From c7f21816612879efdae3d9b67f024a6671494c62 Mon Sep 17 00:00:00 2001 From: Hanna Czenczek Date: Mon, 16 Oct 2023 15:42:37 +0200 Subject: vhost-user.rst: Improve [GS]ET_VRING_BASE doc GET_VRING_BASE does not mention that it stops the respective ring. Fix that. Furthermore, it is not fully clear what the "base offset" these commands' documentation refers to is; an offset could be many things. Be more precise and verbose about it, especially given that these commands use different payload structures depending on whether the vring is split or packed. Signed-off-by: Hanna Czenczek Message-Id: <20231016134243.68248-2-hreitz@redhat.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- docs/interop/vhost-user.rst | 77 ++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 73 insertions(+), 4 deletions(-) diff --git a/docs/interop/vhost-user.rst b/docs/interop/vhost-user.rst index 768fb5c28c..9202b167dd 100644 --- a/docs/interop/vhost-user.rst +++ b/docs/interop/vhost-user.rst @@ -108,6 +108,43 @@ A vring state description :num: a 32-bit number +A vring descriptor index for split virtqueues +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + ++-------------+---------------------+ +| vring index | index in avail ring | ++-------------+---------------------+ + +:vring index: 32-bit index of the respective virtqueue + +:index in avail ring: 32-bit value, of which currently only the lower 16 + bits are used: + + - Bits 0–15: Index of the next *Available Ring* descriptor that the + back-end will process. This is a free-running index that is not + wrapped by the ring size. + - Bits 16–31: Reserved (set to zero) + +Vring descriptor indices for packed virtqueues +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + ++-------------+--------------------+ +| vring index | descriptor indices | ++-------------+--------------------+ + +:vring index: 32-bit index of the respective virtqueue + +:descriptor indices: 32-bit value: + + - Bits 0–14: Index of the next *Available Ring* descriptor that the + back-end will process. This is a free-running index that is not + wrapped by the ring size. + - Bit 15: Driver (Available) Ring Wrap Counter + - Bits 16–30: Index of the entry in the *Used Ring* where the back-end + will place the next descriptor. This is a free-running index that + is not wrapped by the ring size. + - Bit 31: Device (Used) Ring Wrap Counter + A vring address description ^^^^^^^^^^^^^^^^^^^^^^^^^^^ @@ -1042,18 +1079,50 @@ Front-end message types ``VHOST_USER_SET_VRING_BASE`` :id: 10 :equivalent ioctl: ``VHOST_SET_VRING_BASE`` - :request payload: vring state description + :request payload: vring descriptor index/indices :reply payload: N/A - Sets the base offset in the available vring. + Sets the next index to use for descriptors in this vring: + + * For a split virtqueue, sets only the next descriptor index to + process in the *Available Ring*. The device is supposed to read the + next index in the *Used Ring* from the respective vring structure in + guest memory. + + * For a packed virtqueue, both indices are supplied, as they are not + explicitly available in memory. + + Consequently, the payload type is specific to the type of virt queue + (*a vring descriptor index for split virtqueues* vs. *vring descriptor + indices for packed virtqueues*). ``VHOST_USER_GET_VRING_BASE`` :id: 11 :equivalent ioctl: ``VHOST_USER_GET_VRING_BASE`` :request payload: vring state description - :reply payload: vring state description + :reply payload: vring descriptor index/indices + + Stops the vring and returns the current descriptor index or indices: + + * For a split virtqueue, returns only the 16-bit next descriptor + index to process in the *Available Ring*. Note that this may + differ from the available ring index in the vring structure in + memory, which points to where the driver will put new available + descriptors. For the *Used Ring*, the device only needs the next + descriptor index at which to put new descriptors, which is the + value in the vring structure in memory, so this value is not + covered by this message. + + * For a packed virtqueue, neither index is explicitly available to + read from memory, so both indices (as maintained by the device) are + returned. + + Consequently, the payload type is specific to the type of virt queue + (*a vring descriptor index for split virtqueues* vs. *vring descriptor + indices for packed virtqueues*). - Get the available vring base offset. + The request payload’s *num* field is currently reserved and must be + set to 0. ``VHOST_USER_SET_VRING_KICK`` :id: 12 -- cgit v1.2.3 From eae69cc36b278a3e84bb013909135e138b37855d Mon Sep 17 00:00:00 2001 From: Hanna Czenczek Date: Mon, 16 Oct 2023 15:42:38 +0200 Subject: vhost-user.rst: Clarify enabling/disabling vrings Currently, the vhost-user documentation says that rings are to be initialized in a disabled state when VHOST_USER_F_PROTOCOL_FEATURES is negotiated. However, by the time of feature negotiation, all rings have already been initialized, so it is not entirely clear what this means. At least the vhost-user-backend Rust crate's implementation interpreted it to mean that whenever this feature is negotiated, all rings are to put into a disabled state, which means that every SET_FEATURES call would disable all rings, effectively halting the device. This is problematic because the VHOST_F_LOG_ALL feature is also set or cleared this way, which happens during migration. Doing so should not halt the device. Other implementations have interpreted this to mean that the device is to be initialized with all rings disabled, and a subsequent SET_FEATURES call that does not set VHOST_USER_F_PROTOCOL_FEATURES will enable all of them. Here, SET_FEATURES will never disable any ring. This interpretation does not suffer the problem of unintentionally halting the device whenever features are set or cleared, so it seems better and more reasonable. We can clarify this in the documentation by making it explicit that the enabled/disabled state is tracked even while the vring is stopped. Every vring is initialized in a disabled state, and SET_FEATURES without VHOST_USER_F_PROTOCOL_FEATURES simply becomes one way to enable all vrings. Reviewed-by: Stefan Hajnoczi Signed-off-by: Hanna Czenczek Message-Id: <20231016134243.68248-3-hreitz@redhat.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- docs/interop/vhost-user.rst | 32 +++++++++++++++++--------------- 1 file changed, 17 insertions(+), 15 deletions(-) diff --git a/docs/interop/vhost-user.rst b/docs/interop/vhost-user.rst index 9202b167dd..e5a04c04ed 100644 --- a/docs/interop/vhost-user.rst +++ b/docs/interop/vhost-user.rst @@ -411,31 +411,33 @@ negotiation. Ring states ----------- -Rings can be in one of three states: +Rings have two independent states: started/stopped, and enabled/disabled. -* stopped: the back-end must not process the ring at all. +* While a ring is stopped, the back-end must not process the ring at + all, regardless of whether it is enabled or disabled. The + enabled/disabled state should still be tracked, though, so it can come + into effect once the ring is started. -* started but disabled: the back-end must process the ring without +* started and disabled: The back-end must process the ring without causing any side effects. For example, for a networking device, in the disabled state the back-end must not supply any new RX packets, but must process and discard any TX packets. -* started and enabled. +* started and enabled: The back-end must process the ring normally, i.e. + process all requests and execute them. -Each ring is initialized in a stopped state. The back-end must start -ring upon receiving a kick (that is, detecting that file descriptor is -readable) on the descriptor specified by ``VHOST_USER_SET_VRING_KICK`` -or receiving the in-band message ``VHOST_USER_VRING_KICK`` if negotiated, -and stop ring upon receiving ``VHOST_USER_GET_VRING_BASE``. +Each ring is initialized in a stopped and disabled state. The back-end +must start a ring upon receiving a kick (that is, detecting that file +descriptor is readable) on the descriptor specified by +``VHOST_USER_SET_VRING_KICK`` or receiving the in-band message +``VHOST_USER_VRING_KICK`` if negotiated, and stop a ring upon receiving +``VHOST_USER_GET_VRING_BASE``. Rings can be enabled or disabled by ``VHOST_USER_SET_VRING_ENABLE``. -If ``VHOST_USER_F_PROTOCOL_FEATURES`` has not been negotiated, the -ring starts directly in the enabled state. - -If ``VHOST_USER_F_PROTOCOL_FEATURES`` has been negotiated, the ring is -initialized in a disabled state and is enabled by -``VHOST_USER_SET_VRING_ENABLE`` with parameter 1. +In addition, upon receiving a ``VHOST_USER_SET_FEATURES`` message from +the front-end without ``VHOST_USER_F_PROTOCOL_FEATURES`` set, the +back-end must enable all rings immediately. While processing the rings (whether they are enabled or not), the back-end must support changing some configuration aspects on the fly. -- cgit v1.2.3 From a6e76dd3c3b98db86802b4b08984f2ac5ec1faea Mon Sep 17 00:00:00 2001 From: Hanna Czenczek Date: Mon, 16 Oct 2023 15:42:39 +0200 Subject: vhost-user.rst: Introduce suspended state In vDPA, GET_VRING_BASE does not stop the queried vring, which is why SUSPEND was introduced so that the returned index would be stable. In vhost-user, it does stop the vring, so under the same reasoning, it can get away without SUSPEND. Still, we do want to clarify that if the device is completely stopped, i.e. all vrings are stopped, the back-end should cease to modify any state relating to the guest. Do this by calling it "suspended". Suggested-by: Stefan Hajnoczi Reviewed-by: Stefan Hajnoczi Signed-off-by: Hanna Czenczek Message-Id: <20231016134243.68248-4-hreitz@redhat.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- docs/interop/vhost-user.rst | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) diff --git a/docs/interop/vhost-user.rst b/docs/interop/vhost-user.rst index e5a04c04ed..035a23ed35 100644 --- a/docs/interop/vhost-user.rst +++ b/docs/interop/vhost-user.rst @@ -442,6 +442,19 @@ back-end must enable all rings immediately. While processing the rings (whether they are enabled or not), the back-end must support changing some configuration aspects on the fly. +.. _suspended_device_state: + +Suspended device state +^^^^^^^^^^^^^^^^^^^^^^ + +While all vrings are stopped, the device is *suspended*. In addition to +not processing any vring (because they are stopped), the device must: + +* not write to any guest memory regions, +* not send any notifications to the guest, +* not send any messages to the front-end, +* still process and reply to messages from the front-end. + Multiple queue support ---------------------- @@ -529,7 +542,8 @@ ancillary data, it may be used to inform the front-end that the log has been modified. Once the source has finished migration, rings will be stopped by the -source. No further update must be done before rings are restarted. +source (:ref:`Suspended device state `). No +further update must be done before rings are restarted. In postcopy migration the back-end is started before all the memory has been received from the source host, and care must be taken to avoid @@ -1123,6 +1137,10 @@ Front-end message types (*a vring descriptor index for split virtqueues* vs. *vring descriptor indices for packed virtqueues*). + When and as long as all of a device’s vrings are stopped, it is + *suspended*, see :ref:`Suspended device state + `. + The request payload’s *num* field is currently reserved and must be set to 0. -- cgit v1.2.3 From 019233096c03b826e0e677115b6e3c550a54a48d Mon Sep 17 00:00:00 2001 From: Hanna Czenczek Date: Mon, 16 Oct 2023 15:42:40 +0200 Subject: vhost-user.rst: Migrating back-end-internal state For vhost-user devices, qemu can migrate the virtio state, but not the back-end's internal state. To do so, we need to be able to transfer this internal state between front-end (qemu) and back-end. At this point, this new feature is added for the purpose of virtio-fs migration. Because virtiofsd's internal state will not be too large, we believe it is best to transfer it as a single binary blob after the streaming phase. These are the additions to the protocol: - New vhost-user protocol feature VHOST_USER_PROTOCOL_F_DEVICE_STATE - SET_DEVICE_STATE_FD function: Front-end and back-end negotiate a file descriptor over which to transfer the state. - CHECK_DEVICE_STATE: After the state has been transferred through the file descriptor, the front-end invokes this function to verify success. There is no in-band way (through the file descriptor) to indicate failure, so we need to check explicitly. Once the transfer FD has been established via SET_DEVICE_STATE_FD (which includes establishing the direction of transfer and migration phase), the sending side writes its data into it, and the reading side reads it until it sees an EOF. Then, the front-end will check for success via CHECK_DEVICE_STATE, which on the destination side includes checking for integrity (i.e. errors during deserialization). Reviewed-by: Stefan Hajnoczi Signed-off-by: Hanna Czenczek Message-Id: <20231016134243.68248-5-hreitz@redhat.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- docs/interop/vhost-user.rst | 172 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 172 insertions(+) diff --git a/docs/interop/vhost-user.rst b/docs/interop/vhost-user.rst index 035a23ed35..9f1103f85a 100644 --- a/docs/interop/vhost-user.rst +++ b/docs/interop/vhost-user.rst @@ -322,6 +322,32 @@ VhostUserShared :UUID: 16 bytes UUID, whose first three components (a 32-bit value, then two 16-bit values) are stored in big endian. +Device state transfer parameters +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + ++--------------------+-----------------+ +| transfer direction | migration phase | ++--------------------+-----------------+ + +:transfer direction: a 32-bit enum, describing the direction in which + the state is transferred: + + - 0: Save: Transfer the state from the back-end to the front-end, + which happens on the source side of migration + - 1: Load: Transfer the state from the front-end to the back-end, + which happens on the destination side of migration + +:migration phase: a 32-bit enum, describing the state in which the VM + guest and devices are: + + - 0: Stopped (in the period after the transfer of memory-mapped + regions before switch-over to the destination): The VM guest is + stopped, and the vhost-user device is suspended (see + :ref:`Suspended device state `). + + In the future, additional phases might be added e.g. to allow + iterative migration while the device is running. + C structure ----------- @@ -381,6 +407,7 @@ in the ancillary data: * ``VHOST_USER_SET_VRING_ERR`` * ``VHOST_USER_SET_BACKEND_REQ_FD`` (previous name ``VHOST_USER_SET_SLAVE_REQ_FD``) * ``VHOST_USER_SET_INFLIGHT_FD`` (if ``VHOST_USER_PROTOCOL_F_INFLIGHT_SHMFD``) +* ``VHOST_USER_SET_DEVICE_STATE_FD`` If *front-end* is unable to send the full message or receives a wrong reply it will close the connection. An optional reconnection mechanism @@ -555,6 +582,80 @@ it performs WAKE ioctl's on the userfaultfd to wake the stalled back-end. The front-end indicates support for this via the ``VHOST_USER_PROTOCOL_F_PAGEFAULT`` feature. +.. _migrating_backend_state: + +Migrating back-end state +^^^^^^^^^^^^^^^^^^^^^^^^ + +Migrating device state involves transferring the state from one +back-end, called the source, to another back-end, called the +destination. After migration, the destination transparently resumes +operation without requiring the driver to re-initialize the device at +the VIRTIO level. If the migration fails, then the source can +transparently resume operation until another migration attempt is made. + +Generally, the front-end is connected to a virtual machine guest (which +contains the driver), which has its own state to transfer between source +and destination, and therefore will have an implementation-specific +mechanism to do so. The ``VHOST_USER_PROTOCOL_F_DEVICE_STATE`` feature +provides functionality to have the front-end include the back-end's +state in this transfer operation so the back-end does not need to +implement its own mechanism, and so the virtual machine may have its +complete state, including vhost-user devices' states, contained within a +single stream of data. + +To do this, the back-end state is transferred from back-end to front-end +on the source side, and vice versa on the destination side. This +transfer happens over a channel that is negotiated using the +``VHOST_USER_SET_DEVICE_STATE_FD`` message. This message has two +parameters: + +* Direction of transfer: On the source, the data is saved, transferring + it from the back-end to the front-end. On the destination, the data + is loaded, transferring it from the front-end to the back-end. + +* Migration phase: Currently, the only supported phase is the period + after the transfer of memory-mapped regions before switch-over to the + destination, when both the source and destination devices are + suspended (:ref:`Suspended device state `). + In the future, additional phases might be supported to allow iterative + migration while the device is running. + +The nature of the channel is implementation-defined, but it must +generally behave like a pipe: The writing end will write all the data it +has into it, signalling the end of data by closing its end. The reading +end must read all of this data (until encountering the end of file) and +process it. + +* When saving, the writing end is the source back-end, and the reading + end is the source front-end. After reading the state data from the + channel, the source front-end must transfer it to the destination + front-end through an implementation-defined mechanism. + +* When loading, the writing end is the destination front-end, and the + reading end is the destination back-end. After reading the state data + from the channel, the destination back-end must deserialize its + internal state from that data and set itself up to allow the driver to + seamlessly resume operation on the VIRTIO level. + +Seamlessly resuming operation means that the migration must be +transparent to the guest driver, which operates on the VIRTIO level. +This driver will not perform any re-initialization steps, but continue +to use the device as if no migration had occurred. The vhost-user +front-end, however, will re-initialize the vhost state on the +destination, following the usual protocol for establishing a connection +to a vhost-user back-end: This includes, for example, setting up memory +mappings and kick and call FDs as necessary, negotiating protocol +features, or setting the initial vring base indices (to the same value +as on the source side, so that operation can resume). + +Both on the source and on the destination side, after the respective +front-end has seen all data transferred (when the transfer FD has been +closed), it sends the ``VHOST_USER_CHECK_DEVICE_STATE`` message to +verify that data transfer was successful in the back-end, too. The +back-end responds once it knows whether the transfer and processing was +successful or not. + Memory access ------------- @@ -949,6 +1050,7 @@ Protocol features #define VHOST_USER_PROTOCOL_F_STATUS 16 #define VHOST_USER_PROTOCOL_F_XEN_MMAP 17 #define VHOST_USER_PROTOCOL_F_SHARED_OBJECT 18 + #define VHOST_USER_PROTOCOL_F_DEVICE_STATE 19 Front-end message types ----------------------- @@ -1553,6 +1655,76 @@ Front-end message types the requested UUID. Back-end will reply passing the fd when the operation is successful, or no fd otherwise. +``VHOST_USER_SET_DEVICE_STATE_FD`` + :id: 42 + :equivalent ioctl: N/A + :request payload: device state transfer parameters + :reply payload: ``u64`` + + Front-end and back-end negotiate a channel over which to transfer the + back-end’s internal state during migration. Either side (front-end or + back-end) may create the channel. The nature of this channel is not + restricted or defined in this document, but whichever side creates it + must create a file descriptor that is provided to the respectively + other side, allowing access to the channel. This FD must behave as + follows: + + * For the writing end, it must allow writing the whole back-end state + sequentially. Closing the file descriptor signals the end of + transfer. + + * For the reading end, it must allow reading the whole back-end state + sequentially. The end of file signals the end of the transfer. + + For example, the channel may be a pipe, in which case the two ends of + the pipe fulfill these requirements respectively. + + Initially, the front-end creates a channel along with such an FD. It + passes the FD to the back-end as ancillary data of a + ``VHOST_USER_SET_DEVICE_STATE_FD`` message. The back-end may create a + different transfer channel, passing the respective FD back to the + front-end as ancillary data of the reply. If so, the front-end must + then discard its channel and use the one provided by the back-end. + + Whether the back-end should decide to use its own channel is decided + based on efficiency: If the channel is a pipe, both ends will most + likely need to copy data into and out of it. Any channel that allows + for more efficient processing on at least one end, e.g. through + zero-copy, is considered more efficient and thus preferred. If the + back-end can provide such a channel, it should decide to use it. + + The request payload contains parameters for the subsequent data + transfer, as described in the :ref:`Migrating back-end state + ` section. + + The value returned is both an indication for success, and whether a + file descriptor for a back-end-provided channel is returned: Bits 0–7 + are 0 on success, and non-zero on error. Bit 8 is the invalid FD + flag; this flag is set when there is no file descriptor returned. + When this flag is not set, the front-end must use the returned file + descriptor as its end of the transfer channel. The back-end must not + both indicate an error and return a file descriptor. + + Using this function requires prior negotiation of the + ``VHOST_USER_PROTOCOL_F_DEVICE_STATE`` feature. + +``VHOST_USER_CHECK_DEVICE_STATE`` + :id: 43 + :equivalent ioctl: N/A + :request payload: N/A + :reply payload: ``u64`` + + After transferring the back-end’s internal state during migration (see + the :ref:`Migrating back-end state ` + section), check whether the back-end was able to successfully fully + process the state. + + The value returned indicates success or error; 0 is success, any + non-zero value is an error. + + Using this function requires prior negotiation of the + ``VHOST_USER_PROTOCOL_F_DEVICE_STATE`` feature. + Back-end message types ---------------------- -- cgit v1.2.3 From cda83adc62b6108afc8a82d0f54d9a9a861e7aa8 Mon Sep 17 00:00:00 2001 From: Hanna Czenczek Date: Mon, 16 Oct 2023 15:42:41 +0200 Subject: vhost-user: Interface for migration state transfer Add the interface for transferring the back-end's state during migration as defined previously in vhost-user.rst. Reviewed-by: Stefan Hajnoczi Signed-off-by: Hanna Czenczek Message-Id: <20231016134243.68248-6-hreitz@redhat.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/virtio/vhost-user.c | 146 ++++++++++++++++++++++++++++++++++++++ hw/virtio/vhost.c | 37 ++++++++++ include/hw/virtio/vhost-backend.h | 24 +++++++ include/hw/virtio/vhost-user.h | 1 + include/hw/virtio/vhost.h | 78 ++++++++++++++++++++ 5 files changed, 286 insertions(+) diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c index 7b42ae8aae..f214df804b 100644 --- a/hw/virtio/vhost-user.c +++ b/hw/virtio/vhost-user.c @@ -103,6 +103,8 @@ typedef enum VhostUserRequest { VHOST_USER_SET_STATUS = 39, VHOST_USER_GET_STATUS = 40, VHOST_USER_GET_SHARED_OBJECT = 41, + VHOST_USER_SET_DEVICE_STATE_FD = 42, + VHOST_USER_CHECK_DEVICE_STATE = 43, VHOST_USER_MAX } VhostUserRequest; @@ -201,6 +203,12 @@ typedef struct { uint32_t size; /* the following payload size */ } QEMU_PACKED VhostUserHeader; +/* Request payload of VHOST_USER_SET_DEVICE_STATE_FD */ +typedef struct VhostUserTransferDeviceState { + uint32_t direction; + uint32_t phase; +} VhostUserTransferDeviceState; + typedef union { #define VHOST_USER_VRING_IDX_MASK (0xff) #define VHOST_USER_VRING_NOFD_MASK (0x1 << 8) @@ -216,6 +224,7 @@ typedef union { VhostUserVringArea area; VhostUserInflight inflight; VhostUserShared object; + VhostUserTransferDeviceState transfer_state; } VhostUserPayload; typedef struct VhostUserMsg { @@ -2855,6 +2864,140 @@ static void vhost_user_reset_status(struct vhost_dev *dev) } } +static bool vhost_user_supports_device_state(struct vhost_dev *dev) +{ + return virtio_has_feature(dev->protocol_features, + VHOST_USER_PROTOCOL_F_DEVICE_STATE); +} + +static int vhost_user_set_device_state_fd(struct vhost_dev *dev, + VhostDeviceStateDirection direction, + VhostDeviceStatePhase phase, + int fd, + int *reply_fd, + Error **errp) +{ + int ret; + struct vhost_user *vu = dev->opaque; + VhostUserMsg msg = { + .hdr = { + .request = VHOST_USER_SET_DEVICE_STATE_FD, + .flags = VHOST_USER_VERSION, + .size = sizeof(msg.payload.transfer_state), + }, + .payload.transfer_state = { + .direction = direction, + .phase = phase, + }, + }; + + *reply_fd = -1; + + if (!vhost_user_supports_device_state(dev)) { + close(fd); + error_setg(errp, "Back-end does not support migration state transfer"); + return -ENOTSUP; + } + + ret = vhost_user_write(dev, &msg, &fd, 1); + close(fd); + if (ret < 0) { + error_setg_errno(errp, -ret, + "Failed to send SET_DEVICE_STATE_FD message"); + return ret; + } + + ret = vhost_user_read(dev, &msg); + if (ret < 0) { + error_setg_errno(errp, -ret, + "Failed to receive SET_DEVICE_STATE_FD reply"); + return ret; + } + + if (msg.hdr.request != VHOST_USER_SET_DEVICE_STATE_FD) { + error_setg(errp, + "Received unexpected message type, expected %d, received %d", + VHOST_USER_SET_DEVICE_STATE_FD, msg.hdr.request); + return -EPROTO; + } + + if (msg.hdr.size != sizeof(msg.payload.u64)) { + error_setg(errp, + "Received bad message size, expected %zu, received %" PRIu32, + sizeof(msg.payload.u64), msg.hdr.size); + return -EPROTO; + } + + if ((msg.payload.u64 & 0xff) != 0) { + error_setg(errp, "Back-end did not accept migration state transfer"); + return -EIO; + } + + if (!(msg.payload.u64 & VHOST_USER_VRING_NOFD_MASK)) { + *reply_fd = qemu_chr_fe_get_msgfd(vu->user->chr); + if (*reply_fd < 0) { + error_setg(errp, + "Failed to get back-end-provided transfer pipe FD"); + *reply_fd = -1; + return -EIO; + } + } + + return 0; +} + +static int vhost_user_check_device_state(struct vhost_dev *dev, Error **errp) +{ + int ret; + VhostUserMsg msg = { + .hdr = { + .request = VHOST_USER_CHECK_DEVICE_STATE, + .flags = VHOST_USER_VERSION, + .size = 0, + }, + }; + + if (!vhost_user_supports_device_state(dev)) { + error_setg(errp, "Back-end does not support migration state transfer"); + return -ENOTSUP; + } + + ret = vhost_user_write(dev, &msg, NULL, 0); + if (ret < 0) { + error_setg_errno(errp, -ret, + "Failed to send CHECK_DEVICE_STATE message"); + return ret; + } + + ret = vhost_user_read(dev, &msg); + if (ret < 0) { + error_setg_errno(errp, -ret, + "Failed to receive CHECK_DEVICE_STATE reply"); + return ret; + } + + if (msg.hdr.request != VHOST_USER_CHECK_DEVICE_STATE) { + error_setg(errp, + "Received unexpected message type, expected %d, received %d", + VHOST_USER_CHECK_DEVICE_STATE, msg.hdr.request); + return -EPROTO; + } + + if (msg.hdr.size != sizeof(msg.payload.u64)) { + error_setg(errp, + "Received bad message size, expected %zu, received %" PRIu32, + sizeof(msg.payload.u64), msg.hdr.size); + return -EPROTO; + } + + if (msg.payload.u64 != 0) { + error_setg(errp, "Back-end failed to process its internal state"); + return -EIO; + } + + return 0; +} + const VhostOps user_ops = { .backend_type = VHOST_BACKEND_TYPE_USER, .vhost_backend_init = vhost_user_backend_init, @@ -2890,4 +3033,7 @@ const VhostOps user_ops = { .vhost_set_inflight_fd = vhost_user_set_inflight_fd, .vhost_dev_start = vhost_user_dev_start, .vhost_reset_status = vhost_user_reset_status, + .vhost_supports_device_state = vhost_user_supports_device_state, + .vhost_set_device_state_fd = vhost_user_set_device_state_fd, + .vhost_check_device_state = vhost_user_check_device_state, }; diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c index 9c9ae7109e..4db9dbfd64 100644 --- a/hw/virtio/vhost.c +++ b/hw/virtio/vhost.c @@ -2159,3 +2159,40 @@ int vhost_reset_device(struct vhost_dev *hdev) return -ENOSYS; } + +bool vhost_supports_device_state(struct vhost_dev *dev) +{ + if (dev->vhost_ops->vhost_supports_device_state) { + return dev->vhost_ops->vhost_supports_device_state(dev); + } + + return false; +} + +int vhost_set_device_state_fd(struct vhost_dev *dev, + VhostDeviceStateDirection direction, + VhostDeviceStatePhase phase, + int fd, + int *reply_fd, + Error **errp) +{ + if (dev->vhost_ops->vhost_set_device_state_fd) { + return dev->vhost_ops->vhost_set_device_state_fd(dev, direction, phase, + fd, reply_fd, errp); + } + + error_setg(errp, + "vhost transport does not support migration state transfer"); + return -ENOSYS; +} + +int vhost_check_device_state(struct vhost_dev *dev, Error **errp) +{ + if (dev->vhost_ops->vhost_check_device_state) { + return dev->vhost_ops->vhost_check_device_state(dev, errp); + } + + error_setg(errp, + "vhost transport does not support migration state transfer"); + return -ENOSYS; +} diff --git a/include/hw/virtio/vhost-backend.h b/include/hw/virtio/vhost-backend.h index 96ccc18cd3..a86d103f82 100644 --- a/include/hw/virtio/vhost-backend.h +++ b/include/hw/virtio/vhost-backend.h @@ -26,6 +26,18 @@ typedef enum VhostSetConfigType { VHOST_SET_CONFIG_TYPE_MIGRATION = 1, } VhostSetConfigType; +typedef enum VhostDeviceStateDirection { + /* Transfer state from back-end (device) to front-end */ + VHOST_TRANSFER_STATE_DIRECTION_SAVE = 0, + /* Transfer state from front-end to back-end (device) */ + VHOST_TRANSFER_STATE_DIRECTION_LOAD = 1, +} VhostDeviceStateDirection; + +typedef enum VhostDeviceStatePhase { + /* The device (and all its vrings) is stopped */ + VHOST_TRANSFER_STATE_PHASE_STOPPED = 0, +} VhostDeviceStatePhase; + struct vhost_inflight; struct vhost_dev; struct vhost_log; @@ -129,6 +141,15 @@ typedef int (*vhost_set_config_call_op)(struct vhost_dev *dev, typedef void (*vhost_reset_status_op)(struct vhost_dev *dev); +typedef bool (*vhost_supports_device_state_op)(struct vhost_dev *dev); +typedef int (*vhost_set_device_state_fd_op)(struct vhost_dev *dev, + VhostDeviceStateDirection direction, + VhostDeviceStatePhase phase, + int fd, + int *reply_fd, + Error **errp); +typedef int (*vhost_check_device_state_op)(struct vhost_dev *dev, Error **errp); + typedef struct VhostOps { VhostBackendType backend_type; vhost_backend_init vhost_backend_init; @@ -176,6 +197,9 @@ typedef struct VhostOps { vhost_force_iommu_op vhost_force_iommu; vhost_set_config_call_op vhost_set_config_call; vhost_reset_status_op vhost_reset_status; + vhost_supports_device_state_op vhost_supports_device_state; + vhost_set_device_state_fd_op vhost_set_device_state_fd; + vhost_check_device_state_op vhost_check_device_state; } VhostOps; int vhost_backend_update_device_iotlb(struct vhost_dev *dev, diff --git a/include/hw/virtio/vhost-user.h b/include/hw/virtio/vhost-user.h index 20b69d8e85..d7c09ffd34 100644 --- a/include/hw/virtio/vhost-user.h +++ b/include/hw/virtio/vhost-user.h @@ -31,6 +31,7 @@ enum VhostUserProtocolFeature { VHOST_USER_PROTOCOL_F_STATUS = 16, /* Feature 17 reserved for VHOST_USER_PROTOCOL_F_XEN_MMAP. */ VHOST_USER_PROTOCOL_F_SHARED_OBJECT = 18, + VHOST_USER_PROTOCOL_F_DEVICE_STATE = 19, VHOST_USER_PROTOCOL_F_MAX }; diff --git a/include/hw/virtio/vhost.h b/include/hw/virtio/vhost.h index 5e8183f64a..b6ee6da6ce 100644 --- a/include/hw/virtio/vhost.h +++ b/include/hw/virtio/vhost.h @@ -351,4 +351,82 @@ static inline int vhost_reset_device(struct vhost_dev *hdev) } #endif /* CONFIG_VHOST */ +/** + * vhost_supports_device_state(): Checks whether the back-end supports + * transferring internal device state for the purpose of migration. + * Support for this feature is required for vhost_set_device_state_fd() + * and vhost_check_device_state(). + * + * @dev: The vhost device + * + * Returns true if the device supports these commands, and false if it + * does not. + */ +bool vhost_supports_device_state(struct vhost_dev *dev); + +/** + * vhost_set_device_state_fd(): Begin transfer of internal state from/to + * the back-end for the purpose of migration. Data is to be transferred + * over a pipe according to @direction and @phase. The sending end must + * only write to the pipe, and the receiving end must only read from it. + * Once the sending end is done, it closes its FD. The receiving end + * must take this as the end-of-transfer signal and close its FD, too. + * + * @fd is the back-end's end of the pipe: The write FD for SAVE, and the + * read FD for LOAD. This function transfers ownership of @fd to the + * back-end, i.e. closes it in the front-end. + * + * The back-end may optionally reply with an FD of its own, if this + * improves efficiency on its end. In this case, the returned FD is + * stored in *reply_fd. The back-end will discard the FD sent to it, + * and the front-end must use *reply_fd for transferring state to/from + * the back-end. + * + * @dev: The vhost device + * @direction: The direction in which the state is to be transferred. + * For outgoing migrations, this is SAVE, and data is read + * from the back-end and stored by the front-end in the + * migration stream. + * For incoming migrations, this is LOAD, and data is read + * by the front-end from the migration stream and sent to + * the back-end to restore the saved state. + * @phase: Which migration phase we are in. Currently, there is only + * STOPPED (device and all vrings are stopped), in the future, + * more phases such as PRE_COPY or POST_COPY may be added. + * @fd: Back-end's end of the pipe through which to transfer state; note + * that ownership is transferred to the back-end, so this function + * closes @fd in the front-end. + * @reply_fd: If the back-end wishes to use a different pipe for state + * transfer, this will contain an FD for the front-end to + * use. Otherwise, -1 is stored here. + * @errp: Potential error description + * + * Returns 0 on success, and -errno on failure. + */ +int vhost_set_device_state_fd(struct vhost_dev *dev, + VhostDeviceStateDirection direction, + VhostDeviceStatePhase phase, + int fd, + int *reply_fd, + Error **errp); + +/** + * vhost_set_device_state_fd(): After transferring state from/to the + * back-end via vhost_set_device_state_fd(), i.e. once the sending end + * has closed the pipe, inquire the back-end to report any potential + * errors that have occurred on its side. This allows to sense errors + * like: + * - During outgoing migration, when the source side had already started + * to produce its state, something went wrong and it failed to finish + * - During incoming migration, when the received state is somehow + * invalid and cannot be processed by the back-end + * + * @dev: The vhost device + * @errp: Potential error description + * + * Returns 0 when the back-end reports successful state transfer and + * processing, and -errno when an error occurred somewhere. + */ +int vhost_check_device_state(struct vhost_dev *dev, Error **errp); + #endif -- cgit v1.2.3 From 4a00d5d7f4b65ba99b33d5a0d6f8c563895839ea Mon Sep 17 00:00:00 2001 From: Hanna Czenczek Date: Mon, 16 Oct 2023 15:42:42 +0200 Subject: vhost: Add high-level state save/load functions vhost_save_backend_state() and vhost_load_backend_state() can be used by vhost front-ends to easily save and load the back-end's state to/from the migration stream. Because we do not know the full state size ahead of time, vhost_save_backend_state() simply reads the data in 1 MB chunks, and writes each chunk consecutively into the migration stream, prefixed by its length. EOF is indicated by a 0-length chunk. Reviewed-by: Stefan Hajnoczi Signed-off-by: Hanna Czenczek Message-Id: <20231016134243.68248-7-hreitz@redhat.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/virtio/vhost.c | 204 ++++++++++++++++++++++++++++++++++++++++++++++ include/hw/virtio/vhost.h | 35 ++++++++ 2 files changed, 239 insertions(+) diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c index 4db9dbfd64..2c9ac79468 100644 --- a/hw/virtio/vhost.c +++ b/hw/virtio/vhost.c @@ -2196,3 +2196,207 @@ int vhost_check_device_state(struct vhost_dev *dev, Error **errp) "vhost transport does not support migration state transfer"); return -ENOSYS; } + +int vhost_save_backend_state(struct vhost_dev *dev, QEMUFile *f, Error **errp) +{ + /* Maximum chunk size in which to transfer the state */ + const size_t chunk_size = 1 * 1024 * 1024; + g_autofree void *transfer_buf = NULL; + g_autoptr(GError) g_err = NULL; + int pipe_fds[2], read_fd = -1, write_fd = -1, reply_fd = -1; + int ret; + + /* [0] for reading (our end), [1] for writing (back-end's end) */ + if (!g_unix_open_pipe(pipe_fds, FD_CLOEXEC, &g_err)) { + error_setg(errp, "Failed to set up state transfer pipe: %s", + g_err->message); + ret = -EINVAL; + goto fail; + } + + read_fd = pipe_fds[0]; + write_fd = pipe_fds[1]; + + /* + * VHOST_TRANSFER_STATE_PHASE_STOPPED means the device must be stopped. + * Ideally, it is suspended, but SUSPEND/RESUME currently do not exist for + * vhost-user, so just check that it is stopped at all. + */ + assert(!dev->started); + + /* Transfer ownership of write_fd to the back-end */ + ret = vhost_set_device_state_fd(dev, + VHOST_TRANSFER_STATE_DIRECTION_SAVE, + VHOST_TRANSFER_STATE_PHASE_STOPPED, + write_fd, + &reply_fd, + errp); + if (ret < 0) { + error_prepend(errp, "Failed to initiate state transfer: "); + goto fail; + } + + /* If the back-end wishes to use a different pipe, switch over */ + if (reply_fd >= 0) { + close(read_fd); + read_fd = reply_fd; + } + + transfer_buf = g_malloc(chunk_size); + + while (true) { + ssize_t read_ret; + + read_ret = RETRY_ON_EINTR(read(read_fd, transfer_buf, chunk_size)); + if (read_ret < 0) { + ret = -errno; + error_setg_errno(errp, -ret, "Failed to receive state"); + goto fail; + } + + assert(read_ret <= chunk_size); + qemu_put_be32(f, read_ret); + + if (read_ret == 0) { + /* EOF */ + break; + } + + qemu_put_buffer(f, transfer_buf, read_ret); + } + + /* + * Back-end will not really care, but be clean and close our end of the pipe + * before inquiring the back-end about whether transfer was successful + */ + close(read_fd); + read_fd = -1; + + /* Also, verify that the device is still stopped */ + assert(!dev->started); + + ret = vhost_check_device_state(dev, errp); + if (ret < 0) { + goto fail; + } + + ret = 0; +fail: + if (read_fd >= 0) { + close(read_fd); + } + + return ret; +} + +int vhost_load_backend_state(struct vhost_dev *dev, QEMUFile *f, Error **errp) +{ + size_t transfer_buf_size = 0; + g_autofree void *transfer_buf = NULL; + g_autoptr(GError) g_err = NULL; + int pipe_fds[2], read_fd = -1, write_fd = -1, reply_fd = -1; + int ret; + + /* [0] for reading (back-end's end), [1] for writing (our end) */ + if (!g_unix_open_pipe(pipe_fds, FD_CLOEXEC, &g_err)) { + error_setg(errp, "Failed to set up state transfer pipe: %s", + g_err->message); + ret = -EINVAL; + goto fail; + } + + read_fd = pipe_fds[0]; + write_fd = pipe_fds[1]; + + /* + * VHOST_TRANSFER_STATE_PHASE_STOPPED means the device must be stopped. + * Ideally, it is suspended, but SUSPEND/RESUME currently do not exist for + * vhost-user, so just check that it is stopped at all. + */ + assert(!dev->started); + + /* Transfer ownership of read_fd to the back-end */ + ret = vhost_set_device_state_fd(dev, + VHOST_TRANSFER_STATE_DIRECTION_LOAD, + VHOST_TRANSFER_STATE_PHASE_STOPPED, + read_fd, + &reply_fd, + errp); + if (ret < 0) { + error_prepend(errp, "Failed to initiate state transfer: "); + goto fail; + } + + /* If the back-end wishes to use a different pipe, switch over */ + if (reply_fd >= 0) { + close(write_fd); + write_fd = reply_fd; + } + + while (true) { + size_t this_chunk_size = qemu_get_be32(f); + ssize_t write_ret; + const uint8_t *transfer_pointer; + + if (this_chunk_size == 0) { + /* End of state */ + break; + } + + if (transfer_buf_size < this_chunk_size) { + transfer_buf = g_realloc(transfer_buf, this_chunk_size); + transfer_buf_size = this_chunk_size; + } + + if (qemu_get_buffer(f, transfer_buf, this_chunk_size) < + this_chunk_size) + { + error_setg(errp, "Failed to read state"); + ret = -EINVAL; + goto fail; + } + + transfer_pointer = transfer_buf; + while (this_chunk_size > 0) { + write_ret = RETRY_ON_EINTR( + write(write_fd, transfer_pointer, this_chunk_size) + ); + if (write_ret < 0) { + ret = -errno; + error_setg_errno(errp, -ret, "Failed to send state"); + goto fail; + } else if (write_ret == 0) { + error_setg(errp, "Failed to send state: Connection is closed"); + ret = -ECONNRESET; + goto fail; + } + + assert(write_ret <= this_chunk_size); + this_chunk_size -= write_ret; + transfer_pointer += write_ret; + } + } + + /* + * Close our end, thus ending transfer, before inquiring the back-end about + * whether transfer was successful + */ + close(write_fd); + write_fd = -1; + + /* Also, verify that the device is still stopped */ + assert(!dev->started); + + ret = vhost_check_device_state(dev, errp); + if (ret < 0) { + goto fail; + } + + ret = 0; +fail: + if (write_fd >= 0) { + close(write_fd); + } + + return ret; +} diff --git a/include/hw/virtio/vhost.h b/include/hw/virtio/vhost.h index b6ee6da6ce..05d7204a08 100644 --- a/include/hw/virtio/vhost.h +++ b/include/hw/virtio/vhost.h @@ -429,4 +429,39 @@ int vhost_set_device_state_fd(struct vhost_dev *dev, */ int vhost_check_device_state(struct vhost_dev *dev, Error **errp); +/** + * vhost_save_backend_state(): High-level function to receive a vhost + * back-end's state, and save it in @f. Uses + * `vhost_set_device_state_fd()` to get the data from the back-end, and + * stores it in consecutive chunks that are each prefixed by their + * respective length (be32). The end is marked by a 0-length chunk. + * + * Must only be called while the device and all its vrings are stopped + * (`VHOST_TRANSFER_STATE_PHASE_STOPPED`). + * + * @dev: The vhost device from which to save the state + * @f: Migration stream in which to save the state + * @errp: Potential error message + * + * Returns 0 on success, and -errno otherwise. + */ +int vhost_save_backend_state(struct vhost_dev *dev, QEMUFile *f, Error **errp); + +/** + * vhost_load_backend_state(): High-level function to load a vhost + * back-end's state from @f, and send it over to the back-end. Reads + * the data from @f in the format used by `vhost_save_state()`, and uses + * `vhost_set_device_state_fd()` to transfer it to the back-end. + * + * Must only be called while the device and all its vrings are stopped + * (`VHOST_TRANSFER_STATE_PHASE_STOPPED`). + * + * @dev: The vhost device to which to send the sate + * @f: Migration stream from which to load the state + * @errp: Potential error message + * + * Returns 0 on success, and -errno otherwise. + */ +int vhost_load_backend_state(struct vhost_dev *dev, QEMUFile *f, Error **errp); + #endif -- cgit v1.2.3 From bca3e2a13814253b4ed878a3313688554edd1b66 Mon Sep 17 00:00:00 2001 From: Hanna Czenczek Date: Mon, 16 Oct 2023 15:42:43 +0200 Subject: vhost-user-fs: Implement internal migration A virtio-fs device's VM state consists of: - the virtio device (vring) state (VMSTATE_VIRTIO_DEVICE) - the back-end's (virtiofsd's) internal state We get/set the latter via the new vhost operations to transfer migratory state. It is its own dedicated subsection, so that for external migration, it can be disabled. Reviewed-by: Stefan Hajnoczi Signed-off-by: Hanna Czenczek Message-Id: <20231016134243.68248-8-hreitz@redhat.