The vector_use callback is used to enable vector that is unmasked in
guest. The kernel used to only support static MSI-X allocation. When
allocating a new interrupt using "static MSI-X allocation" kernels,
QEMU first disables all previously allocated vectors and then
re-allocates all including the new one. The nr_vectors of VFIOPCIDevice
indicates that all vectors from 0 to nr_vectors are allocated (and may
be enabled), which is used to loop all the possibly used vectors when
e.g., disabling MSI-X interrupts.
Extend the vector_use function to support dynamic MSI-X allocation when
host supports the capability. QEMU therefore can individually allocate
and enable a new interrupt without affecting others or causing interrupts
lost during runtime.
Utilize nr_vectors to calculate the upper bound of enabled vectors in
dynamic MSI-X allocation mode since looping all msix_entries_nr is not
efficient and unnecessary.
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Tested-by: Reinette Chatre <reinette.chatre@intel.com>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Kernel provides the guidance of dynamic MSI-X allocation support of
passthrough device, by clearing the VFIO_IRQ_INFO_NORESIZE flag to
guide user space.
Fetch the flags from host to determine if dynamic MSI-X allocation is
supported.
Originally-by: Reinette Chatre <reinette.chatre@intel.com>
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
vfio_put_device() is a VFIO PCI specific function, rename it with
'vfio_pci' prefix to avoid confusing.
No functional change.
Suggested-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
The below referenced commit renames scanout_width/height to
backing_width/height, but also promotes these fields in various portions
of the egl interface. Meanwhile vfio dmabuf support has never used the
previous scanout fields and is therefore missed in the update. This
results in a black screen when transitioning from ramfb to dmabuf display
when using Intel vGPU with these features.
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1891
Link: https://lists.gnu.org/archive/html/qemu-devel/2023-08/msg02726.html
Fixes: 9ac06df8b6 ("virtio-gpu-udmabuf: correct naming of QemuDmaBuf size properties")
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Tested-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Allow a client to request a subset of negotiated meta contexts. For
example, a client may ask to use a single connection to learn about
both block status and dirty bitmaps, but where the dirty bitmap
queries only need to be performed on a subset of the disk; forcing the
server to compute that information on block status queries in the rest
of the disk is wasted effort (both at the server, and on the amount of
traffic sent over the wire to be parsed and ignored by the client).
Qemu as an NBD client never requests to use more than one meta
context, so it has no need to use block status payloads. Testing this
instead requires support from libnbd, which CAN access multiple meta
contexts in parallel from a single NBD connection; an interop test
submitted to the libnbd project at the same time as this patch
demonstrates the feature working, as well as testing some corner cases
(for example, when the payload length is longer than the export
length), although other corner cases (like passing the same id
duplicated) requires a protocol fuzzer because libnbd is not wired up
to break the protocol that badly.
This also includes tweaks to 'qemu-nbd --list' to show when a server
is advertising the capability, and to the testsuite to reflect the
addition to that output.
Of note: qemu will always advertise the new feature bit during
NBD_OPT_INFO if extended headers have alreay been negotiated
(regardless of whether any NBD_OPT_SET_META_CONTEXT negotiation has
occurred); but for NBD_OPT_GO, qemu only advertises the feature if
block status is also enabled (that is, if the client does not
negotiate any contexts, then NBD_CMD_BLOCK_STATUS cannot be used, so
the feature is not advertised).
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-ID: <20230925192229.3186470-26-eblake@redhat.com>
[eblake: fix logic to reject unnegotiated contexts]
Signed-off-by: Eric Blake <eblake@redhat.com>
The next commit will add support for the optional extension
NBD_CMD_FLAG_PAYLOAD during NBD_CMD_BLOCK_STATUS, where the client can
request that the server only return a subset of negotiated contexts,
rather than all contexts. To make that task easier, this patch
populates the list of contexts to return on a per-command basis (for
now, identical to the full set of negotiated contexts).
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-ID: <20230925192229.3186470-25-eblake@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Peform several minor refactorings of how the list of negotiated meta
contexts is managed, to make upcoming patches easier: Promote the
internal type NBDExportMetaContexts to the public opaque type
NBDMetaContexts, and mark exp const. Use a shorter member name in
NBDClient. Hoist calls to nbd_check_meta_context() earlier in their
callers, as the number of negotiated contexts may impact the flags
exposed in regards to an export, which in turn requires a new
parameter. Drop a redundant parameter to nbd_negotiate_meta_queries.
