Re: [PATCH v0 0/2] virtio-blk and vhost-user-blk cross-device migration

[Eduardo Habkost]

On Tue, Oct 05, 2021 at 03:01:05PM +0100, Dr. David Alan Gilbert wrote:
> * Michael S. Tsirkin (m...@redhat.com) wrote:
> > On Tue, Oct 05, 2021 at 02:18:40AM +0300, Roman Kagan wrote:
> > > On Mon, Oct 04, 2021 at 11:11:00AM -0400, Michael S. Tsirkin wrote:
> > > > On Mon, Oct 04, 2021 at 06:07:29PM +0300, Denis Plotnikov wrote:
> > > > > It might be useful for the cases when a slow block layer should be 
> > > > > replaced
> > > > > with a more performant one on running VM without stopping, i.e. with 
> > > > > very low
> > > > > downtime comparable with the one on migration.
> > > > > 
> > > > > It's possible to achive that for two reasons:
> > > > > 
> > > > > 1.The VMStates of "virtio-blk" and "vhost-user-blk" are almost the 
> > > > > same.
> > > > >   They consist of the identical VMSTATE_VIRTIO_DEVICE and differs from
> > > > >   each other in the values of migration service fields only.
> > > > > 2.The device driver used in the guest is the same: virtio-blk
> > > > > 
> > > > > In the series cross-migration is achieved by adding a new type.
> > > > > The new type uses virtio-blk VMState instead of vhost-user-blk 
> > > > > specific
> > > > > VMstate, also it implements migration save/load callbacks to be 
> > > > > compatible
> > > > > with migration stream produced by "virtio-blk" device.
> > > > > 
> > > > > Adding the new type instead of modifying the existing one is 
> > > > > convenent.
> > > > > It ease to differ the new virtio-blk-compatible vhost-user-blk
> > > > > device from the existing non-compatible one using qemu machinery 
> > > > > without any
> > > > > other modifiactions. That gives all the variety of qemu device related
> > > > > constraints out of box.
> > > > 
> > > > Hmm I'm not sure I understand. What is the advantage for the user?
> > > > What if vhost-user-blk became an alias for vhost-user-virtio-blk?
> > > > We could add some hacks to make it compatible for old machine types.
> > > 
> > > The point is that virtio-blk and vhost-user-blk are not
> > > migration-compatible ATM.  OTOH they are the same device from the guest
> > > POV so there's nothing fundamentally preventing the migration between
> > > the two.  In particular, we see it as a means to switch between the
> > > storage backend transports via live migration without disrupting the
> > > guest.
> > > 
> > > Migration-wise virtio-blk and vhost-user-blk have in common
> > > 
> > > - the content of the VMState -- VMSTATE_VIRTIO_DEVICE
> > > 
> > > The two differ in
> > > 
> > > - the name and the version of the VMStateDescription
> > > 
> > > - virtio-blk has an extra migration section (via .save/.load callbacks
> > >   on VirtioDeviceClass) containing requests in flight
> > > 
> > > It looks like to become migration-compatible with virtio-blk,
> > > vhost-user-blk has to start using VMStateDescription of virtio-blk and
> > > provide compatible .save/.load callbacks.  It isn't entirely obvious how
> > > to make this machine-type-dependent, so we came up with a simpler idea
> > > of defining a new device that shares most of the implementation with the
> > > original vhost-user-blk except for the migration stuff.  We're certainly
> > > open to suggestions on how to reconcile this under a single
> > > vhost-user-blk device, as this would be more user-friendly indeed.
> > > 
> > > We considered using a class property for this and defining the
> > > respective compat clause, but IIUC the class constructors (where .vmsd
> > > and .save/.load are defined) are not supposed to depend on class
> > > properties.
> > > 
> > > Thanks,
> > > Roman.
> > 
> > So the question is how to make vmsd depend on machine type.
> > CC Eduardo who poked at this kind of compat stuff recently,
> > paolo who looked at qom things most recently and dgilbert
> > for advice on migration.
> 
> I don't think I've seen anyone change vmsd name dependent on machine
> type; making fields appear/disappear is easy - that just ends up as a
> property on the device that's checked;  I guess if that property is
> global (rather than per instance) then you can check it in
> vhost_user_blk_class_init and swing the dc->vmsd pointer?

class_init can be called very early during QEMU initialization,
so it's too early to make decisions based on machine type.

Making a specific vmsd appear/disappear based on machine
configuration or state is "easy", by implementing
VMStateDescription.needed.  But this would require registering
both vmsds (one of them would need to be registered manually
instead of using DeviceClass.vmsd).

I don't remember what are the consequences of not using
DeviceClass.vmsd to register a vmsd, I only remember it was
subtle.  See commit b170fce3dd06 ("cpu: Register
VMStateDescription through CPUState") and related threads.  CCing
Philippe, who might remember the details here.

If that's an important use case, I would suggest allowing devices
to implement a DeviceClass.get_vmsd method, which would override
DeviceClass.vmsd if necessary.  Is the 

Re: [PATCH for-6.2 05/12] [automated] Move QOM typedefs and add missing includes

[Eduardo Habkost]

On Tue, Aug 10, 2021 at 02:01:40PM +0200, Juan Quintela wrote:
> Eduardo Habkost  wrote:
> > Some typedefs and macros are defined after the type check macros.
> > This makes it difficult to automatically replace their
> > definitions with OBJECT_DECLARE_TYPE.
> >
> > Patch generated using:
> >
> >  $ ./scripts/codeconverter/converter.py -i --pattern=MoveSymbols \
> > $(git grep -l '' -- '*.[ch]')
> >
> > which will:
> > - split "typdef struct { ... } TypedefName" declarations
> > - move the typedefs and #defines above the type check macros
> > - add missing #include "qom/object.h" lines if necessary
> >
> > Signed-off-by: Eduardo Habkost 
> 
> Reviewed-by: Juan Quintela 

Thanks!

> 
> Just curious, how did my name ended on the CC'd list?  I don't see any
> file that I touched or that is migration related.

include/hw/vmstate-if.h is in the migration section in
MAINTAINERS.

-- 
Eduardo

[PATCH for-6.2 12/12] [automated] Use OBJECT_DECLARE_SIMPLE_TYPE when possible

[Eduardo Habkost]

Replace typedef + DECLARE_INSTANCE_CHECKER with
equivalent OBJECT_DECLARE_SIMPLE_TYPE macro.

Generated using:

$ ./scripts/codeconverter/converter.py -i \
  --pattern=AddObjectDeclareSimpleType $(git grep -l '' -- '*.[ch]')

Signed-off-by: Eduardo Habkost 
---
Cc: Thomas Huth 
Cc: Paul Burton 
Cc: Aleksandar Rikalo 
Cc: "Philippe Mathieu-Daudé" 
Cc: Aurelien Jarno 
Cc: Jiaxun Yang 
Cc: Havard Skinnemoen 
Cc: Tyrone Ting 
Cc: Pavel Pisa 
Cc: Vikram Garhwal 
Cc: Jason Wang 
Cc: Keith Busch 
Cc: Klaus Jensen 
Cc: "Michael S. Tsirkin" 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Cornelia Huck 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Gerd Hoffmann 
Cc: Vijai Kumar K 
Cc: Alistair Francis 
Cc: Bin Meng 
Cc: Palmer Dabbelt 
Cc: "Edgar E. Iglesias" 
Cc: Peter Maydell 
Cc: Laurent Vivier 
Cc: "Cédric Le Goater" 
Cc: Andrew Jeffery 
Cc: Joel Stanley 
Cc: Andrew Baumann 
Cc: Francisco Iglesias 
Cc: David Gibson 
Cc: Greg Kurz 
Cc: Bastian Koppelmann 
Cc: qemu-de...@nongnu.org
Cc: qemu-...@nongnu.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
Cc: qemu-ri...@nongnu.org
Cc: qemu-...@nongnu.org
---
 hw/nvme/nvme.h  | 10 +++---
 hw/usb/hcd-xhci-pci.h   |  4 +---
 hw/usb/hcd-xhci-sysbus.h|  4 +---
 include/hw/adc/npcm7xx_adc.h|  4 +---
 include/hw/char/shakti_uart.h   |  4 +---
 include/hw/dma/sifive_pdma.h|  4 +---
 include/hw/dma/xlnx_csu_dma.h   |  4 +---
 include/hw/gpio/sifive_gpio.h   |  4 +---
 include/hw/intc/m68k_irqc.h |  4 +---
 include/hw/intc/sifive_clint.h  |  4 +---
 include/hw/intc/sifive_plic.h   |  4 +---
 include/hw/misc/aspeed_lpc.h|  4 +---
 include/hw/misc/bcm2835_cprman_internals.h  | 12 
 include/hw/misc/led.h   |  3 +--
 include/hw/misc/mchp_pfsoc_dmc.h|  8 ++--
 include/hw/misc/mchp_pfsoc_ioscb.h  |  4 +---
 include/hw/misc/mchp_pfsoc_sysreg.h |  4 +---
 include/hw/misc/npcm7xx_clk.h   |  3 +--
 include/hw/misc/npcm7xx_gcr.h   |  4 +---
 include/hw/misc/npcm7xx_mft.h   |  4 +---
 include/hw/misc/npcm7xx_pwm.h   |  3 +--
 include/hw/misc/sifive_e_prci.h |  4 +---
 include/hw/misc/sifive_test.h   |  4 +---
 include/hw/misc/sifive_u_otp.h  |  4 +---
 include/hw/misc/sifive_u_prci.h |  4 +---
 include/hw/misc/xlnx-versal-xramc.h |  4 +---
 include/hw/net/npcm7xx_emc.h|  4 +---
 include/hw/net/xlnx-zynqmp-can.h|  4 +---
 include/hw/ppc/spapr_drc.h  |  4 +---
 include/hw/register.h   |  3 +--
 include/hw/riscv/microchip_pfsoc.h  |  4 +---
 include/hw/riscv/shakti_c.h |  8 ++--
 include/hw/riscv/sifive_e.h |  4 +---
 include/hw/riscv/sifive_u.h |  4 +---
 include/hw/sd/cadence_sdhci.h   |  4 +---
 include/hw/ssi/sifive_spi.h |  4 +---
 include/hw/timer/npcm7xx_timer.h|  3 +--
 include/hw/tricore/tricore_testdevice.h |  4 +---
 include/hw/usb/hcd-dwc3.h   |  4 +---
 include/hw/usb/xlnx-versal-usb2-ctrl-regs.h |  4 +---
 hw/m68k/mcf5206.c   |  4 +---
 hw/mips/boston.c|  4 +---
 hw/misc/npcm7xx_clk.c   |  9 +++--
 hw/net/can/ctucan_pci.c |  4 +---
 hw/s390x/vhost-user-fs-ccw.c|  4 +---
 hw/sensor/adm1272.c |  4 +---
 hw/sensor/max34451.c|  4 +---
 47 files changed, 56 insertions(+), 154 deletions(-)

diff --git a/hw/nvme/nvme.h b/hw/nvme/nvme.h
index c4c43da5c17..a9ee5c4f1de 100644
--- a/hw/nvme/nvme.h
+++ b/hw/nvme/nvme.h
@@ -42,9 +42,7 @@ typedef struct NvmeBus {
 } NvmeBus;
 
 #define TYPE_NVME_SUBSYS "nvme-subsys"
-typedef struct NvmeSubsystem NvmeSubsystem;
-DECLARE_INSTANCE_CHECKER(NvmeSubsystem, NVME_SUBSYS,
- TYPE_NVME_SUBSYS)
+OBJECT_DECLARE_SIMPLE_TYPE(NvmeSubsystem, NVME_SUBSYS)
 
 struct NvmeSubsystem {
 DeviceState parent_obj;
@@ -83,8 +81,7 @@ static inline NvmeNamespace *nvme_subsys_ns(NvmeSubsystem 
*subsys,
 }
 
 #define TYPE_NVME_NS "nvme-ns"
-DECLARE_INSTANCE_CHECKER(NvmeNamespace, NVME_NS,
- TYPE_NVME_NS)
+OBJECT_DECLARE_SIMPLE_TYPE(NvmeNamespace, NVME_NS)
 
 typedef struct NvmeZone {
 NvmeZoneDescr   d;
@@ -377,8 +374,7 @@ typedef struct NvmeCQueue {
 } NvmeCQueue;
 
 #define TYPE_NVME "nvme"
-DECLARE_INSTANCE_CHECKER(NvmeCtrl, NVME,
- TYPE_NVME)
+OBJECT_DECLARE_SIMPLE_TYPE(NvmeCtrl, NVME)
 
 typedef struct NvmeParams {
 char *serial;
diff --git a/hw/usb/hcd-xhci-pci.h b/hw/usb/hcd-xhci-pci.h
index d83aad82e04

[PATCH for-6.2 11/12] [automated] Use OBJECT_DECLARE_TYPE when possible

[Eduardo Habkost]

Replace typedefs + DECLARE_OBJ_CHECKERS with equivalent
OBJECT_DECLARE_TYPE macro.

Generated using:

$ ./scripts/codeconverter/converter.py -i \
  --pattern=AddObjectDeclareType $(git grep -l '' -- '*.[ch]')

Signed-off-by: Eduardo Habkost 
---
Cc: John Snow 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Gerd Hoffmann 
Cc: "Daniel P. Berrangé" 
Cc: David Gibson 
Cc: Greg Kurz 
Cc: "Cédric Le Goater" 
Cc: Richard Henderson 
Cc: Paolo Bonzini 
Cc: qemu-block@nongnu.org
Cc: qemu-de...@nongnu.org
Cc: qemu-...@nongnu.org
---
 hw/usb/u2f.h| 6 +-
 include/crypto/tlscreds.h   | 5 +
 include/hw/ppc/spapr_drc.h  | 5 +
 include/hw/ppc/spapr_xive.h | 5 +
 include/qemu/accel.h| 4 +---
 hw/block/fdc-sysbus.c   | 5 +
 6 files changed, 6 insertions(+), 24 deletions(-)

diff --git a/hw/usb/u2f.h b/hw/usb/u2f.h
index 5767ca8fac9..e6e3ead40f1 100644
--- a/hw/usb/u2f.h
+++ b/hw/usb/u2f.h
@@ -32,13 +32,10 @@
 #define U2FHID_PACKET_SIZE 64
 #define U2FHID_PENDING_IN_NUM 32
 
-typedef struct U2FKeyState U2FKeyState;
 typedef struct U2FKeyInfo U2FKeyInfo;
 
 #define TYPE_U2F_KEY "u2f-key"
-typedef struct U2FKeyClass U2FKeyClass;
-DECLARE_OBJ_CHECKERS(U2FKeyState, U2FKeyClass,
- U2F_KEY, TYPE_U2F_KEY)
+OBJECT_DECLARE_TYPE(U2FKeyState, U2FKeyClass, U2F_KEY)
 
 /*
  * Callbacks to be used by the U2F key base device (i.e. hw/u2f.c)
@@ -69,7 +66,6 @@ struct U2FKeyState {
 uint8_t pending_in_end;
 uint8_t pending_in_num;
 };
-typedef struct U2FKeyState U2FKeyState;
 
 /*
  * API to be used by the U2F key device variants (i.e. hw/u2f-*.c)
diff --git a/include/crypto/tlscreds.h b/include/crypto/tlscreds.h
index 2a8a8570109..617979a3986 100644
--- a/include/crypto/tlscreds.h
+++ b/include/crypto/tlscreds.h
@@ -25,10 +25,7 @@
 #include "qom/object.h"
 
 #define TYPE_QCRYPTO_TLS_CREDS "tls-creds"
-typedef struct QCryptoTLSCreds QCryptoTLSCreds;
-typedef struct QCryptoTLSCredsClass QCryptoTLSCredsClass;
-DECLARE_OBJ_CHECKERS(QCryptoTLSCreds, QCryptoTLSCredsClass, QCRYPTO_TLS_CREDS,
- TYPE_QCRYPTO_TLS_CREDS)
+OBJECT_DECLARE_TYPE(QCryptoTLSCreds, QCryptoTLSCredsClass, QCRYPTO_TLS_CREDS)
 
 
 #define QCRYPTO_TLS_CREDS_DH_PARAMS "dh-params.pem"
diff --git a/include/hw/ppc/spapr_drc.h b/include/hw/ppc/spapr_drc.h
index 540439812f0..ff876fd74ca 100644
--- a/include/hw/ppc/spapr_drc.h
+++ b/include/hw/ppc/spapr_drc.h
@@ -20,10 +20,7 @@
 #include "qapi/error.h"
 
 #define TYPE_SPAPR_DR_CONNECTOR "spapr-dr-connector"
-typedef struct SpaprDrc SpaprDrc;
-typedef struct SpaprDrcClass SpaprDrcClass;
-DECLARE_OBJ_CHECKERS(SpaprDrc, SpaprDrcClass,
- SPAPR_DR_CONNECTOR, TYPE_SPAPR_DR_CONNECTOR)
+OBJECT_DECLARE_TYPE(SpaprDrc, SpaprDrcClass, SPAPR_DR_CONNECTOR)
 
 #define TYPE_SPAPR_DRC_PHYSICAL "spapr-drc-physical"
 typedef struct SpaprDrcPhysical SpaprDrcPhysical;
diff --git a/include/hw/ppc/spapr_xive.h b/include/hw/ppc/spapr_xive.h
index 333095c3fd9..9e7c46c801f 100644
--- a/include/hw/ppc/spapr_xive.h
+++ b/include/hw/ppc/spapr_xive.h
@@ -15,10 +15,7 @@
 #include "qom/object.h"
 
 #define TYPE_SPAPR_XIVE "spapr-xive"
-typedef struct SpaprXive SpaprXive;
-typedef struct SpaprXiveClass SpaprXiveClass;
-DECLARE_OBJ_CHECKERS(SpaprXive, SpaprXiveClass,
- SPAPR_XIVE, TYPE_SPAPR_XIVE)
+OBJECT_DECLARE_TYPE(SpaprXive, SpaprXiveClass, SPAPR_XIVE)
 
 struct SpaprXive {
 XiveRouterparent;
diff --git a/include/qemu/accel.h b/include/qemu/accel.h
index 8dc4ccc39ef..1b7696f9fbc 100644
--- a/include/qemu/accel.h
+++ b/include/qemu/accel.h
@@ -54,15 +54,13 @@ struct AccelClass {
  */
 GPtrArray *compat_props;
 };
-typedef struct AccelClass AccelClass;
 
 #define TYPE_ACCEL_BASE "accel"
 
 #define ACCEL_CLASS_SUFFIX  "-" TYPE_ACCEL_BASE
 #define ACCEL_CLASS_NAME(a) (a ACCEL_CLASS_SUFFIX)
 
-DECLARE_OBJ_CHECKERS(AccelState, AccelClass,
- ACCEL_BASE, TYPE_ACCEL_BASE)
+OBJECT_DECLARE_TYPE(AccelState, AccelClass, ACCEL_BASE)
 
 AccelClass *accel_find(const char *opt_name);
 AccelState *current_accel(void);
diff --git a/hw/block/fdc-sysbus.c b/hw/block/fdc-sysbus.c
index 57fc8773f12..ecce27db34e 100644
--- a/hw/block/fdc-sysbus.c
+++ b/hw/block/fdc-sysbus.c
@@ -33,10 +33,7 @@
 #include "trace.h"
 
 #define TYPE_SYSBUS_FDC "base-sysbus-fdc"
-typedef struct FDCtrlSysBusClass FDCtrlSysBusClass;
-typedef struct FDCtrlSysBus FDCtrlSysBus;
-DECLARE_OBJ_CHECKERS(FDCtrlSysBus, FDCtrlSysBusClass,
- SYSBUS_FDC, TYPE_SYSBUS_FDC)
+OBJECT_DECLARE_TYPE(FDCtrlSysBus, FDCtrlSysBusClass, SYSBUS_FDC)
 
 struct FDCtrlSysBusClass {
 /*< private >*/
-- 
2.31.1

[PATCH for-6.2 07/12] [automated] Use DECLARE_*CHECKER* macros when possible

[Eduardo Habkost]

Converting existing QOM types to OBJECT_DECLARE_TYPE is not
always trivial (due to inconsistent type/macro naming schemes),
but at least converting existing manual QOM type checking macros
to use DECLARE_*CHECKER* is a simpler process, and should at
least discourage people from defining new QOM type checker macros
manually.

Generated using:

  $ ./scripts/codeconverter/converter.py -i \
--pattern=TypeCheckMacro $(git grep -l '' -- '*.[ch]')

Signed-off-by: Eduardo Habkost 
---
Cc: "Marc-André Lureau" 
Cc: Paolo Bonzini 
Cc: Patrick Venture 
Cc: Thomas Huth 
Cc: "Michael S. Tsirkin" 
Cc: Igor Mammedov 
Cc: Alexander Bulekov 
Cc: Bandan Das 
Cc: Stefan Hajnoczi 
Cc: Keith Busch 
Cc: Klaus Jensen 
Cc: Cornelia Huck 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Gerd Hoffmann 
Cc: Havard Skinnemoen 
Cc: Tyrone Ting 
Cc: Vijai Kumar K 
Cc: Alistair Francis 
Cc: Bin Meng 
Cc: Palmer Dabbelt 
Cc: "Edgar E. Iglesias" 
Cc: Peter Maydell 
Cc: Laurent Vivier 
Cc: "Cédric Le Goater" 
Cc: Andrew Jeffery 
Cc: Joel Stanley 
Cc: Jason Wang 
Cc: Vikram Garhwal 
Cc: Francisco Iglesias 
Cc: David Gibson 
Cc: Greg Kurz 
Cc: Bastian Koppelmann 
Cc: Taylor Simpson 
Cc: qemu-de...@nongnu.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
Cc: qemu-...@nongnu.org
Cc: qemu-ri...@nongnu.org
Cc: qemu-...@nongnu.org
---
 hw/nvme/nvme.h  | 12 ++--
 hw/usb/hcd-xhci-pci.h   |  4 ++--
 hw/usb/hcd-xhci-sysbus.h|  4 ++--
 hw/usb/u2f.h|  8 ++--
 include/hw/adc/npcm7xx_adc.h|  4 ++--
 include/hw/arm/npcm7xx.h| 15 ---
 include/hw/char/shakti_uart.h   |  4 ++--
 include/hw/dma/sifive_pdma.h|  4 ++--
 include/hw/dma/xlnx_csu_dma.h   |  4 ++--
 include/hw/gpio/npcm7xx_gpio.h  |  4 ++--
 include/hw/i2c/npcm7xx_smbus.h  |  4 ++--
 include/hw/intc/m68k_irqc.h |  4 ++--
 include/hw/intc/sifive_clint.h  |  4 ++--
 include/hw/mem/npcm7xx_mc.h |  3 ++-
 include/hw/misc/aspeed_lpc.h|  3 ++-
 include/hw/misc/mchp_pfsoc_dmc.h| 10 --
 include/hw/misc/mchp_pfsoc_ioscb.h  |  4 ++--
 include/hw/misc/mchp_pfsoc_sysreg.h |  5 ++---
 include/hw/misc/npcm7xx_clk.h   |  3 ++-
 include/hw/misc/npcm7xx_gcr.h   |  3 ++-
 include/hw/misc/npcm7xx_mft.h   |  4 ++--
 include/hw/misc/npcm7xx_pwm.h   |  4 ++--
 include/hw/misc/npcm7xx_rng.h   |  3 ++-
 include/hw/misc/xlnx-versal-xramc.h |  4 ++--
 include/hw/net/npcm7xx_emc.h|  4 ++--
 include/hw/net/xlnx-zynqmp-can.h|  4 ++--
 include/hw/nvram/npcm7xx_otp.h  |  3 ++-
 include/hw/ppc/spapr_drc.h  | 13 -
 include/hw/ppc/spapr_xive.h |  7 ++-
 include/hw/riscv/microchip_pfsoc.h  |  9 -
 include/hw/riscv/shakti_c.h |  8 
 include/hw/riscv/sifive_e.h |  8 
 include/hw/riscv/sifive_u.h |  8 
 include/hw/sd/cadence_sdhci.h   |  4 ++--
 include/hw/ssi/npcm7xx_fiu.h|  3 ++-
 include/hw/ssi/sifive_spi.h |  3 ++-
 include/hw/timer/npcm7xx_timer.h|  4 ++--
 include/hw/tricore/tricore_testdevice.h |  4 ++--
 include/hw/usb/hcd-dwc3.h   |  4 ++--
 include/hw/usb/xlnx-usb-subsystem.h |  4 ++--
 include/hw/usb/xlnx-versal-usb2-ctrl-regs.h |  4 ++--
 include/hw/watchdog/sbsa_gwdt.h |  4 ++--
 include/qemu/accel.h|  8 ++--
 target/hexagon/cpu.h|  8 ++--
 chardev/char-parallel.c |  6 ++
 hw/i2c/i2c_mux_pca954x.c|  4 ++--
 hw/m68k/mcf5206.c   |  3 ++-
 hw/mem/sparse-mem.c |  3 ++-
 hw/misc/sbsa_ec.c   |  3 ++-
 hw/s390x/vhost-user-fs-ccw.c|  4 ++--
 hw/sensor/adm1272.c |  3 ++-
 hw/sensor/max34451.c|  3 ++-
 hw/usb/u2f-emulated.c   |  4 ++--
 hw/usb/u2f-passthru.c   |  4 ++--
 54 files changed, 126 insertions(+), 146 deletions(-)

diff --git a/hw/nvme/nvme.h b/hw/nvme/nvme.h
index 0840f585d6e..c4c43da5c17 100644
--- a/hw/nvme/nvme.h
+++ b/hw/nvme/nvme.h
@@ -43,8 +43,8 @@ typedef struct NvmeBus {
 
 #define TYPE_NVME_SUBSYS "nvme-subsys"
 typedef struct NvmeSubsystem NvmeSubsystem;
-#define NVME_SUBSYS(obj) \
-OBJECT_CHECK(NvmeSubsystem, (obj), TYPE_NVME_SUBSYS)
+DECLARE_INSTANCE_CHECKER(NvmeSubsystem, NVME_SUBSYS,
+ TYPE_NVME_SUBSYS)
 
 struct NvmeSubsystem {
 DeviceState parent_

[PATCH for-6.2 06/12] [automated] Split QOM "typedef struct T { ... } T" declarations

[Eduardo Habkost]

Automatically split struct definition and typedef declaration in
separate declarations, using a codeconverter rule.  The rule will
only touch declarations of structs/typedefs actually used by QOM
types.

This will make automated changes to use OBJECT_DECLARE* macros
easier to implement, because automated removal of typedef lines
will be easier and safer.

Generated using:

  $ ./scripts/codeconverter/converter.py -i \
--pattern=QOMStructTypedefSplit $(git grep -l '' -- '*.[ch]')

Signed-off-by: Eduardo Habkost 
---
Cc: "Marc-André Lureau" 
Cc: Paolo Bonzini 
Cc: Patrick Venture 
Cc: Thomas Huth 
Cc: Keith Busch 
Cc: Klaus Jensen 
Cc: "Michael S. Tsirkin" 
Cc: Cornelia Huck 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Gerd Hoffmann 
Cc: Havard Skinnemoen 
Cc: Tyrone Ting 
Cc: Alistair Francis 
Cc: Bin Meng 
Cc: Palmer Dabbelt 
Cc: "Edgar E. Iglesias" 
Cc: Peter Maydell 
Cc: Andrew Baumann 
Cc: "Philippe Mathieu-Daudé" 
Cc: "Cédric Le Goater" 
Cc: David Gibson 
Cc: Greg Kurz 
Cc: Laurent Vivier 
Cc: qemu-de...@nongnu.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
Cc: qemu-...@nongnu.org
Cc: qemu-ri...@nongnu.org
Cc: qemu-...@nongnu.org
---
 hw/nvme/nvme.h  | 10 ++
 hw/usb/hcd-uhci.h   |  5 +++--
 hw/usb/u2f.h|  5 +++--
 include/hw/adc/npcm7xx_adc.h|  5 +++--
 include/hw/arm/npcm7xx.h| 10 ++
 include/hw/core/accel-cpu.h |  5 +++--
 include/hw/dma/sifive_pdma.h|  5 +++--
 include/hw/dma/xlnx_csu_dma.h   |  5 +++--
 include/hw/gpio/npcm7xx_gpio.h  |  5 +++--
 include/hw/i2c/npcm7xx_smbus.h  |  5 +++--
 include/hw/mem/npcm7xx_mc.h |  5 +++--
 include/hw/misc/bcm2835_cprman.h| 20 
 include/hw/misc/mchp_pfsoc_dmc.h| 10 ++
 include/hw/misc/mchp_pfsoc_ioscb.h  |  5 +++--
 include/hw/misc/mchp_pfsoc_sysreg.h |  5 +++--
 include/hw/misc/npcm7xx_clk.h   | 15 +--
 include/hw/misc/npcm7xx_gcr.h   |  5 +++--
 include/hw/misc/npcm7xx_mft.h   |  5 +++--
 include/hw/misc/npcm7xx_rng.h   |  5 +++--
 include/hw/nvram/npcm7xx_otp.h  |  5 +++--
 include/hw/riscv/microchip_pfsoc.h  | 10 ++
 include/hw/riscv/sifive_e.h |  5 +++--
 include/hw/sd/cadence_sdhci.h   |  5 +++--
 include/qemu/accel.h| 10 ++
 chardev/char-parallel.c | 10 ++
 hw/i2c/i2c_mux_pca954x.c|  5 +++--
 hw/m68k/mcf5206.c   |  5 +++--
 hw/misc/sbsa_ec.c   |  5 +++--
 hw/s390x/vhost-user-fs-ccw.c|  5 +++--
 tests/qtest/pnv-xscom-test.c|  5 +++--
 30 files changed, 123 insertions(+), 82 deletions(-)

diff --git a/hw/nvme/nvme.h b/hw/nvme/nvme.h
index 5ddcd783055..0840f585d6e 100644
--- a/hw/nvme/nvme.h
+++ b/hw/nvme/nvme.h
@@ -118,7 +118,7 @@ typedef struct NvmeNamespaceParams {
 uint32_t zd_extension_size;
 } NvmeNamespaceParams;
 
-typedef struct NvmeNamespace {
+struct NvmeNamespace {
 DeviceState  parent_obj;
 BlockConfblkconf;
 int32_t  bootindex;
@@ -153,7 +153,8 @@ typedef struct NvmeNamespace {
 struct {
 uint32_t err_rec;
 } features;
-} NvmeNamespace;
+};
+typedef struct NvmeNamespace NvmeNamespace;
 
 static inline uint32_t nvme_nsid(NvmeNamespace *ns)
 {
@@ -395,7 +396,7 @@ typedef struct NvmeParams {
 bool legacy_cmb;
 } NvmeParams;
 
-typedef struct NvmeCtrl {
+struct NvmeCtrl {
 PCIDeviceparent_obj;
 MemoryRegion bar0;
 MemoryRegion iomem;
@@ -462,7 +463,8 @@ typedef struct NvmeCtrl {
 };
 uint32_tasync_config;
 } features;
-} NvmeCtrl;
+};
+typedef struct NvmeCtrl NvmeCtrl;
 
 static inline NvmeNamespace *nvme_ns(NvmeCtrl *n, uint32_t nsid)
 {
diff --git a/hw/usb/hcd-uhci.h b/hw/usb/hcd-uhci.h
index 57d0d574644..2597a6d2eb6 100644
--- a/hw/usb/hcd-uhci.h
+++ b/hw/usb/hcd-uhci.h
@@ -43,7 +43,7 @@ typedef struct UHCIPort {
 uint16_t ctrl;
 } UHCIPort;
 
-typedef struct UHCIState {
+struct UHCIState {
 PCIDevice dev;
 MemoryRegion io_bar;
 USBBus bus; /* Note unused when we're a companion controller */
@@ -73,7 +73,8 @@ typedef struct UHCIState {
 char *masterbus;
 uint32_t firstport;
 uint32_t maxframes;
-} UHCIState;
+};
+typedef struct UHCIState UHCIState;
 
 #define TYPE_UHCI "pci-uhci-usb"
 DECLARE_INSTANCE_CHECKER(UHCIState, UHCI, TYPE_UHCI)
diff --git a/hw/usb/u2f.h b/hw/usb/u2f.h
index 705d5c43ce6..7191a23ee1a 100644
--- a/hw/usb/u2f.h
+++ b/hw/usb/u2f.h
@@ -62,7 +62,7 @@ struct U2FKeyClass {
 /*
  * State of the U2F key base device (i.e. hw/u2f.c)
  */
-typedef struct U2FKeyState {
+struct U2FKeyState {
 USBDevice dev;
 USBEndpoint *ep;
 uint8_t idle;
@@ -72,7 +72,8 @@ typedef struct U2FKeyState {
 uint8_t pending_in_start;
 uint8_t pending_in_end;
 uint8_t pen

[PATCH for-6.2 05/12] [automated] Move QOM typedefs and add missing includes

[Eduardo Habkost]

Some typedefs and macros are defined after the type check macros.
This makes it difficult to automatically replace their
definitions with OBJECT_DECLARE_TYPE.

Patch generated using:

 $ ./scripts/codeconverter/converter.py -i --pattern=MoveSymbols \
$(git grep -l '' -- '*.[ch]')

which will:
- split "typdef struct { ... } TypedefName" declarations
- move the typedefs and #defines above the type check macros
- add missing #include "qom/object.h" lines if necessary

Signed-off-by: Eduardo Habkost 
---
Cc: Richard Henderson 
Cc: Paolo Bonzini 
Cc: "Marc-André Lureau" 
Cc: Peter Maydell 
Cc: Andrew Baumann 
Cc: "Philippe Mathieu-Daudé" 
Cc: Thomas Huth 
Cc: "Michael S. Tsirkin" 
Cc: Igor Mammedov 
Cc: Alexander Bulekov 
Cc: Bandan Das 
Cc: Stefan Hajnoczi 
Cc: Huacai Chen 
Cc: Jiaxun Yang 
Cc: Aurelien Jarno 
Cc: Aleksandar Rikalo 
Cc: Havard Skinnemoen 
Cc: Tyrone Ting 
Cc: Pavel Pisa 
Cc: Vikram Garhwal 
Cc: Jason Wang 
Cc: Keith Busch 
Cc: Klaus Jensen 
Cc: Cornelia Huck 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: David Hildenbrand 
Cc: Gerd Hoffmann 
Cc: Vijai Kumar K 
Cc: Alistair Francis 
Cc: Bin Meng 
Cc: Palmer Dabbelt 
Cc: "Edgar E. Iglesias" 
Cc: Laurent Vivier 
Cc: Corey Minyard 
Cc: "Cédric Le Goater" 
Cc: Andrew Jeffery 
Cc: Joel Stanley 
Cc: Francisco Iglesias 
Cc: David Gibson 
Cc: Greg Kurz 
Cc: Alexey Kardashevskiy 
Cc: "Hervé Poussineau" 
Cc: Bastian Koppelmann 
Cc: Juan Quintela 
Cc: "Dr. David Alan Gilbert" 
Cc: "Daniel P. Berrangé" 
Cc: Eduardo Habkost 
Cc: Stefan Berger 
Cc: Taylor Simpson 
Cc: qemu-de...@nongnu.org
Cc: qemu-...@nongnu.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
Cc: qemu-ri...@nongnu.org
Cc: qemu-...@nongnu.org
---
 hw/nvme/nvme.h  |  6 --
 hw/usb/hcd-uhci.h   |  1 +
 hw/usb/hcd-xhci-pci.h   |  6 --
 hw/usb/hcd-xhci-sysbus.h|  6 --
 hw/usb/u2f.h|  6 --
 include/hw/acpi/acpi_dev_interface.h|  2 +-
 include/hw/adc/npcm7xx_adc.h|  1 +
 include/hw/arm/linux-boot-if.h  |  2 +-
 include/hw/arm/npcm7xx.h| 11 +++
 include/hw/char/shakti_uart.h   |  6 --
 include/hw/core/accel-cpu.h |  1 +
 include/hw/dma/sifive_pdma.h|  1 +
 include/hw/dma/xlnx_csu_dma.h   |  1 +
 include/hw/fw-path-provider.h   |  2 +-
 include/hw/gpio/npcm7xx_gpio.h  |  1 +
 include/hw/hotplug.h|  2 +-
 include/hw/i2c/npcm7xx_smbus.h  |  1 +
 include/hw/intc/intc.h  |  2 +-
 include/hw/intc/m68k_irqc.h |  6 --
 include/hw/intc/sifive_clint.h  |  6 --
 include/hw/ipmi/ipmi.h  |  2 +-
 include/hw/mem/memory-device.h  |  2 +-
 include/hw/mem/npcm7xx_mc.h |  1 +
 include/hw/misc/aspeed_lpc.h|  6 --
 include/hw/misc/bcm2835_cprman.h|  1 +
 include/hw/misc/bcm2835_cprman_internals.h  |  1 +
 include/hw/misc/mchp_pfsoc_dmc.h|  1 +
 include/hw/misc/mchp_pfsoc_ioscb.h  |  1 +
 include/hw/misc/mchp_pfsoc_sysreg.h |  1 +
 include/hw/misc/npcm7xx_clk.h   |  1 +
 include/hw/misc/npcm7xx_gcr.h   |  1 +
 include/hw/misc/npcm7xx_pwm.h   |  1 +
 include/hw/misc/npcm7xx_rng.h   |  1 +
 include/hw/misc/xlnx-versal-xramc.h |  6 --
 include/hw/net/npcm7xx_emc.h|  1 +
 include/hw/net/xlnx-zynqmp-can.h|  6 --
 include/hw/nmi.h|  2 +-
 include/hw/nvram/npcm7xx_otp.h  |  1 +
 include/hw/ppc/spapr_drc.h  | 15 +--
 include/hw/ppc/spapr_xive.h | 11 +++
 include/hw/ppc/vof.h|  1 +
 include/hw/rdma/rdma.h  |  2 +-
 include/hw/riscv/microchip_pfsoc.h  |  1 +
 include/hw/riscv/shakti_c.h | 11 +++
 include/hw/riscv/sifive_e.h |  6 --
 include/hw/riscv/sifive_u.h | 11 +++
 include/hw/rtc/m48t59.h |  2 +-
 include/hw/sd/cadence_sdhci.h   |  1 +
 include/hw/ssi/npcm7xx_fiu.h|  1 +
 include/hw/ssi/sifive_spi.h |  6 --
 include/hw/stream.h |  2 +-
 include/hw/timer/npcm7xx_timer.h|  1 +
 include/hw/tricore/tricore_testdevice.h |  6 --
 include/hw/usb/hcd-dwc3.h   |  6 --
 include/hw/usb/msd.h|  1 +
 include/hw/usb/xlnx-usb-subsystem.h |  6 --
 include/hw/usb/xlnx-versal-usb2-ctrl-regs.h |  6 --
 include/hw/vmstate-if.h |  2 +-
 include/hw/watchdog/sbsa_gwdt.

Re: [PATCH v4 01/34] modules: add modinfo macros

[Eduardo Habkost]

On Thu, Jun 24, 2021 at 12:38:03PM +0200, Gerd Hoffmann wrote:
> Add macros for module info annotations.
> 
> Instead of having that module meta-data stored in lists in util/module.c
> place directly in the module source code.
> 
[...]
> +/* module implements QOM type  */
> +#define module_obj(name) modinfo(obj, name)

Can we make OBJECT_DEFINE_TYPE*() use this macro automatically?

-- 
Eduardo

Re: [PATCH] virtio-blk: drop deprecated scsi=on|off property

[Eduardo Habkost]

On Tue, May 04, 2021 at 03:32:55PM +0100, Stefan Hajnoczi wrote:
> On Thu, Apr 29, 2021 at 02:03:52PM -0400, Eduardo Habkost wrote:
> > On Thu, Apr 29, 2021 at 04:52:21PM +0100, Stefan Hajnoczi wrote:
> > > Live migrating old guests from an old QEMU with the SCSI feature bit
> > > enabled will fail with "Features 0x... unsupported. Allowed features:
> > > 0x...". We've followed the QEMU deprecation policy so users have been
> > > warned...
> > > 
> > 
> > Were they really warned, though?  People running
> > "-machine pc-i440fx-2.4" might be completely unaware that it was
> > silently enabling a deprecated feature.
> > 
> > Can we have this documented in a more explicit way?  Maybe just a
> > comment at hw_compat_2_4 would be enough, to warn people doing
> > backports and rebases downstream.
> > 
> > Can we make QEMU refuse to start if using pc-2.4 + virtio-blk
> > together, just to be sure?
> 
> On second thought, do we really want to break pc-2.4 user's QEMU
> command-lines if they have a virtio-blk device?

It depends which command line you are talking about.

I believe we _must_ break the following:
"-machine pc-i440fx-2.4 -device virtio-blk", and
"-machine pc-i440fx-2.4 -device virtio-blk,scsi=on".
Your patch breaks only the latter.

Your patch also breaks the following:
"-machine pc-i440fx-2.4 -device virtio-blk,scsi=off",
which I don't think we should break.

