John Snow [Fri, 15 Mar 2024 15:22:56 +0000 (16:22 +0100)]
qapi/parser.py: assert member.info is present in connect_member
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-21-armbru@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:55 +0000 (16:22 +0100)]
qapi/parser: demote QAPIExpression to Dict[str, Any]
Dict[str, object] is a stricter type, but with the way that code is
currently arranged, it is infeasible to enforce this strictness.
In particular, although expr.py's entire raison d'être is normalization
and type-checking of QAPI Expressions, that type information is not
"remembered" in any meaningful way by mypy because each individual
expression is not downcast to a specific expression type that holds all
the details of each expression's unique form.
As a result, all of the code in schema.py that deals with actually
creating type-safe specialized structures has no guarantee (myopically)
that the data it is being passed is correct.
There are two ways to solve this:
(1) Re-assert that the incoming data is in the shape we expect it to be, or
(2) Disable type checking for this data.
(1) is appealing to my sense of strictness, but I gotta concede that it
is asinine to re-check the shape of a QAPIExpression in schema.py when
expr.py has just completed that work at length. The duplication of code
and the nightmare thought of needing to update both locations if and
when we change the shape of these structures makes me extremely
reluctant to go down this route.
(2) allows us the chance to miss updating types in the case that types
are updated in expr.py, but it *is* an awful lot simpler and,
importantly, gets us closer to type checking schema.py *at
all*. Something is better than nothing, I'd argue.
So, do the simpler dumber thing and worry about future strictness
improvements later.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-20-armbru@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:54 +0000 (16:22 +0100)]
qapi/schema: assert inner type of QAPISchemaVariants in check_clash()
QAPISchemaVariant's "variants" field is typed as
List[QAPISchemaVariant], where the typing for QAPISchemaVariant allows
its type field to be any QAPISchemaType.
However, QAPISchemaVariant expects that all of its variants contain the
narrower QAPISchemaObjectType. This relationship is enforced at runtime
in QAPISchemaVariants.check(). This relationship is not embedded in the
type system though, so QAPISchemaVariants.check_clash() needs to
re-assert this property in order to call
QAPISchemaVariant.type.check_clash().
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-19-armbru@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:53 +0000 (16:22 +0100)]
qapi/schema: fix typing for QAPISchemaVariants.tag_member
There are two related changes here:
(1) We need to perform type narrowing for resolving the type of
tag_member during check(), and
(2) tag_member is a delayed initialization field, but we can hide it
behind a property that raises an Exception if it's called too
early. This simplifies the typing in quite a few places and avoids
needing to assert that the "tag_member is not None" at a dozen
callsites, which can be confusing and suggest the wrong thing to a
drive-by contributor.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-18-armbru@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:52 +0000 (16:22 +0100)]
qapi/schema: Don't initialize "members" with `None`
Declare, but don't initialize the "members" field with type
List[QAPISchemaObjectTypeMember].
This simplifies the typing from what would otherwise be
Optional[List[T]] to merely List[T]. This removes the need to add
assertions to several callsites that this value is not None - which it
never will be after the delayed initialization in check() anyway.
The type declaration without initialization trick will cause accidental
uses of this field prior to full initialization to raise an
AttributeError.
(Note that it is valid to have an empty members list, see the internal
q_empty object as an example. For this reason, we cannot use the empty
list as a replacement test for full initialization and instead rely on
the _checked/_check_complete fields.)
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-17-armbru@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:51 +0000 (16:22 +0100)]
qapi/schema: add _check_complete flag
Instead of using the None value for the members field, use a dedicated
flag to detect recursive misconfigurations.
This is intended to assist with subsequent patches that seek to remove
the "None" value from the members field (which can never hold that value
after the final call to check()) in order to simplify the static typing
of that field; avoiding the need of assertions littered at many
callsites to eliminate the possibility of the None value.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-16-armbru@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:50 +0000 (16:22 +0100)]
qapi/schema: assert info is present when necessary
QAPISchemaInfo arguments can often be None because built-in definitions
don't have such information. The type hint can only be
Optional[QAPISchemaInfo] then. But, mypy gets upset about all the
places where we exploit that it can't actually be None there. Add
assertions that will help mypy over the hump, to enable adding type
hints in a forthcoming commit.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-15-armbru@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:49 +0000 (16:22 +0100)]
qapi/schema: fix QAPISchemaArrayType.check's call to resolve_type
Adjust the expression at the callsite to work around mypy's weak type
introspection that believes this expression can resolve to
QAPISourceInfo; it cannot.
(Fundamentally: self.info only resolves to false in a boolean expression
when it is None; therefore this expression may only ever produce
Optional[str]. mypy does not know that 'info', when it is a
QAPISourceInfo object, cannot ever be false.)
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-14-armbru@redhat.com>
QAPISchema.lookup_type('FOO') returns a QAPISchemaType when type 'FOO'
exists, else None. It won't return None for built-in types like
'int'.
Since mypy can't see that, it'll complain that we assign the
Optional[QAPISchemaType] returned by .lookup_type() to QAPISchemaType
variables.
Add assertions to help it over the hump.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-13-armbru@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:47 +0000 (16:22 +0100)]
qapi/schema: assert resolve_type has 'info' and 'what' args on error
resolve_type() is generally used to resolve configuration-provided type
names into type objects, and generally requires valid 'info' and 'what'
parameters.
In some cases, such as with QAPISchemaArrayType.check(), resolve_type
may be used to resolve built-in types and as such will not have an
'info' argument, but also must not fail in this scenario.
Use an assertion to sate mypy that we will indeed have 'info' and 'what'
parameters for the error pathway in resolve_type.
Note: there are only three callsites to resolve_type at present where
"info" is perceived by mypy to be possibly None:
Of those three, only the first actually ever passes None; the other two
are limited by their base class initializers which accept info=None, but
neither subclass actually use a None value in practice, currently.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-12-armbru@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:46 +0000 (16:22 +0100)]
qapi/schema: add type narrowing to lookup_type()
This function is a bit hard to type as-is; mypy needs some assertions to
assist with the type narrowing.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-11-armbru@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:45 +0000 (16:22 +0100)]
qapi/schema: adjust type narrowing for mypy's benefit
We already take care to perform some type narrowing for arg_type and
ret_type, but not in a way where mypy can utilize the result once we add
type hints, e.g.:
qapi/schema.py:833: error: Incompatible types in assignment (expression
has type "QAPISchemaType", variable has type
"Optional[QAPISchemaObjectType]") [assignment]
qapi/schema.py:893: error: Incompatible types in assignment (expression
has type "QAPISchemaType", variable has type
"Optional[QAPISchemaObjectType]") [assignment]
A simple change to use a temporary variable helps the medicine go down.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-10-armbru@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:44 +0000 (16:22 +0100)]
qapi/schema: make c_type() and json_type() abstract methods
These methods should always return a str, it's only the default abstract
implementation that doesn't. They can be marked "abstract", which
requires subclasses to override the method with the proper return type.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-9-armbru@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:43 +0000 (16:22 +0100)]
qapi/schema: declare type for QAPISchemaArrayType.element_type
A QAPISchemaArrayType's element type gets resolved only during .check().