No semantic change intended on the success path; on the failure path,
dropping context in nbd_check_meta_export even when reporting an error
is safer.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Message-ID: <20230925192229.3186470-24-eblake@redhat.com>
All the pieces are in place for a client to finally request extended
headers. Note that we must not request extended headers when qemu-nbd
is used to connect to the kernel module (as nbd.ko does not expect
them, but expects us to do the negotiation in userspace before handing
the socket over to the kernel), but there is no harm in all other
clients requesting them.
Extended headers are not essential to the information collected during
'qemu-nbd --list', but probing for it gives us one more piece of
information in that output. Update the iotests affected by the new
line of output.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Message-ID: <20230925192229.3186470-23-eblake@redhat.com>
Once extended mode is enabled, we need to accept 64-bit status replies
(even for replies that don't exceed a 32-bit length). It is easier to
normalize narrow replies into wide format so that the rest of our code
only has to handle one width. Although a server is non-compliant if
it sends a 64-bit reply in compact mode, or a 32-bit reply in extended
mode, it is still easy enough to tolerate these mismatches.
In normal execution, we are only requesting "base:allocation" which
never exceeds 32 bits for flag values. But during testing with
x-dirty-bitmap, we can force qemu to connect to some other context
that might have 64-bit status bit; however, we ignore those upper bits
(other than mapping qemu:allocation-depth into something that
'qemu-img map --output=json' can expose), and since that only affects
testing, we really don't bother with checking whether more than the
two least-significant bits are set.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Message-ID: <20230925192229.3186470-22-eblake@redhat.com>
Update the client code to be able to send an extended request, and
parse an extended header from the server. Note that since we reject
any structured reply with a too-large payload, we can always normalize
a valid header back into the compact form, so that the caller need not
deal with two branches of a union. Still, until a later patch lets
the client negotiate extended headers, the code added here should not
be reached. Note that because of the different magic numbers, it is
just as easy to trace and then tolerate a non-compliant server sending
the wrong header reply as it would be to insist that the server is
compliant.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-ID: <20230925192229.3186470-21-eblake@redhat.com>
[eblake: fix trace format]
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Instead of ignoring the low-level error just to refabricate our own
message to pass to the caller, we can just plumb the caller's errp
down to the low level.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-ID: <20230925192229.3186470-20-eblake@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Time to start supporting clients that request extended headers. Now
we can finally reach the code added across several previous patches.
Even though the NBD spec has been altered to allow us to accept
NBD_CMD_READ larger than the max payload size (provided our response
is a hole or broken up over more than one data chunk), we are not
planning to take advantage of that, and continue to cap NBD_CMD_READ
to 32M regardless of header size.
For NBD_CMD_WRITE_ZEROES and NBD_CMD_TRIM, the block layer already
supports 64-bit operations without any effort on our part. For
NBD_CMD_BLOCK_STATUS, the client's length is a hint, and the previous
patch took care of implementing the required
NBD_REPLY_TYPE_BLOCK_STATUS_EXT.
We do not yet support clients that want to do request payload
filtering of NBD_CMD_BLOCK_STATUS; that will be added in later
patches, but is not essential for qemu as a client since qemu only
requests the single context base:allocation.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Message-ID: <20230925192229.3186470-19-eblake@redhat.com>
The NBD spec states that if the client negotiates extended headers,
the server must avoid NBD_REPLY_TYPE_BLOCK_STATUS and instead use
NBD_REPLY_TYPE_BLOCK_STATUS_EXT which supports 64-bit lengths, even if
the reply does not need more than 32 bits. As of this patch,
client->mode is still never NBD_MODE_EXTENDED, so the code added here
does not take effect until the next patch enables negotiation.
For now, all metacontexts that we know how to export never populate
more than 32 bits of information, so we don't have to worry about
NBD_REP_ERR_EXT_HEADER_REQD or filtering during handshake, and we
always send all zeroes for the upper 32 bits of status during
NBD_CMD_BLOCK_STATUS.