> 
> BTW Peter mentioned libvirt avoids the unnecessary scsi=off:
> https://gitlab.com/libvirt/libvirt/-/commit/ec69f0190be731d12faeac08dbf63325836509a9
> 
> Stefan



-- 
Eduardo

Re: [PATCH] virtio-blk: drop deprecated scsi=on|off property

[Eduardo Habkost]

On Thu, Apr 29, 2021 at 04:52:21PM +0100, Stefan Hajnoczi wrote:
> The scsi=on|off property was deprecated in QEMU 5.0 and can be removed
> completely at this point.
> 
> Drop the scsi=on|off option. It was only available on Legacy virtio-blk
> devices. Linux v5.6 already dropped support for it.
> 
> Remove the hw_compat_2_4[] property assignment since scsi=on|off no
> longer exists. Old guests with Legacy virtio-blk devices no longer see
> the SCSI host features bit.
> 

This means pc-2.4 will now break guest ABI if using virtio-blk
devices, correct?

This looks like a sign we should have deprecated pc-2.4 a long
time ago.

> Live migrating old guests from an old QEMU with the SCSI feature bit
> enabled will fail with "Features 0x... unsupported. Allowed features:
> 0x...". We've followed the QEMU deprecation policy so users have been
> warned...
> 

Were they really warned, though?  People running
"-machine pc-i440fx-2.4" might be completely unaware that it was
silently enabling a deprecated feature.

Can we have this documented in a more explicit way?  Maybe just a
comment at hw_compat_2_4 would be enough, to warn people doing
backports and rebases downstream.

Can we make QEMU refuse to start if using pc-2.4 + virtio-blk
together, just to be sure?

An alternative would be keeping the property (and the
hw_compat_2_4 entry) just to keep pc-2.4 working (until pc-2.4 is
deprecated and removed), but refusing to realize the device if
the feature is enabled.


> I have tested that libvirt still works when the property is absent. It
> no longer adds scsi=on|off to the command-line.
> 
> Cc: Markus Armbruster 
> Cc: Christoph Hellwig 
> Cc: Peter Krempa 
> Cc: Dr. David Alan Gilbert 
> Signed-off-by: Stefan Hajnoczi 
> ---
[...]
> diff --git a/hw/core/machine.c b/hw/core/machine.c
> index 40def78183..286f18ec6d 100644
> --- a/hw/core/machine.c
> +++ b/hw/core/machine.c
> @@ -194,8 +194,6 @@ GlobalProperty hw_compat_2_5[] = {
>  const size_t hw_compat_2_5_len = G_N_ELEMENTS(hw_compat_2_5);
>  
>  GlobalProperty hw_compat_2_4[] = {
> -/* Optional because the 'scsi' property is Linux-only */
> -{ "virtio-blk-device", "scsi", "true", .optional = true },
>  { "e1000", "extra_mac_registers", "off" },
>  { "virtio-pci", "x-disable-pcie", "on" },
>  { "virtio-pci", "migrate-extra", "off" },


-- 
Eduardo

Re: [PATCH 1/6] qdev: introduce qapi/hmp command for kick/call event

[Eduardo Habkost]

On Thu, Mar 25, 2021 at 10:44:28PM -0700, Dongli Zhang wrote:
> The virtio device/driver (e.g., vhost-scsi or vhost-net) may hang due to
> the loss of doorbell kick, e.g.,
> 
> https://lists.gnu.org/archive/html/qemu-devel/2018-12/msg01711.html
> 
> ... or due to the loss of IRQ, e.g., as fixed by linux kernel commit
> fe200ae48ef5 ("genirq: Mark polled irqs and defer the real handler").
> 
> This patch introduces a new debug interface 'DeviceEvent' to DeviceClass
> to help narrow down if the issue is due to loss of irq/kick. So far the new
> interface handles only two events: 'call' and 'kick'. Any device (e.g.,
> virtio/vhost or VFIO) may implement the interface (e.g., via eventfd, MSI-X
> or legacy IRQ).
> 
> The 'call' is to inject irq on purpose by admin for a specific device (e.g.,
> vhost-scsi) from QEMU/host to VM, while the 'kick' is to kick the doorbell
> on purpose by admin at QEMU/host side for a specific device.
> 
> Signed-off-by: Dongli Zhang 
[...]
> diff --git a/include/monitor/hmp.h b/include/monitor/hmp.h
> index 605d57287a..c7795d4ba5 100644
> --- a/include/monitor/hmp.h
> +++ b/include/monitor/hmp.h
> @@ -129,5 +129,6 @@ void hmp_info_replay(Monitor *mon, const QDict *qdict);
>  void hmp_replay_break(Monitor *mon, const QDict *qdict);
>  void hmp_replay_delete_break(Monitor *mon, const QDict *qdict);
>  void hmp_replay_seek(Monitor *mon, const QDict *qdict);
> +void hmp_x_debug_device_event(Monitor *mon, const QDict *qdict);
>  
>  #endif
> diff --git a/qapi/qdev.json b/qapi/qdev.json
> index b83178220b..711c4a297a 100644
> --- a/qapi/qdev.json
> +++ b/qapi/qdev.json
> @@ -124,3 +124,33 @@
>  ##
>  { 'event': 'DEVICE_DELETED',
>'data': { '*device': 'str', 'path': 'str' } }
> +
> +##
> +# @x-debug-device-event:
> +#
> +# Generate device event for a specific device queue
> +#
> +# @dev: device path
> +#
> +# @event: event (e.g., kick or call) to trigger

Any specific reason to not use an enum here?

In addition to making the QAPI schema and documentation more
descriptive, it would save you the work of manually defining the
DEVICE_EVENT_* constants and implementing get_device_event().


> +#
> +# @queue: queue id
> +#
> +# Returns: Nothing on success
> +#
> +# Since: 6.1
> +#
> +# Notes: This is used to debug VM driver hang issue. The 'kick' event is to
> +#send notification to QEMU/vhost while the 'call' event is to
> +#interrupt VM on purpose.
> +#
> +# Example:
> +#
> +# -> { "execute": "x-debug-device_event",
> +#  "arguments": { "dev": "/machine/peripheral/vscsi0", "event": "kick",
> +# "queue": 1 } }
> +# <- { "return": {} }
> +#
> +##
> +{ 'command': 'x-debug-device-event',
> +  'data': {'dev': 'str', 'event': 'str', 'queue': 'int'} }
[...]

-- 
Eduardo

Re: [PATCH RFC 0/1] QOM type names and QAPI

[Eduardo Habkost]

On Fri, Jan 29, 2021 at 02:25:56PM +0100, Paolo Bonzini wrote:
> On 29/01/21 13:17, Daniel P. Berrangé wrote:
> > > On this one, my vote would be "no". "Versioned machine names
> > > include the QEMU version number" is pretty well entrenched,
> > > and requiring users to remember that when they want version 4.2
> > > they need to remember some other way of writing it than "4.2"
> > > seems rather unfriendly. And 550 uses of '.' is a lot.
> > We can't make  keyval_parse() accept "/" instead of ".", but can
> > we make it accept "/" in addition to ".", and then encourage "/"  ?
> > 
> > People simply wouldnt be able to use "." as keyval separator if
> > they're using typenames containing "." (or would have to escape
> > the typename.
> 
> '.' is much more common than '/', and is shared by about all programming
> languages that have JSON-ish data structures natively.  So using '/' seems
> decidedly worse to me.

Worse than what, exactly?

Accepting "/" when "." is ambiguous seems decidedly better than
the following alternatives:
- renaming machine types to names like "q35-5-0"; or
- having to escape "." in the command line.

-- 
Eduardo

Re: [PATCH v4 23/32] qdev: Move dev->realized check to qdev_property_set()

[Eduardo Habkost]

On Mon, Dec 14, 2020 at 03:55:30PM +0100, Igor Mammedov wrote:
> On Fri, 11 Dec 2020 17:05:20 -0500
> Eduardo Habkost  wrote:
> 
> > Every single qdev property setter function manually checks
> > dev->realized.  We can just check dev->realized inside
> > qdev_property_set() instead.
> > 
> > The check is being added as a separate function
> > (qdev_prop_allow_set()) because it will become a callback later.
> 
> is callback added within this series?
> and I'd add here what's the purpose of it.

It will be added in part 2 of the series.  See v3:
https://lore.kernel.org/qemu-devel/20201112214350.872250-35-ehabk...@redhat.com/

I don't know what else I could say about its purpose, in addition
to what I wrote above, and the comment below[1].

If you are just curious about the callback and confused because
it is not anywhere in this series, I can just remove the
paragraph above from the commit message.  Would that be enough?

> 
[...]
> > +/* returns: true if property is allowed to be set, false otherwise */

[1] ^^^

> > +static bool qdev_prop_allow_set(Object *obj, const char *name,
> > +Error **errp)
> > +{
> > +DeviceState *dev = DEVICE(obj);
> > +
> > +if (dev->realized) {
> > +qdev_prop_set_after_realize(dev, name, errp);
> > +return false;
> > +}
> > +return true;
> > +}
> > +

-- 
Eduardo

[PATCH v4 30/32] qdev: Rename qdev_get_prop_ptr() to object_field_prop_ptr()

[Eduardo Habkost]

The function will be moved to common QOM code, as it is not
specific to TYPE_DEVICE anymore.

Reviewed-by: Stefan Berger 
Signed-off-by: Eduardo Habkost 
---
Changes v1 -> v2:
* Rename to object_field_prop_ptr() instead of object_static_prop_ptr()
---
Cc: Stefan Berger 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: Eduardo Habkost 
Cc: Cornelia Huck 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Thomas Huth 
Cc: Matthew Rosato 
Cc: Alex Williamson 
Cc: qemu-de...@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
---
 include/hw/qdev-properties.h |  2 +-
 backends/tpm/tpm_util.c  |  6 ++--
 hw/block/xen-block.c |  4 +--
 hw/core/qdev-properties-system.c | 50 +-
 hw/core/qdev-properties.c| 60 
 hw/s390x/css.c   |  4 +--
 hw/s390x/s390-pci-bus.c  |  4 +--
 hw/vfio/pci-quirks.c |  4 +--
 8 files changed, 67 insertions(+), 67 deletions(-)

diff --git a/include/hw/qdev-properties.h b/include/hw/qdev-properties.h
index 90222822f1..97bb9494ae 100644
--- a/include/hw/qdev-properties.h
+++ b/include/hw/qdev-properties.h
@@ -193,7 +193,7 @@ void qdev_prop_set_macaddr(DeviceState *dev, const char 
*name,
const uint8_t *value);
 void qdev_prop_set_enum(DeviceState *dev, const char *name, int value);
 
-void *qdev_get_prop_ptr(Object *obj, Property *prop);
+void *object_field_prop_ptr(Object *obj, Property *prop);
 
 void qdev_prop_register_global(GlobalProperty *prop);
 const GlobalProperty *qdev_find_global_prop(Object *obj,
diff --git a/backends/tpm/tpm_util.c b/backends/tpm/tpm_util.c
index 39b45fa46d..a6e6d3e72f 100644
--- a/backends/tpm/tpm_util.c
+++ b/backends/tpm/tpm_util.c
@@ -35,7 +35,7 @@
 static void get_tpm(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-TPMBackend **be = qdev_get_prop_ptr(obj, opaque);
+TPMBackend **be = object_field_prop_ptr(obj, opaque);
 char *p;
 
 p = g_strdup(*be ? (*be)->id : "");
@@ -47,7 +47,7 @@ static void set_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 Error **errp)
 {
 Property *prop = opaque;
-TPMBackend *s, **be = qdev_get_prop_ptr(obj, prop);
+TPMBackend *s, **be = object_field_prop_ptr(obj, prop);
 char *str;
 
 if (!visit_type_str(v, name, , errp)) {
@@ -67,7 +67,7 @@ static void set_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 static void release_tpm(Object *obj, const char *name, void *opaque)
 {
 Property *prop = opaque;
-TPMBackend **be = qdev_get_prop_ptr(obj, prop);
+TPMBackend **be = object_field_prop_ptr(obj, prop);
 
 if (*be) {
 tpm_backend_reset(*be);
diff --git a/hw/block/xen-block.c b/hw/block/xen-block.c
index bd1aef63a7..718d886e5c 100644
--- a/hw/block/xen-block.c
+++ b/hw/block/xen-block.c
@@ -336,7 +336,7 @@ static void xen_block_get_vdev(Object *obj, Visitor *v, 
const char *name,
void *opaque, Error **errp)
 {
 Property *prop = opaque;
-XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
+XenBlockVdev *vdev = object_field_prop_ptr(obj, prop);
 char *str;
 
 switch (vdev->type) {
@@ -396,7 +396,7 @@ static void xen_block_set_vdev(Object *obj, Visitor *v, 
const char *name,
void *opaque, Error **errp)
 {
 Property *prop = opaque;
-XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
+XenBlockVdev *vdev = object_field_prop_ptr(obj, prop);
 char *str, *p;
 const char *end;
 
diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index 590c5f3d97..e6d378a34e 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -62,7 +62,7 @@ static void get_drive(Object *obj, Visitor *v, const char 
*name, void *opaque,
   Error **errp)
 {
 Property *prop = opaque;
-void **ptr = qdev_get_prop_ptr(obj, prop);
+void **ptr = object_field_prop_ptr(obj, prop);
 const char *value;
 char *p;
 
@@ -88,7 +88,7 @@ static void set_drive_helper(Object *obj, Visitor *v, const 
char *name,
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-void **ptr = qdev_get_prop_ptr(obj, prop);
+void **ptr = object_field_prop_ptr(obj, prop);
 char *str;
 BlockBackend *blk;
 bool blk_created = false;
@@ -181,7 +181,7 @@ static void release_drive(Object *obj, const char *name, 
void *opaque)
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-BlockBackend **ptr = qdev_get_prop_ptr(obj, prop);
+BlockBackend **ptr = object_field_prop_ptr(obj, prop);
 
 if (*ptr) {
 AioContext *ctx = blk_get_aio_co

[PATCH v4 23/32] qdev: Move dev->realized check to qdev_property_set()

[Eduardo Habkost]

Every single qdev property setter function manually checks
dev->realized.  We can just check dev->realized inside
qdev_property_set() instead.

The check is being added as a separate function
(qdev_prop_allow_set()) because it will become a callback later.

Reviewed-by: Stefan Berger 
Signed-off-by: Eduardo Habkost 
---
Changes v1 -> v2:
* Removed unused variable at xen_block_set_vdev()
* Redone patch after changes in the previous patches in the
  series
---
Cc: Stefan Berger 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: Eduardo Habkost 
Cc: Cornelia Huck 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Thomas Huth 
Cc: Matthew Rosato 
Cc: Alex Williamson 
Cc: Mark Cave-Ayland 
Cc: Artyom Tarasenko 
Cc: qemu-de...@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
---
 backends/tpm/tpm_util.c  |   6 --
 hw/block/xen-block.c |   6 --
 hw/core/qdev-properties-system.c |  70 --
 hw/core/qdev-properties.c| 100 ++-
 hw/s390x/css.c   |   6 --
 hw/s390x/s390-pci-bus.c  |   6 --
 hw/vfio/pci-quirks.c |   6 --
 target/sparc/cpu.c   |   6 --
 8 files changed, 18 insertions(+), 188 deletions(-)

diff --git a/backends/tpm/tpm_util.c b/backends/tpm/tpm_util.c
index a5d997e7dc..39b45fa46d 100644
--- a/backends/tpm/tpm_util.c
+++ b/backends/tpm/tpm_util.c
@@ -46,16 +46,10 @@ static void get_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 static void set_tpm(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 TPMBackend *s, **be = qdev_get_prop_ptr(obj, prop);
 char *str;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
diff --git a/hw/block/xen-block.c b/hw/block/xen-block.c
index 905e4acd97..bd1aef63a7 100644
--- a/hw/block/xen-block.c
+++ b/hw/block/xen-block.c
@@ -395,17 +395,11 @@ static int vbd_name_to_disk(const char *name, const char 
**endp,
 static void xen_block_set_vdev(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
 char *str, *p;
 const char *end;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index 42529c3b65..f31aea3de1 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -94,11 +94,6 @@ static void set_drive_helper(Object *obj, Visitor *v, const 
char *name,
 bool blk_created = false;
 int ret;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
@@ -230,17 +225,11 @@ static void get_chr(Object *obj, Visitor *v, const char 
*name, void *opaque,
 static void set_chr(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 CharBackend *be = qdev_get_prop_ptr(obj, prop);
 Chardev *s;
 char *str;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
@@ -311,18 +300,12 @@ static void get_mac(Object *obj, Visitor *v, const char 
*name, void *opaque,
 static void set_mac(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 MACAddr *mac = qdev_get_prop_ptr(obj, prop);
 int i, pos;
 char *str;
 const char *p;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
@@ -390,7 +373,6 @@ static void get_netdev(Object *obj, Visitor *v, const char 
*name,
 static void set_netdev(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 NICPeers *peers_ptr = qdev_get_prop_ptr(obj, prop);
 NetClientState **ncs = peers_ptr->ncs;
@@ -398,11 +380,6 @@ static void set_netdev(Object *obj, Visitor *v, const char 
*name,
 int queues, err = 0, i = 0;
   

[PATCH v4 09/32] qdev: Make qdev_get_prop_ptr() get Object* arg

[Eduardo Habkost]

Make the code more generic and not specific to TYPE_DEVICE.

Reviewed-by: Marc-André Lureau 
Reviewed-by: Cornelia Huck  #s390 parts
Signed-off-by: Eduardo Habkost 
---
Changes v1 -> v2:
- Fix build error with CONFIG_XEN
  I took the liberty of keeping the Reviewed-by line from
  Marc-André as the build fix is a trivial one line change
---
Cc: Stefan Berger 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: Eduardo Habkost 
Cc: Cornelia Huck 
Cc: Thomas Huth 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Matthew Rosato 
Cc: Alex Williamson 
Cc: qemu-de...@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
---
 include/hw/qdev-properties.h |  2 +-
 backends/tpm/tpm_util.c  |  8 ++--
 hw/block/xen-block.c |  5 +-
 hw/core/qdev-properties-system.c | 57 +-
 hw/core/qdev-properties.c| 82 +---
 hw/s390x/css.c   |  5 +-
 hw/s390x/s390-pci-bus.c  |  4 +-
 hw/vfio/pci-quirks.c |  5 +-
 8 files changed, 68 insertions(+), 100 deletions(-)

diff --git a/include/hw/qdev-properties.h b/include/hw/qdev-properties.h
index 0ea822e6a7..0b92cfc761 100644
--- a/include/hw/qdev-properties.h
+++ b/include/hw/qdev-properties.h
@@ -302,7 +302,7 @@ void qdev_prop_set_macaddr(DeviceState *dev, const char 
*name,
const uint8_t *value);
 void qdev_prop_set_enum(DeviceState *dev, const char *name, int value);
 
-void *qdev_get_prop_ptr(DeviceState *dev, Property *prop);
+void *qdev_get_prop_ptr(Object *obj, Property *prop);
 
 void qdev_prop_register_global(GlobalProperty *prop);
 const GlobalProperty *qdev_find_global_prop(DeviceState *dev,
diff --git a/backends/tpm/tpm_util.c b/backends/tpm/tpm_util.c
index e6aeb63587..3973105658 100644
--- a/backends/tpm/tpm_util.c
+++ b/backends/tpm/tpm_util.c
@@ -35,8 +35,7 @@
 static void get_tpm(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
-TPMBackend **be = qdev_get_prop_ptr(dev, opaque);
+TPMBackend **be = qdev_get_prop_ptr(obj, opaque);
 char *p;
 
 p = g_strdup(*be ? (*be)->id : "");
@@ -49,7 +48,7 @@ static void set_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-TPMBackend *s, **be = qdev_get_prop_ptr(dev, prop);
+TPMBackend *s, **be = qdev_get_prop_ptr(obj, prop);
 char *str;
 
 if (dev->realized) {
@@ -73,9 +72,8 @@ static void set_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 
 static void release_tpm(Object *obj, const char *name, void *opaque)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-TPMBackend **be = qdev_get_prop_ptr(dev, prop);
+TPMBackend **be = qdev_get_prop_ptr(obj, prop);
 
 if (*be) {
 tpm_backend_reset(*be);
diff --git a/hw/block/xen-block.c b/hw/block/xen-block.c
index 8a7a3f5452..905e4acd97 100644
--- a/hw/block/xen-block.c
+++ b/hw/block/xen-block.c
@@ -335,9 +335,8 @@ static char *disk_to_vbd_name(unsigned int disk)
 static void xen_block_get_vdev(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-XenBlockVdev *vdev = qdev_get_prop_ptr(dev, prop);
+XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
 char *str;
 
 switch (vdev->type) {
@@ -398,7 +397,7 @@ static void xen_block_set_vdev(Object *obj, Visitor *v, 
const char *name,
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-XenBlockVdev *vdev = qdev_get_prop_ptr(dev, prop);
+XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
 char *str, *p;
 const char *end;
 
diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index 77b31eb9dc..9ac9b95852 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -59,9 +59,8 @@ static bool check_prop_still_unset(DeviceState *dev, const 
char *name,
 static void get_drive(Object *obj, Visitor *v, const char *name, void *opaque,
   Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-void **ptr = qdev_get_prop_ptr(dev, prop);
+void **ptr = qdev_get_prop_ptr(obj, prop);
 const char *value;
 char *p;
 
@@ -87,7 +86,7 @@ static void set_drive_helper(Object *obj, Visitor *v, const 
char *name,
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-void **ptr = qdev_get_prop_ptr(dev, prop);
+void **ptr = qdev_get_prop_ptr(obj, prop);
 char *str;
 BlockBackend *blk;
 bool blk_created = false;
@@ -185,7 +184,7 @@ static void release_dri

Re: [PATCH 00/18] qapi/qom: QAPIfy object-add

[Eduardo Habkost]

On Thu, Dec 03, 2020 at 07:10:37PM +0100, Paolo Bonzini wrote:
> On 03/12/20 18:52, Eduardo Habkost wrote:
> > On Thu, Dec 03, 2020 at 05:50:46PM +0100, Paolo Bonzini wrote:
> > > On 03/12/20 16:15, Kevin Wolf wrote:
> > > > I don't think this is an intermediate state like Eduardo wants to have.
> > > > Creating the object, then setting properties, then realize [1] will fail
> > > > after your change. But keeping it working was the whole point of the
> > > > exercise.
> > > 
> > > With the sample code, you must remove object_class_property_set calls at 
> > > the
> > 
> > Do you mean object_property_set()?
> 
> Yes.
> 
> > > same time as you remove the setters.  Usually that'd be when you convert 
> > > to
> > > QAPI and oc->configure, but it doesn't have to be that way if there are 
> > > good
> > > reasons not to do so.
> > 
> > Having two (or more) similar but incompatible APIs to do exactly
> > the same thing is a mistake we did before, and I wouldn't like us
> > to repeat it.
> > 
> > If we can keep qdev_new() + object_property_set() + realize
> > working after the device is converted, we should.  I believe we
> > can.
> 
> You can.  If you want to do that, you have to give up on removing the
> setters; but that's not so beneficial for devices because they already use
> static properties anyway.  They have much less boilerplate than -object
> objects.

Understood.

We can also get rid of most setters in -object backends using
field properties.  Maybe not a necessary step, but a useful
intermediate step in case the new API takes time to be ready.

> 
> > If we can make object_new_configure() work with all (or most)
> > device types before we manually convert them to the new system,
> > we should.  I believe we can.
> 
> Yup, object_new_configure() is the low-level visitor-based API and therefore
> it supports both properties and oc->configure.

Perfect.  That part was not clear yet to me (I just skimmed to
the example code you posted on the wiki).

> 
> > We may be able avoid these questions with -object because
> > converting all backends at the same time is doable.  With
> > devices, API usability and maintainability during the transition
> > period (which could be very long) needs to be taken into account.
> 
> I think we're in violent agreement. :)
> 
> Paolo
> 

-- 
Eduardo

Re: [PATCH 00/18] qapi/qom: QAPIfy object-add

[Eduardo Habkost]

On Thu, Dec 03, 2020 at 05:50:46PM +0100, Paolo Bonzini wrote:
> On 03/12/20 16:15, Kevin Wolf wrote:
> > I don't think this is an intermediate state like Eduardo wants to have.
> > Creating the object, then setting properties, then realize [1] will fail
> > after your change. But keeping it working was the whole point of the
> > exercise.
> 
> With the sample code, you must remove object_class_property_set calls at the

Do you mean object_property_set()?

> same time as you remove the setters.  Usually that'd be when you convert to
> QAPI and oc->configure, but it doesn't have to be that way if there are good
> reasons not to do so.

Having two (or more) similar but incompatible APIs to do exactly
the same thing is a mistake we did before, and I wouldn't like us
to repeat it.

If we can keep qdev_new() + object_property_set() + realize
working after the device is converted, we should.  I believe we
can.

If we can make object_new_configure() work with all (or most)
device types before we manually convert them to the new system,
we should.  I believe we can.

We may be able avoid these questions with -object because
converting all backends at the same time is doable.  With
devices, API usability and maintainability during the transition
period (which could be very long) needs to be taken into account.

-- 
Eduardo

Re: [PATCH 00/18] qapi/qom: QAPIfy object-add

[Eduardo Habkost]

On Thu, Dec 03, 2020 at 07:46:29AM +0100, Gerd Hoffmann wrote:
>   Hi,
> 
> > It would be much nicer to do the wrapper the other way round, i.e.
> > setting properties before the device is realized would update a
> > configuration struct and realize would then call .create() with that
> > struct. To me, this sounds much harder, though also a more useful state.
> 
> Well, in some places we already have separate config structs.  We have
> NICConf for example, which is typically used like this:
> 
>   struct USBNetState {
>  USBDevice dev;
>  [ ... ]
>  NICConf conf;
>  [ ... ]
>   };
> 
> and
> 
>   static Property net_properties[] = {
>   DEFINE_NIC_PROPERTIES(USBNetState, conf),
>   DEFINE_PROP_END_OF_LIST(),
>   };
> 
> So I think we could:
> 
>   (1) move *all* properties into structs.
>   (2) generate those structs from qapi schemas.
>   (3) generate Property lists (or functions with
>   object_class_property_add_*() calls) from qapi
>   schema.
> 
> We could then convert devices one-by-one without breaking anything
> or needing two code paths essentially doing the same thing in two
> different ways.

Sounds great to me.

This can also work the other way around for devices that weren't
converted yet: we should be able to generate a QAPI schema from
the property lists.

-- 
Eduardo

Re: [PATCH 00/18] qapi/qom: QAPIfy object-add

[Eduardo Habkost]

On Wed, Dec 02, 2020 at 06:35:06PM +0100, Kevin Wolf wrote:
> Am 02.12.2020 um 17:05 hat Eduardo Habkost geschrieben:
> > > > Looks nice as end goal.  Then, these are a few questions I would
> > > > have about the transition plan:
> > > > 
> > > > Would it require changing both device implementation and device
> > > > users in lockstep?  Should we have a compatibility layer to allow
> > > > existing qdev_new()+qdev_prop_set_*() code to keep working after
> > > > the devices are converted to the new system?  If not, why not?
> > > 
> > > Technically, it doesn't strictly require a lockstep update. You can have
> > > two code paths leading to a fully initialised device, one being the
> > > traditional way of setting properties and finally calling dc->realize,
> > > the other being a new method that takes the configuration in its
> > > parameters and also sets dev->realized = true at the end.
> > > 
> > > If at all possible, I would however prefer a lockstep update because
> > > having two paths is a weird intermediate state and the code paths could
> > > easily diverge. Keeping the old way around for a device also means that
> > > property setters are still doing two different jobs (initial
> > > configuration and updates at runtime).
> > 
> > I'd like to understand better how that intermediate state would
> > look like and why there's risk of separate code paths diverging.
> >
> > Could we have an intermediate state that doesn't require any
> > duplication and thus have no separate code paths that could
> > diverge?
> 
> The one requirement we have for an intermediate state is that it
> supports both interfaces: The well-know create/set properties/realize
> dance, and a new DeviceClass method, say .create(), that takes the
> configuration in parameters instead of relying on previously set
> properties.

I agree completely.

> 
> I assumed two separate implementations of transferring the configuration
> into the internal state. On second thought, this assumption is maybe
> wrong.
> 
> You can implement the new method as wrapper around the old way: It could
> just set all the properties and call realize. Of course, you don't win
> much in terms of improving the class implementation this way, but just
> support the new interface, but I guess it can be a reasonable
> intermediate step to resolve complicated dependencies etc.
> 
> It would be much nicer to do the wrapper the other way round, i.e.
> setting properties before the device is realized would update a
> configuration struct and realize would then call .create() with that
> struct. To me, this sounds much harder, though also a more useful state.

Comment about this below (look for [1]).

> 
> As you have worked a lot with properties recently, maybe you have a good
> idea how we could get an intermediate state closer to this?

I'd have to re-read this whole thread and think about it.

> 
> > > > If we add a compatibility layer, is the end goal to convert all
> > > > existing qdev_new() users to the new system?  If yes, why?  If
> > > > not, why not?
> > > 
> > > My personal goal is covering -object and -device, i.e. the external
> > > interfaces. Converting purely internally created devices is not as
> > > interesting (especially as long as we focus only on object creation),
> > > but might be desirable for consistency.
> > 
> > I wonder how much consistency we will lose and how much confusion
> > we'll cause if we end up with two completely separate methods for
> > creating devices.
> 
> I do think we should follow through and convert everything. It's just
> not my main motivation, and if the people who work more with qdev think
> it's better to leave that part unchanged (or that it won't make much of
> a difference), I won't insist.

This worries me.  Converting thousands of lines of code that
don't involve user-visible interfaces seems complicated and maybe
pointless.  On the other hand, having two separate APIs for
creating objects internally would cause confusion.

Maybe we should accept the fact that the 2 APIs will exist, and
address the confusion part: we should guarantee the two APIs to
be 100% equivalent, except for the fact that the newer one gives
us type safety in the C code.

I'd like to avoid a case like qdev vs QOM APIs, where they have
similar but slightly different features, and nobody knows which
one to use.

> 
> > > > What about subclasses?  Would base classes need to be converted
> > > > in lockstep with all subclasses?  How would the transition
> > > > process

Re: [PATCH 00/18] qapi/qom: QAPIfy object-add

[Eduardo Habkost]

On Wed, Dec 02, 2020 at 04:17:13PM +0100, Kevin Wolf wrote:
> Am 02.12.2020 um 14:54 hat Eduardo Habkost geschrieben:
> > On Wed, Dec 02, 2020 at 02:26:44PM +0100, Paolo Bonzini wrote:
> > > On 02/12/20 13:51, Eduardo Habkost wrote:
> > > > > > I'm liking the direction this is taking.  However, I would still
> > > > > > like to have a clearer and feasible plan that would work for
> > > > > > -device, -machine, and -cpu.
> > > > > 
> > > > > -cpu is not a problem since it's generally created with a static
> > > > > configuration (now done with global properties, in the future it 
> > > > > could be a
> > > > > struct).
> > > > 
> > > > It is a problem if it requires manually converting existing code
> > > > defining CPU properties and we don't have a transition plan.
> > > 
> > > We do not have to convert everything _if_ for some objects there are good
> > > reasons to do programmatically-generated properties.  CPUs might be one of
> > > those cases (or if we decide to convert them, they might endure some more
> > > code duplication than other devices because they have so many properties).
> > 
> > OK, we just need to agree on what the transition will look like
> > when we do it.  I think we should put as much care into
> > transition/glue infrastructure as we put into the new
> > infrastructure.
> > 
> > 
> > > 
> > > > Would a -device conversion also involve non-user-creatable
> > > > devices, or would we keep existing internal usage of QOM
> > > > properties?
> > > > 
> > > > Even if it's just for user-creatable devices, getting rid of QOM
> > > > property usage in devices sounds like a very ambitious goal.  I'd
> > > > like us to have a good transition plan, in addition to declaring
> > > > what's our ideal end goal.
> > > 
> > > For user-creatable objects Kevin is doing work in lockstep on all classes;
> > > but once we have the infrastructure for QAPI object configuration schemas 
> > > we
> > > can proceed in the other direction and operate on one device at a time.
> > > 
> > > With some handwaving, something like (see create_unimplemented_device)
> > > 
> > > DeviceState *dev = qdev_new(TYPE_UNIMPLEMENTED_DEVICE);
> > > 
> > > qdev_prop_set_string(dev, "name", name);
> > > qdev_prop_set_uint64(dev, "size", size);
> > > sysbus_realize_and_unref(SYS_BUS_DEVICE(dev), _fatal);
> > > 
> > > might become something like
> > > 
> > > { 'object': 'unimplemented-device',
> > >   'config': {
> > >  'name': 'str',
> > >  'size': 'size'
> > >   },
> > > }
> > > 
> > > DeviceState *dev = qapi_Unimplemented_new(&(
> > >  (UnimplementedDeviceOptions) {
> > >  .name = name,
> > >  .size = size
> > >  }, _fatal);
> > > object_unref(dev);
> > > 
> > > (i.e. all typesafe!) and the underlying code would be something like
> > [...]
> > > 
> > 
> > Looks nice as end goal.  Then, these are a few questions I would
> > have about the transition plan:
> > 
> > Would it require changing both device implementation and device
> > users in lockstep?  Should we have a compatibility layer to allow
> > existing qdev_new()+qdev_prop_set_*() code to keep working after
> > the devices are converted to the new system?  If not, why not?
> 
> Technically, it doesn't strictly require a lockstep update. You can have
> two code paths leading to a fully initialised device, one being the
> traditional way of setting properties and finally calling dc->realize,
> the other being a new method that takes the configuration in its
> parameters and also sets dev->realized = true at the end.
> 
> If at all possible, I would however prefer a lockstep update because
> having two paths is a weird intermediate state and the code paths could
> easily diverge. Keeping the old way around for a device also means that
> property setters are still doing two different jobs (initial
> configuration and updates at runtime).

I'd like to understand better how that intermediate state would
look like and why there's risk of separate code paths diverging.

Could we have an intermediate state that doesn't require any
duplication and thus have no separate code paths that could
diverge?


> 
> >

Re: [PATCH 00/18] qapi/qom: QAPIfy object-add

[Eduardo Habkost]

On Wed, Dec 02, 2020 at 02:26:44PM +0100, Paolo Bonzini wrote:
> On 02/12/20 13:51, Eduardo Habkost wrote:
> > > > I'm liking the direction this is taking.  However, I would still
> > > > like to have a clearer and feasible plan that would work for
> > > > -device, -machine, and -cpu.
> > > 
> > > -cpu is not a problem since it's generally created with a static
> > > configuration (now done with global properties, in the future it could be 
> > > a
> > > struct).
> > 
> > It is a problem if it requires manually converting existing code
> > defining CPU properties and we don't have a transition plan.
> 
> We do not have to convert everything _if_ for some objects there are good
> reasons to do programmatically-generated properties.  CPUs might be one of
> those cases (or if we decide to convert them, they might endure some more
> code duplication than other devices because they have so many properties).

OK, we just need to agree on what the transition will look like
when we do it.  I think we should put as much care into
transition/glue infrastructure as we put into the new
infrastructure.


> 
> > Would a -device conversion also involve non-user-creatable
> > devices, or would we keep existing internal usage of QOM
> > properties?
> > 
> > Even if it's just for user-creatable devices, getting rid of QOM
> > property usage in devices sounds like a very ambitious goal.  I'd
> > like us to have a good transition plan, in addition to declaring
> > what's our ideal end goal.
> 
> For user-creatable objects Kevin is doing work in lockstep on all classes;
> but once we have the infrastructure for QAPI object configuration schemas we
> can proceed in the other direction and operate on one device at a time.
> 
> With some handwaving, something like (see create_unimplemented_device)
> 
> DeviceState *dev = qdev_new(TYPE_UNIMPLEMENTED_DEVICE);
> 
> qdev_prop_set_string(dev, "name", name);
> qdev_prop_set_uint64(dev, "size", size);
> sysbus_realize_and_unref(SYS_BUS_DEVICE(dev), _fatal);
> 
> might become something like
> 
> { 'object': 'unimplemented-device',
>   'config': {
>  'name': 'str',
>  'size': 'size'
>   },
> }
> 
> DeviceState *dev = qapi_Unimplemented_new(&(
>  (UnimplementedDeviceOptions) {
>  .name = name,
>  .size = size
>  }, _fatal);
> object_unref(dev);
> 
> (i.e. all typesafe!) and the underlying code would be something like
[...]
> 

Looks nice as end goal.  Then, these are a few questions I would
have about the transition plan:

Would it require changing both device implementation and device
users in lockstep?  Should we have a compatibility layer to allow
existing qdev_new()+qdev_prop_set_*() code to keep working after
the devices are converted to the new system?  If not, why not?

If we add a compatibility layer, is the end goal to convert all
existing qdev_new() users to the new system?  If yes, why?  If
not, why not?

What about subclasses?  Would base classes need to be converted
in lockstep with all subclasses?  How would the transition
process of (e.g.) PCI devices look like?

-- 
Eduardo

Re: [PATCH 00/18] qapi/qom: QAPIfy object-add

[Eduardo Habkost]

On Wed, Dec 02, 2020 at 10:30:11AM +0100, Paolo Bonzini wrote:
> On 01/12/20 23:08, Eduardo Habkost wrote:
> > > Properties are only a useful concept if they have a use.  If
> > > -object/object_add/object-add can do the same job without properties,
> > > properties are not needed anymore.
> > 
> > Do you mean "not needed for -object anymore"?  Properties are
> > still used by internal C code (esp. board code),
> > -device/device_add, -machine, -cpu, and debugging commands (like
> > "info qtree" and qom-list/qom-get/qom-set).
> 
> Yes.
> 
> > > Right now QOM is all about exposing properties, and having multiple
> > > interfaces to set them (by picking a different visitor).  But in practice
> > > most QOM objects have a lifetime that consists of 1) set properties 2) 
> > > flip
> > > a switch (realized/complete/open) 3) let the object live on its own.  1+2
> > > are a single monitor command or CLI option; during 3 you access the object
> > > through monitor commands, not properties.
> > 
> > I agree with this, except for the word "all" in "QOM is all
> > about".  QOM is also an extensively used internal QEMU API,
> > including internal usage of the QOM property system.
> 
> Yeah, "all about exposing properties" includes internal usage.  And you're
> right that some "phase 3" monitor commands do work at the property level
> (mostly "info qtree", but also "qom-get" because there are some cases of
> public run-time properties).

I still disagree on the "all about" part even for internal usage.
But this shouldn't really matter for this discussion, I guess.

> 
> > I'm liking the direction this is taking.  However, I would still
> > like to have a clearer and feasible plan that would work for
> > -device, -machine, and -cpu.
> 
> -cpu is not a problem since it's generally created with a static
> configuration (now done with global properties, in the future it could be a
> struct).

It is a problem if it requires manually converting existing code
defining CPU properties and we don't have a transition plan.

> 
> -machine and -device in principle could be done the same way as -object,
> just through a different registry (_not_ a huge struct; that's an acceptable
> stopgap for -object but that's it).  The static aka field properties would
> remain as read-only, with defaults moved to instance_init or realize.  But
> there would be again "triplication" with a trivial conversion:

> 
> 1) in the QAPI schema, e.g. 'num_queues': 'int16'
> 
> 2) in the struct, "int16_t num_queues;"
> 
> 3) in the realize function,
> 
> s->num_queues = cfg->has_num_queues ? cfg->num_queues : 8;
> 
> So having a mechanism for defaults in the QAPI schema would be good. Maybe
> 'num_queues': { 'type': 'int16', 'default': '8' }?
> 

Would a -device conversion also involve non-user-creatable
devices, or would we keep existing internal usage of QOM
properties?

Even if it's just for user-creatable devices, getting rid of QOM
property usage in devices sounds like a very ambitious goal.  I'd
like us to have a good transition plan, in addition to declaring
what's our ideal end goal.


> I also need to review more the part of this code with respect to the
> application of global properties.  I wonder if there are visitor tricks that
> we can do, so that global properties keep working but correspond to QAPI
> fields instead of QOM properties.
> 
> Paolo
> 

-- 
Eduardo

Re: [PATCH 00/18] qapi/qom: QAPIfy object-add

[Eduardo Habkost]

On Tue, Dec 01, 2020 at 10:23:57PM +0100, Paolo Bonzini wrote:
> On 01/12/20 20:35, Kevin Wolf wrote:
> > Am 01.12.2020 um 18:16 hat Paolo Bonzini geschrieben:
> > I don't think this is actually a new things. We already have types and
> > commands declared with things like 'if': 'defined(TARGET_S390X)'.
> > As far as I understand, QAPI generated files are already built per
> > target, so not compiling things into all binaries should be entirely
> > possible.
> 
> There is some complication due to having different discriminators per
> target.  So yes it should be possible.  But probably best left after objects
> because it's so much bigger a task and because objects have a bit more
> freedom for experimentation (less ties to other qdev-specific concepts, e.g.
> the magic "bus" property).
> 
> > So maybe only the abstract base class that actually defines the machine
> > properties (like generic-pc-machine) should be described in QAPI, and
> > then the concrete machine types would inherit from it without being
> > described in QAPI themselves?
> 
> Yes, maybe.
> 
> > > 1) whether to generate _all_ boilerplate or only properties
> > 
> > I would like to generate as much boilerplate as possible. That is, I
> > don't want to constrain us to only properties, but at the same time, I'm
> > not sure if it's possible to get rid of all boilerplate.
> > 
> > Basically, the vision I have in mind is that QAPI would generate code
> > that is the same for most instances, and then provide an option that
> > prevents code generation for a specific part for more complicated cases,
> > so that the respective function can (and must) be provided in the C
> > source.
> 
> Ok, so that's a bit different from what I am thinking of.  I don't care very
> much about the internal boilerplate, only the external interface for
> configuration.  So I don't care about type registration, dynamic cast macros
> etc., only essentially the part that leads to ucc->complete.
> 
> > > 2) whether we want to introduce a separation between configuration
> > > schema and run-time state
> > 
> > You mean the internal run-time state? How is this separation not already
> > present with getter/setter functions for each property? In many cases
> > they just directly access the run-time state, but there are other cases
> > where they actually do things.
> 
> I mean moving more towards the blockdev/chardev way of doing things,
> increasing the separation very much by having separate configuration structs
> that have (potentially) no link to the run-time state struct.
> 
> > > 3) in the latter case, whether properties will survive at all---iothread 
> > > and
> > > throttle-groups don't really need them even if they're writable after
> > > creation.
> > 
> > How do you define properties, i.e. at which point would they stop
> > existing and what would be a non-property alternative?
> 
> Properties are only a useful concept if they have a use.  If
> -object/object_add/object-add can do the same job without properties,
> properties are not needed anymore.