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/virtio/vhost-user-fs.c | 101 +++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 100 insertions(+), 1 deletion(-) diff --git a/hw/virtio/vhost-user-fs.c b/hw/virtio/vhost-user-fs.c index 49d699ffc2..eb91723855 100644 --- a/hw/virtio/vhost-user-fs.c +++ b/hw/virtio/vhost-user-fs.c @@ -298,9 +298,108 @@ static struct vhost_dev *vuf_get_vhost(VirtIODevice *vdev) return &fs->vhost_dev; } +/** + * Fetch the internal state from virtiofsd and save it to `f`. + */ +static int vuf_save_state(QEMUFile *f, void *pv, size_t size, + const VMStateField *field, JSONWriter *vmdesc) +{ + VirtIODevice *vdev = pv; + VHostUserFS *fs = VHOST_USER_FS(vdev); + Error *local_error = NULL; + int ret; + + ret = vhost_save_backend_state(&fs->vhost_dev, f, &local_error); + if (ret < 0) { + error_reportf_err(local_error, + "Error saving back-end state of %s device %s " + "(tag: \"%s\"): ", + vdev->name, vdev->parent_obj.canonical_path, + fs->conf.tag ?: ""); + return ret; + } + + return 0; +} + +/** + * Load virtiofsd's internal state from `f` and send it over to virtiofsd. + */ +static int vuf_load_state(QEMUFile *f, void *pv, size_t size, + const VMStateField *field) +{ + VirtIODevice *vdev = pv; + VHostUserFS *fs = VHOST_USER_FS(vdev); + Error *local_error = NULL; + int ret; + + ret = vhost_load_backend_state(&fs->vhost_dev, f, &local_error); + if (ret < 0) { + error_reportf_err(local_error, + "Error loading back-end state of %s device %s " + "(tag: \"%s\"): ", + vdev->name, vdev->parent_obj.canonical_path, + fs->conf.tag ?: ""); + return ret; + } + + return 0; +} + +static bool vuf_is_internal_migration(void *opaque) +{ + /* TODO: Return false when an external migration is requested */ + return true; +} + +static int vuf_check_migration_support(void *opaque) +{ + VirtIODevice *vdev = opaque; + VHostUserFS *fs = VHOST_USER_FS(vdev); + + if (!vhost_supports_device_state(&fs->vhost_dev)) { + error_report("Back-end of %s device %s (tag: \"%s\") does not support " + "migration through qemu", + vdev->name, vdev->parent_obj.canonical_path, + fs->conf.tag ?: ""); + return -ENOTSUP; + } + + return 0; +} + +static const VMStateDescription vuf_backend_vmstate; + static const VMStateDescription vuf_vmstate = { .name = "vhost-user-fs", - .unmigratable = 1, + .version_id = 0, + .fields = (VMStateField[]) { + VMSTATE_VIRTIO_DEVICE, + VMSTATE_END_OF_LIST() + }, + .subsections = (const VMStateDescription * []) { + &vuf_backend_vmstate, + NULL, + } +}; + +static const VMStateDescription vuf_backend_vmstate = { + .name = "vhost-user-fs-backend", + .version_id = 0, + .needed = vuf_is_internal_migration, + .pre_load = vuf_check_migration_support, + .pre_save = vuf_check_migration_support, + .fields = (VMStateField[]) { + { + .name = "back-end", + .info = &(const VMStateInfo) { + .name = "virtio-fs back-end state", + .get = vuf_load_state, + .put = vuf_save_state, + }, + }, + VMSTATE_END_OF_LIST() + }, }; static Property vuf_properties[] = { -- cgit v1.2.3 From 2880e676c000a62828d3d9ece7b2ec7a513560a2 Mon Sep 17 00:00:00 2001 From: Manos Pitsidianakis Date: Mon, 23 Oct 2023 15:03:18 +0300 Subject: Add virtio-sound device stub MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add a new VIRTIO device for the virtio sound device id. Functionality will be added in the following commits. Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin Signed-off-by: Anton Yakovlev Signed-off-by: Manos Pitsidianakis Reviewed-by: Alex Bennée Tested-by: Alex Bennée Message-Id: Acked-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- MAINTAINERS | 7 ++ hw/audio/Kconfig | 5 + hw/audio/meson.build | 1 + hw/audio/trace-events | 9 ++ hw/audio/virtio-snd.c | 233 ++++++++++++++++++++++++++++++++++++++++++ include/hw/audio/virtio-snd.h | 79 ++++++++++++++ system/qdev-monitor.c | 1 + 7 files changed, 335 insertions(+) create mode 100644 hw/audio/virtio-snd.c create mode 100644 include/hw/audio/virtio-snd.h diff --git a/MAINTAINERS b/MAINTAINERS index 8e8a7d5be5..d3ee463d21 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2310,6 +2310,13 @@ F: hw/virtio/virtio-mem-pci.h F: hw/virtio/virtio-mem-pci.c F: include/hw/virtio/virtio-mem.h +virtio-snd +M: Gerd Hoffmann +R: Manos Pitsidianakis +S: Supported +F: hw/audio/virtio-snd.c +F: include/hw/audio/virtio-snd.h + nvme M: Keith Busch M: Klaus Jensen diff --git a/hw/audio/Kconfig b/hw/audio/Kconfig index d0993514a1..daf060e1be 100644 --- a/hw/audio/Kconfig +++ b/hw/audio/Kconfig @@ -50,3 +50,8 @@ config CS4231 config ASC bool + +config VIRTIO_SND + bool + default y + depends on VIRTIO diff --git a/hw/audio/meson.build b/hw/audio/meson.build index 8805322f5c..7a503be1fd 100644 --- a/hw/audio/meson.build +++ b/hw/audio/meson.build @@ -13,3 +13,4 @@ system_ss.add(when: 'CONFIG_PL041', if_true: files('pl041.c', 'lm4549.c')) system_ss.add(when: 'CONFIG_SB16', if_true: files('sb16.c')) system_ss.add(when: 'CONFIG_VT82C686', if_true: files('via-ac97.c')) system_ss.add(when: 'CONFIG_WM8750', if_true: files('wm8750.c')) +system_ss.add(when: ['CONFIG_VIRTIO_SND', 'CONFIG_VIRTIO'], if_true: files('virtio-snd.c')) diff --git a/hw/audio/trace-events b/hw/audio/trace-events index 059ce451f5..525ced2b34 100644 --- a/hw/audio/trace-events +++ b/hw/audio/trace-events @@ -38,3 +38,12 @@ asc_write_fifo(const char fifo, int reg, unsigned size, int wrptr, int cnt, uint asc_write_reg(int reg, unsigned size, uint64_t value) "reg=0x%03x size=%u value=0x%"PRIx64 asc_write_extreg(const char fifo, int reg, unsigned size, uint64_t value) "fifo %c reg=0x%03x size=%u value=0x%"PRIx64 asc_update_irq(int irq, int a, int b) "set IRQ to %d (A: 0x%x B: 0x%x)" + +#virtio-snd.c +virtio_snd_get_config(void *vdev, uint32_t jacks, uint32_t streams, uint32_t chmaps) "snd %p: get_config jacks=%"PRIu32" streams=%"PRIu32" chmaps=%"PRIu32"" +virtio_snd_set_config(void *vdev, uint32_t jacks, uint32_t new_jacks, uint32_t streams, uint32_t new_streams, uint32_t chmaps, uint32_t new_chmaps) "snd %p: set_config jacks from %"PRIu32"->%"PRIu32", streams from %"PRIu32"->%"PRIu32", chmaps from %"PRIu32"->%"PRIu32 +virtio_snd_get_features(void *vdev, uint64_t features) "snd %p: get_features 0x%"PRIx64 +virtio_snd_vm_state_running(void) "vm state running" +virtio_snd_vm_state_stopped(void) "vm state stopped" +virtio_snd_realize(void *snd) "snd %p: realize" +virtio_snd_unrealize(void *snd) "snd %p: unrealize" diff --git a/hw/audio/virtio-snd.c b/hw/audio/virtio-snd.c new file mode 100644 index 0000000000..3a4b441c20 --- /dev/null +++ b/hw/audio/virtio-snd.c @@ -0,0 +1,233 @@ +/* + * VIRTIO Sound Device conforming to + * + * "Virtual I/O Device (VIRTIO) Version 1.2 + * Committee Specification Draft 01 + * 09 May 2022" + * + * + * + * Copyright (c) 2023 Emmanouil Pitsidianakis + * Copyright (C) 2019 OpenSynergy GmbH + * + * This work is licensed under the terms of the GNU GPL, version 2 or + * (at your option) any later version. See the COPYING file in the + * top-level directory. + */ + +#include "qemu/osdep.h" +#include "qemu/iov.h" +#include "qemu/log.h" +#include "qemu/error-report.h" +#include "include/qemu/lockable.h" +#include "sysemu/runstate.h" +#include "trace.h" +#include "qapi/error.h" +#include "hw/audio/virtio-snd.h" +#include "hw/core/cpu.h" + +#define VIRTIO_SOUND_VM_VERSION 1 +#define VIRTIO_SOUND_JACK_DEFAULT 0 +#define VIRTIO_SOUND_STREAM_DEFAULT 1 +#define VIRTIO_SOUND_CHMAP_DEFAULT 0 +#define VIRTIO_SOUND_HDA_FN_NID 0 + +static const VMStateDescription vmstate_virtio_snd_device = { + .name = TYPE_VIRTIO_SND, + .version_id = VIRTIO_SOUND_VM_VERSION, + .minimum_version_id = VIRTIO_SOUND_VM_VERSION, +}; + +static const VMStateDescription vmstate_virtio_snd = { + .name = TYPE_VIRTIO_SND, + .minimum_version_id = VIRTIO_SOUND_VM_VERSION, + .version_id = VIRTIO_SOUND_VM_VERSION, + .fields = (VMStateField[]) { + VMSTATE_VIRTIO_DEVICE, + VMSTATE_END_OF_LIST() + }, +}; + +static Property virtio_snd_properties[] = { + DEFINE_AUDIO_PROPERTIES(VirtIOSound, card), + DEFINE_PROP_UINT32("jacks", VirtIOSound, snd_conf.jacks, + VIRTIO_SOUND_JACK_DEFAULT), + DEFINE_PROP_UINT32("streams", VirtIOSound, snd_conf.streams, + VIRTIO_SOUND_STREAM_DEFAULT), + DEFINE_PROP_UINT32("chmaps", VirtIOSound, snd_conf.chmaps, + VIRTIO_SOUND_CHMAP_DEFAULT), + DEFINE_PROP_END_OF_LIST(), +}; + +static void +virtio_snd_get_config(VirtIODevice *vdev, uint8_t *config) +{ + VirtIOSound *s = VIRTIO_SND(vdev); + virtio_snd_config *sndconfig = + (virtio_snd_config *)config; + trace_virtio_snd_get_config(vdev, + s->snd_conf.jacks, + s->snd_conf.streams, + s->snd_conf.chmaps); + + memcpy(sndconfig, &s->snd_conf, sizeof(s->snd_conf)); + cpu_to_le32s(&sndconfig->jacks); + cpu_to_le32s(&sndconfig->streams); + cpu_to_le32s(&sndconfig->chmaps); + +} + +static void +virtio_snd_set_config(VirtIODevice *vdev, const uint8_t *config) +{ + VirtIOSound *s = VIRTIO_SND(vdev); + const virtio_snd_config *sndconfig = + (const virtio_snd_config *)config; + + + trace_virtio_snd_set_config(vdev, + s->snd_conf.jacks, + sndconfig->jacks, + s->snd_conf.streams, + sndconfig->streams, + s->snd_conf.chmaps, + sndconfig->chmaps); + + memcpy(&s->snd_conf, sndconfig, sizeof(virtio_snd_config)); + le32_to_cpus(&s->snd_conf.jacks); + le32_to_cpus(&s->snd_conf.streams); + le32_to_cpus(&s->snd_conf.chmaps); + +} + +/* + * Queue handler stub. + * + * @vdev: VirtIOSound device + * @vq: virtqueue + */ +static void virtio_snd_handle_queue(VirtIODevice *vdev, VirtQueue *vq) {} + +static uint64_t get_features(VirtIODevice *vdev, uint64_t features, + Error **errp) +{ + /* + * virtio-v1.2-csd01, 5.14.3, + * Feature Bits + * None currently defined. + */ + VirtIOSound *s = VIRTIO_SND(vdev); + features |= s->features; + + trace_virtio_snd_get_features(vdev, features); + + return features; +} + +static void +virtio_snd_vm_state_change(void *opaque, bool running, + RunState state) +{ + if (running) { + trace_virtio_snd_vm_state_running(); + } else { + trace_virtio_snd_vm_state_stopped(); + } +} + +static void virtio_snd_realize(DeviceState *dev, Error **errp) +{ + ERRP_GUARD(); + VirtIOSound *vsnd = VIRTIO_SND(dev); + VirtIODevice *vdev = VIRTIO_DEVICE(dev); + + vsnd->vmstate = + qemu_add_vm_change_state_handler(virtio_snd_vm_state_change, vsnd); + + trace_virtio_snd_realize(vsnd); + + virtio_init(vdev, VIRTIO_ID_SOUND, sizeof(virtio_snd_config)); + virtio_add_feature(&vsnd->features, VIRTIO_F_VERSION_1); + + /* set number of jacks and streams */ + if (vsnd->snd_conf.jacks > 8) { + error_setg(errp, + "Invalid number of jacks: %"PRIu32, + vsnd->snd_conf.jacks); + return; + } + if (vsnd->snd_conf.streams < 1 || vsnd->snd_conf.streams > 10) { + error_setg(errp, + "Invalid number of streams: %"PRIu32, + vsnd->snd_conf.streams); + return; + } + + if (vsnd->snd_conf.chmaps > VIRTIO_SND_CHMAP_MAX_SIZE) { + error_setg(errp, + "Invalid number of channel maps: %"PRIu32, + vsnd->snd_conf.chmaps); + return; + } + + AUD_register_card("virtio-sound", &vsnd->card, errp); + + vsnd->queues[VIRTIO_SND_VQ_CONTROL] = + virtio_add_queue(vdev, 64, virtio_snd_handle_queue); + vsnd->queues[VIRTIO_SND_VQ_EVENT] = + virtio_add_queue(vdev, 64, virtio_snd_handle_queue); + vsnd->queues[VIRTIO_SND_VQ_TX] = + virtio_add_queue(vdev, 64, virtio_snd_handle_queue); + vsnd->queues[VIRTIO_SND_VQ_RX] = + virtio_add_queue(vdev, 64, virtio_snd_handle_queue); +} + +static void virtio_snd_unrealize(DeviceState *dev) +{ + VirtIODevice *vdev = VIRTIO_DEVICE(dev); + VirtIOSound *vsnd = VIRTIO_SND(dev); + + qemu_del_vm_change_state_handler(vsnd->vmstate); + trace_virtio_snd_unrealize(vsnd); + + AUD_remove_card(&vsnd->card); + virtio_delete_queue(vsnd->queues[VIRTIO_SND_VQ_CONTROL]); + virtio_delete_queue(vsnd->queues[VIRTIO_SND_VQ_EVENT]); + virtio_delete_queue(vsnd->queues[VIRTIO_SND_VQ_TX]); + virtio_delete_queue(vsnd->queues[VIRTIO_SND_VQ_RX]); + virtio_cleanup(vdev); +} + + +static void virtio_snd_reset(VirtIODevice *vdev) {} + +static void virtio_snd_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass); + + + set_bit(DEVICE_CATEGORY_SOUND, dc->categories); + device_class_set_props(dc, virtio_snd_properties); + + dc->vmsd = &vmstate_virtio_snd; + vdc->vmsd = &vmstate_virtio_snd_device; + vdc->realize = virtio_snd_realize; + vdc->unrealize = virtio_snd_unrealize; + vdc->get_config = virtio_snd_get_config; + vdc->set_config = virtio_snd_set_config; + vdc->get_features = get_features; + vdc->reset = virtio_snd_reset; + vdc->legacy_features = 0; +} + +static const TypeInfo virtio_snd_types[] = { + { + .name = TYPE_VIRTIO_SND, + .parent = TYPE_VIRTIO_DEVICE, + .instance_size = sizeof(VirtIOSound), + .class_init = virtio_snd_class_init, + } +}; + +DEFINE_TYPES(virtio_snd_types) diff --git a/include/hw/audio/virtio-snd.h b/include/hw/audio/virtio-snd.h new file mode 100644 index 0000000000..d08065941c --- /dev/null +++ b/include/hw/audio/virtio-snd.h @@ -0,0 +1,79 @@ +/* + * VIRTIO Sound Device conforming to + * + * "Virtual I/O Device (VIRTIO) Version 1.2 + * Committee Specification Draft 01 + * 09 May 2022" + * + * Copyright (c) 2023 Emmanouil Pitsidianakis + * Copyright (C) 2019 OpenSynergy GmbH + * + * This work is licensed under the terms of the GNU GPL, version 2 or + * (at your option) any later version. See the COPYING file in the + * top-level directory. + */ + +#ifndef QEMU_VIRTIO_SOUND_H +#define QEMU_VIRTIO_SOUND_H + +#include "hw/virtio/virtio.h" +#include "audio/audio.h" +#include "standard-headers/linux/virtio_ids.h" +#include "standard-headers/linux/virtio_snd.h" + +#define TYPE_VIRTIO_SND "virtio-sound-device" +#define VIRTIO_SND(obj) \ + OBJECT_CHECK(VirtIOSound, (obj), TYPE_VIRTIO_SND) + +/* CONFIGURATION SPACE */ + +typedef struct virtio_snd_config virtio_snd_config; + +/* COMMON DEFINITIONS */ + +/* common header for request/response*/ +typedef struct virtio_snd_hdr virtio_snd_hdr; + +/* event notification */ +typedef struct virtio_snd_event virtio_snd_event; + +/* common control request to query an item information */ +typedef struct virtio_snd_query_info virtio_snd_query_info; + +/* JACK CONTROL MESSAGES */ + +typedef struct virtio_snd_jack_hdr virtio_snd_jack_hdr; + +/* jack information structure */ +typedef struct virtio_snd_jack_info virtio_snd_jack_info; + +/* jack remapping control request */ +typedef struct virtio_snd_jack_remap virtio_snd_jack_remap; + +/* + * PCM CONTROL MESSAGES + */ +typedef struct virtio_snd_pcm_hdr virtio_snd_pcm_hdr; + +/* PCM stream info structure */ +typedef struct virtio_snd_pcm_info virtio_snd_pcm_info; + +/* set PCM stream params */ +typedef struct virtio_snd_pcm_set_params virtio_snd_pcm_set_params; + +/* I/O request header */ +typedef struct virtio_snd_pcm_xfer virtio_snd_pcm_xfer; + +/* I/O request status */ +typedef struct virtio_snd_pcm_status virtio_snd_pcm_status; + +typedef struct VirtIOSound { + VirtIODevice parent_obj; + + VirtQueue *queues[VIRTIO_SND_VQ_MAX]; + uint64_t features; + QEMUSoundCard card; + VMChangeStateEntry *vmstate; + virtio_snd_config snd_conf; +} VirtIOSound; +#endif diff --git a/system/qdev-monitor.c b/system/qdev-monitor.c index 1b8005ae55..3c0aabec4b 100644 --- a/system/qdev-monitor.c +++ b/system/qdev-monitor.c @@ -111,6 +111,7 @@ static const QDevAlias qdev_alias_table[] = { { "virtio-serial-device", "virtio-serial", QEMU_ARCH_VIRTIO_MMIO }, { "virtio-serial-ccw", "virtio-serial", QEMU_ARCH_VIRTIO_CCW }, { "virtio-serial-pci", "virtio-serial", QEMU_ARCH_VIRTIO_PCI}, + { "virtio-sound-device", "virtio-sound", QEMU_ARCH_VIRTIO_MMIO }, { "virtio-tablet-device", "virtio-tablet", QEMU_ARCH_VIRTIO_MMIO }, { "virtio-tablet-ccw", "virtio-tablet", QEMU_ARCH_VIRTIO_CCW }, { "virtio-tablet-pci", "virtio-tablet", QEMU_ARCH_VIRTIO_PCI }, -- cgit v1.2.3 From 2426908590c1d8d9ffb741c5552ad45d9d3ba9f8 Mon Sep 17 00:00:00 2001 From: Manos Pitsidianakis Date: Mon, 23 Oct 2023 15:03:19 +0300 Subject: Add virtio-sound-pci device MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch adds a PCI wrapper device for the virtio-sound device. It is necessary to instantiate a virtio-snd device in a guest. All sound logic will be added to the virtio-snd device in the following commits. To add this device with a guest, you'll need a >=5.13 kernel compiled with CONFIG_SND_VIRTIO=y, which at the time of writing most distros have off by default. Use with following flags in the invocation: Pulseaudio: -audio driver=pa,model=virtio or -audio driver=pa,model=virtio,server=/run/user/1000/pulse/native sdl: -audio driver=sdl,model=virtio coreaudio (macos/darwin): -audio driver=coreaudio,model=virtio etc. Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin Signed-off-by: Anton Yakovlev Signed-off-by: Manos Pitsidianakis Reviewed-by: Alex Bennée Tested-by: Alex Bennée Message-Id: Acked-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- MAINTAINERS | 1 + hw/audio/meson.build | 1 + hw/audio/virtio-snd-pci.c | 93 +++++++++++++++++++++++++++++++++++++++++++++++ system/qdev-monitor.c | 1 + 4 files changed, 96 insertions(+) create mode 100644 hw/audio/virtio-snd-pci.c diff --git a/MAINTAINERS b/MAINTAINERS index d3ee463d21..c09bb8cf41 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2315,6 +2315,7 @@ M: Gerd Hoffmann R: Manos Pitsidianakis S: Supported F: hw/audio/virtio-snd.c +F: hw/audio/virtio-snd-pci.c F: include/hw/audio/virtio-snd.h nvme diff --git a/hw/audio/meson.build b/hw/audio/meson.build index 7a503be1fd..2990974449 100644 --- a/hw/audio/meson.build +++ b/hw/audio/meson.build @@ -14,3 +14,4 @@ system_ss.add(when: 'CONFIG_SB16', if_true: files('sb16.c')) system_ss.add(when: 'CONFIG_VT82C686', if_true: files('via-ac97.c')) system_ss.add(when: 'CONFIG_WM8750', if_true: files('wm8750.c')) system_ss.add(when: ['CONFIG_VIRTIO_SND', 'CONFIG_VIRTIO'], if_true: files('virtio-snd.c')) +system_ss.add(when: ['CONFIG_VIRTIO_SND', 'CONFIG_VIRTIO', 'CONFIG_VIRTIO_PCI'], if_true: files('virtio-snd-pci.c')) diff --git a/hw/audio/virtio-snd-pci.c b/hw/audio/virtio-snd-pci.c new file mode 100644 index 0000000000..0f92e0752b --- /dev/null +++ b/hw/audio/virtio-snd-pci.c @@ -0,0 +1,93 @@ +/* + * VIRTIO Sound Device PCI Bindings + * + * Copyright (c) 2023 Emmanouil Pitsidianakis + * + * This work is licensed under the terms of the GNU GPL, version 2 or + * (at your option) any later version. See the COPYING file in the + * top-level directory. + */ + +#include "qemu/osdep.h" +#include "qom/object.h" +#include "qapi/error.h" +#include "hw/audio/soundhw.h" +#include "hw/virtio/virtio-pci.h" +#include "hw/audio/virtio-snd.h" + +/* + * virtio-snd-pci: This extends VirtioPCIProxy. + */ +#define TYPE_VIRTIO_SND_PCI "virtio-sound-pci" +OBJECT_DECLARE_SIMPLE_TYPE(VirtIOSoundPCI, VIRTIO_SND_PCI) + +struct VirtIOSoundPCI { + VirtIOPCIProxy parent_obj; + + VirtIOSound vdev; +}; + +static Property virtio_snd_pci_properties[] = { + DEFINE_PROP_BIT("ioeventfd", VirtIOPCIProxy, flags, + VIRTIO_PCI_FLAG_USE_IOEVENTFD_BIT, true), + DEFINE_PROP_UINT32("vectors", VirtIOPCIProxy, nvectors, 2), + DEFINE_PROP_END_OF_LIST(), +}; + +static void virtio_snd_pci_realize(VirtIOPCIProxy *vpci_dev, Error **errp) +{ + VirtIOSoundPCI *dev = VIRTIO_SND_PCI(vpci_dev); + DeviceState *vdev = DEVICE(&dev->vdev); + + virtio_pci_force_virtio_1(vpci_dev); + qdev_realize(vdev, BUS(&vpci_dev->bus), errp); +} + +static void virtio_snd_pci_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + VirtioPCIClass *vpciklass = VIRTIO_PCI_CLASS(klass); + + device_class_set_props(dc, virtio_snd_pci_properties); + dc->desc = "Virtio Sound"; + set_bit(DEVICE_CATEGORY_SOUND, dc->categories); + + vpciklass->realize = virtio_snd_pci_realize; +} + +static void virtio_snd_pci_instance_init(Object *obj) +{ + VirtIOSoundPCI *dev = VIRTIO_SND_PCI(obj); + + virtio_instance_init_common(obj, &dev->vdev, sizeof(dev->vdev), + TYPE_VIRTIO_SND); +} + +static const VirtioPCIDeviceTypeInfo virtio_snd_pci_info = { + .generic_name = TYPE_VIRTIO_SND_PCI, + .instance_size = sizeof(VirtIOSoundPCI), + .instance_init = virtio_snd_pci_instance_init, + .class_init = virtio_snd_pci_class_init, +}; + +/* Create a Virtio Sound PCI device, so '-audio driver,model=virtio' works. */ +static int virtio_snd_pci_init(PCIBus *bus, const char *audiodev) +{ + DeviceState *vdev = NULL; + VirtIOSoundPCI *dev = NULL; + + vdev = qdev_new(TYPE_VIRTIO_SND_PCI); + assert(vdev); + dev = VIRTIO_SND_PCI(vdev); + qdev_prop_set_string(DEVICE(&dev->vdev), "audiodev", audiodev); + qdev_realize_and_unref(vdev, BUS(bus), &error_fatal); + return 0; +} + +static void virtio_snd_pci_register(void) +{ + virtio_pci_types_register(&virtio_snd_pci_info); + pci_register_soundhw("virtio", "Virtio Sound", virtio_snd_pci_init); +} + +type_init(virtio_snd_pci_register); diff --git a/system/qdev-monitor.c b/system/qdev-monitor.c index 3c0aabec4b..a13db763e5 100644 --- a/system/qdev-monitor.c +++ b/system/qdev-monitor.c @@ -112,6 +112,7 @@ static const QDevAlias qdev_alias_table[] = { { "virtio-serial-ccw", "virtio-serial", QEMU_ARCH_VIRTIO_CCW }, { "virtio-serial-pci", "virtio-serial", QEMU_ARCH_VIRTIO_PCI}, { "virtio-sound-device", "virtio-sound", QEMU_ARCH_VIRTIO_MMIO }, + { "virtio-sound-pci", "virtio-sound", QEMU_ARCH_VIRTIO_PCI }, { "virtio-tablet-device", "virtio-tablet", QEMU_ARCH_VIRTIO_MMIO }, { "virtio-tablet-ccw", "virtio-tablet", QEMU_ARCH_VIRTIO_CCW }, { "virtio-tablet-pci", "virtio-tablet", QEMU_ARCH_VIRTIO_PCI }, -- cgit v1.2.3 From eb9ad377bb94d2d98d18cfccc431a028361385c6 Mon Sep 17 00:00:00 2001 From: Manos Pitsidianakis Date: Mon, 23 Oct 2023 15:03:20 +0300 Subject: virtio-sound: handle control messages and streams MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Receive guest requests in the control (CTRL) queue of the virtio sound device and reply with a NOT SUPPORTED error to all control commands. The receiving handler is virtio_snd_handle_ctrl(). It stores all control messages in the queue in the device's command queue. Then it calls virtio_snd_process_cmdq() to handle each message. The handler is process_cmd() which replies with VIRTIO_SND_S_NOT_SUPP. Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin Signed-off-by: Anton Yakovlev Signed-off-by: Manos Pitsidianakis Reviewed-by: Alex Bennée Tested-by: Alex Bennée Message-Id: <3224aff87e7c4f2777bfe1bbbbca93b72525992c.1698062525.git.manos.pitsidianakis@linaro.org> Acked-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/audio/trace-events | 4 + hw/audio/virtio-snd.c | 487 +++++++++++++++++++++++++++++++++++++++++- include/hw/audio/virtio-snd.h | 113 +++++++++- 3 files changed, 595 insertions(+), 9 deletions(-) diff --git a/hw/audio/trace-events b/hw/audio/trace-events index 525ced2b34..122d1403ef 100644 --- a/hw/audio/trace-events +++ b/hw/audio/trace-events @@ -47,3 +47,7 @@ virtio_snd_vm_state_running(void) "vm state running" virtio_snd_vm_state_stopped(void) "vm state stopped" virtio_snd_realize(void *snd) "snd %p: realize" virtio_snd_unrealize(void *snd) "snd %p: unrealize" +virtio_snd_handle_ctrl(void *vdev, void *vq) "snd %p: handle ctrl event for queue %p" +virtio_snd_handle_code(uint32_t val, const char *code) "ctrl code msg val = %"PRIu32" == %s" +virtio_snd_handle_chmap_info(void) "VIRTIO_SND_CHMAP_INFO called" +virtio_snd_handle_event(void) "event queue callback called" diff --git a/hw/audio/virtio-snd.c b/hw/audio/virtio-snd.c index 3a4b441c20..4eae76f638 100644 --- a/hw/audio/virtio-snd.c +++ b/hw/audio/virtio-snd.c @@ -32,6 +32,29 @@ #define VIRTIO_SOUND_CHMAP_DEFAULT 0 #define VIRTIO_SOUND_HDA_FN_NID 0 +static uint32_t supported_formats = BIT(VIRTIO_SND_PCM_FMT_S8) + | BIT(VIRTIO_SND_PCM_FMT_U8) + | BIT(VIRTIO_SND_PCM_FMT_S16) + | BIT(VIRTIO_SND_PCM_FMT_U16) + | BIT(VIRTIO_SND_PCM_FMT_S32) + | BIT(VIRTIO_SND_PCM_FMT_U32) + | BIT(VIRTIO_SND_PCM_FMT_FLOAT); + +static uint32_t supported_rates = BIT(VIRTIO_SND_PCM_RATE_5512) + | BIT(VIRTIO_SND_PCM_RATE_8000) + | BIT(VIRTIO_SND_PCM_RATE_11025) + | BIT(VIRTIO_SND_PCM_RATE_16000) + | BIT(VIRTIO_SND_PCM_RATE_22050) + | BIT(VIRTIO_SND_PCM_RATE_32000) + | BIT(VIRTIO_SND_PCM_RATE_44100) + | BIT(VIRTIO_SND_PCM_RATE_48000) + | BIT(VIRTIO_SND_PCM_RATE_64000) + | BIT(VIRTIO_SND_PCM_RATE_88200) + | BIT(VIRTIO_SND_PCM_RATE_96000) + | BIT(VIRTIO_SND_PCM_RATE_176400) + | BIT(VIRTIO_SND_PCM_RATE_192000) + | BIT(VIRTIO_SND_PCM_RATE_384000); + static const VMStateDescription vmstate_virtio_snd_device = { .name = TYPE_VIRTIO_SND, .version_id = VIRTIO_SOUND_VM_VERSION, @@ -100,13 +123,397 @@ virtio_snd_set_config(VirtIODevice *vdev, const uint8_t *config) } +static void +virtio_snd_ctrl_cmd_free(virtio_snd_ctrl_command *cmd) +{ + g_free(cmd->elem); + g_free(cmd); +} + +/* + * Get a specific stream from the virtio sound card device. + * Returns NULL if @stream_id is invalid or not allocated. + * + * @s: VirtIOSound device + * @stream_id: stream id + */ +static VirtIOSoundPCMStream *virtio_snd_pcm_get_stream(VirtIOSound *s, + uint32_t stream_id) +{ + return stream_id >= s->snd_conf.streams ? NULL : + s->pcm->streams[stream_id]; +} + +/* + * Get params for a specific stream. + * + * @s: VirtIOSound device + * @stream_id: stream id + */ +static virtio_snd_pcm_set_params *virtio_snd_pcm_get_params(VirtIOSound *s, + uint32_t stream_id) +{ + return stream_id >= s->snd_conf.streams ? NULL + : &s->pcm->pcm_params[stream_id]; +} + +/* + * Set the given stream params. + * Called by both virtio_snd_handle_pcm_set_params and during device + * initialization. + * Returns the response status code. (VIRTIO_SND_S_*). + * + * @s: VirtIOSound device + * @params: The PCM params as defined in the virtio specification + */ +static +uint32_t virtio_snd_set_pcm_params(VirtIOSound *s, + uint32_t stream_id, + virtio_snd_pcm_set_params *params) +{ + virtio_snd_pcm_set_params *st_params; + + if (stream_id >= s->snd_conf.streams || s->pcm->pcm_params == NULL) { + /* + * TODO: do we need to set DEVICE_NEEDS_RESET? + */ + virtio_error(VIRTIO_DEVICE(s), "Streams have not been initialized.\n"); + return cpu_to_le32(VIRTIO_SND_S_BAD_MSG); + } + + st_params = virtio_snd_pcm_get_params(s, stream_id); + + if (params->channels < 1 || params->channels > AUDIO_MAX_CHANNELS) { + error_report("Number of channels is not supported."); + return cpu_to_le32(VIRTIO_SND_S_NOT_SUPP); + } + if (!(supported_formats & BIT(params->format))) { + error_report("Stream format is not supported."); + return cpu_to_le32(VIRTIO_SND_S_NOT_SUPP); + } + if (!(supported_rates & BIT(params->rate))) { + error_report("Stream rate is not supported."); + return cpu_to_le32(VIRTIO_SND_S_NOT_SUPP); + } + + st_params->buffer_bytes = le32_to_cpu(params->buffer_bytes); + st_params->period_bytes = le32_to_cpu(params->period_bytes); + st_params->features = le32_to_cpu(params->features); + /* the following are uint8_t, so there's no need to bswap the values. */ + st_params->channels = params->channels; + st_params->format = params->format; + st_params->rate = params->rate; + + return cpu_to_le32(VIRTIO_SND_S_OK); +} + +/* + * Get a QEMU Audiosystem compatible format value from a VIRTIO_SND_PCM_FMT_* + */ +static AudioFormat virtio_snd_get_qemu_format(uint32_t format) +{ + #define CASE(FMT) \ + case VIRTIO_SND_PCM_FMT_##FMT: \ + return AUDIO_FORMAT_##FMT; + + switch (format) { + CASE(U8) + CASE(S8) + CASE(U16) + CASE(S16) + CASE(U32) + CASE(S32) + case VIRTIO_SND_PCM_FMT_FLOAT: + return AUDIO_FORMAT_F32; + default: + g_assert_not_reached(); + } + + #undef CASE +} + +/* + * Get a QEMU Audiosystem compatible frequency value from a + * VIRTIO_SND_PCM_RATE_* + */ +static uint32_t virtio_snd_get_qemu_freq(uint32_t rate) +{ + #define CASE(RATE) \ + case VIRTIO_SND_PCM_RATE_##RATE: \ + return RATE; + + switch (rate) { + CASE(5512) + CASE(8000) + CASE(11025) + CASE(16000) + CASE(22050) + CASE(32000) + CASE(44100) + CASE(48000) + CASE(64000) + CASE(88200) + CASE(96000) + CASE(176400) + CASE(192000) + CASE(384000) + default: + g_assert_not_reached(); + } + + #undef CASE +} + +/* + * Get QEMU Audiosystem compatible audsettings from virtio based pcm stream + * params. + */ +static void virtio_snd_get_qemu_audsettings(audsettings *as, + virtio_snd_pcm_set_params *params) +{ + as->nchannels = MIN(AUDIO_MAX_CHANNELS, params->channels); + as->fmt = virtio_snd_get_qemu_format(params->format); + as->freq = virtio_snd_get_qemu_freq(params->rate); + as->endianness = target_words_bigendian() ? 1 : 0; +} + +/* + * Close a stream and free all its resources. + * + * @stream: VirtIOSoundPCMStream *stream + */ +static void virtio_snd_pcm_close(VirtIOSoundPCMStream *stream) +{ +} + /* - * Queue handler stub. + * Prepares a VirtIOSound card stream. + * Returns the response status code. (VIRTIO_SND_S_*). + * + * @s: VirtIOSound device + * @stream_id: stream id + */ +static uint32_t virtio_snd_pcm_prepare(VirtIOSound *s, uint32_t stream_id) +{ + audsettings as; + virtio_snd_pcm_set_params *params; + VirtIOSoundPCMStream *stream; + + if (s->pcm->streams == NULL || + s->pcm->pcm_params == NULL || + stream_id >= s->snd_conf.streams) { + return cpu_to_le32(VIRTIO_SND_S_BAD_MSG); + } + + params = virtio_snd_pcm_get_params(s, stream_id); + if (params == NULL) { + return cpu_to_le32(VIRTIO_SND_S_BAD_MSG); + } + + stream = virtio_snd_pcm_get_stream(s, stream_id); + if (stream == NULL) { + stream = g_new0(VirtIOSoundPCMStream, 1); + stream->active = false; + stream->id = stream_id; + stream->pcm = s->pcm; + stream->s = s; + + /* + * stream_id >= s->snd_conf.streams was checked before so this is + * in-bounds + */ + s->pcm->streams[stream_id] = stream; + } + + virtio_snd_get_qemu_audsettings(&as, params); + stream->info.direction = stream_id < s->snd_conf.streams / 2 + + (s->snd_conf.streams & 1) ? VIRTIO_SND_D_OUTPUT : VIRTIO_SND_D_INPUT; + stream->info.hdr.hda_fn_nid = VIRTIO_SOUND_HDA_FN_NID; + stream->info.features = 0; + stream->info.channels_min = 1; + stream->info.channels_max = as.nchannels; + stream->info.formats = supported_formats; + stream->info.rates = supported_rates; + stream->params = *params; + + stream->positions[0] = VIRTIO_SND_CHMAP_FL; + stream->positions[1] = VIRTIO_SND_CHMAP_FR; + stream->as = as; + + return cpu_to_le32(VIRTIO_SND_S_OK); +} + +static const char *print_code(uint32_t code) +{ + #define CASE(CODE) \ + case VIRTIO_SND_R_##CODE: \ + return "VIRTIO_SND_R_"#CODE + + switch (code) { + CASE(JACK_INFO); + CASE(JACK_REMAP); + CASE(PCM_INFO); + CASE(PCM_SET_PARAMS); + CASE(PCM_PREPARE); + CASE(PCM_RELEASE); + CASE(PCM_START); + CASE(PCM_STOP); + CASE(CHMAP_INFO); + default: + return "invalid code"; + } + + #undef CASE +}; + +/* + * The actual processing done in virtio_snd_process_cmdq(). + * + * @s: VirtIOSound device + * @cmd: control command request + */ +static inline void +process_cmd(VirtIOSound *s, virtio_snd_ctrl_command *cmd) +{ + uint32_t code; + size_t msg_sz = iov_to_buf(cmd->elem->out_sg, + cmd->elem->out_num, + 0, + &cmd->ctrl, + sizeof(virtio_snd_hdr)); + + if (msg_sz != sizeof(virtio_snd_hdr)) { + /* + * TODO: do we need to set DEVICE_NEEDS_RESET? + */ + qemu_log_mask(LOG_GUEST_ERROR, + "%s: virtio-snd command size incorrect %zu vs \ + %zu\n", __func__, msg_sz, sizeof(virtio_snd_hdr)); + return; + } + + code = le32_to_cpu(cmd->ctrl.code); + + trace_virtio_snd_handle_code(code, print_code(code)); + + switch (code) { + case VIRTIO_SND_R_JACK_INFO: + case VIRTIO_SND_R_JACK_REMAP: + qemu_log_mask(LOG_UNIMP, + "virtio_snd: jack functionality is unimplemented.\n"); + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_NOT_SUPP); + break; + case VIRTIO_SND_R_PCM_INFO: + case VIRTIO_SND_R_PCM_SET_PARAMS: + case VIRTIO_SND_R_PCM_PREPARE: + case VIRTIO_SND_R_PCM_START: + case VIRTIO_SND_R_PCM_STOP: + case VIRTIO_SND_R_PCM_RELEASE: + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_NOT_SUPP); + break; + case VIRTIO_SND_R_CHMAP_INFO: + qemu_log_mask(LOG_UNIMP, + "virtio_snd: chmap info functionality is unimplemented.\n"); + trace_virtio_snd_handle_chmap_info(); + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_NOT_SUPP); + break; + default: + /* error */ + error_report("virtio snd header not recognized: %"PRIu32, code); + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_BAD_MSG); + } + + iov_from_buf(cmd->elem->in_sg, + cmd->elem->in_num, + 0, + &cmd->resp, + sizeof(virtio_snd_hdr)); + virtqueue_push(cmd->vq, cmd->elem, sizeof(virtio_snd_hdr)); + virtio_notify(VIRTIO_DEVICE(s), cmd->vq); +} + +/* + * Consume all elements in command queue. + * + * @s: VirtIOSound device + */ +static void virtio_snd_process_cmdq(VirtIOSound *s) +{ + virtio_snd_ctrl_command *cmd; + + if (unlikely(qatomic_read(&s->processing_cmdq))) { + return; + } + + WITH_QEMU_LOCK_GUARD(&s->cmdq_mutex) { + qatomic_set(&s->processing_cmdq, true); + while (!QTAILQ_EMPTY(&s->cmdq)) { + cmd = QTAILQ_FIRST(&s->cmdq); + + /* process command */ + process_cmd(s, cmd); + + QTAILQ_REMOVE(&s->cmdq, cmd, next); + + virtio_snd_ctrl_cmd_free(cmd); + } + qatomic_set(&s->processing_cmdq, false); + } +} + +/* + * The control message handler. Pops an element from the control virtqueue, + * and stores them to VirtIOSound's cmdq queue and finally calls + * virtio_snd_process_cmdq() for processing. + * + * @vdev: VirtIOSound device + * @vq: Control virtqueue + */ +static void virtio_snd_handle_ctrl(VirtIODevice *vdev, VirtQueue *vq) +{ + VirtIOSound *s = VIRTIO_SND(vdev); + VirtQueueElement *elem; + virtio_snd_ctrl_command *cmd; + + trace_virtio_snd_handle_ctrl(vdev, vq); + + if (!virtio_queue_ready(vq)) { + return; + } + + elem = virtqueue_pop(vq, sizeof(VirtQueueElement)); + while (elem) { + cmd = g_new0(virtio_snd_ctrl_command, 1); + cmd->elem = elem; + cmd->vq = vq; + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_OK); + QTAILQ_INSERT_TAIL(&s->cmdq, cmd, next); + elem = virtqueue_pop(vq, sizeof(VirtQueueElement)); + } + + virtio_snd_process_cmdq(s); +} + +/* + * The event virtqueue handler. + * Not implemented yet. + * + * @vdev: VirtIOSound device + * @vq: event vq + */ +static void virtio_snd_handle_event(VirtIODevice *vdev, VirtQueue *vq) +{ + qemu_log_mask(LOG_UNIMP, "virtio_snd: event queue is unimplemented.\n"); + trace_virtio_snd_handle_event(); +} + +/* + * Stub buffer virtqueue handler. * * @vdev: VirtIOSound device * @vq: virtqueue */ -static void virtio_snd_handle_queue(VirtIODevice *vdev, VirtQueue *vq) {} +static void virtio_snd_handle_xfer(VirtIODevice *vdev, VirtQueue *vq) {} static uint64_t get_features(VirtIODevice *vdev, uint64_t features, Error **errp) @@ -140,12 +547,22 @@ static void virtio_snd_realize(DeviceState *dev, Error **errp) ERRP_GUARD(); VirtIOSound *vsnd = VIRTIO_SND(dev); VirtIODevice *vdev = VIRTIO_DEVICE(dev); + virtio_snd_pcm_set_params default_params = { 0 }; + uint32_t status; + vsnd->pcm = NULL; vsnd->vmstate = qemu_add_vm_change_state_handler(virtio_snd_vm_state_change, vsnd); trace_virtio_snd_realize(vsnd); + vsnd->pcm = g_new0(VirtIOSoundPCM, 1); + vsnd->pcm->snd = vsnd; + vsnd->pcm->streams = + g_new0(VirtIOSoundPCMStream *, vsnd->snd_conf.streams); + vsnd->pcm->pcm_params = + g_new0(virtio_snd_pcm_set_params, vsnd->snd_conf.streams); + virtio_init(vdev, VIRTIO_ID_SOUND, sizeof(virtio_snd_config)); virtio_add_feature(&vsnd->features, VIRTIO_F_VERSION_1); @@ -172,25 +589,69 @@ static void virtio_snd_realize(DeviceState *dev, Error **errp) AUD_register_card("virtio-sound", &vsnd->card, errp); + /* set default params for all streams */ + default_params.features = 0; + default_params.buffer_bytes = cpu_to_le32(8192); + default_params.period_bytes = cpu_to_le32(2048); + default_params.channels = 2; + default_params.format = VIRTIO_SND_PCM_FMT_S16; + default_params.rate = VIRTIO_SND_PCM_RATE_48000; vsnd->queues[VIRTIO_SND_VQ_CONTROL] = - virtio_add_queue(vdev, 64, virtio_snd_handle_queue); + virtio_add_queue(vdev, 64, virtio_snd_handle_ctrl); vsnd->queues[VIRTIO_SND_VQ_EVENT] = - virtio_add_queue(vdev, 64, virtio_snd_handle_queue); + virtio_add_queue(vdev, 64, virtio_snd_handle_event); vsnd->queues[VIRTIO_SND_VQ_TX] = - virtio_add_queue(vdev, 64, virtio_snd_handle_queue); + virtio_add_queue(vdev, 64, virtio_snd_handle_xfer); vsnd->queues[VIRTIO_SND_VQ_RX] = - virtio_add_queue(vdev, 64, virtio_snd_handle_queue); + virtio_add_queue(vdev, 64, virtio_snd_handle_xfer); + qemu_mutex_init(&vsnd->cmdq_mutex); + QTAILQ_INIT(&vsnd->cmdq); + + for (uint32_t i = 0; i < vsnd->snd_conf.streams; i++) { + status = virtio_snd_set_pcm_params(vsnd, i, &default_params); + if (status != cpu_to_le32(VIRTIO_SND_S_OK)) { + error_setg(errp, + "Can't initalize stream params, device responded with %s.", + print_code(status)); + return; + } + status = virtio_snd_pcm_prepare(vsnd, i); + if (status != cpu_to_le32(VIRTIO_SND_S_OK)) { + error_setg(errp, + "Can't prepare streams, device responded with %s.", + print_code(status)); + return; + } + } } static void virtio_snd_unrealize(DeviceState *dev) { VirtIODevice *vdev = VIRTIO_DEVICE(dev); VirtIOSound *vsnd = VIRTIO_SND(dev); + VirtIOSoundPCMStream *stream; qemu_del_vm_change_state_handler(vsnd->vmstate); trace_virtio_snd_unrealize(vsnd); + if (vsnd->pcm) { + if (vsnd->pcm->streams) { + for (uint32_t i = 0; i < vsnd->snd_conf.streams; i++) { + stream = vsnd->pcm->streams[i]; + if (stream) { + virtio_snd_process_cmdq(stream->s); + virtio_snd_pcm_close(stream); + g_free(stream); + } + } + g_free(vsnd->pcm->streams); + } + g_free(vsnd->pcm->pcm_params); + g_free(vsnd->pcm); + vsnd->pcm = NULL; + } AUD_remove_card(&vsnd->card); + qemu_mutex_destroy(&vsnd->cmdq_mutex); virtio_delete_queue(vsnd->queues[VIRTIO_SND_VQ_CONTROL]); virtio_delete_queue(vsnd->queues[VIRTIO_SND_VQ_EVENT]); virtio_delete_queue(vsnd->queues[VIRTIO_SND_VQ_TX]); @@ -199,7 +660,19 @@ static void virtio_snd_unrealize(DeviceState *dev) } -static void virtio_snd_reset(VirtIODevice *vdev) {} +static void virtio_snd_reset(VirtIODevice *vdev) +{ + VirtIOSound *s = VIRTIO_SND(vdev); + virtio_snd_ctrl_command *cmd; + + WITH_QEMU_LOCK_GUARD(&s->cmdq_mutex) { + while (!QTAILQ_EMPTY(&s->cmdq)) { + cmd = QTAILQ_FIRST(&s->cmdq); + QTAILQ_REMOVE(&s->cmdq, cmd, next); + virtio_snd_ctrl_cmd_free(cmd); + } + } +} static void virtio_snd_class_init(ObjectClass *klass, void *data) { diff --git a/include/hw/audio/virtio-snd.h b/include/hw/audio/virtio-snd.h index d08065941c..cc14c875ed 100644 --- a/include/hw/audio/virtio-snd.h +++ b/include/hw/audio/virtio-snd.h @@ -67,13 +67,122 @@ typedef struct virtio_snd_pcm_xfer virtio_snd_pcm_xfer; /* I/O request status */ typedef struct virtio_snd_pcm_status virtio_snd_pcm_status; -typedef struct VirtIOSound { +/* device structs */ + +typedef struct VirtIOSound VirtIOSound; + +typedef struct VirtIOSoundPCMStream VirtIOSoundPCMStream; + +typedef struct virtio_snd_ctrl_command virtio_snd_ctrl_command; + +typedef struct VirtIOSoundPCM VirtIOSoundPCM; + +struct VirtIOSoundPCM { + VirtIOSound *snd; + /* + * PCM parameters are a separate field instead of a VirtIOSoundPCMStream + * field, because the operation of PCM control requests is first + * VIRTIO_SND_R_PCM_SET_PARAMS and then VIRTIO_SND_R_PCM_PREPARE; this + * means that some times we get parameters without having an allocated + * stream yet. + */ + virtio_snd_pcm_set_params *pcm_params; + VirtIOSoundPCMStream **streams; +}; + +struct VirtIOSoundPCMStream { + VirtIOSoundPCM *pcm; + virtio_snd_pcm_info info; + virtio_snd_pcm_set_params params; + uint32_t id; + /* channel position values (VIRTIO_SND_CHMAP_XXX) */ + uint8_t positions[VIRTIO_SND_CHMAP_MAX_SIZE]; + VirtIOSound *s; + bool flushing; + audsettings as; + union { + SWVoiceIn *in; + SWVoiceOut *out; + } voice; + bool active; +}; + +/* + * PCM stream state machine. + * ------------------------- + * + * 5.14.6.6.1 PCM Command Lifecycle + * ================================ + * + * A PCM stream has the following command lifecycle: + * - `SET PARAMETERS` + * The driver negotiates the stream parameters (format, transport, etc) with + * the device. + * Possible valid transitions: `SET PARAMETERS`, `PREPARE`. + * - `PREPARE` + * The device prepares the stream (allocates resources, etc). + * Possible valid transitions: `SET PARAMETERS`, `PREPARE`, `START`, + * `RELEASE`. Output only: the driver transfers data for pre-buffing. + * - `START` + * The device starts the stream (unmute, putting into running state, etc). + * Possible valid transitions: `STOP`. + * The driver transfers data to/from the stream. + * - `STOP` + * The device stops the stream (mute, putting into non-running state, etc). + * Possible valid transitions: `START`, `RELEASE`. + * - `RELEASE` + * The device releases the stream (frees resources, etc). + * Possible valid transitions: `SET PARAMETERS`, `PREPARE`. + * + * +---------------+ +---------+ +---------+ +-------+ +-------+ + * | SetParameters | | Prepare | | Release | | Start | | Stop | + * +---------------+ +---------+ +---------+ +-------+ +-------+ + * |- | | | | + * || | | | | + * |< | | | | + * |------------->| | | | + * |<-------------| | | | + * | |- | | | + * | || | | | + * | |< | | | + * | |--------------------->| | + * | |---------->| | | + * | | | |-------->| + * | | | |<--------| + * | | |<-------------------| + * |<-------------------------| | | + * | |<----------| | | + * + * CTRL in the VirtIOSound device + * ============================== + * + * The control messages that affect the state of a stream arrive in the + * `virtio_snd_handle_ctrl()` queue callback and are of type `struct + * virtio_snd_ctrl_command`. They are stored in a queue field in the device + * type, `VirtIOSound`. This allows deferring the CTRL request completion if + * it's not immediately possible due to locking/state reasons. + * + * The CTRL message is finally handled in `process_cmd()`. + */ +struct VirtIOSound { VirtIODevice parent_obj; VirtQueue *queues[VIRTIO_SND_VQ_MAX]; uint64_t features; + VirtIOSoundPCM *pcm; QEMUSoundCard card; VMChangeStateEntry *vmstate; virtio_snd_config snd_conf; -} VirtIOSound; + QemuMutex cmdq_mutex; + QTAILQ_HEAD(, virtio_snd_ctrl_command) cmdq; + bool processing_cmdq; +}; + +struct virtio_snd_ctrl_command { + VirtQueueElement *elem; + VirtQueue *vq; + virtio_snd_hdr ctrl; + virtio_snd_hdr resp; + QTAILQ_ENTRY(virtio_snd_ctrl_command) next; +}; #endif -- cgit v1.2.3 From 0ff05dd209f153f037a48e4039a033885b9ab5e3 Mon Sep 17 00:00:00 2001 From: Manos Pitsidianakis Date: Mon, 23 Oct 2023 15:03:21 +0300 Subject: virtio-sound: handle VIRTIO_SND_R_PCM_INFO request MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Respond to the VIRTIO_SND_R_PCM_INFO control request with the parameters of each requested PCM stream. Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin Signed-off-by: Anton Yakovlev Signed-off-by: Manos Pitsidianakis Reviewed-by: Alex Bennée Tested-by: Alex Bennée Message-Id: <5ecea6ba2fb0e3957d7d90bc4dbac521a3d1f678.1698062525.git.manos.pitsidianakis@linaro.org> Acked-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/audio/trace-events | 1 + hw/audio/virtio-snd.c | 82 +++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 83 insertions(+) diff --git a/hw/audio/trace-events b/hw/audio/trace-events index 122d1403ef..6def414f96 100644 --- a/hw/audio/trace-events +++ b/hw/audio/trace-events @@ -48,6 +48,7 @@ virtio_snd_vm_state_stopped(void) "vm state stopped" virtio_snd_realize(void *snd) "snd %p: realize" virtio_snd_unrealize(void *snd) "snd %p: unrealize" virtio_snd_handle_ctrl(void *vdev, void *vq) "snd %p: handle ctrl event for queue %p" +virtio_snd_handle_pcm_info(uint32_t stream) "VIRTIO_SND_R_PCM_INFO called for stream %"PRIu32 virtio_snd_handle_code(uint32_t val, const char *code) "ctrl code msg val = %"PRIu32" == %s" virtio_snd_handle_chmap_info(void) "VIRTIO_SND_CHMAP_INFO called" virtio_snd_handle_event(void) "event queue callback called" diff --git a/hw/audio/virtio-snd.c b/hw/audio/virtio-snd.c index 4eae76f638..fd0c50de6e 100644 --- a/hw/audio/virtio-snd.c +++ b/hw/audio/virtio-snd.c @@ -157,6 +157,86 @@ static virtio_snd_pcm_set_params *virtio_snd_pcm_get_params(VirtIOSound *s, : &s->pcm->pcm_params[stream_id]; } +/* + * Handle the VIRTIO_SND_R_PCM_INFO request. + * The function writes the info structs to the request element. + * + * @s: VirtIOSound device + * @cmd: The request command queue element from VirtIOSound cmdq field + */ +static void virtio_snd_handle_pcm_info(VirtIOSound *s, + virtio_snd_ctrl_command *cmd) +{ + uint32_t stream_id, start_id, count, size; + virtio_snd_pcm_info val; + virtio_snd_query_info req; + VirtIOSoundPCMStream *stream = NULL; + g_autofree virtio_snd_pcm_info *pcm_info = NULL; + size_t msg_sz = iov_to_buf(cmd->elem->out_sg, + cmd->elem->out_num, + 0, + &req, + sizeof(virtio_snd_query_info)); + + if (msg_sz != sizeof(virtio_snd_query_info)) { + /* + * TODO: do we need to set DEVICE_NEEDS_RESET? + */ + qemu_log_mask(LOG_GUEST_ERROR, + "%s: virtio-snd command size incorrect %zu vs \ + %zu\n", __func__, msg_sz, sizeof(virtio_snd_query_info)); + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_BAD_MSG); + return; + } + + start_id = le32_to_cpu(req.start_id); + count = le32_to_cpu(req.count); + size = le32_to_cpu(req.size); + + if (iov_size(cmd->elem->in_sg, cmd->elem->in_num) < + sizeof(virtio_snd_hdr) + size * count) { + /* + * TODO: do we need to set DEVICE_NEEDS_RESET? + */ + error_report("pcm info: buffer too small, got: %zu, needed: %zu", + iov_size(cmd->elem->in_sg, cmd->elem->in_num), + sizeof(virtio_snd_pcm_info)); + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_BAD_MSG); + return; + } + + pcm_info = g_new0(virtio_snd_pcm_info, count); + for (uint32_t i = 0; i < count; i++) { + stream_id = i + start_id; + trace_virtio_snd_handle_pcm_info(stream_id); + stream = virtio_snd_pcm_get_stream(s, stream_id); + if (!stream) { + error_report("Invalid stream id: %"PRIu32, stream_id); + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_BAD_MSG); + return; + } + val = stream->info; + val.hdr.hda_fn_nid = cpu_to_le32(val.hdr.hda_fn_nid); + val.features = cpu_to_le32(val.features); + val.formats = cpu_to_le64(val.formats); + val.rates = cpu_to_le64(val.rates); + /* + * 5.14.6.6.2.1 Device Requirements: Stream Information The device MUST + * NOT set undefined feature, format, rate and direction values. The + * device MUST initialize the padding bytes to 0. + */ + pcm_info[i] = val; + memset(&pcm_info[i].padding, 0, 5); + } + + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_OK); + iov_from_buf(cmd->elem->in_sg, + cmd->elem->in_num, + sizeof(virtio_snd_hdr), + pcm_info, + sizeof(virtio_snd_pcm_info) * count); +} + /* * Set the given stream params. * Called by both virtio_snd_handle_pcm_set_params and during device @@ -404,6 +484,8 @@ process_cmd(VirtIOSound *s, virtio_snd_ctrl_command *cmd) cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_NOT_SUPP); break; case VIRTIO_SND_R_PCM_INFO: + virtio_snd_handle_pcm_info(s, cmd); + break; case VIRTIO_SND_R_PCM_SET_PARAMS: case VIRTIO_SND_R_PCM_PREPARE: case VIRTIO_SND_R_PCM_START: -- cgit v1.2.3 From fa131d4a8277ef3522384d33d3b6b7963b5e7de9 Mon Sep 17 00:00:00 2001 From: Manos Pitsidianakis Date: Mon, 23 Oct 2023 15:03:22 +0300 Subject: virtio-sound: handle VIRTIO_SND_R_PCM_{START,STOP} MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Handle the start and stop control messages for a stream_id. This request does nothing at the moment except for replying to it. Audio playback or capture will be started/stopped here in follow-up commits. Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin Signed-off-by: Anton Yakovlev Signed-off-by: Manos Pitsidianakis Reviewed-by: Alex Bennée Tested-by: Alex Bennée Message-Id: <9657dbfe3cb4a48ceb033ceb5977dc08669dfefd.1698062525.git.manos.pitsidianakis@linaro.org> Acked-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/audio/trace-events | 1 + hw/audio/virtio-snd.c | 49 +++++++++++++++++++++++++++++++++++++++++++++++-- 2 files changed, 48 insertions(+), 2 deletions(-) diff --git a/hw/audio/trace-events b/hw/audio/trace-events index 6def414f96..db48ff04fe 100644 --- a/hw/audio/trace-events +++ b/hw/audio/trace-events @@ -49,6 +49,7 @@ virtio_snd_realize(void *snd) "snd %p: realize" virtio_snd_unrealize(void *snd) "snd %p: unrealize" virtio_snd_handle_ctrl(void *vdev, void *vq) "snd %p: handle ctrl event for queue %p" virtio_snd_handle_pcm_info(uint32_t stream) "VIRTIO_SND_R_PCM_INFO called for stream %"PRIu32 +virtio_snd_handle_pcm_start_stop(const char *code, uint32_t stream) "%s called for stream %"PRIu32 virtio_snd_handle_code(uint32_t val, const char *code) "ctrl code msg val = %"PRIu32" == %s" virtio_snd_handle_chmap_info(void) "VIRTIO_SND_CHMAP_INFO called" virtio_snd_handle_event(void) "event queue callback called" diff --git a/hw/audio/virtio-snd.c b/hw/audio/virtio-snd.c index fd0c50de6e..e6791de6c6 100644 --- a/hw/audio/virtio-snd.c +++ b/hw/audio/virtio-snd.c @@ -446,6 +446,47 @@ static const char *print_code(uint32_t code) #undef CASE }; +/* + * Handles VIRTIO_SND_R_PCM_START. + * + * @s: VirtIOSound device + * @cmd: The request command queue element from VirtIOSound cmdq field + * @start: whether to start or stop the device + */ +static void virtio_snd_handle_pcm_start_stop(VirtIOSound *s, + virtio_snd_ctrl_command *cmd, + bool start) +{ + VirtIOSoundPCMStream *stream; + virtio_snd_pcm_hdr req; + uint32_t stream_id; + size_t msg_sz = iov_to_buf(cmd->elem->out_sg, + cmd->elem->out_num, + 0, + &req, + sizeof(virtio_snd_pcm_hdr)); + + if (msg_sz != sizeof(virtio_snd_pcm_hdr)) { + qemu_log_mask(LOG_GUEST_ERROR, + "%s: virtio-snd command size incorrect %zu vs \ + %zu\n", __func__, msg_sz, sizeof(virtio_snd_pcm_hdr)); + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_BAD_MSG); + return; + } + + stream_id = le32_to_cpu(req.stream_id); + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_OK); + trace_virtio_snd_handle_pcm_start_stop(start ? "VIRTIO_SND_R_PCM_START" : + "VIRTIO_SND_R_PCM_STOP", stream_id); + stream = virtio_snd_pcm_get_stream(s, stream_id); + if (stream == NULL) { + error_report("Invalid stream id: %"PRIu32, req.stream_id); + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_BAD_MSG); + return; + } + stream->active = start; +} + /* * The actual processing done in virtio_snd_process_cmdq(). * @@ -486,10 +527,14 @@ process_cmd(VirtIOSound *s, virtio_snd_ctrl_command *cmd) case VIRTIO_SND_R_PCM_INFO: virtio_snd_handle_pcm_info(s, cmd); break; - case VIRTIO_SND_R_PCM_SET_PARAMS: - case VIRTIO_SND_R_PCM_PREPARE: case VIRTIO_SND_R_PCM_START: + virtio_snd_handle_pcm_start_stop(s, cmd, true); + break; case VIRTIO_SND_R_PCM_STOP: + virtio_snd_handle_pcm_start_stop(s, cmd, false); + break; + case VIRTIO_SND_R_PCM_SET_PARAMS: + case VIRTIO_SND_R_PCM_PREPARE: case VIRTIO_SND_R_PCM_RELEASE: cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_NOT_SUPP); break; -- cgit v1.2.3 From 64704ce04b0a62184d49f899e7f300f67480a2a6 Mon Sep 17 00:00:00 2001 From: Manos Pitsidianakis Date: Mon, 23 Oct 2023 15:03:23 +0300 Subject: virtio-sound: handle VIRTIO_SND_R_PCM_SET_PARAMS MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Handle the set parameters control request. It reconfigures a stream based on a guest's preference if the values are valid and supported. Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin Signed-off-by: Anton Yakovlev Signed-off-by: Manos Pitsidianakis Reviewed-by: Alex Bennée Tested-by: Alex Bennée Message-Id: Acked-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/audio/trace-events | 1 + hw/audio/virtio-snd.c | 34 ++++++++++++++++++++++++++++++++++ 2 files changed, 35 insertions(+) diff --git a/hw/audio/trace-events b/hw/audio/trace-events index db48ff04fe..3badcab2e8 100644 --- a/hw/audio/trace-events +++ b/hw/audio/trace-events @@ -47,6 +47,7 @@ virtio_snd_vm_state_running(void) "vm state running" virtio_snd_vm_state_stopped(void) "vm state stopped" virtio_snd_realize(void *snd) "snd %p: realize" virtio_snd_unrealize(void *snd) "snd %p: unrealize" +virtio_snd_handle_pcm_set_params(uint32_t stream) "VIRTIO_SND_PCM_SET_PARAMS called for stream %"PRIu32 virtio_snd_handle_ctrl(void *vdev, void *vq) "snd %p: handle ctrl event for queue %p" virtio_snd_handle_pcm_info(uint32_t stream) "VIRTIO_SND_R_PCM_INFO called for stream %"PRIu32 virtio_snd_handle_pcm_start_stop(const char *code, uint32_t stream) "%s called for stream %"PRIu32 diff --git a/hw/audio/virtio-snd.c b/hw/audio/virtio-snd.c index e6791de6c6..084890e52b 100644 --- a/hw/audio/virtio-snd.c +++ b/hw/audio/virtio-snd.c @@ -287,6 +287,38 @@ uint32_t virtio_snd_set_pcm_params(VirtIOSound *s, return cpu_to_le32(VIRTIO_SND_S_OK); } +/* + * Handles the VIRTIO_SND_R_PCM_SET_PARAMS request. + * + * @s: VirtIOSound device + * @cmd: The request command queue element from VirtIOSound cmdq field + */ +static void virtio_snd_handle_pcm_set_params(VirtIOSound *s, + virtio_snd_ctrl_command *cmd) +{ + virtio_snd_pcm_set_params req = { 0 }; + uint32_t stream_id; + size_t msg_sz = iov_to_buf(cmd->elem->out_sg, + cmd->elem->out_num, + 0, + &req, + sizeof(virtio_snd_pcm_set_params)); + + if (msg_sz != sizeof(virtio_snd_pcm_set_params)) { + /* + * TODO: do we need to set DEVICE_NEEDS_RESET? + */ + qemu_log_mask(LOG_GUEST_ERROR, + "%s: virtio-snd command size incorrect %zu vs \ + %zu\n", __func__, msg_sz, sizeof(virtio_snd_pcm_set_params)); + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_BAD_MSG); + return; + } + stream_id = le32_to_cpu(req.hdr.stream_id); + trace_virtio_snd_handle_pcm_set_params(stream_id); + cmd->resp.code = virtio_snd_set_pcm_params(s, stream_id, &req); +} + /* * Get a QEMU Audiosystem compatible format value from a VIRTIO_SND_PCM_FMT_* */ @@ -534,6 +566,8 @@ process_cmd(VirtIOSound *s, virtio_snd_ctrl_command *cmd) virtio_snd_handle_pcm_start_stop(s, cmd, false); break; case VIRTIO_SND_R_PCM_SET_PARAMS: + virtio_snd_handle_pcm_set_params(s, cmd); + break; case VIRTIO_SND_R_PCM_PREPARE: case VIRTIO_SND_R_PCM_RELEASE: cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_NOT_SUPP); -- cgit v1.2.3 From e5788b8fbf9361c978eb8347471d8f855580777f Mon Sep 17 00:00:00 2001 From: Manos Pitsidianakis Date: Mon, 23 Oct 2023 15:03:24 +0300 Subject: virtio-sound: handle VIRTIO_SND_R_PCM_PREPARE MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Handles the PCM prepare control request. It initializes a PCM stream when the guests asks for it. Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin Signed-off-by: Anton Yakovlev Signed-off-by: Manos Pitsidianakis Reviewed-by: Alex Bennée Tested-by: Alex Bennée Message-Id: Acked-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/audio/virtio-snd.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) diff --git a/hw/audio/virtio-snd.c b/hw/audio/virtio-snd.c index 084890e52b..31a1942754 100644 --- a/hw/audio/virtio-snd.c +++ b/hw/audio/virtio-snd.c @@ -478,6 +478,28 @@ static const char *print_code(uint32_t code) #undef CASE }; +/* + * Handles VIRTIO_SND_R_PCM_PREPARE. + * + * @s: VirtIOSound device + * @cmd: The request command queue element from VirtIOSound cmdq field + */ +static void virtio_snd_handle_pcm_prepare(VirtIOSound *s, + virtio_snd_ctrl_command *cmd) +{ + uint32_t stream_id; + size_t msg_sz = iov_to_buf(cmd->elem->out_sg, + cmd->elem->out_num, + sizeof(virtio_snd_hdr), + &stream_id, + sizeof(stream_id)); + + stream_id = le32_to_cpu(stream_id); + cmd->resp.code = msg_sz == sizeof(stream_id) + ? virtio_snd_pcm_prepare(s, stream_id) + : cpu_to_le32(VIRTIO_SND_S_BAD_MSG); +} + /* * Handles VIRTIO_SND_R_PCM_START. * @@ -569,6 +591,8 @@ process_cmd(VirtIOSound *s, virtio_snd_ctrl_command *cmd) virtio_snd_handle_pcm_set_params(s, cmd); break; case VIRTIO_SND_R_PCM_PREPARE: + virtio_snd_handle_pcm_prepare(s, cmd); + break; case VIRTIO_SND_R_PCM_RELEASE: cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_NOT_SUPP); break; -- cgit v1.2.3 From d48800d740d1eaea5be65accbf37d82f0effcfb5 Mon Sep 17 00:00:00 2001 From: Manos Pitsidianakis Date: Mon, 23 Oct 2023 15:03:25 +0300 Subject: virtio-sound: handle VIRTIO_SND_R_PCM_RELEASE MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Handle the PCM release control request, which is necessary for flushing pending sound IO. No IO is handled yet so currently it only replies to the request. Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin Signed-off-by: Anton Yakovlev Signed-off-by: Manos Pitsidianakis Reviewed-by: Alex Bennée Tested-by: Alex Bennée Message-Id: Acked-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/audio/trace-events | 1 + hw/audio/virtio-snd.c | 48 +++++++++++++++++++++++++++++++++++++++++++++++- 2 files changed, 48 insertions(+), 1 deletion(-) diff --git a/hw/audio/trace-events b/hw/audio/trace-events index 3badcab2e8..33e24d0011 100644 --- a/hw/audio/trace-events +++ b/hw/audio/trace-events @@ -51,6 +51,7 @@ virtio_snd_handle_pcm_set_params(uint32_t stream) "VIRTIO_SND_PCM_SET_PARAMS cal virtio_snd_handle_ctrl(void *vdev, void *vq) "snd %p: handle ctrl event for queue %p" virtio_snd_handle_pcm_info(uint32_t stream) "VIRTIO_SND_R_PCM_INFO called for stream %"PRIu32 virtio_snd_handle_pcm_start_stop(const char *code, uint32_t stream) "%s called for stream %"PRIu32 +virtio_snd_handle_pcm_release(uint32_t stream) "VIRTIO_SND_PCM_RELEASE called for stream %"PRIu32 virtio_snd_handle_code(uint32_t val, const char *code) "ctrl code msg val = %"PRIu32" == %s" virtio_snd_handle_chmap_info(void) "VIRTIO_SND_CHMAP_INFO called" virtio_snd_handle_event(void) "event queue callback called" diff --git a/hw/audio/virtio-snd.c b/hw/audio/virtio-snd.c index 31a1942754..9cff724f62 100644 --- a/hw/audio/virtio-snd.c +++ b/hw/audio/virtio-snd.c @@ -541,6 +541,52 @@ static void virtio_snd_handle_pcm_start_stop(VirtIOSound *s, stream->active = start; } +/* + * Handles VIRTIO_SND_R_PCM_RELEASE. Releases the buffer resources allocated to + * a stream. + * + * @s: VirtIOSound device + * @cmd: The request command queue element from VirtIOSound cmdq field + */ +static void virtio_snd_handle_pcm_release(VirtIOSound *s, + virtio_snd_ctrl_command *cmd) +{ + uint32_t stream_id; + VirtIOSoundPCMStream *stream; + size_t msg_sz = iov_to_buf(cmd->elem->out_sg, + cmd->elem->out_num, + sizeof(virtio_snd_hdr), + &stream_id, + sizeof(stream_id)); + + if (msg_sz != sizeof(stream_id)) { + /* + * TODO: do we need to set DEVICE_NEEDS_RESET? + */ + qemu_log_mask(LOG_GUEST_ERROR, + "%s: virtio-snd command size incorrect %zu vs \ + %zu\n", __func__, msg_sz, sizeof(stream_id)); + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_BAD_MSG); + return; + } + + stream_id = le32_to_cpu(stream_id); + trace_virtio_snd_handle_pcm_release(stream_id); + stream = virtio_snd_pcm_get_stream(s, stream_id); + if (stream == NULL) { + /* + * TODO: do we need to set DEVICE_NEEDS_RESET? + */ + error_report("already released stream %"PRIu32, stream_id); + virtio_error(VIRTIO_DEVICE(s), + "already released stream %"PRIu32, + stream_id); + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_BAD_MSG); + return; + } + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_OK); +} + /* * The actual processing done in virtio_snd_process_cmdq(). * @@ -594,7 +640,7 @@ process_cmd(VirtIOSound *s, virtio_snd_ctrl_command *cmd) virtio_snd_handle_pcm_prepare(s, cmd); break; case VIRTIO_SND_R_PCM_RELEASE: - cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_NOT_SUPP); + virtio_snd_handle_pcm_release(s, cmd); break; case VIRTIO_SND_R_CHMAP_INFO: qemu_log_mask(LOG_UNIMP, -- cgit v1.2.3 From 18a752810f04e3f1933682addd0a6c978efc5cdf Mon Sep 17 00:00:00 2001 From: Manos Pitsidianakis Date: Mon, 23 Oct 2023 15:03:26 +0300 Subject: virtio-sound: implement audio output (TX) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Handle output IO messages in the transmit (TX) virtqueue. It allocates a VirtIOSoundPCMBuffer for each IO message and copies the data buffer to it. When the IO buffer is written to the host's sound card, the guest will be notified that it has been consumed. The lifetime of an IO message is: 1. Guest sends IO message to TX virtqueue. 2. QEMU adds it to the appropriate stream's IO buffer queue. 3. Sometime later, the host audio backend calls the output callback, virtio_snd_pcm_out_cb(), which is defined with an AUD_open_out() call. The callback gets an available number of bytes the backend can receive. Then it writes data from the IO buffer queue to the backend. If at any time a buffer is exhausted, it is returned to the guest as completed. 4. If the guest releases the stream, its buffer queue is flushed by attempting to write any leftover data to the audio backend and releasing all IO messages back to the guest. This is how according to the spec the guest knows the release was successful. Based-on: https://github.com/OpenSynergy/qemu/commit/5a2f350eec5d157b90d9c7b40a8e603f4da92471 Signed-off-by: Igor Skalkin Signed-off-by: Anton Yakovlev Signed-off-by: Manos Pitsidianakis Reviewed-by: Alex Bennée Tested-by: Alex Bennée Message-Id: Acked-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/audio/trace-events | 2 + hw/audio/virtio-snd.c | 288 +++++++++++++++++++++++++++++++++++++++++- include/hw/audio/virtio-snd.h | 47 +++++++ 3 files changed, 332 insertions(+), 5 deletions(-) diff --git a/hw/audio/trace-events b/hw/audio/trace-events index 33e24d0011..884108129b 100644 --- a/hw/audio/trace-events +++ b/hw/audio/trace-events @@ -55,3 +55,5 @@ virtio_snd_handle_pcm_release(uint32_t stream) "VIRTIO_SND_PCM_RELEASE called fo virtio_snd_handle_code(uint32_t val, const char *code) "ctrl code msg val = %"PRIu32" == %s" virtio_snd_handle_chmap_info(void) "VIRTIO_SND_CHMAP_INFO called" virtio_snd_handle_event(void) "event queue callback called" +virtio_snd_pcm_stream_flush(uint32_t stream) "flushing stream %"PRIu32 +virtio_snd_handle_xfer(void) "tx/rx queue callback called" diff --git a/hw/audio/virtio-snd.c b/hw/audio/virtio-snd.c index 9cff724f62..6c91d0a740 100644 --- a/hw/audio/virtio-snd.c +++ b/hw/audio/virtio-snd.c @@ -32,6 +32,10 @@ #define VIRTIO_SOUND_CHMAP_DEFAULT 0 #define VIRTIO_SOUND_HDA_FN_NID 0 +static void virtio_snd_pcm_out_cb(void *data, int available); +static void virtio_snd_process_cmdq(VirtIOSound *s); +static void virtio_snd_pcm_flush(VirtIOSoundPCMStream *stream); + static uint32_t supported_formats = BIT(VIRTIO_SND_PCM_FMT_S8) | BIT(VIRTIO_SND_PCM_FMT_U8) | BIT(VIRTIO_SND_PCM_FMT_S16) @@ -123,6 +127,13 @@ virtio_snd_set_config(VirtIODevice *vdev, const uint8_t *config) } +static void +virtio_snd_pcm_buffer_free(VirtIOSoundPCMBuffer *buffer) +{ + g_free(buffer->elem); + g_free(buffer); +} + static void virtio_snd_ctrl_cmd_free(virtio_snd_ctrl_command *cmd) { @@ -396,6 +407,13 @@ static void virtio_snd_get_qemu_audsettings(audsettings *as, */ static void virtio_snd_pcm_close(VirtIOSoundPCMStream *stream) { + if (stream) { + if (stream->info.direction == VIRTIO_SND_D_OUTPUT) { + virtio_snd_pcm_flush(stream); + AUD_close_out(&stream->pcm->snd->card, stream->voice.out); + stream->voice.out = NULL; + } + } } /* @@ -429,6 +447,9 @@ static uint32_t virtio_snd_pcm_prepare(VirtIOSound *s, uint32_t stream_id) stream->id = stream_id; stream->pcm = s->pcm; stream->s = s; + qemu_mutex_init(&stream->queue_mutex); + QSIMPLEQ_INIT(&stream->queue); + QSIMPLEQ_INIT(&stream->invalid); /* * stream_id >= s->snd_conf.streams was checked before so this is @@ -452,6 +473,18 @@ static uint32_t virtio_snd_pcm_prepare(VirtIOSound *s, uint32_t stream_id) stream->positions[1] = VIRTIO_SND_CHMAP_FR; stream->as = as; + if (stream->info.direction == VIRTIO_SND_D_OUTPUT) { + stream->voice.out = AUD_open_out(&s->card, + stream->voice.out, + "virtio-sound.out", + stream, + virtio_snd_pcm_out_cb, + &as); + AUD_set_volume_out(stream->voice.out, 0, 255, 255); + } else { + qemu_log_mask(LOG_UNIMP, "virtio_snd: input/capture is unimplemented."); + } + return cpu_to_le32(VIRTIO_SND_S_OK); } @@ -532,9 +565,17 @@ static void virtio_snd_handle_pcm_start_stop(VirtIOSound *s, cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_OK); trace_virtio_snd_handle_pcm_start_stop(start ? "VIRTIO_SND_R_PCM_START" : "VIRTIO_SND_R_PCM_STOP", stream_id); + stream = virtio_snd_pcm_get_stream(s, stream_id); - if (stream == NULL) { - error_report("Invalid stream id: %"PRIu32, req.stream_id); + if (stream) { + WITH_QEMU_LOCK_GUARD(&stream->queue_mutex) { + stream->active = start; + } + if (stream->info.direction == VIRTIO_SND_D_OUTPUT) { + AUD_set_active_out(stream->voice.out, start); + } + } else { + error_report("Invalid stream id: %"PRIu32, stream_id); cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_BAD_MSG); return; } @@ -542,8 +583,28 @@ static void virtio_snd_handle_pcm_start_stop(VirtIOSound *s, } /* - * Handles VIRTIO_SND_R_PCM_RELEASE. Releases the buffer resources allocated to - * a stream. + * Returns the number of I/O messages that are being processed. + * + * @stream: VirtIOSoundPCMStream + */ +static size_t virtio_snd_pcm_get_io_msgs_count(VirtIOSoundPCMStream *stream) +{ + VirtIOSoundPCMBuffer *buffer, *next; + size_t count = 0; + + WITH_QEMU_LOCK_GUARD(&stream->queue_mutex) { + QSIMPLEQ_FOREACH_SAFE(buffer, &stream->queue, entry, next) { + count += 1; + } + QSIMPLEQ_FOREACH_SAFE(buffer, &stream->invalid, entry, next) { + count += 1; + } + } + return count; +} + +/* + * Handles VIRTIO_SND_R_PCM_RELEASE. * * @s: VirtIOSound device * @cmd: The request command queue element from VirtIOSound cmdq field @@ -584,6 +645,21 @@ static void virtio_snd_handle_pcm_release(VirtIOSound *s, cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_BAD_MSG); return; } + + if (virtio_snd_pcm_get_io_msgs_count(stream)) { + /* + * virtio-v1.2-csd01, 5.14.6.6.5.1, + * Device Requirements: Stream Release + * + * - The device MUST complete all pending I/O messages for the + * specified stream ID. + * - The device MUST NOT complete the control request while there + * are pending I/O messages for the specified stream ID. + */ + trace_virtio_snd_pcm_stream_flush(stream_id); + virtio_snd_pcm_flush(stream); + } + cmd->resp.code = cpu_to_le32(VIRTIO_SND_S_OK); } @@ -738,6 +814,108 @@ static void virtio_snd_handle_event(VirtIODevice *vdev, VirtQueue *vq) trace_virtio_snd_handle_event(); } +/* + * The tx virtqueue handler. Makes the buffers available to their respective + * streams for consumption. + * + * @vdev: VirtIOSound device + * @vq: tx virtqueue + */ +static void virtio_snd_handle_tx(VirtIODevice *vdev, VirtQueue *vq) +{ + VirtIOSound *s = VIRTIO_SND(vdev); + VirtIOSoundPCMStream *stream = NULL; + VirtIOSoundPCMBuffer *buffer; + VirtQueueElement *elem; + size_t msg_sz, size; + virtio_snd_pcm_xfer hdr; + virtio_snd_pcm_status resp = { 0 }; + uint32_t stream_id; + /* + * If any of the I/O messages are invalid, put them in stream->invalid and + * return them after the for loop. + */ + bool must_empty_invalid_queue = false; + + if (!virtio_queue_ready(vq)) { + return; + } + trace_virtio_snd_handle_xfer(); + + for (;;) { + elem = virtqueue_pop(vq, sizeof(VirtQueueElement)); + if (!elem) { + break; + } + /* get the message hdr object */ + msg_sz = iov_to_buf(elem->out_sg, + elem->out_num, + 0, + &hdr, + sizeof(virtio_snd_pcm_xfer)); + if (msg_sz != sizeof(virtio_snd_pcm_xfer)) { + goto tx_err; + } + stream_id = le32_to_cpu(hdr.stream_id); + + if (stream_id >= s->snd_conf.streams + || s->pcm->streams[stream_id] == NULL) { + goto tx_err; + } + + stream = s->pcm->streams[stream_id]; + if (stream->info.direction != VIRTIO_SND_D_OUTPUT) { + goto tx_err; + } + + WITH_QEMU_LOCK_GUARD(&stream->queue_mutex) { + size = iov_size(elem->out_sg, elem->out_num) - msg_sz; + + buffer = g_malloc0(sizeof(VirtIOSoundPCMBuffer) + size); + buffer->elem = elem; + buffer->populated = false; + buffer->vq = vq; + buffer->size = size; + buffer->offset = 0; + + QSIMPLEQ_INSERT_TAIL(&stream->queue, buffer, entry); + } + continue; + +tx_err: + WITH_QEMU_LOCK_GUARD(&stream->queue_mutex) { + must_empty_invalid_queue = true; + buffer = g_malloc0(sizeof(VirtIOSoundPCMBuffer)); + buffer->elem = elem; + buffer->vq = vq; + QSIMPLEQ_INSERT_TAIL(&stream->invalid, buffer, entry); + } + } + + if (must_empty_invalid_queue) { + WITH_QEMU_LOCK_GUARD(&stream->queue_mutex) { + while (!QSIMPLEQ_EMPTY(&stream->invalid)) { + buffer = QSIMPLEQ_FIRST(&stream->invalid); + + resp.status = cpu_to_le32(VIRTIO_SND_S_BAD_MSG); + iov_from_buf(buffer->elem->in_sg, + buffer->elem->in_num, + 0, + &resp, + sizeof(virtio_snd_pcm_status)); + virtqueue_push(vq, buffer->elem, sizeof(virtio_snd_pcm_status)); + QSIMPLEQ_REMOVE_HEAD(&stream->invalid, entry); + virtio_snd_pcm_buffer_free(buffer); + } + /* + * Notify vq about virtio_snd_pcm_status responses. + * Buffer responses must be notified separately later. + */ + virtio_notify(vdev, vq); + } + } +} + /* * Stub buffer virtqueue handler. * @@ -832,7 +1010,7 @@ static void virtio_snd_realize(DeviceState *dev, Error **errp) vsnd->queues[VIRTIO_SND_VQ_EVENT] = virtio_add_queue(vdev, 64, virtio_snd_handle_event); vsnd->queues[VIRTIO_SND_VQ_TX] = - virtio_add_queue(vdev, 64, virtio_snd_handle_xfer); + virtio_add_queue(vdev, 64, virtio_snd_handle_tx); vsnd->queues[VIRTIO_SND_VQ_RX] = virtio_add_queue(vdev, 64, virtio_snd_handle_xfer); qemu_mutex_init(&vsnd->cmdq_mutex); @@ -856,6 +1034,105 @@ static void virtio_snd_realize(DeviceState *dev, Error **errp) } } +static inline void return_tx_buffer(VirtIOSoundPCMStream *stream, + VirtIOSoundPCMBuffer *buffer) +{ + virtio_snd_pcm_status resp = { 0 }; + resp.status = cpu_to_le32(VIRTIO_SND_S_OK); + resp.latency_bytes = cpu_to_le32((uint32_t)buffer->size); + iov_from_buf(buffer->elem->in_sg, + buffer->elem->in_num, + 0, + &resp, + sizeof(virtio_snd_pcm_status)); + virtqueue_push(buffer->vq, + buffer->elem, + sizeof(virtio_snd_pcm_status)); + virtio_notify(VIRTIO_DEVICE(stream->s), buffer->vq); + QSIMPLEQ_REMOVE(&stream->queue, + buffer, + VirtIOSoundPCMBuffer, + entry); + virtio_snd_pcm_buffer_free(buffer); +} + +/* + * AUD_* output callback. + * + * @data: VirtIOSoundPCMStream stream + * @available: number of bytes that can be written with AUD_write() + */ +static void virtio_snd_pcm_out_cb(void *data, int available) +{ + VirtIOSoundPCMStream *stream = data; + VirtIOSoundPCMBuffer *buffer; + size_t size; + + WITH_QEMU_LOCK_GUARD(&stream->queue_mutex) { + while (!QSIMPLEQ_EMPTY(&stream->queue)) { + buffer = QSIMPLEQ_FIRST(&stream->queue); + if (!virtio_queue_ready(buffer->vq)) { + return; + } + if (!stream->active) { + /* Stream has stopped, so do not perform AUD_write. */ + return_tx_buffer(stream, buffer); + continue; + } + if (!buffer->populated) { + iov_to_buf(buffer->elem->out_sg, + buffer->elem->out_num, + sizeof(virtio_snd_pcm_xfer), + buffer->data, + buffer->size); + buffer->populated = true; + } + for (;;) { + size = AUD_write(stream->voice.out, + buffer->data + buffer->offset, + MIN(buffer->size, available)); + assert(size <= MIN(buffer->size, available)); + if (size == 0) { + /* break out of both loops */ + available = 0; + break; + } + buffer->size -= size; + buffer->offset += size; + available -= size; + if (buffer->size < 1) { + return_tx_buffer(stream, buffer); + break; + } + if (!available) { + break; + } + } + if (!available) { + break; + } + } + } +} + +/* + * Flush all buffer data from this stream's queue into the driver's virtual + * queue. + * + * @stream: VirtIOSoundPCMStream *stream + */ +static inline void virtio_snd_pcm_flush(VirtIOSoundPCMStream *stream) +{ + VirtIOSoundPCMBuffer *buffer; + + WITH_QEMU_LOCK_GUARD(&stream->queue_mutex) { + while (!QSIMPLEQ_EMPTY(&stream->queue)) { + buffer = QSIMPLEQ_FIRST(&stream->queue); + return_tx_buffer(stream, buffer); + } + } +} + static void virtio_snd_unrealize(DeviceState *dev) { VirtIODevice *vdev = VIRTIO_DEVICE(dev); @@ -872,6 +1149,7 @@ static void virtio_snd_unrealize(DeviceState *dev) if (stream) { virtio_snd_process_cmdq(stream->s); virtio_snd_pcm_close(stream); + qemu_mutex_destroy(&stream->queue_mutex); g_free(stream); } } diff --git a/include/hw/audio/virtio-snd.h b/include/hw/audio/virtio-snd.h index cc14c875ed..c3767f442b 100644 --- a/include/hw/audio/virtio-snd.h +++ b/include/hw/audio/virtio-snd.h @@ -77,6 +77,50 @@ typedef struct virtio_snd_ctrl_command virtio_snd_ctrl_command; typedef struct VirtIOSoundPCM VirtIOSoundPCM; +typedef struct VirtIOSoundPCMBuffer VirtIOSoundPCMBuffer; + +/* + * The VirtIO sound spec reuses layouts and values from the High Definition + * Audio spec (virtio/v1.2: 5.14 Sound Device). This struct handles each I/O + * message's buffer (virtio/v1.2: 5.14.6.8 PCM I/O Messages). + * + * In the case of TX (i.e. playback) buffers, we defer reading the raw PCM data + * from the virtqueue until QEMU's sound backsystem calls the output callback. + * This is tracked by the `bool populated;` field, which is set to true when + * data has been read into our own buffer for consumption. + * + * VirtIOSoundPCMBuffer has a dynamic size since it includes the raw PCM data + * in its allocation. It must be initialized and destroyed as follows: + * + * size_t size = [[derived from owned VQ element descriptor sizes]]; + * buffer = g_malloc0(sizeof(VirtIOSoundPCMBuffer) + size); + * buffer->elem = [[owned VQ element]]; + * + * [..] + * + * g_free(buffer->elem); + * g_free(buffer); + */ +struct VirtIOSoundPCMBuffer { + QSIMPLEQ_ENTRY(VirtIOSoundPCMBuffer) entry; + VirtQueueElement *elem; + VirtQueue *vq; + size_t size; + /* + * In TX / Plaback, `offset` represents the first unused position inside + * `data`. If `offset == size` then there are no unused data left. + */ + uint64_t offset; + /* Used for the TX queue for lazy I/O copy from `elem` */ + bool populated; + /* + * VirtIOSoundPCMBuffer is an unsized type because it ends with an array of + * bytes. The size of `data` is determined from the I/O message's read-only + * or write-only size when allocating VirtIOSoundPCMBuffer. + */ + uint8_t data[]; +}; + struct VirtIOSoundPCM { VirtIOSound *snd; /* @@ -104,7 +148,10 @@ struct VirtIOSoundPCMStream { SWVoiceIn *in; SWVoiceOut *out; } voice; + QemuMutex queue_mutex; bool active; + QSIMPLEQ_HEAD(, VirtIOSoundPCMBuffer) queue; + QSIMPLEQ_HEAD(, VirtIOSoundPCMBuffer) invalid; }; /* -- cgit v1.2.3 From d8d64acbecfa131153e5b96fb24b4f95f55be881 Mon Sep 17 00:00:00 2001 From: Manos Pitsidianakis Date: Mon, 23 Oct 2023 15:03:27 +0300 Subject: virtio-sound: implement audio capture (RX) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit To perform audio capture we duplicate the TX logic of the previous commit with the following difference: we receive data from the QEMU audio backend and write it in the virt queue IO buffers the guest sends to QEMU. When they are full (i.e. they have `period_bytes` amount of data) or when recording stops in QEMU's audio backend, the buffer is returned to the guest by notifying it. Signed-off-by: Manos Pitsidianakis Tested-by: Alex Bennée Reviewed-by: Alex Bennée Message-Id: Acked-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/audio/trace-events | 3 +- hw/audio/virtio-snd.c | 262 +++++++++++++++++++++++++++++++++++++++++++------- 2 files changed, 230 insertions(+), 35 deletions(-) diff --git a/hw/audio/trace-events b/hw/audio/trace-events index 884108129b..b1870ff224 100644 --- a/hw/audio/trace-events +++ b/hw/audio/trace-events @@ -56,4 +56,5 @@ virtio_snd_handle_code(uint32_t val, const char *code) "ctrl code msg val = %"PR virtio_snd_handle_chmap_info(void) "VIRTIO_SND_CHMAP_INFO called" virtio_snd_handle_event(void) "event queue callback called" virtio_snd_pcm_stream_flush(uint32_t stream) "flushing stream %"PRIu32 -virtio_snd_handle_xfer(void) "tx/rx queue callback called" +virtio_snd_handle_tx_xfer(void) "tx queue callback called" +virtio_snd_handle_rx_xfer(void) "rx queue callback called" diff --git a/hw/audio/virtio-snd.c b/hw/audio/virtio-snd.c index 6c91d0a740..a18a9949a7 100644 --- a/hw/audio/virtio-snd.c +++ b/hw/audio/virtio-snd.c @@ -28,13 +28,14 @@ #define VIRTIO_SOUND_VM_VERSION 1 #define VIRTIO_SOUND_JACK_DEFAULT 0 -#define VIRTIO_SOUND_STREAM_DEFAULT 1 +#define VIRTIO_SOUND_STREAM_DEFAULT 2 #define VIRTIO_SOUND_CHMAP_DEFAULT 0 #define VIRTIO_SOUND_HDA_FN_NID 0 static void virtio_snd_pcm_out_cb(void *data, int available); static void virtio_snd_process_cmdq(VirtIOSound *s); static void virtio_snd_pcm_flush(VirtIOSoundPCMStream *stream); +static void virtio_snd_pcm_in_cb(void *data, int available); static uint32_t supported_formats = BIT(VIRTIO_SND_PCM_FMT_S8) | BIT(VIRTIO_SND_PCM_FMT_U8) @@ -408,10 +409,13 @@ static void virtio_snd_get_qemu_audsettings(audsettings *as, static void virtio_snd_pcm_close(VirtIOSoundPCMStream *stream) { if (stream) { + virtio_snd_pcm_flush(stream); if (stream->info.direction == VIRTIO_SND_D_OUTPUT) { - virtio_snd_pcm_flush(stream); AUD_close_out(&stream->pcm->snd->card, stream->voice.out); stream->voice.out = NULL; + } else if (stream->info.direction == VIRTIO_SND_D_INPUT) { + AUD_close_in(&stream->pcm->snd->card, stream->voice.in); + stream->voice.in = NULL; } } } @@ -482,7 +486,13 @@ static uint32_t virtio_snd_pcm_prepare(VirtIOSound *s, uint32_t stream_id) &as); AUD_set_volume_out(stream->voice.out, 0, 255, 255); } else { - qemu_log_mask(LOG_UNIMP, "virtio_snd: input/capture is unimplemented."); + stream->voice.in = AUD_open_in(&s->card, + stream->voice.in, + "virtio-sound.in", + stream, + virtio_snd_pcm_in_cb, + &as); + AUD_set_volume_in(stream->voice.in, 0, 255, 255); } return cpu_to_le32(VIRTIO_SND_S_OK); @@ -573,6 +583,8 @@ static void virtio_snd_handle_pcm_start_stop(VirtIOSound *s, } if (stream->info.direction == VIRTIO_SND_D_OUTPUT) { AUD_set_active_out(stream->voice.out, start); + } else { + AUD_set_active_in(stream->voice.in, start); } } else { error_report("Invalid stream id: %"PRIu32, stream_id); @@ -814,6 +826,49 @@ static void virtio_snd_handle_event(VirtIODevice *vdev, VirtQueue *vq) trace_virtio_snd_handle_event(); } +static inline void empty_invalid_queue(VirtIODevice *vdev, VirtQueue *vq) +{ + VirtIOSoundPCMBuffer *buffer = NULL; + VirtIOSoundPCMStream *stream = NULL; + virtio_snd_pcm_status resp = { 0 }; + VirtIOSound *vsnd = VIRTIO_SND(vdev); + bool any = false; + + for (uint32_t i = 0; i < vsnd->snd_conf.streams; i++) { + stream = vsnd->pcm->streams[i]; + if (stream) { + any = false; + WITH_QEMU_LOCK_GUARD(&stream->queue_mutex) { + while (!QSIMPLEQ_EMPTY(&stream->invalid)) { + buffer = QSIMPLEQ_FIRST(&stream->invalid); + if (buffer->vq != vq) { + break; + } + any = true; + resp.status = cpu_to_le32(VIRTIO_SND_S_BAD_MSG); + iov_from_buf(buffer->elem->in_sg, + buffer->elem->in_num, + 0, + &resp, + sizeof(virtio_snd_pcm_status)); + virtqueue_push(vq, + buffer->elem, + sizeof(virtio_snd_pcm_status)); + QSIMPLEQ_REMOVE_HEAD(&stream->invalid, entry); + virtio_snd_pcm_buffer_free(buffer); + } + if (any) { + /* + * Notify vq about virtio_snd_pcm_status responses. + * Buffer responses must be notified separately later. + */ + virtio_notify(vdev, vq); + } + } + } + } +} + /* * The tx virtqueue handler. Makes the buffers available to their respective * streams for consumption. @@ -821,7 +876,7 @@ static void virtio_snd_handle_event(VirtIODevice *vdev, VirtQueue *vq) * @vdev: VirtIOSound device * @vq: tx virtqueue */ -static void virtio_snd_handle_tx(VirtIODevice *vdev, VirtQueue *vq) +static void virtio_snd_handle_tx_xfer(VirtIODevice *vdev, VirtQueue *vq) { VirtIOSound *s = VIRTIO_SND(vdev); VirtIOSoundPCMStream *stream = NULL; @@ -829,7 +884,6 @@ static void virtio_snd_handle_tx(VirtIODevice *vdev, VirtQueue *vq) VirtQueueElement *elem; size_t msg_sz, size; virtio_snd_pcm_xfer hdr; - virtio_snd_pcm_status resp = { 0 }; uint32_t stream_id; /* * If any of the I/O messages are invalid, put them in stream->invalid and @@ -840,7 +894,7 @@ static void virtio_snd_handle_tx(VirtIODevice *vdev, VirtQueue *vq) if (!virtio_queue_ready(vq)) { return; } - trace_virtio_snd_handle_xfer(); + trace_virtio_snd_handle_tx_xfer(); for (;;) { elem = virtqueue_pop(vq, sizeof(VirtQueueElement)); @@ -893,36 +947,88 @@ tx_err: } if (must_empty_invalid_queue) { - WITH_QEMU_LOCK_GUARD(&stream->queue_mutex) { - while (!QSIMPLEQ_EMPTY(&stream->invalid)) { - buffer = QSIMPLEQ_FIRST(&stream->invalid); - - resp.status = cpu_to_le32(VIRTIO_SND_S_BAD_MSG); - iov_from_buf(buffer->elem->in_sg, - buffer->elem->in_num, - 0, - &resp, - sizeof(virtio_snd_pcm_status)); - virtqueue_push(vq, buffer->elem, sizeof(virtio_snd_pcm_status)); - QSIMPLEQ_REMOVE_HEAD(&stream->invalid, entry); - virtio_snd_pcm_buffer_free(buffer); - } - /* - * Notify vq about virtio_snd_pcm_status responses. - * Buffer responses must be notified separately later. - */ - virtio_notify(vdev, vq); - } + empty_invalid_queue(vdev, vq); } } /* - * Stub buffer virtqueue handler. + * The rx virtqueue handler. Makes the buffers available to their respective + * streams for consumption. * * @vdev: VirtIOSound device - * @vq: virtqueue + * @vq: rx virtqueue */ -static void virtio_snd_handle_xfer(VirtIODevice *vdev, VirtQueue *vq) {} +static void virtio_snd_handle_rx_xfer(VirtIODevice *vdev, VirtQueue *vq) +{ + VirtIOSound *s = VIRTIO_SND(vdev); + VirtIOSoundPCMStream *stream = NULL; + VirtIOSoundPCMBuffer *buffer; + VirtQueueElement *elem; + size_t msg_sz, size; + virtio_snd_pcm_xfer hdr; + uint32_t stream_id; + /* + * if any of the I/O messages are invalid, put them in stream->invalid and + * return them after the for loop. + */ + bool must_empty_invalid_queue = false; + + if (!virtio_queue_ready(vq)) { + return; + } + trace_virtio_snd_handle_rx_xfer(); + + for (;;) { + elem = virtqueue_pop(vq, sizeof(VirtQueueElement)); + if (!elem) { + break; + } + /* get the message hdr object */ + msg_sz = iov_to_buf(elem->out_sg, + elem->out_num, + 0, + &hdr, + sizeof(virtio_snd_pcm_xfer)); + if (msg_sz != sizeof(virtio_snd_pcm_xfer)) { + goto rx_err; + } + stream_id = le32_to_cpu(hdr.stream_id); + + if (stream_id >= s->snd_conf.streams + || !s->pcm->streams[stream_id]) { + goto rx_err; + } + + stream = s->pcm->streams[stream_id]; + if (stream == NULL || stream->info.direction != VIRTIO_SND_D_INPUT) { + goto rx_err; + } + WITH_QEMU_LOCK_GUARD(&stream->queue_mutex) { + size = iov_size(elem->in_sg, elem->in_num) - + sizeof(virtio_snd_pcm_status); + buffer = g_malloc0(sizeof(VirtIOSoundPCMBuffer) + size); + buffer->elem = elem; + buffer->vq = vq; + buffer->size = 0; + buffer->offset = 0; + QSIMPLEQ_INSERT_TAIL(&stream->queue, buffer, entry); + } + continue; + +rx_err: + WITH_QEMU_LOCK_GUARD(&stream->queue_mutex) { + must_empty_invalid_queue = true; + buffer = g_malloc0(sizeof(VirtIOSoundPCMBuffer)); + buffer->elem = elem; + buffer->vq = vq; + QSIMPLEQ_INSERT_TAIL(&stream->invalid, buffer, entry); + } + } + + if (must_empty_invalid_queue) { + empty_invalid_queue(vdev, vq); + } +} static uint64_t get_features(VirtIODevice *vdev, uint64_t features, Error **errp) @@ -1010,9 +1116,9 @@ static void virtio_snd_realize(DeviceState *dev, Error **errp) vsnd->queues[VIRTIO_SND_VQ_EVENT] = virtio_add_queue(vdev, 64, virtio_snd_handle_event); vsnd->queues[VIRTIO_SND_VQ_TX] = - virtio_add_queue(vdev, 64, virtio_snd_handle_tx); + virtio_add_queue(vdev, 64, virtio_snd_handle_tx_xfer); vsnd->queues[VIRTIO_SND_VQ_RX] = - virtio_add_queue(vdev, 64, virtio_snd_handle_xfer); + virtio_add_queue(vdev, 64, virtio_snd_handle_rx_xfer); qemu_mutex_init(&vsnd->cmdq_mutex); QTAILQ_INIT(&vsnd->cmdq); @@ -1116,19 +1222,107 @@ static void virtio_snd_pcm_out_cb(void *data, int available) } /* - * Flush all buffer data from this stream's queue into the driver's virtual - * queue. + * Flush all buffer data from this input stream's queue into the driver's + * virtual queue. + * + * @stream: VirtIOSoundPCMStream *stream + */ +static inline void return_rx_buffer(VirtIOSoundPCMStream *stream, + VirtIOSoundPCMBuffer *buffer) +{ + virtio_snd_pcm_status resp = { 0 }; + resp.status = cpu_to_le32(VIRTIO_SND_S_OK); + resp.latency_bytes = 0; + /* Copy data -if any- to guest */ + iov_from_buf(buffer->elem->in_sg, + buffer->elem->in_num, + 0, + buffer->data, + buffer->size); + iov_from_buf(buffer->elem->in_sg, + buffer->elem->in_num, + buffer->size, + &resp, + sizeof(virtio_snd_pcm_status)); + virtqueue_push(buffer->vq, + buffer->elem, + sizeof(virtio_snd_pcm_status) + buffer->size); + virtio_notify(VIRTIO_DEVICE(stream->s), buffer->vq); + QSIMPLEQ_REMOVE(&stream->queue, + buffer, + VirtIOSoundPCMBuffer, + entry); + virtio_snd_pcm_buffer_free(buffer); +} + + +/* + * AUD_* input callback. + * + * @data: VirtIOSoundPCMStream stream + * @available: number of bytes that can be read with AUD_read() + */ +static void virtio_snd_pcm_in_cb(void *data, int available) +{ + VirtIOSoundPCMStream *stream = data; + VirtIOSoundPCMBuffer *buffer; + size_t size; + + WITH_QEMU_LOCK_GUARD(&stream->queue_mutex) { + while (!QSIMPLEQ_EMPTY(&stream->queue)) { + buffer = QSIMPLEQ_FIRST(&stream->queue); + if (!virtio_queue_ready(buffer->vq)) { + return; + } + if (!stream->active) { + /* Stream has stopped, so do not perform AUD_read. */ + return_rx_buffer(stream, buffer); + continue; + } + + for (;;) { + size = AUD_read(stream->voice.in, + buffer->data + buffer->size, + MIN(available, (stream->params.period_bytes - + buffer->size))); + if (!size) { + available = 0; + break; + } + buffer->size += size; + available -= size; + if (buffer->size >= stream->params.period_bytes) { + return_rx_buffer(stream, buffer); + break; + } + if (!available) { + break; + } + } + if (!available) { + break; + } + } + } +} + +/* + * Flush all buffer data from this output stream's queue into the driver's + * virtual queue. * * @stream: VirtIOSoundPCMStream *stream */ static inline void virtio_snd_pcm_flush(VirtIOSoundPCMStream *stream) { VirtIOSoundPCMBuffer *buffer; + void (*cb)(VirtIOSoundPCMStream *, VirtIOSoundPCMBuffer *) = + (stream->info.direction == VIRTIO_SND_D_OUTPUT) ? return_tx_buffer : + return_rx_buffer; WITH_QEMU_LOCK_GUARD(&stream->queue_mutex) { while (!QSIMPLEQ_EMPTY(&stream->queue)) { buffer = QSIMPLEQ_FIRST(&stream->queue); - return_tx_buffer(stream, buffer); + cb(stream, buffer); } } } -- cgit v1.2.3 From f54fea113cc48000836b95681f3b9fdbe43c7d96 Mon Sep 17 00:00:00 2001 From: Manos Pitsidianakis Date: Mon, 23 Oct 2023 15:03:28 +0300 Subject: docs/system: add basic virtio-snd documentation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This commit adds basic documentation for using virtio-snd. Signed-off-by: Manos Pitsidianakis Reviewed-by: Alex Bennée Tested-by: Alex Bennée Message-Id: Acked-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- MAINTAINERS | 1 + docs/system/device-emulation.rst | 1 + docs/system/devices/virtio-snd.rst | 49 ++++++++++++++++++++++++++++++++++++++ 3 files changed, 51 insertions(+) create mode 100644 docs/system/devices/virtio-snd.rst diff --git a/MAINTAINERS b/MAINTAINERS index c09bb8cf41..4fad272d73 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2317,6 +2317,7 @@ S: Supported F: hw/audio/virtio-snd.c F: hw/audio/virtio-snd-pci.c F: include/hw/audio/virtio-snd.h +F: docs/system/devices/virtio-snd.rst nvme M: Keith Busch diff --git a/docs/system/device-emulation.rst b/docs/system/device-emulation.rst index 1167f3a9f2..d1f3277cb0 100644 --- a/docs/system/device-emulation.rst +++ b/docs/system/device-emulation.rst @@ -93,6 +93,7 @@ Emulated Devices devices/vhost-user.rst devices/virtio-gpu.rst devices/virtio-pmem.rst + devices/virtio-snd.rst devices/vhost-user-rng.rst devices/canokey.rst devices/usb-u2f.rst diff --git a/docs/system/devices/virtio-snd.rst b/docs/system/devices/virtio-snd.rst new file mode 100644 index 0000000000..2a9187fd70 --- /dev/null +++ b/docs/system/devices/virtio-snd.rst @@ -0,0 +1,49 @@ +virtio sound +============ + +This document explains the setup and usage of the Virtio sound device. +The Virtio sound device is a paravirtualized sound card device. + +Linux kernel support +-------------------- + +Virtio sound requires a guest Linux kernel built with the +``CONFIG_SND_VIRTIO`` option. + +Description +----------- + +Virtio sound implements capture and playback from inside a guest using the +configured audio backend of the host machine. + +Device properties +----------------- + +The Virtio sound device can be configured with the following properties: + + * ``jacks`` number of physical jacks (Unimplemented). + * ``streams`` number of PCM streams. At the moment, no stream configuration is supported: the first one will always be a playback stream, an optional second will always be a capture stream. Adding more will cycle stream directions from playback to capture. + * ``chmaps`` number of channel maps (Unimplemented). + +All streams are stereo and have the default channel positions ``Front left, right``. + +Examples +-------- + +Add an audio device and an audio backend at once with ``-audio`` and ``model=virtio``: + + * pulseaudio: ``-audio driver=pa,model=virtio`` + or ``-audio driver=pa,model=virtio,server=/run/user/1000/pulse/native`` + * sdl: ``-audio driver=sdl,model=virtio`` + * coreaudio: ``-audio driver=coreaudio,model=virtio`` + +etc. + +To specifically add virtualized sound devices, you have to specify a PCI device +and an audio backend listed with ``-audio driver=help`` that works on your host +machine, e.g.: + +:: + + -device virtio-sound-pci,audiodev=my_audiodev \ + -audiodev alsa,id=my_audiodev -- cgit v1.2.3 From 8b98c15f22fca34f23846e70752a70e4a2469317 Mon Sep 17 00:00:00 2001 From: Hawkins Jiawei Date: Wed, 25 Oct 2023 09:02:24 +0800 Subject: vdpa: Restore hash calculation state MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch introduces vhost_vdpa_net_load_rss() to restore the hash calculation state at device's startup. Signed-off-by: Hawkins Jiawei Message-Id: Acked-by: Eugenio Pérez Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- net/vhost-vdpa.c | 91 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 91 insertions(+) diff --git a/net/vhost-vdpa.c b/net/vhost-vdpa.c index 7a226c93bc..e59d40b8ae 100644 --- a/net/vhost-vdpa.c +++ b/net/vhost-vdpa.c @@ -818,6 +818,88 @@ static int vhost_vdpa_net_load_mac(VhostVDPAState *s, const VirtIONet *n, return 0; } +static int vhost_vdpa_net_load_rss(VhostVDPAState *s, const VirtIONet *n, + struct iovec *out_cursor, + struct iovec *in_cursor) +{ + struct virtio_net_rss_config cfg = {}; + ssize_t r; + g_autofree uint16_t *table = NULL; + + /* + * According to VirtIO standard, "Initially the device has all hash + * types disabled and reports only VIRTIO_NET_HASH_REPORT_NONE.". + * + * Therefore, there is no need to send this CVQ command if the + * driver disables the all hash types, which aligns with + * the device's defaults. + * + * Note that the device's defaults can mismatch the driver's + * configuration only at live migration. + */ + if (!n->rss_data.enabled || + n->rss_data.hash_types == VIRTIO_NET_HASH_REPORT_NONE) { + return 0; + } + + table = g_malloc_n(n->rss_data.indirections_len, + sizeof(n->rss_data.indirections_table[0])); + cfg.hash_types = cpu_to_le32(n->rss_data.hash_types); + + /* + * According to VirtIO standard, "Field reserved MUST contain zeroes. + * It is defined to make the structure to match the layout of + * virtio_net_rss_config structure, defined in 5.1.6.5.7.". + * + * Therefore, we need to zero the fields in + * struct virtio_net_rss_config, which corresponds to the + * `reserved` field in struct virtio_net_hash_config. + * + * Note that all other fields are zeroed at their definitions, + * except for the `indirection_table` field, where the actual data + * is stored in the `table` variable to ensure compatibility + * with RSS case. Therefore, we need to zero the `table` variable here. + */ + table[0] = 0; + + /* + * Considering that virtio_net_handle_rss() currently does not restore + * the hash key length parsed from the CVQ command sent from the guest + * into n->rss_data and uses the maximum key length in other code, so + * we also employ the maximum key length here. + */ + cfg.hash_key_length = sizeof(n->rss_data.key); + + const struct iovec data[] = { + { + .iov_base = &cfg, + .iov_len = offsetof(struct virtio_net_rss_config, + indirection_table), + }, { + .iov_base = table, + .iov_len = n->rss_data.indirections_len * + sizeof(n->rss_data.indirections_table[0]), + }, { + .iov_base = &cfg.max_tx_vq, + .iov_len = offsetof(struct virtio_net_rss_config, hash_key_data) - + offsetof(struct virtio_net_rss_config, max_tx_vq), + }, { + .iov_base = (void *)n->rss_data.key, + .iov_len = sizeof(n->rss_data.key), + } + }; + + r = vhost_vdpa_net_load_cmd(s, out_cursor, in_cursor, + VIRTIO_NET_CTRL_MQ, + VIRTIO_NET_CTRL_MQ_HASH_CONFIG, + data, ARRAY_SIZE(data)); + if (unlikely(r < 0)) { + return r; + } + + return 0; +} + static int vhost_vdpa_net_load_mq(VhostVDPAState *s, const VirtIONet *n, struct iovec *out_cursor, @@ -843,6 +925,15 @@ static int vhost_vdpa_net_load_mq(VhostVDPAState *s, return r; } + if (!virtio_vdev_has_feature(&n->parent_obj, VIRTIO_NET_F_HASH_REPORT)) { + return 0; + } + + r = vhost_vdpa_net_load_rss(s, n, out_cursor, in_cursor); + if (unlikely(r < 0)) { + return r; + } + return 0; } -- cgit v1.2.3 From 556b67d413a699431eadb71642033864649ea934 Mon Sep 17 00:00:00 2001 From: Hawkins Jiawei Date: Wed, 25 Oct 2023 09:02:25 +0800 Subject: vdpa: Allow VIRTIO_NET_F_HASH_REPORT in SVQ MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Enable SVQ with VIRTIO_NET_F_HASH_REPORT feature. Signed-off-by: Hawkins Jiawei Message-Id: Acked-by: Eugenio Pérez Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- net/vhost-vdpa.