Note that we previously had some interesting size-juggling on call
chains, such as:
nbd_co_send_block_status(uint32_t length)
-> blockstatus_to_extents(uint32_t bytes)
-> bdrv_block_status_above(bytes, &uint64_t num)
-> nbd_extent_array_add(uint64_t num)
-> store num in 32-bit length
But we were lucky that it never overflowed: bdrv_block_status_above
never sets num larger than bytes, and we had previously been capping
'bytes' at 32 bits (since the protocol does not allow sending a larger
request without extended headers). This patch adds some assertions
that ensure we continue to avoid overflowing 32 bits for a narrow
client, while fully utilizing 64-bits all the way through when the
client understands that. Even in 64-bit math, overflow is not an
issue, because all lengths are coming from the block layer, and we
know that the block layer does not support images larger than off_t
(if lengths were coming from the network, the story would be
different).
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Message-ID: <20230925192229.3186470-18-eblake@redhat.com>
Although extended mode is not yet enabled, once we do turn it on, we
need to reply with extended headers to all messages. Update the low
level entry points necessary so that all other callers automatically
get the right header based on the current mode.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Message-ID: <20230925192229.3186470-17-eblake@redhat.com>
Although extended mode is not yet enabled, once we do turn it on, we
need to accept extended requests for all messages. Previous patches
have already taken care of supporting 64-bit lengths, now we just need
to read it off the wire.
Note that this implementation will block indefinitely on a buggy
client that sends a non-extended payload (that is, we try to read a
full packet before we ever check the magic number, but a client that
mistakenly sends a simple request after negotiating extended headers
doesn't send us enough bytes), but it's no different from any other
client that stops talking to us partway through a packet and thus not
worth coding around.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Message-ID: <20230925192229.3186470-16-eblake@redhat.com>
Upcoming additions to support NBD 64-bit effect lengths allow for the
possibility to distinguish between payload length (capped at 32M) and
effect length (64 bits, although we generally assume 63 bits because
of off_t limitations). Without that extension, only the NBD_CMD_WRITE
request has a payload; but with the extension, it makes sense to allow
at least NBD_CMD_BLOCK_STATUS to have both a payload and effect length
in a future patch (where the payload is a limited-size struct that in
turn gives the real effect length as well as a subset of known ids for
which status is requested). Other future NBD commands may also have a
request payload, so the 64-bit extension introduces a new
NBD_CMD_FLAG_PAYLOAD_LEN that distinguishes between whether the header
length is a payload length or an effect length, rather than
hard-coding the decision based on the command.
According to the spec, a client should never send a command with a
payload without the negotiation phase proving such extension is
available. So in the unlikely event the bit is set or cleared
incorrectly, the client is already at fault; if the client then
provides the payload, we can gracefully consume it off the wire and
fail the command with NBD_EINVAL (subsequent checks for magic numbers
ensure we are still in sync), while if the client fails to send
payload we block waiting for it (basically deadlocking our connection
to the bad client, but not negatively impacting our ability to service
other clients, so not a security risk). Note that we do not support
the payload version of BLOCK_STATUS yet.
This patch also fixes a latent bug introduced in b2578459: once
request->len can be 64 bits, assigning it to a 32-bit payload_len can
cause wraparound to 0 which then sets req->complete prematurely;
thankfully, the bug was not possible back then (it takes this and
later patches to even allow request->len larger than 32 bits; and
since previously the only 'payload_len = request->len' assignment was
in NBD_CMD_WRITE which also sets check_length, which in turn rejects
lengths larger than 32M before relying on any possibly-truncated value
stored in payload_len).
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-ID: <20230925192229.3186470-15-eblake@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
[eblake: enhance comment on handling client error, fix type bug]
Signed-off-by: Eric Blake <eblake@redhat.com>
This fixes authorship of commits 5cbd51a5 and friends, where the
qemu-ppc mailing list rewrote the "From:" field in the corresponding
patches. See commit 3bd2608db7 ("maint: Add .mailmap entries for
patches claiming list authorship") for explanation.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-ID: <20230927143815.3397386-8-eblake@redhat.com>
Documenting that we should not add new lines to work around SPF
rewrites sounds foreboding; the intent is instead that new lines here
are okay, but indicate a second problem elsewhere in our build process
that we should also consider fixing at the same time, to keep the
section from growing without bounds. While we have been doing that
for qemu-devel for a while, we jut recently fixed that for qemu-block:
https://git.linaro.org/people/pmaydell/misc-scripts.git/commit/?id=f9a317392
Mentioning DMARC alongside SPF may also help people grep for this
scenario, as well as documenting the 'git config' workaround that can
be used by submitters to avoid the munging issue in the first place.