Do you mean "not needed for -object anymore"?  Properties are
still used by internal C code (esp. board code),
-device/device_add, -machine, -cpu, and debugging commands (like
"info qtree" and qom-list/qom-get/qom-set).

> 
> Right now QOM is all about exposing properties, and having multiple
> interfaces to set them (by picking a different visitor).  But in practice
> most QOM objects have a lifetime that consists of 1) set properties 2) flip
> a switch (realized/complete/open) 3) let the object live on its own.  1+2
> are a single monitor command or CLI option; during 3 you access the object
> through monitor commands, not properties.

I agree with this, except for the word "all" in "QOM is all
about".  QOM is also an extensively used internal QEMU API,
including internal usage of the QOM property system.

> 
> > So in summary, it seems to me that the QOM way is more flexible because
> > you can get both models out of it. Whether we actually need this
> > flexibility I can't say.
> 
> I'm thinking there's no need for it, but maybe I'm overly optimistic.
> 
> > * Configuration options are described in the QAPI schema. This is mainly
> >for object creation, but runtime modifiable properties are a subset of
> >this.
> > 
> > * Properties are generated for each option. By default, the getter
> >just returns the value from the configuration at creation time, though
> >generation of it can be disabled so that it can be overridden. Also,
> >setters just return an error by default.
> > 
> > * Property setters aren't called for object creation. Instead, the
> >relevant ObjectOptions branch is made available to some init method.
> > 
> > * Runtime modifiable properties (declared as such in the schema) don't
> >get the default setter, so you have to provide an implementation for
> >them.
> 
> I wouldn't bother with properties at all in the QAPI schema.  Just do the
> first and third 

Re: [PATCH 00/18] qapi/qom: QAPIfy object-add

[Eduardo Habkost]

On Tue, Dec 01, 2020 at 06:16:14PM +0100, Paolo Bonzini wrote:
> On 01/12/20 17:20, Kevin Wolf wrote:
[...]
> > BlockdevOptions is about external interfaces, not about
> > implementation details. Same thing as QOM properties are external
> > interfaces, not implementation details. There may be fields in the
> > internal state that correspond 1:1 to the externally visible QAPI
> > type/QOM property, but it's not necessarily the case.
> 
> Right.  It may well be that we decide that, as a result of this series,
> QOM's property interface is declared essentially a failed experiment.  I
> wouldn't be surprised, and that's why I want to understand better where we
> want to go.
> 
> For example, Eduardo is focusing specifically on external interfaces that
> correspond 1:1 to the internal implementation.  If we decide that
> non-1:1-mappings and checks on mandatory properties are an important part of
> QOM, that may make large parts of his work moot.  [...]

Whatever we decide, my first and biggest worry is to have a
reasonable migration path for any new API.

We could describe in detail what's the API we _want_, but we also
need a reasonable migration path for the code we already _have_.
If a new API that requires manual conversion of existing devices,
it will probably coexist with the existing qdev property API for
years.

(Note that I haven't read this whole thread yet.)

>[...]  If we decide that most QOM
> objects need no properties at all, then we don't want to move more
> qdev-specific stuff from to QOM.

I don't understand what "move more qdev-specific stuff to QOM"
means.  I consider the QOM field property API valuable even if we
decide the only user of the API will be legacy qdev code, because
it separates the core mechanism (that's code that already
existed) from qdev-specific policies.

> 
> > QAPI is already here and it's going to stay. QOM doesn't have to
> > duplicate input validation that existing code can already perform.
> > 
> > I'm not sure which complexity you think I'm introducing: QAPI is already
> > there. I'm adding the schema, which you agree is valuable documentation,
> > so we want to have it either case. The actual change to QOM that we have
> > in this series is this:
> 
> The complexity is that properties used to be split in two places, and now
> they're split in three places.
> 
> It may very well be that this is a good first step to at least have classes
> described in the QAPI schema.  But since _in the short term_ there are
> things that the series makes worse (and has a risk of bringing things out of
> sync), I'd like to understand the long term plan and ensure that the QAPI
> maintainers are on board with it.
> 
> Can you at least add a comment to all UserCreatable classes that says "if
> you add a property, remember to modify ... as well in the QAPI schema"?
> 
> > > Are there any validation bugs that you're fixing?  Is that
> > > something that cannot be fixed elsewhere, or are you papering over bad QOM
> > > coding?  (Again, I'm not debating that QOM properties are hard to write
> > > correctly).
> > 
> > Yes, I found bugs that the QAPI schema would have prevented. They were
> > generally about not checking whether mandatory propertes are actually
> > set.
> 
> Okay, I found your series at
> https://patchew.org/QEMU/20201130105615.21799-1-kw...@redhat.com/ too, good
> to know.
> 
> So that's another useful thing that can be chalked to this series at least
> if -object and object_add are converted (and also, another thing against QOM
> properties and 1:1 mappings between configuration schema and run-time
> state).
> 
> Paolo
> 

-- 
Eduardo

[PATCH v2 6/6] Rename arch_init.h to arch_type.h

[Eduardo Habkost]

The only declarations in arch_init.h are related to the arch_type
variable (which is a useful feature that allows us to simplify
command line option handling).  Rename the header to reflect its
purpose.

Signed-off-by: Eduardo Habkost 
---
Cc: Markus Armbruster 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: Eduardo Habkost 
Cc: qemu-block@nongnu.org
Cc: qemu-de...@nongnu.org
---
 include/sysemu/{arch_init.h => arch_type.h} | 0
 blockdev.c  | 2 +-
 softmmu/arch_init.c | 2 +-
 softmmu/qdev-monitor.c  | 2 +-
 softmmu/vl.c| 2 +-
 stubs/arch_type.c   | 2 +-
 6 files changed, 5 insertions(+), 5 deletions(-)
 rename include/sysemu/{arch_init.h => arch_type.h} (100%)

diff --git a/include/sysemu/arch_init.h b/include/sysemu/arch_type.h
similarity index 100%
rename from include/sysemu/arch_init.h
rename to include/sysemu/arch_type.h
diff --git a/blockdev.c b/blockdev.c
index fe6fb5dc1d..46c10b2609 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -56,7 +56,7 @@
 #include "sysemu/iothread.h"
 #include "block/block_int.h"
 #include "block/trace.h"
-#include "sysemu/arch_init.h"
+#include "sysemu/arch_type.h"
 #include "sysemu/qtest.h"
 #include "sysemu/runstate.h"
 #include "sysemu/replay.h"
diff --git a/softmmu/arch_init.c b/softmmu/arch_init.c
index f4770931f5..5a9bc56387 100644
--- a/softmmu/arch_init.c
+++ b/softmmu/arch_init.c
@@ -24,7 +24,7 @@
 #include "qemu/osdep.h"
 #include "cpu.h"
 #include "sysemu/sysemu.h"
-#include "sysemu/arch_init.h"
+#include "sysemu/arch_type.h"
 #include "hw/pci/pci.h"
 #include "hw/audio/soundhw.h"
 #include "qapi/error.h"
diff --git a/softmmu/qdev-monitor.c b/softmmu/qdev-monitor.c
index bf79d0bbcd..c8b7fb27dc 100644
--- a/softmmu/qdev-monitor.c
+++ b/softmmu/qdev-monitor.c
@@ -22,7 +22,7 @@
 #include "monitor/hmp.h"
 #include "monitor/monitor.h"
 #include "monitor/qdev.h"
-#include "sysemu/arch_init.h"
+#include "sysemu/arch_type.h"
 #include "qapi/error.h"
 #include "qapi/qapi-commands-qdev.h"
 #include "qapi/qmp/qdict.h"
diff --git a/softmmu/vl.c b/softmmu/vl.c
index 74b6ebf1e4..1dd63b2782 100644
--- a/softmmu/vl.c
+++ b/softmmu/vl.c
@@ -95,7 +95,7 @@
 #include "trace/control.h"
 #include "qemu/plugin.h"
 #include "qemu/queue.h"
-#include "sysemu/arch_init.h"
+#include "sysemu/arch_type.h"
 
 #include "ui/qemu-spice.h"
 #include "qapi/string-input-visitor.h"
diff --git a/stubs/arch_type.c b/stubs/arch_type.c
index fc5423bc98..603a49deec 100644
--- a/stubs/arch_type.c
+++ b/stubs/arch_type.c
@@ -1,4 +1,4 @@
 #include "qemu/osdep.h"
-#include "sysemu/arch_init.h"
+#include "sysemu/arch_type.h"
 
 const uint32_t arch_type = QEMU_ARCH_NONE;
-- 
2.28.0

[PATCH v3 23/53] qdev: Move dev->realized check to qdev_property_set()

[Eduardo Habkost]

Every single qdev property setter function manually checks
dev->realized.  We can just check dev->realized inside
qdev_property_set() instead.

The check is being added as a separate function
(qdev_prop_allow_set()) because it will become a callback later.

Reviewed-by: Stefan Berger 
Signed-off-by: Eduardo Habkost 
---
Changes v1 -> v2:
* Removed unused variable at xen_block_set_vdev()
* Redone patch after changes in the previous patches in the
  series
---
Cc: Stefan Berger 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: Eduardo Habkost 
Cc: Cornelia Huck 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Thomas Huth 
Cc: Matthew Rosato 
Cc: Alex Williamson 
Cc: Mark Cave-Ayland 
Cc: Artyom Tarasenko 
Cc: qemu-de...@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
---
 backends/tpm/tpm_util.c  |   6 --
 hw/block/xen-block.c |   6 --
 hw/core/qdev-properties-system.c |  70 --
 hw/core/qdev-properties.c| 100 ++-
 hw/s390x/css.c   |   6 --
 hw/s390x/s390-pci-bus.c  |   6 --
 hw/vfio/pci-quirks.c |   6 --
 target/sparc/cpu.c   |   6 --
 8 files changed, 18 insertions(+), 188 deletions(-)

diff --git a/backends/tpm/tpm_util.c b/backends/tpm/tpm_util.c
index dba2f6b04a..0b07cf55ea 100644
--- a/backends/tpm/tpm_util.c
+++ b/backends/tpm/tpm_util.c
@@ -46,16 +46,10 @@ static void get_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 static void set_tpm(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 TPMBackend *s, **be = qdev_get_prop_ptr(obj, prop);
 char *str;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
diff --git a/hw/block/xen-block.c b/hw/block/xen-block.c
index 905e4acd97..bd1aef63a7 100644
--- a/hw/block/xen-block.c
+++ b/hw/block/xen-block.c
@@ -395,17 +395,11 @@ static int vbd_name_to_disk(const char *name, const char 
**endp,
 static void xen_block_set_vdev(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
 char *str, *p;
 const char *end;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index 202abd0e4b..0d3e57bba0 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -94,11 +94,6 @@ static void set_drive_helper(Object *obj, Visitor *v, const 
char *name,
 bool blk_created = false;
 int ret;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
@@ -230,17 +225,11 @@ static void get_chr(Object *obj, Visitor *v, const char 
*name, void *opaque,
 static void set_chr(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 CharBackend *be = qdev_get_prop_ptr(obj, prop);
 Chardev *s;
 char *str;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
@@ -311,18 +300,12 @@ static void get_mac(Object *obj, Visitor *v, const char 
*name, void *opaque,
 static void set_mac(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 MACAddr *mac = qdev_get_prop_ptr(obj, prop);
 int i, pos;
 char *str;
 const char *p;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
@@ -390,7 +373,6 @@ static void get_netdev(Object *obj, Visitor *v, const char 
*name,
 static void set_netdev(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 NICPeers *peers_ptr = qdev_get_prop_ptr(obj, prop);
 NetClientState **ncs = peers_ptr->ncs;
@@ -398,11 +380,6 @@ static void set_netdev(Object *obj, Visitor *v, const char 
*name,
 int queues, err = 0, i = 0;
   

[PATCH v3 30/53] qdev: Rename qdev_get_prop_ptr() to object_field_prop_ptr()

[Eduardo Habkost]

The function will be moved to common QOM code, as it is not
specific to TYPE_DEVICE anymore.

Reviewed-by: Stefan Berger 
Signed-off-by: Eduardo Habkost 
---
Changes v1 -> v2:
* Rename to object_field_prop_ptr() instead of object_static_prop_ptr()
---
Cc: Stefan Berger 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: Eduardo Habkost 
Cc: Cornelia Huck 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Thomas Huth 
Cc: Matthew Rosato 
Cc: Alex Williamson 
Cc: qemu-de...@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
---
 include/hw/qdev-properties.h |  2 +-
 backends/tpm/tpm_util.c  |  6 ++--
 hw/block/xen-block.c |  4 +--
 hw/core/qdev-properties-system.c | 50 +-
 hw/core/qdev-properties.c| 60 
 hw/s390x/css.c   |  4 +--
 hw/s390x/s390-pci-bus.c  |  4 +--
 hw/vfio/pci-quirks.c |  4 +--
 8 files changed, 67 insertions(+), 67 deletions(-)

diff --git a/include/hw/qdev-properties.h b/include/hw/qdev-properties.h
index 90222822f1..97bb9494ae 100644
--- a/include/hw/qdev-properties.h
+++ b/include/hw/qdev-properties.h
@@ -193,7 +193,7 @@ void qdev_prop_set_macaddr(DeviceState *dev, const char 
*name,
const uint8_t *value);
 void qdev_prop_set_enum(DeviceState *dev, const char *name, int value);
 
-void *qdev_get_prop_ptr(Object *obj, Property *prop);
+void *object_field_prop_ptr(Object *obj, Property *prop);
 
 void qdev_prop_register_global(GlobalProperty *prop);
 const GlobalProperty *qdev_find_global_prop(Object *obj,
diff --git a/backends/tpm/tpm_util.c b/backends/tpm/tpm_util.c
index 0b07cf55ea..bb1ab34a75 100644
--- a/backends/tpm/tpm_util.c
+++ b/backends/tpm/tpm_util.c
@@ -35,7 +35,7 @@
 static void get_tpm(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-TPMBackend **be = qdev_get_prop_ptr(obj, opaque);
+TPMBackend **be = object_field_prop_ptr(obj, opaque);
 char *p;
 
 p = g_strdup(*be ? (*be)->id : "");
@@ -47,7 +47,7 @@ static void set_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 Error **errp)
 {
 Property *prop = opaque;
-TPMBackend *s, **be = qdev_get_prop_ptr(obj, prop);
+TPMBackend *s, **be = object_field_prop_ptr(obj, prop);
 char *str;
 
 if (!visit_type_str(v, name, , errp)) {
@@ -67,7 +67,7 @@ static void set_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 static void release_tpm(Object *obj, const char *name, void *opaque)
 {
 Property *prop = opaque;
-TPMBackend **be = qdev_get_prop_ptr(obj, prop);
+TPMBackend **be = object_field_prop_ptr(obj, prop);
 
 if (*be) {
 tpm_backend_reset(*be);
diff --git a/hw/block/xen-block.c b/hw/block/xen-block.c
index bd1aef63a7..718d886e5c 100644
--- a/hw/block/xen-block.c
+++ b/hw/block/xen-block.c
@@ -336,7 +336,7 @@ static void xen_block_get_vdev(Object *obj, Visitor *v, 
const char *name,
void *opaque, Error **errp)
 {
 Property *prop = opaque;
-XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
+XenBlockVdev *vdev = object_field_prop_ptr(obj, prop);
 char *str;
 
 switch (vdev->type) {
@@ -396,7 +396,7 @@ static void xen_block_set_vdev(Object *obj, Visitor *v, 
const char *name,
void *opaque, Error **errp)
 {
 Property *prop = opaque;
-XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
+XenBlockVdev *vdev = object_field_prop_ptr(obj, prop);
 char *str, *p;
 const char *end;
 
diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index 96a0bc5109..8781b856d3 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -62,7 +62,7 @@ static void get_drive(Object *obj, Visitor *v, const char 
*name, void *opaque,
   Error **errp)
 {
 Property *prop = opaque;
-void **ptr = qdev_get_prop_ptr(obj, prop);
+void **ptr = object_field_prop_ptr(obj, prop);
 const char *value;
 char *p;
 
@@ -88,7 +88,7 @@ static void set_drive_helper(Object *obj, Visitor *v, const 
char *name,
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-void **ptr = qdev_get_prop_ptr(obj, prop);
+void **ptr = object_field_prop_ptr(obj, prop);
 char *str;
 BlockBackend *blk;
 bool blk_created = false;
@@ -181,7 +181,7 @@ static void release_drive(Object *obj, const char *name, 
void *opaque)
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-BlockBackend **ptr = qdev_get_prop_ptr(obj, prop);
+BlockBackend **ptr = object_field_prop_ptr(obj, prop);
 
 if (*ptr) {
 AioContext *ctx = blk_get_aio_co

[PATCH v3 09/53] qdev: Make qdev_get_prop_ptr() get Object* arg

[Eduardo Habkost]

Make the code more generic and not specific to TYPE_DEVICE.

Reviewed-by: Marc-André Lureau 
Signed-off-by: Eduardo Habkost 
---
Changes v1 -> v2:
- Fix build error with CONFIG_XEN
  I took the liberty of keeping the Reviewed-by line from
  Marc-André as the build fix is a trivial one line change
---
Cc: Stefan Berger 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: Eduardo Habkost 
Cc: Cornelia Huck 
Cc: Thomas Huth 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Matthew Rosato 
Cc: Alex Williamson 
Cc: qemu-de...@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
---
 include/hw/qdev-properties.h |  2 +-
 backends/tpm/tpm_util.c  |  8 ++--
 hw/block/xen-block.c |  5 +-
 hw/core/qdev-properties-system.c | 57 +-
 hw/core/qdev-properties.c| 82 +---
 hw/s390x/css.c   |  5 +-
 hw/s390x/s390-pci-bus.c  |  4 +-
 hw/vfio/pci-quirks.c |  5 +-
 8 files changed, 68 insertions(+), 100 deletions(-)

diff --git a/include/hw/qdev-properties.h b/include/hw/qdev-properties.h
index 0ea822e6a7..0b92cfc761 100644
--- a/include/hw/qdev-properties.h
+++ b/include/hw/qdev-properties.h
@@ -302,7 +302,7 @@ void qdev_prop_set_macaddr(DeviceState *dev, const char 
*name,
const uint8_t *value);
 void qdev_prop_set_enum(DeviceState *dev, const char *name, int value);
 
-void *qdev_get_prop_ptr(DeviceState *dev, Property *prop);
+void *qdev_get_prop_ptr(Object *obj, Property *prop);
 
 void qdev_prop_register_global(GlobalProperty *prop);
 const GlobalProperty *qdev_find_global_prop(DeviceState *dev,
diff --git a/backends/tpm/tpm_util.c b/backends/tpm/tpm_util.c
index b58d298c1a..e91c21dd4a 100644
--- a/backends/tpm/tpm_util.c
+++ b/backends/tpm/tpm_util.c
@@ -35,8 +35,7 @@
 static void get_tpm(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
-TPMBackend **be = qdev_get_prop_ptr(dev, opaque);
+TPMBackend **be = qdev_get_prop_ptr(obj, opaque);
 char *p;
 
 p = g_strdup(*be ? (*be)->id : "");
@@ -49,7 +48,7 @@ static void set_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-TPMBackend *s, **be = qdev_get_prop_ptr(dev, prop);
+TPMBackend *s, **be = qdev_get_prop_ptr(obj, prop);
 char *str;
 
 if (dev->realized) {
@@ -73,9 +72,8 @@ static void set_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 
 static void release_tpm(Object *obj, const char *name, void *opaque)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-TPMBackend **be = qdev_get_prop_ptr(dev, prop);
+TPMBackend **be = qdev_get_prop_ptr(obj, prop);
 
 if (*be) {
 tpm_backend_reset(*be);
diff --git a/hw/block/xen-block.c b/hw/block/xen-block.c
index 8a7a3f5452..905e4acd97 100644
--- a/hw/block/xen-block.c
+++ b/hw/block/xen-block.c
@@ -335,9 +335,8 @@ static char *disk_to_vbd_name(unsigned int disk)
 static void xen_block_get_vdev(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-XenBlockVdev *vdev = qdev_get_prop_ptr(dev, prop);
+XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
 char *str;
 
 switch (vdev->type) {
@@ -398,7 +397,7 @@ static void xen_block_set_vdev(Object *obj, Visitor *v, 
const char *name,
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-XenBlockVdev *vdev = qdev_get_prop_ptr(dev, prop);
+XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
 char *str, *p;
 const char *end;
 
diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index d0fb063a49..c8c73c371b 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -59,9 +59,8 @@ static bool check_prop_still_unset(DeviceState *dev, const 
char *name,
 static void get_drive(Object *obj, Visitor *v, const char *name, void *opaque,
   Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-void **ptr = qdev_get_prop_ptr(dev, prop);
+void **ptr = qdev_get_prop_ptr(obj, prop);
 const char *value;
 char *p;
 
@@ -87,7 +86,7 @@ static void set_drive_helper(Object *obj, Visitor *v, const 
char *name,
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-void **ptr = qdev_get_prop_ptr(dev, prop);
+void **ptr = qdev_get_prop_ptr(obj, prop);
 char *str;
 BlockBackend *blk;
 bool blk_created = false;
@@ -185,7 +184,7 @@ static void release_drive(Object *obj, const char *name, 
void *

[PATCH 6/7] qom: Add FIELD_PTR, a type-safe wrapper for object_field_prop_ptr()

[Eduardo Habkost]

Introduce a FIELD_PTR macro that will ensure the size of the area
we are accessing has the correct size, and will return a pointer
of the correct type.

Signed-off-by: Eduardo Habkost 
---
Cc: Stefan Berger 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: Eduardo Habkost 
Cc: Cornelia Huck 
Cc: Thomas Huth 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Matthew Rosato 
Cc: Alex Williamson 
Cc: qemu-de...@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
---
 include/qom/field-property.h | 21 ++-
 backends/tpm/tpm_util.c  |  6 ++--
 hw/block/xen-block.c |  4 +--
 hw/core/qdev-properties-system.c | 50 +-
 hw/s390x/css.c   |  4 +--
 hw/s390x/s390-pci-bus.c  |  4 +--
 hw/vfio/pci-quirks.c |  4 +--
 qom/field-property.c |  3 +-
 qom/property-types.c | 60 +---
 9 files changed, 89 insertions(+), 67 deletions(-)

diff --git a/include/qom/field-property.h b/include/qom/field-property.h
index 1d3bf9699b..58baaca160 100644
--- a/include/qom/field-property.h
+++ b/include/qom/field-property.h
@@ -125,6 +125,25 @@ object_class_property_add_field(ObjectClass *oc, const 
char *name,
 Property *prop,
 ObjectPropertyAllowSet allow_set);
 
-void *object_field_prop_ptr(Object *obj, Property *prop);
+/**
+ * object_field_prop_ptr: Get pointer to property field
+ * @obj: the object instance
+ * @prop: field property definition
+ * @expected_size: expected size of struct field
+ *
+ * Don't use this function directly, use the FIELD_PTR() macro instead.
+ */
+void *object_field_prop_ptr(Object *obj, Property *prop, size_t expected_size);
+
+/**
+ * FIELD_PTR: Get pointer to struct field for property
+ *
+ * This returns a pointer to type @type, pointing to the struct
+ * field containing the property value.
+ *
+ * @type must match the expected type for the property.
+ */
+#define FIELD_PTR(obj, prop, type) \
+((type *)object_field_prop_ptr((obj), (prop), sizeof(type)))
 
 #endif
diff --git a/backends/tpm/tpm_util.c b/backends/tpm/tpm_util.c
index 556e21388c..da80379404 100644
--- a/backends/tpm/tpm_util.c
+++ b/backends/tpm/tpm_util.c
@@ -35,7 +35,7 @@
 static void get_tpm(Object *obj, Visitor *v, const char *name,
 Property *prop, Error **errp)
 {
-TPMBackend **be = object_field_prop_ptr(obj, prop);
+TPMBackend **be = FIELD_PTR(obj, prop, TPMBackend *);
 char *p;
 
 p = g_strdup(*be ? (*be)->id : "");
@@ -46,7 +46,7 @@ static void get_tpm(Object *obj, Visitor *v, const char *name,
 static void set_tpm(Object *obj, Visitor *v, const char *name,
 Property *prop, Error **errp)
 {
-TPMBackend *s, **be = object_field_prop_ptr(obj, prop);
+TPMBackend *s, **be = FIELD_PTR(obj, prop, TPMBackend *);
 char *str;
 
 if (!visit_type_str(v, name, , errp)) {
@@ -65,7 +65,7 @@ static void set_tpm(Object *obj, Visitor *v, const char *name,
 
 static void release_tpm(Object *obj, const char *name, Property *prop)
 {
-TPMBackend **be = object_field_prop_ptr(obj, prop);
+TPMBackend **be = FIELD_PTR(obj, prop, TPMBackend *);
 
 if (*be) {
 tpm_backend_reset(*be);
diff --git a/hw/block/xen-block.c b/hw/block/xen-block.c
index c1ee634639..390bf417ab 100644
--- a/hw/block/xen-block.c
+++ b/hw/block/xen-block.c
@@ -335,7 +335,7 @@ static char *disk_to_vbd_name(unsigned int disk)
 static void xen_block_get_vdev(Object *obj, Visitor *v, const char *name,
 Property *prop, Error **errp)
 {
-XenBlockVdev *vdev = object_field_prop_ptr(obj, prop);
+XenBlockVdev *vdev = FIELD_PTR(obj, prop, XenBlockVdev);
 char *str;
 
 switch (vdev->type) {
@@ -394,7 +394,7 @@ static int vbd_name_to_disk(const char *name, const char 
**endp,
 static void xen_block_set_vdev(Object *obj, Visitor *v, const char *name,
 Property *prop, Error **errp)
 {
-XenBlockVdev *vdev = object_field_prop_ptr(obj, prop);
+XenBlockVdev *vdev = FIELD_PTR(obj, prop, XenBlockVdev);
 char *str, *p;
 const char *end;
 
diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index 2fdd5863bb..1ec64514b9 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -61,7 +61,7 @@ static bool check_prop_still_unset(Object *obj, const char 
*name,
 static void get_drive(Object *obj, Visitor *v, const char *name,
   Property *prop, Error **errp)
 {
-void **ptr = object_field_prop_ptr(obj, prop);
+void **ptr = FIELD_PTR(obj, prop, void *);
 const char *value;
 char *p;
 
@@ -87,7 +87,7 @@ static void set_drive_h

[PATCH 4/7] qom: Replace void* parameter with Property* on field getters/setters

[Eduardo Habkost]

All field property getters and setters must interpret the fourth
argument as Property*.  Change the function signature of field
property getters and setters to indicate that.

Signed-off-by: Eduardo Habkost 
---
Cc: Stefan Berger 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: Eduardo Habkost 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Cornelia Huck 
Cc: Thomas Huth 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Matthew Rosato 
Cc: Alex Williamson 
Cc: Mark Cave-Ayland 
Cc: Artyom Tarasenko 
Cc: qemu-de...@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
---
 include/qom/field-property-internal.h |   8 +-
 include/qom/field-property.h  |  26 ---
 backends/tpm/tpm_util.c   |  11 ++-
 hw/block/xen-block.c  |   6 +-
 hw/core/qdev-properties-system.c  |  86 +-
 hw/s390x/css.c|   6 +-
 hw/s390x/s390-pci-bus.c   |   6 +-
 hw/vfio/pci-quirks.c  |  10 +--
 qom/property-types.c  | 102 +-
 target/sparc/cpu.c|   4 +-
 10 files changed, 105 insertions(+), 160 deletions(-)

diff --git a/include/qom/field-property-internal.h 
b/include/qom/field-property-internal.h
index 7aa27ce836..bc7d25033d 100644
--- a/include/qom/field-property-internal.h
+++ b/include/qom/field-property-internal.h
@@ -9,9 +9,9 @@
 #define QOM_STATIC_PROPERTY_INTERNAL_H
 
 void field_prop_get_enum(Object *obj, Visitor *v, const char *name,
- void *opaque, Error **errp);
+ Property *prop, Error **errp);
 void field_prop_set_enum(Object *obj, Visitor *v, const char *name,
- void *opaque, Error **errp);
+ Property *prop, Error **errp);
 
 void field_prop_set_default_value_enum(ObjectProperty *op,
const Property *prop);
@@ -21,9 +21,9 @@ void field_prop_set_default_value_uint(ObjectProperty *op,
const Property *prop);
 
 void field_prop_get_int32(Object *obj, Visitor *v, const char *name,
-  void *opaque, Error **errp);
+  Property *prop, Error **errp);
 void field_prop_get_size32(Object *obj, Visitor *v, const char *name,
-   void *opaque, Error **errp);
+   Property *prop, Error **errp);
 
 /**
  * object_property_add_field: Add a field property to an object instance
diff --git a/include/qom/field-property.h b/include/qom/field-property.h
index e64a2b3c07..438bb25896 100644
--- a/include/qom/field-property.h
+++ b/include/qom/field-property.h
@@ -54,6 +54,18 @@ struct Property {
 const char   *link_type;
 };
 
+/**
+ * typedef FieldAccessor: a field property getter or setter function
+ * @obj: the object instance
+ * @v: the visitor that contains the property data
+ * @name: the name of the property
+ * @prop: Field property definition
+ * @errp: pointer to error information
+ */
+typedef void FieldAccessor(Object *obj, Visitor *v,
+   const char *name, Property *prop,
+   Error **errp);
+
 /**
  * struct PropertyInfo: information on a specific QOM property type
  */
@@ -71,16 +83,10 @@ struct PropertyInfo {
 /** @create: Optional callback for creation of property */
 ObjectProperty *(*create)(ObjectClass *oc, const char *name,
   Property *prop);
-/**
- * @get: Property getter.  The opaque parameter will point to
- *the  struct for the property.
- */
-ObjectPropertyAccessor *get;
-/**
- * @set: Property setter.  The opaque parameter will point to
- *the  struct for the property.
- */
-ObjectPropertyAccessor *set;
+/** @get: Property getter */
+FieldAccessor *get;
+/** @set: Property setter */
+FieldAccessor *set;
 /**
  * @release: Optional release function, called when the object
  * is destroyed
diff --git a/backends/tpm/tpm_util.c b/backends/tpm/tpm_util.c
index bb1ab34a75..e8837938e5 100644
--- a/backends/tpm/tpm_util.c
+++ b/backends/tpm/tpm_util.c
@@ -32,10 +32,10 @@
 
 /* tpm backend property */
 
-static void get_tpm(Object *obj, Visitor *v, const char *name, void *opaque,
-Error **errp)
+static void get_tpm(Object *obj, Visitor *v, const char *name,
+Property *prop, Error **errp)
 {
-TPMBackend **be = object_field_prop_ptr(obj, opaque);
+TPMBackend **be = object_field_prop_ptr(obj, prop);
 char *p;
 
 p = g_strdup(*be ? (*be)->id : "");
@@ -43,10 +43,9 @@ static void get_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 g_free(p);
 }
 
-static void set_tpm(Object *obj, Visitor *v,

[PATCH v2 36/44] qdev: Rename qdev_get_prop_ptr() to object_field_prop_ptr()

[Eduardo Habkost]

The function will be moved to common QOM code, as it is not
specific to TYPE_DEVICE anymore.

Signed-off-by: Eduardo Habkost 
---
Changes v1 -> v2:
* Rename to object_field_prop_ptr() instead of object_static_prop_ptr()
---
Cc: Stefan Berger 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: Eduardo Habkost 
Cc: Cornelia Huck 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Thomas Huth 
Cc: Matthew Rosato 
Cc: Alex Williamson 
Cc: qemu-de...@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
---
 include/hw/qdev-properties.h |  2 +-
 backends/tpm/tpm_util.c  |  6 ++--
 hw/block/xen-block.c |  4 +--
 hw/core/qdev-properties-system.c | 50 +-
 hw/core/qdev-properties.c| 60 
 hw/s390x/css.c   |  4 +--
 hw/s390x/s390-pci-bus.c  |  4 +--
 hw/vfio/pci-quirks.c |  4 +--
 8 files changed, 67 insertions(+), 67 deletions(-)

diff --git a/include/hw/qdev-properties.h b/include/hw/qdev-properties.h
index 7f8d5fc206..2bec65c8e5 100644
--- a/include/hw/qdev-properties.h
+++ b/include/hw/qdev-properties.h
@@ -223,7 +223,7 @@ void qdev_prop_set_macaddr(DeviceState *dev, const char 
*name,
const uint8_t *value);
 void qdev_prop_set_enum(DeviceState *dev, const char *name, int value);
 
-void *qdev_get_prop_ptr(Object *obj, Property *prop);
+void *object_field_prop_ptr(Object *obj, Property *prop);
 
 void qdev_prop_register_global(GlobalProperty *prop);
 const GlobalProperty *qdev_find_global_prop(Object *obj,
diff --git a/backends/tpm/tpm_util.c b/backends/tpm/tpm_util.c
index 0b07cf55ea..bb1ab34a75 100644
--- a/backends/tpm/tpm_util.c
+++ b/backends/tpm/tpm_util.c
@@ -35,7 +35,7 @@
 static void get_tpm(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-TPMBackend **be = qdev_get_prop_ptr(obj, opaque);
+TPMBackend **be = object_field_prop_ptr(obj, opaque);
 char *p;
 
 p = g_strdup(*be ? (*be)->id : "");
@@ -47,7 +47,7 @@ static void set_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 Error **errp)
 {
 Property *prop = opaque;
-TPMBackend *s, **be = qdev_get_prop_ptr(obj, prop);
+TPMBackend *s, **be = object_field_prop_ptr(obj, prop);
 char *str;
 
 if (!visit_type_str(v, name, , errp)) {
@@ -67,7 +67,7 @@ static void set_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 static void release_tpm(Object *obj, const char *name, void *opaque)
 {
 Property *prop = opaque;
-TPMBackend **be = qdev_get_prop_ptr(obj, prop);
+TPMBackend **be = object_field_prop_ptr(obj, prop);
 
 if (*be) {
 tpm_backend_reset(*be);
diff --git a/hw/block/xen-block.c b/hw/block/xen-block.c
index bd1aef63a7..718d886e5c 100644
--- a/hw/block/xen-block.c
+++ b/hw/block/xen-block.c
@@ -336,7 +336,7 @@ static void xen_block_get_vdev(Object *obj, Visitor *v, 
const char *name,
void *opaque, Error **errp)
 {
 Property *prop = opaque;
-XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
+XenBlockVdev *vdev = object_field_prop_ptr(obj, prop);
 char *str;
 
 switch (vdev->type) {
@@ -396,7 +396,7 @@ static void xen_block_set_vdev(Object *obj, Visitor *v, 
const char *name,
void *opaque, Error **errp)
 {
 Property *prop = opaque;
-XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
+XenBlockVdev *vdev = object_field_prop_ptr(obj, prop);
 char *str, *p;
 const char *end;
 
diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index 96a0bc5109..8781b856d3 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -62,7 +62,7 @@ static void get_drive(Object *obj, Visitor *v, const char 
*name, void *opaque,
   Error **errp)
 {
 Property *prop = opaque;
-void **ptr = qdev_get_prop_ptr(obj, prop);
+void **ptr = object_field_prop_ptr(obj, prop);
 const char *value;
 char *p;
 
@@ -88,7 +88,7 @@ static void set_drive_helper(Object *obj, Visitor *v, const 
char *name,
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-void **ptr = qdev_get_prop_ptr(obj, prop);
+void **ptr = object_field_prop_ptr(obj, prop);
 char *str;
 BlockBackend *blk;
 bool blk_created = false;
@@ -181,7 +181,7 @@ static void release_drive(Object *obj, const char *name, 
void *opaque)
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-BlockBackend **ptr = qdev_get_prop_ptr(obj, prop);
+BlockBackend **ptr = object_field_prop_ptr(obj, prop);
 
 if (*ptr) {
 AioContext *ctx = blk_get_aio_context(*ptr);
@@ -214,7

[PATCH v2 22/44] qdev: Move dev->realized check to qdev_property_set()

[Eduardo Habkost]

Every single qdev property setter function manually checks
dev->realized.  We can just check dev->realized inside
qdev_property_set() instead.

The check is being added as a separate function
(qdev_prop_allow_set()) because it will become a callback later.

Signed-off-by: Eduardo Habkost 
---
Changes v1 -> v2:
* Removed unused variable at xen_block_set_vdev()
* Redone patch after changes in the previous patches in the
  series
---
Cc: Stefan Berger 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: Eduardo Habkost 
Cc: Cornelia Huck 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Thomas Huth 
Cc: Matthew Rosato 
Cc: Alex Williamson 
Cc: Mark Cave-Ayland 
Cc: Artyom Tarasenko 
Cc: qemu-de...@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
---
 backends/tpm/tpm_util.c  |   6 --
 hw/block/xen-block.c |   6 --
 hw/core/qdev-properties-system.c |  70 --
 hw/core/qdev-properties.c| 100 ++-
 hw/s390x/css.c   |   6 --
 hw/s390x/s390-pci-bus.c  |   6 --
 hw/vfio/pci-quirks.c |   6 --
 target/sparc/cpu.c   |   6 --
 8 files changed, 18 insertions(+), 188 deletions(-)

diff --git a/backends/tpm/tpm_util.c b/backends/tpm/tpm_util.c
index dba2f6b04a..0b07cf55ea 100644
--- a/backends/tpm/tpm_util.c
+++ b/backends/tpm/tpm_util.c
@@ -46,16 +46,10 @@ static void get_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 static void set_tpm(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 TPMBackend *s, **be = qdev_get_prop_ptr(obj, prop);
 char *str;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
diff --git a/hw/block/xen-block.c b/hw/block/xen-block.c
index 905e4acd97..bd1aef63a7 100644
--- a/hw/block/xen-block.c
+++ b/hw/block/xen-block.c
@@ -395,17 +395,11 @@ static int vbd_name_to_disk(const char *name, const char 
**endp,
 static void xen_block_set_vdev(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
 char *str, *p;
 const char *end;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index 202abd0e4b..0d3e57bba0 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -94,11 +94,6 @@ static void set_drive_helper(Object *obj, Visitor *v, const 
char *name,
 bool blk_created = false;
 int ret;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
@@ -230,17 +225,11 @@ static void get_chr(Object *obj, Visitor *v, const char 
*name, void *opaque,
 static void set_chr(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 CharBackend *be = qdev_get_prop_ptr(obj, prop);
 Chardev *s;
 char *str;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
@@ -311,18 +300,12 @@ static void get_mac(Object *obj, Visitor *v, const char 
*name, void *opaque,
 static void set_mac(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 MACAddr *mac = qdev_get_prop_ptr(obj, prop);
 int i, pos;
 char *str;
 const char *p;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
@@ -390,7 +373,6 @@ static void get_netdev(Object *obj, Visitor *v, const char 
*name,
 static void set_netdev(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 NICPeers *peers_ptr = qdev_get_prop_ptr(obj, prop);
 NetClientState **ncs = peers_ptr->ncs;
@@ -398,11 +380,6 @@ static void set_netdev(Object *obj, Visitor *v, const char 
*name,
 int queues, err = 0, i = 0;
 char *str;
 
-if (dev->realized) {

[PATCH v2 09/44] qdev: Make qdev_get_prop_ptr() get Object* arg

[Eduardo Habkost]

Make the code more generic and not specific to TYPE_DEVICE.