We have QAPISchemaArrayType.__init__() initialize self.element_type =
None, and .check() assign the actual type. Using .element_type before
.check() is wrong, and hopefully crashes due to the value being None.
Works.
However, it makes for awkward typing. With .element_type:
Optional[QAPISchemaType], mypy is of course unable to see that it's None
before .check(), and a QAPISchemaType after. To help it over the hump,
we'd have to assert self.element_type is not None before all the (valid)
uses. The assertion catches invalid uses, but only at run time; mypy
can't flag them.
Instead, declare .element_type in .__init__() as QAPISchemaType
*without* initializing it. Using .element_type before .check() now
certainly crashes, which is an improvement. Mypy still can't flag
invalid uses, but that's okay.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-8-armbru@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:42 +0000 (16:22 +0100)]
qapi/schema: declare type for QAPISchemaObjectTypeMember.type
A QAPISchemaObjectTypeMember's type gets resolved only during .check().
We have QAPISchemaObjectTypeMember.__init__() initialize self.type =
None, and .check() assign the actual type. Using .type before .check()
is wrong, and hopefully crashes due to the value being None. Works.
However, it makes for awkward typing. With .type:
Optional[QAPISchemaType], mypy is of course unable to see that it's None
before .check(), and a QAPISchemaType after. To help it over the hump,
we'd have to assert self.type is not None before all the (valid) uses.
The assertion catches invalid uses, but only at run time; mypy can't
flag them.
Instead, declare .type in .__init__() as QAPISchemaType *without*
initializing it. Using .type before .check() now certainly crashes,
which is an improvement. Mypy still can't flag invalid uses, but that's
okay.
Addresses typing errors such as these:
qapi/schema.py:657: error: "None" has no attribute "alternate_qtype" [attr-defined]
qapi/schema.py:662: error: "None" has no attribute "describe" [attr-defined]
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-7-armbru@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:41 +0000 (16:22 +0100)]
qapi: create QAPISchemaDefinition
Include entities don't have names, but we generally expect "entities" to
have names. Reclassify all entities with names as *definitions*, leaving
the nameless include entities as QAPISchemaEntity instances.
This is primarily to help simplify typing around expectations of what
callers expect for properties of an "entity".
Suggested-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-6-armbru@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:40 +0000 (16:22 +0100)]
qapi/schema: add pylint suppressions
With this patch, pylint is happy with the file, so enable it in the
configuration.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-5-armbru@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:39 +0000 (16:22 +0100)]
qapi: sort pylint suppressions
Suggested-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-4-armbru@redhat.com>
John Snow [Fri, 15 Mar 2024 15:22:38 +0000 (16:22 +0100)]
qapi/parser: shush up pylint
Shhh!
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240315152301.3621858-3-armbru@redhat.com>
Kevin Wolf [Fri, 12 Apr 2024 14:42:02 +0000 (16:42 +0200)]
usb-storage: Fix BlockConf defaults
Commit 30896374 started to pass the full BlockConf from usb-storage to
scsi-disk, while previously only a few select properties would be
forwarded. This enables the user to set more properties, e.g. the block
size, that are actually taking effect.
However, now the calls to blkconf_apply_backend_options() and
blkconf_blocksizes() in usb_msd_storage_realize() that modify some of
these properties take effect, too, instead of being silently ignored.
This means at least that the block sizes get an unconditional default of
512 bytes before the configuration is passed to scsi-disk.
Before commit 30896374, the property wouldn't be set for scsi-disk and
therefore the device dependent defaults would apply - 512 for scsi-hd,
but 2048 for scsi-cd. The latter default has now become 512, too, which
makes at least Windows 11 installation fail when installing from
usb-storage.
Fix this by simply not calling these functions any more in usb-storage
and passing BlockConf on unmodified (except for the BlockBackend). The
same functions are called by the SCSI code anyway and it sets the right
defaults for the actual media type.
Fixes: 308963746169 ('scsi: Don't ignore most usb-storage properties')
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2260 Reported-by: Jonas Svensson Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Hanna Czenczek <hreitz@redhat.com>
Message-id: 20240412144202.13786-1-kwolf@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
* tag 'hw-misc-20240415' of https://github.com/philmd/qemu:
hw/pci-host/ppc440_pcix: Do not expose a bridge device on PCI bus
hw/isa/vt82c686: Keep track of PIRQ/PINT pins separately
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
hw/pci-host/ppc440_pcix: Do not expose a bridge device on PCI bus
Real 460EX SoC apparently does not expose a bridge device and having
it appear on PCI bus confuses an AmigaOS file system driver that uses
this to detect which machine it is running on.
Cc: qemu-stable@nongnu.org Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240411192443.B4D644E6026@zero.eik.bme.hu> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
hw/isa/vt82c686: Keep track of PIRQ/PINT pins separately
Move calculation of mask after the switch which sets the function
number for PIRQ/PINT pins to make sure the state of these pins are
kept track of separately and IRQ is raised if any of them is active.
Cc: qemu-stable@nongnu.org Fixes: 7e01bd80c1 hw/isa/vt82c686: Bring back via_isa_set_irq() Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240410222543.0EA534E6005@zero.eik.bme.hu> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Cindy Lu [Fri, 12 Apr 2024 06:26:55 +0000 (14:26 +0800)]
virtio-pci: fix use of a released vector
During the booting process of the non-standard image, the behavior of the
called function in qemu is as follows:
1. vhost_net_stop() was triggered by guest image. This will call the function
virtio_pci_set_guest_notifiers() with assgin= false,
virtio_pci_set_guest_notifiers(） will release the irqfd for vector 0
2. virtio_reset() was triggered, this will set configure vector to VIRTIO_NO_VECTOR
3.vhost_net_start() was called (at this time, the configure vector is
still VIRTIO_NO_VECTOR) and then call virtio_pci_set_guest_notifiers() with
assgin=true, so the irqfd for vector 0 is still not "init" during this process
4. The system continues to boot and sets the vector back to 0. After that
msix_fire_vector_notifier() was triggered to unmask the vector 0 and meet the crash
To fix the issue, we need to support changing the vector after VIRTIO_CONFIG_S_DRIVER_OK is set.