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/vhost-vdpa.c b/net/vhost-vdpa.c index e59d40b8ae..54f748d49d 100644 --- a/net/vhost-vdpa.c +++ b/net/vhost-vdpa.c @@ -121,6 +121,7 @@ static const uint64_t vdpa_svq_device_features = BIT_ULL(VIRTIO_NET_F_CTRL_MAC_ADDR) | /* VHOST_F_LOG_ALL is exposed by SVQ */ BIT_ULL(VHOST_F_LOG_ALL) | + BIT_ULL(VIRTIO_NET_F_HASH_REPORT) | BIT_ULL(VIRTIO_NET_F_RSC_EXT) | BIT_ULL(VIRTIO_NET_F_STANDBY) | BIT_ULL(VIRTIO_NET_F_SPEED_DUPLEX); -- cgit v1.2.3 From d1fd2d3118a42f36d86efae7ffbdce79f7024584 Mon Sep 17 00:00:00 2001 From: Hawkins Jiawei Date: Wed, 25 Oct 2023 09:08:04 +0800 Subject: vdpa: Add SetSteeringEBPF method for NetClientState MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit At present, to enable the VIRTIO_NET_F_RSS feature, eBPF must be loaded for the vhost backend. Given that vhost-vdpa is one of the vhost backend, we need to implement the SetSteeringEBPF method to support RSS for vhost-vdpa, even if vhost-vdpa calculates the rss hash in the hardware device instead of in the kernel by eBPF. Although this requires QEMU to be compiled with `--enable-bpf` configuration even if the vdpa device does not use eBPF to calculate the rss hash, this can avoid adding the specific conditional statements for vDPA case to enable the VIRTIO_NET_F_RSS feature, which reduces code maintainbility. Suggested-by: Eugenio Pérez Signed-off-by: Hawkins Jiawei Message-Id: <280e20ddce55b6de60f1552ba0865bffffe909b2.1698195059.git.yin31149@gmail.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- net/vhost-vdpa.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/net/vhost-vdpa.c b/net/vhost-vdpa.c index 54f748d49d..3466936b87 100644 --- a/net/vhost-vdpa.c +++ b/net/vhost-vdpa.c @@ -241,6 +241,12 @@ static void vhost_vdpa_cleanup(NetClientState *nc) } } +/** Dummy SetSteeringEBPF to support RSS for vhost-vdpa backend */ +static bool vhost_vdpa_set_steering_ebpf(NetClientState *nc, int prog_fd) +{ + return true; +} + static bool vhost_vdpa_has_vnet_hdr(NetClientState *nc) { assert(nc->info->type == NET_CLIENT_DRIVER_VHOST_VDPA); @@ -423,6 +429,7 @@ static NetClientInfo net_vhost_vdpa_info = { .has_vnet_hdr = vhost_vdpa_has_vnet_hdr, .has_ufo = vhost_vdpa_has_ufo, .check_peer_type = vhost_vdpa_check_peer_type, + .set_steering_ebpf = vhost_vdpa_set_steering_ebpf, }; static int64_t vhost_vdpa_get_vring_group(int device_fd, unsigned vq_index, @@ -1258,6 +1265,7 @@ static NetClientInfo net_vhost_vdpa_cvq_info = { .has_vnet_hdr = vhost_vdpa_has_vnet_hdr, .has_ufo = vhost_vdpa_has_ufo, .check_peer_type = vhost_vdpa_check_peer_type, + .set_steering_ebpf = vhost_vdpa_set_steering_ebpf, }; /* -- cgit v1.2.3 From b3c09106559f9d84a940d2a27c4cfc81c9e15347 Mon Sep 17 00:00:00 2001 From: Hawkins Jiawei Date: Wed, 25 Oct 2023 09:08:05 +0800 Subject: vdpa: Restore receive-side scaling state MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch reuses vhost_vdpa_net_load_rss() with some refactorings to restore the receive-side scaling state at device's startup. Signed-off-by: Hawkins Jiawei Message-Id: Acked-by: Eugenio Pérez Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- net/vhost-vdpa.c | 67 +++++++++++++++++++++++++++++++++++++------------------- 1 file changed, 44 insertions(+), 23 deletions(-) diff --git a/net/vhost-vdpa.c b/net/vhost-vdpa.c index 3466936b87..a4cc1381fc 100644 --- a/net/vhost-vdpa.c +++ b/net/vhost-vdpa.c @@ -828,7 +828,7 @@ static int vhost_vdpa_net_load_mac(VhostVDPAState *s, const VirtIONet *n, static int vhost_vdpa_net_load_rss(VhostVDPAState *s, const VirtIONet *n, struct iovec *out_cursor, - struct iovec *in_cursor) + struct iovec *in_cursor, bool do_rss) { struct virtio_net_rss_config cfg = {}; ssize_t r; @@ -854,21 +854,35 @@ static int vhost_vdpa_net_load_rss(VhostVDPAState *s, const VirtIONet *n, sizeof(n->rss_data.indirections_table[0])); cfg.hash_types = cpu_to_le32(n->rss_data.hash_types); - /* - * According to VirtIO standard, "Field reserved MUST contain zeroes. - * It is defined to make the structure to match the layout of - * virtio_net_rss_config structure, defined in 5.1.6.5.7.". - * - * Therefore, we need to zero the fields in - * struct virtio_net_rss_config, which corresponds to the - * `reserved` field in struct virtio_net_hash_config. - * - * Note that all other fields are zeroed at their definitions, - * except for the `indirection_table` field, where the actual data - * is stored in the `table` variable to ensure compatibility - * with RSS case. Therefore, we need to zero the `table` variable here. - */ - table[0] = 0; + if (do_rss) { + /* + * According to VirtIO standard, "Number of entries in indirection_table + * is (indirection_table_mask + 1)". + */ + cfg.indirection_table_mask = cpu_to_le16(n->rss_data.indirections_len - + 1); + cfg.unclassified_queue = cpu_to_le16(n->rss_data.default_queue); + for (int i = 0; i < n->rss_data.indirections_len; ++i) { + table[i] = cpu_to_le16(n->rss_data.indirections_table[i]); + } + cfg.max_tx_vq = cpu_to_le16(n->curr_queue_pairs); + } else { + /* + * According to VirtIO standard, "Field reserved MUST contain zeroes. + * It is defined to make the structure to match the layout of + * virtio_net_rss_config structure, defined in 5.1.6.5.7.". + * + * Therefore, we need to zero the fields in + * struct virtio_net_rss_config, which corresponds to the + * `reserved` field in struct virtio_net_hash_config. + * + * Note that all other fields are zeroed at their definitions, + * except for the `indirection_table` field, where the actual data + * is stored in the `table` variable to ensure compatibility + * with RSS case. Therefore, we need to zero the `table` variable here. + */ + table[0] = 0; + } /* * Considering that virtio_net_handle_rss() currently does not restore @@ -899,6 +913,7 @@ static int vhost_vdpa_net_load_rss(VhostVDPAState *s, const VirtIONet *n, r = vhost_vdpa_net_load_cmd(s, out_cursor, in_cursor, VIRTIO_NET_CTRL_MQ, + do_rss ? VIRTIO_NET_CTRL_MQ_RSS_CONFIG : VIRTIO_NET_CTRL_MQ_HASH_CONFIG, data, ARRAY_SIZE(data)); if (unlikely(r < 0)) { @@ -933,13 +948,19 @@ static int vhost_vdpa_net_load_mq(VhostVDPAState *s, return r; } - if (!virtio_vdev_has_feature(&n->parent_obj, VIRTIO_NET_F_HASH_REPORT)) { - return 0; - } - - r = vhost_vdpa_net_load_rss(s, n, out_cursor, in_cursor); - if (unlikely(r < 0)) { - return r; + if (virtio_vdev_has_feature(&n->parent_obj, VIRTIO_NET_F_RSS)) { + /* load the receive-side scaling state */ + r = vhost_vdpa_net_load_rss(s, n, out_cursor, in_cursor, true); + if (unlikely(r < 0)) { + return r; + } + } else if (virtio_vdev_has_feature(&n->parent_obj, + VIRTIO_NET_F_HASH_REPORT)) { + /* load the hash calculation state */ + r = vhost_vdpa_net_load_rss(s, n, out_cursor, in_cursor, false); + if (unlikely(r < 0)) { + return r; + } } return 0; -- cgit v1.2.3 From 07eba9493d14b3ba371e64392effc384246892ea Mon Sep 17 00:00:00 2001 From: Hawkins Jiawei Date: Wed, 25 Oct 2023 09:08:06 +0800 Subject: vdpa: Allow VIRTIO_NET_F_RSS in SVQ MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Enable SVQ with VIRTIO_NET_F_RSS feature. Signed-off-by: Hawkins Jiawei Message-Id: <626449eb303207de408126b3dc7c155cd72b028b.1698195059.git.yin31149@gmail.com> Acked-by: Eugenio Pérez Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- net/vhost-vdpa.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/vhost-vdpa.c b/net/vhost-vdpa.c index a4cc1381fc..d0614d7954 100644 --- a/net/vhost-vdpa.c +++ b/net/vhost-vdpa.c @@ -122,6 +122,7 @@ static const uint64_t vdpa_svq_device_features = /* VHOST_F_LOG_ALL is exposed by SVQ */ BIT_ULL(VHOST_F_LOG_ALL) | BIT_ULL(VIRTIO_NET_F_HASH_REPORT) | + BIT_ULL(VIRTIO_NET_F_RSS) | BIT_ULL(VIRTIO_NET_F_RSC_EXT) | BIT_ULL(VIRTIO_NET_F_STANDBY) | BIT_ULL(VIRTIO_NET_F_SPEED_DUPLEX); -- cgit v1.2.3 From 7d5936791ea349d9cef856babcce28470966bfa4 Mon Sep 17 00:00:00 2001 From: Zhao Liu Date: Mon, 23 Oct 2023 17:46:20 +0800 Subject: tests: test-smp-parse: Add the test for cores/threads per socket helpers Use the different ways to calculate cores/threads per socket, so that the new CPU topology levels won't be missed in these 2 helpes: * machine_topo_get_cores_per_socket() * machine_topo_get_threads_per_socket() Test the commit a1d027be95bc3 ("machine: Add helpers to get cores/ threads per socket"). Suggested-by: Igor Mammedov Signed-off-by: Zhao Liu Acked-by: Igor Mammedov Reviewed-by: Michael S. Tsirkin Message-Id: <20231023094635.1588282-2-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/unit/test-smp-parse.c | 67 ++++++++++++++++++++++++++++++++++++--------- 1 file changed, 54 insertions(+), 13 deletions(-) diff --git a/tests/unit/test-smp-parse.c b/tests/unit/test-smp-parse.c index fdc39a846c..24972666a7 100644 --- a/tests/unit/test-smp-parse.c +++ b/tests/unit/test-smp-parse.c @@ -394,20 +394,47 @@ static char *smp_config_to_string(const SMPConfiguration *config) config->has_maxcpus ? "true" : "false", config->maxcpus); } -static char *cpu_topology_to_string(const CpuTopology *topo) +/* Use the different calculation than machine_topo_get_threads_per_socket(). */ +static unsigned int cpu_topology_get_threads_per_socket(const CpuTopology *topo) +{ + /* Check the divisor to avoid invalid topology examples causing SIGFPE. */ + if (!topo->sockets) { + return 0; + } else { + return topo->max_cpus / topo->sockets; + } +} + +/* Use the different calculation than machine_topo_get_cores_per_socket(). */ +static unsigned int cpu_topology_get_cores_per_socket(const CpuTopology *topo) +{ + /* Check the divisor to avoid invalid topology examples causing SIGFPE. */ + if (!topo->threads) { + return 0; + } else { + return cpu_topology_get_threads_per_socket(topo) / topo->threads; + } +} + +static char *cpu_topology_to_string(const CpuTopology *topo, + unsigned int threads_per_socket, + unsigned int cores_per_socket) { return g_strdup_printf( "(CpuTopology) {\n" - " .cpus = %u,\n" - " .sockets = %u,\n" - " .dies = %u,\n" - " .clusters = %u,\n" - " .cores = %u,\n" - " .threads = %u,\n" - " .max_cpus = %u,\n" + " .cpus = %u,\n" + " .sockets = %u,\n" + " .dies = %u,\n" + " .clusters = %u,\n" + " .cores = %u,\n" + " .threads = %u,\n" + " .max_cpus = %u,\n" + " .threads_per_socket = %u,\n" + " .cores_per_socket = %u,\n" "}", topo->cpus, topo->sockets, topo->dies, topo->clusters, - topo->cores, topo->threads, topo->max_cpus); + topo->cores, topo->threads, topo->max_cpus, + threads_per_socket, cores_per_socket); } static void check_parse(MachineState *ms, const SMPConfiguration *config, @@ -415,14 +442,26 @@ static void check_parse(MachineState *ms, const SMPConfiguration *config, bool is_valid) { g_autofree char *config_str = smp_config_to_string(config); - g_autofree char *expect_topo_str = cpu_topology_to_string(expect_topo); - g_autofree char *output_topo_str = NULL; + g_autofree char *expect_topo_str = NULL, *output_topo_str = NULL; + unsigned int expect_threads_per_socket, expect_cores_per_socket; + unsigned int ms_threads_per_socket, ms_cores_per_socket; Error *err = NULL; + expect_threads_per_socket = + cpu_topology_get_threads_per_socket(expect_topo); + expect_cores_per_socket = + cpu_topology_get_cores_per_socket(expect_topo); + expect_topo_str = cpu_topology_to_string(expect_topo, + expect_threads_per_socket, + expect_cores_per_socket); + /* call the generic parser */ machine_parse_smp_config(ms, config, &err); - output_topo_str = cpu_topology_to_string(&ms->smp); + ms_threads_per_socket = machine_topo_get_threads_per_socket(ms); + ms_cores_per_socket = machine_topo_get_cores_per_socket(ms); + output_topo_str = cpu_topology_to_string(&ms->smp, ms_threads_per_socket, + ms_cores_per_socket); /* when the configuration is supposed to be valid */ if (is_valid) { @@ -433,7 +472,9 @@ static void check_parse(MachineState *ms, const SMPConfiguration *config, (ms->smp.clusters == expect_topo->clusters) && (ms->smp.cores == expect_topo->cores) && (ms->smp.threads == expect_topo->threads) && - (ms->smp.max_cpus == expect_topo->max_cpus)) { + (ms->smp.max_cpus == expect_topo->max_cpus) && + (ms_threads_per_socket == expect_threads_per_socket) && + (ms_cores_per_socket == expect_cores_per_socket)) { return; } -- cgit v1.2.3 From 6c7937ece909d3df0d2b4e94909f57388f5ed666 Mon Sep 17 00:00:00 2001 From: Zhao Liu Date: Mon, 23 Oct 2023 17:46:21 +0800 Subject: tests: bios-tables-test: Prepare the ACPI table change for smbios type4 count test Following the guidelines in tests/qtest/bios-tables-test.c, this is step 1 - 3. List the ACPI tables that will be added to test the type 4 count. Signed-off-by: Zhao Liu Reviewed-by: Michael S. Tsirkin Message-Id: <20231023094635.1588282-3-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/data/acpi/q35/APIC.type4-count | 0 tests/data/acpi/q35/DSDT.type4-count | 0 tests/data/acpi/q35/FACP.type4-count | 0 tests/qtest/bios-tables-test-allowed-diff.h | 3 +++ 4 files changed, 3 insertions(+) create mode 100644 tests/data/acpi/q35/APIC.type4-count create mode 100644 tests/data/acpi/q35/DSDT.type4-count create mode 100644 tests/data/acpi/q35/FACP.type4-count diff --git a/tests/data/acpi/q35/APIC.type4-count b/tests/data/acpi/q35/APIC.type4-count new file mode 100644 index 0000000000..e69de29bb2 diff --git a/tests/data/acpi/q35/DSDT.type4-count b/tests/data/acpi/q35/DSDT.type4-count new file mode 100644 index 0000000000..e69de29bb2 diff --git a/tests/data/acpi/q35/FACP.type4-count b/tests/data/acpi/q35/FACP.type4-count new file mode 100644 index 0000000000..e69de29bb2 diff --git a/tests/qtest/bios-tables-test-allowed-diff.h b/tests/qtest/bios-tables-test-allowed-diff.h index dfb8523c8b..0ce6f8fc72 100644 --- a/tests/qtest/bios-tables-test-allowed-diff.h +++ b/tests/qtest/bios-tables-test-allowed-diff.h @@ -1 +1,4 @@ /* List of comma-separated changed AML files to ignore */ +"tests/data/acpi/q35/APIC.type4-count", +"tests/data/acpi/q35/DSDT.type4-count", +"tests/data/acpi/q35/FACP.type4-count", -- cgit v1.2.3 From df210963a10e5ca884908addeb07aa081c5b7ad2 Mon Sep 17 00:00:00 2001 From: Zhao Liu Date: Mon, 23 Oct 2023 17:46:22 +0800 Subject: tests: bios-tables-test: Add test for smbios type4 count This tests the commit d79a284a44bb7 ("hw/smbios: Fix smbios_smp_sockets calculation"). In smbios_get_tables() (hw/smbios/smbios.c), smbios type4 table is built for each socket, so the count of type4 tables should be equal to the number of sockets. Thus for the topology in this case, there're the following considerations: 1. The topology should include multiple sockets to ensure smbios could create type4 tables for each socket. 2. In addition to sockets, for the more general topology, we should also configure as many topology levels as possible (multiple dies, no module since x86 hasn't supported it), to ensure that smbios is able to exclude the effect of other topology levels to create the type4 tables only for sockets. 3. The original miscalculation bug also misused "smp.cpus", so it's necessary to configure "cpus" (presented threads for machine) and "maxcpus" (total threads for machine) as well to make sure that configuring unpluged CPUs in smp (cpus < maxcpus) does not affect the correctness of the count of type4 tables. Based on these considerations, select the topology as the follow: -smp cpus=100,maxcpus=120,sockets=5,dies=2,cores=4,threads=3 The expected count of type4 tables = sockets (5). Suggested-by: Igor Mammedov Signed-off-by: Zhao Liu Reviewed-by: Michael S. Tsirkin Message-Id: <20231023094635.1588282-4-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/qtest/bios-tables-test.c | 33 ++++++++++++++++++++++++++++++++- 1 file changed, 32 insertions(+), 1 deletion(-) diff --git a/tests/qtest/bios-tables-test.c b/tests/qtest/bios-tables-test.c index 9f4bc15aab..cdbfb51559 100644 --- a/tests/qtest/bios-tables-test.c +++ b/tests/qtest/bios-tables-test.c @@ -97,6 +97,7 @@ typedef struct { uint16_t smbios_core_count2; uint8_t *required_struct_types; int required_struct_types_len; + int type4_count; QTestState *qts; } test_data; @@ -673,12 +674,21 @@ static void smbios_cpu_test(test_data *data, uint32_t addr, } } +static void smbios_type4_count_test(test_data *data, int type4_count) +{ + int expected_type4_count = data->type4_count; + + if (expected_type4_count) { + g_assert_cmpuint(type4_count, ==, expected_type4_count); + } +} + static void test_smbios_structs(test_data *data, SmbiosEntryPointType ep_type) { DECLARE_BITMAP(struct_bitmap, SMBIOS_MAX_TYPE+1) = { 0 }; SmbiosEntryPoint *ep_table = &data->smbios_ep_table; - int i = 0, len, max_len = 0; + int i = 0, len, max_len = 0, type4_count = 0; uint8_t type, prv, crt; uint64_t addr; @@ -704,6 +714,7 @@ static void test_smbios_structs(test_data *data, SmbiosEntryPointType ep_type) if (type == 4) { smbios_cpu_test(data, addr, ep_type); + type4_count++; } /* seek to end of unformatted string area of this struct ("\0\0") */ @@ -747,6 +758,8 @@ static void test_smbios_structs(test_data *data, SmbiosEntryPointType ep_type) for (i = 0; i < data->required_struct_types_len; i++) { g_assert(test_bit(data->required_struct_types[i], struct_bitmap)); } + + smbios_type4_count_test(data, type4_count); } static void test_acpi_load_tables(test_data *data) @@ -970,6 +983,22 @@ static void test_acpi_q35_tcg(void) free_test_data(&data); } +static void test_acpi_q35_tcg_type4_count(void) +{ + test_data data = { + .machine = MACHINE_Q35, + .variant = ".type4-count", + .required_struct_types = base_required_struct_types, + .required_struct_types_len = ARRAY_SIZE(base_required_struct_types), + .type4_count = 5, + }; + + test_acpi_one("-machine smbios-entry-point-type=64 " + "-smp cpus=100,maxcpus=120,sockets=5," + "dies=2,cores=4,threads=3", &data); + free_test_data(&data); +} + static void test_acpi_q35_tcg_core_count2(void) { test_data data = { @@ -2147,6 +2176,8 @@ int main(int argc, char *argv[]) if (has_kvm) { qtest_add_func("acpi/q35/kvm/xapic", test_acpi_q35_kvm_xapic); qtest_add_func("acpi/q35/kvm/dmar", test_acpi_q35_kvm_dmar); + qtest_add_func("acpi/q35/type4-count", + test_acpi_q35_tcg_type4_count); qtest_add_func("acpi/q35/core-count2", test_acpi_q35_tcg_core_count2); } -- cgit v1.2.3 From c1cd1d360db857d6f76071e8de6af97cdab2bdfe Mon Sep 17 00:00:00 2001 From: Zhao Liu Date: Mon, 23 Oct 2023 17:46:23 +0800 Subject: tests: bios-tables-test: Add ACPI table binaries for smbios type4 count test Following the guidelines in tests/qtest/bios-tables-test.c, this is step 5 and 6. Changes in the tables: FACP: +/* + * Intel ACPI Component Architecture + * AML/ASL+ Disassembler version 20200925 (64-bit version) + * Copyright (c) 2000 - 2020 Intel Corporation + * + * Disassembly of /tmp/aml-W37791, Wed Aug 23 10:36:32 2023 + * + * ACPI Data Table [FACP] + * + * Format: [HexOffset DecimalOffset ByteLength] FieldName : FieldValue + */ + +[000h 0000 4] Signature : "FACP" [Fixed ACPI Description Table (FADT)] +[004h 0004 4] Table Length : 000000F4 +[008h 0008 1] Revision : 03 +[009h 0009 1] Checksum : B3 +[00Ah 0010 6] Oem ID : "BOCHS " +[010h 0016 8] Oem Table ID : "BXPC " +[018h 0024 4] Oem Revision : 00000001 +[01Ch 0028 4] Asl Compiler ID : "BXPC" +[020h 0032 4] Asl Compiler Revision : 00000001 + +[024h 0036 4] FACS Address : 00000000 +[028h 0040 4] DSDT Address : 00000000 +[02Ch 0044 1] Model : 01 +[02Dh 0045 1] PM Profile : 00 [Unspecified] +[02Eh 0046 2] SCI Interrupt : 0009 +[030h 0048 4] SMI Command Port : 000000B2 +[034h 0052 1] ACPI Enable Value : 02 +[035h 0053 1] ACPI Disable Value : 03 +[036h 0054 1] S4BIOS Command : 00 +[037h 0055 1] P-State Control : 00 +[038h 0056 4] PM1A Event Block Address : 00000600 +[03Ch 0060 4] PM1B Event Block Address : 00000000 +[040h 0064 4] PM1A Control Block Address : 00000604 +[044h 0068 4] PM1B Control Block Address : 00000000 +[048h 0072 4] PM2 Control Block Address : 00000000 +[04Ch 0076 4] PM Timer Block Address : 00000608 +[050h 0080 4] GPE0 Block Address : 00000620 +[054h 0084 4] GPE1 Block Address : 00000000 +[058h 0088 1] PM1 Event Block Length : 04 +[059h 0089 1] PM1 Control Block Length : 02 +[05Ah 0090 1] PM2 Control Block Length : 00 +[05Bh 0091 1] PM Timer Block Length : 04 +[05Ch 0092 1] GPE0 Block Length : 10 +[05Dh 0093 1] GPE1 Block Length : 00 +[05Eh 0094 1] GPE1 Base Offset : 00 +[05Fh 0095 1] _CST Support : 00 +[060h 0096 2] C2 Latency : 0FFF +[062h 0098 2] C3 Latency : 0FFF +[064h 0100 2] CPU Cache Size : 0000 +[066h 0102 2] Cache Flush Stride : 0000 +[068h 0104 1] Duty Cycle Offset : 00 +[069h 0105 1] Duty Cycle Width : 00 +[06Ah 0106 1] RTC Day Alarm Index : 00 +[06Bh 0107 1] RTC Month Alarm Index : 00 +[06Ch 0108 1] RTC Century Index : 32 +[06Dh 0109 2] Boot Flags (decoded below) : 0002 + Legacy Devices Supported (V2) : 0 + 8042 Present on ports 60/64 (V2) : 1 + VGA Not Present (V4) : 0 + MSI Not Supported (V4) : 0 + PCIe ASPM Not Supported (V4) : 0 + CMOS RTC Not Present (V5) : 0 +[06Fh 0111 1] Reserved : 00 +[070h 0112 4] Flags (decoded below) : 000484A5 + WBINVD instruction is operational (V1) : 1 + WBINVD flushes all caches (V1) : 0 + All CPUs support C1 (V1) : 1 + C2 works on MP system (V1) : 0 + Control Method Power Button (V1) : 0 + Control Method Sleep Button (V1) : 1 + RTC wake not in fixed reg space (V1) : 0 + RTC can wake system from S4 (V1) : 1 + 32-bit PM Timer (V1) : 0 + Docking Supported (V1) : 0 + Reset Register Supported (V2) : 1 + Sealed Case (V3) : 0 + Headless - No Video (V3) : 0 + Use native instr after SLP_TYPx (V3) : 0 + PCIEXP_WAK Bits Supported (V4) : 0 + Use Platform Timer (V4) : 1 + RTC_STS valid on S4 wake (V4) : 0 + Remote Power-on capable (V4) : 0 + Use APIC Cluster Model (V4) : 1 + Use APIC Physical Destination Mode (V4) : 0 + Hardware Reduced (V5) : 0 + Low Power S0 Idle (V5) : 0 + +[074h 0116 12] Reset Register : [Generic Address Structure] +[074h 0116 1] Space ID : 01 [SystemIO] +[075h 0117 1] Bit Width : 08 +[076h 0118 1] Bit Offset : 00 +[077h 0119 1] Encoded Access Width : 00 [Undefined/Legacy] +[078h 0120 8] Address : 0000000000000CF9 + +[080h 0128 1] Value to cause reset : 0F +[081h 0129 2] ARM Flags (decoded below) : 0000 + PSCI Compliant : 0 + Must use HVC for PSCI : 0 + +[083h 0131 1] FADT Minor Revision : 00 +[084h 0132 8] FACS Address : 0000000000000000 +[08Ch 0140 8] DSDT Address : 0000000000000000 +[094h 0148 12] PM1A Event Block : [Generic Address Structure] +[094h 0148 1] Space ID : 01 [SystemIO] +[095h 0149 1] Bit Width : 20 +[096h 0150 1] Bit Offset : 00 +[097h 0151 1] Encoded Access Width : 00 [Undefined/Legacy] +[098h 0152 8] Address : 0000000000000600 + +[0A0h 0160 12] PM1B Event Block : [Generic Address Structure] +[0A0h 0160 1] Space ID : 00 [SystemMemory] +[0A1h 0161 1] Bit Width : 00 +[0A2h 0162 1] Bit Offset : 00 +[0A3h 0163 1] Encoded Access Width : 00 [Undefined/Legacy] +[0A4h 0164 8] Address : 0000000000000000 + +[0ACh 0172 12] PM1A Control Block : [Generic Address Structure] +[0ACh 0172 1] Space ID : 01 [SystemIO] +[0ADh 0173 1] Bit Width : 10 +[0AEh 0174 1] Bit Offset : 00 +[0AFh 0175 1] Encoded Access Width : 00 [Undefined/Legacy] +[0B0h 0176 8] Address : 0000000000000604 + +[0B8h 0184 12] PM1B Control Block : [Generic Address Structure] +[0B8h 0184 1] Space ID : 00 [SystemMemory] +[0B9h 0185 1] Bit Width : 00 +[0BAh 0186 1] Bit Offset : 00 +[0BBh 0187 1] Encoded Access Width : 00 [Undefined/Legacy] +[0BCh 0188 8] Address : 0000000000000000 + +[0C4h 0196 12] PM2 Control Block : [Generic Address Structure] +[0C4h 0196 1] Space ID : 00 [SystemMemory] +[0C5h 0197 1] Bit Width : 00 +[0C6h 0198 1] Bit Offset : 00 +[0C7h 0199 1] Encoded Access Width : 00 [Undefined/Legacy] +[0C8h 0200 8] Address : 0000000000000000 + +[0D0h 0208 12] PM Timer Block : [Generic Address Structure] +[0D0h 0208 1] Space ID : 01 [SystemIO] +[0D1h 0209 1] Bit Width : 20 +[0D2h 0210 1] Bit Offset : 00 +[0D3h 0211 1] Encoded Access Width : 00 [Undefined/Legacy] +[0D4h 0212 8] Address : 0000000000000608 + +[0DCh 0220 12] GPE0 Block : [Generic Address Structure] +[0DCh 0220 1] Space ID : 01 [SystemIO] +[0DDh 0221 1] Bit Width : 80 +[0DEh 0222 1] Bit Offset : 00 +[0DFh 0223 1] Encoded Access Width : 00 [Undefined/Legacy] +[0E0h 0224 8] Address : 0000000000000620 + +[0E8h 0232 12] GPE1 Block : [Generic Address Structure] +[0E8h 0232 1] Space ID : 00 [SystemMemory] +[0E9h 0233 1] Bit Width : 00 +[0EAh 0234 1] Bit Offset : 00 +[0EBh 0235 1] Encoded Access Width : 00 [Undefined/Legacy] +[0ECh 0236 8] Address : 0000000000000000 + ... APIC: +/* + * Intel ACPI Component Architecture + * AML/ASL+ Disassembler version 20200925 (64-bit version) + * Copyright (c) 2000 - 2020 Intel Corporation + * + * Disassembly of /tmp/aml-687791, Wed Aug 23 10:36:32 2023 + * + * ACPI Data Table [APIC] + * + * Format: [HexOffset DecimalOffset ByteLength] FieldName : FieldValue + */ + +[000h 0000 4] Signature : "APIC" [Multiple APIC Description Table (MADT)] +[004h 0004 4] Table Length : 00000430 +[008h 0008 1] Revision : 03 +[009h 0009 1] Checksum : C5 +[00Ah 0010 6] Oem ID : "BOCHS " +[010h 0016 8] Oem Table ID : "BXPC " +[018h 0024 4] Oem Revision : 00000001 +[01Ch 0028 4] Asl Compiler ID : "BXPC" +[020h 0032 4] Asl Compiler Revision : 00000001 + +[024h 0036 4] Local Apic Address : FEE00000 +[028h 0040 4] Flags (decoded below) : 00000001 + PC-AT Compatibility : 1 + +[02Ch 0044 1] Subtable Type : 00 [Processor Local APIC] +[02Dh 0045 1] Length : 08 +[02Eh 0046 1] Processor ID : 00 +[02Fh 0047 1] Local Apic ID : 00 +[030h 0048 4] Flags (decoded below) : 00000001 + Processor Enabled : 1 + Runtime Online Capable : 0 + +[034h 0052 1] Subtable Type : 00 [Processor Local APIC] +[035h 0053 1] Length : 08 +[036h 0054 1] Processor ID : 01 +[037h 0055 1] Local Apic ID : 01 +[038h 0056 4] Flags (decoded below) : 00000001 + Processor Enabled : 1 + Runtime Online Capable : 0 [snip] +[3E4h 0996 1] Subtable Type : 00 [Processor Local APIC] +[3E5h 0997 1] Length : 08 +[3E6h 0998 1] Processor ID : 77 +[3E7h 0999 1] Local Apic ID : 9E +[3E8h 1000 4] Flags (decoded below) : 00000000 + Processor Enabled : 0 + Runtime Online Capable : 0 + +[3ECh 1004 1] Subtable Type : 01 [I/O APIC] +[3EDh 1005 1] Length : 0C +[3EEh 1006 1] I/O Apic ID : 00 +[3EFh 1007 1] Reserved : 00 +[3F0h 1008 4] Address : FEC00000 +[3F4h 1012 4] Interrupt : 00000000 + +[3F8h 1016 1] Subtable Type : 02 [Interrupt Source Override] +[3F9h 1017 1] Length : 0A +[3FAh 1018 1] Bus : 00 +[3FBh 1019 1] Source : 00 +[3FCh 1020 4] Interrupt : 00000002 +[400h 1024 2] Flags (decoded below) : 0000 + Polarity : 0 + Trigger Mode : 0 + +[402h 1026 1] Subtable Type : 02 [Interrupt Source Override] +[403h 1027 1] Length : 0A +[404h 1028 1] Bus : 00 +[405h 1029 1] Source : 05 +[406h 1030 4] Interrupt : 00000005 +[40Ah 1034 2] Flags (decoded below) : 000D + Polarity : 1 + Trigger Mode : 3 + +[40Ch 1036 1] Subtable Type : 02 [Interrupt Source Override] +[40Dh 1037 1] Length : 0A +[40Eh 1038 1] Bus : 00 +[40Fh 1039 1] Source : 09 +[410h 1040 4] Interrupt : 00000009 +[414h 1044 2] Flags (decoded below) : 000D + Polarity : 1 + Trigger Mode : 3 + +[416h 1046 1] Subtable Type : 02 [Interrupt Source Override] +[417h 1047 1] Length : 0A +[418h 1048 1] Bus : 00 +[419h 1049 1] Source : 0A +[41Ah 1050 4] Interrupt : 0000000A +[41Eh 1054 2] Flags (decoded below) : 000D + Polarity : 1 + Trigger Mode : 3 + +[420h 1056 1] Subtable Type : 02 [Interrupt Source Override] +[421h 1057 1] Length : 0A +[422h 1058 1] Bus : 00 +[423h 1059 1] Source : 0B +[424h 1060 4] Interrupt : 0000000B +[428h 1064 2] Flags (decoded below) : 000D + Polarity : 1 + Trigger Mode : 3 + +[42Ah 1066 1] Subtable Type : 04 [Local APIC NMI] +[42Bh 1067 1] Length : 06 +[42Ch 1068 1] Processor ID : FF +[42Dh 1069 2] Flags (decoded below) : 0000 + Polarity : 0 + Trigger Mode : 0 +[42Fh 1071 1] Interrupt Input LINT : 01 + ... DSDT: +/* + * Intel ACPI Component Architecture + * AML/ASL+ Disassembler version 20200925 (64-bit version) + * Copyright (c) 2000 - 2020 Intel Corporation + * + * Disassembling to symbolic ASL+ operators + * + * Disassembly of /tmp/aml-8G8791, Wed Aug 23 10:36:32 2023 + * + * Original Table Header: + * Signature "DSDT" + * Length 0x0000489D (18589) + * Revision 0x01 **** 32-bit table (V1), no 64-bit math support + * Checksum 0xDB + * OEM ID "BOCHS " + * OEM Table ID "BXPC " + * OEM Revision 0x00000001 (1) + * Compiler ID "BXPC" + * Compiler Version 0x00000001 (1) + */ +DefinitionBlock ("", "DSDT", 1, "BOCHS ", "BXPC ", 0x00000001) +{ + Scope (\) + { + OperationRegion (DBG, SystemIO, 0x0402, One) + Field (DBG, ByteAcc, NoLock, Preserve) + { + DBGB, 8 + } + + Method (DBUG, 1, NotSerialized) + { + ToHexString (Arg0, Local0) + ToBuffer (Local0, Local0) + Local1 = (SizeOf (Local0) - One) + Local2 = Zero + While ((Local2 < Local1)) + { + DBGB = DerefOf (Local0 [Local2]) + Local2++ + } + + DBGB = 0x0A + } + } + [snip] + + Processor (C000, 0x00, 0x00000000, 0x00) + { + Method (_STA, 0, Serialized) // _STA: Status + { + Return (CSTA (Zero)) + } + + Name (_MAT, Buffer (0x08) // _MAT: Multiple APIC Table Entry + { + 0x00, 0x08, 0x00, 0x00, 0x01, 0x00, 0x00, 0x00 // ........ + }) + Method (_OST, 3, Serialized) // _OST: OSPM Status Indication + { + COST (Zero, Arg0, Arg1, Arg2) + } + } + + Processor (C001, 0x01, 0x00000000, 0x00) + { + Method (_STA, 0, Serialized) // _STA: Status + { + Return (CSTA (One)) + } + + Name (_MAT, Buffer (0x08) // _MAT: Multiple APIC Table Entry + { + 0x00, 0x08, 0x01, 0x01, 0x01, 0x00, 0x00, 0x00 // ........ + }) + Method (_EJ0, 1, NotSerialized) // _EJx: Eject Device, x=0-9 + { + CEJ0 (One) + } + + Method (_OST, 3, Serialized) // _OST: OSPM Status Indication + { + COST (One, Arg0, Arg1, Arg2) + } + } [snip] + Processor (C077, 0x77, 0x00000000, 0x00) + { + Method (_STA, 0, Serialized) // _STA: Status + { + Return (CSTA (0x77)) + } + + Name (_MAT, Buffer (0x08) // _MAT: Multiple APIC Table Entry + { + 0x00, 0x08, 0x77, 0x9E, 0x01, 0x00, 0x00, 0x00 // ..w..... + }) + Method (_EJ0, 1, NotSerialized) // _EJx: Eject Device, x=0-9 + { + CEJ0 (0x77) + } + + Method (_OST, 3, Serialized) // _OST: OSPM Status Indication + { + COST (0x77, Arg0, Arg1, Arg2) + } + } + } + } + ... Signed-off-by: Zhao Liu Reviewed-by: Michael S. Tsirkin Message-Id: <20231023094635.1588282-5-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/data/acpi/q35/APIC.type4-count | Bin 0 -> 1072 bytes tests/data/acpi/q35/DSDT.type4-count | Bin 0 -> 18589 bytes tests/data/acpi/q35/FACP.type4-count | Bin 0 -> 244 bytes tests/qtest/bios-tables-test-allowed-diff.h | 3 --- 4 files changed, 3 deletions(-) diff --git a/tests/data/acpi/q35/APIC.type4-count b/tests/data/acpi/q35/APIC.type4-count index e69de29bb2..ab60a6ef06 100644 Binary files a/tests/data/acpi/q35/APIC.type4-count and b/tests/data/acpi/q35/APIC.type4-count differ diff --git a/tests/data/acpi/q35/DSDT.type4-count b/tests/data/acpi/q35/DSDT.type4-count index e69de29bb2..edc23198cd 100644 Binary files a/tests/data/acpi/q35/DSDT.type4-count and b/tests/data/acpi/q35/DSDT.type4-count differ diff --git a/tests/data/acpi/q35/FACP.type4-count b/tests/data/acpi/q35/FACP.type4-count index e69de29bb2..31fa5dd19c 100644 Binary files a/tests/data/acpi/q35/FACP.type4-count and b/tests/data/acpi/q35/FACP.type4-count differ diff --git a/tests/qtest/bios-tables-test-allowed-diff.h b/tests/qtest/bios-tables-test-allowed-diff.h index 0ce6f8fc72..dfb8523c8b 100644 --- a/tests/qtest/bios-tables-test-allowed-diff.h +++ b/tests/qtest/bios-tables-test-allowed-diff.h @@ -1,4 +1 @@ /* List of comma-separated changed AML files to ignore */ -"tests/data/acpi/q35/APIC.type4-count", -"tests/data/acpi/q35/DSDT.type4-count", -"tests/data/acpi/q35/FACP.type4-count", -- cgit v1.2.3 From 623d26ad9affaeca03633880623961434e60e77b Mon Sep 17 00:00:00 2001 From: Zhao Liu Date: Mon, 23 Oct 2023 17:46:24 +0800 Subject: tests: bios-tables-test: Prepare the ACPI table change for smbios type4 core count test Following the guidelines in tests/qtest/bios-tables-test.c, this is step 1 - 3. List the ACPI tables that will be added to test the type 4 core count field. Signed-off-by: Zhao Liu Reviewed-by: Michael S. Tsirkin Message-Id: <20231023094635.1588282-6-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/data/acpi/q35/APIC.core-count | 0 tests/data/acpi/q35/DSDT.core-count | 0 tests/data/acpi/q35/FACP.core-count | 0 tests/qtest/bios-tables-test-allowed-diff.h | 3 +++ 4 files changed, 3 insertions(+) create mode 100644 tests/data/acpi/q35/APIC.core-count create mode 100644 tests/data/acpi/q35/DSDT.core-count create mode 100644 tests/data/acpi/q35/FACP.core-count diff --git a/tests/data/acpi/q35/APIC.core-count b/tests/data/acpi/q35/APIC.core-count new file mode 100644 index 0000000000..e69de29bb2 diff --git a/tests/data/acpi/q35/DSDT.core-count b/tests/data/acpi/q35/DSDT.core-count new file mode 100644 index 0000000000..e69de29bb2 diff --git a/tests/data/acpi/q35/FACP.core-count b/tests/data/acpi/q35/FACP.core-count new file mode 100644 index 0000000000..e69de29bb2 diff --git a/tests/qtest/bios-tables-test-allowed-diff.h b/tests/qtest/bios-tables-test-allowed-diff.h index dfb8523c8b..b9bc196130 100644 --- a/tests/qtest/bios-tables-test-allowed-diff.h +++ b/tests/qtest/bios-tables-test-allowed-diff.h @@ -1 +1,4 @@ /* List of comma-separated changed AML files to ignore */ +"tests/data/acpi/q35/APIC.core-count", +"tests/data/acpi/q35/DSDT.core-count", +"tests/data/acpi/q35/FACP.core-count", -- cgit v1.2.3 From 148a8a1d5fdbdb0ba2a9883c6182e3135cb417b2 Mon Sep 17 00:00:00 2001 From: Zhao Liu Date: Mon, 23 Oct 2023 17:46:25 +0800 Subject: tests: bios-tables-test: Add test for smbios type4 core count This tests the commit 196ea60a734c3 ("hw/smbios: Fix core count in type4"). In smbios_build_type_4_table() (hw/smbios/smbios.c), if the number of cores in the socket is not more than 255, then smbios type4 table encodes cores per socket into the core count field. So for the topology in this case, there're the following considerations: 1. cores per socket should be not more than 255 to ensure we could cover the core count field. 2. The original bug was that cores per socket was miscalculated, so now we should include as many topology levels as possible (mutiple sockets & dies, no module since x86 hasn't supported it) to cover more general topology scenarios, to ensure that the cores per socket encoded in the core count field is correct. Based on these considerations, select the topology with multiple sockets and dies: -smp 54,sockets=2,dies=3,cores=3,threads=3 The expected core count = cores per socket = cores (3) * dies (3) = 9. Suggested-by: Igor Mammedov Signed-off-by: Zhao Liu Reviewed-by: Michael S. Tsirkin Message-Id: <20231023094635.1588282-7-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/qtest/bios-tables-test.