Reviewed-by: Marc-André Lureau 
Signed-off-by: Eduardo Habkost 
---
Changes v1 -> v2:
- Fix build error with CONFIG_XEN
  I took the liberty of keeping the Reviewed-by line from
  Marc-André as the build fix is a trivial one line change
---
Cc: Stefan Berger 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: Eduardo Habkost 
Cc: Cornelia Huck 
Cc: Thomas Huth 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Matthew Rosato 
Cc: Alex Williamson 
Cc: qemu-de...@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
---
 include/hw/qdev-properties.h |  2 +-
 backends/tpm/tpm_util.c  |  8 ++--
 hw/block/xen-block.c |  5 +-
 hw/core/qdev-properties-system.c | 57 +-
 hw/core/qdev-properties.c| 82 +---
 hw/s390x/css.c   |  5 +-
 hw/s390x/s390-pci-bus.c  |  4 +-
 hw/vfio/pci-quirks.c |  5 +-
 8 files changed, 68 insertions(+), 100 deletions(-)

diff --git a/include/hw/qdev-properties.h b/include/hw/qdev-properties.h
index 0ea822e6a7..0b92cfc761 100644
--- a/include/hw/qdev-properties.h
+++ b/include/hw/qdev-properties.h
@@ -302,7 +302,7 @@ void qdev_prop_set_macaddr(DeviceState *dev, const char 
*name,
const uint8_t *value);
 void qdev_prop_set_enum(DeviceState *dev, const char *name, int value);
 
-void *qdev_get_prop_ptr(DeviceState *dev, Property *prop);
+void *qdev_get_prop_ptr(Object *obj, Property *prop);
 
 void qdev_prop_register_global(GlobalProperty *prop);
 const GlobalProperty *qdev_find_global_prop(DeviceState *dev,
diff --git a/backends/tpm/tpm_util.c b/backends/tpm/tpm_util.c
index b58d298c1a..e91c21dd4a 100644
--- a/backends/tpm/tpm_util.c
+++ b/backends/tpm/tpm_util.c
@@ -35,8 +35,7 @@
 static void get_tpm(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
-TPMBackend **be = qdev_get_prop_ptr(dev, opaque);
+TPMBackend **be = qdev_get_prop_ptr(obj, opaque);
 char *p;
 
 p = g_strdup(*be ? (*be)->id : "");
@@ -49,7 +48,7 @@ static void set_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-TPMBackend *s, **be = qdev_get_prop_ptr(dev, prop);
+TPMBackend *s, **be = qdev_get_prop_ptr(obj, prop);
 char *str;
 
 if (dev->realized) {
@@ -73,9 +72,8 @@ static void set_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 
 static void release_tpm(Object *obj, const char *name, void *opaque)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-TPMBackend **be = qdev_get_prop_ptr(dev, prop);
+TPMBackend **be = qdev_get_prop_ptr(obj, prop);
 
 if (*be) {
 tpm_backend_reset(*be);
diff --git a/hw/block/xen-block.c b/hw/block/xen-block.c
index 8a7a3f5452..905e4acd97 100644
--- a/hw/block/xen-block.c
+++ b/hw/block/xen-block.c
@@ -335,9 +335,8 @@ static char *disk_to_vbd_name(unsigned int disk)
 static void xen_block_get_vdev(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-XenBlockVdev *vdev = qdev_get_prop_ptr(dev, prop);
+XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
 char *str;
 
 switch (vdev->type) {
@@ -398,7 +397,7 @@ static void xen_block_set_vdev(Object *obj, Visitor *v, 
const char *name,
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-XenBlockVdev *vdev = qdev_get_prop_ptr(dev, prop);
+XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
 char *str, *p;
 const char *end;
 
diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index d0fb063a49..c8c73c371b 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -59,9 +59,8 @@ static bool check_prop_still_unset(DeviceState *dev, const 
char *name,
 static void get_drive(Object *obj, Visitor *v, const char *name, void *opaque,
   Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-void **ptr = qdev_get_prop_ptr(dev, prop);
+void **ptr = qdev_get_prop_ptr(obj, prop);
 const char *value;
 char *p;
 
@@ -87,7 +86,7 @@ static void set_drive_helper(Object *obj, Visitor *v, const 
char *name,
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-void **ptr = qdev_get_prop_ptr(dev, prop);
+void **ptr = qdev_get_prop_ptr(obj, prop);
 char *str;
 BlockBackend *blk;
 bool blk_created = false;
@@ -185,7 +184,7 @@ static void release_drive(Object *obj, const char *name, 
void *

--enable-xen on gitlab CI? (was Re: [PATCH 09/36] qdev: Make qdev_get_prop_ptr() get Object* arg)

[Eduardo Habkost]

On Fri, Oct 30, 2020 at 11:29:25AM +0400, Marc-André Lureau wrote:
> On Fri, Oct 30, 2020 at 2:07 AM Eduardo Habkost  wrote:
> 
> > Make the code more generic and not specific to TYPE_DEVICE.
> >
> > Signed-off-by: Eduardo Habkost 
> >
> 
> Nice cleanup!, but fails to build atm
> 
> ../hw/block/xen-block.c:403:9: error: ‘dev’ undeclared (first use in this
> function); did you mean ‘vdev’?
>   403 | if (dev->realized) {

Thanks for catching it!

What is necessary to make sure we have a CONFIG_XEN=y job in
gitlab CI?  Maybe just including xen-devel in some of the
container images is enough?

-- 
Eduardo

[PATCH 14/36] qdev: Move dev->realized check to qdev_property_set()

[Eduardo Habkost]

Every single qdev property setter function manually checks
dev->realized.  We can just check dev->realized inside
qdev_property_set() instead.

The check is being added as a separate function
(qdev_prop_allow_set()) because it will become a callback later.

Signed-off-by: Eduardo Habkost 
---
Cc: Stefan Berger 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: Eduardo Habkost 
Cc: Cornelia Huck 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Thomas Huth 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Matthew Rosato 
Cc: Alex Williamson 
Cc: Mark Cave-Ayland 
Cc: Artyom Tarasenko 
Cc: qemu-de...@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
---
 backends/tpm/tpm_util.c  |   6 --
 hw/block/xen-block.c |   5 --
 hw/core/qdev-properties-system.c |  64 ---
 hw/core/qdev-properties.c| 106 ++-
 hw/s390x/css.c   |   6 --
 hw/s390x/s390-pci-bus.c  |   6 --
 hw/vfio/pci-quirks.c |   6 --
 target/sparc/cpu.c   |   6 --
 8 files changed, 18 insertions(+), 187 deletions(-)

diff --git a/backends/tpm/tpm_util.c b/backends/tpm/tpm_util.c
index e91c21dd4a..042cacfcca 100644
--- a/backends/tpm/tpm_util.c
+++ b/backends/tpm/tpm_util.c
@@ -46,16 +46,10 @@ static void get_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 static void set_tpm(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 TPMBackend *s, **be = qdev_get_prop_ptr(obj, prop);
 char *str;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
diff --git a/hw/block/xen-block.c b/hw/block/xen-block.c
index 1ba9981c08..bd1aef63a7 100644
--- a/hw/block/xen-block.c
+++ b/hw/block/xen-block.c
@@ -400,11 +400,6 @@ static void xen_block_set_vdev(Object *obj, Visitor *v, 
const char *name,
 char *str, *p;
 const char *end;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index fca1b694ca..60a45f5620 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -92,11 +92,6 @@ static void set_drive_helper(Object *obj, Visitor *v, const 
char *name,
 bool blk_created = false;
 int ret;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
@@ -228,17 +223,11 @@ static void get_chr(Object *obj, Visitor *v, const char 
*name, void *opaque,
 static void set_chr(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 CharBackend *be = qdev_get_prop_ptr(obj, prop);
 Chardev *s;
 char *str;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
@@ -309,18 +298,12 @@ static void get_mac(Object *obj, Visitor *v, const char 
*name, void *opaque,
 static void set_mac(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 MACAddr *mac = qdev_get_prop_ptr(obj, prop);
 int i, pos;
 char *str;
 const char *p;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
@@ -388,7 +371,6 @@ static void get_netdev(Object *obj, Visitor *v, const char 
*name,
 static void set_netdev(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
 NICPeers *peers_ptr = qdev_get_prop_ptr(obj, prop);
 NetClientState **ncs = peers_ptr->ncs;
@@ -396,11 +378,6 @@ static void set_netdev(Object *obj, Visitor *v, const char 
*name,
 int queues, err = 0, i = 0;
 char *str;
 
-if (dev->realized) {
-qdev_prop_set_after_realize(dev, name, errp);
-return;
-}
-
 if (!visit_type_str(v, name, , errp)) {
 return;
 }
@@ -467,18 +444,12 @@ static void get_audiodev(Object *obj, Visitor *v, const 
char* name,
 static void set_audiodev(Object *obj, Visitor *v, const char* name,
  void *opaque, Error **errp)
 {
-DeviceState *dev = DEVI

[PATCH 25/36] qdev: Rename qdev_get_prop_ptr() to object_static_prop_ptr()

[Eduardo Habkost]

The function will be moved to common QOM code, as it is not
specific to TYPE_DEVICE anymore.

Signed-off-by: Eduardo Habkost 
---
Cc: Stefan Berger 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: Eduardo Habkost 
Cc: Cornelia Huck 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Thomas Huth 
Cc: Matthew Rosato 
Cc: Alex Williamson 
Cc: qemu-de...@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
---
 include/hw/qdev-properties.h |  2 +-
 backends/tpm/tpm_util.c  |  6 +--
 hw/block/xen-block.c |  4 +-
 hw/core/qdev-properties-system.c | 46 +++
 hw/core/qdev-properties.c| 64 
 hw/s390x/css.c   |  4 +-
 hw/s390x/s390-pci-bus.c  |  4 +-
 hw/vfio/pci-quirks.c |  4 +-
 8 files changed, 67 insertions(+), 67 deletions(-)

diff --git a/include/hw/qdev-properties.h b/include/hw/qdev-properties.h
index 0acc92ae2b..4146dac281 100644
--- a/include/hw/qdev-properties.h
+++ b/include/hw/qdev-properties.h
@@ -332,7 +332,7 @@ void qdev_prop_set_macaddr(DeviceState *dev, const char 
*name,
const uint8_t *value);
 void qdev_prop_set_enum(DeviceState *dev, const char *name, int value);
 
-void *qdev_get_prop_ptr(Object *obj, Property *prop);
+void *object_static_prop_ptr(Object *obj, Property *prop);
 
 void qdev_prop_register_global(GlobalProperty *prop);
 const GlobalProperty *qdev_find_global_prop(Object *obj,
diff --git a/backends/tpm/tpm_util.c b/backends/tpm/tpm_util.c
index 042cacfcca..2b5f788861 100644
--- a/backends/tpm/tpm_util.c
+++ b/backends/tpm/tpm_util.c
@@ -35,7 +35,7 @@
 static void get_tpm(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-TPMBackend **be = qdev_get_prop_ptr(obj, opaque);
+TPMBackend **be = object_static_prop_ptr(obj, opaque);
 char *p;
 
 p = g_strdup(*be ? (*be)->id : "");
@@ -47,7 +47,7 @@ static void set_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 Error **errp)
 {
 Property *prop = opaque;
-TPMBackend *s, **be = qdev_get_prop_ptr(obj, prop);
+TPMBackend *s, **be = object_static_prop_ptr(obj, prop);
 char *str;
 
 if (!visit_type_str(v, name, , errp)) {
@@ -67,7 +67,7 @@ static void set_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 static void release_tpm(Object *obj, const char *name, void *opaque)
 {
 Property *prop = opaque;
-TPMBackend **be = qdev_get_prop_ptr(obj, prop);
+TPMBackend **be = object_static_prop_ptr(obj, prop);
 
 if (*be) {
 tpm_backend_reset(*be);
diff --git a/hw/block/xen-block.c b/hw/block/xen-block.c
index bd1aef63a7..20985c465a 100644
--- a/hw/block/xen-block.c
+++ b/hw/block/xen-block.c
@@ -336,7 +336,7 @@ static void xen_block_get_vdev(Object *obj, Visitor *v, 
const char *name,
void *opaque, Error **errp)
 {
 Property *prop = opaque;
-XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
+XenBlockVdev *vdev = object_static_prop_ptr(obj, prop);
 char *str;
 
 switch (vdev->type) {
@@ -396,7 +396,7 @@ static void xen_block_set_vdev(Object *obj, Visitor *v, 
const char *name,
void *opaque, Error **errp)
 {
 Property *prop = opaque;
-XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
+XenBlockVdev *vdev = object_static_prop_ptr(obj, prop);
 char *str, *p;
 const char *end;
 
diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index d9355053d2..448d77ecab 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -60,7 +60,7 @@ static void get_drive(Object *obj, Visitor *v, const char 
*name, void *opaque,
   Error **errp)
 {
 Property *prop = opaque;
-void **ptr = qdev_get_prop_ptr(obj, prop);
+void **ptr = object_static_prop_ptr(obj, prop);
 const char *value;
 char *p;
 
@@ -86,7 +86,7 @@ static void set_drive_helper(Object *obj, Visitor *v, const 
char *name,
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-void **ptr = qdev_get_prop_ptr(obj, prop);
+void **ptr = object_static_prop_ptr(obj, prop);
 char *str;
 BlockBackend *blk;
 bool blk_created = false;
@@ -179,7 +179,7 @@ static void release_drive(Object *obj, const char *name, 
void *opaque)
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-BlockBackend **ptr = qdev_get_prop_ptr(obj, prop);
+BlockBackend **ptr = object_static_prop_ptr(obj, prop);
 
 if (*ptr) {
 AioContext *ctx = blk_get_aio_context(*ptr);
@@ -212,7 +212,7 @@ const PropertyInfo qdev_prop_drive_iothread = {
 static void get_chr(Object *obj, Visitor 

[PATCH 09/36] qdev: Make qdev_get_prop_ptr() get Object* arg

[Eduardo Habkost]

Make the code more generic and not specific to TYPE_DEVICE.

Signed-off-by: Eduardo Habkost 
---
Cc: Stefan Berger 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: Eduardo Habkost 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Cornelia Huck 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Thomas Huth 
Cc: Matthew Rosato 
Cc: Alex Williamson 
Cc: qemu-de...@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-block@nongnu.org
Cc: qemu-s3...@nongnu.org
---
 include/hw/qdev-properties.h |  2 +-
 backends/tpm/tpm_util.c  |  8 ++--
 hw/block/xen-block.c |  6 +--
 hw/core/qdev-properties-system.c | 57 +-
 hw/core/qdev-properties.c| 82 +---
 hw/s390x/css.c   |  5 +-
 hw/s390x/s390-pci-bus.c  |  4 +-
 hw/vfio/pci-quirks.c |  5 +-
 8 files changed, 68 insertions(+), 101 deletions(-)

diff --git a/include/hw/qdev-properties.h b/include/hw/qdev-properties.h
index 0ea822e6a7..0b92cfc761 100644
--- a/include/hw/qdev-properties.h
+++ b/include/hw/qdev-properties.h
@@ -302,7 +302,7 @@ void qdev_prop_set_macaddr(DeviceState *dev, const char 
*name,
const uint8_t *value);
 void qdev_prop_set_enum(DeviceState *dev, const char *name, int value);
 
-void *qdev_get_prop_ptr(DeviceState *dev, Property *prop);
+void *qdev_get_prop_ptr(Object *obj, Property *prop);
 
 void qdev_prop_register_global(GlobalProperty *prop);
 const GlobalProperty *qdev_find_global_prop(DeviceState *dev,
diff --git a/backends/tpm/tpm_util.c b/backends/tpm/tpm_util.c
index b58d298c1a..e91c21dd4a 100644
--- a/backends/tpm/tpm_util.c
+++ b/backends/tpm/tpm_util.c
@@ -35,8 +35,7 @@
 static void get_tpm(Object *obj, Visitor *v, const char *name, void *opaque,
 Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
-TPMBackend **be = qdev_get_prop_ptr(dev, opaque);
+TPMBackend **be = qdev_get_prop_ptr(obj, opaque);
 char *p;
 
 p = g_strdup(*be ? (*be)->id : "");
@@ -49,7 +48,7 @@ static void set_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-TPMBackend *s, **be = qdev_get_prop_ptr(dev, prop);
+TPMBackend *s, **be = qdev_get_prop_ptr(obj, prop);
 char *str;
 
 if (dev->realized) {
@@ -73,9 +72,8 @@ static void set_tpm(Object *obj, Visitor *v, const char 
*name, void *opaque,
 
 static void release_tpm(Object *obj, const char *name, void *opaque)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-TPMBackend **be = qdev_get_prop_ptr(dev, prop);
+TPMBackend **be = qdev_get_prop_ptr(obj, prop);
 
 if (*be) {
 tpm_backend_reset(*be);
diff --git a/hw/block/xen-block.c b/hw/block/xen-block.c
index 8a7a3f5452..1ba9981c08 100644
--- a/hw/block/xen-block.c
+++ b/hw/block/xen-block.c
@@ -335,9 +335,8 @@ static char *disk_to_vbd_name(unsigned int disk)
 static void xen_block_get_vdev(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-XenBlockVdev *vdev = qdev_get_prop_ptr(dev, prop);
+XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
 char *str;
 
 switch (vdev->type) {
@@ -396,9 +395,8 @@ static int vbd_name_to_disk(const char *name, const char 
**endp,
 static void xen_block_set_vdev(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-XenBlockVdev *vdev = qdev_get_prop_ptr(dev, prop);
+XenBlockVdev *vdev = qdev_get_prop_ptr(obj, prop);
 char *str, *p;
 const char *end;
 
diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index d0fb063a49..c8c73c371b 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -59,9 +59,8 @@ static bool check_prop_still_unset(DeviceState *dev, const 
char *name,
 static void get_drive(Object *obj, Visitor *v, const char *name, void *opaque,
   Error **errp)
 {
-DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-void **ptr = qdev_get_prop_ptr(dev, prop);
+void **ptr = qdev_get_prop_ptr(obj, prop);
 const char *value;
 char *p;
 
@@ -87,7 +86,7 @@ static void set_drive_helper(Object *obj, Visitor *v, const 
char *name,
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = opaque;
-void **ptr = qdev_get_prop_ptr(dev, prop);
+void **ptr = qdev_get_prop_ptr(obj, prop);
 char *str;
 BlockBackend *blk;
 bool blk_created = false;
@@ -185,7 +184,7 @@ static void release_drive(Object *obj, const char *name, 
void *opaque)
 {
 DeviceState *dev = DEVICE(obj);
 Property *prop = op

Re: [PATCH v3 00/11] user-mode: Prune build dependencies (part 3)

[Eduardo Habkost]

On Wed, Sep 30, 2020 at 07:24:24PM +0200, Paolo Bonzini wrote:
> On 30/09/20 19:15, Eduardo Habkost wrote:
> > On Wed, Sep 30, 2020 at 06:49:38PM +0200, Philippe Mathieu-Daudé wrote:
> >> This is the third part of a series reducing user-mode
> >> dependencies. By stripping out unused code, the build
> >> and testing time is reduced (as is space used by objects).
> > I'm queueing patches 2-9 on machine-next.  Thanks!
> > 
> > Markus, Eric: I can merge the QAPI patches (1, 11) if I get an
> > Acked-by.
> > 
> > I'll send separate comments on patch 10.
> > 
> 
> 1-8 is fine, but I think 9-11 is too much complication (especially not
> really future-proof) for the benefit.

I'll dequeue patch 9 while this is discussed.

-- 
Eduardo

Re: [PATCH v3 10/11] target/i386: Restrict X86CPUFeatureWord to X86 targets

[Eduardo Habkost]

On Wed, Sep 30, 2020 at 06:49:48PM +0200, Philippe Mathieu-Daudé wrote:
> Only qemu-system-FOO and qemu-storage-daemon provide QMP
> monitors, therefore such declarations and definitions are
> irrelevant for user-mode emulation.
> 
> Restricting the x86-specific commands to machine-target.json
> pulls less QAPI-generated code into user-mode.
> 
> Add a stub to satisfy linking in user-mode:
> 
>   /usr/bin/ld: libqemu-i386-linux-user.fa.p/target_i386_cpu.c.o: in function 
> `x86_cpu_get_feature_words':
>   target/i386/cpu.c:4643: undefined reference to 
> `visit_type_X86CPUFeatureWordInfoList'
>   collect2: error: ld returned 1 exit status
>   make: *** [Makefile.ninja:1125: qemu-i386] Error 1
> 

If you don't want the QAPI definitions in user mode, there's no
reason to register the properties in user mode.  Wrapping #ifdef
around "feature-words" and "filtered-features" registration would
be simpler than adding a stub.

> Acked-by: Richard Henderson 
> Signed-off-by: Philippe Mathieu-Daudé 
> ---
> v3: Reworded + Meson rebase
> ---
>  qapi/machine-target.json   | 45 ++
>  qapi/machine.json  | 42 ---
>  target/i386/cpu.c  |  2 +-
>  target/i386/feature-stub.c | 23 +++
>  target/i386/meson.build|  1 +
>  5 files changed, 70 insertions(+), 43 deletions(-)
>  create mode 100644 target/i386/feature-stub.c
> 
> diff --git a/qapi/machine-target.json b/qapi/machine-target.json
> index 698850cc78..b4d769a53b 100644
> --- a/qapi/machine-target.json
> +++ b/qapi/machine-target.json
> @@ -4,6 +4,51 @@
>  # This work is licensed under the terms of the GNU GPL, version 2 or later.
>  # See the COPYING file in the top-level directory.
>  
> +##
> +# @X86CPURegister32:
> +#
> +# A X86 32-bit register
> +#
> +# Since: 1.5
> +##
> +{ 'enum': 'X86CPURegister32',
> +  'data': [ 'EAX', 'EBX', 'ECX', 'EDX', 'ESP', 'EBP', 'ESI', 'EDI' ],
> +  'if': 'defined(TARGET_I386)' }
> +
> +##
> +# @X86CPUFeatureWordInfo:
> +#
> +# Information about a X86 CPU feature word
> +#
> +# @cpuid-input-eax: Input EAX value for CPUID instruction for that feature 
> word
> +#
> +# @cpuid-input-ecx: Input ECX value for CPUID instruction for that
> +#   feature word
> +#
> +# @cpuid-register: Output register containing the feature bits
> +#
> +# @features: value of output register, containing the feature bits
> +#
> +# Since: 1.5
> +##
> +{ 'struct': 'X86CPUFeatureWordInfo',
> +  'data': { 'cpuid-input-eax': 'int',
> +'*cpuid-input-ecx': 'int',
> +'cpuid-register': 'X86CPURegister32',
> +'features': 'int' },
> +  'if': 'defined(TARGET_I386)' }
> +
> +##
> +# @DummyForceArrays:
> +#
> +# Not used by QMP; hack to let us use X86CPUFeatureWordInfoList internally
> +#
> +# Since: 2.5
> +##
> +{ 'struct': 'DummyForceArrays',
> +  'data': { 'unused': ['X86CPUFeatureWordInfo'] },
> +  'if': 'defined(TARGET_I386)' }
> +
>  ##
>  # @CpuModelInfo:
>  #
> diff --git a/qapi/machine.json b/qapi/machine.json
> index 72f014bb5b..cb878acdac 100644
> --- a/qapi/machine.json
> +++ b/qapi/machine.json
> @@ -544,48 +544,6 @@
> 'dst': 'uint16',
> 'val': 'uint8' }}
>  
> -##
> -# @X86CPURegister32:
> -#
> -# A X86 32-bit register
> -#
> -# Since: 1.5
> -##
> -{ 'enum': 'X86CPURegister32',
> -  'data': [ 'EAX', 'EBX', 'ECX', 'EDX', 'ESP', 'EBP', 'ESI', 'EDI' ] }
> -
> -##
> -# @X86CPUFeatureWordInfo:
> -#
> -# Information about a X86 CPU feature word
> -#
> -# @cpuid-input-eax: Input EAX value for CPUID instruction for that feature 
> word
> -#
> -# @cpuid-input-ecx: Input ECX value for CPUID instruction for that
> -#   feature word
> -#
> -# @cpuid-register: Output register containing the feature bits
> -#
> -# @features: value of output register, containing the feature bits
> -#
> -# Since: 1.5
> -##
> -{ 'struct': 'X86CPUFeatureWordInfo',
> -  'data': { 'cpuid-input-eax': 'int',
> -'*cpuid-input-ecx': 'int',
> -'cpuid-register': 'X86CPURegister32',
> -'features': 'int' } }
> -
> -##
> -# @DummyForceArrays:
> -#
> -# Not used by QMP; hack to let us use X86CPUFeatureWordInfoList internally
> -#
> -# Since: 2.5
> -##
> -{ 'struct': 'DummyForceArrays',
> -  'data': { 'unused': ['X86CPUFeatureWordInfo'] } }
> -
>  ##
>  # @NumaCpuOptions:
>  #
> diff --git a/target/i386/cpu.c b/target/i386/cpu.c
> index 3ffd877dd5..d45fa217cc 100644
> --- a/target/i386/cpu.c
> +++ b/target/i386/cpu.c
> @@ -38,7 +38,7 @@
>  #include "qemu/option.h"
>  #include "qemu/config-file.h"
>  #include "qapi/error.h"
> -#include "qapi/qapi-visit-machine.h"
> +#include "qapi/qapi-visit-machine-target.h"
>  #include "qapi/qapi-visit-run-state.h"
>  #include "qapi/qmp/qdict.h"
>  #include "qapi/qmp/qerror.h"
> diff --git a/target/i386/feature-stub.c b/target/i386/feature-stub.c
> new file mode 100644
> index 00..787c3c7fa1
> --- /dev/null
> +++ b/target/i386/feature-stub.c
> @@ -0,0 

Re: [PATCH v3 00/11] user-mode: Prune build dependencies (part 3)

[Eduardo Habkost]

On Wed, Sep 30, 2020 at 06:49:38PM +0200, Philippe Mathieu-Daudé wrote:
> This is the third part of a series reducing user-mode
> dependencies. By stripping out unused code, the build
> and testing time is reduced (as is space used by objects).

I'm queueing patches 2-9 on machine-next.  Thanks!

Markus, Eric: I can merge the QAPI patches (1, 11) if I get an
Acked-by.

I'll send separate comments on patch 10.

> 
> Part 3:
> - Extract code not related to user-mode from hw/core/qdev-properties.c
> - Reduce user-mode QAPI generated files
> 
> Since v2:
> - Fixed UuidInfo placed in incorrect json
> - Rebased on Meson
> - Include X86CPUFeatureWord unmerged from part 2
> 
> Since v1:
> - Addressed Richard and Paolo review comments
> 
> Patches missing review: QAPI ones :)
> - #1  'qapi: Restrict query-uuid command to block code'
> - #11 'qapi: Restrict code generated for user-mode'
> 
> Green CI: https://gitlab.com/philmd/qemu/-/pipelines/196505787
> 
> v2: https://www.mail-archive.com/qemu-devel@nongnu.org/msg688879.html
> v1: https://www.mail-archive.com/qemu-devel@nongnu.org/msg688486.html
> 
> Philippe Mathieu-Daudé (11):
>   qapi: Restrict query-uuid command to block code
>   hw/core/qdev-properties: Use qemu_strtol() in set_mac() handler
>   hw/core/qdev-properties: Use qemu_strtoul() in set_pci_host_devaddr()
>   hw/core/qdev-properties: Fix code style
>   hw/core/qdev-properties: Export enum-related functions
>   hw/core/qdev-properties: Export qdev_prop_enum
>   hw/core/qdev-properties: Export some integer-related functions
>   hw/core/qdev-properties: Extract system-mode specific properties
>   hw/core: Add qdev stub for user-mode
>   target/i386: Restrict X86CPUFeatureWord to X86 targets
>   qapi: Restrict code generated for user-mode
> 
>  qapi/block.json  |  30 ++
>  qapi/machine-target.json |  45 ++
>  qapi/machine.json|  72 ---
>  hw/core/qdev-prop-internal.h |  30 ++
>  include/hw/qdev-properties.h |   1 +
>  block/iscsi.c|   2 +-
>  hw/core/qdev-properties-system.c | 687 -
>  hw/core/qdev-properties.c| 735 ++-
>  stubs/qdev-system.c  |  24 +
>  stubs/uuid.c |   2 +-
>  target/i386/cpu.c|   2 +-
>  target/i386/feature-stub.c   |  23 +
>  qapi/meson.build |  51 ++-
>  stubs/meson.build|   5 +-
>  target/i386/meson.build  |   1 +
>  15 files changed, 915 insertions(+), 795 deletions(-)
>  create mode 100644 hw/core/qdev-prop-internal.h
>  create mode 100644 stubs/qdev-system.c
>  create mode 100644 target/i386/feature-stub.c
> 
> -- 
> 2.26.2
> 

-- 
Eduardo

Re: [PATCH 0/5] qom: Convert more declarations to OBJECT_DECLARE*

[Eduardo Habkost]

On Wed, Sep 16, 2020 at 02:25:14PM -0400, Eduardo Habkost wrote:
> This converts many QOM types to use OBJECT_DECLARE* instead of
> manually using DECLARE*_CHECKER*.
> 
> Before doing that, I'm simplifying the OBJECT_DECLARE* API to
> make it easier to use and more difficult to misuse.  The
> module_obj_name and ParentClassType parameters were removed
> because they are not needed.

I'm queueing this on machine-next.

> 
> Eduardo Habkost (5):
>   scripts/codeconverter: Update to latest version
>   qom: Remove ParentClassType argument from OBJECT_DECLARE_SIMPLE_TYPE
>   qom: Remove module_obj_name parameter from OBJECT_DECLARE* macros
>   [automated] Use OBJECT_DECLARE_TYPE when possible
>   [automated] Use OBJECT_DECLARE_SIMPLE_TYPE when possible
> 
>  hw/9pfs/virtio-9p.h   |   4 +-
>  hw/audio/intel-hda.h  |   6 +-
>  hw/display/ati_int.h  |   4 +-
>  hw/display/qxl.h  |   4 +-
>  hw/display/virtio-vga.h   |   2 +-
>  hw/i386/amd_iommu.h   |   4 +-
>  hw/misc/tmp105.h  |   4 +-
>  hw/net/fsl_etsec/etsec.h  |   3 +-
>  hw/net/tulip.h|   4 +-
>  hw/ppc/e500-ccsr.h|   4 +-
>  hw/ppc/e500.h |   5 +-
>  hw/ppc/mac.h  |   4 +-
>  hw/s390x/ccw-device.h |   4 +-
>  hw/s390x/ipl.h|   4 +-
>  hw/s390x/s390-pci-bus.h   |  16 +-
>  hw/s390x/virtio-ccw.h |  57 +-
>  hw/usb/ccid.h |   5 +-
>  hw/usb/hcd-dwc2.h |   3 +-
>  hw/usb/hcd-ehci.h |  13 +-
>  hw/usb/hcd-ohci.h |   4 +-
>  hw/usb/hcd-xhci.h |   4 +-
>  hw/vfio/pci.h |   4 +-
>  hw/virtio/virtio-pci.h|   5 +-
>  hw/xen/xen_pt.h   |   4 +-
>  include/authz/base.h  |   2 +-
>  include/authz/list.h  |   4 +-
>  include/authz/listfile.h  |   4 +-
>  include/authz/pamacct.h   |   4 +-
>  include/authz/simple.h|   4 +-
>  include/block/throttle-groups.h   |   4 +-
>  include/chardev/char.h|   4 +-
>  include/crypto/secret_common.h|   2 +-
>  include/crypto/secret_keyring.h   |   4 +-
>  include/hw/acpi/generic_event_device.h|   4 +-
>  include/hw/acpi/vmgenid.h |   4 +-
>  include/hw/adc/stm32f2xx_adc.h|   4 +-
>  include/hw/arm/allwinner-a10.h|   4 +-
>  include/hw/arm/allwinner-h3.h |   4 +-
>  include/hw/arm/armsse.h   |   2 +-
>  include/hw/arm/armv7m.h   |   8 +-
>  include/hw/arm/aspeed_soc.h   |   5 +-
>  include/hw/arm/bcm2835_peripherals.h  |   4 +-
>  include/hw/arm/bcm2836.h  |   5 +-
>  include/hw/arm/digic.h|   4 +-
>  include/hw/arm/exynos4210.h   |   4 +-
>  include/hw/arm/fsl-imx25.h|   4 +-
>  include/hw/arm/fsl-imx31.h|   4 +-
>  include/hw/arm/fsl-imx6.h |   4 +-
>  include/hw/arm/fsl-imx6ul.h   |   4 +-
>  include/hw/arm/fsl-imx7.h |   4 +-
>  include/hw/arm/msf2-soc.h |   4 +-
>  include/hw/arm/nrf51_soc.h|   4 +-
>  include/hw/arm/omap.h |   4 +-
>  include/hw/arm/pxa.h  |  15 +-
>  include/hw/arm/smmu-common.h  |   5 +-
>  include/hw/arm/smmuv3.h   |   5 +-
>  include/hw/arm/stm32f205_soc.h|   4 +-
>  include/hw/arm/stm32f405_soc.h|   4 +-
>  include/hw/arm/virt.h |   5 +-
>  include/hw/arm/xlnx-versal.h  |   4 +-
>  include/hw/arm/xlnx-zynqmp.h  |   4 +-
>  include/hw/block/flash.h  |   8 +-
>  include/hw/block/swim.h   |  12 +-
>  include/hw/boards.h   |   3 +-
>  include/hw/char/avr_usart.h   |   4 +-
>  include/hw/char/bcm2835_aux.h |   4 +-
>  include/hw/char/cadence_uart.h|   4 +-
>  include/hw/char/cmsdk-apb-uart.h  |   4 +-
>  include/hw/char/digic-uart.h   

[PATCH 4/5] [automated] Use OBJECT_DECLARE_TYPE when possible

[Eduardo Habkost]

This converts existing DECLARE_OBJ_CHECKERS usage to
OBJECT_DECLARE_TYPE when possible.

 $ ./scripts/codeconverter/converter.py -i \
   --pattern=AddObjectDeclareType $(git grep -l '' -- '*.[ch]')

Signed-off-by: Eduardo Habkost 
---
Cc: Peter Maydell 
Cc: Andrzej Zaborowski 
Cc: Alistair Francis 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: Mark Cave-Ayland 
Cc: David Gibson 
Cc: Richard Henderson 
Cc: David Hildenbrand 
Cc: Cornelia Huck 
Cc: Thomas Huth 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: "Michael S. Tsirkin" 
Cc: Paolo Bonzini 
Cc: Fam Zheng 
Cc: Dmitry Fleytman 
Cc: Gerd Hoffmann 
Cc: "Marc-André Lureau" 
Cc: "Cédric Le Goater" 
Cc: Andrew Jeffery 
Cc: Joel Stanley 
Cc: Andrew Baumann 
Cc: "Philippe Mathieu-Daudé" 
Cc: Eric Auger 
Cc: Eduardo Habkost 
Cc: Marcel Apfelbaum 
Cc: Laurent Vivier 
Cc: Sergio Lopez 
Cc: John Snow 
Cc: Xiao Guangrong 
Cc: Peter Chubb 
Cc: "Daniel P. Berrangé" 
Cc: Beniamino Galvani 
Cc: "Edgar E. Iglesias" 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Jason Wang 
Cc: qemu-...@nongnu.org
Cc: qemu-de...@nongnu.org
Cc: qemu-block@nongnu.org
Cc: qemu-...@nongnu.org
Cc: qemu-s3...@nongnu.org
Cc: xen-de...@lists.xenproject.org
---
 hw/ppc/e500.h | 5 +
 hw/s390x/ccw-device.h | 4 +---
 hw/s390x/virtio-ccw.h | 5 +
 hw/usb/ccid.h | 5 +
 hw/usb/hcd-dwc2.h | 3 +--
 hw/usb/hcd-ehci.h | 5 +
 hw/virtio/virtio-pci.h| 5 +
 include/chardev/char.h| 4 +---
 include/hw/arm/aspeed_soc.h   | 5 +
 include/hw/arm/bcm2836.h  | 5 +
 include/hw/arm/smmu-common.h  | 5 +
 include/hw/arm/smmuv3.h   | 5 +
 include/hw/arm/virt.h | 5 +
 include/hw/boards.h   | 3 +--
 include/hw/display/macfb.h| 5 +
 include/hw/gpio/aspeed_gpio.h | 5 +
 include/hw/i2c/aspeed_i2c.h   | 5 +
 include/hw/i386/ioapic_internal.h | 5 +
 include/hw/i386/microvm.h | 5 +
 include/hw/i386/pc.h  | 4 +---
 include/hw/i386/x86-iommu.h   | 5 +
 include/hw/i386/x86.h | 5 +
 include/hw/ide/internal.h | 4 +---
 include/hw/input/adb.h| 4 +---
 include/hw/isa/i8259_internal.h   | 5 +
 include/hw/isa/isa.h  | 4 +---
 include/hw/mem/nvdimm.h   | 5 +
 include/hw/misc/aspeed_scu.h  | 5 +
 include/hw/misc/aspeed_sdmc.h | 5 +
 include/hw/misc/imx_ccm.h | 5 +
 include/hw/misc/mos6522.h | 5 +
 include/hw/pci-host/pnv_phb4.h| 5 +
 include/hw/pci/pci.h  | 4 +---
 include/hw/pci/pci_host.h | 4 +---
 include/hw/pcmcia.h   | 5 +
 include/hw/ppc/spapr.h| 5 +
 include/hw/qdev-core.h| 4 +---
 include/hw/rtc/allwinner-rtc.h| 5 +
 include/hw/s390x/3270-ccw.h   | 5 +
 include/hw/s390x/s390-virtio-ccw.h| 5 +
 include/hw/s390x/storage-attributes.h | 5 +
 include/hw/s390x/storage-keys.h   | 5 +
 include/hw/s390x/tod.h| 5 +
 include/hw/scsi/scsi.h| 4 +---
 include/hw/sd/allwinner-sdhost.h  | 5 +
 include/hw/sd/sd.h| 5 +
 include/hw/ssi/aspeed_smc.h   | 5 +
 include/hw/ssi/xilinx_spips.h | 4 +---
 include/hw/timer/aspeed_timer.h   | 5 +
 include/hw/timer/i8254.h  | 5 +
 include/hw/usb.h  | 4 +---
 include/hw/virtio/virtio.h| 4 +---
 include/hw/watchdog/wdt_aspeed.h  | 5 +
 include/hw/xen/xen-block.h| 4 +---
 include/hw/xen/xen-bus.h  | 4 +---
 include/net/can_host.h| 5 +
 include/net/filter.h  | 4 +---
 include/ui/console.h  | 4 +---
 hw/arm/mps2-tz.c  | 5 +
 hw/arm/mps2.c | 5 +
 hw/arm/musca.c| 5 +
 hw/arm/spitz.c| 5 +
 hw/arm/vexpress.c | 5 +
 hw/block/m25p80.c | 5 +
 hw/input/adb-kbd.c| 5 +
 hw/input/adb-mouse.c  | 5 +
 hw/misc/tmp421.c  | 5 +
 hw/scsi/scsi-disk.c   | 5 +
 hw/scsi/vmw_pvscsi.c  | 5 +
 69 files changed, 69 insertions(+), 255 deletions(-)

diff --git a/hw/ppc/e500.h b/hw/ppc/e500.h
index 63870751ff..1e5853b032 100644
--- a/hw/ppc/e500.h
+++ b/hw/ppc/e500.h
@@ -14,7 +14,6 @@ struct PPCE500MachineState {
  */
 PlatformBusDevice *pbus_dev;
 };
-typedef struct PPCE500MachineState PPCE500MachineState;
 
 struct PPCE500MachineClass {
 /*< private &

[PATCH 3/5] qom: Remove module_obj_name parameter from OBJECT_DECLARE* macros

[Eduardo Habkost]

One of the goals of having less boilerplate on QOM declarations
is to avoid human error.  Requiring an extra argument that is
never used is an opportunity for mistakes.

Remove the unused argument from OBJECT_DECLARE_TYPE and
OBJECT_DECLARE_SIMPLE_TYPE.

