/tags/fixes-20200825-pull-request' into staging

meson: keymap fixes

# gpg: Signature made Tue 25 Aug 2020 07:19:15 BST
# gpg:                using RSA key 4CB6D8EED3E87138
# gpg: Good signature from "Gerd Hoffmann (work) <kraxel@redhat.com>" [full]
# gpg:                 aka "Gerd Hoffmann <gerd@kraxel.org>" [full]
# gpg:                 aka "Gerd Hoffmann (private) <kraxel@gmail.com>" [full]
# Primary key fingerprint: A032 8CFF B93A 17A7 9901  FE7D 4CB6 D8EE D3E8 7138

* remotes/kraxel/tags/fixes-20200825-pull-request:
  meson: avoid compiling qemu-keymap by default
  meson: move xkbcommon to meson
  meson: drop keymaps symlink

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
14 months agoqcow2: Assert that expand_zero_clusters_in_l1() does not support subclusters
Alberto Garcia [Fri, 10 Jul 2020 16:13:15 +0000 (18:13 +0200)] 
qcow2: Assert that expand_zero_clusters_in_l1() does not support subclusters

This function is only used by qcow2_expand_zero_clusters() to
downgrade a qcow2 image to a previous version. This would require
transforming all extended L2 entries into normal L2 entries but this
is not a simple task and there are no plans to implement this at the
moment.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <15e65112b4144381b4d8c0bdf8fb76b0d813e3d1.1594396418.git.berto@igalia.com>
[mreitz: Fixed comment style]
Signed-off-by: Max Reitz <mreitz@redhat.com>
14 months agoqcow2: Allow preallocation and backing files if extended_l2 is set
Alberto Garcia [Fri, 10 Jul 2020 16:13:14 +0000 (18:13 +0200)] 
qcow2: Allow preallocation and backing files if extended_l2 is set

Traditional qcow2 images don't allow preallocation if a backing file
is set. This is because once a cluster is allocated there is no way to
tell that its data should be read from the backing file.

Extended L2 entries have individual allocation bits for each
subcluster, and therefore it is perfectly possible to have an
allocated cluster with all its subclusters unallocated.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <6d5b0f38e7dc5f2f31d8cab1cb92044e9909aece.1594396418.git.berto@igalia.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
14 months agoqcow2: Add the 'extended_l2' option and the QCOW2_INCOMPAT_EXTL2 bit
Alberto Garcia [Fri, 10 Jul 2020 16:13:13 +0000 (18:13 +0200)] 
qcow2: Add the 'extended_l2' option and the QCOW2_INCOMPAT_EXTL2 bit

Now that the implementation of subclusters is complete we can finally
add the necessary options to create and read images with this feature,
which we call "extended L2 entries".

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <6476caaa73216bd05b7bb2d504a20415e1665176.1594396418.git.berto@igalia.com>
[mreitz: %s/5\.1/5.2/; fixed 302's and 303's reference output]
Signed-off-by: Max Reitz <mreitz@redhat.com>
14 months agoqcow2: Add prealloc field to QCowL2Meta
Alberto Garcia [Fri, 10 Jul 2020 16:13:12 +0000 (18:13 +0200)] 
qcow2: Add prealloc field to QCowL2Meta

This field allows us to indicate that the L2 metadata update does not
come from a write request with actual data but from a preallocation
request.

For traditional images this does not make any difference, but for
images with extended L2 entries this means that the clusters are
allocated normally in the L2 table but individual subclusters are
marked as unallocated.

This will allow preallocating images that have a backing file.

There is one special case: when we resize an existing image we can
also request that the new clusters are preallocated. If the image
already had a backing file then we have to hide any possible stale
data and zero out the new clusters (see commit 955c7d6687 for more
details).

In this case the subclusters cannot be left as unallocated so the L2
bitmap must be updated.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <960d4c444a4f5a870e2b47e5da322a73cd9a2f5a.1594396418.git.berto@igalia.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
14 months agoqcow2: Add subcluster support to qcow2_measure()
Alberto Garcia [Fri, 10 Jul 2020 16:13:11 +0000 (18:13 +0200)] 
qcow2: Add subcluster support to qcow2_measure()

Extended L2 entries are bigger than normal L2 entries so this has an
impact on the amount of metadata needed for a qcow2 file.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <7efae2efd5e36b42d2570743a12576d68ce53685.1594396418.git.berto@igalia.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
14 months agoqcow2: Add subcluster support to qcow2_co_pwrite_zeroes()
Alberto Garcia [Fri, 10 Jul 2020 16:13:10 +0000 (18:13 +0200)] 
qcow2: Add subcluster support to qcow2_co_pwrite_zeroes()

This works now at the subcluster level and pwrite_zeroes_alignment is
updated accordingly.

qcow2_cluster_zeroize() is turned into qcow2_subcluster_zeroize() with
the following changes:

   - The request can now be subcluster-aligned.