(gdb) bt
0 __pthread_kill_implementation (threadid=<optimized out>, signo=signo@entry=6, no_tid=no_tid@entry=0)
at pthread_kill.c:44
1 0x00007fc87148ec53 in __pthread_kill_internal (signo=6, threadid=<optimized out>) at pthread_kill.c:78
2 0x00007fc87143e956 in __GI_raise (sig=sig@entry=6) at ../sysdeps/posix/raise.c:26
3 0x00007fc8714287f4 in __GI_abort () at abort.c:79
4 0x00007fc87142871b in __assert_fail_base
(fmt=0x7fc8715bbde0 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=0x5606413efd53 "ret == 0", file=0x5606413ef87d "../accel/kvm/kvm-all.c", line=1837, function=<optimized out>) at assert.c:92
5 0x00007fc871437536 in __GI___assert_fail
(assertion=0x5606413efd53 "ret == 0", file=0x5606413ef87d "../accel/kvm/kvm-all.c", line=1837, function=0x5606413f06f0 <__PRETTY_FUNCTION__.19> "kvm_irqchip_commit_routes") at assert.c:101
6 0x0000560640f884b5 in kvm_irqchip_commit_routes (s=0x560642cae1f0) at ../accel/kvm/kvm-all.c:1837
7 0x0000560640c98f8e in virtio_pci_one_vector_unmask
(proxy=0x560643c65f00, queue_no=4294967295, vector=0, msg=..., n=0x560643c6e4c8)
at ../hw/virtio/virtio-pci.c:1005
8 0x0000560640c99201 in virtio_pci_vector_unmask (dev=0x560643c65f00, vector=0, msg=...)
at ../hw/virtio/virtio-pci.c:1070
9 0x0000560640bc402e in msix_fire_vector_notifier (dev=0x560643c65f00, vector=0, is_masked=false)
at ../hw/pci/msix.c:120
10 0x0000560640bc40f1 in msix_handle_mask_update (dev=0x560643c65f00, vector=0, was_masked=true)
at ../hw/pci/msix.c:140
11 0x0000560640bc4503 in msix_table_mmio_write (opaque=0x560643c65f00, addr=12, val=0, size=4)
at ../hw/pci/msix.c:231
12 0x0000560640f26d83 in memory_region_write_accessor
(mr=0x560643c66540, addr=12, value=0x7fc86b7bc628, size=4, shift=0, mask=4294967295, attrs=...)
at ../system/memory.c:497
13 0x0000560640f270a6 in access_with_adjusted_size
(addr=12, value=0x7fc86b7bc628, size=4, access_size_min=1, access_size_max=4, access_fn=0x560640f26c8d <memory_region_write_accessor>, mr=0x560643c66540, attrs=...) at ../system/memory.c:573
14 0x0000560640f2a2b5 in memory_region_dispatch_write (mr=0x560643c66540, addr=12, data=0, op=MO_32, attrs=...)
at ../system/memory.c:1521
15 0x0000560640f37bac in flatview_write_continue
(fv=0x7fc65805e0b0, addr=4273803276, attrs=..., ptr=0x7fc871e9c028, len=4, addr1=12, l=4, mr=0x560643c66540)
at ../system/physmem.c:2714
16 0x0000560640f37d0f in flatview_write
(fv=0x7fc65805e0b0, addr=4273803276, attrs=..., buf=0x7fc871e9c028, len=4) at ../system/physmem.c:2756
17 0x0000560640f380bf in address_space_write
(as=0x560642161ae0 <address_space_memory>, addr=4273803276, attrs=..., buf=0x7fc871e9c028, len=4)
at ../system/physmem.c:2863
18 0x0000560640f3812c in address_space_rw
(as=0x560642161ae0 <address_space_memory>, addr=4273803276, attrs=..., buf=0x7fc871e9c028, len=4, is_write=true) at ../system/physmem.c:2873
--Type <RET> for more, q to quit, c to continue without paging--
19 0x0000560640f8aa55 in kvm_cpu_exec (cpu=0x560642f205e0) at ../accel/kvm/kvm-all.c:2915
20 0x0000560640f8d731 in kvm_vcpu_thread_fn (arg=0x560642f205e0) at ../accel/kvm/kvm-accel-ops.c:51
21 0x00005606411949f4 in qemu_thread_start (args=0x560642f292b0) at ../util/qemu-thread-posix.c:541
22 0x00007fc87148cdcd in start_thread (arg=<optimized out>) at pthread_create.c:442
23 0x00007fc871512630 in clone3 () at ../sysdeps/unix/sysv/linux/x86_64/clone3.S:81
(gdb)
MST: coding style and typo fixups
Fixes: f9a09ca3ea ("vhost: add support for configure interrupt") Cc: qemu-stable@nongnu.org Signed-off-by: Cindy Lu <lulu@redhat.com>
Message-ID: <2321ade5f601367efe7380c04e3f61379c59b48f.1713173550.git.mst@redhat.com> Cc: Lei Yang <leiyang@redhat.com> Cc: Jason Wang <jasowang@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Tested-by: Cindy Lu <lulu@redhat.com>
target/sparc: Use GET_ASI_CODE for ASI_KERNELTXT and ASI_USERTXT
Reads are done with execute access. It is not clear whether writes
are legal at all -- for now, leave helper_st_asi unchanged, so that
we continue to raise an mmu fault.
This generalizes the exiting code for ASI_KERNELTXT to be usable for
ASI_USERTXT as well, by passing down the MemOpIdx to use.
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2281
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2059
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1609
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1166 Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Acked-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Tested-by: M Bazz <bazz@bazz1.com>
* tag 'for-upstream' of https://gitlab.com/bonzini/qemu:
meson.build: Disable -fzero-call-used-regs on OpenBSD
Makefile: fix use of -j without an argument
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Thomas Huth [Thu, 11 Apr 2024 12:08:19 +0000 (14:08 +0200)]
meson.build: Disable -fzero-call-used-regs on OpenBSD
QEMU currently does not work on OpenBSD since the -fzero-call-used-regs
option that we added to meson.build recently does not work with the
"retguard" extension from OpenBSD's Clang. Thus let's disable the
-fzero-call-used-regs here until there's a better solution available.
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2278 Signed-off-by: Thomas Huth <thuth@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240411120819.56417-1-thuth@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Our Makefile massages the given make arguments to invoke ninja
accordingly. One key difference is that ninja will parallelize by
default, whereas make only does so with -j<n> or -j. The make man page
says that "if the -j option is given without an argument, make will not
limit the number of jobs that can run simultaneously". We use to support
that by replacing -j with "" (empty string) when calling ninja, so that
it would do its auto-parallelization based on the number of CPU cores.
This was accidentally broken at d1ce2cc95b (Makefile: preserve
--jobserver-auth argument when calling ninja, 2024-04-02),
causing `make -j` to fail:
$ make -j V=1
/usr/bin/ninja -v -j -d keepdepfile all | cat
make -C contrib/plugins/ V="1" TARGET_DIR="contrib/plugins/" all
ninja: fatal: invalid -j parameter
make: *** [Makefile:161: run-ninja] Error
Let's fix that and indent the touched code for better readability.
Signed-off-by: Matheus Tavares Bernardino <quic_mathbern@quicinc.com> Fixes: d1ce2cc95b ("Makefile: preserve --jobserver-auth argument when calling ninja", 2024-04-02) Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
* tag 'hw-misc-20240410' of https://github.com/philmd/qemu:
hw/audio/virtio-snd: Remove unused assignment
hw/net/net_tx_pkt: Fix overrun in update_sctp_checksum()
hw/sd/sdhci: Do not update TRNMOD when Command Inhibit (DAT) is set
hw/net/lan9118: Fix overflow in MIL TX FIFO
hw/net/lan9118: Replace magic '2048' value by MIL_TXFIFO_SIZE definition
backends/cryptodev: Do not abort for invalid session ID
hw/misc/applesmc: Fix memory leak in reset() handler
hw/misc/applesmc: Do not call DeviceReset from DeviceRealize
hw/block/nand: Fix out-of-bound access in NAND block buffer
hw/block/nand: Have blk_load() take unsigned offset and return boolean
hw/block/nand: Factor nand_load_iolen() method out
qemu-options: Fix CXL Fixed Memory Window interleave-granularity typo
hw/virtio/virtio-crypto: Protect from DMA re-entrancy bugs
hw/char/virtio-serial-bus: Protect from DMA re-entrancy bugs
hw/display/virtio-gpu: Protect from DMA re-entrancy bugs
hw/virtio: Introduce virtio_bh_new_guarded() helper
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
>>> CID 1542933: Code maintainability issues (UNUSED_VALUE)
>>> CID 1542934: Code maintainability issues (UNUSED_VALUE)
>>> Assigning value "NULL" to "stream" here, but that stored
value is overwritten before it can be used.
hw/sd/sdhci: Do not update TRNMOD when Command Inhibit (DAT) is set
Per "SD Host Controller Standard Specification Version 3.00":
* 2.2.5 Transfer Mode Register (Offset 00Ch)
Writes to this register shall be ignored when the Command
Inhibit (DAT) in the Present State register is 1.