Coccinelle patch used to convert all users of the macros:

  @@
  declarer name OBJECT_DECLARE_TYPE;
  identifier InstanceType, ClassType, lowercase, UPPERCASE;
  @@
   OBJECT_DECLARE_TYPE(InstanceType, ClassType,
  -lowercase,
   UPPERCASE);

  @@
  declarer name OBJECT_DECLARE_SIMPLE_TYPE;
  identifier InstanceType, lowercase, UPPERCASE;
  @@
   OBJECT_DECLARE_SIMPLE_TYPE(InstanceType,
  -lowercase,
   UPPERCASE);

Signed-off-by: Eduardo Habkost 
---
Cc: "Marc-André Lureau" 
Cc: Gerd Hoffmann 
Cc: "Michael S. Tsirkin" 
Cc: "Daniel P. Berrangé" 
Cc: Peter Maydell 
Cc: Corey Minyard 
Cc: "Cédric Le Goater" 
Cc: David Gibson 
Cc: Cornelia Huck 
Cc: Thomas Huth 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: "Philippe Mathieu-Daudé" 
Cc: Alistair Francis 
Cc: David Hildenbrand 
Cc: Laurent Vivier 
Cc: Amit Shah 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Paolo Bonzini 
Cc: Eduardo Habkost 
Cc: Fam Zheng 
Cc: "Gonglei (Arei)" 
Cc: Igor Mammedov 
Cc: Stefan Berger 
Cc: Richard Henderson 
Cc: Michael Rolnik 
Cc: Sarah Harris 
Cc: "Edgar E. Iglesias" 
Cc: Michael Walle 
Cc: Aleksandar Markovic 
Cc: Aurelien Jarno 
Cc: Jiaxun Yang 
Cc: Aleksandar Rikalo 
Cc: Anthony Green 
Cc: Chris Wulff 
Cc: Marek Vasut 
Cc: Stafford Horne 
Cc: Palmer Dabbelt 
Cc: Sagar Karandikar 
Cc: Bastian Koppelmann 
Cc: Yoshinori Sato 
Cc: Mark Cave-Ayland 
Cc: Artyom Tarasenko 
Cc: Guan Xuetao 
Cc: Max Filippov 
Cc: qemu-de...@nongnu.org
Cc: qemu-...@nongnu.org
Cc: qemu-...@nongnu.org
Cc: qemu-s3...@nongnu.org
Cc: qemu-block@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-ri...@nongnu.org
---
 hw/audio/intel-hda.h| 2 +-
 hw/display/virtio-vga.h | 2 +-
 include/authz/base.h| 2 +-
 include/authz/list.h| 2 +-
 include/authz/listfile.h| 2 +-
 include/authz/pamacct.h | 2 +-
 include/authz/simple.h  | 2 +-
 include/crypto/secret_common.h  | 2 +-
 include/crypto/secret_keyring.h | 2 +-
 include/hw/arm/armsse.h | 2 +-
 include/hw/hyperv/vmbus.h   | 2 +-
 include/hw/i2c/i2c.h| 2 +-
 include/hw/i2c/smbus_slave.h| 2 +-
 include/hw/ipack/ipack.h| 2 +-
 include/hw/ipmi/ipmi.h  | 2 +-
 include/hw/mem/pc-dimm.h| 2 +-
 include/hw/ppc/pnv.h| 2 +-
 include/hw/ppc/pnv_core.h   | 2 +-
 include/hw/ppc/pnv_homer.h  | 2 +-
 include/hw/ppc/pnv_occ.h| 2 +-
 include/hw/ppc/pnv_psi.h| 2 +-
 include/hw/ppc/pnv_xive.h   | 2 +-
 include/hw/ppc/spapr_cpu_core.h | 2 +-
 include/hw/ppc/spapr_vio.h  | 2 +-
 include/hw/ppc/xics.h   | 2 +-
 include/hw/ppc/xive.h   | 2 +-
 include/hw/s390x/event-facility.h   | 2 +-
 include/hw/s390x/s390_flic.h| 2 +-
 include/hw/s390x/sclp.h | 2 +-
 include/hw/sd/sd.h  | 2 +-
 include/hw/ssi/ssi.h| 2 +-
 include/hw/sysbus.h | 2 +-
 include/hw/virtio/virtio-gpu.h  | 2 +-
 include/hw/virtio/virtio-input.h| 2 +-
 include/hw/virtio/virtio-mem.h  | 2 +-
 include/hw/virtio/virtio-pmem.h | 2 +-
 include/hw/virtio/virtio-serial.h   | 2 +-
 include/hw/xen/xen-bus.h| 2 +-
 include/io/channel.h| 2 +-
 include/io/dns-resolver.h   | 2 +-
 include/io/net-listener.h   | 2 +-
 include/qom/object.h| 6 ++
 include/scsi/pr-manager.h   | 2 +-
 include/sysemu/cryptodev.h  | 2 +-
 include/sysemu/hostmem.h| 2 +-
 include/sysemu/rng.h| 2 +-
 include/sysemu/tpm_backend.h| 2 +-
 include/sysemu/vhost-user-backend.h | 2 +-
 target/alpha/cpu-qom.h  | 2 +-
 target/arm/cpu-qom.h| 2 +-
 target/avr/cpu-qom.h| 2 +-
 target/cris/cpu-qom.h   | 2 +-
 target/hppa/cpu-qom.h   | 2 +-
 target/i386/cpu-qom.h   | 2 +-
 target/lm32/cpu-qom.h   | 2 +-
 target/m68k/cpu-qom.h   | 2 +-
 target/microblaze/cpu-qom.h | 2 +-
 target/mips/cpu-qom.h   | 2 +-
 target/moxie/cpu.h  | 2 +-
 target/nios2/cpu.h  | 2 +-
 target/openrisc/cpu.h   | 2 +-
 target/ppc/cpu-qom.h| 2 +-
 target/riscv/cpu.h  | 2 +-
 target/rx/cpu-qom.h | 2 +-
 target/s390x/cpu-qom.h  | 2 +-
 target/sh4/cpu-qom.h| 2 +-
 target/sparc/cpu-qom.h  | 2 +-
 target/tilegx

Re: [PATCH 00/63] qom: Rename macros for consistency

[Eduardo Habkost]

On Wed, Sep 02, 2020 at 06:42:08PM -0400, Eduardo Habkost wrote:
> Reducing boilerplate QOM code using the new OBJECT_DEFINE_* and
> OBJECT_DECLARE_* macros is quite difficult when there are
> multiple ways a TYPE_* constant name is written.
> 
> This series renames many type checking macros and/or TYPE_*
> constants to make sure they are consistent.
> 
> This series is based on machine-next and can be fetched from:
> 
>   https://github.com/ehabkost/qemu-hacks work/qom-rename-macros

I'm queueing:

[PATCH 01/63] gpex: Fix type checking function name
[PATCH 02/63] chardev: Rename TYPE_CHARDEV_* to TYPE_*_CHARDEV
[PATCH 05/63] ap-device: Rename AP_DEVICE_TYPE to TYPE_AP_DEVICE
[PATCH 06/63] dev-smartcard-reader: Rename CCID_DEV_NAME to TYPE_USB_CCID_DEV
[PATCH 08/63] vfio: Rename VFIO_AP_DEVICE_TYPE to TYPE_VFIO_AP_DEVICE
[PATCH 10/63] vmgenid: Rename VMGENID_DEVICE to TYPE_VMGENID
[PATCH 32/63] ahci: Rename ICH_AHCI to ICH9_AHCI
[PATCH 39/63] esp: Rename ESP_STATE to ESP
[PATCH 40/63] filter-rewriter: Rename FILTER_COLO_REWRITER to FILTER_REWRITER
[PATCH 47/63] rs6000_mc: Rename RS6000MC_DEVICE to RS6000MC
[PATCH 48/63] sabre: Rename SABRE_DEVICE to SABRE
[PATCH 53/63] usb: Rename USB_SERIAL_DEV to USB_SERIAL
[PATCH 55/63] vfio: Rename PCI_VFIO to VFIO_PCI

-- 
Eduardo

[PATCH 32/63] ahci: Rename ICH_AHCI to ICH9_AHCI

[Eduardo Habkost]

Make the type checking macro name consistent with the TYPE_*
constant.

Signed-off-by: Eduardo Habkost 
---
Cc: John Snow 
Cc: qemu-block@nongnu.org
Cc: qemu-de...@nongnu.org
---
 include/hw/ide/ahci.h | 2 +-
 hw/ide/ahci.c | 4 ++--
 hw/ide/ich.c  | 8 
 3 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/include/hw/ide/ahci.h b/include/hw/ide/ahci.h
index 4cf6813d80..da3cddcc65 100644
--- a/include/hw/ide/ahci.h
+++ b/include/hw/ide/ahci.h
@@ -54,7 +54,7 @@ typedef struct AHCIState {
 typedef struct AHCIPCIState AHCIPCIState;
 
 #define TYPE_ICH9_AHCI "ich9-ahci"
-DECLARE_INSTANCE_CHECKER(AHCIPCIState, ICH_AHCI,
+DECLARE_INSTANCE_CHECKER(AHCIPCIState, ICH9_AHCI,
  TYPE_ICH9_AHCI)
 
 int32_t ahci_get_num_ports(PCIDevice *dev);
diff --git a/hw/ide/ahci.c b/hw/ide/ahci.c
index b696c6291a..ee1d47ff75 100644
--- a/hw/ide/ahci.c
+++ b/hw/ide/ahci.c
@@ -1819,7 +1819,7 @@ type_init(sysbus_ahci_register_types)
 
 int32_t ahci_get_num_ports(PCIDevice *dev)
 {
-AHCIPCIState *d = ICH_AHCI(dev);
+AHCIPCIState *d = ICH9_AHCI(dev);
 AHCIState *ahci = >ahci;
 
 return ahci->ports;
@@ -1827,7 +1827,7 @@ int32_t ahci_get_num_ports(PCIDevice *dev)
 
 void ahci_ide_create_devs(PCIDevice *dev, DriveInfo **hd)
 {
-AHCIPCIState *d = ICH_AHCI(dev);
+AHCIPCIState *d = ICH9_AHCI(dev);
 AHCIState *ahci = >ahci;
 int i;
 
diff --git a/hw/ide/ich.c b/hw/ide/ich.c
index eff3188fff..51cd2f38b7 100644
--- a/hw/ide/ich.c
+++ b/hw/ide/ich.c
@@ -91,14 +91,14 @@ static const VMStateDescription vmstate_ich9_ahci = {
 
 static void pci_ich9_reset(DeviceState *dev)
 {
-AHCIPCIState *d = ICH_AHCI(dev);
+AHCIPCIState *d = ICH9_AHCI(dev);
 
 ahci_reset(>ahci);
 }
 
 static void pci_ich9_ahci_init(Object *obj)
 {
-struct AHCIPCIState *d = ICH_AHCI(obj);
+struct AHCIPCIState *d = ICH9_AHCI(obj);
 
 ahci_init(>ahci, DEVICE(obj));
 }
@@ -108,7 +108,7 @@ static void pci_ich9_ahci_realize(PCIDevice *dev, Error 
**errp)
 struct AHCIPCIState *d;
 int sata_cap_offset;
 uint8_t *sata_cap;
-d = ICH_AHCI(dev);
+d = ICH9_AHCI(dev);
 int ret;
 
 ahci_realize(>ahci, DEVICE(dev), pci_get_address_space(dev), 6);
@@ -154,7 +154,7 @@ static void pci_ich9_ahci_realize(PCIDevice *dev, Error 
**errp)
 static void pci_ich9_uninit(PCIDevice *dev)
 {
 struct AHCIPCIState *d;
-d = ICH_AHCI(dev);
+d = ICH9_AHCI(dev);
 
 msi_uninit(dev);
 ahci_uninit(>ahci);
-- 
2.26.2

Re: [PATCH v4 00/18] qom: Automated conversion of type checking boilerplate

[Eduardo Habkost]

On Mon, Aug 31, 2020 at 05:07:22PM -0400, Eduardo Habkost wrote:
> Latest version of this series can be found at the branch:
> https://github.com/ehabkost/qemu-hacks/tree/work/qom-macros-autoconvert
> 
> This is an extension of the series previously submitted by
> Daniel[1], including a script that will convert existing type
> checker macros automatically.
> 
> [1] https://lore.kernel.org/qemu-devel/20200723181410.3145233-1-berrange@redh=
> at.com/
> 
> Link to series v4:
> https://github.com/ehabkost/qemu-hacks/tree/work/qom-macros-autoconvert-v4

I'm queueing this on machine-next.

-- 
Eduardo

[PATCH v4 15/18] [automated] Use OBJECT_DECLARE_TYPE where possible

[Eduardo Habkost]

Replace DECLARE_OBJ_CHECKERS with OBJECT_DECLARE_TYPE where the
typedefs can be safely removed.

Generated running:

$ ./scripts/codeconverter/converter.py -i \
  --pattern=DeclareObjCheckers $(git grep -l '' -- '*.[ch]')

Reviewed-by: Daniel P. Berrangé 
Signed-off-by: Eduardo Habkost 
---
Changes v3 -> v4: none

Changes v2 -> v3:
* Removed hunks due to rebase conflict: include/hw/ppc/xive.h
  include/hw/arm/armsse.h
* Reviewed-by line from Daniel was kept, as no additional hunks
  are introduced in this version

Changes v1 -> v2:
* Script re-run after typedefs and macros were moved, and now the
  patch also touches:
  - TYPE_ARM_SSE
  - TYPE_SD_BUS

Signed-off-by: Eduardo Habkost 

---
Cc: "Marc-André Lureau" 
Cc: Gerd Hoffmann 
Cc: "Michael S. Tsirkin" 
Cc: "Daniel P. Berrangé" 
Cc: Peter Maydell 
Cc: Corey Minyard 
Cc: "Cédric Le Goater" 
Cc: David Gibson 
Cc: Cornelia Huck 
Cc: Thomas Huth 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: "Philippe Mathieu-Daudé" 
Cc: Alistair Francis 
Cc: David Hildenbrand 
Cc: Laurent Vivier 
Cc: Amit Shah 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Paolo Bonzini 
Cc: Fam Zheng 
Cc: "Gonglei (Arei)" 
Cc: Eduardo Habkost 
Cc: Igor Mammedov 
Cc: Stefan Berger 
Cc: Richard Henderson 
Cc: Michael Rolnik 
Cc: Sarah Harris 
Cc: "Edgar E. Iglesias" 
Cc: Michael Walle 
Cc: Aleksandar Markovic 
Cc: Aurelien Jarno 
Cc: Jiaxun Yang 
Cc: Aleksandar Rikalo 
Cc: Anthony Green 
Cc: Chris Wulff 
Cc: Marek Vasut 
Cc: Stafford Horne 
Cc: Palmer Dabbelt 
Cc: Sagar Karandikar 
Cc: Bastian Koppelmann 
Cc: Yoshinori Sato 
Cc: Mark Cave-Ayland 
Cc: Artyom Tarasenko 
Cc: Guan Xuetao 
Cc: Max Filippov 
Cc: qemu-de...@nongnu.org
Cc: qemu-...@nongnu.org
Cc: qemu-...@nongnu.org
Cc: qemu-s3...@nongnu.org
Cc: qemu-block@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-ri...@nongnu.org

Signed-off-by: Eduardo Habkost 
---
 hw/audio/intel-hda.h| 6 ++
 hw/display/virtio-vga.h | 6 ++
 include/authz/base.h| 6 ++
 include/authz/list.h| 6 ++
 include/authz/listfile.h| 6 ++
 include/authz/pamacct.h | 6 ++
 include/authz/simple.h  | 6 ++
 include/crypto/secret_common.h  | 6 ++
 include/crypto/secret_keyring.h | 6 ++
 include/hw/hyperv/vmbus.h   | 6 ++
 include/hw/i2c/i2c.h| 6 ++
 include/hw/i2c/smbus_slave.h| 6 ++
 include/hw/ipack/ipack.h| 6 ++
 include/hw/ipmi/ipmi.h  | 6 ++
 include/hw/mem/pc-dimm.h| 6 ++
 include/hw/ppc/pnv.h| 6 ++
 include/hw/ppc/pnv_core.h   | 6 ++
 include/hw/ppc/pnv_homer.h  | 6 ++
 include/hw/ppc/pnv_occ.h| 6 ++
 include/hw/ppc/pnv_psi.h| 6 ++
 include/hw/ppc/pnv_xive.h   | 6 ++
 include/hw/ppc/spapr_cpu_core.h | 6 ++
 include/hw/ppc/spapr_drc.h  | 6 ++
 include/hw/ppc/spapr_vio.h  | 6 ++
 include/hw/ppc/spapr_xive.h | 6 ++
 include/hw/ppc/xics.h   | 6 ++
 include/hw/s390x/event-facility.h   | 6 ++
 include/hw/s390x/s390_flic.h| 6 ++
 include/hw/s390x/sclp.h | 6 ++
 include/hw/sd/sd.h  | 6 ++
 include/hw/ssi/ssi.h| 6 ++
 include/hw/sysbus.h | 6 ++
 include/hw/virtio/virtio-gpu.h  | 6 ++
 include/hw/virtio/virtio-input.h| 6 ++
 include/hw/virtio/virtio-mem.h  | 6 ++
 include/hw/virtio/virtio-pmem.h | 6 ++
 include/hw/virtio/virtio-serial.h   | 6 ++
 include/hw/xen/xen-bus.h| 6 ++
 include/io/channel.h| 6 ++
 include/io/dns-resolver.h   | 6 ++
 include/io/net-listener.h   | 6 ++
 include/scsi/pr-manager.h   | 6 ++
 include/sysemu/cryptodev.h  | 6 ++
 include/sysemu/hostmem.h| 6 ++
 include/sysemu/rng.h| 6 ++
 include/sysemu/tpm_backend.h| 6 ++
 include/sysemu/vhost-user-backend.h | 6 ++
 target/alpha/cpu-qom.h  | 6 ++
 target/arm/cpu-qom.h| 6 ++
 target/avr/cpu-qom.h| 6 ++
 target/cris/cpu-qom.h   | 6 ++
 target/hppa/cpu-qom.h   | 6 ++
 target/i386/cpu-qom.h   | 6 ++
 target/lm32/cpu-qom.h   | 6 ++
 target/m68k/cpu-qom.h   | 6 ++
 target/microblaze/cpu-qom.h | 6 ++
 target/mips/cpu-qom.h   | 6 ++
 target/moxie/cpu.h  | 6 ++
 target/nios2/cpu.h  | 6 ++
 target/openrisc/cpu.h   | 6 ++
 target/ppc/cpu-qom.h| 6 ++
 target/riscv/cpu.h  | 6 ++
 target/s390x/cpu-qom.h  | 6

[PATCH v4 14/18] [semi-automated] Use DECLARE_*CHECKER* when possible (--force mode)

[Eduardo Habkost]

Separate run of the TypeCheckMacro converter using the --force
flag, for the cases where typedefs weren't found in the same
header nor in typedefs.h.

Generated initially using:

 $ ./scripts/codeconverter/converter.py --force -i \
   --pattern=TypeCheckMacro $(git grep -l '' -- '*.[ch]')

Then each case was manually reviewed, and a comment was added
indicating what's unusual about those type checking
macros/functions.  Despite not following the usual pattern, the
changes in this patch were found to be safe.

Reviewed-by: Daniel P. Berrangé 
Signed-off-by: Eduardo Habkost 
---
Changes v3 -> v4: none

Changes v2 -> v3: none

Changes v1 -> v2:
* Most of the old changes in this patch are now being handled by
  the regular TypeCheckMacro patch (without --force mode)
* Added comments added explaining why these unusual changes
  remain

---
Cc: "Michael S. Tsirkin" 
Cc: Paolo Bonzini 
Cc: Peter Maydell 
Cc: Beniamino Galvani 
Cc: Andrew Baumann 
Cc: "Philippe Mathieu-Daudé" 
Cc: Andrzej Zaborowski 
Cc: David Gibson 
Cc: qemu-de...@nongnu.org
Cc: qemu-...@nongnu.org
Cc: qemu-block@nongnu.org
Cc: qemu-...@nongnu.org
---
 include/hw/intc/arm_gic.h   | 9 +++--
 include/hw/intc/arm_gicv3.h | 8 +++-
 include/hw/ppc/xics_spapr.h | 4 +++-
 include/hw/virtio/virtio-mmio.h | 9 +++--
 hw/intc/apic.c  | 5 +++--
 hw/intc/arm_gic_kvm.c   | 9 +++--
 hw/intc/arm_gicv3_its_kvm.c | 8 +++-
 hw/intc/arm_gicv3_kvm.c | 9 +++--
 hw/sd/allwinner-sdhost.c| 5 +++--
 hw/sd/bcm2835_sdhost.c  | 5 +++--
 hw/sd/pxa2xx_mmci.c | 4 +++-
 hw/sd/sdhci.c   | 4 +++-
 12 files changed, 36 insertions(+), 43 deletions(-)

diff --git a/include/hw/intc/arm_gic.h b/include/hw/intc/arm_gic.h
index 704ef2b751..116ccbb5a9 100644
--- a/include/hw/intc/arm_gic.h
+++ b/include/hw/intc/arm_gic.h
@@ -74,12 +74,9 @@
 
 #define TYPE_ARM_GIC "arm_gic"
 typedef struct ARMGICClass ARMGICClass;
-#define ARM_GIC(obj) \
- OBJECT_CHECK(GICState, (obj), TYPE_ARM_GIC)
-#define ARM_GIC_CLASS(klass) \
- OBJECT_CLASS_CHECK(ARMGICClass, (klass), TYPE_ARM_GIC)
-#define ARM_GIC_GET_CLASS(obj) \
- OBJECT_GET_CLASS(ARMGICClass, (obj), TYPE_ARM_GIC)
+/* This is reusing the GICState typedef from TYPE_ARM_GIC_COMMON */
+DECLARE_OBJ_CHECKERS(GICState, ARMGICClass,
+ ARM_GIC, TYPE_ARM_GIC)
 
 struct ARMGICClass {
 /*< private >*/
diff --git a/include/hw/intc/arm_gicv3.h b/include/hw/intc/arm_gicv3.h
index 58e9131a33..a81a6ae7ec 100644
--- a/include/hw/intc/arm_gicv3.h
+++ b/include/hw/intc/arm_gicv3.h
@@ -17,11 +17,9 @@
 
 #define TYPE_ARM_GICV3 "arm-gicv3"
 typedef struct ARMGICv3Class ARMGICv3Class;
-#define ARM_GICV3(obj) OBJECT_CHECK(GICv3State, (obj), TYPE_ARM_GICV3)
-#define ARM_GICV3_CLASS(klass) \
- OBJECT_CLASS_CHECK(ARMGICv3Class, (klass), TYPE_ARM_GICV3)
-#define ARM_GICV3_GET_CLASS(obj) \
- OBJECT_GET_CLASS(ARMGICv3Class, (obj), TYPE_ARM_GICV3)
+/* This is reusing the GICState typedef from TYPE_ARM_GICV3_COMMON */
+DECLARE_OBJ_CHECKERS(GICv3State, ARMGICv3Class,
+ ARM_GICV3, TYPE_ARM_GICV3)
 
 struct ARMGICv3Class {
 /*< private >*/
diff --git a/include/hw/ppc/xics_spapr.h b/include/hw/ppc/xics_spapr.h
index 09e428de4e..0b8182e40b 100644
--- a/include/hw/ppc/xics_spapr.h
+++ b/include/hw/ppc/xics_spapr.h
@@ -31,7 +31,9 @@
 #include "qom/object.h"
 
 #define TYPE_ICS_SPAPR "ics-spapr"
-#define ICS_SPAPR(obj) OBJECT_CHECK(ICSState, (obj), TYPE_ICS_SPAPR)
+/* This is reusing the ICSState typedef from TYPE_ICS */
+DECLARE_INSTANCE_CHECKER(ICSState, ICS_SPAPR,
+ TYPE_ICS_SPAPR)
 
 int xics_kvm_connect(SpaprInterruptController *intc, uint32_t nr_servers,
  Error **errp);
diff --git a/include/hw/virtio/virtio-mmio.h b/include/hw/virtio/virtio-mmio.h
index dca651fd14..6a1c2c20d4 100644
--- a/include/hw/virtio/virtio-mmio.h
+++ b/include/hw/virtio/virtio-mmio.h
@@ -28,12 +28,9 @@
 /* QOM macros */
 /* virtio-mmio-bus */
 #define TYPE_VIRTIO_MMIO_BUS "virtio-mmio-bus"
-#define VIRTIO_MMIO_BUS(obj) \
-OBJECT_CHECK(VirtioBusState, (obj), TYPE_VIRTIO_MMIO_BUS)
-#define VIRTIO_MMIO_BUS_GET_CLASS(obj) \
-OBJECT_GET_CLASS(VirtioBusClass, (obj), TYPE_VIRTIO_MMIO_BUS)
-#define VIRTIO_MMIO_BUS_CLASS(klass) \
-OBJECT_CLASS_CHECK(VirtioBusClass, (klass), TYPE_VIRTIO_MMIO_BUS)
+/* This is reusing the VirtioBusState typedef from TYPE_VIRTIO_BUS */
+DECLARE_OBJ_CHECKERS(VirtioBusState, VirtioBusClass,
+ VIRTIO_MMIO_BUS, TYPE_VIRTIO_MMIO_BUS)
 
 /* virtio-mmio */
 #define TYPE_VIRTIO_MMIO "virtio-mmio"
diff --git a/hw/intc/apic.c b/hw/intc/apic.c
index e055bb3af2..b6a05e5439 100644
--- a/hw/intc/apic.c
+++ b/hw/intc/apic.c
@@ -40,8 +40,9 @@
 static APICCommonState *local_apics[MAX_APICS + 1];
 
 #define TYPE_APIC "apic"
-

Re: [PATCH v3 00/74] qom: Automated conversion of type checking boilerplate

[Eduardo Habkost]

On Tue, Aug 25, 2020 at 03:19:56PM -0400, Eduardo Habkost wrote:
> git tree for this series:
> https://github.com/ehabkost/qemu-hacks/tree/work/qom-macros-autoconvert
> 
> This is an extension of the series previously submitted by
> Daniel[1], including a script that will convert existing type
> checker macros automatically.
> 
> [1] https://lore.kernel.org/qemu-devel/20200723181410.3145233-1-berrange@redh=
> at.com/

I'm queueing the following patches on machine-next:

[PATCH v3 01/74] e1000: Rename QOM class cast macros
[PATCH v3 02/74] megasas: Rename QOM class cast macros
[PATCH v3 03/74] vmw_pvscsi: Rename QOM class cast macros
[PATCH v3 04/74] pl110: Rename pl110_version enum values
[PATCH v3 05/74] allwinner-h3: Rename memmap enum constants
[PATCH v3 06/74] aspeed_soc: Rename memmap/irqmap enum constants
[PATCH v3 07/74] opentitan: Rename memmap enum constants
[PATCH v3 10/74] aspeed_timer: Fix ASPEED_TIMER macro definition
[PATCH v3 11/74] versatile: Fix typo in PCI_VPB_HOST definition
[PATCH v3 12/74] virtio-ccw: Fix definition of VIRTIO_CCW_BUS_GET_CLASS
[PATCH v3 13/74] hvf: Add missing include
[PATCH v3 14/74] hcd-dwc2: Rename USB_*CLASS macros for consistency
[PATCH v3 15/74] tulip: Move TulipState typedef to header
[PATCH v3 16/74] throttle-groups: Move ThrottleGroup typedef to header
[PATCH v3 17/74] pci: Move PCIBusClass typedef to pci.h
[PATCH v3 18/74] i8254: Move PITCommonState/PITCommonClass typedefs to i8254.h
[PATCH v3 19/74] hvf: Move HVFState typedef to hvf.h
[PATCH v3 20/74] mcf_fec: Move mcf_fec_state typedef to header
[PATCH v3 21/74] s390_flic: Move KVMS390FLICState typedef to header
[PATCH v3 22/74] can_emu: Delete macros for non-existing typedef
[PATCH v3 23/74] nubus: Delete unused NUBUS_BRIDGE macro
[PATCH v3 24/74] platform-bus: Delete macros for non-existing typedef
[PATCH v3 25/74] armsse: Rename QOM macros to avoid conflicts
[PATCH v3 26/74] xen-legacy-backend: Add missing typedef XenLegacyDevice
[PATCH v3 27/74] spapr: Move typedef SpaprMachineState to spapr.h
[PATCH v3 28/74] s390x: Move typedef SCLPEventFacility to event-facility.h
[PATCH v3 29/74] vhost-user-gpu: Move QOM macro to header
[PATCH v3 30/74] ahci: Move QOM macros to header
[PATCH v3 31/74] i8257: Move QOM macro to header
[PATCH v3 32/74] ahci: Move QOM macro to header
[PATCH v3 33/74] pckbd: Move QOM macro to header
[PATCH v3 34/74] vmbus: Move QOM macros to vmbus.h
[PATCH v3 35/74] virtio-serial-bus: Move QOM macros to header
[PATCH v3 36/74] piix: Move QOM macros to header
[PATCH v3 37/74] auxbus: Move QOM macros to header
[PATCH v3 38/74] rocker: Move QOM macros to header
[PATCH v3 39/74] pxa2xx: Move QOM macros to header
[PATCH v3 40/74] mptsas: Move QOM macros to header
[PATCH v3 41/74] kvm: Move QOM macros to kvm.h
[PATCH v3 42/74] vfio/pci: Move QOM macros to header
[PATCH v3 43/74] nubus: Rename class type checking macros
[PATCH v3 48/74] s390-virtio-ccw: Rename S390_MACHINE_CLASS macro
[PATCH v3 49/74] swim: Rename struct SWIM to Swim
[PATCH v3 50/74] migration: Rename class type checking macros

-- 
Eduardo

Re: [PATCH v3 72/74] [automated] Remove redundant instance_size/class_size fields

[Eduardo Habkost]

On Tue, Aug 25, 2020 at 03:21:08PM -0400, Eduardo Habkost wrote:
> This will remove instance_size/class_size fields from TypeInfo
> variables when the value is exactly the same as the one in the
> parent class.
> 
> Generated by:
> 
>  $ ./scripts/codeconverter/converter.py -i \
>--pattern=RedundantTypeSizes $(git grep -l TypeInfo -- '*.[ch]')
> 
> Signed-off-by: Eduardo Habkost 

This patch triggers an assert due to code outside QOM core using
TypeInfo.instance_size directly.  Please ignore it by now.

Thanks to Roman Bolshakov for reporting the problem.

-- 
Eduardo

[PATCH v3 72/74] [automated] Remove redundant instance_size/class_size fields

[Eduardo Habkost]

This will remove instance_size/class_size fields from TypeInfo
variables when the value is exactly the same as the one in the
parent class.

Generated by:

 $ ./scripts/codeconverter/converter.py -i \
   --pattern=RedundantTypeSizes $(git grep -l TypeInfo -- '*.[ch]')

Signed-off-by: Eduardo Habkost 
---
Changes series v2 -> v3: this is a new patch in series v3

---
Cc: "Marc-André Lureau" 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: "Cédric Le Goater" 
Cc: Peter Maydell 
Cc: Andrew Jeffery 
Cc: Joel Stanley 
Cc: Jan Kiszka 
Cc: Eduardo Habkost 
Cc: "Michael S. Tsirkin" 
Cc: Marcel Apfelbaum 
Cc: Richard Henderson 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Gerd Hoffmann 
Cc: Aleksandar Markovic 
Cc: "Philippe Mathieu-Daudé" 
Cc: Aurelien Jarno 
Cc: Jiaxun Yang 
Cc: Aleksandar Rikalo 
Cc: Laurent Vivier 
Cc: Mark Cave-Ayland 
Cc: David Gibson 
Cc: Cornelia Huck 
Cc: Thomas Huth 
Cc: David Hildenbrand 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Fam Zheng 
Cc: Beniamino Galvani 
Cc: Andrew Baumann 
Cc: Michael Walle 
Cc: Andrzej Zaborowski 
Cc: Yoshinori Sato 
Cc: Alex Williamson 
Cc: qemu-de...@nongnu.org
Cc: qemu-...@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-...@nongnu.org
Cc: qemu-s3...@nongnu.org
Cc: qemu-block@nongnu.org

---
Cc: "Marc-André Lureau" 
Cc: Paolo Bonzini 
Cc: "Daniel P. Berrangé" 
Cc: "Cédric Le Goater" 
Cc: Peter Maydell 
Cc: Andrew Jeffery 
Cc: Joel Stanley 
Cc: Jan Kiszka 
Cc: Eduardo Habkost 
Cc: "Michael S. Tsirkin" 
Cc: Marcel Apfelbaum 
Cc: Richard Henderson 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Gerd Hoffmann 
Cc: Aleksandar Markovic 
Cc: "Philippe Mathieu-Daudé" 
Cc: Aurelien Jarno 
Cc: Jiaxun Yang 
Cc: Aleksandar Rikalo 
Cc: Laurent Vivier 
Cc: Mark Cave-Ayland 
Cc: David Gibson 
Cc: Cornelia Huck 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: Thomas Huth 
Cc: David Hildenbrand 
Cc: Fam Zheng 
Cc: Beniamino Galvani 
Cc: Andrew Baumann 
Cc: Michael Walle 
Cc: Andrzej Zaborowski 
Cc: Yoshinori Sato 
Cc: Alex Williamson 
Cc: qemu-de...@nongnu.org
Cc: qemu-...@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-...@nongnu.org
Cc: qemu-s3...@nongnu.org
Cc: qemu-block@nongnu.org
---
 chardev/char-null.c| 1 -
 crypto/tls-cipher-suites.c | 2 --
 hw/arm/aspeed_ast2600.c| 2 --
 hw/arm/aspeed_soc.c| 2 --
 hw/arm/musicpal.c  | 1 -
 hw/core/sysbus.c   | 1 -
 hw/i386/kvm/apic.c | 1 -
 hw/i386/pc_piix.c  | 1 -
 hw/i386/xen/xen_apic.c | 1 -
 hw/input/virtio-input-hid.c| 3 ---
 hw/intc/apic.c | 1 -
 hw/intc/ioapic.c   | 1 -
 hw/isa/isa-bus.c   | 1 -
 hw/mips/gt64xxx_pci.c  | 1 -
 hw/misc/aspeed_scu.c   | 3 ---
 hw/misc/ivshmem.c  | 2 --
 hw/nubus/nubus-bridge.c| 1 -
 hw/pci-bridge/dec.c| 2 --
 hw/pci-bridge/pci_bridge_dev.c | 1 -
 hw/pci-host/grackle.c  | 1 -
 hw/pci-host/uninorth.c | 4 
 hw/pci-host/versatile.c| 1 -
 hw/pci-host/xen_igd_pt.c   | 1 -
 hw/ppc/pnv_homer.c | 2 --
 hw/ppc/pnv_occ.c   | 2 --
 hw/ppc/ppc4xx_pci.c| 1 -
 hw/s390x/s390-skeys-kvm.c  | 2 --
 hw/s390x/sclpcpu.c | 2 --
 hw/s390x/sclpquiesce.c | 2 --
 hw/s390x/tod-kvm.c | 2 --
 hw/s390x/tod-qemu.c| 2 --
 hw/s390x/virtio-ccw-input.c| 3 ---
 hw/scsi/scsi-generic.c | 1 -
 hw/sd/allwinner-sdhost.c   | 1 -
 hw/sd/bcm2835_sdhost.c | 1 -
 hw/sd/milkymist-memcard.c  | 1 -
 hw/sd/pl181.c  | 1 -
 hw/sd/pxa2xx_mmci.c| 1 -
 hw/sd/sdhci.c  | 1 -
 hw/sh4/sh_pci.c| 1 -
 hw/timer/pxa2xx_timer.c| 2 --
 hw/vfio/pci.c  | 1 -
 hw/virtio/virtio-mmio.c| 1 -
 hw/watchdog/wdt_aspeed.c   | 3 ---
 hw/xen/xen-legacy-backend.c| 1 -
 45 files changed, 69 deletions(-)

diff --git a/chardev/char-null.c b/chardev/char-null.c
index ce43ccdda6..2736b2ff2b 100644
--- a/chardev/char-null.c
+++ b/chardev/char-null.c
@@ -44,7 +44,6 @@ static void char_null_class_init(ObjectClass *oc, void *data)
 static const TypeInfo char_null_type_info = {
 .name = TYPE_CHARDEV_NULL,
 .parent = TYPE_CHARDEV,
-.instance_size = sizeof(Chardev),
 .class_init = char_null_class_init,
 };
 TYPE_INFO(char_null_type_info)
diff --git a/crypto/tls-cipher-suites.c b/crypto/tls-cipher-suites.c
index e92a380a24..0c9713d301 100644
--- a/crypto/tls-cipher-suites.c
+++ b/crypto/tls-cipher-suites.c
@@ -108,8 +108,6 @@ static void 
qcrypto_tls_cipher_suites_class_init(ObjectClass *oc, void *data)
 static const TypeInfo qcrypto_tls_cipher_suites_info = {
 .parent = TYPE_QCRYPTO_TLS_CREDS,
 .name = TYPE_QCRYPTO_TLS_CIPHER_SUITES,
-.ins

[PATCH v3 69/74] [automated] Use OBJECT_DECLARE_TYPE where possible

[Eduardo Habkost]

Replace DECLARE_OBJ_CHECKERS with OBJECT_DECLARE_TYPE where the
typedefs can be safely removed.

Generated running:

$ ./scripts/codeconverter/converter.py -i \
  --pattern=DeclareObjCheckers $(git grep -l '' -- '*.[ch]')

Reviewed-by: Daniel P. Berrangé 
Signed-off-by: Eduardo Habkost 
---
Changes v2 -> v3:
* Removed hunk due to rebase conflict: include/hw/ppc/xive.h
* Reviewed-by line from Daniel was kept, as no additional hunks
  are introduced in this version

Changes v1 -> v2:
* Script re-run after typedefs and macros were moved, and now the
  patch also touches:
  - TYPE_ARM_SSE
  - TYPE_SD_BUS

Signed-off-by: Eduardo Habkost 

---
Cc: "Marc-André Lureau" 
Cc: Gerd Hoffmann 
Cc: "Michael S. Tsirkin" 
Cc: "Daniel P. Berrangé" 
Cc: Peter Maydell 
Cc: Corey Minyard 
Cc: "Cédric Le Goater" 
Cc: David Gibson 
Cc: Cornelia Huck 
Cc: Thomas Huth 
Cc: Halil Pasic 
Cc: Christian Borntraeger 
Cc: "Philippe Mathieu-Daudé" 
Cc: Alistair Francis 
Cc: David Hildenbrand 
Cc: Laurent Vivier 
Cc: Amit Shah 
Cc: Stefano Stabellini 
Cc: Anthony Perard 
Cc: Paul Durrant 
Cc: Paolo Bonzini 
Cc: Fam Zheng 
Cc: "Gonglei (Arei)" 
Cc: Eduardo Habkost 
Cc: Igor Mammedov 
Cc: Stefan Berger 
Cc: Richard Henderson 
Cc: Michael Rolnik 
Cc: Sarah Harris 
Cc: "Edgar E. Iglesias" 
Cc: Michael Walle 
Cc: Aleksandar Markovic 
Cc: Aurelien Jarno 
Cc: Jiaxun Yang 
Cc: Aleksandar Rikalo 
Cc: Anthony Green 
Cc: Chris Wulff 
Cc: Marek Vasut 
Cc: Stafford Horne 
Cc: Palmer Dabbelt 
Cc: Sagar Karandikar 
Cc: Bastian Koppelmann 
Cc: Yoshinori Sato 
Cc: Mark Cave-Ayland 
Cc: Artyom Tarasenko 
Cc: Guan Xuetao 
Cc: Max Filippov 
Cc: qemu-de...@nongnu.org
Cc: qemu-...@nongnu.org
Cc: qemu-...@nongnu.org
Cc: qemu-s3...@nongnu.org
Cc: qemu-block@nongnu.org
Cc: xen-de...@lists.xenproject.org
Cc: qemu-ri...@nongnu.org
---
 hw/audio/intel-hda.h| 6 ++
 hw/display/virtio-vga.h | 6 ++
 include/authz/base.h| 6 ++
 include/authz/list.h| 6 ++
 include/authz/listfile.h| 6 ++
 include/authz/pamacct.h | 6 ++
 include/authz/simple.h  | 6 ++
 include/crypto/secret_common.h  | 6 ++
 include/crypto/secret_keyring.h | 6 ++
 include/hw/arm/armsse.h | 6 ++
 include/hw/hyperv/vmbus.h   | 6 ++
 include/hw/i2c/i2c.h| 6 ++
 include/hw/i2c/smbus_slave.h| 6 ++
 include/hw/ipack/ipack.h| 6 ++
 include/hw/ipmi/ipmi.h  | 6 ++
 include/hw/mem/pc-dimm.h| 6 ++
 include/hw/ppc/pnv.h| 6 ++
 include/hw/ppc/pnv_core.h   | 6 ++
 include/hw/ppc/pnv_homer.h  | 6 ++
 include/hw/ppc/pnv_occ.h| 6 ++
 include/hw/ppc/pnv_psi.h| 6 ++
 include/hw/ppc/pnv_xive.h   | 6 ++
 include/hw/ppc/spapr_cpu_core.h | 6 ++
 include/hw/ppc/spapr_drc.h  | 6 ++
 include/hw/ppc/spapr_vio.h  | 6 ++
 include/hw/ppc/spapr_xive.h | 6 ++
 include/hw/ppc/xics.h   | 6 ++
 include/hw/s390x/event-facility.h   | 6 ++
 include/hw/s390x/s390_flic.h| 6 ++
 include/hw/s390x/sclp.h | 6 ++
 include/hw/sd/sd.h  | 6 ++
 include/hw/ssi/ssi.h| 6 ++
 include/hw/sysbus.h | 6 ++
 include/hw/virtio/virtio-gpu.h  | 6 ++
 include/hw/virtio/virtio-input.h| 6 ++
 include/hw/virtio/virtio-mem.h  | 6 ++
 include/hw/virtio/virtio-pmem.h | 6 ++
 include/hw/virtio/virtio-serial.h   | 6 ++
 include/hw/xen/xen-bus.h| 6 ++
 include/io/channel.h| 6 ++
 include/io/dns-resolver.h   | 6 ++
 include/io/net-listener.h   | 6 ++
 include/scsi/pr-manager.h   | 6 ++
 include/sysemu/cryptodev.h  | 6 ++
 include/sysemu/hostmem.h| 6 ++
 include/sysemu/rng.h| 6 ++
 include/sysemu/tpm_backend.h| 6 ++
 include/sysemu/vhost-user-backend.h | 6 ++
 target/alpha/cpu-qom.h  | 6 ++
 target/arm/cpu-qom.h| 6 ++
 target/avr/cpu-qom.h| 6 ++
 target/cris/cpu-qom.h   | 6 ++
 target/hppa/cpu-qom.h   | 6 ++
 target/i386/cpu-qom.h   | 6 ++
 target/lm32/cpu-qom.h   | 6 ++
 target/m68k/cpu-qom.h   | 6 ++
 target/microblaze/cpu-qom.h | 6 ++
 target/mips/cpu-qom.h   | 6 ++
 target/moxie/cpu.h  | 6 ++
 target/nios2/cpu.h  | 6 ++
 target/openrisc/cpu.h   | 6 ++
 target/ppc/cpu-qom.h| 6 ++
 target/riscv/cpu.h  | 6 ++
 target/s390x/cpu-qom.h  | 6 ++
 target/sh4/cpu-qom.h| 6

[PATCH v3 49/74] swim: Rename struct SWIM to Swim

[Eduardo Habkost]

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 
---
Changes series v2 -> v3: new patch added to series v3

---
Cc: Laurent Vivier 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: qemu-block@nongnu.org
Cc: qemu-de...@nongnu.org

Signed-off-by: Eduardo Habkost 
---
 include/hw/block/swim.h |  6 +++---
 hw/block/swim.c | 10 +-
 2 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/include/hw/block/swim.h b/include/hw/block/swim.h
index 6add3499d0..9d8b65c561 100644
--- a/include/hw/block/swim.h
+++ b/include/hw/block/swim.h
@@ -67,10 +67,10 @@ struct SWIMCtrl {
 };
 
 #define TYPE_SWIM "swim"
-#define SWIM(obj) OBJECT_CHECK(SWIM, (obj), TYPE_SWIM)
+#define SWIM(obj) OBJECT_CHECK(Swim, (obj), TYPE_SWIM)
 
-typedef struct SWIM {
+typedef struct Swim {
 SysBusDevice parent_obj;
 SWIMCtrl ctrl;
-} SWIM;
+} Swim;
 #endif
diff --git a/hw/block/swim.c b/hw/block/swim.c
index 74f56e8f46..20133a814c 100644
--- a/hw/block/swim.c
+++ b/hw/block/swim.c
@@ -387,7 +387,7 @@ static const MemoryRegionOps swimctrl_mem_ops = {
 
 static void sysbus_swim_reset(DeviceState *d)
 {
-SWIM *sys = SWIM(d);
+Swim *sys = SWIM(d);
 SWIMCtrl *ctrl = >ctrl;
 int i;
 
@@ -408,7 +408,7 @@ static void sysbus_swim_reset(DeviceState *d)
 static void sysbus_swim_init(Object *obj)
 {
 SysBusDevice *sbd = SYS_BUS_DEVICE(obj);
-SWIM *sbs = SWIM(obj);
+Swim *sbs = SWIM(obj);
 SWIMCtrl *swimctrl = >ctrl;
 
 memory_region_init_io(>iomem, obj, _mem_ops, swimctrl,
@@ -418,7 +418,7 @@ static void sysbus_swim_init(Object *obj)
 
 static void sysbus_swim_realize(DeviceState *dev, Error **errp)
 {
-SWIM *sys = SWIM(dev);
+Swim *sys = SWIM(dev);
 SWIMCtrl *swimctrl = >ctrl;
 
 qbus_create_inplace(>bus, sizeof(SWIMBus), TYPE_SWIM_BUS, dev,
@@ -460,7 +460,7 @@ static const VMStateDescription vmstate_sysbus_swim = {
 .name = "SWIM",
 .version_id = 1,
 .fields = (VMStateField[]) {
-VMSTATE_STRUCT(ctrl, SWIM, 0, vmstate_swim, SWIMCtrl),
+VMSTATE_STRUCT(ctrl, Swim, 0, vmstate_swim, SWIMCtrl),
 VMSTATE_END_OF_LIST()
 }
 };
@@ -477,7 +477,7 @@ static void sysbus_swim_class_init(ObjectClass *oc, void 
*data)
 static const TypeInfo sysbus_swim_info = {
 .name  = TYPE_SWIM,
 .parent= TYPE_SYS_BUS_DEVICE,
-.instance_size = sizeof(SWIM),
+.instance_size = sizeof(Swim),
 .instance_init = sysbus_swim_init,
 .class_init= sysbus_swim_class_init,
 };
-- 
2.26.2

[PATCH v3 30/74] ahci: Move QOM macros to header

[Eduardo Habkost]

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 
Signed-off-by: Eduardo Habkost 
---
Changes v2 -> v3: none

Changes series v1 -> v2: new patch in series v2

Cc: John Snow 
Cc: qemu-block@nongnu.org
Cc: qemu-de...@nongnu.org
---
 hw/ide/ahci_internal.h | 5 -
 include/hw/ide/ahci.h  | 3 +++
 2 files changed, 3 insertions(+), 5 deletions(-)

diff --git a/hw/ide/ahci_internal.h b/hw/ide/ahci_internal.h
index bab0459774..ac9bdead7b 100644
--- a/hw/ide/ahci_internal.h
+++ b/hw/ide/ahci_internal.h
@@ -332,9 +332,6 @@ struct AHCIPCIState {
 AHCIState ahci;
 };
 
-#define ICH_AHCI(obj) \
-OBJECT_CHECK(AHCIPCIState, (obj), TYPE_ICH9_AHCI)
-
 extern const VMStateDescription vmstate_ahci;
 
 #define VMSTATE_AHCI(_field, _state) {   \
@@ -394,6 +391,4 @@ void ahci_uninit(AHCIState *s);
 
 void ahci_reset(AHCIState *s);
 
-#define SYSBUS_AHCI(obj) OBJECT_CHECK(SysbusAHCIState, (obj), TYPE_SYSBUS_AHCI)
-
 #endif /* HW_IDE_AHCI_INTERNAL_H */
diff --git a/include/hw/ide/ahci.h b/include/hw/ide/ahci.h
index b44e3000cf..ce2bf8a5f8 100644
--- a/include/hw/ide/ahci.h
+++ b/include/hw/ide/ahci.h
@@ -53,11 +53,14 @@ typedef struct AHCIState {
 typedef struct AHCIPCIState AHCIPCIState;
 
 #define TYPE_ICH9_AHCI "ich9-ahci"
+#define ICH_AHCI(obj) \
+OBJECT_CHECK(AHCIPCIState, (obj), TYPE_ICH9_AHCI)
 
 int32_t ahci_get_num_ports(PCIDevice *dev);
 void ahci_ide_create_devs(PCIDevice *dev, DriveInfo **hd);
 
 #define TYPE_SYSBUS_AHCI "sysbus-ahci"
+#define SYSBUS_AHCI(obj) OBJECT_CHECK(SysbusAHCIState, (obj), TYPE_SYSBUS_AHCI)
 
 typedef struct SysbusAHCIState {
 /*< private >*/
-- 
2.26.2

[PATCH v3 68/74] [semi-automated] Use DECLARE_*CHECKER* when possible (--force mode)

[Eduardo Habkost]

Separate run of the TypeCheckMacro converter using the --force
flag, for the cases where typedefs weren't found in the same
header nor in typedefs.h.