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/tests/qtest/bios-tables-test.c b/tests/qtest/bios-tables-test.c index cdbfb51559..c20f6f73d0 100644 --- a/tests/qtest/bios-tables-test.c +++ b/tests/qtest/bios-tables-test.c @@ -999,6 +999,23 @@ static void test_acpi_q35_tcg_type4_count(void) free_test_data(&data); } +static void test_acpi_q35_tcg_core_count(void) +{ + test_data data = { + .machine = MACHINE_Q35, + .variant = ".core-count", + .required_struct_types = base_required_struct_types, + .required_struct_types_len = ARRAY_SIZE(base_required_struct_types), + .smbios_core_count = 9, + .smbios_core_count2 = 9, + }; + + test_acpi_one("-machine smbios-entry-point-type=64 " + "-smp 54,sockets=2,dies=3,cores=3,threads=3", + &data); + free_test_data(&data); +} + static void test_acpi_q35_tcg_core_count2(void) { test_data data = { @@ -2178,6 +2195,8 @@ int main(int argc, char *argv[]) qtest_add_func("acpi/q35/kvm/dmar", test_acpi_q35_kvm_dmar); qtest_add_func("acpi/q35/type4-count", test_acpi_q35_tcg_type4_count); + qtest_add_func("acpi/q35/core-count", + test_acpi_q35_tcg_core_count); qtest_add_func("acpi/q35/core-count2", test_acpi_q35_tcg_core_count2); } -- cgit v1.2.3 From 61ace1d77280b26f0d6ed32b27a1f532f85af9f5 Mon Sep 17 00:00:00 2001 From: Zhao Liu Date: Mon, 23 Oct 2023 17:46:26 +0800 Subject: tests: bios-tables-test: Add ACPI table binaries for smbios type4 core count test Following the guidelines in tests/qtest/bios-tables-test.c, this is step 5 and 6. Changes in the tables: FACP: +/* + * Intel ACPI Component Architecture + * AML/ASL+ Disassembler version 20200925 (64-bit version) + * Copyright (c) 2000 - 2020 Intel Corporation + * + * Disassembly of /tmp/aml-Y6WW91, Wed Aug 23 15:43:43 2023 + * + * ACPI Data Table [FACP] + * + * Format: [HexOffset DecimalOffset ByteLength] FieldName : FieldValue + */ + +[000h 0000 4] Signature : "FACP" [Fixed ACPI Description Table (FADT)] +[004h 0004 4] Table Length : 000000F4 +[008h 0008 1] Revision : 03 +[009h 0009 1] Checksum : B3 +[00Ah 0010 6] Oem ID : "BOCHS " +[010h 0016 8] Oem Table ID : "BXPC " +[018h 0024 4] Oem Revision : 00000001 +[01Ch 0028 4] Asl Compiler ID : "BXPC" +[020h 0032 4] Asl Compiler Revision : 00000001 + +[024h 0036 4] FACS Address : 00000000 +[028h 0040 4] DSDT Address : 00000000 +[02Ch 0044 1] Model : 01 +[02Dh 0045 1] PM Profile : 00 [Unspecified] +[02Eh 0046 2] SCI Interrupt : 0009 +[030h 0048 4] SMI Command Port : 000000B2 +[034h 0052 1] ACPI Enable Value : 02 +[035h 0053 1] ACPI Disable Value : 03 +[036h 0054 1] S4BIOS Command : 00 +[037h 0055 1] P-State Control : 00 +[038h 0056 4] PM1A Event Block Address : 00000600 +[03Ch 0060 4] PM1B Event Block Address : 00000000 +[040h 0064 4] PM1A Control Block Address : 00000604 +[044h 0068 4] PM1B Control Block Address : 00000000 +[048h 0072 4] PM2 Control Block Address : 00000000 +[04Ch 0076 4] PM Timer Block Address : 00000608 +[050h 0080 4] GPE0 Block Address : 00000620 +[054h 0084 4] GPE1 Block Address : 00000000 +[058h 0088 1] PM1 Event Block Length : 04 +[059h 0089 1] PM1 Control Block Length : 02 +[05Ah 0090 1] PM2 Control Block Length : 00 +[05Bh 0091 1] PM Timer Block Length : 04 +[05Ch 0092 1] GPE0 Block Length : 10 +[05Dh 0093 1] GPE1 Block Length : 00 +[05Eh 0094 1] GPE1 Base Offset : 00 +[05Fh 0095 1] _CST Support : 00 +[060h 0096 2] C2 Latency : 0FFF +[062h 0098 2] C3 Latency : 0FFF +[064h 0100 2] CPU Cache Size : 0000 +[066h 0102 2] Cache Flush Stride : 0000 +[068h 0104 1] Duty Cycle Offset : 00 +[069h 0105 1] Duty Cycle Width : 00 +[06Ah 0106 1] RTC Day Alarm Index : 00 +[06Bh 0107 1] RTC Month Alarm Index : 00 +[06Ch 0108 1] RTC Century Index : 32 +[06Dh 0109 2] Boot Flags (decoded below) : 0002 + Legacy Devices Supported (V2) : 0 + 8042 Present on ports 60/64 (V2) : 1 + VGA Not Present (V4) : 0 + MSI Not Supported (V4) : 0 + PCIe ASPM Not Supported (V4) : 0 + CMOS RTC Not Present (V5) : 0 +[06Fh 0111 1] Reserved : 00 +[070h 0112 4] Flags (decoded below) : 000484A5 + WBINVD instruction is operational (V1) : 1 + WBINVD flushes all caches (V1) : 0 + All CPUs support C1 (V1) : 1 + C2 works on MP system (V1) : 0 + Control Method Power Button (V1) : 0 + Control Method Sleep Button (V1) : 1 + RTC wake not in fixed reg space (V1) : 0 + RTC can wake system from S4 (V1) : 1 + 32-bit PM Timer (V1) : 0 + Docking Supported (V1) : 0 + Reset Register Supported (V2) : 1 + Sealed Case (V3) : 0 + Headless - No Video (V3) : 0 + Use native instr after SLP_TYPx (V3) : 0 + PCIEXP_WAK Bits Supported (V4) : 0 + Use Platform Timer (V4) : 1 + RTC_STS valid on S4 wake (V4) : 0 + Remote Power-on capable (V4) : 0 + Use APIC Cluster Model (V4) : 1 + Use APIC Physical Destination Mode (V4) : 0 + Hardware Reduced (V5) : 0 + Low Power S0 Idle (V5) : 0 + +[074h 0116 12] Reset Register : [Generic Address Structure] +[074h 0116 1] Space ID : 01 [SystemIO] +[075h 0117 1] Bit Width : 08 +[076h 0118 1] Bit Offset : 00 +[077h 0119 1] Encoded Access Width : 00 [Undefined/Legacy] +[078h 0120 8] Address : 0000000000000CF9 + +[080h 0128 1] Value to cause reset : 0F +[081h 0129 2] ARM Flags (decoded below) : 0000 + PSCI Compliant : 0 + Must use HVC for PSCI : 0 + +[083h 0131 1] FADT Minor Revision : 00 +[084h 0132 8] FACS Address : 0000000000000000 +[08Ch 0140 8] DSDT Address : 0000000000000000 +[094h 0148 12] PM1A Event Block : [Generic Address Structure] +[094h 0148 1] Space ID : 01 [SystemIO] +[095h 0149 1] Bit Width : 20 +[096h 0150 1] Bit Offset : 00 +[097h 0151 1] Encoded Access Width : 00 [Undefined/Legacy] +[098h 0152 8] Address : 0000000000000600 + +[0A0h 0160 12] PM1B Event Block : [Generic Address Structure] +[0A0h 0160 1] Space ID : 00 [SystemMemory] +[0A1h 0161 1] Bit Width : 00 +[0A2h 0162 1] Bit Offset : 00 +[0A3h 0163 1] Encoded Access Width : 00 [Undefined/Legacy] +[0A4h 0164 8] Address : 0000000000000000 + +[0ACh 0172 12] PM1A Control Block : [Generic Address Structure] +[0ACh 0172 1] Space ID : 01 [SystemIO] +[0ADh 0173 1] Bit Width : 10 +[0AEh 0174 1] Bit Offset : 00 +[0AFh 0175 1] Encoded Access Width : 00 [Undefined/Legacy] +[0B0h 0176 8] Address : 0000000000000604 + +[0B8h 0184 12] PM1B Control Block : [Generic Address Structure] +[0B8h 0184 1] Space ID : 00 [SystemMemory] +[0B9h 0185 1] Bit Width : 00 +[0BAh 0186 1] Bit Offset : 00 +[0BBh 0187 1] Encoded Access Width : 00 [Undefined/Legacy] +[0BCh 0188 8] Address : 0000000000000000 + +[0C4h 0196 12] PM2 Control Block : [Generic Address Structure] +[0C4h 0196 1] Space ID : 00 [SystemMemory] +[0C5h 0197 1] Bit Width : 00 +[0C6h 0198 1] Bit Offset : 00 +[0C7h 0199 1] Encoded Access Width : 00 [Undefined/Legacy] +[0C8h 0200 8] Address : 0000000000000000 + +[0D0h 0208 12] PM Timer Block : [Generic Address Structure] +[0D0h 0208 1] Space ID : 01 [SystemIO] +[0D1h 0209 1] Bit Width : 20 +[0D2h 0210 1] Bit Offset : 00 +[0D3h 0211 1] Encoded Access Width : 00 [Undefined/Legacy] +[0D4h 0212 8] Address : 0000000000000608 + +[0DCh 0220 12] GPE0 Block : [Generic Address Structure] +[0DCh 0220 1] Space ID : 01 [SystemIO] +[0DDh 0221 1] Bit Width : 80 +[0DEh 0222 1] Bit Offset : 00 +[0DFh 0223 1] Encoded Access Width : 00 [Undefined/Legacy] +[0E0h 0224 8] Address : 0000000000000620 + +[0E8h 0232 12] GPE1 Block : [Generic Address Structure] +[0E8h 0232 1] Space ID : 00 [SystemMemory] +[0E9h 0233 1] Bit Width : 00 +[0EAh 0234 1] Bit Offset : 00 +[0EBh 0235 1] Encoded Access Width : 00 [Undefined/Legacy] +[0ECh 0236 8] Address : 0000000000000000 ... APIC: +/* + * Intel ACPI Component Architecture + * AML/ASL+ Disassembler version 20200925 (64-bit version) + * Copyright (c) 2000 - 2020 Intel Corporation + * + * Disassembly of /tmp/aml-FFXW91, Wed Aug 23 15:43:43 2023 + * + * ACPI Data Table [APIC] + * + * Format: [HexOffset DecimalOffset ByteLength] FieldName : FieldValue + */ + +[000h 0000 4] Signature : "APIC" [Multiple APIC Description Table (MADT)] +[004h 0004 4] Table Length : 00000220 +[008h 0008 1] Revision : 03 +[009h 0009 1] Checksum : 3C +[00Ah 0010 6] Oem ID : "BOCHS " +[010h 0016 8] Oem Table ID : "BXPC " +[018h 0024 4] Oem Revision : 00000001 +[01Ch 0028 4] Asl Compiler ID : "BXPC" +[020h 0032 4] Asl Compiler Revision : 00000001 + +[024h 0036 4] Local Apic Address : FEE00000 +[028h 0040 4] Flags (decoded below) : 00000001 + PC-AT Compatibility : 1 + +[02Ch 0044 1] Subtable Type : 00 [Processor Local APIC] +[02Dh 0045 1] Length : 08 +[02Eh 0046 1] Processor ID : 00 +[02Fh 0047 1] Local Apic ID : 00 +[030h 0048 4] Flags (decoded below) : 00000001 + Processor Enabled : 1 + Runtime Online Capable : 0 [snip] +[1D4h 0468 1] Subtable Type : 00 [Processor Local APIC] +[1D5h 0469 1] Length : 08 +[1D6h 0470 1] Processor ID : 35 +[1D7h 0471 1] Local Apic ID : 6A +[1D8h 0472 4] Flags (decoded below) : 00000001 + Processor Enabled : 1 + Runtime Online Capable : 0 + +[1DCh 0476 1] Subtable Type : 01 [I/O APIC] +[1DDh 0477 1] Length : 0C +[1DEh 0478 1] I/O Apic ID : 00 +[1DFh 0479 1] Reserved : 00 +[1E0h 0480 4] Address : FEC00000 +[1E4h 0484 4] Interrupt : 00000000 + +[1E8h 0488 1] Subtable Type : 02 [Interrupt Source Override] +[1E9h 0489 1] Length : 0A +[1EAh 0490 1] Bus : 00 +[1EBh 0491 1] Source : 00 +[1ECh 0492 4] Interrupt : 00000002 +[1F0h 0496 2] Flags (decoded below) : 0000 + Polarity : 0 + Trigger Mode : 0 + +[1F2h 0498 1] Subtable Type : 02 [Interrupt Source Override] +[1F3h 0499 1] Length : 0A +[1F4h 0500 1] Bus : 00 +[1F5h 0501 1] Source : 05 +[1F6h 0502 4] Interrupt : 00000005 +[1FAh 0506 2] Flags (decoded below) : 000D + Polarity : 1 + Trigger Mode : 3 + +[1FCh 0508 1] Subtable Type : 02 [Interrupt Source Override] +[1FDh 0509 1] Length : 0A +[1FEh 0510 1] Bus : 00 +[1FFh 0511 1] Source : 09 +[200h 0512 4] Interrupt : 00000009 +[204h 0516 2] Flags (decoded below) : 000D + Polarity : 1 + Trigger Mode : 3 + +[206h 0518 1] Subtable Type : 02 [Interrupt Source Override] +[207h 0519 1] Length : 0A +[208h 0520 1] Bus : 00 +[209h 0521 1] Source : 0A +[20Ah 0522 4] Interrupt : 0000000A +[20Eh 0526 2] Flags (decoded below) : 000D + Polarity : 1 + Trigger Mode : 3 + +[210h 0528 1] Subtable Type : 02 [Interrupt Source Override] +[211h 0529 1] Length : 0A +[212h 0530 1] Bus : 00 +[213h 0531 1] Source : 0B +[214h 0532 4] Interrupt : 0000000B +[218h 0536 2] Flags (decoded below) : 000D + Polarity : 1 + Trigger Mode : 3 + +[21Ah 0538 1] Subtable Type : 04 [Local APIC NMI] +[21Bh 0539 1] Length : 06 +[21Ch 0540 1] Processor ID : FF +[21Dh 0541 2] Flags (decoded below) : 0000 + Polarity : 0 + Trigger Mode : 0 +[21Fh 0543 1] Interrupt Input LINT : 01 ... DSDT: +/* + * Intel ACPI Component Architecture + * AML/ASL+ Disassembler version 20200925 (64-bit version) + * Copyright (c) 2000 - 2020 Intel Corporation + * + * Disassembling to symbolic ASL+ operators + * + * Disassembly of /tmp/aml-9ZXW91, Wed Aug 23 15:43:43 2023 + * + * Original Table Header: + * Signature "DSDT" + * Length 0x00003271 (12913) + * Revision 0x01 **** 32-bit table (V1), no 64-bit math support + * Checksum 0xAF + * OEM ID "BOCHS " + * OEM Table ID "BXPC " + * OEM Revision 0x00000001 (1) + * Compiler ID "BXPC" + * Compiler Version 0x00000001 (1) + */ +DefinitionBlock ("", "DSDT", 1, "BOCHS ", "BXPC ", 0x00000001) +{ + Scope (\) + { + OperationRegion (DBG, SystemIO, 0x0402, One) + Field (DBG, ByteAcc, NoLock, Preserve) + { + DBGB, 8 + } + + Method (DBUG, 1, NotSerialized) + { + ToHexString (Arg0, Local0) + ToBuffer (Local0, Local0) + Local1 = (SizeOf (Local0) - One) + Local2 = Zero + While ((Local2 < Local1)) + { + DBGB = DerefOf (Local0 [Local2]) + Local2++ + } + + DBGB = 0x0A + } + } [snip] + Device (\_SB.CPUS) + { + Name (_HID, "ACPI0010" /* Processor Container Device */) // _HID: Hardware ID + Name (_CID, EisaId ("PNP0A05") /* Generic Container Device */) // _CID: Compatible ID + Method (CTFY, 2, NotSerialized) + { + If ((Arg0 == Zero)) + { + Notify (C000, Arg1) + } + + If ((Arg0 == One)) + { + Notify (C001, Arg1) + } [snip] + If ((Arg0 == 0x35)) + { + Notify (C035, Arg1) + } + } [snip] + Processor (C000, 0x00, 0x00000000, 0x00) + { + Method (_STA, 0, Serialized) // _STA: Status + { + Return (CSTA (Zero)) + } + + Name (_MAT, Buffer (0x08) // _MAT: Multiple APIC Table Entry + { + 0x00, 0x08, 0x00, 0x00, 0x01, 0x00, 0x00, 0x00 // ........ + }) + Method (_OST, 3, Serialized) // _OST: OSPM Status Indication + { + COST (Zero, Arg0, Arg1, Arg2) + } + } + + Processor (C001, 0x01, 0x00000000, 0x00) + { + Method (_STA, 0, Serialized) // _STA: Status + { + Return (CSTA (One)) + } + + Name (_MAT, Buffer (0x08) // _MAT: Multiple APIC Table Entry + { + 0x00, 0x08, 0x01, 0x01, 0x01, 0x00, 0x00, 0x00 // ........ + }) + Method (_EJ0, 1, NotSerialized) // _EJx: Eject Device, x=0-9 + { + CEJ0 (One) + } + + Method (_OST, 3, Serialized) // _OST: OSPM Status Indication + { + COST (One, Arg0, Arg1, Arg2) + } + } [snip] + Processor (C035, 0x35, 0x00000000, 0x00) + { + Method (_STA, 0, Serialized) // _STA: Status + { + Return (CSTA (0x35)) + } + + Name (_MAT, Buffer (0x08) // _MAT: Multiple APIC Table Entry + { + 0x00, 0x08, 0x35, 0x6A, 0x01, 0x00, 0x00, 0x00 // ..5j.... + }) + Method (_EJ0, 1, NotSerialized) // _EJx: Eject Device, x=0-9 + { + CEJ0 (0x35) + } + + Method (_OST, 3, Serialized) // _OST: OSPM Status Indication + { + COST (0x35, Arg0, Arg1, Arg2) + } + } ... Signed-off-by: Zhao Liu Reviewed-by: Michael S. Tsirkin Message-Id: <20231023094635.1588282-8-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/data/acpi/q35/APIC.core-count | Bin 0 -> 544 bytes tests/data/acpi/q35/DSDT.core-count | Bin 0 -> 12913 bytes tests/data/acpi/q35/FACP.core-count | Bin 0 -> 244 bytes tests/qtest/bios-tables-test-allowed-diff.h | 3 --- 4 files changed, 3 deletions(-) diff --git a/tests/data/acpi/q35/APIC.core-count b/tests/data/acpi/q35/APIC.core-count index e69de29bb2..d9d7ca9a89 100644 Binary files a/tests/data/acpi/q35/APIC.core-count and b/tests/data/acpi/q35/APIC.core-count differ diff --git a/tests/data/acpi/q35/DSDT.core-count b/tests/data/acpi/q35/DSDT.core-count index e69de29bb2..a24b04cbdb 100644 Binary files a/tests/data/acpi/q35/DSDT.core-count and b/tests/data/acpi/q35/DSDT.core-count differ diff --git a/tests/data/acpi/q35/FACP.core-count b/tests/data/acpi/q35/FACP.core-count index e69de29bb2..31fa5dd19c 100644 Binary files a/tests/data/acpi/q35/FACP.core-count and b/tests/data/acpi/q35/FACP.core-count differ diff --git a/tests/qtest/bios-tables-test-allowed-diff.h b/tests/qtest/bios-tables-test-allowed-diff.h index b9bc196130..dfb8523c8b 100644 --- a/tests/qtest/bios-tables-test-allowed-diff.h +++ b/tests/qtest/bios-tables-test-allowed-diff.h @@ -1,4 +1 @@ /* List of comma-separated changed AML files to ignore */ -"tests/data/acpi/q35/APIC.core-count", -"tests/data/acpi/q35/DSDT.core-count", -"tests/data/acpi/q35/FACP.core-count", -- cgit v1.2.3 From c63fcb2c10810e8294712f6e09657ef5bdd78833 Mon Sep 17 00:00:00 2001 From: Zhao Liu Date: Mon, 23 Oct 2023 17:46:27 +0800 Subject: tests: bios-tables-test: Prepare the ACPI table change for smbios type4 core count2 test Following the guidelines in tests/qtest/bios-tables-test.c, this is step 1 - 3. List the ACPI tables that will be changed about the type 4 core count2 test case. Signed-off-by: Zhao Liu Reviewed-by: Michael S. Tsirkin Message-Id: <20231023094635.1588282-9-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/qtest/bios-tables-test-allowed-diff.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/qtest/bios-tables-test-allowed-diff.h b/tests/qtest/bios-tables-test-allowed-diff.h index dfb8523c8b..0f95d1344b 100644 --- a/tests/qtest/bios-tables-test-allowed-diff.h +++ b/tests/qtest/bios-tables-test-allowed-diff.h @@ -1 +1,3 @@ /* List of comma-separated changed AML files to ignore */ +"tests/data/acpi/q35/APIC.core-count2", +"tests/data/acpi/q35/DSDT.core-count2", -- cgit v1.2.3 From 6dc82e32226dd3455a1b82bb2ef37d938570084f Mon Sep 17 00:00:00 2001 From: Zhao Liu Date: Mon, 23 Oct 2023 17:46:28 +0800 Subject: tests: bios-tables-test: Extend smbios core count2 test to cover general topology The commit 196ea60a734c3 ("hw/smbios: Fix core count in type4") fixed the miscalculation of cores per socket. The original core count2 test (with the topology configured by "-smp 275") didn't recognize that topology-related but because it just created a special topology with only one socket and one die by default, ignoring the effect of more topology levels (between socket and core) on the cores per socket calculation. So for the topology in this case, there're the following considerations: 1. cores per socket should be more than 255 to ensure we could cover the core count2 field. 2. The original bug was that cores per socket was miscalculated, so now we should include as many topology levels as possible (multiple sockets or dies, no module since x86 hasn't supported it) to cover more general topology scenarios, to ensure that the cores per socket encoded in the core count2 field is correct. Based on these considerations, select the topology with multiple dies: -smp 260,dies=2,cores=130,threads=1 Note, here we doesn't configure multiple sockets to avoid the error ("kvm_init_vcpu: kvm_get_vcpu failed (*): Too many open files") if user uses the default ulimit seeting on his machine. And the cores per socket calculation for multiple sockets has already been covered by the core count test case, so that only multiple dies configuration is enough. The expected core count2 = cores per socket = cores (130) * dies (2) = 260. Suggested-by: Igor Mammedov Signed-off-by: Zhao Liu Acked-by: Igor Mammedov Reviewed-by: Michael S. Tsirkin Message-Id: <20231023094635.1588282-10-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/qtest/bios-tables-test.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/tests/qtest/bios-tables-test.c b/tests/qtest/bios-tables-test.c index c20f6f73d0..f3af20cf2c 100644 --- a/tests/qtest/bios-tables-test.c +++ b/tests/qtest/bios-tables-test.c @@ -1024,10 +1024,12 @@ static void test_acpi_q35_tcg_core_count2(void) .required_struct_types = base_required_struct_types, .required_struct_types_len = ARRAY_SIZE(base_required_struct_types), .smbios_core_count = 0xFF, - .smbios_core_count2 = 275, + .smbios_core_count2 = 260, }; - test_acpi_one("-machine smbios-entry-point-type=64 -smp 275", &data); + test_acpi_one("-machine smbios-entry-point-type=64 " + "-smp 260,dies=2,cores=130,threads=1", + &data); free_test_data(&data); } -- cgit v1.2.3 From f03359a85b4c29858c95d043961288c5e6dc662b Mon Sep 17 00:00:00 2001 From: Zhao Liu Date: Mon, 23 Oct 2023 17:46:29 +0800 Subject: tests: bios-tables-test: Update ACPI table binaries for smbios core count2 test Change the core count2 from 275 to 260. Following the guidelines in tests/qtest/bios-tables-test.c, this is step 5 and 6. Changes in the tables: APIC: /* * Intel ACPI Component Architecture * AML/ASL+ Disassembler version 20200925 (64-bit version) * Copyright (c) 2000 - 2020 Intel Corporation * - * Disassembly of tests/data/acpi/q35/APIC.core-count2, Wed Aug 23 16:29:51 2023 + * Disassembly of /tmp/aml-KQDX91, Wed Aug 23 16:29:51 2023 * * ACPI Data Table [APIC] * * Format: [HexOffset DecimalOffset ByteLength] FieldName : FieldValue */ [000h 0000 4] Signature : "APIC" [Multiple APIC Description Table (MADT)] -[004h 0004 4] Table Length : 000009AE +[004h 0004 4] Table Length : 00000CA6 [008h 0008 1] Revision : 03 -[009h 0009 1] Checksum : CE +[009h 0009 1] Checksum : FA [00Ah 0010 6] Oem ID : "BOCHS " [010h 0016 8] Oem Table ID : "BXPC " [018h 0024 4] Oem Revision : 00000001 [01Ch 0028 4] Asl Compiler ID : "BXPC" [020h 0032 4] Asl Compiler Revision : 00000001 [024h 0036 4] Local Apic Address : FEE00000 [028h 0040 4] Flags (decoded below) : 00000001 PC-AT Compatibility : 1 [02Ch 0044 1] Subtable Type : 00 [Processor Local APIC] [02Dh 0045 1] Length : 08 [02Eh 0046 1] Processor ID : 00 [02Fh 0047 1] Local Apic ID : 00 [030h 0048 4] Flags (decoded below) : 00000001 Processor Enabled : 1 @@ -1051,1256 +1051,1136 @@ [42Ch 1068 1] Subtable Type : 00 [Processor Local APIC] [42Dh 1069 1] Length : 08 [42Eh 1070 1] Processor ID : 80 [42Fh 1071 1] Local Apic ID : 80 [430h 1072 4] Flags (decoded below) : 00000001 Processor Enabled : 1 Runtime Online Capable : 0 [434h 1076 1] Subtable Type : 00 [Processor Local APIC] [435h 1077 1] Length : 08 [436h 1078 1] Processor ID : 81 [437h 1079 1] Local Apic ID : 81 [438h 1080 4] Flags (decoded below) : 00000001 Processor Enabled : 1 Runtime Online Capable : 0 -[43Ch 1084 1] Subtable Type : 00 [Processor Local APIC] -[43Dh 1085 1] Length : 08 -[43Eh 1086 1] Processor ID : 82 -[43Fh 1087 1] Local Apic ID : 82 -[440h 1088 4] Flags (decoded below) : 00000001 - Processor Enabled : 1 - Runtime Online Capable : 0 - -[444h 1092 1] Subtable Type : 00 [Processor Local APIC] -[445h 1093 1] Length : 08 -[446h 1094 1] Processor ID : 83 -[447h 1095 1] Local Apic ID : 83 -[448h 1096 4] Flags (decoded below) : 00000001 - Processor Enabled : 1 - Runtime Online Capable : 0 [snip] - -[964h 2404 1] Subtable Type : 01 [I/O APIC] -[965h 2405 1] Length : 0C -[966h 2406 1] I/O Apic ID : 00 -[967h 2407 1] Reserved : 00 -[968h 2408 4] Address : FEC00000 -[96Ch 2412 4] Interrupt : 00000000 - -[970h 2416 1] Subtable Type : 02 [Interrupt Source Override] -[971h 2417 1] Length : 0A -[972h 2418 1] Bus : 00 -[973h 2419 1] Source : 00 -[974h 2420 4] Interrupt : 00000002 -[978h 2424 2] Flags (decoded below) : 0000 +[43Ch 1084 1] Subtable Type : 09 [Processor Local x2APIC] +[43Dh 1085 1] Length : 10 +[43Eh 1086 2] Reserved : 0000 +[440h 1088 4] Processor x2Apic ID : 00000100 +[444h 1092 4] Flags (decoded below) : 00000001 + Processor Enabled : 1 +[448h 1096 4] Processor UID : 00000082 + +[44Ch 1100 1] Subtable Type : 09 [Processor Local x2APIC] +[44Dh 1101 1] Length : 10 +[44Eh 1102 2] Reserved : 0000 +[450h 1104 4] Processor x2Apic ID : 00000101 +[454h 1108 4] Flags (decoded below) : 00000001 + Processor Enabled : 1 +[458h 1112 4] Processor UID : 00000083 + [snip] + +[C68h 3176 1] Subtable Type : 02 [Interrupt Source Override] +[C69h 3177 1] Length : 0A +[C6Ah 3178 1] Bus : 00 +[C6Bh 3179 1] Source : 00 +[C6Ch 3180 4] Interrupt : 00000002 +[C70h 3184 2] Flags (decoded below) : 0000 Polarity : 0 Trigger Mode : 0 -[97Ah 2426 1] Subtable Type : 02 [Interrupt Source Override] -[97Bh 2427 1] Length : 0A -[97Ch 2428 1] Bus : 00 -[97Dh 2429 1] Source : 05 -[97Eh 2430 4] Interrupt : 00000005 -[982h 2434 2] Flags (decoded below) : 000D +[C72h 3186 1] Subtable Type : 02 [Interrupt Source Override] +[C73h 3187 1] Length : 0A +[C74h 3188 1] Bus : 00 +[C75h 3189 1] Source : 05 +[C76h 3190 4] Interrupt : 00000005 +[C7Ah 3194 2] Flags (decoded below) : 000D Polarity : 1 Trigger Mode : 3 -[984h 2436 1] Subtable Type : 02 [Interrupt Source Override] -[985h 2437 1] Length : 0A -[986h 2438 1] Bus : 00 -[987h 2439 1] Source : 09 -[988h 2440 4] Interrupt : 00000009 -[98Ch 2444 2] Flags (decoded below) : 000D +[C7Ch 3196 1] Subtable Type : 02 [Interrupt Source Override] +[C7Dh 3197 1] Length : 0A +[C7Eh 3198 1] Bus : 00 +[C7Fh 3199 1] Source : 09 +[C80h 3200 4] Interrupt : 00000009 +[C84h 3204 2] Flags (decoded below) : 000D Polarity : 1 Trigger Mode : 3 -[98Eh 2446 1] Subtable Type : 02 [Interrupt Source Override] -[98Fh 2447 1] Length : 0A -[990h 2448 1] Bus : 00 -[991h 2449 1] Source : 0A -[992h 2450 4] Interrupt : 0000000A -[996h 2454 2] Flags (decoded below) : 000D +[C86h 3206 1] Subtable Type : 02 [Interrupt Source Override] +[C87h 3207 1] Length : 0A +[C88h 3208 1] Bus : 00 +[C89h 3209 1] Source : 0A +[C8Ah 3210 4] Interrupt : 0000000A +[C8Eh 3214 2] Flags (decoded below) : 000D Polarity : 1 Trigger Mode : 3 -[998h 2456 1] Subtable Type : 02 [Interrupt Source Override] -[999h 2457 1] Length : 0A -[99Ah 2458 1] Bus : 00 -[99Bh 2459 1] Source : 0B -[99Ch 2460 4] Interrupt : 0000000B -[9A0h 2464 2] Flags (decoded below) : 000D +[C90h 3216 1] Subtable Type : 02 [Interrupt Source Override] +[C91h 3217 1] Length : 0A +[C92h 3218 1] Bus : 00 +[C93h 3219 1] Source : 0B +[C94h 3220 4] Interrupt : 0000000B +[C98h 3224 2] Flags (decoded below) : 000D Polarity : 1 Trigger Mode : 3 -[9A2h 2466 1] Subtable Type : 0A [Local x2APIC NMI] -[9A3h 2467 1] Length : 0C -[9A4h 2468 2] Flags (decoded below) : 0000 +[C9Ah 3226 1] Subtable Type : 0A [Local x2APIC NMI] +[C9Bh 3227 1] Length : 0C +[C9Ch 3228 2] Flags (decoded below) : 0000 Polarity : 0 Trigger Mode : 0 -[9A6h 2470 4] Processor UID : FFFFFFFF -[9AAh 2474 1] Interrupt Input LINT : 01 -[9ABh 2475 3] Reserved : 000000 +[C9Eh 3230 4] Processor UID : FFFFFFFF +[CA2h 3234 1] Interrupt Input LINT : 01 +[CA3h 3235 3] Reserved : 000000 ... DSDT: /* * Intel ACPI Component Architecture * AML/ASL+ Disassembler version 20200925 (64-bit version) * Copyright (c) 2000 - 2020 Intel Corporation * * Disassembling to symbolic ASL+ operators * - * Disassembly of tests/data/acpi/q35/DSDT.core-count2, Wed Aug 23 16:29:51 2023 + * Disassembly of /tmp/aml-6DDX91, Wed Aug 23 16:29:51 2023 * * Original Table Header: * Signature "DSDT" - * Length 0x00007EEF (32495) + * Length 0x000083EA (33770) * Revision 0x01 **** 32-bit table (V1), no 64-bit math support - * Checksum 0x52 + * Checksum 0x01 * OEM ID "BOCHS " * OEM Table ID "BXPC " * OEM Revision 0x00000001 (1) * Compiler ID "BXPC" * Compiler Version 0x00000001 (1) */ DefinitionBlock ("", "DSDT", 1, "BOCHS ", "BXPC ", 0x00000001) { Scope (\) { OperationRegion (DBG, SystemIO, 0x0402, One) Field (DBG, ByteAcc, NoLock, Preserve) { DBGB, 8 } @@ -4196,107 +4196,32 @@ } If ((Arg0 == 0x0101)) { Notify (C101, Arg1) } If ((Arg0 == 0x0102)) { Notify (C102, Arg1) } If ((Arg0 == 0x0103)) { Notify (C103, Arg1) } - - If ((Arg0 == 0x0104)) - { - Notify (C104, Arg1) - } - - If ((Arg0 == 0x0105)) - { - Notify (C105, Arg1) - } - - If ((Arg0 == 0x0106)) - { - Notify (C106, Arg1) - } - [snip] - If ((Arg0 == 0x0112)) - { - Notify (C112, Arg1) - } } Method (CSTA, 1, Serialized) { Acquire (\_SB.PCI0.PRES.CPLK, 0xFFFF) \_SB.PCI0.PRES.CSEL = Arg0 Local0 = Zero If ((\_SB.PCI0.PRES.CPEN == One)) { Local0 = 0x0F } Release (\_SB.PCI0.PRES.CPLK) Return (Local0) } @@ -4306,33 +4231,33 @@ \_SB.PCI0.PRES.CSEL = Arg0 \_SB.PCI0.PRES.CEJ0 = One Release (\_SB.PCI0.PRES.CPLK) } Method (CSCN, 0, Serialized) { Acquire (\_SB.PCI0.PRES.CPLK, 0xFFFF) Name (CNEW, Package (0xFF) {}) Local3 = Zero Local4 = One While ((Local4 == One)) { Local4 = Zero Local0 = One Local1 = Zero - While (((Local0 == One) && (Local3 < 0x0113))) + While (((Local0 == One) && (Local3 < 0x0104))) { Local0 = Zero \_SB.PCI0.PRES.CSEL = Local3 \_SB.PCI0.PRES.CCMD = Zero If ((\_SB.PCI0.PRES.CDAT < Local3)) { Break } If ((Local1 == 0xFF)) { Local4 = One Break } Local3 = \_SB.PCI0.PRES.CDAT @@ -7220,3281 +7145,3281 @@ Name (_MAT, Buffer (0x08) // _MAT: Multiple APIC Table Entry { 0x00, 0x08, 0x81, 0x81, 0x01, 0x00, 0x00, 0x00 // ........ }) Method (_EJ0, 1, NotSerialized) // _EJx: Eject Device, x=0-9 { CEJ0 (0x81) } Method (_OST, 3, Serialized) // _OST: OSPM Status Indication { COST (0x81, Arg0, Arg1, Arg2) } } - Processor (C082, 0x82, 0x00000000, 0x00) + Device (C082) { + Name (_HID, "ACPI0007" /* Processor Device */) // _HID: Hardware ID + Name (_UID, 0x82) // _UID: Unique ID Method (_STA, 0, Serialized) // _STA: Status { Return (CSTA (0x82)) } - Name (_MAT, Buffer (0x08) // _MAT: Multiple APIC Table Entry + Name (_MAT, Buffer (0x10) // _MAT: Multiple APIC Table Entry { - 0x00, 0x08, 0x82, 0x82, 0x01, 0x00, 0x00, 0x00 // ........ + /* 0000 */ 0x09, 0x10, 0x00, 0x00, 0x00, 0x01, 0x00, 0x00, // ........ + /* 0008 */ 0x01, 0x00, 0x00, 0x00, 0x82, 0x00, 0x00, 0x00 // ........ }) Method (_EJ0, 1, NotSerialized) // _EJx: Eject Device, x=0-9 { CEJ0 (0x82) } Method (_OST, 3, Serialized) // _OST: OSPM Status Indication { COST (0x82, Arg0, Arg1, Arg2) } } - Processor (C083, 0x83, 0x00000000, 0x00) + Device (C083) { + Name (_HID, "ACPI0007" /* Processor Device */) // _HID: Hardware ID + Name (_UID, 0x83) // _UID: Unique ID Method (_STA, 0, Serialized) // _STA: Status { Return (CSTA (0x83)) } - Name (_MAT, Buffer (0x08) // _MAT: Multiple APIC Table Entry + Name (_MAT, Buffer (0x10) // _MAT: Multiple APIC Table Entry { - 0x00, 0x08, 0x83, 0x83, 0x01, 0x00, 0x00, 0x00 // ........ + /* 0000 */ 0x09, 0x10, 0x00, 0x00, 0x01, 0x01, 0x00, 0x00, // ........ + /* 0008 */ 0x01, 0x00, 0x00, 0x00, 0x83, 0x00, 0x00, 0x00 // ........ }) Method (_EJ0, 1, NotSerialized) // _EJx: Eject Device, x=0-9 { CEJ0 (0x83) } Method (_OST, 3, Serialized) // _OST: OSPM Status Indication { COST (0x83, Arg0, Arg1, Arg2) } } - Processor (C084, 0x84, 0x00000000, 0x00) + Device (C084) { + Name (_HID, "ACPI0007" /* Processor Device */) // _HID: Hardware ID + Name (_UID, 0x84) // _UID: Unique ID Method (_STA, 0, Serialized) // _STA: Status { Return (CSTA (0x84)) } - Name (_MAT, Buffer (0x08) // _MAT: Multiple APIC Table Entry + Name (_MAT, Buffer (0x10) // _MAT: Multiple APIC Table Entry { - 0x00, 0x08, 0x84, 0x84, 0x01, 0x00, 0x00, 0x00 // ........ + /* 0000 */ 0x09, 0x10, 0x00, 0x00, 0x02, 0x01, 0x00, 0x00, // ........ + /* 0008 */ 0x01, 0x00, 0x00, 0x00, 0x84, 0x00, 0x00, 0x00 // ........ }) Method (_EJ0, 1, NotSerialized) // _EJx: Eject Device, x=0-9 { CEJ0 (0x84) } Method (_OST, 3, Serialized) // _OST: OSPM Status Indication { COST (0x84, Arg0, Arg1, Arg2) } } [snip] - Processor (C0FE, 0xFE, 0x00000000, 0x00) + Device (C0FE) { + Name (_HID, "ACPI0007" /* Processor Device */) // _HID: Hardware ID + Name (_UID, 0xFE) // _UID: Unique ID Method (_STA, 0, Serialized) // _STA: Status { Return (CSTA (0xFE)) } - Name (_MAT, Buffer (0x08) // _MAT: Multiple APIC Table Entry + Name (_MAT, Buffer (0x10) // _MAT: Multiple APIC Table Entry { - 0x00, 0x08, 0xFE, 0xFE, 0x01, 0x00, 0x00, 0x00 // ........ + /* 0000 */ 0x09, 0x10, 0x00, 0x00, 0x7C, 0x01, 0x00, 0x00, // ....|... + /* 0008 */ 0x01, 0x00, 0x00, 0x00, 0xFE, 0x00, 0x00, 0x00 // ........ }) Method (_EJ0, 1, NotSerialized) // _EJx: Eject Device, x=0-9 { CEJ0 (0xFE) } Method (_OST, 3, Serialized) // _OST: OSPM Status Indication { COST (0xFE, Arg0, Arg1, Arg2) } } ... Signed-off-by: Zhao Liu Reviewed-by: Michael S. Tsirkin Message-Id: <20231023094635.1588282-11-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/data/acpi/q35/APIC.core-count2 | Bin 2478 -> 3238 bytes tests/data/acpi/q35/DSDT.core-count2 | Bin 32495 -> 33770 bytes tests/qtest/bios-tables-test-allowed-diff.h | 2 -- 3 files changed, 2 deletions(-) diff --git a/tests/data/acpi/q35/APIC.core-count2 b/tests/data/acpi/q35/APIC.core-count2 index f5da2eb1e8..4f24284434 100644 Binary files a/tests/data/acpi/q35/APIC.core-count2 and b/tests/data/acpi/q35/APIC.core-count2 differ diff --git a/tests/data/acpi/q35/DSDT.core-count2 b/tests/data/acpi/q35/DSDT.core-count2 index b47891ec10..3a0cb8c581 100644 Binary files a/tests/data/acpi/q35/DSDT.core-count2 and b/tests/data/acpi/q35/DSDT.core-count2 differ diff --git a/tests/qtest/bios-tables-test-allowed-diff.h b/tests/qtest/bios-tables-test-allowed-diff.h index 0f95d1344b..dfb8523c8b 100644 --- a/tests/qtest/bios-tables-test-allowed-diff.h +++ b/tests/qtest/bios-tables-test-allowed-diff.h @@ -1,3 +1 @@ /* List of comma-separated changed AML files to ignore */ -"tests/data/acpi/q35/APIC.core-count2", -"tests/data/acpi/q35/DSDT.core-count2", -- cgit v1.2.3 From 85ccbe1275b58efe9e30d229c942d33144b5ef6f Mon Sep 17 00:00:00 2001 From: Zhao Liu Date: Mon, 23 Oct 2023 17:46:30 +0800 Subject: tests: bios-tables-test: Prepare the ACPI table change for smbios type4 thread count test Following the guidelines in tests/qtest/bios-tables-test.c, this is step 1 - 3. List the ACPI tables that will be added to test the thread count field of smbios type4 table. Signed-off-by: Zhao Liu Reviewed-by: Michael S. Tsirkin Message-Id: <20231023094635.1588282-12-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/data/acpi/q35/APIC.thread-count | 0 tests/data/acpi/q35/DSDT.thread-count | 0 tests/data/acpi/q35/FACP.thread-count | 0 tests/qtest/bios-tables-test-allowed-diff.h | 3 +++ 4 files changed, 3 insertions(+) create mode 100644 tests/data/acpi/q35/APIC.thread-count create mode 100644 tests/data/acpi/q35/DSDT.thread-count create mode 100644 tests/data/acpi/q35/FACP.thread-count diff --git a/tests/data/acpi/q35/APIC.thread-count b/tests/data/acpi/q35/APIC.thread-count new file mode 100644 index 0000000000..e69de29bb2 diff --git a/tests/data/acpi/q35/DSDT.thread-count b/tests/data/acpi/q35/DSDT.thread-count new file mode 100644 index 0000000000..e69de29bb2 diff --git a/tests/data/acpi/q35/FACP.thread-count b/tests/data/acpi/q35/FACP.thread-count new file mode 100644 index 0000000000..e69de29bb2 diff --git a/tests/qtest/bios-tables-test-allowed-diff.h b/tests/qtest/bios-tables-test-allowed-diff.h index dfb8523c8b..4d139d7f6b 100644 --- a/tests/qtest/bios-tables-test-allowed-diff.h +++ b/tests/qtest/bios-tables-test-allowed-diff.h @@ -1 +1,4 @@ /* List of comma-separated changed AML files to ignore */ +"tests/data/acpi/q35/APIC.thread-count", +"tests/data/acpi/q35/DSDT.thread-count", +"tests/data/acpi/q35/FACP.thread-count", -- cgit v1.2.3 From 7ee18dcef14540c79892381d7af0a2c54ec11484 Mon Sep 17 00:00:00 2001 From: Zhao Liu Date: Mon, 23 Oct 2023 17:46:31 +0800 Subject: tests: bios-tables-test: Add test for smbios type4 thread count This tests the commit 7298fd7de5551 ("hw/smbios: Fix thread count in type4"). In smbios_build_type_4_table() (hw/smbios/smbios.c), if the number of threads in the socket is not more than 255, then smbios type4 table encodes threads per socket into the thread count field. So for the topology in this case, there're the following considerations: 1. threads per socket should be not more than 255 to ensure we could cover the thread count field. 2. The original bug was that threads per socket was miscalculated, so now we should configure as many topology levels as possible (multiple sockets & dies, no module since x86 hasn't supported it) to cover more general topology scenarios, to ensure that the threads per socket encoded in the thread count field is correct. 3. For the more general topology, we should also add "cpus" (presented threads for machine) and "maxcpus" (total threads for machine) to make sure that configuring unpluged CPUs in smp (cpus < maxcpus) does not affect the correctness of threads per socket for thread count field. Based on these considerations, select the topology as the follow: -smp cpus=15,maxcpus=54,sockets=2,dies=3,cores=3,threads=3 The expected thread count = threads per socket = threads (3) * cores (3) * dies (3) = 27. Suggested-by: Igor Mammedov Signed-off-by: Zhao Liu Reviewed-by: Michael S. Tsirkin Message-Id: <20231023094635.1588282-13-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/qtest/bios-tables-test.c | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) diff --git a/tests/qtest/bios-tables-test.c b/tests/qtest/bios-tables-test.c index f3af20cf2c..395ed7f9ff 100644 --- a/tests/qtest/bios-tables-test.c +++ b/tests/qtest/bios-tables-test.c @@ -95,6 +95,7 @@ typedef struct { uint16_t smbios_cpu_curr_speed; uint8_t smbios_core_count; uint16_t smbios_core_count2; + uint8_t smbios_thread_count; uint8_t *required_struct_types; int required_struct_types_len; int type4_count; @@ -640,6 +641,7 @@ static void smbios_cpu_test(test_data *data, uint32_t addr, SmbiosEntryPointType ep_type) { uint8_t core_count, expected_core_count = data->smbios_core_count; + uint8_t thread_count, expected_thread_count = data->smbios_thread_count; uint16_t speed, expected_speed[2]; uint16_t core_count2, expected_core_count2 = data->smbios_core_count2; int offset[2]; @@ -663,6 +665,13 @@ static void smbios_cpu_test(test_data *data, uint32_t addr, g_assert_cmpuint(core_count, ==, expected_core_count); } + thread_count = qtest_readb(data->qts, + addr + offsetof(struct smbios_type_4, thread_count)); + + if (expected_thread_count) { + g_assert_cmpuint(thread_count, ==, expected_thread_count); + } + if (ep_type == SMBIOS_ENTRY_POINT_TYPE_64) { core_count2 = qtest_readw(data->qts, addr + offsetof(struct smbios_type_4, core_count2)); @@ -1033,6 +1042,22 @@ static void test_acpi_q35_tcg_core_count2(void) free_test_data(&data); } +static void test_acpi_q35_tcg_thread_count(void) +{ + test_data data = { + .machine = MACHINE_Q35, + .variant = ".thread-count", + .required_struct_types = base_required_struct_types, + .required_struct_types_len = ARRAY_SIZE(base_required_struct_types), + .smbios_thread_count = 27, + }; + + test_acpi_one("-machine smbios-entry-point-type=64 " + "-smp cpus=15,maxcpus=54,sockets=2,dies=3,cores=3,threads=3", + &data); + free_test_data(&data); +} + static void test_acpi_q35_tcg_bridge(void) { test_data data = {}; @@ -2201,6 +2226,8 @@ int main(int argc, char *argv[]) test_acpi_q35_tcg_core_count); qtest_add_func("acpi/q35/core-count2", test_acpi_q35_tcg_core_count2); + qtest_add_func("acpi/q35/thread-count", + test_acpi_q35_tcg_thread_count); } if (qtest_has_device("virtio-iommu-pci")) { qtest_add_func("acpi/q35/viot", test_acpi_q35_viot); -- cgit v1.2.3 From a775cb191e85c86a1b43f0d082cf9fa7f6d80464 Mon Sep 17 00:00:00 2001 From: Zhao Liu Date: Mon, 23 Oct 2023 17:46:32 +0800 Subject: tests: bios-tables-test: Add ACPI table binaries for smbios type4 thread count test Following the guidelines in tests/qtest/bios-tables-test.c, this is step 5 and 6. Changes in the tables: FACP: +/* + * Intel ACPI Component Architecture + * AML/ASL+ Disassembler version 20200925 (64-bit version) + * Copyright (c) 2000 - 2020 Intel Corporation + * + * Disassembly of /tmp/aml-1NP791, Wed Aug 23 21:51:31 2023 + * + * ACPI Data Table [FACP] + * + * Format: [HexOffset DecimalOffset ByteLength] FieldName : FieldValue + */ + +[000h 0000 4] Signature : "FACP" [Fixed ACPI Description Table (FADT)] +[004h 0004 4] Table Length : 000000F4 +[008h 0008 1] Revision : 03 +[009h 0009 1] Checksum : B3 +[00Ah 0010 6] Oem ID : "BOCHS " +[010h 0016 8] Oem Table ID : "BXPC " +[018h 0024 4] Oem Revision : 00000001 +[01Ch 0028 4] Asl Compiler ID : "BXPC" +[020h 0032 4] Asl Compiler Revision : 00000001 + +[024h 0036 4] FACS Address : 00000000 +[028h 0040 4] DSDT Address : 00000000 +[02Ch 0044 1] Model : 01 +[02Dh 0045 1] PM Profile : 00 [Unspecified] +[02Eh 0046 2] SCI Interrupt : 0009 +[030h 0048 4] SMI Command Port : 000000B2 +[034h 0052 1] ACPI Enable Value : 02 +[035h 0053 1] ACPI Disable Value : 03 +[036h 0054 1] S4BIOS Command : 00 +[037h 0055 1] P-State Control : 00 +[038h 0056 4] PM1A Event Block Address : 00000600 +[03Ch 0060 4] PM1B Event Block Address : 00000000 +[040h 0064 4] PM1A Control Block Address : 00000604 +[044h 0068 4] PM1B Control Block Address : 00000000 +[048h 0072 4] PM2 Control Block Address : 00000000 +[04Ch 0076 4] PM Timer Block Address : 00000608 +[050h 0080 4] GPE0 Block Address : 00000620 +[054h 0084 4] GPE1 Block Address : 00000000 +[058h 0088 1] PM1 Event Block Length : 04 +[059h 0089 1] PM1 Control Block Length : 02 +[05Ah 0090 1] PM2 Control Block Length : 00 +[05Bh 0091 1] PM Timer Block Length : 04 +[05Ch 0092 1] GPE0 Block Length : 10 +[05Dh 0093 1] GPE1 Block Length : 00 +[05Eh 0094 1] GPE1 Base Offset : 00 +[05Fh 0095 1] _CST Support : 00 +[060h 0096 2] C2 Latency : 0FFF +[062h 0098 2] C3 Latency : 0FFF +[064h 0100 2] CPU Cache Size : 0000 +[066h 0102 2] Cache Flush Stride : 0000 +[068h 0104 1] Duty Cycle Offset : 00 +[069h 0105 1] Duty Cycle Width : 00 +[06Ah 0106 1] RTC Day Alarm Index : 00 +[06Bh 0107 1] RTC Month Alarm Index : 00 +[06Ch 0108 1] RTC Century Index : 32 +[06Dh 0109 2] Boot Flags (decoded below) : 0002 + Legacy Devices Supported (V2) : 0 + 8042 Present on ports 60/64 (V2) : 1 + VGA Not Present (V4) : 0 + MSI Not Supported (V4) : 0 + PCIe ASPM Not Supported (V4) : 0 + CMOS RTC Not Present (V5) : 0 +[06Fh 0111 1] Reserved : 00 +[070h 0112 4] Flags (decoded below) : 000484A5 + WBINVD instruction is operational (V1) : 1 + WBINVD flushes all caches (V1) : 0 + All CPUs support C1 (V1) : 1 + C2 works on MP system (V1) : 0 + Control Method Power Button (V1) : 0 + Control Method Sleep Button (V1) : 1 + RTC wake not in fixed reg space (V1) : 0 + RTC can wake system from S4 (V1) : 1 + 32-bit PM Timer (V1) : 0 + Docking Supported (V1) : 0 + Reset Register Supported (V2) : 1 + Sealed Case (V3) : 0 + Headless - No Video (V3) : 0 + Use native instr after SLP_TYPx (V3) : 0 + PCIEXP_WAK Bits Supported (V4) : 0 + Use Platform Timer (V4) : 1 + RTC_STS valid on S4 wake (V4) : 0 + Remote Power-on capable (V4) : 0 + Use APIC Cluster Model (V4) : 1 + Use APIC Physical Destination Mode (V4) : 0 + Hardware Reduced (V5) : 0 + Low Power S0 Idle (V5) : 0 + +[074h 0116 12] Reset Register : [Generic Address Structure] +[074h 0116 1] Space ID : 01 [SystemIO] +[075h 0117 1] Bit Width : 08 +[076h 0118 1] Bit Offset : 00 +[077h 0119 1] Encoded Access Width : 00 [Undefined/Legacy] +[078h 0120 8] Address : 0000000000000CF9 + +[080h 0128 1] Value to cause reset : 0F +[081h 0129 2] ARM Flags (decoded below) : 0000 + PSCI Compliant : 0 + Must use HVC for PSCI : 0 + +[083h 0131 1] FADT Minor Revision : 00 +[084h 0132 8] FACS Address : 0000000000000000 +[08Ch 0140 8] DSDT Address : 0000000000000000 +[094h 0148 12] PM1A Event Block : [Generic Address Structure] +[094h 0148 1] Space ID : 01 [SystemIO] +[095h 0149 1] Bit Width : 20 +[096h 0150 1] Bit Offset : 00 +[097h 0151 1] Encoded Access Width : 00 [Undefined/Legacy] +[098h 0152 8] Address : 0000000000000600 + +[0A0h 0160 12] PM1B Event Block : [Generic Address Structure] +[0A0h 0160 1] Space ID : 00 [SystemMemory] +[0A1h 0161 1] Bit Width : 00 +[0A2h 0162 1] Bit Offset : 00 +[0A3h 0163 1] Encoded Access Width : 00 [Undefined/Legacy] +[0A4h 0164 8] Address : 0000000000000000 + +[0ACh 0172 12] PM1A Control Block : [Generic Address Structure] +[0ACh 0172 1] Space ID : 01 [SystemIO] +[0ADh 0173 1] Bit Width : 10 +[0AEh 0174 1] Bit Offset : 00 +[0AFh 0175 1] Encoded Access Width : 00 [Undefined/Legacy] +[0B0h 0176 8] Address : 0000000000000604 + +[0B8h 0184 12] PM1B Control Block : [Generic Address Structure] +[0B8h 0184 1] Space ID : 00 [SystemMemory] +[0B9h 0185 1] Bit Width : 00 +[0BAh 0186 1] Bit Offset : 00 +[0BBh 0187 1] Encoded Access Width : 00 [Undefined/Legacy] +[0BCh 0188 8] Address : 0000000000000000 + +[0C4h 0196 12] PM2 Control Block : [Generic Address Structure] +[0C4h 0196 1] Space ID : 00 [SystemMemory] +[0C5h 0197 1] Bit Width : 00 +[0C6h 0198 1] Bit Offset : 00 +[0C7h 0199 1] Encoded Access Width : 00 [Undefined/Legacy] +[0C8h 0200 8] Address : 0000000000000000 + +[0D0h 0208 12] PM Timer Block : [Generic Address Structure] +[0D0h 0208 1] Space ID : 01 [SystemIO] +[0D1h 0209 1] Bit Width : 20 +[0D2h 0210 1] Bit Offset : 00 +[0D3h 0211 1] Encoded Access Width : 00 [Undefined/Legacy] +[0D4h 0212 8] Address : 0000000000000608 + +[0DCh 0220 12] GPE0 Block : [Generic Address Structure] +[0DCh 0220 1] Space ID : 01 [SystemIO] +[0DDh 0221 1] Bit Width : 80 +[0DEh 0222 1] Bit Offset : 00 +[0DFh 0223 1] Encoded Access Width : 00 [Undefined/Legacy] +[0E0h 0224 8] Address : 0000000000000620 + +[0E8h 0232 12] GPE1 Block : [Generic Address Structure] +[0E8h 0232 1] Space ID : 00 [SystemMemory] +[0E9h 0233 1] Bit Width : 00 +[0EAh 0234 1] Bit Offset : 00 +[0EBh 0235 1] Encoded Access Width : 00 [Undefined/Legacy] +[0ECh 0236 8] Address : 0000000000000000 ... APIC: +/* + * Intel ACPI Component Architecture + * AML/ASL+ Disassembler version 20200925 (64-bit version) + * Copyright (c) 2000 - 2020 Intel Corporation + * + * Disassembly of /tmp/aml-2JP791, Wed Aug 23 21:51:31 2023 + * + * ACPI Data Table [APIC] + * + * Format: [HexOffset DecimalOffset ByteLength] FieldName : FieldValue + */ + +[000h 0000 4] Signature : "APIC" [Multiple APIC Description Table (MADT)] +[004h 0004 4] Table Length : 00000220 +[008h 0008 1] Revision : 03 +[009h 0009 1] Checksum : 63 +[00Ah 0010 6] Oem ID : "BOCHS " +[010h 0016 8] Oem Table ID : "BXPC " +[018h 0024 4] Oem Revision : 00000001 +[01Ch 0028 4] Asl Compiler ID : "BXPC" +[020h 0032 4] Asl Compiler Revision : 00000001 + +[024h 0036 4] Local Apic Address : FEE00000 +[028h 0040 4] Flags (decoded below) : 00000001 + PC-AT Compatibility : 1 + +[02Ch 0044 1] Subtable Type : 00 [Processor Local APIC] +[02Dh 0045 1] Length : 08 +[02Eh 0046 1] Processor ID : 00 +[02Fh 0047 1] Local Apic ID : 00 +[030h 0048 4] Flags (decoded below) : 00000001 + Processor Enabled : 1 + Runtime Online Capable : 0 + +[034h 0052 1] Subtable Type : 00 [Processor Local APIC] +[035h 0053 1] Length : 08 +[036h 0054 1] Processor ID : 01 +[037h 0055 1] Local Apic ID : 01 +[038h 0056 4] Flags (decoded below) : 00000001 + Processor Enabled : 1 + Runtime Online Capable : 0 [snip] +[1D4h 0468 1] Subtable Type : 00 [Processor Local APIC] +[1D5h 0469 1] Length : 08 +[1D6h 0470 1] Processor ID : 35 +[1D7h 0471 1] Local Apic ID : 6A +[1D8h 0472 4] Flags (decoded below) : 00000000 + Processor Enabled : 0 + Runtime Online Capable : 0 + +[1DCh 0476 1] Subtable Type : 01 [I/O APIC] +[1DDh 0477 1] Length : 0C +[1DEh 0478 1] I/O Apic ID : 00 +[1DFh 0479 1] Reserved : 00 +[1E0h 0480 4] Address : FEC00000 +[1E4h 0484 4] Interrupt : 00000000 + +[1E8h 0488 1] Subtable Type : 02 [Interrupt Source Override] +[1E9h 0489 1] Length : 0A +[1EAh 0490 1] Bus : 00 +[1EBh 0491 1] Source : 00 +[1ECh 0492 4] Interrupt : 00000002 +[1F0h 0496 2] Flags (decoded below) : 0000 + Polarity : 0 + Trigger Mode : 0 + +[1F2h 0498 1] Subtable Type : 02 [Interrupt Source Override] +[1F3h 0499 1] Length : 0A +[1F4h 0500 1] Bus : 00 +[1F5h 0501 1] Source : 05 +[1F6h 0502 4] Interrupt : 00000005 +[1FAh 0506 2] Flags (decoded below) : 000D + Polarity : 1 + Trigger Mode : 3 + +[1FCh 0508 1] Subtable Type : 02 [Interrupt Source Override] +[1FDh 0509 1] Length : 0A +[1FEh 0510 1] Bus : 00 +[1FFh 0511 1] Source : 09 +[200h 0512 4] Interrupt : 00000009 +[204h 0516 2] Flags (decoded below) : 000D + Polarity : 1 + Trigger Mode : 3 + +[206h 0518 1] Subtable Type : 02 [Interrupt Source Override] +[207h 0519 1] Length : 0A +[208h 0520 1] Bus : 00 +[209h 0521 1] Source : 0A +[20Ah 0522 4] Interrupt : 0000000A +[20Eh 0526 2] Flags (decoded below) : 000D + Polarity : 1 + Trigger Mode : 3 + +[210h 0528 1] Subtable Type : 02 [Interrupt Source Override] +[211h 0529 1] Length : 0A +[212h 0530 1] Bus : 00 +[213h 0531 1] Source : 0B +[214h 0532 4] Interrupt : 0000000B +[218h 0536 2] Flags (decoded below) : 000D + Polarity : 1 + Trigger Mode : 3 + +[21Ah 0538 1] Subtable Type : 04 [Local APIC NMI] +[21Bh 0539 1] Length : 06 +[21Ch 0540 1] Processor ID : FF +[21Dh 0541 2] Flags (decoded below) : 0000 + Polarity : 0 + Trigger Mode : 0 +[21Fh 0543 1] Interrupt Input LINT : 01 ... DSDT: +/* + * Intel ACPI Component Architecture + * AML/ASL+ Disassembler version 20200925 (64-bit version) + * Copyright (c) 2000 - 2020 Intel Corporation + * + * Disassembling to symbolic ASL+ operators + * + * Disassembly of /tmp/aml-00O791, Wed Aug 23 21:51:31 2023 + * + * Original Table Header: + * Signature "DSDT" + * Length 0x00003271 (12913) + * Revision 0x01 **** 32-bit table (V1), no 64-bit math support + * Checksum 0xAF + * OEM ID "BOCHS " + * OEM Table ID "BXPC " + * OEM Revision 0x00000001 (1) + * Compiler ID "BXPC" + * Compiler Version 0x00000001 (1) + */ +DefinitionBlock ("", "DSDT", 1, "BOCHS ", "BXPC ", 0x00000001) +{ + Scope (\) + { + OperationRegion (DBG, SystemIO, 0x0402, One) + Field (DBG, ByteAcc, NoLock, Preserve) + { + DBGB, 8 + } + + Method (DBUG, 1, NotSerialized) + { + ToHexString (Arg0, Local0) + ToBuffer (Local0, Local0) + Local1 = (SizeOf (Local0) - One) + Local2 = Zero + While ((Local2 < Local1)) + { + DBGB = DerefOf (Local0 [Local2]) + Local2++ + } + + DBGB = 0x0A + } + } [snip] + Processor (C000, 0x00, 0x00000000, 0x00) + { + Method (_STA, 0, Serialized) // _STA: Status + { + Return (CSTA (Zero)) + } + + Name (_MAT, Buffer (0x08) // _MAT: Multiple APIC Table Entry + { + 0x00, 0x08, 0x00, 0x00, 0x01, 0x00, 0x00, 0x00 // ........ + }) + Method (_OST, 3, Serialized) // _OST: OSPM Status Indication + { + COST (Zero, Arg0, Arg1, Arg2) + } + } + + Processor (C001, 0x01, 0x00000000, 0x00) + { + Method (_STA, 0, Serialized) // _STA: Status + { + Return (CSTA (One)) + } + + Name (_MAT, Buffer (0x08) // _MAT: Multiple APIC Table Entry + { + 0x00, 0x08, 0x01, 0x01, 0x01, 0x00, 0x00, 0x00 // ........ + }) + Method (_EJ0, 1, NotSerialized) // _EJx: Eject Device, x=0-9 + { + CEJ0 (One) + } + + Method (_OST, 3, Serialized) // _OST: OSPM Status Indication + { + COST (One, Arg0, Arg1, Arg2) + } + } [snip] + Processor (C035, 0x35, 0x00000000, 0x00) + { + Method (_STA, 0, Serialized) // _STA: Status + { + Return (CSTA (0x35)) + } + + Name (_MAT, Buffer (0x08) // _MAT: Multiple APIC Table Entry + { + 0x00, 0x08, 0x35, 0x6A, 0x01, 0x00, 0x00, 0x00 // ..5j.... + }) + Method (_EJ0, 1, NotSerialized) // _EJx: Eject Device, x=0-9 + { + CEJ0 (0x35) + } + + Method (_OST, 3, Serialized) // _OST: OSPM Status Indication + { + COST (0x35, Arg0, Arg1, Arg2) + } + } ... Signed-off-by: Zhao Liu Reviewed-by: Michael S. Tsirkin Message-Id: <20231023094635.1588282-14-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/data/acpi/q35/APIC.thread-count | Bin 0 -> 544 bytes tests/data/acpi/q35/DSDT.thread-count | Bin 0 -> 12913 bytes tests/data/acpi/q35/FACP.thread-count | Bin 0 -> 244 bytes tests/qtest/bios-tables-test-allowed-diff.h | 3 --- 4 files changed, 3 deletions(-) diff --git a/tests/data/acpi/q35/APIC.thread-count b/tests/data/acpi/q35/APIC.thread-count index e69de29bb2..c27e87fcf1 100644 Binary files a/tests/data/acpi/q35/APIC.thread-count and b/tests/data/acpi/q35/APIC.thread-count differ diff --git a/tests/data/acpi/q35/DSDT.thread-count b/tests/data/acpi/q35/DSDT.thread-count index e69de29bb2..a24b04cbdb 100644 Binary files a/tests/data/acpi/q35/DSDT.thread-count and b/tests/data/acpi/q35/DSDT.thread-count differ diff --git a/tests/data/acpi/q35/FACP.thread-count b/tests/data/acpi/q35/FACP.thread-count index e69de29bb2..31fa5dd19c 100644 Binary files a/tests/data/acpi/q35/FACP.thread-count and b/tests/data/acpi/q35/FACP.thread-count differ diff --git a/tests/qtest/bios-tables-test-allowed-diff.h b/tests/qtest/bios-tables-test-allowed-diff.h index 4d139d7f6b..dfb8523c8b 100644 --- a/tests/qtest/bios-tables-test-allowed-diff.h +++ b/tests/qtest/bios-tables-test-allowed-diff.h @@ -1,4 +1 @@ /* List of comma-separated changed AML files to ignore */ -"tests/data/acpi/q35/APIC.thread-count", -"tests/data/acpi/q35/DSDT.thread-count", -"tests/data/acpi/q35/FACP.thread-count", -- cgit v1.2.3 From 7cb953ca1956d4ac5d8ad798321fe3fde765befc Mon Sep 17 00:00:00 2001 From: Zhao Liu Date: Mon, 23 Oct 2023 17:46:33 +0800 Subject: tests: bios-tables-test: Prepare the ACPI table change for smbios type4 thread count2 test Following the guidelines in tests/qtest/bios-tables-test.c, this is step 1 - 3. List the ACPI tables that will be added to test the thread count2 field of smbios type4 table. Signed-off-by: Zhao Liu Reviewed-by: Michael S. Tsirkin Message-Id: <20231023094635.1588282-15-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/data/acpi/q35/APIC.thread-count2 | 0 tests/data/acpi/q35/DSDT.thread-count2 | 0 tests/data/acpi/q35/FACP.thread-count2 | 0 tests/qtest/bios-tables-test-allowed-diff.h | 3 +++ 4 files changed, 3 insertions(+) create mode 100644 tests/data/acpi/q35/APIC.thread-count2 create mode 100644 tests/data/acpi/q35/DSDT.thread-count2 create mode 100644 tests/data/acpi/q35/FACP.thread-count2 diff --git a/tests/data/acpi/q35/APIC.thread-count2 b/tests/data/acpi/q35/APIC.thread-count2 new file mode 100644 index 0000000000..e69de29bb2 diff --git a/tests/data/acpi/q35/DSDT.thread-count2 b/tests/data/acpi/q35/DSDT.thread-count2 new file mode 100644 index 0000000000..e69de29bb2 diff --git a/tests/data/acpi/q35/FACP.thread-count2 b/tests/data/acpi/q35/FACP.thread-count2 new file mode 100644 index 0000000000..e69de29bb2 diff --git a/tests/qtest/bios-tables-test-allowed-diff.h b/tests/qtest/bios-tables-test-allowed-diff.h index dfb8523c8b..d17d80e21a 100644 --- a/tests/qtest/bios-tables-test-allowed-diff.h +++ b/tests/qtest/bios-tables-test-allowed-diff.h @@ -1 +1,4 @@ /* List of comma-separated changed AML files to ignore */ +"tests/data/acpi/q35/APIC.thread-count2", +"tests/data/acpi/q35/DSDT.thread-count2", +"tests/data/acpi/q35/FACP.thread-count2", -- cgit v1.2.3 From 198eee0cc17d74f17ecd54ddf72681f421f7bc43 Mon Sep 17 00:00:00 2001 From: Zhao Liu Date: Mon, 23 Oct 2023 17:46:34 +0800 Subject: tests: bios-tables-test: Add test for smbios type4 thread count2 This tests the commit 7298fd7de5551 ("hw/smbios: Fix thread count in type4"). In smbios_build_type_4_table() (hw/smbios/smbios.c), if the number of threads in the socket is more than 255, then smbios type4 table encodes threads per socket into the thread count2 field. So for the topology in this case, there're the following considerations: 1. threads per socket should be more than 255 to ensure we could cover the thread count2 field. 2. The original bug was that threads per socket was miscalculated, so now we should configure as many topology levels as possible (multiple dies, no module since x86 hasn't supported it) to cover more general topology scenarios, to ensure that the threads per socket encoded in the thread count2 field is correct. 3. For the more general topology, we should also add "cpus" (presented threads for machine) and "maxcpus" (total threads for machine) to make sure that configuring unpluged CPUs in smp (cpus < maxcpus) does not affect the correctness of threads per socket for thread count2 field. Note we don't consider the topology with multiple sockets since this topology would create too many vCPUs (more than 255 threads per socket with at least 2 sockets, which may cause the failure "Number of hotpluggable cpus requested (*) exceeds the maximum cpus supported by KVM (*) socket_accept failed: Resource temporarily unavailable"), and the calculation of threads per socket has already been covered by "thread count" test case. Based on these considerations, select the topology as the follow: -smp cpus=210,maxcpus=260,dies=2,cores=65,threads=2 The expected thread count2 = threads per socket = threads (2) * cores (65) * dies (2) = 260. Suggested-by: Igor Mammedov Signed-off-by: Zhao Liu Message-Id: <20231023094635.1588282-16-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/qtest/bios-tables-test.c | 31 +++++++++++++++++++++++++++++++ 1 file changed, 31 insertions(+) diff --git a/tests/qtest/bios-tables-test.c b/tests/qtest/bios-tables-test.c index 395ed7f9ff..71af5cf69f 100644 --- a/tests/qtest/bios-tables-test.c +++ b/tests/qtest/bios-tables-test.c @@ -96,6 +96,7 @@ typedef struct { uint8_t smbios_core_count; uint16_t smbios_core_count2; uint8_t smbios_thread_count; + uint16_t smbios_thread_count2; uint8_t *required_struct_types; int required_struct_types_len; int type4_count; @@ -644,6 +645,7 @@ static void smbios_cpu_test(test_data *data, uint32_t addr, uint8_t thread_count, expected_thread_count = data->smbios_thread_count; uint16_t speed, expected_speed[2]; uint16_t core_count2, expected_core_count2 = data->smbios_core_count2; + uint16_t thread_count2, expected_thread_count2 = data->smbios_thread_count2; int offset[2]; int i; @@ -680,6 +682,15 @@ static void smbios_cpu_test(test_data *data, uint32_t addr, if (expected_core_count == 0xFF && expected_core_count2) { g_assert_cmpuint(core_count2, ==, expected_core_count2); } + + thread_count2 = qtest_readw(data->qts, + addr + offsetof(struct smbios_type_4, + thread_count2)); + + /* Thread Count has reached its limit, checking Thread Count 2 */ + if (expected_thread_count == 0xFF && expected_thread_count2) { + g_assert_cmpuint(thread_count2, ==, expected_thread_count2); + } } } @@ -1050,6 +1061,7 @@ static void test_acpi_q35_tcg_thread_count(void) .required_struct_types = base_required_struct_types, .required_struct_types_len = ARRAY_SIZE(base_required_struct_types), .smbios_thread_count = 27, + .smbios_thread_count2 = 27, }; test_acpi_one("-machine smbios-entry-point-type=64 " @@ -1058,6 +1070,23 @@ static void test_acpi_q35_tcg_thread_count(void) free_test_data(&data); } +static void test_acpi_q35_tcg_thread_count2(void) +{ + test_data data = { + .machine = MACHINE_Q35, + .variant = ".thread-count2", + .required_struct_types = base_required_struct_types, + .required_struct_types_len = ARRAY_SIZE(base_required_struct_types), + .smbios_thread_count = 0xFF, + .smbios_thread_count2 = 260, + }; + + test_acpi_one("-machine smbios-entry-point-type=64 " + "-smp cpus=210,maxcpus=260,dies=2,cores=65,threads=2", + &data); + free_test_data(&data); +} + static void test_acpi_q35_tcg_bridge(void) { test_data data = {}; @@ -2228,6 +2257,8 @@ int main(int argc, char *argv[]) test_acpi_q35_tcg_core_count2); qtest_add_func("acpi/q35/thread-count", test_acpi_q35_tcg_thread_count); + qtest_add_func("acpi/q35/thread-count2", + test_acpi_q35_tcg_thread_count2); } if (qtest_has_device("virtio-iommu-pci")) { qtest_add_func("acpi/q35/viot", test_acpi_q35_viot); -- cgit v1.2.3 From f58db4eeb1c7f850de86d4efece653d1018f6eae Mon Sep 17 00:00:00 2001 From: Zhao Liu Date: Mon, 23 Oct 2023 17:46:35 +0800 Subject: tests: bios-tables-test: Add ACPI table binaries for smbios type4 thread count2 test Following the guidelines in tests/qtest/bios-tables-test.c, this is step 5 and 6. Changes in the tables: FACP: +/* + * Intel ACPI Component Architecture + * AML/ASL+ Disassembler version 20200925 (64-bit version) + * Copyright (c) 2000 - 2020 Intel Corporation + * + * Disassembly of /tmp/aml-CNE3C2, Mon Oct 23 15:25:01 2023 + * + * ACPI Data Table [FACP] + * + * Format: [HexOffset DecimalOffset ByteLength] FieldName : FieldValue + */ + +[000h 0000 4] Signature : "FACP" [Fixed ACPI Description Table (FADT)] +[004h 0004 4] Table Length : 000000F4 +[008h 0008 1] Revision : 03 +[009h 0009 1] Checksum : B3 +[00Ah 0010 6] Oem ID : "BOCHS " +[010h 0016 8] Oem Table ID : "BXPC " +[018h 0024 4] Oem Revision : 00000001 +[01Ch 0028 4] Asl Compiler ID : "BXPC" +[020h 0032 4] Asl Compiler Revision : 00000001 + +[024h 0036 4] FACS Address : 00000000 +[028h 0040 4] DSDT Address : 00000000 +[02Ch 0044 1] Model : 01 +[02Dh 0045 1] PM Profile : 00 [Unspecified] +[02Eh 0046 2] SCI Interrupt : 0009 +[030h 0048 4] SMI Command Port : 000000B2 +[034h 0052 1] ACPI Enable Value : 02 +[035h 0053 1] ACPI Disable Value : 03 +[036h 0054 1] S4BIOS Command : 00 +[037h 0055 1] P-State Control : 00 +[038h 0056 4] PM1A Event Block Address : 00000600 +[03Ch 0060 4] PM1B Event Block Address : 00000000 +[040h 0064 4] PM1A Control Block Address : 00000604 +[044h 0068 4] PM1B Control Block Address : 00000000 +[048h 0072 4] PM2 Control Block Address : 00000000 +[04Ch 0076 4] PM Timer Block Address : 00000608 +[050h 0080 4] GPE0 Block Address : 00000620 +[054h 0084 4] GPE1 Block Address : 00000000 +[058h 0088 1] PM1 Event Block Length : 04 +[059h 0089 1] PM1 Control Block Length : 02 +[05Ah 0090 1] PM2 Control Block Length : 00 +[05Bh 0091 1] PM Timer Block Length : 04 +[05Ch 0092 1] GPE0 Block Length : 10 +[05Dh 0093 1] GPE1 Block Length : 00 +[05Eh 0094 1] GPE1 Base Offset : 00 +[05Fh 0095 1] _CST Support : 00 +[060h 0096 2] C2 Latency : 0FFF +[062h 0098 2] C3 Latency : 0FFF +[064h 0100 2] CPU Cache Size : 0000 +[066h 0102 2] Cache Flush Stride : 0000 +[068h 0104 1] Duty Cycle Offset : 00 +[069h 0105 1] Duty Cycle Width : 00 +[06Ah 0106 1] RTC Day Alarm Index : 00 +[06Bh 0107 1] RTC Month Alarm Index : 00 +[06Ch 0108 1] RTC Century Index : 32 +[06Dh 0109 2] Boot Flags (decoded below) : 0002 + Legacy Devices Supported (V2) : 0 + 8042 Present on ports 60/64 (V2) : 1 + VGA Not Present (V4) : 0 + MSI Not Supported (V4) : 0 + PCIe ASPM Not Supported (V4) : 0 + CMOS RTC Not Present (V5) : 0 +[06Fh 0111 1] Reserved : 00 +[070h 0112 4] Flags (decoded below) : 000484A5 + WBINVD instruction is operational (V1) : 1 + WBINVD flushes all caches (V1) : 0 + All CPUs support C1 (V1) : 1 + C2 works on MP system (V1) : 0 + Control Method Power Button (V1) : 0 + Control Method Sleep Button (V1) : 1 + RTC wake not in fixed reg space (V1) : 0 + RTC can wake system from S4 (V1) : 1 + 32-bit PM Timer (V1) : 0 + Docking Supported (V1) : 0 + Reset Register Supported (V2) : 1 + Sealed Case (V3) : 0 + Headless - No Video (V3) : 0 + Use native instr after SLP_TYPx (V3) : 0 + PCIEXP_WAK Bits Supported (V4) : 0 + Use Platform Timer (V4) : 1 + RTC_STS valid on S4 wake (V4) : 0 + Remote Power-on capable (V4) : 0 + Use APIC Cluster Model (V4) : 1 + Use APIC Physical Destination Mode (V4) : 0 + Hardware Reduced (V5) : 0 + Low Power S0 Idle (V5) : 0 + +[074h 0116 12] Reset Register : [Generic Address Structure] +[074h 0116 1] Space ID : 01 [SystemIO] +[075h 0117 1] Bit Width : 08 +[076h 0118 1] Bit Offset : 00 +[077h 0119 1] Encoded Access Width : 00 [Undefined/Legacy] +[078h 0120 8] Address : 0000000000000CF9 + +[080h 0128 1] Value to cause reset : 0F +[081h 0129 2] ARM Flags (decoded below) : 0000 + PSCI Compliant : 0 + Must use HVC for PSCI : 0 + +[083h 0131 1] FADT Minor Revision : 00 +[084h 0132 8] FACS Address : 0000000000000000 +[08Ch 0140 8] DSDT Address : 0000000000000000 +[094h 0148 12] PM1A Event Block : [Generic Address Structure] +[094h 0148 1] Space ID : 01 [SystemIO] +[095h 0149 1] Bit Width : 20 +[096h 0150 1] Bit Offset : 00 +[097h 0151 1] Encoded Access Width : 00 [Undefined/Legacy] +[098h 0152 8] Address : 0000000000000600 + +[0A0h 0160 12] PM1B Event Block : [Generic Address Structure] +[0A0h 0160 1] Space ID : 00 [SystemMemory] +[0A1h 0161 1] Bit Width : 00 +[0A2h 0162 1] Bit Offset : 00 +[0A3h 0163 1] Encoded Access Width : 00 [Undefined/Legacy] +[0A4h 0164 8] Address : 0000000000000000 + +[0ACh 0172 12] PM1A Control Block : [Generic Address Structure] +[0ACh 0172 1] Space ID : 01 [SystemIO] +[0ADh 0173 1] Bit Width : 10 +[0AEh 0174 1] Bit Offset : 00 +[0AFh 0175 1] Encoded Access Width : 00 [Undefined/Legacy] +[0B0h 0176 8] Address : 0000000000000604 + +[0B8h 0184 12] PM1B Control Block : [Generic Address Structure] +[0B8h 0184 1] Space ID : 00 [SystemMemory] +[0B9h 0185 1] Bit Width : 00 +[0BAh 0186 1] Bit Offset : 00 +[0BBh 0187 1] Encoded Access Width : 00 [Undefined/Legacy] +[0BCh 0188 8] Address : 0000000000000000 + +[0C4h 0196 12] PM2 Control Block : [Generic Address Structure] +[0C4h 0196 1] Space ID : 00 [SystemMemory] +[0C5h 0197 1] Bit Width : 00 +[0C6h 0198 1] Bit Offset : 00 +[0C7h 0199 1] Encoded Access Width : 00 [Undefined/Legacy] +[0C8h 0200 8] Address : 0000000000000000 + +[0D0h 0208 12] PM Timer Block : [Generic Address Structure] +[0D0h 0208 1] Space ID : 01 [SystemIO] +[0D1h 0209 1] Bit Width : 20 +[0D2h 0210 1] Bit Offset : 00 +[0D3h 0211 1] Encoded Access Width : 00 [Undefined/Legacy] +[0D4h 0212 8] Address : 0000000000000608 + +[0DCh 0220 12] GPE0 Block : [Generic Address Structure] +[0DCh 0220 1] Space ID : 01 [SystemIO] +[0DDh 0221 1] Bit Width : 80 +[0DEh 0222 1] Bit Offset : 00 +[0DFh 0223 1] Encoded Access Width : 00 [Undefined/Legacy] +[0E0h 0224 8] Address : 0000000000000620 + +[0E8h 0232 12] GPE1 Block : [Generic Address Structure] +[0E8h 0232 1] Space ID : 00 [SystemMemory] +[0E9h 0233 1] Bit Width : 00 +[0EAh 0234 1] Bit Offset : 00 +[0EBh 0235 1] Encoded Access Width : 00 [Undefined/Legacy] +[0ECh 0236 8] Address : 0000000000000000 ... APIC: +/* + * Intel ACPI Component Architecture + * AML/ASL+ Disassembler version 20200925 (64-bit version) + * Copyright (c) 2000 - 2020 Intel Corporation + * + * Disassembly of /tmp/aml-WKE3C2, Mon Oct 23 15:25:01 2023 + * + * ACPI Data Table [APIC] + * + * Format: [HexOffset DecimalOffset ByteLength] FieldName : FieldValue + */ + +[000h 0000 4] Signature : "APIC" [Multiple APIC Description Table (MADT)] +[004h 0004 4] Table Length : 00000CA6 +[008h 0008 1] Revision : 03 +[009h 0009 1] Checksum : 2C +[00Ah 0010 6] Oem ID : "BOCHS " +[010h 0016 8] Oem Table ID : "BXPC " +[018h 0024 4] Oem Revision : 00000001 +[01Ch 0028 4] Asl Compiler ID : "BXPC" +[020h 0032 4] Asl Compiler Revision : 00000001 + +[024h 0036 4] Local Apic Address : FEE00000 +[028h 0040 4] Flags (decoded below) : 00000001 + PC-AT Compatibility : 1 + +[02Ch 0044 1] Subtable Type : 00 [Processor Local APIC] +[02Dh 0045 1] Length : 08 +[02Eh 0046 1] Processor ID : 00 +[02Fh 0047 1] Local Apic ID : 00 +[030h 0048 4] Flags (decoded below) : 00000001 + Processor Enabled : 1 + Runtime Online Capable : 0 + +[034h 0052 1] Subtable Type : 00 [Processor Local APIC] +[035h 0053 1] Length : 08 +[036h 0054 1] Processor ID : 01 +[037h 0055 1] Local Apic ID : 01 +[038h 0056 4] Flags (decoded below) : 00000001 + Processor Enabled : 1 + Runtime Online Capable : 0 [snip] +[434h 1076 1] Subtable Type : 00 [Processor Local APIC] +[435h 1077 1] Length : 08 +[436h 1078 1] Processor ID : 81 +[437h 1079 1] Local Apic ID : 81 +[438h 1080 4] Flags (decoded below) : 00000001 + Processor Enabled : 1 + Runtime Online Capable : 0 + +[43Ch 1084 1] Subtable Type : 09 [Processor Local x2APIC] +[43Dh 1085 1] Length : 10 +[43Eh 1086 2] Reserved : 0000 +[440h 1088 4] Processor x2Apic ID : 00000100 +[444h 1092 4] Flags (decoded below) : 00000001 + Processor Enabled : 1 +[448h 1096 4] Processor UID : 00000082 [snip] +[C4Ch 3148 1] Subtable Type : 09 [Processor Local x2APIC] +[C4Dh 3149 1] Length : 10 +[C4Eh 3150 2] Reserved : 0000 +[C50h 3152 4] Processor x2Apic ID : 00000181 +[C54h 3156 4] Flags (decoded below) : 00000000 + Processor Enabled : 0 +[C58h 3160 4] Processor UID : 00000103 + +[C5Ch 3164 1] Subtable Type : 01 [I/O APIC] +[C5Dh 3165 1] Length : 0C +[C5Eh 3166 1] I/O Apic ID : 00 +[C5Fh 3167 1] Reserved : 00 +[C60h 3168 4] Address : FEC00000 +[C64h 3172 4] Interrupt : 00000000 + +[C68h 3176 1] Subtable Type : 02 [Interrupt Source Override] +[C69h 3177 1] Length : 0A +[C6Ah 3178 1] Bus : 00 +[C6Bh 3179 1] Source : 00 +[C6Ch 3180 4] Interrupt : 00000002 +[C70h 3184 2] Flags (decoded below) : 0000 + Polarity : 0 + Trigger Mode : 0 [snip] +[C90h 3216 1] Subtable Type : 02 [Interrupt Source Override] +[C91h 3217 1] Length : 0A +[C92h 3218 1] Bus : 00 +[C93h 3219 1] Source : 0B +[C94h 3220 4] Interrupt : 0000000B +[C98h 3224 2] Flags (decoded below) : 000D + Polarity : 1 + Trigger Mode : 3 + +[C9Ah 3226 1] Subtable Type : 0A [Local x2APIC NMI] +[C9Bh 3227 1] Length : 0C +[C9Ch 3228 2] Flags (decoded below) : 0000 + Polarity : 0 + Trigger Mode : 0 +[C9Eh 3230 4] Processor UID : FFFFFFFF +[CA2h 3234 1] Interrupt Input LINT : 01 +[CA3h 3235 3] Reserved : 000000 ... DSDT: +/* + * Intel ACPI Component Architecture + * AML/ASL+ Disassembler version 20200925 (64-bit version) + * Copyright (c) 2000 - 2020 Intel Corporation + * + * Disassembling to symbolic ASL+ operators + * + * Disassembly of /tmp/aml-CDE3C2, Mon Oct 23 15:25:01 2023 + * + * Original Table Header: + * Signature "DSDT" + * Length 0x000083EA (33770) + * Revision 0x01 **** 32-bit table (V1), no 64-bit math support + * Checksum 0x01 + * OEM ID "BOCHS " + * OEM Table ID "BXPC " + * OEM Revision 0x00000001 (1) + * Compiler ID "BXPC" + * Compiler Version 0x00000001 (1) + */ +DefinitionBlock ("", "DSDT", 1, "BOCHS ", "BXPC ", 0x00000001) +{ + Scope (\) + { + OperationRegion (DBG, SystemIO, 0x0402, One) + Field (DBG, ByteAcc, NoLock, Preserve) + { + DBGB, 8 + } + + Method (DBUG, 1, NotSerialized) + { + ToHexString (Arg0, Local0) + ToBuffer (Local0, Local0) + Local1 = (SizeOf (Local0) - One) + Local2 = Zero + While ((Local2 < Local1)) + { + DBGB = DerefOf (Local0 [Local2]) + Local2++ + } + + DBGB = 0x0A + } + } [snip] + Processor (C000, 0x00, 0x00000000, 0x00) + { + Method (_STA, 0, Serialized) // _STA: Status + { + Return (CSTA (Zero)) + } + + Name (_MAT, Buffer (0x08) // _MAT: Multiple APIC Table Entry + { + 0x00, 0x08, 0x00, 0x00, 0x01, 0x00, 0x00, 0x00 // ........ + }) + Method (_OST, 3, Serialized) // _OST: OSPM Status Indication + { + COST (Zero, Arg0, Arg1, Arg2) + } + } [snip] + Processor (C081, 0x81, 0x00000000, 0x00) + { + Method (_STA, 0, Serialized) // _STA: Status + { + Return (CSTA (0x81)) + } + + Name (_MAT, Buffer (0x08) // _MAT: Multiple APIC Table Entry + { + 0x00, 0x08, 0x81, 0x81, 0x01, 0x00, 0x00, 0x00 // ........ + }) + Method (_EJ0, 1, NotSerialized) // _EJx: Eject Device, x=0-9 + { + CEJ0 (0x81) + } + + Method (_OST, 3, Serialized) // _OST: OSPM Status Indication + { + COST (0x81, Arg0, Arg1, Arg2) + } + } ... Signed-off-by: Zhao Liu Message-Id: <20231023094635.1588282-17-zhao1.liu@linux.intel.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/data/acpi/q35/APIC.thread-count2 | Bin 0 -> 3238 bytes tests/data/acpi/q35/DSDT.thread-count2 | Bin 0 -> 33770 bytes tests/data/acpi/q35/FACP.thread-count2 | Bin 0 -> 244 bytes tests/qtest/bios-tables-test-allowed-diff.h | 3 --- 4 files changed, 3 deletions(-) diff --git a/tests/data/acpi/q35/APIC.thread-count2 b/tests/data/acpi/q35/APIC.thread-count2 index e69de29bb2..ac200ab7aa 100644 Binary files a/tests/data/acpi/q35/APIC.thread-count2 and b/tests/data/acpi/q35/APIC.thread-count2 differ diff --git a/tests/data/acpi/q35/DSDT.thread-count2 b/tests/data/acpi/q35/DSDT.thread-count2 index e69de29bb2..3a0cb8c581 100644 Binary files a/tests/data/acpi/q35/DSDT.thread-count2 and b/tests/data/acpi/q35/DSDT.thread-count2 differ diff --git a/tests/data/acpi/q35/FACP.thread-count2 b/tests/data/acpi/q35/FACP.thread-count2 index e69de29bb2..31fa5dd19c 100644 Binary files a/tests/data/acpi/q35/FACP.thread-count2 and b/tests/data/acpi/q35/FACP.thread-count2 differ diff --git a/tests/qtest/bios-tables-test-allowed-diff.h b/tests/qtest/bios-tables-test-allowed-diff.h index d17d80e21a..dfb8523c8b 100644 --- a/tests/qtest/bios-tables-test-allowed-diff.h +++ b/tests/qtest/bios-tables-test-allowed-diff.h @@ -1,4 +1 @@ /* List of comma-separated changed AML files to ignore */ -"tests/data/acpi/q35/APIC.thread-count2", -"tests/data/acpi/q35/DSDT.thread-count2", -"tests/data/acpi/q35/FACP.thread-count2", -- cgit v1.2.3 From 629df5cc23cc9aec5d115cc9be3456458e2b44fa Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 15:02:06 +0100 Subject: hw/cxl: Use a switch to explicitly check size in caps_reg_read() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Bring this read function inline with the others that do check for unexpected size values. Also reduces line lengths to sub 80 chars. Signed-off-by: Jonathan Cameron Reviewed-by: Philippe Mathieu-Daudé Reviewed-by: Fan Ni Message-Id: <20231023140210.3089-2-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-device-utils.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/hw/cxl/cxl-device-utils.c b/hw/cxl/cxl-device-utils.c index bd68328032..eb7195272e 100644 --- a/hw/cxl/cxl-device-utils.c +++ b/hw/cxl/cxl-device-utils.c @@ -32,10 +32,13 @@ static uint64_t caps_reg_read(void *opaque, hwaddr offset, unsigned size) { CXLDeviceState *cxl_dstate = opaque; - if (size == 4) { - return cxl_dstate->caps_reg_state32[offset / sizeof(*cxl_dstate->caps_reg_state32)]; - } else { - return cxl_dstate->caps_reg_state64[offset / sizeof(*cxl_dstate->caps_reg_state64)]; + switch (size) { + case 4: + return cxl_dstate->caps_reg_state32[offset / size]; + case 8: + return cxl_dstate->caps_reg_state64[offset / size]; + default: + g_assert_not_reached(); } } -- cgit v1.2.3 From 388d6b574e282b02e8180f4cba428316a404deea Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 15:02:07 +0100 Subject: hw/cxl: Use switch statements for read and write of cachemem registers Establishing that only register accesses of size 4 and 8 can occur using these functions requires looking at their callers. Make it easier to see that by using switch statements. Assertions are used to enforce that the register storage is of the matching size, allowing fixed values to be used for divisors of the array indices. Suggested-by: Michael Tokarev Signed-off-by: Jonathan Cameron Reviewed-by: Fan Ni Message-Id: <20231023140210.3089-3-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-component-utils.c | 66 +++++++++++++++++++++++++++++--------------- 1 file changed, 43 insertions(+), 23 deletions(-) diff --git a/hw/cxl/cxl-component-utils.c b/hw/cxl/cxl-component-utils.c index f3bbf0fd13..9d4f4bc8d4 100644 --- a/hw/cxl/cxl-component-utils.c +++ b/hw/cxl/cxl-component-utils.c @@ -67,16 +67,24 @@ static uint64_t cxl_cache_mem_read_reg(void *opaque, hwaddr offset, CXLComponentState *cxl_cstate = opaque; ComponentRegisters *cregs = &cxl_cstate->crb; - if (size == 8) { + switch (size) { + case 4: + if (cregs->special_ops && cregs->special_ops->read) { + return cregs->special_ops->read(cxl_cstate, offset, 4); + } else { + QEMU_BUILD_BUG_ON(sizeof(*cregs->cache_mem_registers) != 4); + return cregs->cache_mem_registers[offset / 4]; + } + case 8: qemu_log_mask(LOG_UNIMP, "CXL 8 byte cache mem registers not implemented\n"); return 0; - } - - if (cregs->special_ops && cregs->special_ops->read) { - return cregs->special_ops->read(cxl_cstate, offset, size); - } else { - return cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)]; + default: + /* + * In line with specifiction limitaions on access sizes, this + * routine is not called with other sizes. + */ + g_assert_not_reached(); } } @@ -117,25 +125,37 @@ static void cxl_cache_mem_write_reg(void *opaque, hwaddr offset, uint64_t value, ComponentRegisters *cregs = &cxl_cstate->crb; uint32_t mask; - if (size == 8) { - qemu_log_mask(LOG_UNIMP, - "CXL 8 byte cache mem registers not implemented\n"); + switch (size) { + case 4: { + QEMU_BUILD_BUG_ON(sizeof(*cregs->cache_mem_regs_write_mask) != 4); + QEMU_BUILD_BUG_ON(sizeof(*cregs->cache_mem_registers) != 4); + mask = cregs->cache_mem_regs_write_mask[offset / 4]; + value &= mask; + /* RO bits should remain constant. Done by reading existing value */ + value |= ~mask & cregs->cache_mem_registers[offset / 4]; + if (cregs->special_ops && cregs->special_ops->write) { + cregs->special_ops->write(cxl_cstate, offset, value, size); + return; + } + + if (offset >= A_CXL_HDM_DECODER_CAPABILITY && + offset <= A_CXL_HDM_DECODER3_TARGET_LIST_HI) { + dumb_hdm_handler(cxl_cstate, offset, value); + } else { + cregs->cache_mem_registers[offset / 4] = value; + } return; } - mask = cregs->cache_mem_regs_write_mask[offset / sizeof(*cregs->cache_mem_regs_write_mask)]; - value &= mask; - /* RO bits should remain constant. Done by reading existing value */ - value |= ~mask & cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)]; - if (cregs->special_ops && cregs->special_ops->write) { - cregs->special_ops->write(cxl_cstate, offset, value, size); + case 8: + qemu_log_mask(LOG_UNIMP, + "CXL 8 byte cache mem registers not implemented\n"); return; - } - - if (offset >= A_CXL_HDM_DECODER_CAPABILITY && - offset <= A_CXL_HDM_DECODER3_TARGET_LIST_HI) { - dumb_hdm_handler(cxl_cstate, offset, value); - } else { - cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)] = value; + default: + /* + * In line with specifiction limitaions on access sizes, this + * routine is not called with other sizes. + */ + g_assert_not_reached(); } } -- cgit v1.2.3 From b34ae3c9064a976e718dc96e454d32c1d8409eba Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 15:02:08 +0100 Subject: hw/cxl: CXLDVSECPortExtensions renamed to CXLDVSECPortExt MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Done to reduce line lengths where this is used. Ext seems sufficiently obvious that it need not be spelt out fully. Signed-off-by: Jonathan Cameron Reviewed-by: Philippe Mathieu-Daudé Reviewed-by: Fan Ni Message-Id: <20231023140210.3089-4-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-component-utils.c | 49 +++++++++++++++++++++++++----------------- hw/pci-bridge/cxl_downstream.c | 2 +- hw/pci-bridge/cxl_root_port.c | 2 +- hw/pci-bridge/cxl_upstream.c | 2 +- include/hw/cxl/cxl_pci.h | 6 +++--- 5 files changed, 35 insertions(+), 26 deletions(-) diff --git a/hw/cxl/cxl-component-utils.c b/hw/cxl/cxl-component-utils.c index 9d4f4bc8d4..1f4ea11640 100644 --- a/hw/cxl/cxl-component-utils.c +++ b/hw/cxl/cxl-component-utils.c @@ -393,26 +393,35 @@ void cxl_component_create_dvsec(CXLComponentState *cxl, case NON_CXL_FUNCTION_MAP_DVSEC: break; /* Not yet implemented */ case EXTENSIONS_PORT_DVSEC: - wmask[offset + offsetof(CXLDVSECPortExtensions, control)] = 0x0F; - wmask[offset + offsetof(CXLDVSECPortExtensions, control) + 1] = 0x40; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_bus_base)] = 0xFF; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_bus_limit)] = 0xFF; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_memory_base)] = 0xF0; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_memory_base) + 1] = 0xFF; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_memory_limit)] = 0xF0; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_memory_limit) + 1] = 0xFF; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base)] = 0xF0; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base) + 1] = 0xFF; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit)] = 0xF0; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit) + 1] = 0xFF; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base_high)] = 0xFF; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base_high) + 1] = 0xFF; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base_high) + 2] = 0xFF; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base_high) + 3] = 0xFF; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit_high)] = 0xFF; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit_high) + 1] = 0xFF; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit_high) + 2] = 0xFF; - wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit_high) + 3] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExt, control)] = 0x0F; + wmask[offset + offsetof(CXLDVSECPortExt, control) + 1] = 0x40; + wmask[offset + offsetof(CXLDVSECPortExt, alt_bus_base)] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExt, alt_bus_limit)] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExt, alt_memory_base)] = 0xF0; + wmask[offset + offsetof(CXLDVSECPortExt, alt_memory_base) + 1] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExt, alt_memory_limit)] = 0xF0; + wmask[offset + offsetof(CXLDVSECPortExt, alt_memory_limit) + 1] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_base)] = 0xF0; + wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_base) + 1] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_limit)] = 0xF0; + wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_limit) + 1] = + 0xFF; + wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_base_high)] = + 0xFF; + wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_base_high) + 1] = + 0xFF; + wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_base_high) + 2] = + 0xFF; + wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_base_high) + 3] = + 0xFF; + wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_limit_high)] = + 0xFF; + wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_limit_high) + 1] = + 0xFF; + wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_limit_high) + 2] = + 0xFF; + wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_limit_high) + 3] = + 0xFF; break; case GPF_PORT_DVSEC: wmask[offset + offsetof(CXLDVSECPortGPF, phase1_ctrl)] = 0x0F; diff --git a/hw/pci-bridge/cxl_downstream.c b/hw/pci-bridge/cxl_downstream.c index 5a2b749c8e..8c0f759add 100644 --- a/hw/pci-bridge/cxl_downstream.c +++ b/hw/pci-bridge/cxl_downstream.c @@ -98,7 +98,7 @@ static void build_dvsecs(CXLComponentState *cxl) { uint8_t *dvsec; - dvsec = (uint8_t *)&(CXLDVSECPortExtensions){ 0 }; + dvsec = (uint8_t *)&(CXLDVSECPortExt){ 0 }; cxl_component_create_dvsec(cxl, CXL2_DOWNSTREAM_PORT, EXTENSIONS_PORT_DVSEC_LENGTH, EXTENSIONS_PORT_DVSEC, diff --git a/hw/pci-bridge/cxl_root_port.c b/hw/pci-bridge/cxl_root_port.c index 7dfd20aa67..8f97697631 100644 --- a/hw/pci-bridge/cxl_root_port.c +++ b/hw/pci-bridge/cxl_root_port.c @@ -107,7 +107,7 @@ static void build_dvsecs(CXLComponentState *cxl) { uint8_t *dvsec; - dvsec = (uint8_t *)&(CXLDVSECPortExtensions){ 0 }; + dvsec = (uint8_t *)&(CXLDVSECPortExt){ 0 }; cxl_component_create_dvsec(cxl, CXL2_ROOT_PORT, EXTENSIONS_PORT_DVSEC_LENGTH, EXTENSIONS_PORT_DVSEC, diff --git a/hw/pci-bridge/cxl_upstream.c b/hw/pci-bridge/cxl_upstream.c index a57806fb31..b81bb5fec9 100644 --- a/hw/pci-bridge/cxl_upstream.c +++ b/hw/pci-bridge/cxl_upstream.c @@ -116,7 +116,7 @@ static void build_dvsecs(CXLComponentState *cxl) { uint8_t *dvsec; - dvsec = (uint8_t *)&(CXLDVSECPortExtensions){ + dvsec = (uint8_t *)&(CXLDVSECPortExt){ .status = 0x1, /* Port Power Management Init Complete */ }; cxl_component_create_dvsec(cxl, CXL2_UPSTREAM_PORT, diff --git a/include/hw/cxl/cxl_pci.h b/include/hw/cxl/cxl_pci.h index 407be95b9e..ddf01a543b 100644 --- a/include/hw/cxl/cxl_pci.h +++ b/include/hw/cxl/cxl_pci.h @@ -86,7 +86,7 @@ typedef struct CXLDVSECDevice { QEMU_BUILD_BUG_ON(sizeof(CXLDVSECDevice) != 0x38); /* CXL 2.0 - 8.1.5 (ID 0003) */ -typedef struct CXLDVSECPortExtensions { +typedef struct CXLDVSECPortExt { DVSECHeader hdr; uint16_t status; uint16_t control; @@ -100,8 +100,8 @@ typedef struct CXLDVSECPortExtensions { uint32_t alt_prefetch_limit_high; uint32_t rcrb_base; uint32_t rcrb_base_high; -} CXLDVSECPortExtensions; -QEMU_BUILD_BUG_ON(sizeof(CXLDVSECPortExtensions) != 0x28); +} CXLDVSECPortExt; +QEMU_BUILD_BUG_ON(sizeof(CXLDVSECPortExt) != 0x28); #define PORT_CONTROL_OFFSET 0xc #define PORT_CONTROL_UNMASK_SBR 1 -- cgit v1.2.3 From b342489ae795f5c2a9f7a565bac8443ccb11b0ce Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 15:02:09 +0100 Subject: hw/cxl: Line length reductions Michael Tsirkin observed that there were some unnecessarily long lines in the CXL code in a recent review. This patch is intended to rectify that where it does not hurt readability. Signed-off-by: Jonathan Cameron Reviewed-by: Michael Tokarev Reviewed-by: Fan Ni Message-Id: <20231023140210.3089-5-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-cdat.c | 3 ++- hw/cxl/cxl-component-utils.c | 14 ++++++++------ hw/cxl/cxl-events.c | 9 ++++++--- hw/cxl/cxl-mailbox-utils.c | 21 ++++++++++++++------- hw/mem/cxl_type3.c | 31 +++++++++++++++++++------------ hw/mem/cxl_type3_stubs.c | 5 +++-- include/hw/cxl/cxl_component.h | 3 ++- include/hw/cxl/cxl_device.h | 5 +++-- include/hw/cxl/cxl_events.h | 3 ++- 9 files changed, 59 insertions(+), 35 deletions(-) diff --git a/hw/cxl/cxl-cdat.c b/hw/cxl/cxl-cdat.c index d246d6885b..639a2db3e1 100644 --- a/hw/cxl/cxl-cdat.c +++ b/hw/cxl/cxl-cdat.c @@ -60,7 +60,8 @@ static void ct3_build_cdat(CDATObject *cdat, Error **errp) return; } - cdat->built_buf_len = cdat->build_cdat_table(&cdat->built_buf, cdat->private); + cdat->built_buf_len = cdat->build_cdat_table(&cdat->built_buf, + cdat->private); if (!cdat->built_buf_len) { /* Build later as not all data available yet */ diff --git a/hw/cxl/cxl-component-utils.c b/hw/cxl/cxl-component-utils.c index 1f4ea11640..5ebd81daf3 100644 --- a/hw/cxl/cxl-component-utils.c +++ b/hw/cxl/cxl-component-utils.c @@ -241,7 +241,8 @@ static void hdm_init_common(uint32_t *reg_state, uint32_t *write_msk, ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_CAPABILITY, TARGET_COUNT, 1); ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_CAPABILITY, INTERLEAVE_256B, 1); ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_CAPABILITY, INTERLEAVE_4K, 1); - ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_CAPABILITY, POISON_ON_ERR_CAP, 0); + ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_CAPABILITY, + POISON_ON_ERR_CAP, 0); ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_GLOBAL_CONTROL, HDM_DECODER_ENABLE, 0); write_msk[R_CXL_HDM_DECODER_GLOBAL_CONTROL] = 0x3; @@ -264,15 +265,16 @@ static void hdm_init_common(uint32_t *reg_state, uint32_t *write_msk, } } -void cxl_component_register_init_common(uint32_t *reg_state, uint32_t *write_msk, +void cxl_component_register_init_common(uint32_t *reg_state, + uint32_t *write_msk, enum reg_type type) { int caps = 0; /* - * In CXL 2.0 the capabilities required for each CXL component are such that, - * with the ordering chosen here, a single number can be used to define - * which capabilities should be provided. + * In CXL 2.0 the capabilities required for each CXL component are such + * that, with the ordering chosen here, a single number can be used to + * define which capabilities should be provided. */ switch (type) { case CXL2_DOWNSTREAM_PORT: @@ -449,7 +451,7 @@ void cxl_component_create_dvsec(CXLComponentState *cxl, default: /* Registers are RO for other component types */ break; } - /* There are rw1cs bits in the status register but never set currently */ + /* There are rw1cs bits in the status register but never set */ break; } diff --git a/hw/cxl/cxl-events.c b/hw/cxl/cxl-events.c index 3ddd6369ad..e2172b94b9 100644 --- a/hw/cxl/cxl-events.c +++ b/hw/cxl/cxl-events.c @@ -170,8 +170,10 @@ CXLRetCode cxl_event_get_records(CXLDeviceState *cxlds, CXLGetEventPayload *pl, if (log->overflow_err_count) { pl->flags |= CXL_GET_EVENT_FLAG_OVERFLOW; pl->overflow_err_count = cpu_to_le16(log->overflow_err_count); - pl->first_overflow_timestamp = cpu_to_le64(log->first_overflow_timestamp); - pl->last_overflow_timestamp = cpu_to_le64(log->last_overflow_timestamp); + pl->first_overflow_timestamp = + cpu_to_le64(log->first_overflow_timestamp); + pl->last_overflow_timestamp = + cpu_to_le64(log->last_overflow_timestamp); } pl->record_count = cpu_to_le16(nr); @@ -180,7 +182,8 @@ CXLRetCode cxl_event_get_records(CXLDeviceState *cxlds, CXLGetEventPayload *pl, return CXL_MBOX_SUCCESS; } -CXLRetCode cxl_event_clear_records(CXLDeviceState *cxlds, CXLClearEventPayload *pl) +CXLRetCode cxl_event_clear_records(CXLDeviceState *cxlds, + CXLClearEventPayload *pl) { CXLEventLog *log; uint8_t log_type; diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c index 434ccc5f6e..ab082ec9de 100644 --- a/hw/cxl/cxl-mailbox-utils.c +++ b/hw/cxl/cxl-mailbox-utils.c @@ -366,9 +366,12 @@ static CXLRetCode cmd_identify_memory_device(struct cxl_cmd *cmd, snprintf(id->fw_revision, 0x10, "BWFW VERSION %02d", 0); - stq_le_p(&id->total_capacity, cxl_dstate->mem_size / CXL_CAPACITY_MULTIPLIER); - stq_le_p(&id->persistent_capacity, cxl_dstate->pmem_size / CXL_CAPACITY_MULTIPLIER); - stq_le_p(&id->volatile_capacity, cxl_dstate->vmem_size / CXL_CAPACITY_MULTIPLIER); + stq_le_p(&id->total_capacity, + cxl_dstate->mem_size / CXL_CAPACITY_MULTIPLIER); + stq_le_p(&id->persistent_capacity, + cxl_dstate->pmem_size / CXL_CAPACITY_MULTIPLIER); + stq_le_p(&id->volatile_capacity, + cxl_dstate->vmem_size / CXL_CAPACITY_MULTIPLIER); stl_le_p(&id->lsa_size, cvc->get_lsa_size(ct3d)); /* 256 poison records */ st24_le_p(id->poison_list_max_mer, 256); @@ -396,13 +399,15 @@ static CXLRetCode cmd_ccls_get_partition_info(struct cxl_cmd *cmd, return CXL_MBOX_INTERNAL_ERROR; } - stq_le_p(&part_info->active_vmem, cxl_dstate->vmem_size / CXL_CAPACITY_MULTIPLIER); + stq_le_p(&part_info->active_vmem, + cxl_dstate->vmem_size / CXL_CAPACITY_MULTIPLIER); /* * When both next_vmem and next_pmem are 0, there is no pending change to * partitioning. */ stq_le_p(&part_info->next_vmem, 0); - stq_le_p(&part_info->active_pmem, cxl_dstate->pmem_size / CXL_CAPACITY_MULTIPLIER); + stq_le_p(&part_info->active_pmem, + cxl_dstate->pmem_size / CXL_CAPACITY_MULTIPLIER); stq_le_p(&part_info->next_pmem, 0); *len = sizeof(*part_info); @@ -681,8 +686,10 @@ static struct cxl_cmd cxl_cmd_set[256][256] = { [FIRMWARE_UPDATE][GET_INFO] = { "FIRMWARE_UPDATE_GET_INFO", cmd_firmware_update_get_info, 0, 0 }, [TIMESTAMP][GET] = { "TIMESTAMP_GET", cmd_timestamp_get, 0, 0 }, - [TIMESTAMP][SET] = { "TIMESTAMP_SET", cmd_timestamp_set, 8, IMMEDIATE_POLICY_CHANGE }, - [LOGS][GET_SUPPORTED] = { "LOGS_GET_SUPPORTED", cmd_logs_get_supported, 0, 0 }, + [TIMESTAMP][SET] = { "TIMESTAMP_SET", cmd_timestamp_set, + 8, IMMEDIATE_POLICY_CHANGE }, + [LOGS][GET_SUPPORTED] = { "LOGS_GET_SUPPORTED", cmd_logs_get_supported, + 0, 0 }, [LOGS][GET_LOG] = { "LOGS_GET_LOG", cmd_logs_get_log, 0x18, 0 }, [IDENTIFY][MEMORY_DEVICE] = { "IDENTIFY_MEMORY_DEVICE", cmd_identify_memory_device, 0, 0 }, diff --git a/hw/mem/cxl_type3.c b/hw/mem/cxl_type3.c index c02be4ce45..18ad853f5b 100644 --- a/hw/mem/cxl_type3.c +++ b/hw/mem/cxl_type3.c @@ -208,10 +208,9 @@ static int ct3_build_cdat_table(CDATSubHeader ***cdat_table, void *priv) } if (nonvolatile_mr) { + uint64_t base = volatile_mr ? memory_region_size(volatile_mr) : 0; rc = ct3_build_cdat_entries_for_mr(&(table[cur_ent]), dsmad_handle++, - nonvolatile_mr, true, - (volatile_mr ? - memory_region_size(volatile_mr) : 0)); + nonvolatile_mr, true, base); if (rc < 0) { goto error_cleanup; } @@ -514,7 +513,8 @@ static void ct3d_reg_write(void *opaque, hwaddr offset, uint64_t value, case A_CXL_RAS_UNC_ERR_STATUS: { uint32_t capctrl = ldl_le_p(cache_mem + R_CXL_RAS_ERR_CAP_CTRL); - uint32_t fe = FIELD_EX32(capctrl, CXL_RAS_ERR_CAP_CTRL, FIRST_ERROR_POINTER); + uint32_t fe = FIELD_EX32(capctrl, CXL_RAS_ERR_CAP_CTRL, + FIRST_ERROR_POINTER); CXLError *cxl_err; uint32_t unc_err; @@ -533,7 +533,8 @@ static void ct3d_reg_write(void *opaque, hwaddr offset, uint64_t value, * closest to behavior of hardware not capable of multiple * header recording. */ - QTAILQ_FOREACH_SAFE(cxl_err, &ct3d->error_list, node, cxl_next) { + QTAILQ_FOREACH_SAFE(cxl_err, &ct3d->error_list, node, + cxl_next) { if ((1 << cxl_err->type) & value) { QTAILQ_REMOVE(&ct3d->error_list, cxl_err, node); g_free(cxl_err); @@ -1072,7 +1073,8 @@ void qmp_cxl_inject_poison(const char *path, uint64_t start, uint64_t length, if (((start >= p->start) && (start < p->start + p->length)) || ((start + length > p->start) && (start + length <= p->start + p->length))) { - error_setg(errp, "Overlap with existing poisoned region not supported"); + error_setg(errp, + "Overlap with existing poisoned region not supported"); return; } } @@ -1085,7 +1087,8 @@ void qmp_cxl_inject_poison(const char *path, uint64_t start, uint64_t length, p = g_new0(CXLPoison, 1); p->length = length; p->start = start; - p->type = CXL_POISON_TYPE_INTERNAL; /* Different from injected via the mbox */ + /* Different from injected via the mbox */ + p->type = CXL_POISON_TYPE_INTERNAL; QLIST_INSERT_HEAD(&ct3d->poison_list, p, node); ct3d->poison_list_cnt++; @@ -1222,7 +1225,8 @@ void qmp_cxl_inject_correctable_error(const char *path, CxlCorErrorType type, return; } /* If the error is masked, nothting to do here */ - if (!((1 << cxl_err_type) & ~ldl_le_p(reg_state + R_CXL_RAS_COR_ERR_MASK))) { + if (!