Merge remote-tracking branch 'remotes/pmaydell/tags/pull-target-arm-20200828' into staging

target-arm queue:
 * target/arm: Cleanup and refactoring preparatory to SVE2
 * armsse: Define ARMSSEClass correctly
 * hw/misc/unimp: Improve information provided in log messages
 * hw/qdev-clock: Avoid calling qdev_connect_clock_in after DeviceRealize
 * hw/arm/xilinx_zynq: Call qdev_connect_clock_in() before DeviceRealize
 * hw/net/allwinner-sun8i-emac: Use AddressSpace for DMA transfers
 * hw/sd/allwinner-sdhost: Use AddressSpace for DMA transfers
 * target/arm: Fill in the WnR syndrome bit in mte_check_fail
 * target/arm: Clarify HCR_EL2 ARMCPRegInfo type
 * hw/arm/musicpal: Use AddressSpace for DMA transfers
 * hw/clock: Minor cleanups
 * hw/arm/sbsa-ref: fix typo breaking PCIe IRQs

# gpg: Signature made Fri 28 Aug 2020 10:23:02 BST
# gpg:                using RSA key E1A5C593CD419DE28E8315CF3C2525ED14360CDE
# gpg:                issuer "[email protected]"
# gpg: Good signature from "Peter Maydell <[email protected]>" [ultimate]
# gpg:                 aka "Peter Maydell <[email protected]>" [ultimate]
# gpg:                 aka "Peter Maydell <[email protected]>" [ultimate]
# Primary key fingerprint: E1A5 C593 CD41 9DE2 8E83  15CF 3C25 25ED 1436 0CDE

* remotes/pmaydell/tags/pull-target-arm-20200828: (35 commits)
  target/arm: Convert sq{, r}dmulh to gvec for aa64 advsimd
  target/arm: Convert integer multiply-add (indexed) to gvec for aa64 advsimd
  target/arm: Convert integer multiply (indexed) to gvec for aa64 advsimd
  target/arm: Generalize inl_qrdmlah_* helper functions
  target/arm: Tidy SVE tszimm shift formats
  target/arm: Split out gen_gvec_ool_zz
  target/arm: Split out gen_gvec_ool_zzz
  target/arm: Split out gen_gvec_ool_zzp
  target/arm: Merge helper_sve_clr_* and helper_sve_movz_*
  target/arm: Split out gen_gvec_ool_zzzp
  target/arm: Use tcg_gen_gvec_bitsel for trans_SEL_pppp
  target/arm: Clean up 4-operand predicate expansion
  target/arm: Merge do_vector2_p into do_mov_p
  target/arm: Rearrange {sve,fp}_check_access assert
  target/arm: Split out gen_gvec_fn_zzz, do_zzz_fn
  target/arm: Split out gen_gvec_fn_zz
  qemu/int128: Add int128_lshift
  armsse: Define ARMSSEClass correctly
  hw/misc/unimp: Display the offset with width of the region size
  hw/misc/unimp: Display the value with width of the access size
  ...

Signed-off-by: Peter Maydell <[email protected]>

Merge remote-tracking branch 'remotes/ehabkost/tags/machine-next-pull-request' into staging

Machine queue + QOM fixes and cleanups

Bug fix:
* numa: hmat: fix cache size check (Igor Mammedov)

QOM fixes and cleanups:
* Move QOM macros and typedefs to header files
* Use TYPE_* constants on TypeInfo structs
* Rename QOM type checking macros for consistency
* Rename enum values and typedefs that conflict with QOM
  type checking amcros
* Fix typos on QOM type checking macros
* Delete unused QOM type checking macros that use
  non-existing typedefs
* hvf: Add missing include
* xen-legacy-backend: Add missing typedef XenLegacyDevice

# gpg: Signature made Thu 27 Aug 2020 20:20:05 BST
# gpg:                using RSA key 5A322FD5ABC4D3DBACCFD1AA2807936F984DC5A6
# gpg:                issuer "[email protected]"
# gpg: Good signature from "Eduardo Habkost <[email protected]>" [full]
# Primary key fingerprint: 5A32 2FD5 ABC4 D3DB ACCF  D1AA 2807 936F 984D C5A6