Generated initially using:

 $ ./scripts/codeconverter/converter.py --force -i \
   --pattern=TypeCheckMacro $(git grep -l '' -- '*.[ch]')

Then each case was manually reviewed, and a comment was added
indicating what's unusual about those type checking
macros/functions.  Despite not following the usual pattern, the
changes in this patch were found to be safe.

Reviewed-by: Daniel P. Berrangé 
Signed-off-by: Eduardo Habkost 
---
Changes v2 -> v3: none

Changes v1 -> v2:
* Most of the old changes in this patch are now being handled by
  the regular TypeCheckMacro patch (without --force mode)
* Added comments added explaining why these unusual changes
  remain

---
Cc: "Michael S. Tsirkin" 
Cc: Paolo Bonzini 
Cc: Peter Maydell 
Cc: Beniamino Galvani 
Cc: Andrew Baumann 
Cc: "Philippe Mathieu-Daudé" 
Cc: Andrzej Zaborowski 
Cc: David Gibson 
Cc: qemu-de...@nongnu.org
Cc: qemu-...@nongnu.org
Cc: qemu-block@nongnu.org
Cc: qemu-...@nongnu.org
---
 include/hw/intc/arm_gic.h   | 9 +++--
 include/hw/intc/arm_gicv3.h | 8 +++-
 include/hw/ppc/xics_spapr.h | 4 +++-
 include/hw/virtio/virtio-mmio.h | 9 +++--
 hw/intc/apic.c  | 5 +++--
 hw/intc/arm_gic_kvm.c   | 9 +++--
 hw/intc/arm_gicv3_its_kvm.c | 8 +++-
 hw/intc/arm_gicv3_kvm.c | 9 +++--
 hw/sd/allwinner-sdhost.c| 5 +++--
 hw/sd/bcm2835_sdhost.c  | 5 +++--
 hw/sd/pxa2xx_mmci.c | 4 +++-
 hw/sd/sdhci.c   | 4 +++-
 12 files changed, 36 insertions(+), 43 deletions(-)

diff --git a/include/hw/intc/arm_gic.h b/include/hw/intc/arm_gic.h
index 704ef2b751..116ccbb5a9 100644
--- a/include/hw/intc/arm_gic.h
+++ b/include/hw/intc/arm_gic.h
@@ -74,12 +74,9 @@
 
 #define TYPE_ARM_GIC "arm_gic"
 typedef struct ARMGICClass ARMGICClass;
-#define ARM_GIC(obj) \
- OBJECT_CHECK(GICState, (obj), TYPE_ARM_GIC)
-#define ARM_GIC_CLASS(klass) \
- OBJECT_CLASS_CHECK(ARMGICClass, (klass), TYPE_ARM_GIC)
-#define ARM_GIC_GET_CLASS(obj) \
- OBJECT_GET_CLASS(ARMGICClass, (obj), TYPE_ARM_GIC)
+/* This is reusing the GICState typedef from TYPE_ARM_GIC_COMMON */
+DECLARE_OBJ_CHECKERS(GICState, ARMGICClass,
+ ARM_GIC, TYPE_ARM_GIC)
 
 struct ARMGICClass {
 /*< private >*/
diff --git a/include/hw/intc/arm_gicv3.h b/include/hw/intc/arm_gicv3.h
index 58e9131a33..a81a6ae7ec 100644
--- a/include/hw/intc/arm_gicv3.h
+++ b/include/hw/intc/arm_gicv3.h
@@ -17,11 +17,9 @@
 
 #define TYPE_ARM_GICV3 "arm-gicv3"
 typedef struct ARMGICv3Class ARMGICv3Class;
-#define ARM_GICV3(obj) OBJECT_CHECK(GICv3State, (obj), TYPE_ARM_GICV3)
-#define ARM_GICV3_CLASS(klass) \
- OBJECT_CLASS_CHECK(ARMGICv3Class, (klass), TYPE_ARM_GICV3)
-#define ARM_GICV3_GET_CLASS(obj) \
- OBJECT_GET_CLASS(ARMGICv3Class, (obj), TYPE_ARM_GICV3)
+/* This is reusing the GICState typedef from TYPE_ARM_GICV3_COMMON */
+DECLARE_OBJ_CHECKERS(GICv3State, ARMGICv3Class,
+ ARM_GICV3, TYPE_ARM_GICV3)
 
 struct ARMGICv3Class {
 /*< private >*/
diff --git a/include/hw/ppc/xics_spapr.h b/include/hw/ppc/xics_spapr.h
index 09e428de4e..0b8182e40b 100644
--- a/include/hw/ppc/xics_spapr.h
+++ b/include/hw/ppc/xics_spapr.h
@@ -31,7 +31,9 @@
 #include "qom/object.h"
 
 #define TYPE_ICS_SPAPR "ics-spapr"
-#define ICS_SPAPR(obj) OBJECT_CHECK(ICSState, (obj), TYPE_ICS_SPAPR)
+/* This is reusing the ICSState typedef from TYPE_ICS */
+DECLARE_INSTANCE_CHECKER(ICSState, ICS_SPAPR,
+ TYPE_ICS_SPAPR)
 
 int xics_kvm_connect(SpaprInterruptController *intc, uint32_t nr_servers,
  Error **errp);
diff --git a/include/hw/virtio/virtio-mmio.h b/include/hw/virtio/virtio-mmio.h
index dca651fd14..6a1c2c20d4 100644
--- a/include/hw/virtio/virtio-mmio.h
+++ b/include/hw/virtio/virtio-mmio.h
@@ -28,12 +28,9 @@
 /* QOM macros */
 /* virtio-mmio-bus */
 #define TYPE_VIRTIO_MMIO_BUS "virtio-mmio-bus"
-#define VIRTIO_MMIO_BUS(obj) \
-OBJECT_CHECK(VirtioBusState, (obj), TYPE_VIRTIO_MMIO_BUS)
-#define VIRTIO_MMIO_BUS_GET_CLASS(obj) \
-OBJECT_GET_CLASS(VirtioBusClass, (obj), TYPE_VIRTIO_MMIO_BUS)
-#define VIRTIO_MMIO_BUS_CLASS(klass) \
-OBJECT_CLASS_CHECK(VirtioBusClass, (klass), TYPE_VIRTIO_MMIO_BUS)
+/* This is reusing the VirtioBusState typedef from TYPE_VIRTIO_BUS */
+DECLARE_OBJ_CHECKERS(VirtioBusState, VirtioBusClass,
+ VIRTIO_MMIO_BUS, TYPE_VIRTIO_MMIO_BUS)
 
 /* virtio-mmio */
 #define TYPE_VIRTIO_MMIO "virtio-mmio"
diff --git a/hw/intc/apic.c b/hw/intc/apic.c
index afbb653497..dadbfd9a75 100644
--- a/hw/intc/apic.c
+++ b/hw/intc/apic.c
@@ -40,8 +40,9 @@
 static APICCommonState *local_apics[MAX_APICS + 1];
 
 #define TYPE_APIC "apic"
-#define APIC(obj) \
-OBJ

[PATCH v3 32/74] ahci: Move QOM macro to header

[Eduardo Habkost]

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 
Signed-off-by: Eduardo Habkost 
---
Changes v2 -> v3: none

Changes series v1 -> v2: new patch in series v2

Cc: John Snow 
Cc: qemu-block@nongnu.org
Cc: qemu-de...@nongnu.org
---
 include/hw/ide/ahci.h   | 2 ++
 hw/ide/ahci-allwinner.c | 3 ---
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/include/hw/ide/ahci.h b/include/hw/ide/ahci.h
index ce2bf8a5f8..41bb517047 100644
--- a/include/hw/ide/ahci.h
+++ b/include/hw/ide/ahci.h
@@ -72,6 +72,8 @@ typedef struct SysbusAHCIState {
 } SysbusAHCIState;
 
 #define TYPE_ALLWINNER_AHCI "allwinner-ahci"
+#define ALLWINNER_AHCI(obj) \
+OBJECT_CHECK(AllwinnerAHCIState, (obj), TYPE_ALLWINNER_AHCI)
 
 #define ALLWINNER_AHCI_MMIO_OFF  0x80
 #define ALLWINNER_AHCI_MMIO_SIZE 0x80
diff --git a/hw/ide/ahci-allwinner.c b/hw/ide/ahci-allwinner.c
index 8536b9eb5a..227e747ba7 100644
--- a/hw/ide/ahci-allwinner.c
+++ b/hw/ide/ahci-allwinner.c
@@ -25,9 +25,6 @@
 
 #include "trace.h"
 
-#define ALLWINNER_AHCI(obj) \
-OBJECT_CHECK(AllwinnerAHCIState, (obj), TYPE_ALLWINNER_AHCI)
-
 #define ALLWINNER_AHCI_BISTAFR((0xa0 - ALLWINNER_AHCI_MMIO_OFF) / 4)
 #define ALLWINNER_AHCI_BISTCR ((0xa4 - ALLWINNER_AHCI_MMIO_OFF) / 4)
 #define ALLWINNER_AHCI_BISTFCTR   ((0xa8 - ALLWINNER_AHCI_MMIO_OFF) / 4)
-- 
2.26.2

[PATCH v3 16/74] throttle-groups: Move ThrottleGroup typedef to header

[Eduardo Habkost]

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 
Reviewed-by: Daniel P. Berrangé 
Signed-off-by: Eduardo Habkost 
---
Changes v2 -> v3: none

Changes v1 -> v2: none

---
Cc: Alberto Garcia 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: qemu-block@nongnu.org
Cc: qemu-de...@nongnu.org
---
 include/block/throttle-groups.h | 1 +
 block/throttle-groups.c | 4 ++--
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/include/block/throttle-groups.h b/include/block/throttle-groups.h
index 712a8e64b4..5e77db700f 100644
--- a/include/block/throttle-groups.h
+++ b/include/block/throttle-groups.h
@@ -59,6 +59,7 @@ typedef struct ThrottleGroupMember {
 } ThrottleGroupMember;
 
 #define TYPE_THROTTLE_GROUP "throttle-group"
+typedef struct ThrottleGroup ThrottleGroup;
 #define THROTTLE_GROUP(obj) OBJECT_CHECK(ThrottleGroup, (obj), 
TYPE_THROTTLE_GROUP)
 
 const char *throttle_group_get_name(ThrottleGroupMember *tgm);
diff --git a/block/throttle-groups.c b/block/throttle-groups.c
index 98fea7fd47..4e28365d8d 100644
--- a/block/throttle-groups.c
+++ b/block/throttle-groups.c
@@ -63,7 +63,7 @@ static void timer_cb(ThrottleGroupMember *tgm, bool is_write);
  * access some other ThrottleGroupMember's timers only after verifying that
  * that ThrottleGroupMember has throttled requests in the queue.
  */
-typedef struct ThrottleGroup {
+struct ThrottleGroup {
 Object parent_obj;
 
 /* refuse individual property change if initialization is complete */
@@ -79,7 +79,7 @@ typedef struct ThrottleGroup {
 
 /* This field is protected by the global QEMU mutex */
 QTAILQ_ENTRY(ThrottleGroup) list;
-} ThrottleGroup;
+};
 
 /* This is protected by the global QEMU mutex */
 static QTAILQ_HEAD(, ThrottleGroup) throttle_groups =
-- 
2.26.2

[PATCH v3 02/74] megasas: Rename QOM class cast macros

[Eduardo Habkost]

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 
Signed-off-by: Eduardo Habkost 
---
Changes v2 -> v3: none

Changes v1 -> v2: none

---
Cc: Paolo Bonzini 
Cc: Fam Zheng 
Cc: Hannes Reinecke 
Cc: qemu-block@nongnu.org
Cc: qemu-de...@nongnu.org
---
 hw/scsi/megasas.c | 12 ++--
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/hw/scsi/megasas.c b/hw/scsi/megasas.c
index 5cfd1bf22e..390c2f2edb 100644
--- a/hw/scsi/megasas.c
+++ b/hw/scsi/megasas.c
@@ -134,9 +134,9 @@ typedef struct MegasasBaseClass {
 #define MEGASAS(obj) \
 OBJECT_CHECK(MegasasState, (obj), TYPE_MEGASAS_BASE)
 
-#define MEGASAS_DEVICE_CLASS(oc) \
+#define MEGASAS_CLASS(oc) \
 OBJECT_CLASS_CHECK(MegasasBaseClass, (oc), TYPE_MEGASAS_BASE)
-#define MEGASAS_DEVICE_GET_CLASS(oc) \
+#define MEGASAS_GET_CLASS(oc) \
 OBJECT_GET_CLASS(MegasasBaseClass, (oc), TYPE_MEGASAS_BASE)
 
 #define MEGASAS_INTR_DISABLED_MASK 0x
@@ -733,7 +733,7 @@ static int megasas_ctrl_get_info(MegasasState *s, 
MegasasCmd *cmd)
 {
 PCIDevice *pci_dev = PCI_DEVICE(s);
 PCIDeviceClass *pci_class = PCI_DEVICE_GET_CLASS(pci_dev);
-MegasasBaseClass *base_class = MEGASAS_DEVICE_GET_CLASS(s);
+MegasasBaseClass *base_class = MEGASAS_GET_CLASS(s);
 struct mfi_ctrl_info info;
 size_t dcmd_size = sizeof(info);
 BusChild *kid;
@@ -1999,7 +1999,7 @@ static uint64_t megasas_mmio_read(void *opaque, hwaddr 
addr,
 {
 MegasasState *s = opaque;
 PCIDevice *pci_dev = PCI_DEVICE(s);
-MegasasBaseClass *base_class = MEGASAS_DEVICE_GET_CLASS(s);
+MegasasBaseClass *base_class = MEGASAS_GET_CLASS(s);
 uint32_t retval = 0;
 
 switch (addr) {
@@ -2322,7 +2322,7 @@ static const struct SCSIBusInfo megasas_scsi_info = {
 static void megasas_scsi_realize(PCIDevice *dev, Error **errp)
 {
 MegasasState *s = MEGASAS(dev);
-MegasasBaseClass *b = MEGASAS_DEVICE_GET_CLASS(s);
+MegasasBaseClass *b = MEGASAS_GET_CLASS(s);
 uint8_t *pci_conf;
 int i, bar_type;
 Error *err = NULL;
@@ -2506,7 +2506,7 @@ static void megasas_class_init(ObjectClass *oc, void 
*data)
 {
 DeviceClass *dc = DEVICE_CLASS(oc);
 PCIDeviceClass *pc = PCI_DEVICE_CLASS(oc);
-MegasasBaseClass *e = MEGASAS_DEVICE_CLASS(oc);
+MegasasBaseClass *e = MEGASAS_CLASS(oc);
 const MegasasInfo *info = data;
 
 pc->realize = megasas_scsi_realize;
-- 
2.26.2

Re: Suspicious QOM types without instance/class size

[Eduardo Habkost]

On Mon, Aug 24, 2020 at 08:06:42PM +0300, Roman Bolshakov wrote:
> On Mon, Aug 24, 2020 at 12:45:52PM -0400, Eduardo Habkost wrote:
> > On Mon, Aug 24, 2020 at 07:41:34PM +0300, Roman Bolshakov wrote:
> > > On Fri, Aug 21, 2020 at 01:48:02PM -0400, Eduardo Habkost wrote:
> > > > Signed-off-by: Eduardo Habkost 
> > > > ---
> > > > diff --git a/target/i386/hvf/hvf.c b/target/i386/hvf/hvf.c
> > > > index d81f569aed..81d1662d06 100644
> > > > --- a/target/i386/hvf/hvf.c
> > > > +++ b/target/i386/hvf/hvf.c
> > > > @@ -878,13 +878,11 @@ static int hvf_accel_init(MachineState *ms)
> > > >  {
> > > >  int x;
> > > >  hv_return_t ret;
> > > > -HVFState *s;
> > > > +HVFState *s = HVF_STATE(ms->accelerator);
> > > 
> > > The file also needs definition of MachineState:
> > > #include "hw/boards.h"
> > > 
> > > >  
> > > >  ret = hv_vm_create(HV_VM_DEFAULT);
> > > >  assert_hvf_ok(ret);
> > > >  
> > > > -s = g_new0(HVFState, 1);
> > > > - 
> > > >  s->num_slots = 32;
> > > >  for (x = 0; x < s->num_slots; ++x) {
> > > >  s->slots[x].size = 0;
> > > > @@ -908,6 +906,7 @@ static void hvf_accel_class_init(ObjectClass *oc, 
> > > > void *data)
> > > >  static const TypeInfo hvf_accel_type = {
> > > >  .name = TYPE_HVF_ACCEL,
> > > >  .parent = TYPE_ACCEL,
> > > > +.instance_size = sizeof(HVFState),
> > > >  .class_init = hvf_accel_class_init,
> > > >  };
> > > >  
> > > >  
> > 
> > However, the hvf patch above shouldn't require it.  You should be
> > able to apply and test it on top of qemu.git master.
> > 
> 
> Yeah, that's correct, thanks.
> 
> With the include fix for hw/boards.h, the patch works:
> Reviewed-By: Roman Bolshakov 
> Tested-By: Roman Bolshakov 
> 
> BTW, am I expected to see the accel in "info qtree" (or qom-tree)? It's
> not there for a reason.

I don't know if you are expect to see it.  I don't think there's
explicit code to attach the accel object to the user-visible QOM
tree.

-- 
Eduardo

Re: Suspicious QOM types without instance/class size

[Eduardo Habkost]

On Mon, Aug 24, 2020 at 07:41:34PM +0300, Roman Bolshakov wrote:
> On Fri, Aug 21, 2020 at 01:48:02PM -0400, Eduardo Habkost wrote:
> > On Fri, Aug 21, 2020 at 01:29:38PM -0400, Eduardo Habkost wrote:
> > > On Fri, Aug 21, 2020 at 01:53:52PM +0300, Roman Bolshakov wrote:
> > > > On Thu, Aug 20, 2020 at 05:55:29PM -0400, Eduardo Habkost wrote:
> > > > > While trying to convert TypeInfo declarations to the new
> > > > > OBJECT_DECLARE* macros, I've stumbled on a few suspicious cases
> > > > > where instance_size or class_size is not set, despite having type
> > > > > checker macros that use a specific type.
> > > > > 
> > > > > The ones with "WARNING" are abstract types (maybe not serious if
> > > > > subclasses set the appropriate sizes).  The ones with "ERROR"
> > > > > don't seem to be abstract types.
> > > > > 
> > > > 
> > > > > ERROR: target/i386/hvf/hvf.c:908:1: instance_size should be set to 
> > > > > sizeof(HVFState)?
> > > > 
> > > 
> > > > BTW, the object definition for hvf seems different from KVM (and perhaps
> > > > wrong?), e.g. HVFState is allocated within init_machine handler and then
> > > > assigned to a global variable:
> > > 
> > > Interesting.  It looks like hvf_state is _not_ the actual QOM
> > > object instance.  The actual TYPE_HVF_ACCEL instance is created
> > > by do_configure_accelerator().  That would explain why the lack
> > > of instance_init never caused any problems.
> > > 
> > > Luckily, no code ever used the HVF_STATE macro.  If
> > > HVF_STATE(hvf_state) got called, it would crash because of
> > > uninitialized object instance data.  If HVF_STATE(machine->accel)
> > > got called, it would return an invalid HVFState pointer (not
> > > hvf_state).
> > > 
> > > I believe the simplest short term solution here is to just delete
> > > the HVF_STATE macro and HVFState::parent field.  We can worry
> > > about actually moving hvf_state to the machine->accel QOM object
> > > later.
> > 
> > Actually, it might be easier to do the full QOM conversion in a
> > single patch instead of deleting the incomplete code.
> > 
> 
> I agree full QOM conversion is better, perhaps we can later move
> certains bits to accel/hvf.c like it's done for kvm/tcg/qtest.
> 
> > Can you check if the following patch works?  I don't have a host
> > where I can test it.
> > 
> 
> Sure, thanks :)
> 
> > Signed-off-by: Eduardo Habkost 
> > ---
> > diff --git a/target/i386/hvf/hvf.c b/target/i386/hvf/hvf.c
> > index d81f569aed..81d1662d06 100644
> > --- a/target/i386/hvf/hvf.c
> > +++ b/target/i386/hvf/hvf.c
> > @@ -878,13 +878,11 @@ static int hvf_accel_init(MachineState *ms)
> >  {
> >  int x;
> >  hv_return_t ret;
> > -HVFState *s;
> > +HVFState *s = HVF_STATE(ms->accelerator);
> 
> The file also needs definition of MachineState:
> #include "hw/boards.h"
> 
> >  
> >  ret = hv_vm_create(HV_VM_DEFAULT);
> >  assert_hvf_ok(ret);
> >  
> > -s = g_new0(HVFState, 1);
> > - 
> >  s->num_slots = 32;
> >  for (x = 0; x < s->num_slots; ++x) {
> >  s->slots[x].size = 0;
> > @@ -908,6 +906,7 @@ static void hvf_accel_class_init(ObjectClass *oc, void 
> > *data)
> >  static const TypeInfo hvf_accel_type = {
> >  .name = TYPE_HVF_ACCEL,
> >  .parent = TYPE_ACCEL,
> > +.instance_size = sizeof(HVFState),
> >  .class_init = hvf_accel_class_init,
> >  };
> >  
> >  
> 
> Unfortunately it fails to start (even without the HVF patch):
> ERROR:../qom/object.c:314:type_initialize: assertion failed: 
> (parent->class_size <= ti->class_size)
> Bail out! ERROR:../qom/object.c:314:type_initialize: assertion failed: 
> (parent->class_size <= ti->class_size)
[...]
> It doesn't seem related to HVF QOM changes 樂
> 
> Bisected it to:
> 
> b717702de21461138ac0e1d6775da0bd0482c013 is the first bad commit
> commit b717702de21461138ac0e1d6775da0bd0482c013
> Author: Daniel P. Berrangé 
> Date:   Wed Aug 19 20:12:35 2020 -0400
> 
> crypto: use QOM macros for declaration/definition of secret types
> 
> This introduces the use of the OBJECT_DEFINE and OBJECT_DECLARE macro
> families in the secret types, in order to eliminate boilerplate code.
> 
> Signed-off-by: Daniel P. Berrangé 
> Message-Id: <20200723181410.3145233-4-berra...@redhat.com>
> [ehabkost: rebase, update to pass additional arguments to macro]
> Signed-off-by: Eduardo Habkost 
> Message-Id: <20200820001236.1284548-58-ehabk...@redhat.com>

Oh, that's a bug in my QOM series.  Thanks for debugging it!  I
will fix it in v3.

However, the hvf patch above shouldn't require it.  You should be
able to apply and test it on top of qemu.git master.

-- 
Eduardo

Re: Suspicious QOM types without instance/class size

[Eduardo Habkost]

On Fri, Aug 21, 2020 at 11:43:35AM +0200, Cornelia Huck wrote:
> On Thu, 20 Aug 2020 17:55:29 -0400
> Eduardo Habkost  wrote:
> 
> > While trying to convert TypeInfo declarations to the new
> > OBJECT_DECLARE* macros, I've stumbled on a few suspicious cases
> > where instance_size or class_size is not set, despite having type
> > checker macros that use a specific type.
> > 
> > The ones with "WARNING" are abstract types (maybe not serious if
> > subclasses set the appropriate sizes).  The ones with "ERROR"
> > don't seem to be abstract types.
> > 
> > ERROR: hw/s390x/virtio-ccw.c:1237:1: class_size should be set to 
> > sizeof(VirtioCcwBusClass)?
> > ERROR: hw/virtio/virtio-pci.c:2101:1: class_size should be set to 
> > sizeof(VirtioPCIBusClass)?
> 
> VirtioCcwBusClass and VirtioPCIBusClass are both simple typedefs of
> VirtioBusClass (it's likely that I copied the ccw definition from the
> pci one). virtio-mmio instead uses VirtioBusClass directly in its
> checker macros.
> 
> I don't see a real reason for the typedefs, maybe ccw and pci should
> use the mmio approach as well?

I think it's OK to keep the typedefs if the code is consistent
(i.e. we set instance_size and class_size just in case the
typedefs are replaced by a real struct one day).

I'm not sure about the TYPE_VIRTIO_MMIO_BUS approach.  If the
code just needs VirtioBusState or VirtioBusClass pointers, it can
already use the VIRTIO_BUS* macros.

The OBJECT_DECLARE_TYPE macro Daniel sent expects each QOM type
to have a separate struct being defined, which isn't true in many
cases.  I'm considering removing the "typedef struct Foo Foo"
lines from OBJECT_DECLARE_TYPE(), to make initial conversion
easier.

-- 
Eduardo

Re: Suspicious QOM types without instance/class size

[Eduardo Habkost]

On Fri, Aug 21, 2020 at 11:40:12AM +0200, David Hildenbrand wrote:
> On 20.08.20 23:55, Eduardo Habkost wrote:
> > While trying to convert TypeInfo declarations to the new
> > OBJECT_DECLARE* macros, I've stumbled on a few suspicious cases
> > where instance_size or class_size is not set, despite having type
> > checker macros that use a specific type.
> > 
> > The ones with "WARNING" are abstract types (maybe not serious if
> > subclasses set the appropriate sizes).  The ones with "ERROR"
> > don't seem to be abstract types.
[...]
> > ERROR: hw/s390x/virtio-ccw.c:1237:1: class_size should be set to 
> > sizeof(VirtioCcwBusClass)?
> 
> The parent of TYPE_VIRTIO_CCW_BUS is TYPE_VIRTIO_BUS.
> 
> typedef struct VirtioBusClass VirtioCcwBusClass;
> 
> So I guess the sizes match? Anyhow, setting doesn't hurt.

Thanks for checking.  Yeah, the sizes match today.

It's a good idea to set it, just in case a real VirtioCcwBusClass
struct gets created one day.

-- 
Eduardo

Re: Suspicious QOM types without instance/class size

[Eduardo Habkost]

On Fri, Aug 21, 2020 at 11:47:32AM +1000, David Gibson wrote:
> On Thu, Aug 20, 2020 at 05:55:29PM -0400, Eduardo Habkost wrote:
> > While trying to convert TypeInfo declarations to the new
> > OBJECT_DECLARE* macros, I've stumbled on a few suspicious cases
> > where instance_size or class_size is not set, despite having type
> > checker macros that use a specific type.
> > 
> > The ones with "WARNING" are abstract types (maybe not serious if
> > subclasses set the appropriate sizes).  The ones with "ERROR"
> > don't seem to be abstract types.
> 
> 
> Comment on the ones within my area:
> > 
> > WARNING: hw/input/adb.c:310:1: class_size should be set to 
> > sizeof(ADBDeviceClass)?
> 
> Yeah, that looks like a bug (though we'll get away with it because
> it's abstract).

Right, luckily we are not touching any ADBDeviceClass field
inside adb_device_class_init().

> 
> > WARNING: hw/ppc/pnv_lpc.c:771:1: instance_size should be set to 
> > sizeof(PnvLpcController)?
> 
> Ditto.

Agreed.

> 
> Should I make fixes for these, or will you?

Please send the fixes, and I will apply them before running the
TypeInfo conversion script.

> 
> > ERROR: hw/ppc/spapr_drc.c:771:1: instance_size should be set to 
> > sizeof(SpaprDrc)?
> 
> I'm confused by this one.  I'm not exactly sure which definition is
> tripping the error, and AFAICT they should all be correctly inheriting
> instance_size from either TYPE_SPAPR_DR_CONNECTOR or
> TYPE_SPAPR_DRC_PHSYICAL.  If anything, it looks like
> TYPE_SPAPR_DRC_PHB could drop it's explicit override of instance_size.

The error is triggered because of this type checking macro at
include/hw/ppc/spapr_drc.h:

#define SPAPR_DRC_PCI(obj) OBJECT_CHECK(SpaprDrc, (obj), \
TYPE_SPAPR_DRC_PCI)

The expectation is that whatever type you use in OBJECT_CHECK
will be the one used for instance_size.  The script also looks at
the parent type, to reduce false positives, but this case was
flagged because SPAPR_DRC_PCI uses SpaprDrc, but the parent type
(SPAPR_DRC_PHYSICAL) uses SpaprDrcPhysical.

Now, I don't understand why we have so many instance checker
macros that use the same typedef (SpaprDrc).  If the code needs a
valid SpaprDrc pointer, it can just use SPAPR_DR_CONNECTOR().

-- 
Eduardo

Re: Suspicious QOM types without instance/class size

[Eduardo Habkost]

On Fri, Aug 21, 2020 at 01:29:38PM -0400, Eduardo Habkost wrote:
> On Fri, Aug 21, 2020 at 01:53:52PM +0300, Roman Bolshakov wrote:
> > On Thu, Aug 20, 2020 at 05:55:29PM -0400, Eduardo Habkost wrote:
> > > While trying to convert TypeInfo declarations to the new
> > > OBJECT_DECLARE* macros, I've stumbled on a few suspicious cases
> > > where instance_size or class_size is not set, despite having type
> > > checker macros that use a specific type.
> > > 
> > > The ones with "WARNING" are abstract types (maybe not serious if
> > > subclasses set the appropriate sizes).  The ones with "ERROR"
> > > don't seem to be abstract types.
> > > 
> > 
> > > ERROR: target/i386/hvf/hvf.c:908:1: instance_size should be set to 
> > > sizeof(HVFState)?
> > 
> > Hi Eduardo,
> > 
> > How do you get the error?
> 
> My script looks for corresponding type checking macros, and check
> if instance_size is set to sizeof(T) with the right type from the
> type checking macro.
> 
> The code is here:
> https://github.com/ehabkost/qemu-hacks/blob/920b2c521ad2a29fa663256854e24ed2059ba9cd/scripts/codeconverter/codeconverter/qom_type_info.py#L136
> 
> 
> > 
> > Given your changes, instance size should really be sizeof(HVFState).
> > 
> 
> The changes I've made shouldn't make any difference (if there's
> an issue, it is there before or after my series).
> 
> > BTW, the object definition for hvf seems different from KVM (and perhaps
> > wrong?), e.g. HVFState is allocated within init_machine handler and then
> > assigned to a global variable:
> 
> Interesting.  It looks like hvf_state is _not_ the actual QOM
> object instance.  The actual TYPE_HVF_ACCEL instance is created
> by do_configure_accelerator().  That would explain why the lack
> of instance_init never caused any problems.
> 
> Luckily, no code ever used the HVF_STATE macro.  If
> HVF_STATE(hvf_state) got called, it would crash because of
> uninitialized object instance data.  If HVF_STATE(machine->accel)
> got called, it would return an invalid HVFState pointer (not
> hvf_state).
> 
> I believe the simplest short term solution here is to just delete
> the HVF_STATE macro and HVFState::parent field.  We can worry
> about actually moving hvf_state to the machine->accel QOM object
> later.

Actually, it might be easier to do the full QOM conversion in a
single patch instead of deleting the incomplete code.

Can you check if the following patch works?  I don't have a host
where I can test it.

Signed-off-by: Eduardo Habkost 
---
diff --git a/target/i386/hvf/hvf.c b/target/i386/hvf/hvf.c
index d81f569aed..81d1662d06 100644
--- a/target/i386/hvf/hvf.c
+++ b/target/i386/hvf/hvf.c
@@ -878,13 +878,11 @@ static int hvf_accel_init(MachineState *ms)
 {
 int x;
 hv_return_t ret;
-HVFState *s;
+HVFState *s = HVF_STATE(ms->accelerator);
 
 ret = hv_vm_create(HV_VM_DEFAULT);
 assert_hvf_ok(ret);
 
-s = g_new0(HVFState, 1);
- 
 s->num_slots = 32;
 for (x = 0; x < s->num_slots; ++x) {
 s->slots[x].size = 0;
@@ -908,6 +906,7 @@ static void hvf_accel_class_init(ObjectClass *oc, void 
*data)
 static const TypeInfo hvf_accel_type = {
 .name = TYPE_HVF_ACCEL,
 .parent = TYPE_ACCEL,
+.instance_size = sizeof(HVFState),
 .class_init = hvf_accel_class_init,
 };
 
 
-- 
Eduardo

Re: Suspicious QOM types without instance/class size

[Eduardo Habkost]

On Fri, Aug 21, 2020 at 01:53:52PM +0300, Roman Bolshakov wrote:
> On Thu, Aug 20, 2020 at 05:55:29PM -0400, Eduardo Habkost wrote:
> > While trying to convert TypeInfo declarations to the new
> > OBJECT_DECLARE* macros, I've stumbled on a few suspicious cases
> > where instance_size or class_size is not set, despite having type
> > checker macros that use a specific type.
> > 
> > The ones with "WARNING" are abstract types (maybe not serious if
> > subclasses set the appropriate sizes).  The ones with "ERROR"
> > don't seem to be abstract types.
> > 
> 
> > ERROR: target/i386/hvf/hvf.c:908:1: instance_size should be set to 
> > sizeof(HVFState)?
> 
> Hi Eduardo,
> 
> How do you get the error?

My script looks for corresponding type checking macros, and check
if instance_size is set to sizeof(T) with the right type from the
type checking macro.

The code is here:
https://github.com/ehabkost/qemu-hacks/blob/920b2c521ad2a29fa663256854e24ed2059ba9cd/scripts/codeconverter/codeconverter/qom_type_info.py#L136


> 
> Given your changes, instance size should really be sizeof(HVFState).
> 

The changes I've made shouldn't make any difference (if there's
an issue, it is there before or after my series).

> BTW, the object definition for hvf seems different from KVM (and perhaps
> wrong?), e.g. HVFState is allocated within init_machine handler and then
> assigned to a global variable:

Interesting.  It looks like hvf_state is _not_ the actual QOM
object instance.  The actual TYPE_HVF_ACCEL instance is created
by do_configure_accelerator().  That would explain why the lack
of instance_init never caused any problems.

Luckily, no code ever used the HVF_STATE macro.  If
HVF_STATE(hvf_state) got called, it would crash because of
uninitialized object instance data.  If HVF_STATE(machine->accel)
got called, it would return an invalid HVFState pointer (not
hvf_state).

I believe the simplest short term solution here is to just delete
the HVF_STATE macro and HVFState::parent field.  We can worry
about actually moving hvf_state to the machine->accel QOM object
later.

> 
> static int hvf_accel_init(MachineState *ms)
> {
> int x;
> hv_return_t ret;
> HVFState *s;
> 
> ret = hv_vm_create(HV_VM_DEFAULT);
> assert_hvf_ok(ret);
> 
> s = g_new0(HVFState, 1);
>  
> s->num_slots = 32;
> for (x = 0; x < s->num_slots; ++x) {
> s->slots[x].size = 0;
> s->slots[x].slot_id = x;
> }
>   
> hvf_state = s;
> cpu_interrupt_handler = hvf_handle_interrupt;
> memory_listener_register(_memory_listener, _space_memory);
> return 0;
> }
> 
> static void hvf_accel_class_init(ObjectClass *oc, void *data)
> {
> AccelClass *ac = ACCEL_CLASS(oc);
> ac->name = "HVF";
> ac->init_machine = hvf_accel_init;
> ac->allowed = _allowed;
> }
> 
> static const TypeInfo hvf_accel_type = {
> .name = TYPE_HVF_ACCEL,
> .parent = TYPE_ACCEL,
> .class_init = hvf_accel_class_init,
> };
> 
> Thanks,
> Roman
> 

-- 
Eduardo

Re: Suspicious QOM types without instance/class size

[Eduardo Habkost]

On Fri, Aug 21, 2020 at 09:06:51AM -0700, Alistair Francis wrote:
> On Thu, Aug 20, 2020 at 2:56 PM Eduardo Habkost  wrote:
> >
> > While trying to convert TypeInfo declarations to the new
> > OBJECT_DECLARE* macros, I've stumbled on a few suspicious cases
> > where instance_size or class_size is not set, despite having type
> > checker macros that use a specific type.
> >
> > The ones with "WARNING" are abstract types (maybe not serious if
> > subclasses set the appropriate sizes).  The ones with "ERROR"
> > don't seem to be abstract types.
> >
> 
> > ERROR: hw/core/register.c:328:1: instance_size should be set to 
> > sizeof(RegisterInfo)?
> 
> I'll send a patch out for this one today.
> 
> If you are fixing all of these as part of a series I'm also happy to
> just let you do that.