((1 << cxl_err_type) & + ~ldl_le_p(reg_state + R_CXL_RAS_COR_ERR_MASK))) { return; } @@ -1372,7 +1376,8 @@ void qmp_cxl_inject_dram_event(const char *path, CxlEventLog log, uint8_t flags, bool has_bank, uint8_t bank, bool has_row, uint32_t row, bool has_column, uint16_t column, - bool has_correction_mask, uint64List *correction_mask, + bool has_correction_mask, + uint64List *correction_mask, Error **errp) { Object *obj = object_resolve_path(path, NULL); @@ -1473,7 +1478,7 @@ void qmp_cxl_inject_memory_module_event(const char *path, CxlEventLog log, int16_t temperature, uint32_t dirty_shutdown_count, uint32_t corrected_volatile_error_count, - uint32_t corrected_persistent_error_count, + uint32_t corrected_persist_error_count, Error **errp) { Object *obj = object_resolve_path(path, NULL); @@ -1513,8 +1518,10 @@ void qmp_cxl_inject_memory_module_event(const char *path, CxlEventLog log, module.life_used = life_used; stw_le_p(&module.temperature, temperature); stl_le_p(&module.dirty_shutdown_count, dirty_shutdown_count); - stl_le_p(&module.corrected_volatile_error_count, corrected_volatile_error_count); - stl_le_p(&module.corrected_persistent_error_count, corrected_persistent_error_count); + stl_le_p(&module.corrected_volatile_error_count, + corrected_volatile_error_count); + stl_le_p(&module.corrected_persistent_error_count, + corrected_persist_error_count); if (cxl_event_insert(cxlds, enc_log, (CXLEventRecordRaw *)&module)) { cxl_event_irq_assert(ct3d); diff --git a/hw/mem/cxl_type3_stubs.c b/hw/mem/cxl_type3_stubs.c index 8ba5d3d1f7..3e1851e32b 100644 --- a/hw/mem/cxl_type3_stubs.c +++ b/hw/mem/cxl_type3_stubs.c @@ -33,7 +33,8 @@ void qmp_cxl_inject_dram_event(const char *path, CxlEventLog log, uint8_t flags, bool has_bank, uint8_t bank, bool has_row, uint32_t row, bool has_column, uint16_t column, - bool has_correction_mask, uint64List *correction_mask, + bool has_correction_mask, + uint64List *correction_mask, Error **errp) {} void qmp_cxl_inject_memory_module_event(const char *path, CxlEventLog log, @@ -45,7 +46,7 @@ void qmp_cxl_inject_memory_module_event(const char *path, CxlEventLog log, int16_t temperature, uint32_t dirty_shutdown_count, uint32_t corrected_volatile_error_count, - uint32_t corrected_persistent_error_count, + uint32_t corrected_persist_error_count, Error **errp) {} void qmp_cxl_inject_poison(const char *path, uint64_t start, uint64_t length, diff --git a/include/hw/cxl/cxl_component.h b/include/hw/cxl/cxl_component.h index 3c795a6278..e52dd8d2b9 100644 --- a/include/hw/cxl/cxl_component.h +++ b/include/hw/cxl/cxl_component.h @@ -175,7 +175,8 @@ HDM_DECODER_INIT(3); (CXL_IDE_REGISTERS_OFFSET + CXL_IDE_REGISTERS_SIZE) #define CXL_SNOOP_REGISTERS_SIZE 0x8 -QEMU_BUILD_BUG_MSG((CXL_SNOOP_REGISTERS_OFFSET + CXL_SNOOP_REGISTERS_SIZE) >= 0x1000, +QEMU_BUILD_BUG_MSG((CXL_SNOOP_REGISTERS_OFFSET + + CXL_SNOOP_REGISTERS_SIZE) >= 0x1000, "No space for registers"); typedef struct component_registers { diff --git a/include/hw/cxl/cxl_device.h b/include/hw/cxl/cxl_device.h index 51cd0d9ce3..007ddaf078 100644 --- a/include/hw/cxl/cxl_device.h +++ b/include/hw/cxl/cxl_device.h @@ -192,7 +192,7 @@ void cxl_device_register_init_common(CXLDeviceState *dev); * Documented as a 128 bit register, but 64 bit accesses and the second * 64 bits are currently reserved. */ -REG64(CXL_DEV_CAP_ARRAY, 0) /* Documented as 128 bit register but 64 byte accesses */ +REG64(CXL_DEV_CAP_ARRAY, 0) FIELD(CXL_DEV_CAP_ARRAY, CAP_ID, 0, 16) FIELD(CXL_DEV_CAP_ARRAY, CAP_VERSION, 16, 8) FIELD(CXL_DEV_CAP_ARRAY, CAP_COUNT, 32, 16) @@ -361,7 +361,8 @@ struct CXLType3Class { uint64_t offset); void (*set_lsa)(CXLType3Dev *ct3d, const void *buf, uint64_t size, uint64_t offset); - bool (*set_cacheline)(CXLType3Dev *ct3d, uint64_t dpa_offset, uint8_t *data); + bool (*set_cacheline)(CXLType3Dev *ct3d, uint64_t dpa_offset, + uint8_t *data); }; MemTxResult cxl_type3_read(PCIDevice *d, hwaddr host_addr, uint64_t *data, diff --git a/include/hw/cxl/cxl_events.h b/include/hw/cxl/cxl_events.h index 089ba2091f..d778487b7e 100644 --- a/include/hw/cxl/cxl_events.h +++ b/include/hw/cxl/cxl_events.h @@ -92,7 +92,8 @@ typedef enum CXLEventIntMode { CXL_INT_RES = 0x03, } CXLEventIntMode; #define CXL_EVENT_INT_MODE_MASK 0x3 -#define CXL_EVENT_INT_SETTING(vector) ((((uint8_t)vector & 0xf) << 4) | CXL_INT_MSI_MSIX) +#define CXL_EVENT_INT_SETTING(vector) \ + ((((uint8_t)vector & 0xf) << 4) | CXL_INT_MSI_MSIX) typedef struct CXLEventInterruptPolicy { uint8_t info_settings; uint8_t warn_settings; -- cgit v1.2.3 From 45234c2dd2920f16f768bda1ec8353bda8c5a929 Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 15:02:10 +0100 Subject: hw/cxl: Fix a QEMU_BUILD_BUG_ON() in switch statement scope issue. As _Static_assert is a declaration, it can't follow a label until C23. Some older versions of GCC trip up on this one. This check has no obvious purpose so just remove it. Reported-by: Jeongtae Park Signed-off-by: Jonathan Cameron Message-Id: <20231023140210.3089-6-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-component-utils.c | 1 - 1 file changed, 1 deletion(-) diff --git a/hw/cxl/cxl-component-utils.c b/hw/cxl/cxl-component-utils.c index 5ebd81daf3..d0245cc55d 100644 --- a/hw/cxl/cxl-component-utils.c +++ b/hw/cxl/cxl-component-utils.c @@ -305,7 +305,6 @@ void cxl_component_register_init_common(uint32_t *reg_state, ARRAY_FIELD_DP32(reg_state, CXL_CAPABILITY_HEADER, ARRAY_SIZE, caps); #define init_cap_reg(reg, id, version) \ - QEMU_BUILD_BUG_ON(CXL_##reg##_REGISTERS_OFFSET == 0); \ do { \ int which = R_CXL_##reg##_CAPABILITY_HEADER; \ reg_state[which] = FIELD_DP32(reg_state[which], \ -- cgit v1.2.3 From a7bb53b1eee6836d29003a8ffbd2c2c4c55a40c2 Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 17:07:50 +0100 Subject: hw/cxl/mbox: Pull the payload out of struct cxl_cmd and make instances constant Putting the pointer in the structure for command handling puts a single variable element inside an otherwise constant structure. Move it out as a directly passed variable and take the cxl_cmd structures constant. Signed-off-by: Jonathan Cameron Reviewed-by: Fan Ni Message-Id: <20231023160806.13206-2-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-mailbox-utils.c | 121 ++++++++++++++++++++++++-------------------- include/hw/cxl/cxl_device.h | 13 +++++ 2 files changed, 78 insertions(+), 56 deletions(-) diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c index ab082ec9de..c02de06943 100644 --- a/hw/cxl/cxl-mailbox-utils.c +++ b/hw/cxl/cxl-mailbox-utils.c @@ -69,18 +69,9 @@ enum { #define CLEAR_POISON 0x2 }; -struct cxl_cmd; -typedef CXLRetCode (*opcode_handler)(struct cxl_cmd *cmd, - CXLDeviceState *cxl_dstate, uint16_t *len); -struct cxl_cmd { - const char *name; - opcode_handler handler; - ssize_t in; - uint16_t effect; /* Reported in CEL */ - uint8_t *payload; -}; -static CXLRetCode cmd_events_get_records(struct cxl_cmd *cmd, +static CXLRetCode cmd_events_get_records(const struct cxl_cmd *cmd, + uint8_t *payload, CXLDeviceState *cxlds, uint16_t *len) { @@ -92,9 +83,9 @@ static CXLRetCode cmd_events_get_records(struct cxl_cmd *cmd, return CXL_MBOX_INVALID_INPUT; } - log_type = *((uint8_t *)cmd->payload); + log_type = payload[0]; - pl = (CXLGetEventPayload *)cmd->payload; + pl = (CXLGetEventPayload *)payload; memset(pl, 0, sizeof(*pl)); max_recs = (cxlds->payload_size - CXL_EVENT_PAYLOAD_HDR_SIZE) / @@ -106,25 +97,27 @@ static CXLRetCode cmd_events_get_records(struct cxl_cmd *cmd, return cxl_event_get_records(cxlds, pl, log_type, max_recs, len); } -static CXLRetCode cmd_events_clear_records(struct cxl_cmd *cmd, +static CXLRetCode cmd_events_clear_records(const struct cxl_cmd *cmd, + uint8_t *payload, CXLDeviceState *cxlds, uint16_t *len) { CXLClearEventPayload *pl; - pl = (CXLClearEventPayload *)cmd->payload; + pl = (CXLClearEventPayload *)payload; *len = 0; return cxl_event_clear_records(cxlds, pl); } -static CXLRetCode cmd_events_get_interrupt_policy(struct cxl_cmd *cmd, +static CXLRetCode cmd_events_get_interrupt_policy(const struct cxl_cmd *cmd, + uint8_t *payload, CXLDeviceState *cxlds, uint16_t *len) { CXLEventInterruptPolicy *policy; CXLEventLog *log; - policy = (CXLEventInterruptPolicy *)cmd->payload; + policy = (CXLEventInterruptPolicy *)payload; memset(policy, 0, sizeof(*policy)); log = &cxlds->event_logs[CXL_EVENT_TYPE_INFO]; @@ -157,7 +150,8 @@ static CXLRetCode cmd_events_get_interrupt_policy(struct cxl_cmd *cmd, return CXL_MBOX_SUCCESS; } -static CXLRetCode cmd_events_set_interrupt_policy(struct cxl_cmd *cmd, +static CXLRetCode cmd_events_set_interrupt_policy(const struct cxl_cmd *cmd, + uint8_t *payload, CXLDeviceState *cxlds, uint16_t *len) { @@ -168,7 +162,7 @@ static CXLRetCode cmd_events_set_interrupt_policy(struct cxl_cmd *cmd, return CXL_MBOX_INVALID_PAYLOAD_LENGTH; } - policy = (CXLEventInterruptPolicy *)cmd->payload; + policy = (CXLEventInterruptPolicy *)payload; log = &cxlds->event_logs[CXL_EVENT_TYPE_INFO]; log->irq_enabled = (policy->info_settings & CXL_EVENT_INT_MODE_MASK) == @@ -200,7 +194,8 @@ static CXLRetCode cmd_events_set_interrupt_policy(struct cxl_cmd *cmd, } /* 8.2.9.2.1 */ -static CXLRetCode cmd_firmware_update_get_info(struct cxl_cmd *cmd, +static CXLRetCode cmd_firmware_update_get_info(const struct cxl_cmd *cmd, + uint8_t *payload, CXLDeviceState *cxl_dstate, uint16_t *len) { @@ -221,7 +216,7 @@ static CXLRetCode cmd_firmware_update_get_info(struct cxl_cmd *cmd, return CXL_MBOX_INTERNAL_ERROR; } - fw_info = (void *)cmd->payload; + fw_info = (void *)payload; memset(fw_info, 0, sizeof(*fw_info)); fw_info->slots_supported = 2; @@ -234,27 +229,29 @@ static CXLRetCode cmd_firmware_update_get_info(struct cxl_cmd *cmd, } /* 8.2.9.3.1 */ -static CXLRetCode cmd_timestamp_get(struct cxl_cmd *cmd, +static CXLRetCode cmd_timestamp_get(const struct cxl_cmd *cmd, + uint8_t *payload, CXLDeviceState *cxl_dstate, uint16_t *len) { uint64_t final_time = cxl_device_get_timestamp(cxl_dstate); - stq_le_p(cmd->payload, final_time); + stq_le_p(payload, final_time); *len = 8; return CXL_MBOX_SUCCESS; } /* 8.2.9.3.2 */ -static CXLRetCode cmd_timestamp_set(struct cxl_cmd *cmd, - CXLDeviceState *cxl_dstate, - uint16_t *len) +static CXLRetCode cmd_timestamp_set(const struct cxl_cmd *cmd, + uint8_t *payload, + CXLDeviceState *cxl_dstate, + uint16_t *len) { cxl_dstate->timestamp.set = true; cxl_dstate->timestamp.last_set = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL); - cxl_dstate->timestamp.host_set = le64_to_cpu(*(uint64_t *)cmd->payload); + cxl_dstate->timestamp.host_set = le64_to_cpu(*(uint64_t *)payload); *len = 0; return CXL_MBOX_SUCCESS; @@ -267,7 +264,8 @@ static const QemuUUID cel_uuid = { }; /* 8.2.9.4.1 */ -static CXLRetCode cmd_logs_get_supported(struct cxl_cmd *cmd, +static CXLRetCode cmd_logs_get_supported(const struct cxl_cmd *cmd, + uint8_t *payload, CXLDeviceState *cxl_dstate, uint16_t *len) { @@ -278,7 +276,7 @@ static CXLRetCode cmd_logs_get_supported(struct cxl_cmd *cmd, QemuUUID uuid; uint32_t size; } log_entries[1]; - } QEMU_PACKED *supported_logs = (void *)cmd->payload; + } QEMU_PACKED *supported_logs = (void *)payload; QEMU_BUILD_BUG_ON(sizeof(*supported_logs) != 0x1c); supported_logs->entries = 1; @@ -290,7 +288,8 @@ static CXLRetCode cmd_logs_get_supported(struct cxl_cmd *cmd, } /* 8.2.9.4.2 */ -static CXLRetCode cmd_logs_get_log(struct cxl_cmd *cmd, +static CXLRetCode cmd_logs_get_log(const struct cxl_cmd *cmd, + uint8_t *payload, CXLDeviceState *cxl_dstate, uint16_t *len) { @@ -298,7 +297,9 @@ static CXLRetCode cmd_logs_get_log(struct cxl_cmd *cmd, QemuUUID uuid; uint32_t offset; uint32_t length; - } QEMU_PACKED QEMU_ALIGNED(16) *get_log = (void *)cmd->payload; + } QEMU_PACKED QEMU_ALIGNED(16) *get_log; + + get_log = (void *)payload; /* * 8.2.9.4.2 @@ -324,14 +325,15 @@ static CXLRetCode cmd_logs_get_log(struct cxl_cmd *cmd, /* Store off everything to local variables so we can wipe out the payload */ *len = get_log->length; - memmove(cmd->payload, cxl_dstate->cel_log + get_log->offset, + memmove(payload, cxl_dstate->cel_log + get_log->offset, get_log->length); return CXL_MBOX_SUCCESS; } /* 8.2.9.5.1.1 */ -static CXLRetCode cmd_identify_memory_device(struct cxl_cmd *cmd, +static CXLRetCode cmd_identify_memory_device(const struct cxl_cmd *cmd, + uint8_t *payload, CXLDeviceState *cxl_dstate, uint16_t *len) { @@ -361,7 +363,7 @@ static CXLRetCode cmd_identify_memory_device(struct cxl_cmd *cmd, return CXL_MBOX_INTERNAL_ERROR; } - id = (void *)cmd->payload; + id = (void *)payload; memset(id, 0, sizeof(*id)); snprintf(id->fw_revision, 0x10, "BWFW VERSION %02d", 0); @@ -382,7 +384,8 @@ static CXLRetCode cmd_identify_memory_device(struct cxl_cmd *cmd, return CXL_MBOX_SUCCESS; } -static CXLRetCode cmd_ccls_get_partition_info(struct cxl_cmd *cmd, +static CXLRetCode cmd_ccls_get_partition_info(const struct cxl_cmd *cmd, + uint8_t *payload, CXLDeviceState *cxl_dstate, uint16_t *len) { @@ -391,7 +394,7 @@ static CXLRetCode cmd_ccls_get_partition_info(struct cxl_cmd *cmd, uint64_t active_pmem; uint64_t next_vmem; uint64_t next_pmem; - } QEMU_PACKED *part_info = (void *)cmd->payload; + } QEMU_PACKED *part_info = (void *)payload; QEMU_BUILD_BUG_ON(sizeof(*part_info) != 0x20); if ((!QEMU_IS_ALIGNED(cxl_dstate->vmem_size, CXL_CAPACITY_MULTIPLIER)) || @@ -414,7 +417,8 @@ static CXLRetCode cmd_ccls_get_partition_info(struct cxl_cmd *cmd, return CXL_MBOX_SUCCESS; } -static CXLRetCode cmd_ccls_get_lsa(struct cxl_cmd *cmd, +static CXLRetCode cmd_ccls_get_lsa(const struct cxl_cmd *cmd, + uint8_t *payload, CXLDeviceState *cxl_dstate, uint16_t *len) { @@ -426,7 +430,7 @@ static CXLRetCode cmd_ccls_get_lsa(struct cxl_cmd *cmd, CXLType3Class *cvc = CXL_TYPE3_GET_CLASS(ct3d); uint32_t offset, length; - get_lsa = (void *)cmd->payload; + get_lsa = (void *)payload; offset = get_lsa->offset; length = get_lsa->length; @@ -439,7 +443,8 @@ static CXLRetCode cmd_ccls_get_lsa(struct cxl_cmd *cmd, return CXL_MBOX_SUCCESS; } -static CXLRetCode cmd_ccls_set_lsa(struct cxl_cmd *cmd, +static CXLRetCode cmd_ccls_set_lsa(const struct cxl_cmd *cmd, + uint8_t *payload, CXLDeviceState *cxl_dstate, uint16_t *len) { @@ -448,7 +453,7 @@ static CXLRetCode cmd_ccls_set_lsa(struct cxl_cmd *cmd, uint32_t rsvd; uint8_t data[]; } QEMU_PACKED; - struct set_lsa_pl *set_lsa_payload = (void *)cmd->payload; + struct set_lsa_pl *set_lsa_payload = (void *)payload; CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate); CXLType3Class *cvc = CXL_TYPE3_GET_CLASS(ct3d); const size_t hdr_len = offsetof(struct set_lsa_pl, data); @@ -474,7 +479,8 @@ static CXLRetCode cmd_ccls_set_lsa(struct cxl_cmd *cmd, * make this stateful. We may want to allow longer poison lists to aid * testing that kernel functionality. */ -static CXLRetCode cmd_media_get_poison_list(struct cxl_cmd *cmd, +static CXLRetCode cmd_media_get_poison_list(const struct cxl_cmd *cmd, + uint8_t *payload, CXLDeviceState *cxl_dstate, uint16_t *len) { @@ -496,8 +502,8 @@ static CXLRetCode cmd_media_get_poison_list(struct cxl_cmd *cmd, } QEMU_PACKED records[]; } QEMU_PACKED; - struct get_poison_list_pl *in = (void *)cmd->payload; - struct get_poison_list_out_pl *out = (void *)cmd->payload; + struct get_poison_list_pl *in = (void *)payload; + struct get_poison_list_out_pl *out = (void *)payload; CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate); uint16_t record_count = 0, i = 0; uint64_t query_start, query_length; @@ -550,7 +556,8 @@ static CXLRetCode cmd_media_get_poison_list(struct cxl_cmd *cmd, return CXL_MBOX_SUCCESS; } -static CXLRetCode cmd_media_inject_poison(struct cxl_cmd *cmd, +static CXLRetCode cmd_media_inject_poison(const struct cxl_cmd *cmd, + uint8_t *payload, CXLDeviceState *cxl_dstate, uint16_t *len_unused) { @@ -560,7 +567,7 @@ static CXLRetCode cmd_media_inject_poison(struct cxl_cmd *cmd, struct inject_poison_pl { uint64_t dpa; }; - struct inject_poison_pl *in = (void *)cmd->payload; + struct inject_poison_pl *in = (void *)payload; uint64_t dpa = ldq_le_p(&in->dpa); CXLPoison *p; @@ -589,7 +596,8 @@ static CXLRetCode cmd_media_inject_poison(struct cxl_cmd *cmd, return CXL_MBOX_SUCCESS; } -static CXLRetCode cmd_media_clear_poison(struct cxl_cmd *cmd, +static CXLRetCode cmd_media_clear_poison(const struct cxl_cmd *cmd, + uint8_t *payload, CXLDeviceState *cxl_dstate, uint16_t *len_unused) { @@ -603,7 +611,7 @@ static CXLRetCode cmd_media_clear_poison(struct cxl_cmd *cmd, CXLPoison *ent; uint64_t dpa; - struct clear_poison_pl *in = (void *)cmd->payload; + struct clear_poison_pl *in = (void *)payload; dpa = ldq_le_p(&in->dpa); if (dpa + CXL_CACHE_LINE_SIZE > cxl_dstate->mem_size) { @@ -673,7 +681,7 @@ static CXLRetCode cmd_media_clear_poison(struct cxl_cmd *cmd, #define IMMEDIATE_POLICY_CHANGE (1 << 3) #define IMMEDIATE_LOG_CHANGE (1 << 4) -static struct cxl_cmd cxl_cmd_set[256][256] = { +static const struct cxl_cmd cxl_cmd_set[256][256] = { [EVENTS][GET_RECORDS] = { "EVENTS_GET_RECORDS", cmd_events_get_records, 1, 0 }, [EVENTS][CLEAR_RECORDS] = { "EVENTS_CLEAR_RECORDS", @@ -709,21 +717,21 @@ static struct cxl_cmd cxl_cmd_set[256][256] = { void cxl_process_mailbox(CXLDeviceState *cxl_dstate) { uint16_t ret = CXL_MBOX_SUCCESS; - struct cxl_cmd *cxl_cmd; - uint64_t status_reg; + const struct cxl_cmd *cxl_cmd; + uint64_t status_reg = 0; opcode_handler h; uint64_t command_reg = cxl_dstate->mbox_reg_state64[R_CXL_DEV_MAILBOX_CMD]; uint8_t set = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, COMMAND_SET); uint8_t cmd = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, COMMAND); uint16_t len = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, LENGTH); - cxl_cmd = &cxl_cmd_set[set][cmd]; + uint8_t *pl = cxl_dstate->mbox_reg_state + A_CXL_DEV_CMD_PAYLOAD; + + cxl_cmd = &cxl_dstate->cxl_cmd_set[set][cmd]; h = cxl_cmd->handler; if (h) { if (len == cxl_cmd->in || cxl_cmd->in == ~0) { - cxl_cmd->payload = cxl_dstate->mbox_reg_state + - A_CXL_DEV_CMD_PAYLOAD; - ret = (*h)(cxl_cmd, cxl_dstate, &len); + ret = (*h)(cxl_cmd, pl, cxl_dstate, &len); assert(len <= cxl_dstate->payload_size); } else { ret = CXL_MBOX_INVALID_PAYLOAD_LENGTH; @@ -752,10 +760,11 @@ void cxl_process_mailbox(CXLDeviceState *cxl_dstate) void cxl_initialize_mailbox(CXLDeviceState *cxl_dstate) { + cxl_dstate->cxl_cmd_set = cxl_cmd_set; for (int set = 0; set < 256; set++) { for (int cmd = 0; cmd < 256; cmd++) { - if (cxl_cmd_set[set][cmd].handler) { - struct cxl_cmd *c = &cxl_cmd_set[set][cmd]; + if (cxl_dstate->cxl_cmd_set[set][cmd].handler) { + const struct cxl_cmd *c = &cxl_dstate->cxl_cmd_set[set][cmd]; struct cel_log *log = &cxl_dstate->cel_log[cxl_dstate->cel_size]; diff --git a/include/hw/cxl/cxl_device.h b/include/hw/cxl/cxl_device.h index 007ddaf078..556953469c 100644 --- a/include/hw/cxl/cxl_device.h +++ b/include/hw/cxl/cxl_device.h @@ -111,6 +111,18 @@ typedef enum { CXL_MBOX_MAX = 0x17 } CXLRetCode; +typedef struct cxl_device_state CXLDeviceState; +struct cxl_cmd; +typedef CXLRetCode (*opcode_handler)(const struct cxl_cmd *cmd, + uint8_t *payload, + CXLDeviceState *cxl_dstate, uint16_t *len); +struct cxl_cmd { + const char *name; + opcode_handler handler; + ssize_t in; + uint16_t effect; /* Reported in CEL */ +}; + typedef struct CXLEvent { CXLEventRecordRaw data; QSIMPLEQ_ENTRY(CXLEvent) node; @@ -178,6 +190,7 @@ typedef struct cxl_device_state { uint64_t pmem_size; uint64_t vmem_size; + const struct cxl_cmd (*cxl_cmd_set)[256]; CXLEventLog event_logs[CXL_EVENT_TYPE_MAX]; } CXLDeviceState; -- cgit v1.2.3 From 6f59274e937576fbb2623b687aa2556e115a712f Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 17:07:51 +0100 Subject: hw/cxl/mbox: Split mailbox command payload into separate input and output New CCI types that will be supported shortly do not have a single buffer used in both directions. As such, split it up. To avoid the complexities of implementing all commands to handle potential aliasing, take a copy of the input before use. Signed-off-by: Jonathan Cameron Message-Id: <20231023160806.13206-3-Jonathan.Cameron@huawei.com> Reviewed-by: Fan Ni Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-events.c | 2 +- hw/cxl/cxl-mailbox-utils.c | 230 ++++++++++++++++++++++++++------------------ include/hw/cxl/cxl_device.h | 7 +- 3 files changed, 140 insertions(+), 99 deletions(-) diff --git a/hw/cxl/cxl-events.c b/hw/cxl/cxl-events.c index e2172b94b9..bee6dfaf14 100644 --- a/hw/cxl/cxl-events.c +++ b/hw/cxl/cxl-events.c @@ -143,7 +143,7 @@ bool cxl_event_insert(CXLDeviceState *cxlds, CXLEventLogType log_type, CXLRetCode cxl_event_get_records(CXLDeviceState *cxlds, CXLGetEventPayload *pl, uint8_t log_type, int max_recs, - uint16_t *len) + size_t *len) { CXLEventLog *log; CXLEvent *entry; diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c index c02de06943..e5ddce37c7 100644 --- a/hw/cxl/cxl-mailbox-utils.c +++ b/hw/cxl/cxl-mailbox-utils.c @@ -71,9 +71,9 @@ enum { static CXLRetCode cmd_events_get_records(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxlds, - uint16_t *len) + uint8_t *payload_in, size_t len_in, + uint8_t *payload_out, size_t *len_out, + CXLDeviceState *cxlds) { CXLGetEventPayload *pl; uint8_t log_type; @@ -83,9 +83,9 @@ static CXLRetCode cmd_events_get_records(const struct cxl_cmd *cmd, return CXL_MBOX_INVALID_INPUT; } - log_type = payload[0]; + log_type = payload_in[0]; - pl = (CXLGetEventPayload *)payload; + pl = (CXLGetEventPayload *)payload_out; memset(pl, 0, sizeof(*pl)); max_recs = (cxlds->payload_size - CXL_EVENT_PAYLOAD_HDR_SIZE) / @@ -94,30 +94,34 @@ static CXLRetCode cmd_events_get_records(const struct cxl_cmd *cmd, max_recs = 0xFFFF; } - return cxl_event_get_records(cxlds, pl, log_type, max_recs, len); + return cxl_event_get_records(cxlds, pl, log_type, max_recs, len_out); } static CXLRetCode cmd_events_clear_records(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxlds, - uint16_t *len) + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLDeviceState *cxlds) { CXLClearEventPayload *pl; - pl = (CXLClearEventPayload *)payload; - *len = 0; + pl = (CXLClearEventPayload *)payload_in; + *len_out = 0; return cxl_event_clear_records(cxlds, pl); } static CXLRetCode cmd_events_get_interrupt_policy(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxlds, - uint16_t *len) + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLDeviceState *cxlds) { CXLEventInterruptPolicy *policy; CXLEventLog *log; - policy = (CXLEventInterruptPolicy *)payload; + policy = (CXLEventInterruptPolicy *)payload_out; memset(policy, 0, sizeof(*policy)); log = &cxlds->event_logs[CXL_EVENT_TYPE_INFO]; @@ -146,23 +150,25 @@ static CXLRetCode cmd_events_get_interrupt_policy(const struct cxl_cmd *cmd, policy->dyn_cap_settings = CXL_INT_MSI_MSIX; } - *len = sizeof(*policy); + *len_out = sizeof(*policy); return CXL_MBOX_SUCCESS; } static CXLRetCode cmd_events_set_interrupt_policy(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxlds, - uint16_t *len) + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLDeviceState *cxlds) { CXLEventInterruptPolicy *policy; CXLEventLog *log; - if (*len < CXL_EVENT_INT_SETTING_MIN_LEN) { + if (len_in < CXL_EVENT_INT_SETTING_MIN_LEN) { return CXL_MBOX_INVALID_PAYLOAD_LENGTH; } - policy = (CXLEventInterruptPolicy *)payload; + policy = (CXLEventInterruptPolicy *)payload_in; log = &cxlds->event_logs[CXL_EVENT_TYPE_INFO]; log->irq_enabled = (policy->info_settings & CXL_EVENT_INT_MODE_MASK) == @@ -181,7 +187,7 @@ static CXLRetCode cmd_events_set_interrupt_policy(const struct cxl_cmd *cmd, CXL_INT_MSI_MSIX; /* DCD is optional */ - if (*len < sizeof(*policy)) { + if (len_in < sizeof(*policy)) { return CXL_MBOX_SUCCESS; } @@ -189,15 +195,17 @@ static CXLRetCode cmd_events_set_interrupt_policy(const struct cxl_cmd *cmd, log->irq_enabled = (policy->dyn_cap_settings & CXL_EVENT_INT_MODE_MASK) == CXL_INT_MSI_MSIX; - *len = sizeof(*policy); + *len_out = 0; return CXL_MBOX_SUCCESS; } /* 8.2.9.2.1 */ static CXLRetCode cmd_firmware_update_get_info(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxl_dstate, - uint16_t *len) + uint8_t *payload_in, + size_t len, + uint8_t *payload_out, + size_t *len_out, + CXLDeviceState *cxl_dstate) { struct { uint8_t slots_supported; @@ -216,7 +224,7 @@ static CXLRetCode cmd_firmware_update_get_info(const struct cxl_cmd *cmd, return CXL_MBOX_INTERNAL_ERROR; } - fw_info = (void *)payload; + fw_info = (void *)payload_out; memset(fw_info, 0, sizeof(*fw_info)); fw_info->slots_supported = 2; @@ -224,36 +232,40 @@ static CXLRetCode cmd_firmware_update_get_info(const struct cxl_cmd *cmd, fw_info->caps = 0; pstrcpy(fw_info->fw_rev1, sizeof(fw_info->fw_rev1), "BWFW VERSION 0"); - *len = sizeof(*fw_info); + *len_out = sizeof(*fw_info); return CXL_MBOX_SUCCESS; } /* 8.2.9.3.1 */ static CXLRetCode cmd_timestamp_get(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxl_dstate, - uint16_t *len) + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLDeviceState *cxl_dstate) { uint64_t final_time = cxl_device_get_timestamp(cxl_dstate); - stq_le_p(payload, final_time); - *len = 8; + stq_le_p(payload_out, final_time); + *len_out = 8; return CXL_MBOX_SUCCESS; } /* 8.2.9.3.2 */ static CXLRetCode cmd_timestamp_set(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxl_dstate, - uint16_t *len) + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLDeviceState *cxl_dstate) { cxl_dstate->timestamp.set = true; cxl_dstate->timestamp.last_set = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL); - cxl_dstate->timestamp.host_set = le64_to_cpu(*(uint64_t *)payload); + cxl_dstate->timestamp.host_set = le64_to_cpu(*(uint64_t *)payload_in); - *len = 0; + *len_out = 0; return CXL_MBOX_SUCCESS; } @@ -265,9 +277,11 @@ static const QemuUUID cel_uuid = { /* 8.2.9.4.1 */ static CXLRetCode cmd_logs_get_supported(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxl_dstate, - uint16_t *len) + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLDeviceState *cxl_dstate) { struct { uint16_t entries; @@ -276,22 +290,24 @@ static CXLRetCode cmd_logs_get_supported(const struct cxl_cmd *cmd, QemuUUID uuid; uint32_t size; } log_entries[1]; - } QEMU_PACKED *supported_logs = (void *)payload; + } QEMU_PACKED *supported_logs = (void *)payload_out; QEMU_BUILD_BUG_ON(sizeof(*supported_logs) != 0x1c); supported_logs->entries = 1; supported_logs->log_entries[0].uuid = cel_uuid; supported_logs->log_entries[0].size = 4 * cxl_dstate->cel_size; - *len = sizeof(*supported_logs); + *len_out = sizeof(*supported_logs); return CXL_MBOX_SUCCESS; } /* 8.2.9.4.2 */ static CXLRetCode cmd_logs_get_log(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxl_dstate, - uint16_t *len) + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLDeviceState *cxl_dstate) { struct { QemuUUID uuid; @@ -299,7 +315,7 @@ static CXLRetCode cmd_logs_get_log(const struct cxl_cmd *cmd, uint32_t length; } QEMU_PACKED QEMU_ALIGNED(16) *get_log; - get_log = (void *)payload; + get_log = (void *)payload_in; /* * 8.2.9.4.2 @@ -323,19 +339,21 @@ static CXLRetCode cmd_logs_get_log(const struct cxl_cmd *cmd, } /* Store off everything to local variables so we can wipe out the payload */ - *len = get_log->length; + *len_out = get_log->length; - memmove(payload, cxl_dstate->cel_log + get_log->offset, - get_log->length); + memmove(payload_out, cxl_dstate->cel_log + get_log->offset, + get_log->length); return CXL_MBOX_SUCCESS; } /* 8.2.9.5.1.1 */ static CXLRetCode cmd_identify_memory_device(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxl_dstate, - uint16_t *len) + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLDeviceState *cxl_dstate) { struct { char fw_revision[0x10]; @@ -363,7 +381,7 @@ static CXLRetCode cmd_identify_memory_device(const struct cxl_cmd *cmd, return CXL_MBOX_INTERNAL_ERROR; } - id = (void *)payload; + id = (void *)payload_out; memset(id, 0, sizeof(*id)); snprintf(id->fw_revision, 0x10, "BWFW VERSION %02d", 0); @@ -380,21 +398,23 @@ static CXLRetCode cmd_identify_memory_device(const struct cxl_cmd *cmd, /* No limit - so limited by main poison record limit */ stw_le_p(&id->inject_poison_limit, 0); - *len = sizeof(*id); + *len_out = sizeof(*id); return CXL_MBOX_SUCCESS; } static CXLRetCode cmd_ccls_get_partition_info(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxl_dstate, - uint16_t *len) + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLDeviceState *cxl_dstate) { struct { uint64_t active_vmem; uint64_t active_pmem; uint64_t next_vmem; uint64_t next_pmem; - } QEMU_PACKED *part_info = (void *)payload; + } QEMU_PACKED *part_info = (void *)payload_out; QEMU_BUILD_BUG_ON(sizeof(*part_info) != 0x20); if ((!QEMU_IS_ALIGNED(cxl_dstate->vmem_size, CXL_CAPACITY_MULTIPLIER)) || @@ -413,14 +433,16 @@ static CXLRetCode cmd_ccls_get_partition_info(const struct cxl_cmd *cmd, cxl_dstate->pmem_size / CXL_CAPACITY_MULTIPLIER); stq_le_p(&part_info->next_pmem, 0); - *len = sizeof(*part_info); + *len_out = sizeof(*part_info); return CXL_MBOX_SUCCESS; } static CXLRetCode cmd_ccls_get_lsa(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxl_dstate, - uint16_t *len) + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLDeviceState *cxl_dstate) { struct { uint32_t offset; @@ -430,46 +452,47 @@ static CXLRetCode cmd_ccls_get_lsa(const struct cxl_cmd *cmd, CXLType3Class *cvc = CXL_TYPE3_GET_CLASS(ct3d); uint32_t offset, length; - get_lsa = (void *)payload; + get_lsa = (void *)payload_in; offset = get_lsa->offset; length = get_lsa->length; if (offset + length > cvc->get_lsa_size(ct3d)) { - *len = 0; + *len_out = 0; return CXL_MBOX_INVALID_INPUT; } - *len = cvc->get_lsa(ct3d, get_lsa, length, offset); + *len_out = cvc->get_lsa(ct3d, payload_out, length, offset); return CXL_MBOX_SUCCESS; } static CXLRetCode cmd_ccls_set_lsa(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxl_dstate, - uint16_t *len) + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLDeviceState *cxl_dstate) { struct set_lsa_pl { uint32_t offset; uint32_t rsvd; uint8_t data[]; } QEMU_PACKED; - struct set_lsa_pl *set_lsa_payload = (void *)payload; + struct set_lsa_pl *set_lsa_payload = (void *)payload_in; CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate); CXLType3Class *cvc = CXL_TYPE3_GET_CLASS(ct3d); const size_t hdr_len = offsetof(struct set_lsa_pl, data); - uint16_t plen = *len; - *len = 0; - if (!plen) { + *len_out = 0; + if (!len_in) { return CXL_MBOX_SUCCESS; } - if (set_lsa_payload->offset + plen > cvc->get_lsa_size(ct3d) + hdr_len) { + if (set_lsa_payload->offset + len_in > cvc->get_lsa_size(ct3d) + hdr_len) { return CXL_MBOX_INVALID_INPUT; } - plen -= hdr_len; + len_in -= hdr_len; - cvc->set_lsa(ct3d, set_lsa_payload->data, plen, set_lsa_payload->offset); + cvc->set_lsa(ct3d, set_lsa_payload->data, len_in, set_lsa_payload->offset); return CXL_MBOX_SUCCESS; } @@ -480,9 +503,11 @@ static CXLRetCode cmd_ccls_set_lsa(const struct cxl_cmd *cmd, * testing that kernel functionality. */ static CXLRetCode cmd_media_get_poison_list(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxl_dstate, - uint16_t *len) + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLDeviceState *cxl_dstate) { struct get_poison_list_pl { uint64_t pa; @@ -502,8 +527,8 @@ static CXLRetCode cmd_media_get_poison_list(const struct cxl_cmd *cmd, } QEMU_PACKED records[]; } QEMU_PACKED; - struct get_poison_list_pl *in = (void *)payload; - struct get_poison_list_out_pl *out = (void *)payload; + struct get_poison_list_pl *in = (void *)payload_in; + struct get_poison_list_out_pl *out = (void *)payload_out; CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate); uint16_t record_count = 0, i = 0; uint64_t query_start, query_length; @@ -552,14 +577,16 @@ static CXLRetCode cmd_media_get_poison_list(const struct cxl_cmd *cmd, stq_le_p(&out->overflow_timestamp, ct3d->poison_list_overflow_ts); } stw_le_p(&out->count, record_count); - *len = out_pl_len; + *len_out = out_pl_len; return CXL_MBOX_SUCCESS; } static CXLRetCode cmd_media_inject_poison(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxl_dstate, - uint16_t *len_unused) + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLDeviceState *cxl_dstate) { CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate); CXLPoisonList *poison_list = &ct3d->poison_list; @@ -567,7 +594,7 @@ static CXLRetCode cmd_media_inject_poison(const struct cxl_cmd *cmd, struct inject_poison_pl { uint64_t dpa; }; - struct inject_poison_pl *in = (void *)payload; + struct inject_poison_pl *in = (void *)payload_in; uint64_t dpa = ldq_le_p(&in->dpa); CXLPoison *p; @@ -592,14 +619,17 @@ static CXLRetCode cmd_media_inject_poison(const struct cxl_cmd *cmd, */ QLIST_INSERT_HEAD(poison_list, p, node); ct3d->poison_list_cnt++; + *len_out = 0; return CXL_MBOX_SUCCESS; } static CXLRetCode cmd_media_clear_poison(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxl_dstate, - uint16_t *len_unused) + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLDeviceState *cxl_dstate) { CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate); CXLPoisonList *poison_list = &ct3d->poison_list; @@ -611,7 +641,7 @@ static CXLRetCode cmd_media_clear_poison(const struct cxl_cmd *cmd, CXLPoison *ent; uint64_t dpa; - struct clear_poison_pl *in = (void *)payload; + struct clear_poison_pl *in = (void *)payload_in; dpa = ldq_le_p(&in->dpa); if (dpa + CXL_CACHE_LINE_SIZE > cxl_dstate->mem_size) { @@ -672,6 +702,7 @@ static CXLRetCode cmd_media_clear_poison(const struct cxl_cmd *cmd, } /* Any fragments have been added, free original entry */ g_free(ent); + *len_out = 0; return CXL_MBOX_SUCCESS; } @@ -724,15 +755,24 @@ void cxl_process_mailbox(CXLDeviceState *cxl_dstate) uint8_t set = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, COMMAND_SET); uint8_t cmd = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, COMMAND); - uint16_t len = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, LENGTH); + uint16_t len_in = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, LENGTH); uint8_t *pl = cxl_dstate->mbox_reg_state + A_CXL_DEV_CMD_PAYLOAD; + /* + * Copy taken to avoid need for individual command handlers to care + * about aliasing. + */ + g_autofree uint8_t *pl_in_copy = NULL; + size_t len_out = 0; + pl_in_copy = g_memdup2(pl, len_in); + /* Avoid stale data - including from earlier commands */ + memset(pl, 0, CXL_MAILBOX_MAX_PAYLOAD_SIZE); cxl_cmd = &cxl_dstate->cxl_cmd_set[set][cmd]; h = cxl_cmd->handler; if (h) { - if (len == cxl_cmd->in || cxl_cmd->in == ~0) { - ret = (*h)(cxl_cmd, pl, cxl_dstate, &len); - assert(len <= cxl_dstate->payload_size); + if (len_in == cxl_cmd->in || cxl_cmd->in == ~0) { + ret = (*h)(cxl_cmd, pl_in_copy, len_in, pl, &len_out, cxl_dstate); + assert(len_out <= cxl_dstate->payload_size); } else { ret = CXL_MBOX_INVALID_PAYLOAD_LENGTH; } @@ -748,7 +788,7 @@ void cxl_process_mailbox(CXLDeviceState *cxl_dstate) /* Set the return length */ command_reg = FIELD_DP64(command_reg, CXL_DEV_MAILBOX_CMD, COMMAND_SET, 0); command_reg = FIELD_DP64(command_reg, CXL_DEV_MAILBOX_CMD, COMMAND, 0); - command_reg = FIELD_DP64(command_reg, CXL_DEV_MAILBOX_CMD, LENGTH, len); + command_reg = FIELD_DP64(command_reg, CXL_DEV_MAILBOX_CMD, LENGTH, len_out); cxl_dstate->mbox_reg_state64[R_CXL_DEV_MAILBOX_CMD] = command_reg; cxl_dstate->mbox_reg_state64[R_CXL_DEV_MAILBOX_STS] = status_reg; diff --git a/include/hw/cxl/cxl_device.h b/include/hw/cxl/cxl_device.h index 556953469c..d7a2c4009e 100644 --- a/include/hw/cxl/cxl_device.h +++ b/include/hw/cxl/cxl_device.h @@ -114,8 +114,9 @@ typedef enum { typedef struct cxl_device_state CXLDeviceState; struct cxl_cmd; typedef CXLRetCode (*opcode_handler)(const struct cxl_cmd *cmd, - uint8_t *payload, - CXLDeviceState *cxl_dstate, uint16_t *len); + uint8_t *payload_in, size_t len_in, + uint8_t *payload_out, size_t *len_out, + CXLDeviceState *cxl_dstate); struct cxl_cmd { const char *name; opcode_handler handler; @@ -390,7 +391,7 @@ bool cxl_event_insert(CXLDeviceState *cxlds, CXLEventLogType log_type, CXLEventRecordRaw *event); CXLRetCode cxl_event_get_records(CXLDeviceState *cxlds, CXLGetEventPayload *pl, uint8_t log_type, int max_recs, - uint16_t *len); + size_t *len); CXLRetCode cxl_event_clear_records(CXLDeviceState *cxlds, CXLClearEventPayload *pl); -- cgit v1.2.3 From cac36a8faffc62ba6b07d8e9dfdc9fbf15c7d1bf Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 17:07:52 +0100 Subject: hw/cxl/mbox: Pull the CCI definition out of the CXLDeviceState Enables having multiple CCIs per devices. Each CCI (mailbox) has it's own state and command list, so they can't share a single structure. Signed-off-by: Jonathan Cameron Message-Id: <20231023160806.13206-4-Jonathan.Cameron@huawei.com> Reviewed-by: Fan Ni Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-device-utils.c | 31 +++++++++++---- hw/cxl/cxl-mailbox-utils.c | 94 +++++++++++++++++++++++++++------------------ hw/mem/cxl_type3.c | 5 ++- include/hw/cxl/cxl_device.h | 35 ++++++++++++----- 4 files changed, 109 insertions(+), 56 deletions(-) diff --git a/hw/cxl/cxl-device-utils.c b/hw/cxl/cxl-device-utils.c index eb7195272e..327949a805 100644 --- a/hw/cxl/cxl-device-utils.c +++ b/hw/cxl/cxl-device-utils.c @@ -62,7 +62,14 @@ static uint64_t dev_reg_read(void *opaque, hwaddr offset, unsigned size) static uint64_t mailbox_reg_read(void *opaque, hwaddr offset, unsigned size) { - CXLDeviceState *cxl_dstate = opaque; + CXLDeviceState *cxl_dstate; + CXLCCI *cci = opaque; + + if (object_dynamic_cast(OBJECT(cci->intf), TYPE_CXL_TYPE3)) { + cxl_dstate = &CXL_TYPE3(cci->intf)->cxl_dstate; + } else { + return 0; + } switch (size) { case 1: @@ -123,7 +130,14 @@ static void mailbox_mem_writeq(uint64_t *reg_state, hwaddr offset, static void mailbox_reg_write(void *opaque, hwaddr offset, uint64_t value, unsigned size) { - CXLDeviceState *cxl_dstate = opaque; + CXLDeviceState *cxl_dstate; + CXLCCI *cci = opaque; + + if (object_dynamic_cast(OBJECT(cci->intf), TYPE_CXL_TYPE3)) { + cxl_dstate = &CXL_TYPE3(cci->intf)->cxl_dstate; + } else { + return; + } if (offset >= A_CXL_DEV_CMD_PAYLOAD) { memcpy(cxl_dstate->mbox_reg_state + offset, &value, size); @@ -143,7 +157,7 @@ static void mailbox_reg_write(void *opaque, hwaddr offset, uint64_t value, if (ARRAY_FIELD_EX32(cxl_dstate->mbox_reg_state32, CXL_DEV_MAILBOX_CTRL, DOORBELL)) { - cxl_process_mailbox(cxl_dstate); + cxl_process_mailbox(cci); } } @@ -223,7 +237,8 @@ static const MemoryRegionOps caps_ops = { }, }; -void cxl_device_register_block_init(Object *obj, CXLDeviceState *cxl_dstate) +void cxl_device_register_block_init(Object *obj, CXLDeviceState *cxl_dstate, + CXLCCI *cci) { /* This will be a BAR, so needs to be rounded up to pow2 for PCI spec */ memory_region_init(&cxl_dstate->device_registers, obj, "device-registers", @@ -233,7 +248,7 @@ void cxl_device_register_block_init(Object *obj, CXLDeviceState *cxl_dstate) "cap-array", CXL_CAPS_SIZE); memory_region_init_io(&cxl_dstate->device, obj, &dev_ops, cxl_dstate, "device-status", CXL_DEVICE_STATUS_REGISTERS_LENGTH); - memory_region_init_io(&cxl_dstate->mailbox, obj, &mailbox_ops, cxl_dstate, + memory_region_init_io(&cxl_dstate->mailbox, obj, &mailbox_ops, cci, "mailbox", CXL_MAILBOX_REGISTERS_LENGTH); memory_region_init_io(&cxl_dstate->memory_device, obj, &mdev_ops, cxl_dstate, "memory device caps", @@ -284,8 +299,9 @@ static void mailbox_reg_init_common(CXLDeviceState *cxl_dstate) static void memdev_reg_init_common(CXLDeviceState *cxl_dstate) { } -void cxl_device_register_init_common(CXLDeviceState *cxl_dstate) +void cxl_device_register_init_t3(CXLType3Dev *ct3d) { + CXLDeviceState *cxl_dstate = &ct3d->cxl_dstate; uint64_t *cap_h = cxl_dstate->caps_reg_state64; const int cap_count = 3; @@ -303,7 +319,8 @@ void cxl_device_register_init_common(CXLDeviceState *cxl_dstate) cxl_device_cap_init(cxl_dstate, MEMORY_DEVICE, 0x4000, 1); memdev_reg_init_common(cxl_dstate); - cxl_initialize_mailbox(cxl_dstate); + cxl_initialize_mailbox_t3(&ct3d->cci, DEVICE(ct3d), + CXL_MAILBOX_MAX_PAYLOAD_SIZE); } uint64_t cxl_device_get_timestamp(CXLDeviceState *cxl_dstate) diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c index e5ddce37c7..5484dfbbf1 100644 --- a/hw/cxl/cxl-mailbox-utils.c +++ b/hw/cxl/cxl-mailbox-utils.c @@ -73,8 +73,9 @@ enum { static CXLRetCode cmd_events_get_records(const struct cxl_cmd *cmd, uint8_t *payload_in, size_t len_in, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxlds) + CXLCCI *cci) { + CXLDeviceState *cxlds = &CXL_TYPE3(cci->d)->cxl_dstate; CXLGetEventPayload *pl; uint8_t log_type; int max_recs; @@ -102,8 +103,9 @@ static CXLRetCode cmd_events_clear_records(const struct cxl_cmd *cmd, size_t len_in, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxlds) + CXLCCI *cci) { + CXLDeviceState *cxlds = &CXL_TYPE3(cci->d)->cxl_dstate; CXLClearEventPayload *pl; pl = (CXLClearEventPayload *)payload_in; @@ -116,8 +118,9 @@ static CXLRetCode cmd_events_get_interrupt_policy(const struct cxl_cmd *cmd, size_t len_in, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxlds) + CXLCCI *cci) { + CXLDeviceState *cxlds = &CXL_TYPE3(cci->d)->cxl_dstate; CXLEventInterruptPolicy *policy; CXLEventLog *log; @@ -159,8 +162,9 @@ static CXLRetCode cmd_events_set_interrupt_policy(const struct cxl_cmd *cmd, size_t len_in, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxlds) + CXLCCI *cci) { + CXLDeviceState *cxlds = &CXL_TYPE3(cci->d)->cxl_dstate; CXLEventInterruptPolicy *policy; CXLEventLog *log; @@ -205,8 +209,9 @@ static CXLRetCode cmd_firmware_update_get_info(const struct cxl_cmd *cmd, size_t len, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxl_dstate) + CXLCCI *cci) { + CXLDeviceState *cxl_dstate = &CXL_TYPE3(cci->d)->cxl_dstate; struct { uint8_t slots_supported; uint8_t slot_info; @@ -242,8 +247,9 @@ static CXLRetCode cmd_timestamp_get(const struct cxl_cmd *cmd, size_t len_in, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxl_dstate) + CXLCCI *cci) { + CXLDeviceState *cxl_dstate = &CXL_TYPE3(cci->d)->cxl_dstate; uint64_t final_time = cxl_device_get_timestamp(cxl_dstate); stq_le_p(payload_out, final_time); @@ -258,8 +264,10 @@ static CXLRetCode cmd_timestamp_set(const struct cxl_cmd *cmd, size_t len_in, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxl_dstate) + CXLCCI *cci) { + CXLDeviceState *cxl_dstate = &CXL_TYPE3(cci->d)->cxl_dstate; + cxl_dstate->timestamp.set = true; cxl_dstate->timestamp.last_set = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL); @@ -281,7 +289,7 @@ static CXLRetCode cmd_logs_get_supported(const struct cxl_cmd *cmd, size_t len_in, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxl_dstate) + CXLCCI *cci) { struct { uint16_t entries; @@ -295,7 +303,7 @@ static CXLRetCode cmd_logs_get_supported(const struct cxl_cmd *cmd, supported_logs->entries = 1; supported_logs->log_entries[0].uuid = cel_uuid; - supported_logs->log_entries[0].size = 4 * cxl_dstate->cel_size; + supported_logs->log_entries[0].size = 4 * cci->cel_size; *len_out = sizeof(*supported_logs); return CXL_MBOX_SUCCESS; @@ -307,7 +315,7 @@ static CXLRetCode cmd_logs_get_log(const struct cxl_cmd *cmd, size_t len_in, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxl_dstate) + CXLCCI *cci) { struct { QemuUUID uuid; @@ -330,7 +338,7 @@ static CXLRetCode cmd_logs_get_log(const struct cxl_cmd *cmd, * the only possible failure would be if the mailbox itself isn't big * enough. */ - if (get_log->offset + get_log->length > cxl_dstate->payload_size) { + if (get_log->offset + get_log->length > cci->payload_max) { return CXL_MBOX_INVALID_INPUT; } @@ -341,8 +349,7 @@ static CXLRetCode cmd_logs_get_log(const struct cxl_cmd *cmd, /* Store off everything to local variables so we can wipe out the payload */ *len_out = get_log->length; - memmove(payload_out, cxl_dstate->cel_log + get_log->offset, - get_log->length); + memmove(payload_out, cci->cel_log + get_log->offset, get_log->length); return CXL_MBOX_SUCCESS; } @@ -353,7 +360,7 @@ static CXLRetCode cmd_identify_memory_device(const struct cxl_cmd *cmd, size_t len_in, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxl_dstate) + CXLCCI *cci) { struct { char fw_revision[0x10]; @@ -372,9 +379,9 @@ static CXLRetCode cmd_identify_memory_device(const struct cxl_cmd *cmd, uint8_t qos_telemetry_caps; } QEMU_PACKED *id; QEMU_BUILD_BUG_ON(sizeof(*id) != 0x43); - - CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate); + CXLType3Dev *ct3d = CXL_TYPE3(cci->d); CXLType3Class *cvc = CXL_TYPE3_GET_CLASS(ct3d); + CXLDeviceState *cxl_dstate = &ct3d->cxl_dstate; if ((!QEMU_IS_ALIGNED(cxl_dstate->vmem_size, CXL_CAPACITY_MULTIPLIER)) || (!QEMU_IS_ALIGNED(cxl_dstate->pmem_size, CXL_CAPACITY_MULTIPLIER))) { @@ -407,8 +414,9 @@ static CXLRetCode cmd_ccls_get_partition_info(const struct cxl_cmd *cmd, size_t len_in, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxl_dstate) + CXLCCI *cci) { + CXLDeviceState *cxl_dstate = &CXL_TYPE3(cci->d)->cxl_dstate; struct { uint64_t active_vmem; uint64_t active_pmem; @@ -442,13 +450,13 @@ static CXLRetCode cmd_ccls_get_lsa(const struct cxl_cmd *cmd, size_t len_in, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxl_dstate) + CXLCCI *cci) { struct { uint32_t offset; uint32_t length; } QEMU_PACKED *get_lsa; - CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate); + CXLType3Dev *ct3d = CXL_TYPE3(cci->d); CXLType3Class *cvc = CXL_TYPE3_GET_CLASS(ct3d); uint32_t