* remotes/ehabkost/tags/machine-next-pull-request: (53 commits)
  dc390: Use TYPE_DC390_DEVICE constant
  ppce500: Use TYPE_PPC_E500_PCI_BRIDGE constant
  tosa: Use TYPE_TOSA_MISC_GPIO constant
  xlnx-zcu102: Use TYPE_ZCU102_MACHINE constant
  sclpconsole: Use TYPE_* constants
  amd_iommu: Use TYPE_AMD_IOMMU_PCI constant
  nios2_iic: Use TYPE_ALTERA_IIC constant
  etsec: Use TYPE_ETSEC_COMMON constant
  migration: Rename class type checking macros
  swim: Rename struct SWIM to Swim
  s390-virtio-ccw: Rename S390_MACHINE_CLASS macro
  nubus: Rename class type checking macros
  vfio/pci: Move QOM macros to header
  kvm: Move QOM macros to kvm.h
  mptsas: Move QOM macros to header
  pxa2xx: Move QOM macros to header
  rocker: Move QOM macros to header
  auxbus: Move QOM macros to header
  piix: Move QOM macros to header
  virtio-serial-bus: Move QOM macros to header
  ...

Signed-off-by: Peter Maydell <[email protected]>

target/arm: Convert sq{, r}dmulh to gvec for aa64 advsimd

Signed-off-by: Richard Henderson <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Convert integer multiply-add (indexed) to gvec for aa64 advsimd

Signed-off-by: Richard Henderson <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Convert integer multiply (indexed) to gvec for aa64 advsimd

Signed-off-by: Richard Henderson <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Generalize inl_qrdmlah_* helper functions

Unify add/sub helpers and add a parameter for rounding.
This will allow saturating non-rounding to reuse this code.

Signed-off-by: Richard Henderson <[email protected]>
[PMM: fixed accidental use of '=' rather than '+=' in do_sqrdmlah_s]
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Tidy SVE tszimm shift formats

Rather than require the user to fill in the immediate (shl or shr),
create full formats that include the immediate.

Signed-off-by: Richard Henderson <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Split out gen_gvec_ool_zz

Signed-off-by: Richard Henderson <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Split out gen_gvec_ool_zzz

Signed-off-by: Richard Henderson <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Split out gen_gvec_ool_zzp

Model after gen_gvec_fn_zzz et al.

Signed-off-by: Richard Henderson <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Merge helper_sve_clr_* and helper_sve_movz_*

The existing clr functions have only one vector argument, and so
can only clear in place.  The existing movz functions have two
vector arguments, and so can clear while moving.  Merge them, with
a flag that controls the sense of active vs inactive elements
being cleared.

Signed-off-by: Richard Henderson <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Split out gen_gvec_ool_zzzp

Model after gen_gvec_fn_zzz et al.

Signed-off-by: Richard Henderson <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Use tcg_gen_gvec_bitsel for trans_SEL_pppp

The gvec operation was added after the initial implementation
of the SEL instruction and was missed in the conversion.

Signed-off-by: Richard Henderson <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Clean up 4-operand predicate expansion

Move the check for !S into do_pppp_flags, which allows to merge in
do_vecop4_p.  Split out gen_gvec_fn_ppp without sve_access_check,
to mirror gen_gvec_fn_zzz.

Signed-off-by: Richard Henderson <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Merge do_vector2_p into do_mov_p

This is the only user of the function.

Signed-off-by: Richard Henderson <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Rearrange {sve,fp}_check_access assert

We want to ensure that access is checked by the time we ask
for a specific fp/vector register.  We want to ensure that
we do not emit two lots of code to raise an exception.

But sometimes it's difficult to cleanly organize the code
such that we never pass through sve_check_access exactly once.
Allow multiple calls so long as the result is true, that is,
no exception to be raised.

Signed-off-by: Richard Henderson <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Split out gen_gvec_fn_zzz, do_zzz_fn

Model gen_gvec_fn_zzz on gen_gvec_fn3 in translate-a64.c, but
indicating which kind of register and in which order.

Model do_zzz_fn on the other do_foo functions that take an
argument set and verify sve enabled.

Signed-off-by: Richard Henderson <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Split out gen_gvec_fn_zz

Model the new function on gen_gvec_fn2 in translate-a64.c, but
indicating which kind of register and in which order.  Since there
is only one user of do_vector2_z, fold it into do_mov_z.

Signed-off-by: Richard Henderson <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

qemu/int128: Add int128_lshift

Add left-shift to match the existing right-shift.