Do not update the TRNMOD register when Command Inhibit (DAT)
bit is set to avoid the present-status register going out of
sync, leading to malicious guest using DMA mode and overflowing
the FIFO buffer:
Stack trace (part):
=================================================================
==89993==ERROR: AddressSanitizer: heap-buffer-overflow on address
0x615000029900 at pc 0x55d5f885700d bp 0x7ffc1e1e9470 sp 0x7ffc1e1e9468
WRITE of size 1 at 0x615000029900 thread T0
#0 0x55d5f885700c in sdhci_write_dataport hw/sd/sdhci.c:564:39
#1 0x55d5f8849150 in sdhci_write hw/sd/sdhci.c:1223:13
#2 0x55d5fa01db63 in memory_region_write_accessor system/memory.c:497:5
#3 0x55d5fa01d245 in access_with_adjusted_size system/memory.c:573:18
#4 0x55d5fa01b1a9 in memory_region_dispatch_write system/memory.c:1521:16
#5 0x55d5fa09f5c9 in flatview_write_continue system/physmem.c:2711:23
#6 0x55d5fa08f78b in flatview_write system/physmem.c:2753:12
#7 0x55d5fa08f258 in address_space_write system/physmem.c:2860:18
...
0x615000029900 is located 0 bytes to the right of 512-byte region
[0x615000029700,0x615000029900) allocated by thread T0 here:
#0 0x55d5f7237b27 in __interceptor_calloc
#1 0x7f9e36dd4c50 in g_malloc0
#2 0x55d5f88672f7 in sdhci_pci_realize hw/sd/sdhci-pci.c:36:5
#3 0x55d5f844b582 in pci_qdev_realize hw/pci/pci.c:2092:9
#4 0x55d5fa2ee74b in device_set_realized hw/core/qdev.c:510:13
#5 0x55d5fa325bfb in property_set_bool qom/object.c:2358:5
#6 0x55d5fa31ea45 in object_property_set qom/object.c:1472:5
#7 0x55d5fa332509 in object_property_set_qobject om/qom-qobject.c:28:10
#8 0x55d5fa31f6ed in object_property_set_bool qom/object.c:1541:15
#9 0x55d5fa2e2948 in qdev_realize hw/core/qdev.c:292:12
#10 0x55d5f8eed3f1 in qdev_device_add_from_qdict system/qdev-monitor.c:719:10
#11 0x55d5f8eef7ff in qdev_device_add system/qdev-monitor.c:738:11
#12 0x55d5f8f211f0 in device_init_func system/vl.c:1200:11
#13 0x55d5fad0877d in qemu_opts_foreach util/qemu-option.c:1135:14
#14 0x55d5f8f0df9c in qemu_create_cli_devices system/vl.c:2638:5
#15 0x55d5f8f0db24 in qmp_x_exit_preconfig system/vl.c:2706:5
#16 0x55d5f8f14dc0 in qemu_init system/vl.c:3737:9
...
SUMMARY: AddressSanitizer: heap-buffer-overflow hw/sd/sdhci.c:564:39
in sdhci_write_dataport
Add assertions to ensure the fifo_buffer[] is not overflowed by
malicious accesses to the Buffer Data Port register.
Fixes: CVE-2024-3447 Cc: qemu-stable@nongnu.org Fixes: d7dfca0807 ("hw/sdhci: introduce standard SD host controller") Buglink: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=58813 Reported-by: Alexander Bulekov <alxndr@bu.edu> Reported-by: Chuhong Yuan <hslester96@gmail.com> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <CAFEAcA9iLiv1XGTGKeopgMa8Y9+8kvptvsb8z2OBeuy+5=NUfg@mail.gmail.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20240409145524.27913-1-philmd@linaro.org>
When the MAC Interface Layer (MIL) transmit FIFO is full,
truncate the packet, and raise the Transmitter Error (TXE)
flag.
Broken since model introduction in commit 2a42499017
("LAN9118 emulation").
When using the reproducer from
https://gitlab.com/qemu-project/qemu/-/issues/2267 we get:
hw/net/lan9118.c:798:17: runtime error:
index 2048 out of bounds for type 'uint8_t[2048]' (aka 'unsigned char[2048]')
#0 0x563ec9a057b1 in tx_fifo_push hw/net/lan9118.c:798:43
#1 0x563ec99fbb28 in lan9118_writel hw/net/lan9118.c:1042:9
#2 0x563ec99f2de2 in lan9118_16bit_mode_write hw/net/lan9118.c:1205:9
#3 0x563ecbf78013 in memory_region_write_accessor system/memory.c:497:5
#4 0x563ecbf776f5 in access_with_adjusted_size system/memory.c:573:18
#5 0x563ecbf75643 in memory_region_dispatch_write system/memory.c:1521:16
#6 0x563ecc01bade in flatview_write_continue_step system/physmem.c:2713:18
#7 0x563ecc01b374 in flatview_write_continue system/physmem.c:2743:19
#8 0x563ecbff1c9b in flatview_write system/physmem.c:2774:12
#9 0x563ecbff1768 in address_space_write system/physmem.c:2894:18
...
[*] LAN9118 DS00002266B.pdf, Table 5.3.3 "INTERRUPT STATUS REGISTER"
Cc: qemu-stable@nongnu.org Reported-by: Will Lester Reported-by: Chuhong Yuan <hslester96@gmail.com> Suggested-by: Peter Maydell <peter.maydell@linaro.org>
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2267 Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20240409133801.23503-3-philmd@linaro.org>
hw/net/lan9118: Replace magic '2048' value by MIL_TXFIFO_SIZE definition
The magic 2048 is explained in the LAN9211 datasheet (DS00002414A)
in chapter 1.4, "10/100 Ethernet MAC":
The MAC Interface Layer (MIL), within the MAC, contains a
2K Byte transmit and a 128 Byte receive FIFO which is separate
from the TX and RX FIFOs. [...]