Feel free to send the fix, and I will include it as part of my
series if necessary.

Note that register_init_block() relies on the fact that
register_init() won't touch any RegisterInfo field except
parent_obj, so this won't be a one line patch.

-- 
Eduardo

Suspicious QOM types without instance/class size

[Eduardo Habkost]

While trying to convert TypeInfo declarations to the new
OBJECT_DECLARE* macros, I've stumbled on a few suspicious cases
where instance_size or class_size is not set, despite having type
checker macros that use a specific type.

The ones with "WARNING" are abstract types (maybe not serious if
subclasses set the appropriate sizes).  The ones with "ERROR"
don't seem to be abstract types.

WARNING: hw/arm/armsse.c:1159:1: class_size should be set to 
sizeof(ARMSSEClass)?
WARNING: hw/audio/hda-codec.c:900:1: instance_size should be set to 
sizeof(HDAAudioState)?
ERROR: hw/core/register.c:328:1: instance_size should be set to 
sizeof(RegisterInfo)?
WARNING: hw/input/adb.c:310:1: class_size should be set to 
sizeof(ADBDeviceClass)?
WARNING: hw/isa/isa-superio.c:181:1: instance_size should be set to 
sizeof(ISASuperIODevice)?
WARNING: hw/ppc/pnv_lpc.c:771:1: instance_size should be set to 
sizeof(PnvLpcController)?
ERROR: hw/ppc/spapr_drc.c:771:1: instance_size should be set to 
sizeof(SpaprDrc)?
WARNING: hw/rtc/m48t59-isa.c:156:1: class_size should be set to 
sizeof(M48txxISADeviceClass)?
WARNING: hw/rtc/m48t59.c:691:1: class_size should be set to 
sizeof(M48txxSysBusDeviceClass)?
ERROR: hw/s390x/virtio-ccw.c:1237:1: class_size should be set to 
sizeof(VirtioCcwBusClass)?
WARNING: hw/ssi/ssi.c:88:1: instance_size should be set to sizeof(SSISlave)?
ERROR: hw/virtio/virtio-pci.c:2101:1: class_size should be set to 
sizeof(VirtioPCIBusClass)?
WARNING: scsi/pr-manager.c:76:1: instance_size should be set to 
sizeof(PRManager)?
ERROR: target/i386/hvf/hvf.c:908:1: instance_size should be set to 
sizeof(HVFState)?

-- 
Eduardo

[PATCH v2 32/58] ahci: Move QOM macro to header

[Eduardo Habkost]

Move the ALLWINNER_AHCI macro close to the TYPE_ALLWINNER_AHCI
define.

This will make future conversion to OBJECT_DECLARE* easier.

Signed-off-by: Eduardo Habkost 
---
Changes series v1 -> v2: new patch in series v2

Cc: John Snow 
Cc: qemu-block@nongnu.org
Cc: qemu-de...@nongnu.org
---
 include/hw/ide/ahci.h   | 2 ++
 hw/ide/ahci-allwinner.c | 3 ---
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/include/hw/ide/ahci.h b/include/hw/ide/ahci.h
index ce2bf8a5f8..41bb517047 100644
--- a/include/hw/ide/ahci.h
+++ b/include/hw/ide/ahci.h
@@ -72,6 +72,8 @@ typedef struct SysbusAHCIState {
 } SysbusAHCIState;
 
 #define TYPE_ALLWINNER_AHCI "allwinner-ahci"
+#define ALLWINNER_AHCI(obj) \
+OBJECT_CHECK(AllwinnerAHCIState, (obj), TYPE_ALLWINNER_AHCI)
 
 #define ALLWINNER_AHCI_MMIO_OFF  0x80
 #define ALLWINNER_AHCI_MMIO_SIZE 0x80
diff --git a/hw/ide/ahci-allwinner.c b/hw/ide/ahci-allwinner.c
index 8536b9eb5a..227e747ba7 100644
--- a/hw/ide/ahci-allwinner.c
+++ b/hw/ide/ahci-allwinner.c
@@ -25,9 +25,6 @@
 
 #include "trace.h"
 
-#define ALLWINNER_AHCI(obj) \
-OBJECT_CHECK(AllwinnerAHCIState, (obj), TYPE_ALLWINNER_AHCI)
-
 #define ALLWINNER_AHCI_BISTAFR((0xa0 - ALLWINNER_AHCI_MMIO_OFF) / 4)
 #define ALLWINNER_AHCI_BISTCR ((0xa4 - ALLWINNER_AHCI_MMIO_OFF) / 4)
 #define ALLWINNER_AHCI_BISTFCTR   ((0xa8 - ALLWINNER_AHCI_MMIO_OFF) / 4)
-- 
2.26.2

[PATCH v2 16/58] throttle-groups: Move ThrottleGroup typedef to header

[Eduardo Habkost]

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 
Reviewed-by: Daniel P. Berrangé 
Signed-off-by: Eduardo Habkost 
---
Changes v1 -> v2: none

---
Cc: Alberto Garcia 
Cc: Kevin Wolf 
Cc: Max Reitz 
Cc: qemu-block@nongnu.org
Cc: qemu-de...@nongnu.org
---
 include/block/throttle-groups.h | 1 +
 block/throttle-groups.c | 4 ++--
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/include/block/throttle-groups.h b/include/block/throttle-groups.h
index 712a8e64b4..5e77db700f 100644
--- a/include/block/throttle-groups.h
+++ b/include/block/throttle-groups.h
@@ -59,6 +59,7 @@ typedef struct ThrottleGroupMember {
 } ThrottleGroupMember;
 
 #define TYPE_THROTTLE_GROUP "throttle-group"
+typedef struct ThrottleGroup ThrottleGroup;
 #define THROTTLE_GROUP(obj) OBJECT_CHECK(ThrottleGroup, (obj), 
TYPE_THROTTLE_GROUP)
 
 const char *throttle_group_get_name(ThrottleGroupMember *tgm);
diff --git a/block/throttle-groups.c b/block/throttle-groups.c
index 98fea7fd47..4e28365d8d 100644
--- a/block/throttle-groups.c
+++ b/block/throttle-groups.c
@@ -63,7 +63,7 @@ static void timer_cb(ThrottleGroupMember *tgm, bool is_write);
  * access some other ThrottleGroupMember's timers only after verifying that
  * that ThrottleGroupMember has throttled requests in the queue.
  */
-typedef struct ThrottleGroup {
+struct ThrottleGroup {
 Object parent_obj;
 
 /* refuse individual property change if initialization is complete */
@@ -79,7 +79,7 @@ typedef struct ThrottleGroup {
 
 /* This field is protected by the global QEMU mutex */
 QTAILQ_ENTRY(ThrottleGroup) list;
-} ThrottleGroup;
+};
 
 /* This is protected by the global QEMU mutex */
 static QTAILQ_HEAD(, ThrottleGroup) throttle_groups =
-- 
2.26.2

[PATCH v2 30/58] ahci: Move QOM macros to header

[Eduardo Habkost]

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.

Signed-off-by: Eduardo Habkost 
---
Changes series v1 -> v2: new patch in series v2

Cc: John Snow 
Cc: qemu-block@nongnu.org
Cc: qemu-de...@nongnu.org
---
 hw/ide/ahci_internal.h | 5 -
 include/hw/ide/ahci.h  | 3 +++
 2 files changed, 3 insertions(+), 5 deletions(-)

diff --git a/hw/ide/ahci_internal.h b/hw/ide/ahci_internal.h
index bab0459774..ac9bdead7b 100644
--- a/hw/ide/ahci_internal.h
+++ b/hw/ide/ahci_internal.h
@@ -332,9 +332,6 @@ struct AHCIPCIState {
 AHCIState ahci;
 };
 
-#define ICH_AHCI(obj) \
-OBJECT_CHECK(AHCIPCIState, (obj), TYPE_ICH9_AHCI)
-
 extern const VMStateDescription vmstate_ahci;
 
 #define VMSTATE_AHCI(_field, _state) {   \
@@ -394,6 +391,4 @@ void ahci_uninit(AHCIState *s);
 
 void ahci_reset(AHCIState *s);
 
-#define SYSBUS_AHCI(obj) OBJECT_CHECK(SysbusAHCIState, (obj), TYPE_SYSBUS_AHCI)
-
 #endif /* HW_IDE_AHCI_INTERNAL_H */
diff --git a/include/hw/ide/ahci.h b/include/hw/ide/ahci.h
index b44e3000cf..ce2bf8a5f8 100644
--- a/include/hw/ide/ahci.h
+++ b/include/hw/ide/ahci.h
@@ -53,11 +53,14 @@ typedef struct AHCIState {
 typedef struct AHCIPCIState AHCIPCIState;
 
 #define TYPE_ICH9_AHCI "ich9-ahci"
+#define ICH_AHCI(obj) \
+OBJECT_CHECK(AHCIPCIState, (obj), TYPE_ICH9_AHCI)
 
 int32_t ahci_get_num_ports(PCIDevice *dev);
 void ahci_ide_create_devs(PCIDevice *dev, DriveInfo **hd);
 
 #define TYPE_SYSBUS_AHCI "sysbus-ahci"
+#define SYSBUS_AHCI(obj) OBJECT_CHECK(SysbusAHCIState, (obj), TYPE_SYSBUS_AHCI)
 
 typedef struct SysbusAHCIState {
 /*< private >*/
-- 
2.26.2

[PATCH v2 02/58] megasas: Rename QOM class cast macros

[Eduardo Habkost]

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 
Signed-off-by: Eduardo Habkost 
---
Changes v1 -> v2: none

---
Cc: Paolo Bonzini 
Cc: Fam Zheng 
Cc: Hannes Reinecke 
Cc: qemu-block@nongnu.org
Cc: qemu-de...@nongnu.org
---
 hw/scsi/megasas.c | 12 ++--
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/hw/scsi/megasas.c b/hw/scsi/megasas.c
index 5cfd1bf22e..390c2f2edb 100644
--- a/hw/scsi/megasas.c
+++ b/hw/scsi/megasas.c
@@ -134,9 +134,9 @@ typedef struct MegasasBaseClass {
 #define MEGASAS(obj) \
 OBJECT_CHECK(MegasasState, (obj), TYPE_MEGASAS_BASE)
 
-#define MEGASAS_DEVICE_CLASS(oc) \
+#define MEGASAS_CLASS(oc) \
 OBJECT_CLASS_CHECK(MegasasBaseClass, (oc), TYPE_MEGASAS_BASE)
-#define MEGASAS_DEVICE_GET_CLASS(oc) \
+#define MEGASAS_GET_CLASS(oc) \
 OBJECT_GET_CLASS(MegasasBaseClass, (oc), TYPE_MEGASAS_BASE)
 
 #define MEGASAS_INTR_DISABLED_MASK 0x
@@ -733,7 +733,7 @@ static int megasas_ctrl_get_info(MegasasState *s, 
MegasasCmd *cmd)
 {
 PCIDevice *pci_dev = PCI_DEVICE(s);
 PCIDeviceClass *pci_class = PCI_DEVICE_GET_CLASS(pci_dev);
-MegasasBaseClass *base_class = MEGASAS_DEVICE_GET_CLASS(s);
+MegasasBaseClass *base_class = MEGASAS_GET_CLASS(s);
 struct mfi_ctrl_info info;
 size_t dcmd_size = sizeof(info);
 BusChild *kid;
@@ -1999,7 +1999,7 @@ static uint64_t megasas_mmio_read(void *opaque, hwaddr 
addr,
 {
 MegasasState *s = opaque;
 PCIDevice *pci_dev = PCI_DEVICE(s);
-MegasasBaseClass *base_class = MEGASAS_DEVICE_GET_CLASS(s);
+MegasasBaseClass *base_class = MEGASAS_GET_CLASS(s);
 uint32_t retval = 0;
 
 switch (addr) {
@@ -2322,7 +2322,7 @@ static const struct SCSIBusInfo megasas_scsi_info = {
 static void megasas_scsi_realize(PCIDevice *dev, Error **errp)
 {
 MegasasState *s = MEGASAS(dev);
-MegasasBaseClass *b = MEGASAS_DEVICE_GET_CLASS(s);
+MegasasBaseClass *b = MEGASAS_GET_CLASS(s);
 uint8_t *pci_conf;
 int i, bar_type;
 Error *err = NULL;
@@ -2506,7 +2506,7 @@ static void megasas_class_init(ObjectClass *oc, void 
*data)
 {
 DeviceClass *dc = DEVICE_CLASS(oc);
 PCIDeviceClass *pc = PCI_DEVICE_CLASS(oc);
-MegasasBaseClass *e = MEGASAS_DEVICE_CLASS(oc);
+MegasasBaseClass *e = MEGASAS_CLASS(oc);
 const MegasasInfo *info = data;
 
 pc->realize = megasas_scsi_realize;
-- 
2.26.2

Re: [PATCH] ide: Get rid of IDEDrive struct

[Eduardo Habkost]

On Wed, Aug 05, 2020 at 09:41:25PM +0100, Peter Maydell wrote:
> On Wed, 5 Aug 2020 at 20:49, Eduardo Habkost  wrote:
> >
> > The struct had a single field (IDEDevice dev), and is only used
> > in the QOM type declarations and property lists.  We can simply
> > use the IDEDevice struct directly instead.
> >
> > Signed-off-by: Eduardo Habkost 
> > @@ -327,7 +323,6 @@ static void ide_hd_class_init(ObjectClass *klass, void 
> > *data)
> >  static const TypeInfo ide_hd_info = {
> >  .name  = "ide-hd",
> >  .parent= TYPE_IDE_DEVICE,
> > -.instance_size = sizeof(IDEDrive),
> >  .class_init= ide_hd_class_init,
> >  };
> 
> This is one of those areas where this change works and reduces
> amount of code, but on the other hand it means the QOM type
> doesn't follow the common pattern for a leaf type of:
>  * it has a struct
>  * it has cast macros that cast to that struct
>  * the typeinfo instance_size is the size of that struct
> (it wasn't exactly following this pattern before, of course).

Is this really a pattern that exists and we want to follow?
I don't see why that pattern would be useful for simple leaf
types.

Also, in this case the code wasn't even following that pattern:
it was using the same IDEDrive struct for all TYPE_IDE_DEVICE
subtypes.

> 
> We define in https://wiki.qemu.org/Documentation/QOMConventions
> (in the 'When to create class types and macros' bit at the bottom)
> what we expect for whether to provide class cast macros/a
> class struct/class_size in the TypeInfo, essentially recommending
> that types follow one of two patterns (simple leaf class with no
> methods or class members, vs everything else) even if in a
> particular case you could take a short-cut and not define
> everything. We haven't really defined similar "this is the
> standard pattern, provide it all even if you don't strictly
> need it" rules for the instance struct/macros. Maybe we should?

I think we should include the instance struct/macros in the
recommendations there, but I would expect those recommendations
to apply only to non-leaf types.

> 
> Just a thought, not a nak; I know we have quite a number
> of types that take this kind of "we don't really need to
> provide all the standard QOM macros/structs/etc" approach
> (some of which I wrote!).
> 
> thanks
> -- PMM
> 

-- 
Eduardo

[PATCH] ide: Get rid of IDEDrive struct

[Eduardo Habkost]

The struct had a single field (IDEDevice dev), and is only used
in the QOM type declarations and property lists.  We can simply
use the IDEDevice struct directly instead.

Signed-off-by: Eduardo Habkost 
---
 hw/ide/qdev.c | 25 +
 1 file changed, 9 insertions(+), 16 deletions(-)

diff --git a/hw/ide/qdev.c b/hw/ide/qdev.c
index 27ff1f7f66..dd3867d8b3 100644
--- a/hw/ide/qdev.c
+++ b/hw/ide/qdev.c
@@ -157,10 +157,6 @@ int ide_get_bios_chs_trans(BusState *bus, int unit)
 
 /* - */
 
-typedef struct IDEDrive {
-IDEDevice dev;
-} IDEDrive;
-
 static void ide_dev_initfn(IDEDevice *dev, IDEDriveKind kind, Error **errp)
 {
 IDEBus *bus = DO_UPCAST(IDEBus, qbus, dev->qdev.parent_bus);
@@ -297,19 +293,19 @@ static void ide_drive_realize(IDEDevice *dev, Error 
**errp)
 }
 
 #define DEFINE_IDE_DEV_PROPERTIES() \
-DEFINE_BLOCK_PROPERTIES(IDEDrive, dev.conf),\
-DEFINE_BLOCK_ERROR_PROPERTIES(IDEDrive, dev.conf),  \
-DEFINE_PROP_STRING("ver",  IDEDrive, dev.version),  \
-DEFINE_PROP_UINT64("wwn",  IDEDrive, dev.wwn, 0),   \
-DEFINE_PROP_STRING("serial",  IDEDrive, dev.serial),\
-DEFINE_PROP_STRING("model", IDEDrive, dev.model)
+DEFINE_BLOCK_PROPERTIES(IDEDevice, conf),\
+DEFINE_BLOCK_ERROR_PROPERTIES(IDEDevice, conf),  \
+DEFINE_PROP_STRING("ver",  IDEDevice, version),  \
+DEFINE_PROP_UINT64("wwn",  IDEDevice, wwn, 0),   \
+DEFINE_PROP_STRING("serial",  IDEDevice, serial),\
+DEFINE_PROP_STRING("model", IDEDevice, model)
 
 static Property ide_hd_properties[] = {
 DEFINE_IDE_DEV_PROPERTIES(),
-DEFINE_BLOCK_CHS_PROPERTIES(IDEDrive, dev.conf),
+DEFINE_BLOCK_CHS_PROPERTIES(IDEDevice, conf),
 DEFINE_PROP_BIOS_CHS_TRANS("bios-chs-trans",
-IDEDrive, dev.chs_trans, BIOS_ATA_TRANSLATION_AUTO),
-DEFINE_PROP_UINT16("rotation_rate", IDEDrive, dev.rotation_rate, 0),
+IDEDevice, chs_trans, BIOS_ATA_TRANSLATION_AUTO),
+DEFINE_PROP_UINT16("rotation_rate", IDEDevice, rotation_rate, 0),
 DEFINE_PROP_END_OF_LIST(),
 };
 
@@ -327,7 +323,6 @@ static void ide_hd_class_init(ObjectClass *klass, void 
*data)
 static const TypeInfo ide_hd_info = {
 .name  = "ide-hd",
 .parent= TYPE_IDE_DEVICE,
-.instance_size = sizeof(IDEDrive),
 .class_init= ide_hd_class_init,
 };
 
@@ -350,7 +345,6 @@ static void ide_cd_class_init(ObjectClass *klass, void 
*data)
 static const TypeInfo ide_cd_info = {
 .name  = "ide-cd",
 .parent= TYPE_IDE_DEVICE,
-.instance_size = sizeof(IDEDrive),
 .class_init= ide_cd_class_init,
 };
 
@@ -373,7 +367,6 @@ static void ide_drive_class_init(ObjectClass *klass, void 
*data)
 static const TypeInfo ide_drive_info = {
 .name  = "ide-drive",
 .parent= TYPE_IDE_DEVICE,
-.instance_size = sizeof(IDEDrive),
 .class_init= ide_drive_class_init,
 };
 
-- 
2.26.2

Re: cleanups with long-term benefits

[Eduardo Habkost]

On Wed, Aug 05, 2020 at 06:23:23PM +0200, Kevin Wolf wrote:
> Am 05.08.2020 um 12:08 hat Daniel P. Berrangé geschrieben:
> > On Wed, Aug 05, 2020 at 11:11:55AM +0200, Cornelia Huck wrote:
> > > On Wed, 5 Aug 2020 10:05:40 +0100
> > > Daniel P. Berrangé  wrote:
> > > 
> > > > On Wed, Aug 05, 2020 at 10:49:35AM +0200, Paolo Bonzini wrote:
> > > > > On 05/08/20 10:39, Dr. David Alan Gilbert wrote:  
> > > > > >> Do you really use "git blame" that much?  "git log -S" does more 
> > > > > >> or less
> > > > > >> the same function (in a different way) and is not affected as much 
> > > > > >> by
> > > > > >> large code movement and transformation patches.  
> > > > > >
> > > > > > I use it a lot!   Following stuff back to find where a change came
> > > > > > from and then asking people.  
> > > > > 
> > > > > Indeed, but I use "git log -S" instead. :)  Another possibility is to
> > > > > just do "git log -p" and search for a relevant line of the code I'm
> > > > > "blaming".  
> > > > 
> > > > I used git blame alot too, but I don't think its a reason to not do the
> > > > cleanups. It is easy enough to just tell blame to use an earlier commit
> > > > if you see it displaying a refactor. I don't think such mild 
> > > > inconvenience
> > > > should stop us making otherwise desirable code changes
> > > 
> > > I don't think people argue that it should block changes; it it simply
> > > another thing to consider when weighing benefits vs. drawbacks.
> > 
> > Actually, I'm saying that including "git blame" in such a weighing is
> > the mechanism for blocking otherwise beneficial change to the code. Or
> > to put it another way, I believe the value assigned to "git blame" in
> > such a comparison is miniscule / effectively zero. The only time
> > "git blame" should win is if the change in question is purely the
> > bike shed colour and has no technical benefits at all.  If there's any
> > technical benefit to making the change it should always win.  We
> > shouldn't preserve technical debt in the code merely to avoid an impact
> > on "git blame".
> 
> We're basically weighing "git blame" against syntax highlighting
> defaults. I don't think the latter has an obviously higher weight.

I think "syntax highlight defaults" is far from being an accurate
description of the motivations behind this discussion.

-- 
Eduardo

Re: [PATCH v6] scripts/simplebench: compare write request performance

[Eduardo Habkost]

On Tue, Jul 14, 2020 at 07:05:05PM +0300, Vladimir Sementsov-Ogievskiy wrote:
> Hi Eduardo!
> 
> Could you please stage this in your branch, to be pulled for 5.2 later?

Queued, thanks!

> 
> 14.07.2020 18:50, Andrey Shinkevich wrote:
> > The script 'bench_write_req.py' allows comparing performances of write
> > request for two qemu-img binary files.
> > An example with (qemu-img binary 1) and without (qemu-img binary 2) the
> > applied patch "qcow2: skip writing zero buffers to empty COW areas"
> > (git commit ID: c8bb23cbdbe32f5) has the following results:
> > 
> > SSD:
> >   ---  ---
> >  
> >0.10 +- 0.00 8.16 +- 0.65
> >   0.10 +- 0.00 7.37 +- 1.10
> >7.40 +- 1.08 21.97 +- 4.19
> >  2.14 +- 0.94 8.48 +- 1.66
> >   ---  ---
> > HDD:
> >   ---  ---
> >  
> >2.30 +- 0.01 6.19 +- 0.06
> >   2.20 +- 0.09 6.20 +- 0.06
> >8.32 +- 0.16 8.26 +- 0.14
> >  8.20 +- 0.05 6.26 +- 0.10
> >   ---  ---
> > 
> > Suggested-by: Denis V. Lunev 
> > Suggested-by: Vladimir Sementsov-Ogievskiy 
> > Signed-off-by: Andrey Shinkevich 
> > Reviewed-by: Vladimir Sementsov-Ogievskiy 
> > ---
> >   scripts/simplebench/bench_write_req.py | 170 
> > +
> >   1 file changed, 170 insertions(+)
> >   create mode 100755 scripts/simplebench/bench_write_req.py
> > 
> > diff --git a/scripts/simplebench/bench_write_req.py 
> > b/scripts/simplebench/bench_write_req.py
> > new file mode 100755
> > index 000..ca1178f
> > --- /dev/null
> > +++ b/scripts/simplebench/bench_write_req.py
> > @@ -0,0 +1,170 @@
> > +#!/usr/bin/env python3
> > +#
> > +# Test to compare performance of write requests for two qemu-img binary 
> > files.
> > +#
> > +# The idea of the test comes from intention to check the benefit of 
> > c8bb23cbdbe
> > +# "qcow2: skip writing zero buffers to empty COW areas".
> > +#
> > +# Copyright (c) 2020 Virtuozzo International GmbH.
> > +#
> > +# This program is free software; you can redistribute it and/or modify
> > +# it under the terms of the GNU General Public License as published by
> > +# the Free Software Foundation; either version 2 of the License, or
> > +# (at your option) any later version.
> > +#
> > +# This program is distributed in the hope that it will be useful,
> > +# but WITHOUT ANY WARRANTY; without even the implied warranty of
> > +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> > +# GNU General Public License for more details.
> > +#
> > +# You should have received a copy of the GNU General Public License
> > +# along with this program.  If not, see .
> > +#
> > +
> > +
> > +import sys
> > +import os
> > +import subprocess
> > +import simplebench
> > +
> > +
> > +def bench_func(env, case):
> > +""" Handle one "cell" of benchmarking table. """
> > +return bench_write_req(env['qemu_img'], env['image_name'],
> > +   case['block_size'], case['block_offset'],
> > +   case['cluster_size'])
> > +
> > +
> > +def qemu_img_pipe(*args):
> > +'''Run qemu-img and return its output'''
> > +subp = subprocess.Popen(list(args),
> > +stdout=subprocess.PIPE,
> > +stderr=subprocess.STDOUT,
> > +universal_newlines=True)
> > +exitcode = subp.wait()
> > +if exitcode < 0:
> > +sys.stderr.write('qemu-img received signal %i: %s\n'
> > + % (-exitcode, ' '.join(list(args
> > +return subp.communicate()[0]
> > +
> > +
> > +def bench_write_req(qemu_img, image_name, block_size, block_offset,
> > +cluster_size):
> > +"""Benchmark write requests
> > +
> > +The function creates a QCOW2 image with the given path/name. Then it 
> > runs
> > +the 'qemu-img bench' command and makes series of write requests on the
> > +image clusters. Finally, it returns the total time of the write 
> > operations
> > +on the disk.
> > +
> > +qemu_img -- path to qemu_img executable file
> > +image_name   -- QCOW2 image name to create
> > +block_size   -- size of a block to write to clusters
> > +block_offset -- offset of the block in clusters
> > +cluster_size -- size of the image cluster
> > +
> > +Returns {'seconds': int} on success and {'error': str} on failure.
> > +Return value is compatible with simplebench lib.
> > +"""
> > +
> > +if not os.path.isfile(qemu_img):
> > +print(f'File not found: {qemu_img}')
> > +sys.exit(1)
> > +
> > +image_dir = os.path.dirname(os.path.abspath(image_name))
> > +if not os.path.isdir(image_dir):
> > +print(f'Path not found: 

Re: [PULL 00/12] Block patches

[Eduardo Habkost]

On Thu, Jul 09, 2020 at 05:02:06PM +0200, Kevin Wolf wrote:
> Am 08.07.2020 um 00:05 hat Eduardo Habkost geschrieben:
> > On Tue, Jul 07, 2020 at 05:28:21PM +0200, Philippe Mathieu-Daudé wrote:
> > > On 6/26/20 12:25 PM, Stefan Hajnoczi wrote:
> > > > On Thu, Jun 25, 2020 at 02:31:14PM +0100, Peter Maydell wrote:
> > > >> On Wed, 24 Jun 2020 at 11:02, Stefan Hajnoczi  
> > > >> wrote:
> > > >>>
> > > >>> The following changes since commit 
> > > >>> 171199f56f5f9bdf1e5d670d09ef1351d8f01bae:
> > > >>>
> > > >>>   Merge remote-tracking branch 
> > > >>> 'remotes/alistair/tags/pull-riscv-to-apply-20200619-3' into staging 
> > > >>> (2020-06-22 14:45:25 +0100)
> > > >>>
> > > >>> are available in the Git repository at:
> > > >>>
> > > >>>   https://github.com/stefanha/qemu.git tags/block-pull-request
> > > >>>
> > > >>> for you to fetch changes up to 
> > > >>> 7838c67f22a81fcf669785cd6c0876438422071a:
> > > >>>
> > > >>>   block/nvme: support nested aio_poll() (2020-06-23 15:46:08 +0100)
> > > >>>
> > > >>> 
> > > >>> Pull request
> > > >>>
> > > >>> 
> > > >>
> > > >> Failure on iotest 030, x86-64 Linux:
> > > >>
> > > >>   TESTiotest-qcow2: 030 [fail]
> > > >> QEMU  --
> > > >> "/home/petmay01/linaro/qemu-for-merges/build/alldbg/tests/qemu-iotests/../../x86_64-softmmu/qemu-system-x86_64"
> > > >> -nodefaults -display none -accel qtest
> > > >> QEMU_IMG  --
> > > >> "/home/petmay01/linaro/qemu-for-merges/build/alldbg/tests/qemu-iotests/../../qemu-img"
> > > >> QEMU_IO   --
> > > >> "/home/petmay01/linaro/qemu-for-merges/build/alldbg/tests/qemu-iotests/../../qemu-io"
> > > >>  --cache writeback --aio threads -f qcow2
> > > >> QEMU_NBD  --
> > > >> "/home/petmay01/linaro/qemu-for-merges/build/alldbg/tests/qemu-iotests/../../qemu-nbd"
> > > >> IMGFMT-- qcow2 (compat=1.1)
> > > >> IMGPROTO  -- file
> > > >> PLATFORM  -- Linux/x86_64 e104462 4.15.0-76-generic
> > > >> TEST_DIR  --
> > > >> /home/petmay01/linaro/qemu-for-merges/build/alldbg/tests/qemu-iotests/scratch
> > > >> SOCK_DIR  -- /tmp/tmp.8tgdDjoZcO
> > > >> SOCKET_SCM_HELPER --
> > > >> /home/petmay01/linaro/qemu-for-merges/build/alldbg/tests/qemu-iotest/socket_scm_helper
> > > >>
> > > >> --- /home/petmay01/linaro/qemu-for-merges/tests/qemu-iotests/030.out
> > > >>  2019-07-15 17:18:35.251364738 +0100
> > > >> +++ 
> > > >> /home/petmay01/linaro/qemu-for-merges/build/alldbg/tests/qemu-iotests/030.out.bad
> > > >>   2020-06-25 14:04:28.500534007 +0100
> > > >> @@ -1,5 +1,17 @@
> > > >> -...
> > > >> +.F.
> > > >> +==
> > > >> +FAIL: test_stream_parallel (__main__.TestParallelOps)
> > > >> +--
> > > >> +Traceback (most recent call last):
> > > >> +  File "030", line 246, in test_stream_parallel
> > > >> +self.assert_qmp(result, 'return', {})
> > > >> +  File 
> > > >> "/home/petmay01/linaro/qemu-for-merges/tests/qemu-iotests/iotests.py",
> > > >> line 848, in assert_qmp
> > > >> +result = self.dictpath(d, path)
> > > >> +  File 
> > > >> "/home/petmay01/linaro/qemu-for-merges/tests/qemu-iotests/iotests.py",
> > > >> line 822, in dictpath
> > > >> +self.fail(f'failed path traversal for "{path}" in "{d}"')
> > > >> +AssertionError: failed path traversal for "return" in "{'error':
> > > >> {'class': 'DeviceNotActive', 'desc': "Block job 'stream-node8' not
> > > >> found"}}"
> > > >> +
> > > >>  --

Re: [PULL 00/12] Block patches

[Eduardo Habkost]

On Tue, Jul 07, 2020 at 05:28:21PM +0200, Philippe Mathieu-Daudé wrote:
> On 6/26/20 12:25 PM, Stefan Hajnoczi wrote:
> > On Thu, Jun 25, 2020 at 02:31:14PM +0100, Peter Maydell wrote:
> >> On Wed, 24 Jun 2020 at 11:02, Stefan Hajnoczi  wrote:
> >>>
> >>> The following changes since commit 
> >>> 171199f56f5f9bdf1e5d670d09ef1351d8f01bae:
> >>>
> >>>   Merge remote-tracking branch 
> >>> 'remotes/alistair/tags/pull-riscv-to-apply-20200619-3' into staging 
> >>> (2020-06-22 14:45:25 +0100)
> >>>
> >>> are available in the Git repository at:
> >>>
> >>>   https://github.com/stefanha/qemu.git tags/block-pull-request
> >>>
> >>> for you to fetch changes up to 7838c67f22a81fcf669785cd6c0876438422071a:
> >>>
> >>>   block/nvme: support nested aio_poll() (2020-06-23 15:46:08 +0100)
> >>>
> >>> 
> >>> Pull request
> >>>
> >>> 
> >>
> >> Failure on iotest 030, x86-64 Linux:
> >>
> >>   TESTiotest-qcow2: 030 [fail]
> >> QEMU  --
> >> "/home/petmay01/linaro/qemu-for-merges/build/alldbg/tests/qemu-iotests/../../x86_64-softmmu/qemu-system-x86_64"
> >> -nodefaults -display none -accel qtest
> >> QEMU_IMG  --
> >> "/home/petmay01/linaro/qemu-for-merges/build/alldbg/tests/qemu-iotests/../../qemu-img"
> >> QEMU_IO   --
> >> "/home/petmay01/linaro/qemu-for-merges/build/alldbg/tests/qemu-iotests/../../qemu-io"
> >>  --cache writeback --aio threads -f qcow2
> >> QEMU_NBD  --
> >> "/home/petmay01/linaro/qemu-for-merges/build/alldbg/tests/qemu-iotests/../../qemu-nbd"
> >> IMGFMT-- qcow2 (compat=1.1)
> >> IMGPROTO  -- file
> >> PLATFORM  -- Linux/x86_64 e104462 4.15.0-76-generic
> >> TEST_DIR  --
> >> /home/petmay01/linaro/qemu-for-merges/build/alldbg/tests/qemu-iotests/scratch
> >> SOCK_DIR  -- /tmp/tmp.8tgdDjoZcO
> >> SOCKET_SCM_HELPER --
> >> /home/petmay01/linaro/qemu-for-merges/build/alldbg/tests/qemu-iotest/socket_scm_helper
> >>
> >> --- /home/petmay01/linaro/qemu-for-merges/tests/qemu-iotests/030.out
> >>  2019-07-15 17:18:35.251364738 +0100
> >> +++ 
> >> /home/petmay01/linaro/qemu-for-merges/build/alldbg/tests/qemu-iotests/030.out.bad
> >>   2020-06-25 14:04:28.500534007 +0100
> >> @@ -1,5 +1,17 @@
> >> -...
> >> +.F.
> >> +==
> >> +FAIL: test_stream_parallel (__main__.TestParallelOps)
> >> +--
> >> +Traceback (most recent call last):
> >> +  File "030", line 246, in test_stream_parallel
> >> +self.assert_qmp(result, 'return', {})
> >> +  File 
> >> "/home/petmay01/linaro/qemu-for-merges/tests/qemu-iotests/iotests.py",
> >> line 848, in assert_qmp
> >> +result = self.dictpath(d, path)
> >> +  File 
> >> "/home/petmay01/linaro/qemu-for-merges/tests/qemu-iotests/iotests.py",
> >> line 822, in dictpath
> >> +self.fail(f'failed path traversal for "{path}" in "{d}"')
> >> +AssertionError: failed path traversal for "return" in "{'error':
> >> {'class': 'DeviceNotActive', 'desc': "Block job 'stream-node8' not
> >> found"}}"
> >> +
> >>  --
> >>  Ran 27 tests
> >>
> >> -OK
> >> +FAILED (failures=1)
> > 
> > Strange, I can't reproduce this failure on my pull request branch or on
> > qemu.git/master.
> > 
> > Is this failure deterministic? Are you sure it is introduced by this
> > pull request?
> 
> Probably not introduced by this pullreq, but I also hit it on FreeBSD:
> https://cirrus-ci.com/task/4620718312783872?command=main#L5803
> 
>   TESTiotest-qcow2: 030 [fail]
> QEMU  --
> "/tmp/cirrus-ci-build/build/tests/qemu-iotests/../../aarch64-softmmu/qemu-system-aarch64"
> -nodefaults -display none -machine virt -accel qtest
> QEMU_IMG  --
> "/tmp/cirrus-ci-build/build/tests/qemu-iotests/../../qemu-img"
> QEMU_IO   --
> "/tmp/cirrus-ci-build/build/tests/qemu-iotests/../../qemu-io"  --cache
> writeback --aio threads -f qcow2
> QEMU_NBD  --
> "/tmp/cirrus-ci-build/build/tests/qemu-iotests/../../qemu-nbd"
> IMGFMT-- qcow2 (compat=1.1)
> IMGPROTO  -- file
> PLATFORM  -- FreeBSD/amd64 cirrus-task-4620718312783872 12.1-RELEASE
> TEST_DIR  -- /tmp/cirrus-ci-build/build/tests/qemu-iotests/scratch
> SOCK_DIR  -- /tmp/tmp.aZ5pxFLF
> SOCKET_SCM_HELPER --
> --- /tmp/cirrus-ci-build/tests/qemu-iotests/030.out   2020-07-07
> 14:48:48.123804000 +
> +++ /tmp/cirrus-ci-build/build/tests/qemu-iotests/030.out.bad 2020-07-07
> 15:05:07.863685000 +
> @@ -1,5 +1,17 @@
> -...
> +.F.
> +==
> +FAIL: test_stream_parallel (__main__.TestParallelOps)
>  --
> +Traceback (most recent call 

Re: [PATCH 0/2] block: propagate discard alignment from format drivers to the guest

[Eduardo Habkost]

On Thu, Jun 11, 2020 at 08:16:06PM +0300, Denis V. Lunev wrote:
> Nowaday SCSI drivers in guests are able to align UNMAP requests before
> sending to the device. Right now QEMU provides an ability to set
> this via "discard_granularity" property of the block device which could
> be used by management layer.
> 
> Though, in particular, from the point of QEMU, there is
> pdiscard_granularity on the format driver level, f.e. on QCOW2 or iSCSI.
> It would be beneficial to pass this value as a default for this
> property.

I assume the value is visible to the guest.  What is supposed to
happen if live migrating and the block backend is a different one
on the destination?

Also, don't we have mechanisms to change the block backend change
at run time?  What should happen in that case?

> 
> Technically this should reduce the amount of use less UNMAP requests
> from the guest to the host. Basic test confirms this. Fedora 31 guest
> during 'fstrim /' on 32 Gb disk has issued 401/415 requests with/without
> proper alignment to QEMU.
> 
> Changes from v2:
> - 172 iotest fixed
> 
> Changes from v1:
> - fixed typos in description
> - added machine type compatibility layer as suggested by Kevin
> 
> Signed-off-by: Denis V. Lunev 
> CC: Kevin Wolf 
> CC: Max Reitz 
> CC: Eduardo Habkost 
> CC: Marcel Apfelbaum 
> CC: John Snow 
> CC: Paolo Bonzini 
> CC: Fam Zheng 
> 
> 

-- 
Eduardo

Re: ping Re: [PATCH for-5.0 v2 0/3] benchmark util

[Eduardo Habkost]

Sorry, this is due to lack of bandwidth of maintainers who can
review those patches.

I have one suggestion: if you make your script self-contained
inside a scripts/ subdirectory, it would be simpler to merge it
without detailed reviews from others.

The python/ subdirectory is supposed to appear on sys.path, so
maybe simplebench.py and qemu/bench_block_job.py can stay there,
but bench-example.py is not a loadable Python module and
shouldn't be there.

I see two possible options:

a) Moving everything to a scripts/simplebench subdirectory.
b) Moving only bench-example.py to scripts/, and do the sys.path
   hacking the other scripts do.

On either case, please add your name to MAINTAINERS as the
maintainer of those new files.