Signed-off-by: Richard Henderson <[email protected]>
Reviewed-by: Peter Maydell <[email protected]>
Message-id: 20200815013145[email protected]
Signed-off-by: Peter Maydell <[email protected]>

armsse: Define ARMSSEClass correctly

TYPE_ARM_SSE is a TYPE_SYS_BUS_DEVICE subclass, but
ARMSSEClass::parent_class is declared as DeviceClass.

It never caused any problems by pure luck:

We were not setting class_size for TYPE_ARM_SSE, so class_size of
TYPE_SYS_BUS_DEVICE was being used (sizeof(SysBusDeviceClass)).
This made the system allocate enough memory for TYPE_ARM_SSE
devices even though ARMSSEClass was too small for a sysbus
device.

Additionally, the ARMSSEClass::info field ended up at the same
offset as SysBusDeviceClass::explicit_ofw_unit_address.  This
would make sysbus_get_fw_dev_path() crash for the device.
Luckily, sysbus_get_fw_dev_path() never gets called for
TYPE_ARM_SSE devices, because qdev_get_fw_dev_path() is only used
by the boot device code, and TYPE_ARM_SSE devices don't appear at
the fw_boot_order list.

Signed-off-by: Eduardo Habkost <[email protected]>
Message-id: 20200826181006.4097163[email protected]
Reviewed-by: Peter Maydell <[email protected]>
Signed-off-by: Peter Maydell <[email protected]>

hw/misc/unimp: Display the offset with width of the region size

To have a better idea of how big is the region where the offset
belongs, display the value with the width of the region size
(i.e. a region of 0x1000 bytes uses 0x000 format).

Reviewed-by: Richard Henderson <[email protected]>
Signed-off-by: Philippe Mathieu-Daudé <[email protected]>
Message-id: 20200812190206[email protected]
Signed-off-by: Peter Maydell <[email protected]>

hw/misc/unimp: Display the value with width of the access size

To quickly notice the access size, display the value with the
width of the access (i.e. 16-bit access is displayed 0x0000,
while 8-bit access 0x00).

Reviewed-by: Richard Henderson <[email protected]>
Signed-off-by: Philippe Mathieu-Daudé <[email protected]>
Message-id: 20200812190206[email protected]
Signed-off-by: Peter Maydell <[email protected]>

hw/misc/unimp: Display value after offset

To better align the read/write accesses, display the value after
the offset (read accesses only display the offset).

Reviewed-by: Richard Henderson <[email protected]>
Signed-off-by: Philippe Mathieu-Daudé <[email protected]>
Message-id: 20200812190206[email protected]
Signed-off-by: Peter Maydell <[email protected]>

hw/qdev-clock: Avoid calling qdev_connect_clock_in after DeviceRealize

Clock canonical name is set in device_set_realized (see the block
added to hw/core/qdev.c in commit 0e6934f264).
If we connect a clock after the device is realized, this code is
not executed. This is currently not a problem as this name is only
used for trace events, however this disrupt tracing.

Add a comment to document qdev_connect_clock_in() must be called
before the device is realized, and assert this condition.

Signed-off-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Alistair Francis <[email protected]>
Message-id: 20200803105647[email protected]
Signed-off-by: Peter Maydell <[email protected]>

hw/qdev-clock: Uninline qdev_connect_clock_in()

We want to assert the device is not realized. To avoid overloading
this header including "hw/qdev-core.h", uninline the function first.

Signed-off-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Alistair Francis <[email protected]>
Message-id: 20200803105647[email protected]
Signed-off-by: Peter Maydell <[email protected]>

hw/arm/xilinx_zynq: Call qdev_connect_clock_in() before DeviceRealize

Clock canonical name is set in device_set_realized (see the block
added to hw/core/qdev.c in commit 0e6934f264).
If we connect a clock after the device is realized, this code is
not executed. This is currently not a problem as this name is only
used for trace events, however this disrupt tracing.

Fix by calling qdev_connect_clock_in() before realizing.

Signed-off-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Alistair Francis <[email protected]>
Message-id: 20200803105647[email protected]
Signed-off-by: Peter Maydell <[email protected]>

hw/arm/xilinx_zynq: Uninline cadence_uart_create()

As we want to call qdev_connect_clock_in() before the device
is realized, we need to uninline cadence_uart_create() first.

Signed-off-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Alistair Francis <[email protected]>
Message-id: 20200803105647[email protected]
Signed-off-by: Peter Maydell <[email protected]>

hw/net/allwinner-sun8i-emac: Use AddressSpace for DMA transfers

Allow the device to execute the DMA transfers in a different
AddressSpace.