Note the subtlety: 'git commit' sets authorship information based on
user.name and user.email (where name is usually unquoted); while 'git
send-email' includes a body 'From:' line only when sendemail.from is
present but differs from authorship information. Hence the use of
quotes in sendemail.from (not a semantic change to email, but enough
of a difference to add the body 'From:').
Fixes: 3bd2608d ("maint: Add .mailmap entries for patches claiming list authorship")
CC: Andrey Drobyshev <andrey.drobyshev@virtuozzo.com>
Cc: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-ID: <20230927143815.3397386-7-eblake@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
This fixes authorship of commits 2848289168, 52b10c9c0c as the mailing
list rewrote the "From:" field in the corresponding patches. See commit
3bd2608db7 ("maint: Add .mailmap entries for patches claiming list
authorship") for explanation.
Signed-off-by: Andrey Drobyshev <andrey.drobyshev@virtuozzo.com>
Message-ID: <20230926102801.512107-1-andrey.drobyshev@virtuozzo.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Eric Blake <eblake@redhat.com>
vdpa:
shadow vq vlan support
net migration with cvq
cxl:
support emulating 4 HDM decoders
serial number extended capability
virtio:
hared dma-buf
Fixes, cleanups all over the place.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-----BEGIN PGP SIGNATURE-----
iQFDBAABCAAtFiEEXQn9CHHI+FuUyooNKB8NuNKNVGkFAmUd4/YPHG1zdEByZWRo
YXQuY29tAAoJECgfDbjSjVRpyM8H/02cRbJcQOjYt7j68zPW6GaDXxBI/UmdWDyG
15LZZbGNOPjyjNd3Vz1M7stQ5rhoKcgo/RdI+0E60a78svgW5JvpXoXR3pksc3Dx
v28B/akXwHUErYFSZQ+2VHNc8OhCd0v2ehxZxbwPEAYIOAj3hcCIVoPGXTnKJmAJ
imr5hjH0wZUc0+xdsmn8Vfdv5NTzpwfVObbGiMZejeJsaoh0y6Rt8RANBMY67KQD
S7/HPlVuDYf/y43t4ZEHNYuV9RaCdZZYlLWwV1scdKaYcofgmtJOKbOdCjHRXgj+
004Afb3rggIoCfnCzOFzhGx+MLDtLjvEn2N4oLEWCLi+k/3huaA=
=GAvH
-----END PGP SIGNATURE-----
Merge tag 'for_upstream' of https://git.kernel.org/pub/scm/virt/kvm/mst/qemu into staging
virtio,pci: features, cleanups
vdpa:
shadow vq vlan support
net migration with cvq
cxl:
support emulating 4 HDM decoders
serial number extended capability
virtio:
hared dma-buf
Fixes, cleanups all over the place.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
* tag 'for_upstream' of https://git.kernel.org/pub/scm/virt/kvm/mst/qemu: (53 commits)
libvhost-user: handle shared_object msg
vhost-user: add shared_object msg
hw/display: introduce virtio-dmabuf
util/uuid: add a hash function
virtio: remove unused next argument from virtqueue_split_read_next_desc()
virtio: remove unnecessary thread fence while reading next descriptor
virtio: use shadow_avail_idx while checking number of heads
libvhost-user.c: add assertion to vu_message_read_default
pcie_sriov: unregister_vfs(): fix error path
hw/i386/pc: improve physical address space bound check for 32-bit x86 systems
amd_iommu: Fix APIC address check
vdpa net: follow VirtIO initialization properly at cvq isolation probing
vdpa net: stop probing if cannot set features
vdpa net: fix error message setting virtio status
hw/pci-bridge/cxl-upstream: Add serial number extended capability support
hw/cxl: Support 4 HDM decoders at all levels of topology
hw/cxl: Fix and use same calculation for HDM decoder block size everywhere
hw/cxl: Add utility functions decoder interleave ways and target count.
hw/cxl: Push cxl_decoder_count_enc() and cxl_decode_ig() into .c
vdpa net: zero vhost_vdpa iova_tree pointer at cleanup
...
Conflicts:
hw/core/machine.c
Context conflict with commit 314e0a84cd ("hw/core: remove needless
includes") because it removed an adjacent #include.
In the libvhost-user library we need to
handle VHOST_USER_GET_SHARED_OBJECT requests,
and add helper functions to allow sending messages
to interact with the virtio shared objects
hash table.
Signed-off-by: Albert Esteve <aesteve@redhat.com>
Message-Id: <20231002065706.94707-5-aesteve@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Add three new vhost-user protocol
`VHOST_USER_BACKEND_SHARED_OBJECT_* messages`.