On Thu, Feb 27, 2020 at 04:18:00PM +0300, Vladimir Sementsov-Ogievskiy wrote:
> Hi!
> 
> Is problem in "S: Odd fixes" in Python section of MAINTAINERS?
> 
> Will it be correct, if I send a patch to MAINTAINERS, proposing
> myself as maintainer of Python scripts and s/Odd fixes/Maintained/ ?
> 
> And then just send pull request with this series, as "nobody minds"?
> 
> 08.02.2020 13:36, Vladimir Sementsov-Ogievskiy wrote:
> > pingg..
> > 
> > Hi! Could it be merged at all?
> > 
> > 20.01.2020 12:10, Vladimir Sementsov-Ogievskiy wrote:
> > > ping
> > > 
> > > 26.11.2019 18:48, Vladimir Sementsov-Ogievskiy wrote:
> > > > Hi all!
> > > > 
> > > > Here is simple benchmarking utility, to generate performance
> > > > comparison tables, like the following:
> > > > 
> > > > --  -  -  -
> > > >  backup-1   backup-2   mirror
> > > > ssd -> ssd  0.43 +- 0.00   4.48 +- 0.06   4.38 +- 0.02
> > > > ssd -> hdd  10.60 +- 0.08  10.69 +- 0.18  10.57 +- 0.05
> > > > ssd -> nbd  33.81 +- 0.37  10.67 +- 0.17  10.07 +- 0.07
> > > > --  -  -  -
> > > > 
> > > > This is a v2, as v1 was inside
> > > >   "[RFC 00/24] backup performance: block_status + async"
> > > > 
> > > > I'll use this benchmark in other series, hope someone
> > > > will like it.
> > > > 
> > > > Vladimir Sementsov-Ogievskiy (3):
> > > >    python: add simplebench.py
> > > >    python: add qemu/bench_block_job.py
> > > >    python: add example usage of simplebench
> > > > 
> > > >   python/bench-example.py    |  80 +
> > > >   python/qemu/bench_block_job.py | 115 +
> > > >   python/simplebench.py  | 128 +
> > > >   3 files changed, 323 insertions(+)
> > > >   create mode 100644 python/bench-example.py
> > > >   create mode 100755 python/qemu/bench_block_job.py
> > > >   create mode 100644 python/simplebench.py
> > > > 
> > > 
> > > 
> > 
> > 
> 
> 
> -- 
> Best regards,
> Vladimir
> 

-- 
Eduardo

Re: [PULL v2 00/46] Python queue 2020-02-07

[Eduardo Habkost]

On Fri, Feb 07, 2020 at 04:11:12PM +0100, Philippe Mathieu-Daudé wrote:
> Hi Peter,
> 
> I prepared this series on behalf of Eduardo and
> Cleber.
> 
> Eduardo already ack'ed yesterday version (2020-02-06) cover:
> https://www.mail-archive.com/qemu-devel@nongnu.org/msg677636.html

Acked-by: Eduardo Habkost 


> 
> Since 2020-02-06 (v1):
> - rebased to cover new iotests #283 (merged yesterday).
> 
> Regards,
> 
> Phil.
> 
> The following changes since commit 863d2ed5823f90c42dcd481687cc99cbc9c4a17c:
> 
>   Merge remote-tracking branch 'remotes/maxreitz/tags/pull-block-2020-02-06' 
> into staging (2020-02-06 16:22:05 +)
> 
> are available in the Git repository at:
> 
>   https://gitlab.com/philmd/qemu.git tags/python-next-20200207
> 
> for you to fetch changes up to 66e7dde18cc4085ca47124be4ca08fa8e6bcdd3a:
> 
>   .readthedocs.yml: specify some minimum python requirements (2020-02-07 
> 15:15:16 +0100)
> 

-- 
Eduardo


signature.asc
Description: PGP signature

Re: Can we retire Python 2 now?

[Eduardo Habkost]

On Fri, Dec 20, 2019 at 05:29:30PM +0100, Markus Armbruster wrote:
> Python 2 EOL is only a few days away[*].  We made configure bitch about
> it in commit e5abf59eae "Deprecate Python 2 support", 2019-07-01.  Any
> objections to retiring it now, i.e. in 5.0?

Thanks for the reminder!

I'll be honest: even if somebody in this list objects to dropping
Python 2 support, I'm not willing to be maintainer of a Python 2
codebase in 2020.  The only reason for not doing it in 4.1 was
the tests/vm/netbsd breakage we took very long to debug and fix.

I have just submitted this pull request:

  Subject: [PULL 0/2] Require Python >= 3.5 to build QEMU
  
https://lore.kernel.org/qemu-devel/20191220165141.2207058-1-ehabk...@redhat.com/

> 
> Cc'ing everyone who appears to be maintaining something that looks like
> a Python script.
> 
> [*] https://pythonclock.org/

-- 
Eduardo

Re: [PATCH v0 1/2] qdev-properties-system: extend set_pionter for unrealized devices

[Eduardo Habkost]

On Fri, Nov 22, 2019 at 11:36:30AM +, Denis Plotnikov wrote:
> 
> 
> On 18.11.2019 21:54, Eduardo Habkost wrote:
> > On Sun, Nov 10, 2019 at 10:03:09PM +0300, Denis Plotnikov wrote:
> >> Some device's property can be changed if the device has been already
> >> realized. For example, it could be "drive" property of a scsi disk device.
> >>
> >> So far, set_pointer could operate only on a relized device. The patch
> >> extends its interface for operation on an unrealized device.
> >>
> >> Signed-off-by: Denis Plotnikov 
> >> ---
> >>   hw/core/qdev-properties-system.c | 32 +---
> >>   1 file changed, 21 insertions(+), 11 deletions(-)
> >>
> >> diff --git a/hw/core/qdev-properties-system.c 
> >> b/hw/core/qdev-properties-system.c
> >> index ba412dd2ca..c534590dcd 100644
> >> --- a/hw/core/qdev-properties-system.c
> >> +++ b/hw/core/qdev-properties-system.c
> >> @@ -38,9 +38,14 @@ static void get_pointer(Object *obj, Visitor *v, 
> >> Property *prop,
> >>   }
> >>   
> >>   static void set_pointer(Object *obj, Visitor *v, Property *prop,
> >> -void (*parse)(DeviceState *dev, const char *str,
> >> -  void **ptr, const char *propname,
> >> -  Error **errp),
> >> +void (*parse_realized)(DeviceState *dev,
> >> +   const char *str, void 
> >> **ptr,
> >> +   const char *propname,
> >> +   Error **errp),
> >> +void (*parse_unrealized)(DeviceState *dev,
> >> + const char *str, void 
> >> **ptr,
> >> + const char *propname,
> >> + Error **errp),
> >>   const char *name, Error **errp)
> > Wouldn't it be simpler to just add a PropertyInfo::allow_set_after_realize
> > bool field, and call the same setter function?  Then you can
> > simply change do_parse_drive() to check if realized is true.
> May be, but I thought It would be more clear to have a separate callback 
> for all the devices supporting the property setting when realized.
> Also the "drive" property setting on realized and non-realized device a 
> little bit different: in the realized case the setter function expects 
> to get
> BlockDriverState only, when in the unrealized case the setter can accept 
> both BlockBackend and BlockDriverState. Also, in the unrealized case the 
> setter function doesn't expect to have a device with an empty BlockBackend.
> I decided that extending do_parse_drive would make it more complex for 
> understanding. That's why I made two separate functions for both cases.

I understand you might want two separate functions in the
specific case of drive.  You can still call different
functions after checking dev->realized inside do_parse_drive().

My point was that you don't need to make set_pointer() require
two separate function pointers just to propagate 1 bit of
information that is already available in DeviceState.  In patch
2/2 you had to create 4 different copies of parse_drive*()
because of this.


> 
> I'd like to mention that I have a few concerns about 
> do_parse_drive_realized (please see the next patch from the series) and 
> I'd like them to be reviewed as well. After that, may be it would be 
> better to go the way you suggested.

In the case if your questions in patch 2/2, I'm afraid I don't
know the answers and we need help from the block maintainers.

-- 
Eduardo

Re: [PATCH v0 1/2] qdev-properties-system: extend set_pionter for unrealized devices

[Eduardo Habkost]

On Sun, Nov 10, 2019 at 10:03:09PM +0300, Denis Plotnikov wrote:
> Some device's property can be changed if the device has been already
> realized. For example, it could be "drive" property of a scsi disk device.
> 
> So far, set_pointer could operate only on a relized device. The patch
> extends its interface for operation on an unrealized device.
> 
> Signed-off-by: Denis Plotnikov 
> ---
>  hw/core/qdev-properties-system.c | 32 +---
>  1 file changed, 21 insertions(+), 11 deletions(-)
> 
> diff --git a/hw/core/qdev-properties-system.c 
> b/hw/core/qdev-properties-system.c
> index ba412dd2ca..c534590dcd 100644
> --- a/hw/core/qdev-properties-system.c
> +++ b/hw/core/qdev-properties-system.c
> @@ -38,9 +38,14 @@ static void get_pointer(Object *obj, Visitor *v, Property 
> *prop,
>  }
>  
>  static void set_pointer(Object *obj, Visitor *v, Property *prop,
> -void (*parse)(DeviceState *dev, const char *str,
> -  void **ptr, const char *propname,
> -  Error **errp),
> +void (*parse_realized)(DeviceState *dev,
> +   const char *str, void **ptr,
> +   const char *propname,
> +   Error **errp),
> +void (*parse_unrealized)(DeviceState *dev,
> + const char *str, void **ptr,
> + const char *propname,
> + Error **errp),
>  const char *name, Error **errp)

Wouldn't it be simpler to just add a PropertyInfo::allow_set_after_realize
bool field, and call the same setter function?  Then you can
simply change do_parse_drive() to check if realized is true.

>  {
>  DeviceState *dev = DEVICE(obj);
> @@ -48,11 +53,6 @@ static void set_pointer(Object *obj, Visitor *v, Property 
> *prop,
>  void **ptr = qdev_get_prop_ptr(dev, prop);
>  char *str;
>  
> -if (dev->realized) {
> -qdev_prop_set_after_realize(dev, name, errp);
> -return;
> -}
> -
>  visit_type_str(v, name, , _err);
>  if (local_err) {
>  error_propagate(errp, local_err);
> @@ -63,7 +63,17 @@ static void set_pointer(Object *obj, Visitor *v, Property 
> *prop,
>  *ptr = NULL;
>  return;
>  }
> -parse(dev, str, ptr, prop->name, errp);
> +
> +if (dev->realized) {
> +if (parse_realized) {
> +parse_realized(dev, str, ptr, prop->name, errp);
> +} else {
> +qdev_prop_set_after_realize(dev, name, errp);
> +}
> +} else {
> +parse_unrealized(dev, str, ptr, prop->name, errp);
> +}
> +
>  g_free(str);
>  }
>  
> @@ -178,13 +188,13 @@ static void get_drive(Object *obj, Visitor *v, const 
> char *name, void *opaque,
>  static void set_drive(Object *obj, Visitor *v, const char *name, void 
> *opaque,
>Error **errp)
>  {
> -set_pointer(obj, v, opaque, parse_drive, name, errp);
> +set_pointer(obj, v, opaque, NULL, parse_drive, name, errp);
>  }
>  
>  static void set_drive_iothread(Object *obj, Visitor *v, const char *name,
> void *opaque, Error **errp)
>  {
> -set_pointer(obj, v, opaque, parse_drive_iothread, name, errp);
> +set_pointer(obj, v, opaque, NULL, parse_drive_iothread, name, errp);
>  }
>  
>  const PropertyInfo qdev_prop_drive = {
> -- 
> 2.17.0
> 

-- 
Eduardo

Re: [PATCH v1 4/4] iotests: add test for virtio-scsi and virtio-blk machine type settings

[Eduardo Habkost]

On Wed, Nov 06, 2019 at 11:04:16AM +0100, Max Reitz wrote:
> On 06.11.19 10:24, Stefan Hajnoczi wrote:
> > On Tue, Nov 05, 2019 at 07:11:05PM +0300, Denis Plotnikov wrote:
> >> It tests proper queue size settings for all available machine types.
> >>
> >> Signed-off-by: Denis Plotnikov 
> >> ---
> >>  tests/qemu-iotests/267 | 154 +
> >>  tests/qemu-iotests/267.out |   1 +
> >>  tests/qemu-iotests/group   |   1 +
> >>  3 files changed, 156 insertions(+)
> >>  create mode 100755 tests/qemu-iotests/267
> >>  create mode 100644 tests/qemu-iotests/267.out
> > 
> > The qemu-iotests maintainers might prefer for this to be at the
> > top-level in tests/ since it's not really an iotest, but the code itself
> > looks fine to me:
> > 
> > Reviewed-by: Stefan Hajnoczi 
> 
> Good question.  I don’t really mind, but it would be weird if started
> adding all kinds of “external” qemu tests (i.e. that use QMP) in the
> iotests directory.
> 
> What is the alternative?  Just putting it in a different directory
> doesn’t sound that appealing to me either, because it would still depend
> on the iotests infrastructure, right?  (i.e., iotests.py and check)

We do have tests/acceptance for simple test cases written in
Python.  What's the reason for this test case to depend on the
iotests infrastructure?

-- 
Eduardo

Re: [PATCH 00/10] image-fuzzer: Port to Python 3

[Eduardo Habkost]

On Thu, Oct 17, 2019 at 06:29:27PM -0300, Eduardo Habkost wrote:
> On Thu, Oct 17, 2019 at 05:11:29PM -0400, John Snow wrote:
> > 
> > 
> > On 10/16/19 3:24 PM, Eduardo Habkost wrote:
> > > This series ports image-fuzzer to Python 3.
> > > 
> > > Eduardo Habkost (10):
> > >   image-fuzzer: Open image files in binary mode
> > >   image-fuzzer: Write bytes instead of string to image file
> > >   image-fuzzer: Explicitly use integer division operator
> > >   image-fuzzer: Use io.StringIO
> > >   image-fuzzer: Use %r for all fiels at Field.__repr__()
> > >   image-fuzzer: Return bytes objects on string fuzzing functions
> > >   image-fuzzer: Use bytes constant for field values
> > >   image-fuzzer: Encode file name and file format to bytes
> > >   image-fuzzer: Run using python3
> > >   image-fuzzer: Use errors parameter of subprocess.Popen()
> > > 
> > >  tests/image-fuzzer/qcow2/__init__.py |  1 -
> > >  tests/image-fuzzer/qcow2/fuzz.py | 54 +-
> > >  tests/image-fuzzer/qcow2/layout.py   | 57 ++--
> > >  tests/image-fuzzer/runner.py | 12 +++---
> > >  4 files changed, 61 insertions(+), 63 deletions(-)
> > > 
> > 
> > When I gave my try at converting this to python3 I noticed that the
> > "except OSError as e" segments used e[1] in a way that was not seemingly
> > supported.
> > 
> > Did you fix that in this series or did I miss it?
> 
> Good catch, I hadn't noticed that.  I didn't fix it.

Separate patch sent for that issue:
https://lore.kernel.org/qemu-devel/20191021214117.18091-1-ehabk...@redhat.com/

-- 
Eduardo

[PATCH] image-fuzzer: Use OSerror.strerror instead of tuple subscript

[Eduardo Habkost]

OSError can't be used like a tuple on Python 3, so change the
code to use `e.sterror` instead of `e[1]`.

Reported-by: John Snow 
Signed-off-by: Eduardo Habkost 
---
 tests/image-fuzzer/runner.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/image-fuzzer/runner.py b/tests/image-fuzzer/runner.py
index 95d84f38f3..1eb2138f94 100755
--- a/tests/image-fuzzer/runner.py
+++ b/tests/image-fuzzer/runner.py
@@ -159,7 +159,7 @@ class TestEnv(object):
 os.makedirs(self.current_dir)
 except OSError as e:
 print("Error: The working directory '%s' cannot be used. Reason: 
%s"\
-% (self.work_dir, e[1]), file=sys.stderr)
+% (self.work_dir, e.strerror), file=sys.stderr)
 raise TestException
 self.log = open(os.path.join(self.current_dir, "test.log"), "w")
 self.parent_log = open(run_log, "a")
@@ -246,7 +246,7 @@ class TestEnv(object):
 except OSError as e:
 multilog("%sError: Start of '%s' failed. Reason: %s\n\n"
  % (test_summary, os.path.basename(current_cmd[0]),
-e[1]),
+e.strerror),
  sys.stderr, self.log, self.parent_log)
 raise TestException
 
-- 
2.21.0

Re: [PATCH 00/10] image-fuzzer: Port to Python 3

[Eduardo Habkost]

On Thu, Oct 17, 2019 at 05:11:29PM -0400, John Snow wrote:
> 
> 
> On 10/16/19 3:24 PM, Eduardo Habkost wrote:
> > This series ports image-fuzzer to Python 3.
> > 
> > Eduardo Habkost (10):
> >   image-fuzzer: Open image files in binary mode
> >   image-fuzzer: Write bytes instead of string to image file
> >   image-fuzzer: Explicitly use integer division operator
> >   image-fuzzer: Use io.StringIO
> >   image-fuzzer: Use %r for all fiels at Field.__repr__()
> >   image-fuzzer: Return bytes objects on string fuzzing functions
> >   image-fuzzer: Use bytes constant for field values
> >   image-fuzzer: Encode file name and file format to bytes
> >   image-fuzzer: Run using python3
> >   image-fuzzer: Use errors parameter of subprocess.Popen()
> > 
> >  tests/image-fuzzer/qcow2/__init__.py |  1 -
> >  tests/image-fuzzer/qcow2/fuzz.py | 54 +-
> >  tests/image-fuzzer/qcow2/layout.py   | 57 ++--
> >  tests/image-fuzzer/runner.py | 12 +++---
> >  4 files changed, 61 insertions(+), 63 deletions(-)
> > 
> 
> When I gave my try at converting this to python3 I noticed that the
> "except OSError as e" segments used e[1] in a way that was not seemingly
> supported.
> 
> Did you fix that in this series or did I miss it?

Good catch, I hadn't noticed that.  I didn't fix it.

-- 
Eduardo

[PATCH 09/10] image-fuzzer: Run using python3

[Eduardo Habkost]

image-fuzzer is now supposed to be ready to run using Python 3.
Remove the __future__ imports and change the interpreter line to
"#!/usr/bin/env python3".

Signed-off-by: Eduardo Habkost 
---
 tests/image-fuzzer/qcow2/__init__.py | 1 -
 tests/image-fuzzer/qcow2/layout.py   | 1 -
 tests/image-fuzzer/runner.py | 3 +--
 3 files changed, 1 insertion(+), 4 deletions(-)

diff --git a/tests/image-fuzzer/qcow2/__init__.py 
b/tests/image-fuzzer/qcow2/__init__.py
index 09ef59821b..ed3af5da86 100644
--- a/tests/image-fuzzer/qcow2/__init__.py
+++ b/tests/image-fuzzer/qcow2/__init__.py
@@ -1,2 +1 @@
-from __future__ import absolute_import
 from .layout import create_image
diff --git a/tests/image-fuzzer/qcow2/layout.py 
b/tests/image-fuzzer/qcow2/layout.py
index 01bff4d05e..57ebe86e9a 100644
--- a/tests/image-fuzzer/qcow2/layout.py
+++ b/tests/image-fuzzer/qcow2/layout.py
@@ -16,7 +16,6 @@
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 #
 
-from __future__ import absolute_import
 import random
 import struct
 from . import fuzz
diff --git a/tests/image-fuzzer/runner.py b/tests/image-fuzzer/runner.py
index 94cab5bd93..0793234815 100755
--- a/tests/image-fuzzer/runner.py
+++ b/tests/image-fuzzer/runner.py
@@ -1,4 +1,4 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 
 # Tool for running fuzz tests
 #
@@ -18,7 +18,6 @@
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 #
 
-from __future__ import print_function
 import sys
 import os
 import signal
-- 
2.21.0

[PATCH 05/10] image-fuzzer: Use %r for all fiels at Field.__repr__()

[Eduardo Habkost]

This makes the formatting code simpler, and safer if we change
the type of self.value from str to bytes.

Signed-off-by: Eduardo Habkost 
---
 tests/image-fuzzer/qcow2/layout.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/image-fuzzer/qcow2/layout.py 
b/tests/image-fuzzer/qcow2/layout.py
index 6501c9fd4b..0adcbd448d 100644
--- a/tests/image-fuzzer/qcow2/layout.py
+++ b/tests/image-fuzzer/qcow2/layout.py
@@ -53,8 +53,8 @@ class Field(object):
 return iter([self.fmt, self.offset, self.value, self.name])
 
 def __repr__(self):
-return "Field(fmt='%s', offset=%d, value=%s, name=%s)" % \
-(self.fmt, self.offset, str(self.value), self.name)
+return "Field(fmt=%r, offset=%r, value=%r, name=%r)" % \
+(self.fmt, self.offset, self.value, self.name)
 
 
 class FieldsList(object):
-- 
2.21.0

[PATCH 04/10] image-fuzzer: Use io.StringIO

[Eduardo Habkost]

StringIO.StringIO is not available on Python 3, but io.StringIO
is available on both Python 2 and 3.  io.StringIO is slightly
different from the Python 2 StringIO module, though, so we need
bytes coming from subprocess.Popen() to be explicitly decoded.

Signed-off-by: Eduardo Habkost 
---
 tests/image-fuzzer/runner.py | 14 +-
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/tests/image-fuzzer/runner.py b/tests/image-fuzzer/runner.py
index 95d84f38f3..94cab5bd93 100755
--- a/tests/image-fuzzer/runner.py
+++ b/tests/image-fuzzer/runner.py
@@ -28,7 +28,7 @@ import shutil
 from itertools import count
 import time
 import getopt
-import StringIO
+import io
 import resource
 
 try:
@@ -84,8 +84,12 @@ def run_app(fd, q_args):
 try:
 out, err = process.communicate()
 signal.alarm(0)
-fd.write(out)
-fd.write(err)
+# fd is a text file, so we need to decode the process output before
+# writing to it.
+# We could be simply using the `errors` parameter of 
subprocess.Popen(),
+# but this will be possible only after migrating to Python 3
+fd.write(out.decode(errors='replace'))
+fd.write(err.decode(errors='replace'))
 fd.flush()
 return process.returncode
 
@@ -183,7 +187,7 @@ class TestEnv(object):
MAX_BACKING_FILE_SIZE) * (1 << 20)
 cmd = self.qemu_img + ['create', '-f', backing_file_fmt,
backing_file_name, str(backing_file_size)]
-temp_log = StringIO.StringIO()
+temp_log = io.StringIO()
 retcode = run_app(temp_log, cmd)
 if retcode == 0:
 temp_log.close()
@@ -240,7 +244,7 @@ class TestEnv(object):
"Backing file: %s\n" \
% (self.seed, " ".join(current_cmd),
   self.current_dir, backing_file_name)
-temp_log = StringIO.StringIO()
+temp_log = io.StringIO()
 try:
 retcode = run_app(temp_log, current_cmd)
 except OSError as e:
-- 
2.21.0

[PATCH 03/10] image-fuzzer: Explicitly use integer division operator

[Eduardo Habkost]

Most of the division expressions in image-fuzzer assume integer
division.  Use the // operator to keep the same behavior when we
move to Python 3.

Signed-off-by: Eduardo Habkost 
---
 tests/image-fuzzer/qcow2/fuzz.py   | 12 -
 tests/image-fuzzer/qcow2/layout.py | 40 +++---
 2 files changed, 26 insertions(+), 26 deletions(-)

diff --git a/tests/image-fuzzer/qcow2/fuzz.py b/tests/image-fuzzer/qcow2/fuzz.py
index abc4f0635d..154dc06cc0 100644
--- a/tests/image-fuzzer/qcow2/fuzz.py
+++ b/tests/image-fuzzer/qcow2/fuzz.py
@@ -27,14 +27,14 @@ UINT64 = 0x
 UINT32_M = 31
 UINT64_M = 63
 # Fuzz vectors
-UINT8_V = [0, 0x10, UINT8/4, UINT8/2 - 1, UINT8/2, UINT8/2 + 1, UINT8 - 1,
+UINT8_V = [0, 0x10, UINT8//4, UINT8//2 - 1, UINT8//2, UINT8//2 + 1, UINT8 - 1,
UINT8]
-UINT16_V = [0, 0x100, 0x1000, UINT16/4, UINT16/2 - 1, UINT16/2, UINT16/2 + 1,
+UINT16_V = [0, 0x100, 0x1000, UINT16//4, UINT16//2 - 1, UINT16//2, UINT16//2 + 
1,
 UINT16 - 1, UINT16]
-UINT32_V = [0, 0x100, 0x1000, 0x1, 0x10, UINT32/4, UINT32/2 - 1,
-UINT32/2, UINT32/2 + 1, UINT32 - 1, UINT32]
-UINT64_V = UINT32_V + [0x100, 0x1000, 0x1, UINT64/4,
-   UINT64/2 - 1, UINT64/2, UINT64/2 + 1, UINT64 - 1,
+UINT32_V = [0, 0x100, 0x1000, 0x1, 0x10, UINT32//4, UINT32//2 - 1,
+UINT32//2, UINT32//2 + 1, UINT32 - 1, UINT32]
+UINT64_V = UINT32_V + [0x100, 0x1000, 0x1, UINT64//4,
+   UINT64//2 - 1, UINT64//2, UINT64//2 + 1, UINT64 - 1,
UINT64]
 STRING_V = ['%s%p%x%d', '.1024d', '%.2049d', '%p%p%p%p', '%x%x%x%x',
 '%d%d%d%d', '%s%s%s%s', '%999s', '%08x', '%%20d', '%%20n',
diff --git a/tests/image-fuzzer/qcow2/layout.py 
b/tests/image-fuzzer/qcow2/layout.py
index fe273d4143..6501c9fd4b 100644
--- a/tests/image-fuzzer/qcow2/layout.py
+++ b/tests/image-fuzzer/qcow2/layout.py
@@ -253,7 +253,7 @@ class Image(object):
 ['>I', self.ext_offset, 0x6803f857, 'ext_magic'],
 # One feature table contains 3 fields and takes 48 bytes
 ['>I', self.ext_offset + UINT32_S,
- len(feature_tables) / 3 * 48, 'ext_length']
+ len(feature_tables) // 3 * 48, 'ext_length']
 ] + feature_tables)
 self.ext_offset = inner_offset
 
@@ -271,7 +271,7 @@ class Image(object):
 def create_l2_entry(host, guest, l2_cluster):
 """Generate one L2 entry."""
 offset = l2_cluster * self.cluster_size
-l2_size = self.cluster_size / UINT64_S
+l2_size = self.cluster_size // UINT64_S
 entry_offset = offset + UINT64_S * (guest % l2_size)
 cluster_descriptor = host * self.cluster_size
 if not self.header['version'][0].value == 2:
@@ -283,8 +283,8 @@ class Image(object):
 
 def create_l1_entry(l2_cluster, l1_offset, guest):
 """Generate one L1 entry."""
-l2_size = self.cluster_size / UINT64_S
-entry_offset = l1_offset + UINT64_S * (guest / l2_size)
+l2_size = self.cluster_size // UINT64_S
+entry_offset = l1_offset + UINT64_S * (guest // l2_size)
 # While snapshots are not supported bit #63 = 1
 entry_val = (1 << 63) + l2_cluster * self.cluster_size
 return ['>Q', entry_offset, entry_val, 'l1_entry']
@@ -298,11 +298,11 @@ class Image(object):
 l2 = []
 else:
 meta_data = self._get_metadata()
-guest_clusters = random.sample(range(self.image_size /
+guest_clusters = random.sample(range(self.image_size //
  self.cluster_size),
len(self.data_clusters))
 # Number of entries in a L1/L2 table
-l_size = self.cluster_size / UINT64_S
+l_size = self.cluster_size // UINT64_S
 # Number of clusters necessary for L1 table
 l1_size = int(ceil((max(guest_clusters) + 1) / float(l_size**2)))
 l1_start = self._get_adjacent_clusters(self.data_clusters |
@@ -318,7 +318,7 @@ class Image(object):
 # L2 entries
 l2 = []
 for host, guest in zip(self.data_clusters, guest_clusters):
-l2_id = guest / l_size
+l2_id = guest // l_size
 if l2_id not in l2_ids:
 l2_ids.append(l2_id)
 l2_clusters.append(self._get_adjacent_clusters(
@@ -339,14 +339,14 @@ class Image(object):
 def allocate_rfc_blocks(data, size):
 """Return indices of clusters allocated for refcount blocks."""
 cluster_ids = set()
-diff = block_ids = set([x / size for x i

[PATCH 10/10] image-fuzzer: Use errors parameter of subprocess.Popen()

[Eduardo Habkost]

Instead of manually encoding stderr and stdout output, use
`errors` parameter of subprocess.Popen().  This will make
process.communicate() return unicode strings instead of bytes
objects.

Signed-off-by: Eduardo Habkost 
---
 tests/image-fuzzer/runner.py | 11 ---
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/tests/image-fuzzer/runner.py b/tests/image-fuzzer/runner.py
index 0793234815..4ba5c79e13 100755
--- a/tests/image-fuzzer/runner.py
+++ b/tests/image-fuzzer/runner.py
@@ -79,16 +79,13 @@ def run_app(fd, q_args):
 devnull = open('/dev/null', 'r+')
 process = subprocess.Popen(q_args, stdin=devnull,
stdout=subprocess.PIPE,
-   stderr=subprocess.PIPE)
+   stderr=subprocess.PIPE,
+   errors='replace')
 try:
 out, err = process.communicate()
 signal.alarm(0)
-# fd is a text file, so we need to decode the process output before
-# writing to it.
-# We could be simply using the `errors` parameter of 
subprocess.Popen(),
-# but this will be possible only after migrating to Python 3
-fd.write(out.decode(errors='replace'))
-fd.write(err.decode(errors='replace'))
+fd.write(out)
+fd.write(err)
 fd.flush()
 return process.returncode
 
-- 
2.21.0

[PATCH 08/10] image-fuzzer: Encode file name and file format to bytes

[Eduardo Habkost]

Callers of create_image() will pass strings as arguments, but the
Image class will expect bytes objects to be provided.  Encode
them inside create_image().

Signed-off-by: Eduardo Habkost 
---
 tests/image-fuzzer/qcow2/layout.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/image-fuzzer/qcow2/layout.py 
b/tests/image-fuzzer/qcow2/layout.py
index a0fd53c7ad..01bff4d05e 100644
--- a/tests/image-fuzzer/qcow2/layout.py
+++ b/tests/image-fuzzer/qcow2/layout.py
@@ -602,8 +602,8 @@ class Image(object):
 def create_image(test_img_path, backing_file_name=None, backing_file_fmt=None,
  fields_to_fuzz=None):
 """Create a fuzzed image and write it to the specified file."""
-image = Image(backing_file_name)
-image.set_backing_file_format(backing_file_fmt)
+image = Image(backing_file_name.encode())
+image.set_backing_file_format(backing_file_fmt.encode())
 image.create_feature_name_table()
 image.set_end_of_extension_area()
 image.create_l_structures()
-- 
2.21.0

[PATCH 07/10] image-fuzzer: Use bytes constant for field values

[Eduardo Habkost]

Field values are supposed to be bytes objects, not unicode
strings.  Change two constants that were declared as strings.

Signed-off-by: Eduardo Habkost 
---
 tests/image-fuzzer/qcow2/layout.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/image-fuzzer/qcow2/layout.py 
b/tests/image-fuzzer/qcow2/layout.py
index 0adcbd448d..a0fd53c7ad 100644
--- a/tests/image-fuzzer/qcow2/layout.py
+++ b/tests/image-fuzzer/qcow2/layout.py
@@ -122,7 +122,7 @@ class Image(object):
 def create_header(self, cluster_bits, backing_file_name=None):
 """Generate a random valid header."""
 meta_header = [
-['>4s', 0, "QFI\xfb", 'magic'],
+['>4s', 0, b"QFI\xfb", 'magic'],
 ['>I', 4, random.randint(2, 3), 'version'],
 ['>Q', 8, 0, 'backing_file_offset'],
 ['>I', 16, 0, 'backing_file_size'],
@@ -231,7 +231,7 @@ class Image(object):
 feature_tables = []
 feature_ids = []
 inner_offset = self.ext_offset + ext_header_len
-feat_name = 'some cool feature'
+feat_name = b'some cool feature'
 while len(feature_tables) < num_fnt_entries * 3:
 feat_type, feat_bit = gen_feat_ids()
 # Remove duplicates
-- 
2.21.0

[PATCH 02/10] image-fuzzer: Write bytes instead of string to image file

[Eduardo Habkost]

This is necessary for Python 3 compatibility.

Signed-off-by: Eduardo Habkost 
---
 tests/image-fuzzer/qcow2/layout.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/image-fuzzer/qcow2/layout.py 
b/tests/image-fuzzer/qcow2/layout.py
index c57418fa15..fe273d4143 100644
--- a/tests/image-fuzzer/qcow2/layout.py
+++ b/tests/image-fuzzer/qcow2/layout.py
@@ -518,7 +518,7 @@ class Image(object):
 rounded = (size + self.cluster_size - 1) & ~(self.cluster_size - 1)
 if rounded > size:
 image_file.seek(rounded - 1)
-image_file.write("\0")
+image_file.write(b'\x00')
 image_file.close()
 
 @staticmethod
-- 
2.21.0

[PATCH 06/10] image-fuzzer: Return bytes objects on string fuzzing functions

[Eduardo Habkost]

No caller of fuzzer functions is interested in unicode string values,
so replace them with bytes sequences.

Signed-off-by: Eduardo Habkost 
---
 tests/image-fuzzer/qcow2/fuzz.py | 42 
 1 file changed, 21 insertions(+), 21 deletions(-)

diff --git a/tests/image-fuzzer/qcow2/fuzz.py b/tests/image-fuzzer/qcow2/fuzz.py
index 154dc06cc0..c58bf11005 100644
--- a/tests/image-fuzzer/qcow2/fuzz.py
+++ b/tests/image-fuzzer/qcow2/fuzz.py
@@ -36,11 +36,11 @@ UINT32_V = [0, 0x100, 0x1000, 0x1, 0x10, UINT32//4, 
UINT32//2 - 1,
 UINT64_V = UINT32_V + [0x100, 0x1000, 0x1, UINT64//4,
UINT64//2 - 1, UINT64//2, UINT64//2 + 1, UINT64 - 1,
UINT64]
-STRING_V = ['%s%p%x%d', '.1024d', '%.2049d', '%p%p%p%p', '%x%x%x%x',
-'%d%d%d%d', '%s%s%s%s', '%999s', '%08x', '%%20d', '%%20n',
-'%%20x', '%%20s', '%s%s%s%s%s%s%s%s%s%s', '%p%p%p%p%p%p%p%p%p%p',
-'%#0123456x%08x%x%s%p%d%n%o%u%c%h%l%q%j%z%Z%t%i%e%g%f%a%C%S%08x%%',
-'%s x 129', '%x x 257']
+BYTES_V = [b'%s%p%x%d', b'.1024d', b'%.2049d', b'%p%p%p%p', b'%x%x%x%x',
+   b'%d%d%d%d', b'%s%s%s%s', b'%999s', b'%08x', b'%%20d', 
b'%%20n',
+   b'%%20x', b'%%20s', b'%s%s%s%s%s%s%s%s%s%s', 
b'%p%p%p%p%p%p%p%p%p%p',
+   b'%#0123456x%08x%x%s%p%d%n%o%u%c%h%l%q%j%z%Z%t%i%e%g%f%a%C%S%08x%%',
+   b'%s x 129', b'%x x 257']
 
 
 def random_from_intervals(intervals):
@@ -76,12 +76,12 @@ def random_bits(bit_ranges):
 return val
 
 
-def truncate_string(strings, length):
-"""Return strings truncated to specified length."""
-if type(strings) == list:
-return [s[:length] for s in strings]
+def truncate_bytes(sequences, length):
+"""Return sequences truncated to specified length."""
+if type(sequences) == list:
+return [s[:length] for s in sequences]
 else:
-return strings[:length]
+return sequences[:length]
 
 
 def validator(current, pick, choices):
@@ -110,12 +110,12 @@ def bit_validator(current, bit_ranges):
 return validator(current, random_bits, bit_ranges)
 
 
-def string_validator(current, strings):
-"""Return a random string value from the list not equal to the current.
+def bytes_validator(current, sequences):
+"""Return a random bytes value from the list not equal to the current.
 
 This function is useful for selection from valid values except current one.
 """
-return validator(current, random.choice, strings)
+return validator(current, random.choice, sequences)
 
 
 def selector(current, constraints, validate=int_validator):
@@ -283,9 +283,9 @@ def header_length(current):
 def bf_name(current):
 """Fuzz the backing file name."""
 constraints = [
-truncate_string(STRING_V, len(current))
+truncate_bytes(BYTES_V, len(current))
 ]
-return selector(current, constraints, string_validator)
+return selector(current, constraints, bytes_validator)
 
 
 def ext_magic(current):
@@ -303,10 +303,10 @@ def ext_length(current):
 def bf_format(current):
 """Fuzz backing file format in the corresponding header extension."""
 constraints = [
-truncate_string(STRING_V, len(current)),
-truncate_string(STRING_V, (len(current) + 7) & ~7)  # Fuzz padding
+truncate_bytes(BYTES_V, len(current)),
+truncate_bytes(BYTES_V, (len(current) + 7) & ~7)  # Fuzz padding
 ]
-return selector(current, constraints, string_validator)
+return selector(current, constraints, bytes_validator)
 
 
 def feature_type(current):
@@ -324,10 +324,10 @@ def feature_bit_number(current):
 def feature_name(current):
 """Fuzz feature name field of a feature name table header extension."""
 constraints = [
-truncate_string(STRING_V, len(current)),
-truncate_string(STRING_V, 46)  # Fuzz padding (field length = 46)
+truncate_bytes(BYTES_V, len(current)),
+truncate_bytes(BYTES_V, 46)  # Fuzz padding (field length = 46)
 ]
-return selector(current, constraints, string_validator)
+return selector(current, constraints, bytes_validator)
 
 
 def l1_entry(current):
-- 
2.21.0

[PATCH 01/10] image-fuzzer: Open image files in binary mode

[Eduardo Habkost]

This probably never caused problems because on Linux there's no
actual newline conversion happening, but on Python 3 the
binary/text distinction is stronger and we must explicitly open
the image file in binary mode.

Signed-off-by: Eduardo Habkost 
---
 tests/image-fuzzer/qcow2/layout.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/image-fuzzer/qcow2/layout.py 
b/tests/image-fuzzer/qcow2/layout.py
index 675877da96..c57418fa15 100644
--- a/tests/image-fuzzer/qcow2/layout.py
+++ b/tests/image-fuzzer/qcow2/layout.py
@@ -503,7 +503,7 @@ class Image(object):
 
 def write(self, filename):
 """Write an entire image to the file."""
-image_file = open(filename, 'w')
+image_file = open(filename, 'wb')
 for field in self:
 image_file.seek(field.offset)
 image_file.write(struct.pack(field.fmt, field.value))
-- 
2.21.0

[PATCH 00/10] image-fuzzer: Port to Python 3

[Eduardo Habkost]

This series ports image-fuzzer to Python 3.

Eduardo Habkost (10):
  image-fuzzer: Open image files in binary mode
  image-fuzzer: Write bytes instead of string to image file
  image-fuzzer: Explicitly use integer division operator
  image-fuzzer: Use io.StringIO
  image-fuzzer: Use %r for all fiels at Field.__repr__()
  image-fuzzer: Return bytes objects on string fuzzing functions
  image-fuzzer: Use bytes constant for field values
  image-fuzzer: Encode file name and file format to bytes
  image-fuzzer: Run using python3
  image-fuzzer: Use errors parameter of subprocess.Popen()

 tests/image-fuzzer/qcow2/__init__.py |  1 -
 tests/image-fuzzer/qcow2/fuzz.py | 54 +-
 tests/image-fuzzer/qcow2/layout.py   | 57 ++--
 tests/image-fuzzer/runner.py | 12 +++---
 4 files changed, 61 insertions(+), 63 deletions(-)

-- 
2.21.0

Re: [PATCH v3 0/8] hw: Convert various reset() handler to DeviceReset

[Eduardo Habkost]

Queueing on machine-next.  Thanks!

On Thu, Oct 10, 2019 at 03:15:19PM +0200, Philippe Mathieu-Daudé wrote:
> Only patch 3/8 is missing review:
> - hw/ide/piix: Convert reset handler to DeviceReset
> 
> Since v2:
> - Fixed PIIX_IDE conversion (Li)
> - Added more R-b tag.
> 
> Since v1:
> - Removed the pci-host devices
> - Removed the vmcoreinfo conversion (elmarco) but add a comment.
> - Added Igor's R-b tag.
> 
> Following the thread discussion between Peter/Markus/Damien about
> reset handlers:
> https://www.mail-archive.com/qemu-devel@nongnu.org/msg617103.html
> I started to remove qemu_register_reset() calls from few qdevified
> devices (the trivial ones).
> 
> Regards,
> 
> Phil.
> 
> v2: https://lists.gnu.org/archive/html/qemu-devel/2019-10/msg01677.html
> v1: https://lists.gnu.org/archive/html/qemu-devel/2019-09/msg06367.html
> 
> Philippe Mathieu-Daudé (8):
>   hw/acpi/piix4: Convert reset handler to DeviceReset
>   hw/isa/piix4: Convert reset handler to DeviceReset
>   hw/ide/piix: Convert reset handler to DeviceReset
>   hw/ide/sii3112: Convert reset handler to DeviceReset
>   hw/ide/via82c: Convert reset handler to DeviceReset
>   hw/isa/vt82c686: Convert reset handler to DeviceReset
>   hw/input/lm832x: Convert reset handler to DeviceReset
>   hw/misc/vmcoreinfo: Add comment about reset handler
> 
>  hw/acpi/piix4.c  |  7 +++
>  hw/ide/piix.c|  9 -
>  hw/ide/sii3112.c |  7 +++
>  hw/ide/via.c | 10 --
>  hw/input/lm832x.c| 12 +---
>  hw/isa/piix4.c   |  7 +++
>  hw/isa/vt82c686.c| 11 ---
>  hw/misc/vmcoreinfo.c |  4 
>  8 files changed, 30 insertions(+), 37 deletions(-)
> 
> -- 
> 2.21.0
> 

-- 
Eduardo