The H3 SoC keeps using the system_memory address space,
but via the proper dma_memory_access() API.

Signed-off-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Richard Henderson <[email protected]>
Reviewed-by: Niek Linnenbank <[email protected]>
Tested-by: Niek Linnenbank <[email protected]>
Message-id: 20200814122907[email protected]
Signed-off-by: Peter Maydell <[email protected]>

hw/sd/allwinner-sdhost: Use AddressSpace for DMA transfers

Allow the device to execute the DMA transfers in a different
AddressSpace.

The A10 and H3 SoC keep using the system_memory address space,
but via the proper dma_memory_access() API.

Signed-off-by: Philippe Mathieu-Daudé <[email protected]>
Tested-by: Niek Linnenbank <[email protected]>
Reviewed-by: Niek Linnenbank <[email protected]>
Message-id: 20200814110057[email protected]
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Fill in the WnR syndrome bit in mte_check_fail

According to AArch64.TagCheckFault, none of the other ISS values are
provided, so we do not need to go so far as merge_syn_data_abort.
But we were missing the WnR bit.

Tested-by: Andrey Konovalov <[email protected]>
Reported-by: Andrey Konovalov <[email protected]>
Signed-off-by: Richard Henderson <[email protected]>
Message-id: 20200813200816.3037186[email protected]
Reviewed-by: Peter Maydell <[email protected]>
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Pass the entire mte descriptor to mte_check_fail

We need more information than just the mmu_idx in order
to create the proper exception syndrome.  Only change the
function signature so far.

Signed-off-by: Richard Henderson <[email protected]>
Message-id: 20200813200816.3037186[email protected]
Reviewed-by: Peter Maydell <[email protected]>
Signed-off-by: Peter Maydell <[email protected]>

target/arm: Clarify HCR_EL2 ARMCPRegInfo type

In commit ce4afed839 ("target/arm: Implement AArch32 HCR and HCR2")
the HCR_EL2 register has been changed from type NO_RAW (no underlying
state and does not support raw access for state saving/loading) to
type CONST (TCG can assume the value to be constant), removing the
read/write accessors.
We forgot to remove the previous type ARM_CP_NO_RAW. This is not
really a problem since the field is overwritten. However it makes
code review confuse, so remove it.

Signed-off-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Edgar E. Iglesias <[email protected]>
Reviewed-by: Richard Henderson <[email protected]>
Message-id: 20200812111223[email protected]
Signed-off-by: Peter Maydell <[email protected]>

hw/arm/musicpal: Use AddressSpace for DMA transfers

Allow the device to execute the DMA transfers in a different
AddressSpace.

We keep using the system_memory address space, but via the
proper dma_memory_access() API.

Signed-off-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Richard Henderson <[email protected]>
Message-id: 20200814125533[email protected]
Signed-off-by: Peter Maydell <[email protected]>

hw/clock: Only propagate clock changes if the clock is changed

Avoid propagating the clock change when the clock does not change.

Signed-off-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Richard Henderson <[email protected]>
Message-id: 20200806123858[email protected]
Signed-off-by: Peter Maydell <[email protected]>

hw/clock: Let clock_set() return boolean value

Let clock_set() return a boolean value whether the clock
has been updated or not.

Signed-off-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Richard Henderson <[email protected]>
Message-id: 20200806123858[email protected]
Signed-off-by: Peter Maydell <[email protected]>

hw/clock: Remove unused clock_init*() functions

clock_init*() inlined funtions are simple wrappers around
clock_set*() and are not used. Remove them in favor of clock_set*().

Signed-off-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Richard Henderson <[email protected]>
Message-id: 20200806123858[email protected]
Signed-off-by: Peter Maydell <[email protected]>

hw/arm/sbsa-ref: fix typo breaking PCIe IRQs

Fixing a typo in a previous patch that translated an "i" to a 1
and therefore breaking the allocation of PCIe interrupts. This was
discovered when virtio-net-pci devices ceased to function correctly.