Note, the use of the constant in lan9118_receive() reveals that
our implementation is using the same buffer for both tx and rx.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20240409133801.23503-2-philmd@linaro.org>
hw/misc/applesmc: Do not call DeviceReset from DeviceRealize
QDev core layer always call DeviceReset() after DeviceRealize(),
no need to do it manually. Remove the extra call.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20240408095217.57239-2-philmd@linaro.org>
=================================================================
==15750==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x61f000000de0
at pc 0x560e61557210 bp 0x7ffcfc4a59f0 sp 0x7ffcfc4a59e8
READ of size 1 at 0x61f000000de0 thread T0
#0 0x560e6155720f in mem_and hw/block/nand.c:101:20
#1 0x560e6155ac9c in nand_blk_write_512 hw/block/nand.c:663:9
#2 0x560e61544200 in nand_command hw/block/nand.c:293:13
#3 0x560e6153cc83 in nand_setio hw/block/nand.c:520:13
#4 0x560e61a0a69e in tc6393xb_nand_writeb hw/display/tc6393xb.c:380:13
#5 0x560e619f9bf7 in tc6393xb_writeb hw/display/tc6393xb.c:524:9
#6 0x560e647c7d03 in memory_region_write_accessor softmmu/memory.c:492:5
#7 0x560e647c7641 in access_with_adjusted_size softmmu/memory.c:554:18
#8 0x560e647c5f66 in memory_region_dispatch_write softmmu/memory.c:1514:16
#9 0x560e6485409e in flatview_write_continue softmmu/physmem.c:2825:23
#10 0x560e648421eb in flatview_write softmmu/physmem.c:2867:12
#11 0x560e64841ca8 in address_space_write softmmu/physmem.c:2963:18
#12 0x560e61170162 in qemu_writeb tests/qtest/videzzo/videzzo_qemu.c:1080:5
#13 0x560e6116eef7 in dispatch_mmio_write tests/qtest/videzzo/videzzo_qemu.c:1227:28
0x61f000000de0 is located 0 bytes to the right of 3424-byte region [0x61f000000080,0x61f000000de0)
allocated by thread T0 here:
#0 0x560e611276cf in malloc /root/llvm-project/compiler-rt/lib/asan/asan_malloc_linux.cpp:145:3
#1 0x7f7959a87e98 in g_malloc (/lib/x86_64-linux-gnu/libglib-2.0.so.0+0x57e98)
#2 0x560e64b98871 in object_new qom/object.c:749:12
#3 0x560e64b5d1a1 in qdev_new hw/core/qdev.c:153:19
#4 0x560e61547ea5 in nand_init hw/block/nand.c:639:11
#5 0x560e619f8772 in tc6393xb_init hw/display/tc6393xb.c:558:16
#6 0x560e6390bad2 in tosa_init hw/arm/tosa.c:250:12
SUMMARY: AddressSanitizer: heap-buffer-overflow hw/block/nand.c:101:20 in mem_and
==15750==ABORTING
Broken since introduction in commit 3e3d5815cb ("NAND Flash memory
emulation and ECC calculation helpers for use by NAND controllers").
Cc: qemu-stable@nongnu.org
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1445
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1446 Reported-by: Qiang Liu <cyruscyliu@gmail.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20240409135944.24997-4-philmd@linaro.org>
hw/block/nand: Have blk_load() take unsigned offset and return boolean
Negative offset is meaningless, use unsigned type.
Return a boolean value indicating success.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20240409135944.24997-3-philmd@linaro.org>
hw/block/nand: Factor nand_load_iolen() method out
Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20240409135944.24997-2-philmd@linaro.org>
Fix the unit typo of interleave-granularity of CXL Fixed Memory
Window in qemu-option.hx.
Fixes: 03b39fcf64 ("hw/cxl: Make the CFMW a machine parameter.") Signed-off-by: Yuquan Wang wangyuquan1236@phytium.com.cn
Message-ID: <20240407083539.1488172-2-wangyuquan1236@phytium.com.cn>
[PMD: Reworded] Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
hw/virtio/virtio-crypto: Protect from DMA re-entrancy bugs
Replace qemu_bh_new_guarded() by virtio_bh_new_guarded()
so the bus and device use the same guard. Otherwise the
DMA-reentrancy protection can be bypassed.
Fixes: CVE-2024-3446 Cc: qemu-stable@nongnu.org Suggested-by: Alexander Bulekov <alxndr@bu.edu> Reviewed-by: Gerd Hoffmann <kraxel@redhat.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20240409105537.18308-5-philmd@linaro.org>
hw/char/virtio-serial-bus: Protect from DMA re-entrancy bugs
Replace qemu_bh_new_guarded() by virtio_bh_new_guarded()
so the bus and device use the same guard. Otherwise the
DMA-reentrancy protection can be bypassed.
Fixes: CVE-2024-3446 Cc: qemu-stable@nongnu.org Suggested-by: Alexander Bulekov <alxndr@bu.edu> Reviewed-by: Gerd Hoffmann <kraxel@redhat.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20240409105537.18308-4-philmd@linaro.org>
hw/display/virtio-gpu: Protect from DMA re-entrancy bugs
Replace qemu_bh_new_guarded() by virtio_bh_new_guarded()
so the bus and device use the same guard. Otherwise the
DMA-reentrancy protection can be bypassed:
0x60d000011178 is located 56 bytes inside of 136-byte region [0x60d000011140,0x60d0000111c8)
freed by thread T0 here:
#0 0x562cc2adb662 in __interceptor_free (qemu-system-i386+0x239e662)
#1 0x562cc3b86b21 in virtio_gpu_reset hw/display/virtio-gpu.c:1524:9
#2 0x562cc416e20e in virtio_reset hw/virtio/virtio.c:2145:9
#3 0x562cc37c5644 in virtio_pci_reset hw/virtio/virtio-pci.c:2249:5
#4 0x562cc4233758 in memory_region_write_accessor system/memory.c:497:5
#5 0x562cc4232eea in access_with_adjusted_size system/memory.c:573:18
previously allocated by thread T0 here:
#0 0x562cc2adb90e in malloc (qemu-system-i386+0x239e90e)
#1 0x7f368405a678 in g_malloc (/lib/x86_64-linux-gnu/libglib-2.0.so.0+0x5a678)
#2 0x562cc4163ffc in virtqueue_split_pop hw/virtio/virtio.c:1612:12
#3 0x562cc4163ffc in virtqueue_pop hw/virtio/virtio.c:1783:16
#4 0x562cc3b91a95 in virtio_gpu_handle_ctrl hw/display/virtio-gpu.c:1112:15
#5 0x562cc4a85514 in aio_bh_call util/async.c:169:5
#6 0x562cc4a85c52 in aio_bh_poll util/async.c:216:13
#7 0x562cc4a1a79b in aio_dispatch util/aio-posix.c:423:5
SUMMARY: AddressSanitizer: heap-use-after-free hw/display/virtio-gpu.c:180:42 in virtio_gpu_ctrl_response
With this change, the same reproducer triggers:
qemu-system-i386: warning: Blocked re-entrant IO on MemoryRegion: virtio-pci-common-virtio-gpu at addr: 0x6
Fixes: CVE-2024-3446 Cc: qemu-stable@nongnu.org Reported-by: Alexander Bulekov <alxndr@bu.edu> Reported-by: Yongkang Jia <kangel@zju.edu.cn> Reported-by: Xiao Lei <nop.leixiao@gmail.com> Reported-by: Yiming Tao <taoym@zju.edu.cn> Buglink: https://bugs.launchpad.net/qemu/+bug/1888606 Reviewed-by: Gerd Hoffmann <kraxel@redhat.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20240409105537.18308-3-philmd@linaro.org>
Introduce virtio_bh_new_guarded(), similar to qemu_bh_new_guarded()
but using the transport memory guard, instead of the device one
(there can only be one virtio device per virtio bus).