These new messages are sent from vhost-user
back-ends to interact with the virtio-dmabuf
table in order to add or remove themselves as
virtio exporters, or lookup for virtio dma-buf
shared objects.
The action taken in the front-end depends
on the type stored in the virtio shared
object hash table.
When the table holds a pointer to a vhost
backend for a given UUID, the front-end sends
a VHOST_USER_GET_SHARED_OBJECT to the
backend holding the shared object.
The messages can only be sent after successfully
negotiating a new VHOST_USER_PROTOCOL_F_SHARED_OBJECT
vhost-user protocol feature bit.
Finally, refactor code to send response message so
that all common parts both for the common REPLY_ACK
case, and other data responses, can call it and
avoid code repetition.
Signed-off-by: Albert Esteve <aesteve@redhat.com>
Message-Id: <20231002065706.94707-4-aesteve@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This API manages objects (in this iteration,
dmabuf fds) that can be shared along different
virtio devices, associated to a UUID.
The API allows the different devices to add,
remove and/or retrieve the objects by simply
invoking the public functions that reside in the
virtio-dmabuf file.
For vhost backends, the API stores the pointer
to the backend holding the object.
Suggested-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Albert Esteve <aesteve@redhat.com>
Message-Id: <20231002065706.94707-3-aesteve@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Add hash function to uuid module using the
djb2 hash algorithm.
Add a couple simple unit tests for the hash
function, checking collisions for similar UUIDs.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Albert Esteve <aesteve@redhat.com>
Message-Id: <20231002065706.94707-2-aesteve@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The 'next' was converted from a local variable to an output parameter
in commit:
412e0e81b1 ("virtio: handle virtqueue_read_next_desc() errors")
But all the actual uses of the 'i/next' as an output were removed a few
months prior in commit:
aa570d6fb6 ("virtio: combine the read of a descriptor")
Remove the unused argument to simplify the code.
Also, adding a comment to the function to describe what it is actually
doing, as it is not obvious that the 'desc' is both an input and an
output argument.
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Message-Id: <20230927140016.2317404-3-i.maximets@ovn.org>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
It was supposed to be a compiler barrier and it was a compiler barrier
initially called 'wmb' when virtio core support was introduced.
Later all the instances of 'wmb' were switched to smp_wmb to fix memory
ordering issues on non-x86 platforms. However, this one doesn't need
to be an actual barrier, as its only purpose was to ensure that the
value is not read twice.
And since commit aa570d6fb6 ("virtio: combine the read of a descriptor")
there is no need for a barrier at all, since we're no longer reading
guest memory here, but accessing a local structure.
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Message-Id: <20230927140016.2317404-2-i.maximets@ovn.org>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
We do not need the most up to date number of heads, we only want to
know if there is at least one.
Use shadow variable as long as it is not equal to the last available
index checked. This avoids expensive qatomic dereference of the
RCU-protected memory region cache as well as the memory access itself.
The change improves performance of the af-xdp network backend by 2-3%.
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Message-Id: <20230927135157.2316982-1-i.maximets@ovn.org>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Explain Coverity that we are not going to overflow vmsg->fds.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Message-Id: <20230925194040.68592-5-vsementsov@yandex-team.ru>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
local_err must be NULL before calling object_property_set_bool(), so we
must clear it on each iteration. Let's also use more convenient
error_reportf_err().
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Message-Id: <20230925194040.68592-8-vsementsov@yandex-team.ru>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
32-bit x86 systems do not have a reserved memory for hole64. On those 32-bit
systems without PSE36 or PAE CPU features, hotplugging memory devices are not
supported by QEMU as QEMU always places hotplugged memory above 4 GiB boundary
which is beyond the physical address space of the processor. Linux guests also
does not support memory hotplug on those systems. Please see Linux
kernel commit b59d02ed08690 ("mm/memory_hotplug: disable the functionality
for 32b") for more details.
Therefore, the maximum limit of the guest physical address in the absence of
additional memory devices effectively coincides with the end of
"above 4G memory space" region for 32-bit x86 without PAE/PSE36. When users
configure additional memory devices, after properly accounting for the
additional device memory region to find the maximum value of the guest
physical address, the address will be outside the range of the processor's
physical address space.
This change adds improvements to take above into consideration.