   - The cluster-aligned body of the request is still zeroized using
     zero_in_l2_slice() as before.

   - The subcluster-aligned head and tail of the request are zeroized
     with the new zero_l2_subclusters() function.

There is just one thing to take into account for a possible future
improvement: compressed clusters cannot be partially zeroized so
zero_l2_subclusters() on the head or the tail can return -ENOTSUP.
This makes the caller repeat the *complete* request and write actual
zeroes to disk. This is sub-optimal because

   1) if the head area was compressed we would still be able to use
      the fast path for the body and possibly the tail.

   2) if the tail area was compressed we are writing zeroes to the
      head and the body areas, which are already zeroized.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <17e05e2ee7e12f10dcf012da81e83ebe27eb3bef.1594396418.git.berto@igalia.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
14 months agoqcow2: Add subcluster support to handle_alloc_space()
Alberto Garcia [Fri, 10 Jul 2020 16:13:09 +0000 (18:13 +0200)] 
qcow2: Add subcluster support to handle_alloc_space()

The bdrv_co_pwrite_zeroes() call here fills complete clusters with
zeroes, but it can happen that some subclusters are not part of the
write request or the copy-on-write. This patch makes sure that only
the affected subclusters are overwritten.

A potential improvement would be to also fill with zeroes the other
subclusters if we can guarantee that we are not overwriting existing
data. However this would waste more disk space, so we should first
evaluate if it's really worth doing.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <b3dc97e8e2240ddb5191a4f930e8fc9653f94621.1594396418.git.berto@igalia.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
14 months agoqcow2: Clear the L2 bitmap when allocating a compressed cluster
Alberto Garcia [Fri, 10 Jul 2020 16:13:08 +0000 (18:13 +0200)] 
qcow2: Clear the L2 bitmap when allocating a compressed cluster

Compressed clusters always have the bitmap part of the extended L2
entry set to 0.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <04455b3de5dfeb9d1cfe1fc7b02d7060a6e09710.1594396418.git.berto@igalia.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
14 months agoqcow2: Update L2 bitmap in qcow2_alloc_cluster_link_l2()
Alberto Garcia [Fri, 10 Jul 2020 16:13:07 +0000 (18:13 +0200)] 
qcow2: Update L2 bitmap in qcow2_alloc_cluster_link_l2()

The L2 bitmap needs to be updated after each write to indicate what
new subclusters are now allocated. This needs to happen even if the
cluster was already allocated and the L2 entry was otherwise valid.

In some cases however a write operation doesn't need change the L2
bitmap (because all affected subclusters were already allocated). This
is detected in calculate_l2_meta(), and qcow2_alloc_cluster_link_l2()
is never called in those cases.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <0875620d49f44320334b6a91c73b3f301f975f38.1594396418.git.berto@igalia.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
14 months agoqcow2: Add subcluster support to check_refcounts_l2()
Alberto Garcia [Fri, 10 Jul 2020 16:13:06 +0000 (18:13 +0200)] 
qcow2: Add subcluster support to check_refcounts_l2()

The offset field of an uncompressed cluster's L2 entry must be aligned
to the cluster size, otherwise it is invalid. If the cluster has no
data then it means that the offset points to a preallocation, so we
can clear the offset field without affecting the guest-visible data.
This is what 'qemu-img check' does when run in repair mode.

On traditional qcow2 images this can only happen when QCOW_OFLAG_ZERO
is set, and repairing such entries turns the clusters from ZERO_ALLOC
into ZERO_PLAIN.

Extended L2 entries have no ZERO_ALLOC clusters and no QCOW_OFLAG_ZERO
but the idea is the same: if none of the subclusters are allocated
then we can clear the offset field and leave the bitmap untouched.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <9f4ed1d0a34b0a545b032c31ecd8c14734065342.1594396418.git.berto@igalia.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
14 months agoqcow2: Add subcluster support to discard_in_l2_slice()
Alberto Garcia [Fri, 10 Jul 2020 16:13:05 +0000 (18:13 +0200)] 
qcow2: Add subcluster support to discard_in_l2_slice()

Two things need to be taken into account here:

1) With full_discard == true the L2 entry must be cleared completely.
   This also includes the L2 bitmap if the image has extended L2
   entries.