Inspired-by: Gerd Hoffmann <kraxel@redhat.com> Reviewed-by: Gerd Hoffmann <kraxel@redhat.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20240409105537.18308-2-philmd@linaro.org>
* tag 'pull-misc-20240409' of https://gitlab.com/rth7680/qemu:
linux-user: Preserve unswapped siginfo_t for strace
accel/tcg: Improve can_do_io management
target/s390x: Use insn_start from DisasContextBase
target/riscv: Use insn_start from DisasContextBase
target/microblaze: Use insn_start from DisasContextBase
target/i386: Preserve DisasContextBase.insn_start across rewind
target/hppa: Use insn_start from DisasContextBase
target/arm: Use insn_start from DisasContextBase
accel/tcg: Add insn_start to DisasContextBase
tcg: Add TCGContext.emit_before_op
target/m68k: Map FPU exceptions to FPSR register
target/sh4: add missing CHECK_NOT_DELAY_SLOT
target/sh4: Fix mac.w with saturation enabled
target/sh4: Fix mac.l with saturation enabled
target/sh4: Merge mach and macl into a union
target/sh4: mac.w: memory accesses are 16-bit words
target/hppa: Fix IIAOQ, IIASQ for pa2.0
linux-user: replace calloc() with g_new0()
linux-user: Fix waitid return of siginfo_t and rusage
tcg/optimize: Do not attempt to constant fold neg_vec
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
linux-user: Preserve unswapped siginfo_t for strace
Passing the tswapped structure to strace means that
our internal si_type is also gone, which then aborts
in print_siginfo.
Fixes: 4d6d8a05a0a ("linux-user: Move tswap_siginfo out of target code") Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
We already attempted to set and clear can_do_io before the first
and last insns, but only used the initial value of max_insns and
the call to translator_io_start to find those insns.
Now that we track insn_start in DisasContextBase, and now that
we have emit_before_op, we can wait until we have finished
translation to identify the true first and last insns and emit
the sets of can_do_io at that time.
This fixes the case of a translation block which crossed a page
boundary, and for which the second page turned out to be mmio.
In this case we truncate the block, and the previous logic for
can_do_io could leave a block with a single insn with can_do_io
set to false, which would fail an assertion in cpu_io_recompile.
Reported-by: Jørgen Hansen <Jorgen.Hansen@wdc.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Tested-by: Jørgen Hansen <Jorgen.Hansen@wdc.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Keith Packard [Thu, 3 Aug 2023 03:52:31 +0000 (20:52 -0700)]
target/m68k: Map FPU exceptions to FPSR register
Add helpers for reading/writing the 68881 FPSR register so that
changes in floating point exception state can be seen by the
application.
Call these helpers in pre_load/post_load hooks to synchronize
exception state.
Signed-off-by: Keith Packard <keithp@keithp.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230803035231.429697-1-keithp@keithp.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
The saturation arithmetic logic in helper_macw is not correct.
I tested and verified this behavior on a SH7091.
Reviewd-by: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Zack Buhman <zack@buhman.org>
Message-Id: <20240405233802.29128-3-zack@buhman.org>
[rth: Reformat helper_macw, add a test case.] Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
linux-user: Fix waitid return of siginfo_t and rusage
The copy back to siginfo_t should be conditional only on arg3,
not the specific values that might have been written.
The copy back to rusage was missing entirely.
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2262 Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Tested-by: Alex Fan <alex.fan.q@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
* tag 'edk2-20240409-pull-request' of https://gitlab.com/kraxel/qemu:
edk2: rebuild binaries with correct version information
edk2/seabios: use common extra version
edk2: commit version info
edk2: get version + date from git submodule
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Gerd Hoffmann [Wed, 27 Mar 2024 10:24:45 +0000 (11:24 +0100)]
edk2: get version + date from git submodule
Turned out hard-coding version and date in the Makefile wasn't a bright
idea. Updating it on edk2 updates is easily forgotten. Fetch the info
from git instead. Store in edk2-version, so this can be committed to
the repo and is present in tarballs too.
Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-ID: <20240327102448.61877-2-kraxel@redhat.com>
* tag 'for_upstream' of https://git.kernel.org/pub/scm/virt/kvm/mst/qemu:
qdev-monitor: fix error message in find_device_state()
vhost-user-blk: simplify and fix vhost_user_blk_handle_config_change
vdpa-dev: Fix the issue of device status not updating when configuration interruption is triggered
hw/virtio: Fix packed virtqueue flush used_idx
virtio-snd: rewrite invalid tx/rx message handling
virtio-snd: Enhance error handling for invalid transfers
Revert "hw/virtio: Add support for VDPA network simulation devices"
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Peter Maydell [Tue, 9 Apr 2024 08:51:00 +0000 (09:51 +0100)]
Merge tag 'for-upstream' of https://gitlab.com/bonzini/qemu into staging
* fall back to non-ioeventfd notification if KVM routing table is full
* support kitware ninja with jobserver support
* nanomips: fix warnings with GCC 14
* tag 'for-upstream' of https://gitlab.com/bonzini/qemu:
kvm: error out of kvm_irqchip_add_msi_route() in case of full route table
nanomips: fix warnings with GCC 14
Makefile: preserve --jobserver-auth argument when calling ninja
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
qdev-monitor: fix error message in find_device_state()
This "hotpluggable" here is misleading. Actually we check is object a
device or not. Let's drop the word.
Suggested-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru> Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20240329183758.3360733-3-vsementsov@yandex-team.ru> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
vhost-user-blk: simplify and fix vhost_user_blk_handle_config_change
Let's not care about what was changed and update the whole config,
reasons:
1. config->geometry should be updated together with capacity, so we fix
a bug.
2. Vhost-user protocol doesn't say anything about config change
limitation. Silent ignore of changes doesn't seem to be correct.
3. vhost-user-vsock reads the whole config
4. on realize we don't do any checks on retrieved config, so no reason
to care here
Comment "valid for resize only" exists since introduction the whole
hw/block/vhost-user-blk.c in commit 00343e4b54ba0685e9ebe928ec5713b0cf7f1d1c
"vhost-user-blk: introduce a new vhost-user-blk host device",
seems it was just an extra limitation.
Also, let's notify guest unconditionally:
1. So does vhost-user-vsock
2. We are going to reuse the functionality in new cases when we do want
to notify the guest unconditionally. So, no reason to create extra
branches in the logic.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru> Acked-by: Raphael Norwitz <raphael.norwitz@nutanix.com>
Message-Id: <20240329183758.3360733-2-vsementsov@yandex-team.ru> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
vdpa-dev: Fix the issue of device status not updating when configuration interruption is triggered
The set_config callback function vhost_vdpa_device_get_config in
vdpa-dev does not fetch the current device status from the hardware
device, causing the guest os to not receive the latest device status
information.