For example, previously this was allowed:
$ ./qemu-system-x86_64 -cpu pentium -m size=10G
With this change now it is no longer allowed:
$ ./qemu-system-x86_64 -cpu pentium -m size=10G
qemu-system-x86_64: Address space limit 0xffffffff < 0x2bfffffff phys-bits too low (32)
However, the following are allowed since on both cases physical address
space of the processor is 36 bits:
$ ./qemu-system-x86_64 -cpu pentium2 -m size=10G
$ ./qemu-system-x86_64 -cpu pentium,pse36=on -m size=10G
For 32-bit, without PAE/PSE36, hotplugging additional memory is no longer allowed.
$ ./qemu-system-i386 -m size=1G,maxmem=3G,slots=2
qemu-system-i386: Address space limit 0xffffffff < 0x1ffffffff phys-bits too low (32)
$ ./qemu-system-i386 -machine q35 -m size=1G,maxmem=3G,slots=2
qemu-system-i386: Address space limit 0xffffffff < 0x1ffffffff phys-bits too low (32)
A new compatibility flag is introduced to make sure pc_max_used_gpa() keeps
returning the old value for machines 8.1 and older.
Therefore, the above is still allowed for older machine types in order to support
compatibility. Hence, the following still works:
$ ./qemu-system-i386 -machine pc-i440fx-8.1 -m size=1G,maxmem=3G,slots=2
$ ./qemu-system-i386 -machine pc-q35-8.1 -m size=1G,maxmem=3G,slots=2
Further, following is also allowed as with PSE36, the processor has 36-bit
address space:
$ ./qemu-system-i386 -cpu 486,pse36=on -m size=1G,maxmem=3G,slots=2
After calling CPUID with EAX=0x80000001, all AMD64 compliant processors
have the longmode-capable-bit turned on in the extended feature flags (bit 29)
in EDX. The absence of CPUID longmode can be used to differentiate between
32-bit and 64-bit processors and is the recommended approach. QEMU takes this
approach elsewhere (for example, please see x86_cpu_realizefn()), With
this change, pc_max_used_gpa() also uses the same method to detect 32-bit
processors.
Unit tests are modified to not run 32-bit x86 tests that use memory hotplug.
Suggested-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Ani Sinha <anisinha@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-Id: <20230922160413.165702-1-anisinha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
An MSI from I/O APIC may not exactly equal to APIC_DEFAULT_ADDRESS. In
fact, Windows 17763.3650 configures I/O APIC to set the dest_mode bit.
Cover the range assigned to APIC.
Fixes: 577c470f43 ("x86_iommu/amd: Prepare for interrupt remap support")
Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Message-Id: <20230921114612.40671-1-akihiko.odaki@daynix.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This patch solves a few issues. The most obvious is that the feature
set was done previous to ACKNOWLEDGE | DRIVER status bit set. Current
vdpa devices are permissive with this, but it is better to follow the
standard.
Fixes: 152128d646 ("vdpa: move CVQ isolation check to net_init_vhost_vdpa")
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <20230915170836.3078172-4-eperezma@redhat.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Otherwise it continues the CVQ isolation probing.
Fixes: 152128d646 ("vdpa: move CVQ isolation check to net_init_vhost_vdpa")
Reported-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <20230915170836.3078172-3-eperezma@redhat.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
It incorrectly prints "error setting features", probably because a copy
paste miss.
Fixes: 152128d646 ("vdpa: move CVQ isolation check to net_init_vhost_vdpa")
Reported-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <20230915170836.3078172-2-eperezma@redhat.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Will be needed so there is a defined serial number for
information queries via the Switch CCI.
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Message-Id: <20230913133615.29876-1-Jonathan.Cameron@huawei.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Support these decoders in CXL host bridges (pxb-cxl), CXL Switch USP
and CXL Type 3 end points.
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Message-Id: <20230913132523.29780-5-Jonathan.Cameron@huawei.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
In order to avoid having the size of the per HDM decoder register block
repeated in lots of places, create the register definitions for HDM
decoder 1 and use the offset between the first registers in HDM decoder 0 and
HDM decoder 1 to establish the offset.
Calculate in each function as this is more obvious and leads to shorter
line lengths than a single #define which would need a long name
to be specific enough.
Note that the code currently only supports one decoder, so the bugs this
fixes don't actually affect anything.
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Fan Ni <fan.ni@samsung.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230913132523.29780-4-Jonathan.Cameron@huawei.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
As an encoded version of these key configuration parameters is available
in a register, provide functions to extract it again so as to avoid
the need for duplicating the storage.