Cc: [email protected]
Fixes: 48ba18e6d3f3 ("hw/arm/sbsa-ref: Simplify by moving the gic in the machine state")
Signed-off-by: Graeme Gregory <[email protected]>
Reviewed-by: Philippe Mathieu-Daudé <[email protected]>
Message-id: 20200821083853[email protected]
Signed-off-by: Peter Maydell <[email protected]>

Merge remote-tracking branch 'remotes/bonzini-gitlab/tags/for-upstream' into staging

meson fixes:
* fix PIE default for Windows
* tweak conditions for requiring SDL2 and pixman
* move NSIS building to Meson
* fix creation of systemtap files
* add --ninja option for Windows builds
* other fixes for stupid bugs and for Windows paths

meson related:
* forward port of dtrace fix for MacOS
* change colorout option to auto

# gpg: Signature made Thu 27 Aug 2020 17:58:21 BST
# gpg:                using RSA key F13338574B662389866C7682BFFBD25F78C7AE83
# gpg:                issuer "[email protected]"
# gpg: Good signature from "Paolo Bonzini <[email protected]>" [full]
# gpg:                 aka "Paolo Bonzini <[email protected]>" [full]
# 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-gitlab/tags/for-upstream:
  ninjatool: Fixes E$$: in generated Makefile.ninja
  meson: set colorout to auto
  meson: Mingw64 gcc doesn't recognize system include_type for sdl2
  meson: fix relpath failure on Win32
  ninjatool: quote dollars in variables
  meson: don't require CONFIG_VTE for the GTK UI
  meson: cleanup xkbcommon detection
  configure: add --ninja option
  meson: skip SDL2 detection if --disable-system
  meson: move pixman detection to meson
  configure: default to PIE disabled on Windows platforms
  meson: Fix chardev-baum.so name
  meson: Fix meson build with --enable-libdaxctl
  scripts/qemu-version.sh: Add missing space before ']'
  meson: Don't make object files for dtrace on macOS
  meson: Build qemu-nbd on macOS again
  trace: fix creation of systemtap files

Signed-off-by: Peter Maydell <[email protected]>

dc390: Use TYPE_DC390_DEVICE constant

This will make future conversion to use OBJECT_DECLARE* easier.

Signed-off-by: Eduardo Habkost <[email protected]>
Reviewed-by: Li Qiang <[email protected]>
Message-Id: <20200826184334.4120620[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

ppce500: Use TYPE_PPC_E500_PCI_BRIDGE constant

This will make future conversion to use OBJECT_DECLARE* easier.

Signed-off-by: Eduardo Habkost <[email protected]>
Acked-by: David Gibson <[email protected]>
Reviewed-by: Li Qiang <[email protected]>
Message-Id: <20200826184334.4120620[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

tosa: Use TYPE_TOSA_MISC_GPIO constant

This will make future conversion to use OBJECT_DECLARE* easier.

Signed-off-by: Eduardo Habkost <[email protected]>
Reviewed-by: Li Qiang <[email protected]>
Message-Id: <20200826184334.4120620[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

xlnx-zcu102: Use TYPE_ZCU102_MACHINE constant

This will make future conversion to use OBJECT_DECLARE* easier.

Signed-off-by: Eduardo Habkost <[email protected]>
Reviewed-by: Alistair Francis <[email protected]>
Reviewed-by: Edgar E. Iglesias <[email protected]>
Reviewed-by: Li Qiang <[email protected]>
Message-Id: <20200826184334.4120620[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

sclpconsole: Use TYPE_* constants

This will make future conversion to use OBJECT_DECLARE* easier.

Signed-off-by: Eduardo Habkost <[email protected]>
Acked-by: Cornelia Huck <[email protected]>
Reviewed-by: Li Qiang <[email protected]>
Reviewed-by: Thomas Huth <[email protected]>
Message-Id: <20200826184334.4120620[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

amd_iommu: Use TYPE_AMD_IOMMU_PCI constant

This will make future conversion to use OBJECT_DECLARE* easier.

Signed-off-by: Eduardo Habkost <[email protected]>
Reviewed-by: Li Qiang <[email protected]>
Reviewed-by: Michael S. Tsirkin <[email protected]>
Message-Id: <20200826184334.4120620[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

nios2_iic: Use TYPE_ALTERA_IIC constant

This will make future conversion to use OBJECT_DECLARE* easier.

Signed-off-by: Eduardo Habkost <[email protected]>
Reviewed-by: Li Qiang <[email protected]>
Message-Id: <20200826184334.4120620[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

etsec: Use TYPE_ETSEC_COMMON constant

This will make future conversion to use OBJECT_DECLARE* easier.