The hardware updates the config status of the vdpa device and then
notifies the os. The guest os receives an interrupt notification,
triggering a get_config access in the kernel, which then enters qemu
internally. Ultimately, the vhost_vdpa_device_get_config function of
vdpa-dev is called
One scenario encountered is when the device needs to bring down the
vdpa net device. After modifying the status field of virtio_net_config
in the hardware, it sends an interrupt notification. However, the guest
os always receives the STATUS field as VIRTIO_NET_S_LINK_UP.
Signed-off-by: Yuxue Liu <yuxue.liu@jaguarmicro.com> Acked-by: Jason Wang <jasowang@redhat.com>
Message-Id: <20240408020003.1979-1-yuxue.liu@jaguarmicro.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
In the event of writing many chains of descriptors, the device must
write just the id of the last buffer in the descriptor chain, skip
forward the number of descriptors in the chain, and then repeat the
operations for the rest of chains.
Current QEMU code writes all the buffer ids consecutively, and then
skips all the buffers altogether. This is a bug, and can be reproduced
with a VirtIONet device with _F_MRG_RXBUB and without
_F_INDIRECT_DESC:
If a virtio-net device has the VIRTIO_NET_F_MRG_RXBUF feature
but not the VIRTIO_RING_F_INDIRECT_DESC feature,
'VirtIONetQueue->rx_vq' will use the merge feature
to store data in multiple 'elems'.
The 'num_buffers' in the virtio header indicates how many elements are merged.
If the value of 'num_buffers' is greater than 1,
all the merged elements will be filled into the descriptor ring.
The 'idx' of the elements should be the value of 'vq->used_idx' plus 'ndescs'.
Fixes: 86044b24e8 ("virtio: basic packed virtqueue support") Acked-by: Eugenio Pérez <eperezma@redhat.com> Signed-off-by: Wafer <wafer@jaguarmicro.com>
Message-Id: <20240407015451.5228-2-wafer@jaguarmicro.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The current handling of invalid virtqueue elements inside the TX/RX virt
queue handlers is wrong.
They are added in a per-stream invalid queue to be processed after the
handler is done examining each message, but the invalid message might
not be specifying any stream_id; which means it's invalid to add it to
any stream->invalid queue since stream could be NULL at this point.
This commit moves the invalid queue to the VirtIOSound struct which
guarantees there will always be a valid temporary place to store them
inside the tx/rx handlers. The queue will be emptied before the handler
returns, so the queue must be empty at any other point of the device's
lifetime.
Signed-off-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org>
Message-Id: <virtio-snd-rewrite-invalid-tx-rx-message-handling-v1.manos.pitsidianakis@linaro.org> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Zheyu Ma [Fri, 22 Mar 2024 11:08:27 +0000 (12:08 +0100)]
virtio-snd: Enhance error handling for invalid transfers
This patch improves error handling in virtio_snd_handle_tx_xfer()
and virtio_snd_handle_rx_xfer() in the VirtIO sound driver. Previously,
'goto' statements were used for error paths, leading to unnecessary
processing and potential null pointer dereferences. Now, 'continue' is
used to skip the rest of the current loop iteration for errors such as
message size discrepancies or null streams, reducing crash risks.
ASAN log illustrating the issue addressed:
ERROR: AddressSanitizer: SEGV on unknown address 0x0000000000b4
#0 0x57cea39967b8 in qemu_mutex_lock_impl qemu/util/qemu-thread-posix.c:92:5
#1 0x57cea128c462 in qemu_mutex_lock qemu/include/qemu/thread.h:122:5
#2 0x57cea128d72f in qemu_lockable_lock qemu/include/qemu/lockable.h:95:5
#3 0x57cea128c294 in qemu_lockable_auto_lock qemu/include/qemu/lockable.h:105:5
#4 0x57cea1285eb2 in virtio_snd_handle_rx_xfer qemu/hw/audio/virtio-snd.c:1026:9
#5 0x57cea2caebbc in virtio_queue_notify_vq qemu/hw/virtio/virtio.c:2268:9
#6 0x57cea2cae412 in virtio_queue_host_notifier_read qemu/hw/virtio/virtio.c:3671:9
#7 0x57cea39822f1 in aio_dispatch_handler qemu/util/aio-posix.c:372:9
#8 0x57cea3979385 in aio_dispatch_handlers qemu/util/aio-posix.c:414:20
#9 0x57cea3978eb1 in aio_dispatch qemu/util/aio-posix.c:424:5
#10 0x57cea3a1eede in aio_ctx_dispatch qemu/util/async.c:360:5
Signed-off-by: Zheyu Ma <zheyuma97@gmail.com> Reviewed-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org>
Message-Id: <20240322110827.568412-1-zheyuma97@gmail.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The patch adds non-upstream code in
include/standard-headers/linux/virtio_pci.h
which would make maintainance harder.
Revert for now.
Suggested-by: Jason Wang <jasowang@redhat.com>
Message-Id: <df6b6b465753e754a19459e8cd61416548f89a42.1712569644.git.mst@redhat.com> Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Igor Mammedov [Mon, 8 Apr 2024 11:09:56 +0000 (13:09 +0200)]
kvm: error out of kvm_irqchip_add_msi_route() in case of full route table
subj is calling kvm_add_routing_entry() which simply extends
KVMState::irq_routes::entries[]
but doesn't check if number of routes goes beyond limit the kernel
is willing to accept. Which later leads toi the assert
typically it happens during guest boot for large enough guest
Reproduced with:
./qemu --enable-kvm -m 8G -smp 64 -machine pc \
`for b in {1..2}; do echo -n "-device pci-bridge,id=pci$b,chassis_nr=$b ";
for i in {0..31}; do touch /tmp/vblk$b$i;
echo -n "-drive file=/tmp/vblk$b$i,if=none,id=drive$b$i,format=raw
-device virtio-blk-pci,drive=drive$b$i,bus=pci$b ";
done; done`
While crash at boot time is bad, the same might happen at hotplug time
which is unacceptable.
So instead calling kvm_add_routing_entry() unconditionally, check first
that number of routes won't exceed KVM_CAP_IRQ_ROUTING. This way virtio
device insteads killin qemu, will gracefully fail to initialize device
as expected with following warnings on console:
virtio-blk failed to set guest notifier (-28), ensure -accel kvm is set.
virtio_bus_start_ioeventfd: failed. Fallback to userspace (slower).
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Message-ID: <20240408110956.451558-1-imammedo@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Mon, 8 Apr 2024 14:10:40 +0000 (16:10 +0200)]
nanomips: fix warnings with GCC 14
GCC 14 shows -Wshadow=local warnings if an enum conflicts with a local
variable (including a parameter). To avoid this, move the problematic
enum and all of its dependencies after the hundreds of functions that
have a parameter named "instruction".
Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
* tag 'pull-target-arm-20240408' of https://git.linaro.org/people/pmaydell/qemu-arm:
target/arm: Use correct SecuritySpace for AArch64 AT ops at EL3
target/arm: Fix CNTPOFF_EL2 trap to missing EL3
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Peter Maydell [Mon, 8 Apr 2024 15:24:04 +0000 (16:24 +0100)]
Merge tag 'migration-20240407-pull-request' of https://gitlab.com/peterx/qemu into staging
Migration pull for 9.0-rc3
- Wei/Lei's fix on a rare postcopy race that can hang the channel (since 8.0)
- Avihai's fix on maintainers file, points to the right doc links
# -----BEGIN PGP SIGNATURE-----
#
# iIgEABYKADAWIQS5GE3CDMRX2s990ak7X8zN86vXBgUCZhLpJBIccGV0ZXJ4QHJl
# ZGhhdC5jb20ACgkQO1/MzfOr1wa87AEAhvXqJyLxYYdlQ5fqp4hVV6O/3N1vNHMu
# kT3d9tmM0jsBAJ5KxK176iGDp+ej5MEyYSm1gG7ivj3y3v3wlPnSmJMJ
# =T1lk
# -----END PGP SIGNATURE-----
# gpg: Signature made Sun 07 Apr 2024 19:42:44 BST
# gpg: using EDDSA key B9184DC20CC457DACF7DD1A93B5FCCCDF3ABD706
# gpg: issuer "peterx@redhat.com"
# gpg: Good signature from "Peter Xu <xzpeter@gmail.com>" [marginal]
# gpg: aka "Peter Xu <peterx@redhat.com>" [marginal]
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg: It is not certain that the signature belongs to the owner.
# Primary key fingerprint: B918 4DC2 0CC4 57DA CF7D D1A9 3B5F CCCD F3AB D706
* tag 'migration-20240407-pull-request' of https://gitlab.com/peterx/qemu:
MAINTAINERS: Adjust migration documentation files
migration/postcopy: ensure preempt channel is ready before loading states
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Peter Maydell [Fri, 5 Apr 2024 18:02:32 +0000 (19:02 +0100)]
target/arm: Use correct SecuritySpace for AArch64 AT ops at EL3
When we do an AT address translation operation, the page table walk
is supposed to be performed in the context of the EL we're doing the
walk for, so for instance an AT S1E2R walk is done for EL2. In the
pseudocode an EL is passed to AArch64.AT(), which calls
SecurityStateAtEL() to find the security state that we should be
doing the walk with.
In ats_write64() we get this wrong, instead using the current
security space always. This is fine for AT operations performed from
EL1 and EL2, because there the current security state and the
security state for the lower EL are the same. But for AT operations
performed from EL3, the current security state is always either
Secure or Root, whereas we want to use the security state defined by
SCR_EL3.{NS,NSE} for the walk. This affects not just guests using
FEAT_RME but also ones where EL3 is Secure state and the EL3 code
is trying to do an AT for a NonSecure EL2 or EL1.
Use arm_security_space_below_el3() to get the SecuritySpace to
pass to do_ats_write() for all AT operations except the
AT S1E3* operations.
Cc: qemu-stable@nongnu.org Fixes: e1ee56ec2383 ("target/arm: Pass security space rather than flag for AT instructions")
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2250 Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20240405180232.3570066-1-peter.maydell@linaro.org
Makefile: preserve --jobserver-auth argument when calling ninja
Qemu wraps its call to ninja in a Makefile. Since ninja, as opposed to
make, utilizes all CPU cores by default, the qemu Makefile translates
the absense of a `-jN` argument into `-j1`. This breaks jobserver
functionality, so update the -jN mangling to take the --jobserver-auth
argument into considerationa too.
Signed-off-by: Martin Hundebøll <martin@geanix.com>
Message-Id: <20240402081738.1051560-1-martin@geanix.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Commit 8cb2f8b172e7 ("docs/migration: Create migration/ directory")
changed migration documentation file structure but forgot to update the
entries in the MAINTAINERS file.
Commit 4c6f8a79ae53 ("docs/migration: Split 'dirty limit'") extracted
dirty limit documentation to a new file without updating dirty limit
section in MAINTAINERS file.
Wei Wang [Fri, 5 Apr 2024 03:40:56 +0000 (11:40 +0800)]
migration/postcopy: ensure preempt channel is ready before loading states
Before loading the guest states, ensure that the preempt channel has been
ready to use, as some of the states (e.g. via virtio_load) might trigger
page faults that will be handled through the preempt channel. So yield to
the main thread in the case that the channel create event hasn't been
dispatched.
EL2 accesses to CNTPOFF_EL2 should only ever trap to EL3 if EL3 is
present, as described by the reference manual (for MRS):
/* ... */
elsif PSTATE.EL == EL2 then
if Halted() && HaveEL(EL3) && /*...*/ then
UNDEFINED;
elsif HaveEL(EL3) && SCR_EL3.ECVEn == '0' then
/* ... */
else
X[t, 64] = CNTPOFF_EL2;
However, the existing implementation of gt_cntpoff_access() always
returns CP_ACCESS_TRAP_EL3 for EL2 accesses with SCR_EL3.ECVEn unset. In
pseudo-code terminology, this corresponds to assuming that HaveEL(EL3)
is always true, which is wrong. As a result, QEMU panics in
access_check_cp_reg() when started without EL3 and running EL2 code
accessing the register (e.g. any recent KVM booting a guest).
Therefore, add the HaveEL(EL3) check to gt_cntpoff_access().
* tag 'qemu-sparc-20240404' of https://github.com/mcayland/qemu:
esp.c: remove explicit setting of DRQ within ESP state machine
esp.c: ensure esp_pdma_write() always calls esp_fifo_push()
esp.c: update esp_fifo_{push, pop}() to call esp_update_drq()
esp.c: introduce esp_update_drq() and update esp_fifo_{push, pop}_buf() to use it
esp.c: move esp_set_phase() and esp_get_phase() towards the beginning of the file
esp.c: prevent cmdfifo overflow in esp_cdb_ready()
esp.c: rework esp_cdb_length() into esp_cdb_ready()
esp.c: don't assert() if FIFO empty when executing non-DMA SELATNS
esp.c: introduce esp_fifo_push_buf() function for pushing to the FIFO
esp.c: change esp_fifo_pop_buf() to take ESPState
esp.c: use esp_fifo_push() instead of fifo8_push()
esp.c: change esp_fifo_pop() to take ESPState
esp.c: change esp_fifo_push() to take ESPState
esp.c: replace cmdfifo use of esp_fifo_pop() in do_message_phase()
esp.c: replace esp_fifo_pop_buf() with esp_fifo8_pop_buf() in do_message_phase()
esp.c: replace esp_fifo_pop_buf() with esp_fifo8_pop_buf() in do_command_phase()
esp.c: move esp_fifo_pop_buf() internals to new esp_fifo8_pop_buf() function
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
This ensures that esp_update_drq() is called via esp_fifo_push() whenever the
host uses PDMA to transfer data to a SCSI device.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20240324191707.623175-17-mark.cave-ayland@ilande.co.uk> Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Mark Cave-Ayland [Sun, 24 Mar 2024 19:17:04 +0000 (19:17 +0000)]
esp.c: update esp_fifo_{push, pop}() to call esp_update_drq()
This ensures that the DRQ line is always set correctly when reading/writing
single bytes to/from the FIFO.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20240324191707.623175-16-mark.cave-ayland@ilande.co.uk> Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
projects / mirror_qemu.git / log