Whilst here update the _enc() function to include additional values
as defined in the CXL 3.0 specification. Whilst they are not
currently used in the emulation, they may be in future and it is
easier to compare with the specification if all values are covered.
Add a spec reference for cxl_interleave_ways_enc() for consistency
with the target count equivalent (and because it's nice to know where
the magic numbers come from).
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Fan Ni <fan.ni@samsung.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Message-Id: <20230913132523.29780-3-Jonathan.Cameron@huawei.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
There is no strong justification for keeping these in the header
so push them down into the associated cxl-component-utils.c file.
Suggested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Fan Ni <fan.ni@samsung.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Message-Id: <20230913132523.29780-2-Jonathan.Cameron@huawei.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Not zeroing it causes a SIGSEGV if the live migration is cancelled, at
net device restart.
This is caused because CVQ tries to reuse the iova_tree that is present
in the first vhost_vdpa device at the end of vhost_vdpa_net_cvq_start.
As a consequence, it tries to access an iova_tree that has been already
free.
Fixes: 00ef422e9f ("vdpa net: move iova tree creation from init to start")
Reported-by: Yanhui Ma <yama@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <20230913123408.2819185-1-eperezma@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Si-Wei Liu <si-wei.liu@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
gcc 13.2.1 emits the following warning:
net/vhost-vdpa.c: In function ‘net_vhost_vdpa_init.constprop’:
net/vhost-vdpa.c:1394:25: error: ‘cvq_isolated’ may be used uninitialized [-Werror=maybe-uninitialized]
1394 | s->cvq_isolated = cvq_isolated;
| ~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~
net/vhost-vdpa.c:1355:9: note: ‘cvq_isolated’ was declared here
1355 | int cvq_isolated;
| ^~~~~~~~~~~~
cc1: all warnings being treated as errors
Cc: Eugenio Pérez <eperezma@redhat.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Jason Wang <jasowang@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20230911215435.4156314-1-stefanha@redhat.com>
Acked-by: Eugenio Pérez <eperezma@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The bit positions of both registers are related. Tracing the registers
independently results in the same offsets across these registers which
eases debugging.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230908084234.17642-9-shentey@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230908084234.17642-8-shentey@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The SMI command port is currently hardcoded by means of the ACPI_PORT_SMI_CMD
macro. This hardcoding is Intel specific and doesn't match VIA, for example.
There is already the AcpiFadtData::smi_cmd attribute which is used when building
the FADT. Let's also use it when building the DSDT which confines SMI command
port determination to just one place. This allows it to become a property later,
thus resolving the Intel assumption.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230908084234.17642-7-shentey@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Now that TYPE_ACPI_GED_X86 doesn't assign AcpiDeviceIfClass::madt_cpu any more
it is the same as TYPE_ACPI_GED.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230908084234.17642-6-shentey@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The "hw/boards.h" is unused since the previous commit. Since its removal
requires include fixes in various unrelated files to keep the code compiling it
has been split in a dedicated commit.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230908084234.17642-5-shentey@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This virtual method was always set to the x86-specific pc_madt_cpu_entry(),
even in piix4 which is also used in MIPS. The previous changes use
pc_madt_cpu_entry() otherwise, so madt_cpu can be dropped.
Since pc_madt_cpu_entry() is now only used in x86-specific code, the stub
in hw/acpi/acpi-x86-stub can be removed as well.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230908084234.17642-4-shentey@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
build_cpus_aml() is architecture independent but needs to create architecture-
specific CPU AML. So far this was achieved by using a virtual method from
TYPE_ACPI_DEVICE_IF. However, build_cpus_aml() would resolve this interface from
global (!) state. This makes it quite incomprehensible where this interface
comes from (TYPE_PIIX4_PM?, TYPE_ICH9_LPC_DEVICE?, TYPE_ACPI_GED_X86?) an can
lead to crashes when the generic code is ported to new architectures.
So far, build_cpus_aml() is only called in architecture-specific code -- and
only in x86. We can therefore simply pass pc_madt_cpu_entry() as callback to
build_cpus_aml(). This is the same callback that would be used through
TYPE_ACPI_DEVICE_IF.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230908084234.17642-3-shentey@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This is x86-specific code, so there is no advantage in using
pc_madt_cpu_entry() behind an architecture-agnostic interface.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230908084234.17642-2-shentey@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