Signed-off-by: Eduardo Habkost <[email protected]>
Acked-by: David Gibson <[email protected]>
Reviewed-by: Li Qiang <[email protected]>
Message-Id: <20200826184334.4120620[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Juan Quintela <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Acked-by: Laurent Vivier <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Acked-by: Cornelia Huck <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Acked-by: Laurent Vivier <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

vfio/pci: Move QOM macros to header

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

mptsas: Move QOM macros to header

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

pxa2xx: Move QOM macros to header

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

rocker: Move QOM macros to header

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

auxbus: Move QOM macros to header

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

piix: Move QOM macros to header

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

virtio-serial-bus: Move QOM macros to header

This will make future conversion to OBJECT_DECLARE* easier.

Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Reviewed-by: Thomas Huth <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

nubus: Delete unused NUBUS_BRIDGE macro

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

Reviewed-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Acked-by: Laurent Vivier <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Reviewed-by: Thomas Huth <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Reviewed-by: Gerd Hoffmann <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Roman Bolshakov <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

virtio-ccw: Fix definition of VIRTIO_CCW_BUS_GET_CLASS

The macro was incorrectly defined using OBJECT_CHECK.

Acked-by: Cornelia Huck <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Reviewed-by: David Hildenbrand <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

versatile: Fix typo in PCI_VPB_HOST definition

Fixes: cd93dbf375bd ("versatile_pci: Update to realize and instance init functions")
Reviewed-by: Philippe Mathieu-Daudé <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Reviewed-by: Cédric Le Goater <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Niek Linnenbank <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Reviewed-by: Daniel P. Berrangé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>
Tested-By: Roman Bolshakov <[email protected]>
Message-Id: <20200825192110.3528606[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

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 <[email protected]>
Message-Id: <20200821100519.1325691[email protected]>
Reviewed-by: Philippe Mathieu-Daudé <[email protected]>
Signed-off-by: Eduardo Habkost <[email protected]>

ninjatool: Fixes E$$: in generated Makefile.ninja

Even though SIMPLE_PATH_RE is used with re.match (which anchors the
match implictly to the beginning of the string) it also needs an
end-of-string anchor in order to match the full path token.

Otherwise, the match would succeed incorrectly for $ and : characters
contained in the path, for example if the path starts with C:/ or E:/.

Signed-off-by: Yonggang Luo <[email protected]>
Tested-by: Mark Cave-Ayland <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>

meson: set colorout to auto

Dunno why the default is set to "always".  IMHO it should be "auto",
i.e. only colorize in case stdout goes to a terminal.  Cluttering
logfiles and confusing compiler message parsers with terminal control
sequences is not nice ...

Signed-off-by: Gerd Hoffmann <[email protected]>
Reviewed-by: Marc-André Lureau <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>

meson: Mingw64 gcc doesn't recognize system include_type for sdl2

Windows paths result in command lines like "-isystemC:/msys64/..." that
are not recognized by GCC.  "include_type: 'system'" was only included
in an attempt to fix the -Wundef warnings in SDL 2.0.8, but it was not
effective.  Therefore we can fix this by remove the include_type.

Signed-off-by: Yonggang Luo <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>

meson: fix relpath failure on Win32

On win32, os.path.relpath can raise an exception when computing
for example C:/msys64/mingw64/x.exe relative to E:/path/qemu-build.
Use try...except to avoid this, just using an absolute path in
this case.

Signed-off-by: Yonggang Luo <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>

ninjatool: quote dollars in variables

Otherwise, dollars (such as in the special $ORIGIN rpath) are
eaten by Make.

Reported-by: Laurent Vivier <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>

meson: don't require CONFIG_VTE for the GTK UI

Prevously CONFIG_VTE was not required to build QEMU with GTK UI support as not
all platforms have VTE available (in particular Windows).

Remove this requirement from the meson build system to enable QEMU to be built
with GTK UI support for Windows once again.

Signed-off-by: Mark Cave-Ayland <[email protected]>
Reviewed-by: Gerd Hoffmann <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>

meson: cleanup xkbcommon detection

Signed-off-by: Paolo Bonzini <[email protected]>

configure: add --ninja option

On Windows it is not possible to invoke a Python script as $NINJA.
If ninja is present use it directly, while if it is not we can
keep using ninjatool.

Reported-by: Mark Cave-Ayland <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>

meson: skip SDL2 detection if --disable-system

SDL is only used for system emulation; avoid spurious warnings for
static --disable-system emulation by skipping the detection of
the library if there are no system emulation targets.

Reported-by: Peter Maydell <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>