2) With full_discard == false we have to make the discarded cluster
   read back as zeroes. With normal L2 entries this is done with the
   QCOW_OFLAG_ZERO bit, whereas with extended L2 entries this is done
   with the individual 'all zeroes' bits for each subcluster.

   Note however that QCOW_OFLAG_ZERO is not supported in v2 qcow2
   images so, if there is a backing file, discard cannot guarantee
   that the image will read back as zeroes. If this is important for
   the caller it should forbid it as qcow2_co_pdiscard() does (see
   80f5c01183 for more details).

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <5ef8274e628aa3ab559bfac467abf488534f2b76.1594396418.git.berto@igalia.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
14 months agoqcow2: Add subcluster support to zero_in_l2_slice()
Alberto Garcia [Fri, 10 Jul 2020 16:13:04 +0000 (18:13 +0200)] 
qcow2: Add subcluster support to zero_in_l2_slice()

The QCOW_OFLAG_ZERO bit that indicates that a cluster reads as
zeroes is only used in standard L2 entries. Extended L2 entries use
individual 'all zeroes' bits for each subcluster.

This must be taken into account when updating the L2 entry and also
when deciding that an existing entry does not need to be updated.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <b61d61606d8c9b367bd641ab37351ddb9172799a.1594396418.git.berto@igalia.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
14 months agoqcow2: Add subcluster support to qcow2_get_host_offset()
Alberto Garcia [Fri, 10 Jul 2020 16:13:03 +0000 (18:13 +0200)] 
qcow2: Add subcluster support to qcow2_get_host_offset()

The logic of this function remains pretty much the same, except that
it uses count_contiguous_subclusters(), which combines the logic of
count_contiguous_clusters() / count_contiguous_clusters_unallocated()
and checks individual subclusters.

qcow2_cluster_to_subcluster_type() is not necessary as a separate
function anymore so it's inlined into its caller.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <d2193fd48653a350d80f0eca1c67b1d9053fb2f3.1594396418.git.berto@igalia.com>
[mreitz: Initialize expected_type to anything]
Signed-off-by: Max Reitz <mreitz@redhat.com>
14 months agoqcow2: Add subcluster support to calculate_l2_meta()
Alberto Garcia [Fri, 10 Jul 2020 16:13:02 +0000 (18:13 +0200)] 
qcow2: Add subcluster support to calculate_l2_meta()

If an image has subclusters then there are more copy-on-write
scenarios that we need to consider. Let's say we have a write request
from the middle of subcluster #3 until the end of the cluster:

1) If we are writing to a newly allocated cluster then we need
   copy-on-write. The previous contents of subclusters #0 to #3 must
   be copied to the new cluster. We can optimize this process by
   skipping all leading unallocated or zero subclusters (the status of
   those skipped subclusters will be reflected in the new L2 bitmap).

2) If we are overwriting an existing cluster:

   2.1) If subcluster #3 is unallocated or has the all-zeroes bit set
        then we need copy-on-write (on subcluster #3 only).

   2.2) If subcluster #3 was already allocated then there is no need
        for any copy-on-write. However we still need to update the L2
        bitmap to reflect possible changes in the allocation status of
        subclusters #4 to #31. Because of this, this function checks
        if all the overwritten subclusters are already allocated and
        in this case it returns without creating a new QCowL2Meta
        structure.

After all these changes l2meta_cow_start() and l2meta_cow_end()
are not necessarily cluster-aligned anymore. We need to update the
calculation of old_start and old_end in handle_dependencies() to
guarantee that no two requests try to write on the same cluster.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <4292dd56e4446d386a2fe307311737a711c00708.1594396418.git.berto@igalia.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
14 months agoqcow2: Handle QCOW2_SUBCLUSTER_UNALLOCATED_ALLOC
Alberto Garcia [Fri, 10 Jul 2020 16:13:01 +0000 (18:13 +0200)] 
qcow2: Handle QCOW2_SUBCLUSTER_UNALLOCATED_ALLOC

When dealing with subcluster types there is a new value called
QCOW2_SUBCLUSTER_UNALLOCATED_ALLOC that has no equivalent in
QCow2ClusterType.

This patch handles that value in all places where subcluster types
are processed.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <bf09e2e2439a468a901bb96ace411eed9ee50295.1594396418.git.berto@igalia.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>