offset, length; @@ -470,7 +478,7 @@ static CXLRetCode cmd_ccls_set_lsa(const struct cxl_cmd *cmd, size_t len_in, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxl_dstate) + CXLCCI *cci) { struct set_lsa_pl { uint32_t offset; @@ -478,7 +486,7 @@ static CXLRetCode cmd_ccls_set_lsa(const struct cxl_cmd *cmd, uint8_t data[]; } QEMU_PACKED; struct set_lsa_pl *set_lsa_payload = (void *)payload_in; - CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate); + CXLType3Dev *ct3d = CXL_TYPE3(cci->d); CXLType3Class *cvc = CXL_TYPE3_GET_CLASS(ct3d); const size_t hdr_len = offsetof(struct set_lsa_pl, data); @@ -507,7 +515,7 @@ static CXLRetCode cmd_media_get_poison_list(const struct cxl_cmd *cmd, size_t len_in, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxl_dstate) + CXLCCI *cci) { struct get_poison_list_pl { uint64_t pa; @@ -529,7 +537,7 @@ static CXLRetCode cmd_media_get_poison_list(const struct cxl_cmd *cmd, struct get_poison_list_pl *in = (void *)payload_in; struct get_poison_list_out_pl *out = (void *)payload_out; - CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate); + CXLType3Dev *ct3d = CXL_TYPE3(cci->d); uint16_t record_count = 0, i = 0; uint64_t query_start, query_length; CXLPoisonList *poison_list = &ct3d->poison_list; @@ -586,9 +594,9 @@ static CXLRetCode cmd_media_inject_poison(const struct cxl_cmd *cmd, size_t len_in, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxl_dstate) + CXLCCI *cci) { - CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate); + CXLType3Dev *ct3d = CXL_TYPE3(cci->d); CXLPoisonList *poison_list = &ct3d->poison_list; CXLPoison *ent; struct inject_poison_pl { @@ -629,9 +637,10 @@ static CXLRetCode cmd_media_clear_poison(const struct cxl_cmd *cmd, size_t len_in, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxl_dstate) + CXLCCI *cci) { - CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate); + CXLType3Dev *ct3d = CXL_TYPE3(cci->d); + CXLDeviceState *cxl_dstate = &ct3d->cxl_dstate; CXLPoisonList *poison_list = &ct3d->poison_list; CXLType3Class *cvc = CXL_TYPE3_GET_CLASS(ct3d); struct clear_poison_pl { @@ -745,12 +754,13 @@ static const struct cxl_cmd cxl_cmd_set[256][256] = { cmd_media_clear_poison, 72, 0 }, }; -void cxl_process_mailbox(CXLDeviceState *cxl_dstate) +void cxl_process_mailbox(CXLCCI *cci) { uint16_t ret = CXL_MBOX_SUCCESS; const struct cxl_cmd *cxl_cmd; uint64_t status_reg = 0; opcode_handler h; + CXLDeviceState *cxl_dstate = &CXL_TYPE3(cci->d)->cxl_dstate; uint64_t command_reg = cxl_dstate->mbox_reg_state64[R_CXL_DEV_MAILBOX_CMD]; uint8_t set = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, COMMAND_SET); @@ -767,12 +777,12 @@ void cxl_process_mailbox(CXLDeviceState *cxl_dstate) pl_in_copy = g_memdup2(pl, len_in); /* Avoid stale data - including from earlier commands */ memset(pl, 0, CXL_MAILBOX_MAX_PAYLOAD_SIZE); - cxl_cmd = &cxl_dstate->cxl_cmd_set[set][cmd]; + cxl_cmd = &cci->cxl_cmd_set[set][cmd]; h = cxl_cmd->handler; if (h) { if (len_in == cxl_cmd->in || cxl_cmd->in == ~0) { - ret = (*h)(cxl_cmd, pl_in_copy, len_in, pl, &len_out, cxl_dstate); - assert(len_out <= cxl_dstate->payload_size); + ret = (*h)(cxl_cmd, pl, len_in, pl, &len_out, cci); + assert(len_out <= cci->payload_max); } else { ret = CXL_MBOX_INVALID_PAYLOAD_LENGTH; } @@ -798,20 +808,30 @@ void cxl_process_mailbox(CXLDeviceState *cxl_dstate) DOORBELL, 0); } -void cxl_initialize_mailbox(CXLDeviceState *cxl_dstate) +void cxl_init_cci(CXLCCI *cci, size_t payload_max) { - cxl_dstate->cxl_cmd_set = cxl_cmd_set; + cci->payload_max = payload_max; for (int set = 0; set < 256; set++) { for (int cmd = 0; cmd < 256; cmd++) { - if (cxl_dstate->cxl_cmd_set[set][cmd].handler) { - const struct cxl_cmd *c = &cxl_dstate->cxl_cmd_set[set][cmd]; + if (cci->cxl_cmd_set[set][cmd].handler) { + const struct cxl_cmd *c = &cci->cxl_cmd_set[set][cmd]; struct cel_log *log = - &cxl_dstate->cel_log[cxl_dstate->cel_size]; + &cci->cel_log[cci->cel_size]; log->opcode = (set << 8) | cmd; log->effect = c->effect; - cxl_dstate->cel_size++; + cci->cel_size++; } } } } + +void cxl_initialize_mailbox_t3(CXLCCI *cci, DeviceState *d, size_t payload_max) +{ + cci->cxl_cmd_set = cxl_cmd_set; + cci->d = d; + + /* No separation for PCI MB as protocol handled in PCI device */ + cci->intf = d; + cxl_init_cci(cci, payload_max); +} diff --git a/hw/mem/cxl_type3.c b/hw/mem/cxl_type3.c index 18ad853f5b..0529745786 100644 --- a/hw/mem/cxl_type3.c +++ b/hw/mem/cxl_type3.c @@ -716,7 +716,8 @@ static void ct3_realize(PCIDevice *pci_dev, Error **errp) pci_dev, CXL_COMPONENT_REG_BAR_IDX, PCI_BASE_ADDRESS_SPACE_MEMORY | PCI_BASE_ADDRESS_MEM_TYPE_64, mr); - cxl_device_register_block_init(OBJECT(pci_dev), &ct3d->cxl_dstate); + cxl_device_register_block_init(OBJECT(pci_dev), &ct3d->cxl_dstate, + &ct3d->cci); pci_register_bar(pci_dev, CXL_DEVICE_REG_BAR_IDX, PCI_BASE_ADDRESS_SPACE_MEMORY | PCI_BASE_ADDRESS_MEM_TYPE_64, @@ -922,7 +923,7 @@ static void ct3d_reset(DeviceState *dev) uint32_t *write_msk = ct3d->cxl_cstate.crb.cache_mem_regs_write_mask; cxl_component_register_init_common(reg_state, write_msk, CXL2_TYPE3_DEVICE); - cxl_device_register_init_common(&ct3d->cxl_dstate); + cxl_device_register_init_t3(ct3d); } static Property ct3_props[] = { diff --git a/include/hw/cxl/cxl_device.h b/include/hw/cxl/cxl_device.h index d7a2c4009e..779ca85319 100644 --- a/include/hw/cxl/cxl_device.h +++ b/include/hw/cxl/cxl_device.h @@ -111,12 +111,13 @@ typedef enum { CXL_MBOX_MAX = 0x17 } CXLRetCode; +typedef struct CXLCCI CXLCCI; typedef struct cxl_device_state CXLDeviceState; struct cxl_cmd; typedef CXLRetCode (*opcode_handler)(const struct cxl_cmd *cmd, uint8_t *payload_in, size_t len_in, uint8_t *payload_out, size_t *len_out, - CXLDeviceState *cxl_dstate); + CXLCCI *cci); struct cxl_cmd { const char *name; opcode_handler handler; @@ -140,6 +141,21 @@ typedef struct CXLEventLog { QSIMPLEQ_HEAD(, CXLEvent) events; } CXLEventLog; +typedef struct CXLCCI { + const struct cxl_cmd (*cxl_cmd_set)[256]; + struct cel_log { + uint16_t opcode; + uint16_t effect; + } cel_log[1 << 16]; + size_t cel_size; + + size_t payload_max; + /* Pointer to device hosting the CCI */ + DeviceState *d; + /* Pointer to the device hosting the protocol conversion */ + DeviceState *intf; +} CXLCCI; + typedef struct cxl_device_state { MemoryRegion device_registers; @@ -173,11 +189,6 @@ typedef struct cxl_device_state { uint32_t mbox_reg_state32[CXL_MAILBOX_REGISTERS_LENGTH / 4]; uint64_t mbox_reg_state64[CXL_MAILBOX_REGISTERS_LENGTH / 8]; }; - struct cel_log { - uint16_t opcode; - uint16_t effect; - } cel_log[1 << 16]; - size_t cel_size; }; struct { @@ -196,10 +207,12 @@ typedef struct cxl_device_state { } CXLDeviceState; /* Initialize the register block for a device */ -void cxl_device_register_block_init(Object *obj, CXLDeviceState *dev); +void cxl_device_register_block_init(Object *obj, CXLDeviceState *dev, + CXLCCI *cci); +typedef struct CXLType3Dev CXLType3Dev; /* Set up default values for the register block */ -void cxl_device_register_init_common(CXLDeviceState *dev); +void cxl_device_register_init_t3(CXLType3Dev *ct3d); /* * CXL 2.0 - 8.2.8.1 including errata F4 @@ -245,8 +258,9 @@ CXL_DEVICE_CAPABILITY_HEADER_REGISTER(MEMORY_DEVICE, CXL_DEVICE_CAP_HDR1_OFFSET + CXL_DEVICE_CAP_REG_SIZE * 2) -void cxl_initialize_mailbox(CXLDeviceState *cxl_dstate); -void cxl_process_mailbox(CXLDeviceState *cxl_dstate); +void cxl_initialize_mailbox_t3(CXLCCI *cci, DeviceState *d, size_t payload_max); +void cxl_init_cci(CXLCCI *cci, size_t payload_max); +void cxl_process_mailbox(CXLCCI *cci); #define cxl_device_cap_init(dstate, reg, cap_id, ver) \ do { \ @@ -347,6 +361,7 @@ struct CXLType3Dev { AddressSpace hostpmem_as; CXLComponentState cxl_cstate; CXLDeviceState cxl_dstate; + CXLCCI cci; /* Primary PCI mailbox CCI */ /* DOE */ DOECap doe_cdat; -- cgit v1.2.3 From c9460561edbd8b2d4adbf1f7c5cb4ad4d210de4c Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 17:07:53 +0100 Subject: hw/cxl/mbox: Generalize the CCI command processing By moving the parts of the mailbox command handling that are CCI type specific out to the caller, make the main handling code generic. Rename it to cxl_process_cci_message() to reflect this new generality. Change the type3 mailbox handling (reused shortly for the switch mailbox CCI) to take a snapshot of the mailbox input data rather than operating on it in place. This reduces the chance of bugs due to aliasing going forwars. Signed-off-by: Jonathan Cameron Message-Id: <20231023160806.13206-5-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-device-utils.c | 44 +++++++++++++++++++++++++++++++++++++- hw/cxl/cxl-mailbox-utils.c | 51 +++++++++------------------------------------ include/hw/cxl/cxl_device.h | 5 ++++- 3 files changed, 57 insertions(+), 43 deletions(-) diff --git a/hw/cxl/cxl-device-utils.c b/hw/cxl/cxl-device-utils.c index 327949a805..eb86634250 100644 --- a/hw/cxl/cxl-device-utils.c +++ b/hw/cxl/cxl-device-utils.c @@ -157,7 +157,49 @@ static void mailbox_reg_write(void *opaque, hwaddr offset, uint64_t value, if (ARRAY_FIELD_EX32(cxl_dstate->mbox_reg_state32, CXL_DEV_MAILBOX_CTRL, DOORBELL)) { - cxl_process_mailbox(cci); + uint64_t command_reg = + cxl_dstate->mbox_reg_state64[R_CXL_DEV_MAILBOX_CMD]; + uint8_t cmd_set = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, + COMMAND_SET); + uint8_t cmd = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, COMMAND); + size_t len_in = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, LENGTH); + uint8_t *pl = cxl_dstate->mbox_reg_state + A_CXL_DEV_CMD_PAYLOAD; + /* + * Copy taken to avoid need for individual command handlers to care + * about aliasing. + */ + g_autofree uint8_t *pl_in_copy = NULL; + size_t len_out = 0; + uint64_t status_reg; + bool bg_started = false; + int rc; + + pl_in_copy = g_memdup2(pl, len_in); + if (len_in == 0 || pl_in_copy) { + /* Avoid stale data - including from earlier cmds */ + memset(pl, 0, CXL_MAILBOX_MAX_PAYLOAD_SIZE); + rc = cxl_process_cci_message(cci, cmd_set, cmd, len_in, pl_in_copy, + &len_out, pl, &bg_started); + } else { + rc = CXL_MBOX_INTERNAL_ERROR; + } + + /* Set bg and the return code */ + status_reg = FIELD_DP64(0, CXL_DEV_MAILBOX_STS, BG_OP, + bg_started ? 1 : 0); + status_reg = FIELD_DP64(status_reg, CXL_DEV_MAILBOX_STS, ERRNO, rc); + /* Set the return length */ + command_reg = FIELD_DP64(0, CXL_DEV_MAILBOX_CMD, COMMAND_SET, cmd_set); + command_reg = FIELD_DP64(command_reg, CXL_DEV_MAILBOX_CMD, + COMMAND, cmd); + command_reg = FIELD_DP64(command_reg, CXL_DEV_MAILBOX_CMD, + LENGTH, len_out); + + cxl_dstate->mbox_reg_state64[R_CXL_DEV_MAILBOX_CMD] = command_reg; + cxl_dstate->mbox_reg_state64[R_CXL_DEV_MAILBOX_STS] = status_reg; + /* Tell the host we're done */ + ARRAY_FIELD_DP32(cxl_dstate->mbox_reg_state32, CXL_DEV_MAILBOX_CTRL, + DOORBELL, 0); } } diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c index 5484dfbbf1..239acc659d 100644 --- a/hw/cxl/cxl-mailbox-utils.c +++ b/hw/cxl/cxl-mailbox-utils.c @@ -754,58 +754,27 @@ static const struct cxl_cmd cxl_cmd_set[256][256] = { cmd_media_clear_poison, 72, 0 }, }; -void cxl_process_mailbox(CXLCCI *cci) +int cxl_process_cci_message(CXLCCI *cci, uint8_t set, uint8_t cmd, + size_t len_in, uint8_t *pl_in, size_t *len_out, + uint8_t *pl_out, bool *bg_started) { - uint16_t ret = CXL_MBOX_SUCCESS; const struct cxl_cmd *cxl_cmd; - uint64_t status_reg = 0; opcode_handler h; - CXLDeviceState *cxl_dstate = &CXL_TYPE3(cci->d)->cxl_dstate; - uint64_t command_reg = cxl_dstate->mbox_reg_state64[R_CXL_DEV_MAILBOX_CMD]; - - uint8_t set = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, COMMAND_SET); - uint8_t cmd = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, COMMAND); - uint16_t len_in = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, LENGTH); - uint8_t *pl = cxl_dstate->mbox_reg_state + A_CXL_DEV_CMD_PAYLOAD; - /* - * Copy taken to avoid need for individual command handlers to care - * about aliasing. - */ - g_autofree uint8_t *pl_in_copy = NULL; - size_t len_out = 0; - pl_in_copy = g_memdup2(pl, len_in); - /* Avoid stale data - including from earlier commands */ - memset(pl, 0, CXL_MAILBOX_MAX_PAYLOAD_SIZE); + *len_out = 0; cxl_cmd = &cci->cxl_cmd_set[set][cmd]; h = cxl_cmd->handler; - if (h) { - if (len_in == cxl_cmd->in || cxl_cmd->in == ~0) { - ret = (*h)(cxl_cmd, pl, len_in, pl, &len_out, cci); - assert(len_out <= cci->payload_max); - } else { - ret = CXL_MBOX_INVALID_PAYLOAD_LENGTH; - } - } else { + if (!h) { qemu_log_mask(LOG_UNIMP, "Command %04xh not implemented\n", set << 8 | cmd); - ret = CXL_MBOX_UNSUPPORTED; + return CXL_MBOX_UNSUPPORTED; } - /* Set the return code */ - status_reg = FIELD_DP64(0, CXL_DEV_MAILBOX_STS, ERRNO, ret); - - /* Set the return length */ - command_reg = FIELD_DP64(command_reg, CXL_DEV_MAILBOX_CMD, COMMAND_SET, 0); - command_reg = FIELD_DP64(command_reg, CXL_DEV_MAILBOX_CMD, COMMAND, 0); - command_reg = FIELD_DP64(command_reg, CXL_DEV_MAILBOX_CMD, LENGTH, len_out); - - cxl_dstate->mbox_reg_state64[R_CXL_DEV_MAILBOX_CMD] = command_reg; - cxl_dstate->mbox_reg_state64[R_CXL_DEV_MAILBOX_STS] = status_reg; + if (len_in != cxl_cmd->in && cxl_cmd->in != ~0) { + return CXL_MBOX_INVALID_PAYLOAD_LENGTH; + } - /* Tell the host we're done */ - ARRAY_FIELD_DP32(cxl_dstate->mbox_reg_state32, CXL_DEV_MAILBOX_CTRL, - DOORBELL, 0); + return (*h)(cxl_cmd, pl_in, len_in, pl_out, len_out, cci); } void cxl_init_cci(CXLCCI *cci, size_t payload_max) diff --git a/include/hw/cxl/cxl_device.h b/include/hw/cxl/cxl_device.h index 779ca85319..6f8040b5ff 100644 --- a/include/hw/cxl/cxl_device.h +++ b/include/hw/cxl/cxl_device.h @@ -260,7 +260,10 @@ CXL_DEVICE_CAPABILITY_HEADER_REGISTER(MEMORY_DEVICE, void cxl_initialize_mailbox_t3(CXLCCI *cci, DeviceState *d, size_t payload_max); void cxl_init_cci(CXLCCI *cci, size_t payload_max); -void cxl_process_mailbox(CXLCCI *cci); +int cxl_process_cci_message(CXLCCI *cci, uint8_t set, uint8_t cmd, + size_t len_in, uint8_t *pl_in, + size_t *len_out, uint8_t *pl_out, + bool *bg_started); #define cxl_device_cap_init(dstate, reg, cap_id, ver) \ do { \ -- cgit v1.2.3 From 2710d49a7c8b9b117a46847c7ace5eb21d48e882 Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 17:07:54 +0100 Subject: hw/pci-bridge/cxl_upstream: Move defintion of device to header. To avoid repetition of switch upstream port specific data in the CXLDeviceState structure it will be necessary to access the switch USP specific data from mailbox callbacks. Hence move it to cxl_device.h so it is no longer an opaque structure. Signed-off-by: Jonathan Cameron Reviewed-by: Fan Ni Message-Id: <20231023160806.13206-6-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/pci-bridge/cxl_upstream.c | 11 +---------- include/hw/pci-bridge/cxl_upstream_port.h | 18 ++++++++++++++++++ 2 files changed, 19 insertions(+), 10 deletions(-) create mode 100644 include/hw/pci-bridge/cxl_upstream_port.h diff --git a/hw/pci-bridge/cxl_upstream.c b/hw/pci-bridge/cxl_upstream.c index b81bb5fec9..36737189c6 100644 --- a/hw/pci-bridge/cxl_upstream.c +++ b/hw/pci-bridge/cxl_upstream.c @@ -14,6 +14,7 @@ #include "hw/pci/msi.h" #include "hw/pci/pcie.h" #include "hw/pci/pcie_port.h" +#include "hw/pci-bridge/cxl_upstream_port.h" /* * Null value of all Fs suggested by IEEE RA guidelines for use of * EU, OUI and CID @@ -30,16 +31,6 @@ #define CXL_UPSTREAM_PORT_DVSEC_OFFSET \ (CXL_UPSTREAM_PORT_SN_OFFSET + PCI_EXT_CAP_DSN_SIZEOF) -typedef struct CXLUpstreamPort { - /*< private >*/ - PCIEPort parent_obj; - - /*< public >*/ - CXLComponentState cxl_cstate; - DOECap doe_cdat; - uint64_t sn; -} CXLUpstreamPort; - CXLComponentState *cxl_usp_to_cstate(CXLUpstreamPort *usp) { return &usp->cxl_cstate; diff --git a/include/hw/pci-bridge/cxl_upstream_port.h b/include/hw/pci-bridge/cxl_upstream_port.h new file mode 100644 index 0000000000..b02aa8f659 --- /dev/null +++ b/include/hw/pci-bridge/cxl_upstream_port.h @@ -0,0 +1,18 @@ + +#ifndef CXL_USP_H +#define CXL_USP_H +#include "hw/pci/pcie.h" +#include "hw/pci/pcie_port.h" +#include "hw/cxl/cxl.h" + +typedef struct CXLUpstreamPort { + /*< private >*/ + PCIEPort parent_obj; + + /*< public >*/ + CXLComponentState cxl_cstate; + DOECap doe_cdat; + uint64_t sn; +} CXLUpstreamPort; + +#endif /* CXL_SUP_H */ -- cgit v1.2.3 From 4a58330343e6a16f6828e225fd0c054c8d1916bd Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 17:07:55 +0100 Subject: hw/cxl: Add a switch mailbox CCI function CXL switch CCIs were added in CXL r3.0. They are a PCI function, identified by class code that provides a CXL mailbox (identical to that previously defined for CXL type 3 memory devices) over which various FM-API commands may be used. Whilst the intent of this feature is enable switch control from a BMC attached to a switch upstream port, it is also useful to allow emulation of this feature on the upstream port connected to a host using the CXL devices as this greatly simplifies testing. Signed-off-by: Jonathan Cameron Message-Id: <20231023160806.13206-7-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-device-utils.c | 27 ++++++++ hw/cxl/cxl-mailbox-utils.c | 18 +++++ hw/cxl/meson.build | 1 + hw/cxl/switch-mailbox-cci.c | 111 ++++++++++++++++++++++++++++++ include/hw/cxl/cxl_component.h | 3 +- include/hw/cxl/cxl_device.h | 15 ++++ include/hw/pci-bridge/cxl_upstream_port.h | 1 + 7 files changed, 175 insertions(+), 1 deletion(-) create mode 100644 hw/cxl/switch-mailbox-cci.c diff --git a/hw/cxl/cxl-device-utils.c b/hw/cxl/cxl-device-utils.c index eb86634250..f8938678c7 100644 --- a/hw/cxl/cxl-device-utils.c +++ b/hw/cxl/cxl-device-utils.c @@ -67,6 +67,9 @@ static uint64_t mailbox_reg_read(void *opaque, hwaddr offset, unsigned size) if (object_dynamic_cast(OBJECT(cci->intf), TYPE_CXL_TYPE3)) { cxl_dstate = &CXL_TYPE3(cci->intf)->cxl_dstate; + } else if (object_dynamic_cast(OBJECT(cci->intf), + TYPE_CXL_SWITCH_MAILBOX_CCI)) { + cxl_dstate = &CXL_SWITCH_MAILBOX_CCI(cci->intf)->cxl_dstate; } else { return 0; } @@ -135,6 +138,9 @@ static void mailbox_reg_write(void *opaque, hwaddr offset, uint64_t value, if (object_dynamic_cast(OBJECT(cci->intf), TYPE_CXL_TYPE3)) { cxl_dstate = &CXL_TYPE3(cci->intf)->cxl_dstate; + } else if (object_dynamic_cast(OBJECT(cci->intf), + TYPE_CXL_SWITCH_MAILBOX_CCI)) { + cxl_dstate = &CXL_SWITCH_MAILBOX_CCI(cci->intf)->cxl_dstate; } else { return; } @@ -365,6 +371,27 @@ void cxl_device_register_init_t3(CXLType3Dev *ct3d) CXL_MAILBOX_MAX_PAYLOAD_SIZE); } +void cxl_device_register_init_swcci(CSWMBCCIDev *sw) +{ + CXLDeviceState *cxl_dstate = &sw->cxl_dstate; + uint64_t *cap_h = cxl_dstate->caps_reg_state64; + const int cap_count = 3; + + /* CXL Device Capabilities Array Register */ + ARRAY_FIELD_DP64(cap_h, CXL_DEV_CAP_ARRAY, CAP_ID, 0); + ARRAY_FIELD_DP64(cap_h, CXL_DEV_CAP_ARRAY, CAP_VERSION, 1); + ARRAY_FIELD_DP64(cap_h, CXL_DEV_CAP_ARRAY, CAP_COUNT, cap_count); + + cxl_device_cap_init(cxl_dstate, DEVICE_STATUS, 1, 2); + device_reg_init_common(cxl_dstate); + + cxl_device_cap_init(cxl_dstate, MAILBOX, 2, 1); + mailbox_reg_init_common(cxl_dstate); + + cxl_device_cap_init(cxl_dstate, MEMORY_DEVICE, 0x4000, 1); + memdev_reg_init_common(cxl_dstate); +} + uint64_t cxl_device_get_timestamp(CXLDeviceState *cxl_dstate) { uint64_t time, delta; diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c index 239acc659d..28ea02fcbe 100644 --- a/hw/cxl/cxl-mailbox-utils.c +++ b/hw/cxl/cxl-mailbox-utils.c @@ -754,6 +754,15 @@ static const struct cxl_cmd cxl_cmd_set[256][256] = { cmd_media_clear_poison, 72, 0 }, }; +static const struct cxl_cmd cxl_cmd_set_sw[256][256] = { + [TIMESTAMP][GET] = { "TIMESTAMP_GET", cmd_timestamp_get, 0, 0 }, + [TIMESTAMP][SET] = { "TIMESTAMP_SET", cmd_timestamp_set, 0, + IMMEDIATE_POLICY_CHANGE }, + [LOGS][GET_SUPPORTED] = { "LOGS_GET_SUPPORTED", cmd_logs_get_supported, 0, + 0 }, + [LOGS][GET_LOG] = { "LOGS_GET_LOG", cmd_logs_get_log, 0x18, 0 }, +}; + int cxl_process_cci_message(CXLCCI *cci, uint8_t set, uint8_t cmd, size_t len_in, uint8_t *pl_in, size_t *len_out, uint8_t *pl_out, bool *bg_started) @@ -795,6 +804,15 @@ void cxl_init_cci(CXLCCI *cci, size_t payload_max) } } +void cxl_initialize_mailbox_swcci(CXLCCI *cci, DeviceState *intf, + DeviceState *d, size_t payload_max) +{ + cci->cxl_cmd_set = cxl_cmd_set_sw; + cci->d = d; + cci->intf = intf; + cxl_init_cci(cci, payload_max); +} + void cxl_initialize_mailbox_t3(CXLCCI *cci, DeviceState *d, size_t payload_max) { cci->cxl_cmd_set = cxl_cmd_set; diff --git a/hw/cxl/meson.build b/hw/cxl/meson.build index e261ff3881..ea0aebf6e3 100644 --- a/hw/cxl/meson.build +++ b/hw/cxl/meson.build @@ -6,6 +6,7 @@ system_ss.add(when: 'CONFIG_CXL', 'cxl-host.c', 'cxl-cdat.c', 'cxl-events.c', + 'switch-mailbox-cci.c', ), if_false: files( 'cxl-host-stubs.c', diff --git a/hw/cxl/switch-mailbox-cci.c b/hw/cxl/switch-mailbox-cci.c new file mode 100644 index 0000000000..ba399c6240 --- /dev/null +++ b/hw/cxl/switch-mailbox-cci.c @@ -0,0 +1,111 @@ +/* + * SPDX-License-Identifier: GPL-2.0-or-later + * + * Emulation of a CXL Switch Mailbox CCI PCIe function. + * + * Copyright (c) 2023 Huawei Technologies. + * + * From www.computeexpresslink.org + * Compute Express Link (CXL) Specification revision 3.0 Version 1.0 + */ +#include "qemu/osdep.h" +#include "hw/pci/pci.h" +#include "hw/pci-bridge/cxl_upstream_port.h" +#include "qapi/error.h" +#include "qemu/log.h" +#include "qemu/module.h" +#include "hw/qdev-properties.h" +#include "hw/cxl/cxl.h" + +static void cswmbcci_reset(DeviceState *dev) +{ + CSWMBCCIDev *cswmb = CXL_SWITCH_MAILBOX_CCI(dev); + cxl_device_register_init_swcci(cswmb); +} + +static void cswbcci_realize(PCIDevice *pci_dev, Error **errp) +{ + CSWMBCCIDev *cswmb = CXL_SWITCH_MAILBOX_CCI(pci_dev); + CXLComponentState *cxl_cstate = &cswmb->cxl_cstate; + CXLDeviceState *cxl_dstate = &cswmb->cxl_dstate; + CXLDVSECRegisterLocator *regloc_dvsec; + CXLUpstreamPort *usp; + + if (!cswmb->target) { + error_setg(errp, "Target not set"); + return; + } + usp = CXL_USP(cswmb->target); + + pcie_endpoint_cap_init(pci_dev, 0x80); + cxl_cstate->dvsec_offset = 0x100; + cxl_cstate->pdev = pci_dev; + cswmb->cci = &usp->swcci; + cxl_device_register_block_init(OBJECT(pci_dev), cxl_dstate, cswmb->cci); + pci_register_bar(pci_dev, 0, + PCI_BASE_ADDRESS_SPACE_MEMORY | + PCI_BASE_ADDRESS_MEM_TYPE_64, + &cxl_dstate->device_registers); + regloc_dvsec = &(CXLDVSECRegisterLocator) { + .rsvd = 0, + .reg0_base_lo = RBI_CXL_DEVICE_REG | 0, + .reg0_base_hi = 0, + }; + cxl_component_create_dvsec(cxl_cstate, CXL3_SWITCH_MAILBOX_CCI, + REG_LOC_DVSEC_LENGTH, REG_LOC_DVSEC, + REG_LOC_DVSEC_REVID, (uint8_t *)regloc_dvsec); + + cxl_initialize_mailbox_swcci(cswmb->cci, DEVICE(pci_dev), + DEVICE(cswmb->target), + CXL_MAILBOX_MAX_PAYLOAD_SIZE); +} + +static void cswmbcci_exit(PCIDevice *pci_dev) +{ + /* Nothing to do here yet */ +} + +static Property cxl_switch_cci_props[] = { + DEFINE_PROP_LINK("target", CSWMBCCIDev, + target, TYPE_CXL_USP, PCIDevice *), + DEFINE_PROP_END_OF_LIST(), +}; + +static void cswmbcci_class_init(ObjectClass *oc, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(oc); + PCIDeviceClass *pc = PCI_DEVICE_CLASS(oc); + + pc->realize = cswbcci_realize; + pc->exit = cswmbcci_exit; + /* Serial bus, CXL Switch CCI */ + pc->class_id = 0x0c0b; + /* + * Huawei Technologies + * CXL Switch Mailbox CCI - DID assigned for emulation only. + * No real hardware will ever use this ID. + */ + pc->vendor_id = 0x19e5; + pc->device_id = 0xa123; + pc->revision = 0; + dc->desc = "CXL Switch Mailbox CCI"; + dc->reset = cswmbcci_reset; + device_class_set_props(dc, cxl_switch_cci_props); +} + +static const TypeInfo cswmbcci_info = { + .name = TYPE_CXL_SWITCH_MAILBOX_CCI, + .parent = TYPE_PCI_DEVICE, + .class_init = cswmbcci_class_init, + .instance_size = sizeof(CSWMBCCIDev), + .interfaces = (InterfaceInfo[]) { + { INTERFACE_PCIE_DEVICE }, + { } + }, +}; + +static void cxl_switch_mailbox_cci_register(void) +{ + type_register_static(&cswmbcci_info); +} +type_init(cxl_switch_mailbox_cci_register); diff --git a/include/hw/cxl/cxl_component.h b/include/hw/cxl/cxl_component.h index e52dd8d2b9..5227a8e833 100644 --- a/include/hw/cxl/cxl_component.h +++ b/include/hw/cxl/cxl_component.h @@ -26,7 +26,8 @@ enum reg_type { CXL2_LOGICAL_DEVICE, CXL2_ROOT_PORT, CXL2_UPSTREAM_PORT, - CXL2_DOWNSTREAM_PORT + CXL2_DOWNSTREAM_PORT, + CXL3_SWITCH_MAILBOX_CCI, }; /* diff --git a/include/hw/cxl/cxl_device.h b/include/hw/cxl/cxl_device.h index 6f8040b5ff..fa73ed03e5 100644 --- a/include/hw/cxl/cxl_device.h +++ b/include/hw/cxl/cxl_device.h @@ -211,8 +211,10 @@ void cxl_device_register_block_init(Object *obj, CXLDeviceState *dev, CXLCCI *cci); typedef struct CXLType3Dev CXLType3Dev; +typedef struct CSWMBCCIDev CSWMBCCIDev; /* Set up default values for the register block */ void cxl_device_register_init_t3(CXLType3Dev *ct3d); +void cxl_device_register_init_swcci(CSWMBCCIDev *sw); /* * CXL 2.0 - 8.2.8.1 including errata F4 @@ -259,6 +261,8 @@ CXL_DEVICE_CAPABILITY_HEADER_REGISTER(MEMORY_DEVICE, CXL_DEVICE_CAP_REG_SIZE * 2) void cxl_initialize_mailbox_t3(CXLCCI *cci, DeviceState *d, size_t payload_max); +void cxl_initialize_mailbox_swcci(CXLCCI *cci, DeviceState *intf, + DeviceState *d, size_t payload_max); void cxl_init_cci(CXLCCI *cci, size_t payload_max); int cxl_process_cci_message(CXLCCI *cci, uint8_t set, uint8_t cmd, size_t len_in, uint8_t *pl_in, @@ -397,6 +401,17 @@ struct CXLType3Class { uint8_t *data); }; +struct CSWMBCCIDev { + PCIDevice parent_obj; + PCIDevice *target; + CXLComponentState cxl_cstate; + CXLDeviceState cxl_dstate; + CXLCCI *cci; +}; + +#define TYPE_CXL_SWITCH_MAILBOX_CCI "cxl-switch-mailbox-cci" +OBJECT_DECLARE_TYPE(CSWMBCCIDev, CSWMBCCIClass, CXL_SWITCH_MAILBOX_CCI) + MemTxResult cxl_type3_read(PCIDevice *d, hwaddr host_addr, uint64_t *data, unsigned size, MemTxAttrs attrs); MemTxResult cxl_type3_write(PCIDevice *d, hwaddr host_addr, uint64_t data, diff --git a/include/hw/pci-bridge/cxl_upstream_port.h b/include/hw/pci-bridge/cxl_upstream_port.h index b02aa8f659..12635139f6 100644 --- a/include/hw/pci-bridge/cxl_upstream_port.h +++ b/include/hw/pci-bridge/cxl_upstream_port.h @@ -11,6 +11,7 @@ typedef struct CXLUpstreamPort { /*< public >*/ CXLComponentState cxl_cstate; + CXLCCI swcci; DOECap doe_cdat; uint64_t sn; } CXLUpstreamPort; -- cgit v1.2.3 From 6cf416c176f11bc093fcc1818d6c9a95136f4c59 Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 17:07:56 +0100 Subject: hw/cxl/mbox: Add Information and Status / Identify command Add this command that is only available via out of band CCIs. It replicates information that can be discovered inband via PCI config space. Signed-off-by: Jonathan Cameron Message-Id: <20231023160806.13206-8-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-mailbox-utils.c | 55 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 55 insertions(+) diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c index 28ea02fcbe..6741698ee7 100644 --- a/hw/cxl/cxl-mailbox-utils.c +++ b/hw/cxl/cxl-mailbox-utils.c @@ -11,6 +11,7 @@ #include "hw/cxl/cxl.h" #include "hw/cxl/cxl_events.h" #include "hw/pci/pci.h" +#include "hw/pci-bridge/cxl_upstream_port.h" #include "qemu/cutils.h" #include "qemu/log.h" #include "qemu/units.h" @@ -44,6 +45,8 @@ */ enum { + INFOSTAT = 0x00, + #define IS_IDENTIFY 0x1 EVENTS = 0x01, #define GET_RECORDS 0x0 #define CLEAR_RECORDS 0x1 @@ -203,6 +206,57 @@ static CXLRetCode cmd_events_set_interrupt_policy(const struct cxl_cmd *cmd, return CXL_MBOX_SUCCESS; } +/* CXL r3.0 section 8.2.9.1.1: Identify (Opcode 0001h) */ +static CXLRetCode cmd_infostat_identify(const struct cxl_cmd *cmd, + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLCCI *cci) +{ + PCIDeviceClass *class = PCI_DEVICE_GET_CLASS(cci->d); + struct { + uint16_t pcie_vid; + uint16_t pcie_did; + uint16_t pcie_subsys_vid; + uint16_t pcie_subsys_id; + uint64_t sn; + uint8_t max_message_size; + uint8_t component_type; + } QEMU_PACKED *is_identify; + QEMU_BUILD_BUG_ON(sizeof(*is_identify) != 18); + + is_identify = (void *)payload_out; + memset(is_identify, 0, sizeof(*is_identify)); + is_identify->pcie_vid = class->vendor_id; + is_identify->pcie_did = class->device_id; + if (object_dynamic_cast(OBJECT(cci->d), TYPE_CXL_USP)) { + is_identify->sn = CXL_USP(cci->d)->sn; + /* Subsystem info not defined for a USP */ + is_identify->pcie_subsys_vid = 0; + is_identify->pcie_subsys_id = 0; + is_identify->component_type = 0x0; /* Switch */ + } else if (object_dynamic_cast(OBJECT(cci->d), TYPE_CXL_TYPE3)) { + PCIDevice *pci_dev = PCI_DEVICE(cci->d); + + is_identify->sn = CXL_TYPE3(cci->d)->sn; + /* + * We can't always use class->subsystem_vendor_id as + * it is not set if the defaults are used. + */ + is_identify->pcie_subsys_vid = + pci_get_word(pci_dev->config + PCI_SUBSYSTEM_VENDOR_ID); + is_identify->pcie_subsys_id = + pci_get_word(pci_dev->config + PCI_SUBSYSTEM_ID); + is_identify->component_type = 0x3; /* Type 3 */ + } + + /* TODO: Allow this to vary across different CCIs */ + is_identify->max_message_size = 9; /* 512 bytes - MCTP_CXL_MAILBOX_BYTES */ + *len_out = sizeof(*is_identify); + return CXL_MBOX_SUCCESS; +} + /* 8.2.9.2.1 */ static CXLRetCode cmd_firmware_update_get_info(const struct cxl_cmd *cmd, uint8_t *payload_in, @@ -755,6 +809,7 @@ static const struct cxl_cmd cxl_cmd_set[256][256] = { }; static const struct cxl_cmd cxl_cmd_set_sw[256][256] = { + [INFOSTAT][IS_IDENTIFY] = { "IDENTIFY", cmd_infostat_identify, 0, 0 }, [TIMESTAMP][GET] = { "TIMESTAMP_GET", cmd_timestamp_get, 0, 0 }, [TIMESTAMP][SET] = { "TIMESTAMP_SET", cmd_timestamp_set, 0, IMMEDIATE_POLICY_CHANGE }, -- cgit v1.2.3 From 3314efd276ada18cc0b8beb70b8943f8deb872b7 Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 17:07:57 +0100 Subject: hw/cxl/mbox: Add Physical Switch Identify command. Enable it for the switch CCI. Signed-off-by: Jonathan Cameron Message-Id: <20231023160806.13206-9-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-mailbox-utils.c | 65 ++++++++++++++++++++++++++++++++++++++++++ hw/pci-bridge/cxl_downstream.c | 4 +-- include/hw/cxl/cxl.h | 6 ++++ 3 files changed, 72 insertions(+), 3 deletions(-) diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c index 6741698ee7..6ada49d37c 100644 --- a/hw/cxl/cxl-mailbox-utils.c +++ b/hw/cxl/cxl-mailbox-utils.c @@ -70,6 +70,8 @@ enum { #define GET_POISON_LIST 0x0 #define INJECT_POISON 0x1 #define CLEAR_POISON 0x2 + PHYSICAL_SWITCH = 0x51, + #define IDENTIFY_SWITCH_DEVICE 0x0 }; @@ -257,6 +259,67 @@ static CXLRetCode cmd_infostat_identify(const struct cxl_cmd *cmd, return CXL_MBOX_SUCCESS; } +static void cxl_set_dsp_active_bm(PCIBus *b, PCIDevice *d, + void *private) +{ + uint8_t *bm = private; + if (object_dynamic_cast(OBJECT(d), TYPE_CXL_DSP)) { + uint8_t port = PCIE_PORT(d)->port; + bm[port / 8] |= 1 << (port % 8); + } +} + +/* CXL r3 8.2.9.1.1 */ +static CXLRetCode cmd_identify_switch_device(const struct cxl_cmd *cmd, + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLCCI *cci) +{ + PCIEPort *usp = PCIE_PORT(cci->d); + PCIBus *bus = &PCI_BRIDGE(cci->d)->sec_bus; + int num_phys_ports = pcie_count_ds_ports(bus); + + struct cxl_fmapi_ident_switch_dev_resp_pl { + uint8_t ingress_port_id; + uint8_t rsvd; + uint8_t num_physical_ports; + uint8_t num_vcss; + uint8_t active_port_bitmask[0x20]; + uint8_t active_vcs_bitmask[0x20]; + uint16_t total_vppbs; + uint16_t bound_vppbs; + uint8_t num_hdm_decoders_per_usp; + } QEMU_PACKED *out; + QEMU_BUILD_BUG_ON(sizeof(*out) != 0x49); + + out = (struct cxl_fmapi_ident_switch_dev_resp_pl *)payload_out; + *out = (struct cxl_fmapi_ident_switch_dev_resp_pl) { + .num_physical_ports = num_phys_ports + 1, /* 1 USP */ + .num_vcss = 1, /* Not yet support multiple VCS - potentialy tricky */ + .active_vcs_bitmask[0] = 0x1, + .total_vppbs = num_phys_ports + 1, + .bound_vppbs = num_phys_ports + 1, + .num_hdm_decoders_per_usp = 4, + }; + + /* Depends on the CCI type */ + if (object_dynamic_cast(OBJECT(cci->intf), TYPE_PCIE_PORT)) { + out->ingress_port_id = PCIE_PORT(cci->intf)->port; + } else { + /* MCTP? */ + out->ingress_port_id = 0; + } + + pci_for_each_device_under_bus(bus, cxl_set_dsp_active_bm, + out->active_port_bitmask); + out->active_port_bitmask[usp->port / 8] |= (1 << usp->port % 8); + + *len_out = sizeof(*out); + + return CXL_MBOX_SUCCESS; +} /* 8.2.9.2.1 */ static CXLRetCode cmd_firmware_update_get_info(const struct cxl_cmd *cmd, uint8_t *payload_in, @@ -816,6 +879,8 @@ static const struct cxl_cmd cxl_cmd_set_sw[256][256] = { [LOGS][GET_SUPPORTED] = { "LOGS_GET_SUPPORTED", cmd_logs_get_supported, 0, 0 }, [LOGS][GET_LOG] = { "LOGS_GET_LOG", cmd_logs_get_log, 0x18, 0 }, + [PHYSICAL_SWITCH][IDENTIFY_SWITCH_DEVICE] = { "IDENTIFY_SWITCH_DEVICE", + cmd_identify_switch_device, 0, 0 }, }; int cxl_process_cci_message(CXLCCI *cci, uint8_t set, uint8_t cmd, diff --git a/hw/pci-bridge/cxl_downstream.c b/hw/pci-bridge/cxl_downstream.c index 8c0f759add..8d99e1e96d 100644 --- a/hw/pci-bridge/cxl_downstream.c +++ b/hw/pci-bridge/cxl_downstream.c @@ -13,6 +13,7 @@ #include "hw/pci/msi.h" #include "hw/pci/pcie.h" #include "hw/pci/pcie_port.h" +#include "hw/cxl/cxl.h" #include "qapi/error.h" typedef struct CXLDownstreamPort { @@ -23,9 +24,6 @@ typedef struct CXLDownstreamPort { CXLComponentState cxl_cstate; } CXLDownstreamPort; -#define TYPE_CXL_DSP "cxl-downstream" -DECLARE_INSTANCE_CHECKER(CXLDownstreamPort, CXL_DSP, TYPE_CXL_DSP) - #define CXL_DOWNSTREAM_PORT_MSI_OFFSET 0x70 #define CXL_DOWNSTREAM_PORT_MSI_NR_VECTOR 1 #define CXL_DOWNSTREAM_PORT_EXP_OFFSET 0x90 diff --git a/include/hw/cxl/cxl.h b/include/hw/cxl/cxl.h index 4944725849..75e47b6864 100644 --- a/include/hw/cxl/cxl.h +++ b/include/hw/cxl/cxl.h @@ -61,4 +61,10 @@ OBJECT_DECLARE_SIMPLE_TYPE(CXLHost, PXB_CXL_HOST) typedef struct CXLUpstreamPort CXLUpstreamPort; DECLARE_INSTANCE_CHECKER(CXLUpstreamPort, CXL_USP, TYPE_CXL_USP) CXLComponentState *cxl_usp_to_cstate(CXLUpstreamPort *usp); + +#define TYPE_CXL_DSP "cxl-downstream" + +typedef struct CXLDownstreamPort CXLDownstreamPort; +DECLARE_INSTANCE_CHECKER(CXLDownstreamPort, CXL_DSP, TYPE_CXL_DSP) + #endif -- cgit v1.2.3 From 314f5033c639ebe8218078a17513935747f15d9d Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 17:07:58 +0100 Subject: hw/pci-bridge/cxl_downstream: Set default link width and link speed Without these being set the PCIE Link Capabilities register has invalid values in these two fields. Signed-off-by: Jonathan Cameron Message-Id: <20231023160806.13206-10-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/pci-bridge/cxl_downstream.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/hw/pci-bridge/cxl_downstream.c b/hw/pci-bridge/cxl_downstream.c index 8d99e1e96d..405a133eef 100644 --- a/hw/pci-bridge/cxl_downstream.c +++ b/hw/pci-bridge/cxl_downstream.c @@ -210,6 +210,19 @@ static void cxl_dsp_exitfn(PCIDevice *d) pci_bridge_exitfn(d); } +static void cxl_dsp_instance_post_init(Object *obj) +{ + PCIESlot *s = PCIE_SLOT(obj); + + if (!s->speed) { + s->speed = QEMU_PCI_EXP_LNK_2_5GT; + } + + if (!s->width) { + s->width = QEMU_PCI_EXP_LNK_X1; + } +} + static void cxl_dsp_class_init(ObjectClass *oc, void *data) { DeviceClass *dc = DEVICE_CLASS(oc); @@ -230,6 +243,7 @@ static const TypeInfo cxl_dsp_info = { .name = TYPE_CXL_DSP, .instance_size = sizeof(CXLDownstreamPort), .parent = TYPE_PCIE_SLOT, + .instance_post_init = cxl_dsp_instance_post_init, .class_init = cxl_dsp_class_init, .interfaces = (InterfaceInfo[]) { { INTERFACE_PCIE_DEVICE }, -- cgit v1.2.3 From 892e3479d7283a38b50e06f42624f5a1ce61d24a Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 17:07:59 +0100 Subject: hw/cxl: Implement Physical Ports status retrieval Add this command for both the Switch CCI in switch upstream ports. Signed-off-by: Jonathan Cameron Message-Id: <20231023160806.13206-11-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-mailbox-utils.c | 128 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 128 insertions(+) diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c index 6ada49d37c..de63fc1a28 100644 --- a/hw/cxl/cxl-mailbox-utils.c +++ b/hw/cxl/cxl-mailbox-utils.c @@ -72,6 +72,7 @@ enum { #define CLEAR_POISON 0x2 PHYSICAL_SWITCH = 0x51, #define IDENTIFY_SWITCH_DEVICE 0x0 + #define GET_PHYSICAL_PORT_STATE 0x1 }; @@ -320,6 +321,131 @@ static CXLRetCode cmd_identify_switch_device(const struct cxl_cmd *cmd, return CXL_MBOX_SUCCESS; } + +/* CXL r3.0 Section 7.6.7.1.2: Get Physical Port State (Opcode 5101h) */ +static CXLRetCode cmd_get_physical_port_state(const struct cxl_cmd *cmd, + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLCCI *cci) +{ + /* CXL r3.0 Table 7-18: Get Physical Port State Request Payload */ + struct cxl_fmapi_get_phys_port_state_req_pl { + uint8_t num_ports; + uint8_t ports[]; + } QEMU_PACKED *in; + + /* + * CXL r3.0 Table 7-20: Get Physical Port State Port Information Block + * Format + */ + struct cxl_fmapi_port_state_info_block { + uint8_t port_id; + uint8_t config_state; + uint8_t connected_device_cxl_version; + uint8_t rsv1; + uint8_t connected_device_type; + uint8_t port_cxl_version_bitmask; + uint8_t max_link_width; + uint8_t negotiated_link_width; + uint8_t supported_link_speeds_vector; + uint8_t max_link_speed; + uint8_t current_link_speed; + uint8_t ltssm_state; + uint8_t first_lane_num; + uint16_t link_state; + uint8_t supported_ld_count; + } QEMU_PACKED; + + /* CXL r3.0 Table 7-19: Get Physical Port State Response Payload */ + struct cxl_fmapi_get_phys_port_state_resp_pl { + uint8_t num_ports; + uint8_t rsv1[3]; + struct cxl_fmapi_port_state_info_block ports[]; + } QEMU_PACKED *out; + PCIBus *bus = &PCI_BRIDGE(cci->d)->sec_bus; + PCIEPort *usp = PCIE_PORT(cci->d); + size_t pl_size; + int i; + + in = (struct cxl_fmapi_get_phys_port_state_req_pl *)payload_in; + out = (struct cxl_fmapi_get_phys_port_state_resp_pl *)payload_out; + + /* Check if what was requested can fit */ + if (sizeof(*out) + sizeof(*out->ports) * in->num_ports > cci->payload_max) { + return CXL_MBOX_INVALID_INPUT; + } + + /* For success there should be a match for each requested */ + out->num_ports = in->num_ports; + + for (i = 0; i < in->num_ports; i++) { + struct cxl_fmapi_port_state_info_block *port; + /* First try to match on downstream port */ + PCIDevice *port_dev; + uint16_t lnkcap, lnkcap2, lnksta; + + port = &out->ports[i]; + + port_dev = pcie_find_port_by_pn(bus, in->ports[i]); + if (port_dev) { /* DSP */ + PCIDevice *ds_dev = pci_bridge_get_sec_bus(PCI_BRIDGE(port_dev)) + ->devices[0]; + port->config_state = 3; + if (ds_dev) { + if (object_dynamic_cast(OBJECT(ds_dev), TYPE_CXL_TYPE3)) { + port->connected_device_type = 5; /* Assume MLD for now */ + } else { + port->connected_device_type = 1; + } + } else { + port->connected_device_type = 0; + } + port->supported_ld_count = 3; + } else if (usp->port == in->ports[i]) { /* USP */ + port_dev = PCI_DEVICE(usp); + port->config_state = 4; + port->connected_device_type = 0; + } else { + return CXL_MBOX_INVALID_INPUT; + } + + port->port_id = in->ports[i]; + /* Information on status of this port in lnksta, lnkcap */ + if (!port_dev->exp.exp_cap) { + return CXL_MBOX_INTERNAL_ERROR; + } + lnksta = port_dev->config_read(port_dev, + port_dev->exp.exp_cap + PCI_EXP_LNKSTA, + sizeof(lnksta)); + lnkcap = port_dev->config_read(port_dev, + port_dev->exp.exp_cap + PCI_EXP_LNKCAP, + sizeof(lnkcap)); + lnkcap2 = port_dev->config_read(port_dev, + port_dev->exp.exp_cap + PCI_EXP_LNKCAP2, + sizeof(lnkcap2)); + + port->max_link_width = (lnkcap & PCI_EXP_LNKCAP_MLW) >> 4; + port->negotiated_link_width = (lnksta & PCI_EXP_LNKSTA_NLW) >> 4; + /* No definition for SLS field in linux/pci_regs.h */ + port->supported_link_speeds_vector = (lnkcap2 & 0xFE) >> 1; + port->max_link_speed = lnkcap & PCI_EXP_LNKCAP_SLS; + port->current_link_speed = lnksta & PCI_EXP_LNKSTA_CLS; + /* TODO: Track down if we can get the rest of the info */ + port->ltssm_state = 0x7; + port->first_lane_num = 0; + port->link_state = 0; + port->port_cxl_version_bitmask = 0x2; + port->connected_device_cxl_version = 0x2; + } + + pl_size = sizeof(*out) + sizeof(*out->ports) * in->num_ports; + *len_out = pl_size; + + return CXL_MBOX_SUCCESS; +} + /* 8.2.9.2.1 */ static CXLRetCode cmd_firmware_update_get_info(const struct cxl_cmd *cmd, uint8_t *payload_in, @@ -881,6 +1007,8 @@ static const struct cxl_cmd cxl_cmd_set_sw[256][256] = { [LOGS][GET_LOG] = { "LOGS_GET_LOG", cmd_logs_get_log, 0x18, 0 }, [PHYSICAL_SWITCH][IDENTIFY_SWITCH_DEVICE] = { "IDENTIFY_SWITCH_DEVICE", cmd_identify_switch_device, 0, 0 }, + [PHYSICAL_SWITCH][GET_PHYSICAL_PORT_STATE] = { "SWITCH_PHYSICAL_PORT_STATS", + cmd_get_physical_port_state, ~0, 0 }, }; int cxl_process_cci_message(CXLCCI *cci, uint8_t set, uint8_t cmd, -- cgit v1.2.3 From 221d2cfbdb5301f8f0cfbf26baf76544a5d71c27 Mon Sep 17 00:00:00 2001 From: Davidlohr Bueso Date: Mon, 23 Oct 2023 17:08:00 +0100 Subject: hw/cxl/mbox: Add support for background operations Support background commands in the mailbox, and update cmd_infostat_bg_op_sts() accordingly. This patch does not implement mbox interrupts upon completion, so the kernel driver must rely on polling to know when the operation is done. Signed-off-by: Davidlohr Bueso Signed-off-by: Jonathan Cameron Message-Id: <20231023160806.13206-12-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-device-utils.c | 24 ++++++++++-- hw/cxl/cxl-mailbox-utils.c | 92 ++++++++++++++++++++++++++++++++++++++++++++- include/hw/cxl/cxl_device.h | 10 +++++ 3 files changed, 122 insertions(+), 4 deletions(-) diff --git a/hw/cxl/cxl-device-utils.c b/hw/cxl/cxl-device-utils.c index f8938678c7..51466a626b 100644 --- a/hw/cxl/cxl-device-utils.c +++ b/hw/cxl/cxl-device-utils.c @@ -82,6 +82,25 @@ static uint64_t mailbox_reg_read(void *opaque, hwaddr offset, unsigned size) case 4: return cxl_dstate->mbox_reg_state32[offset / size]; case 8: + if (offset == A_CXL_DEV_BG_CMD_STS) { + uint64_t bg_status_reg; + bg_status_reg = FIELD_DP64(0, CXL_DEV_BG_CMD_STS, OP, + cci->bg.opcode); + bg_status_reg = FIELD_DP64(bg_status_reg, CXL_DEV_BG_CMD_STS, + PERCENTAGE_COMP, cci->bg.complete_pct); + bg_status_reg = FIELD_DP64(bg_status_reg, CXL_DEV_BG_CMD_STS, + RET_CODE, cci->bg.ret_code); + /* endian? */ + cxl_dstate->mbox_reg_state64[offset / size] = bg_status_reg; + } + if (offset == A_CXL_DEV_MAILBOX_STS) { + uint64_t status_reg = cxl_dstate->mbox_reg_state64[offset / size]; + if (cci->bg.complete_pct) { + status_reg = FIELD_DP64(status_reg, CXL_DEV_MAILBOX_STS, BG_OP, + 0); + cxl_dstate->mbox_reg_state64[offset / size] = status_reg; + } + } return cxl_dstate->mbox_reg_state64[offset / size]; default: g_assert_not_reached(); @@ -114,8 +133,7 @@ static void mailbox_mem_writeq(uint64_t *reg_state, hwaddr offset, case A_CXL_DEV_MAILBOX_CMD: break; case A_CXL_DEV_BG_CMD_STS: - /* BG not supported */ - /* fallthrough */ + break; case A_CXL_DEV_MAILBOX_STS: /* Read only register, will get updated by the state machine */ return; @@ -339,7 +357,7 @@ static void device_reg_init_common(CXLDeviceState *cxl_dstate) static void mailbox_reg_init_common(CXLDeviceState *cxl_dstate) { - /* 2048 payload size, with no interrupt or background support */ + /* 2048 payload size, with no interrupt */ ARRAY_FIELD_DP32(cxl_dstate->mbox_reg_state32, CXL_DEV_MAILBOX_CAP, PAYLOAD_SIZE, CXL_MAILBOX_PAYLOAD_SHIFT); cxl_dstate->payload_size = CXL_MAILBOX_MAX_PAYLOAD_SIZE; diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c index de63fc1a28..f1226f8f39 100644 --- a/hw/cxl/cxl-mailbox-utils.c +++ b/hw/cxl/cxl-mailbox-utils.c @@ -963,6 +963,8 @@ static CXLRetCode cmd_media_clear_poison(const struct cxl_cmd *cmd, #define IMMEDIATE_DATA_CHANGE (1 << 2) #define IMMEDIATE_POLICY_CHANGE (1 << 3) #define IMMEDIATE_LOG_CHANGE (1 << 4) +#define SECURITY_STATE_CHANGE (1 << 5) +#define BACKGROUND_OPERATION (1 << 6) static const struct cxl_cmd cxl_cmd_set[256][256] = { [EVENTS][GET_RECORDS] = { "EVENTS_GET_RECORDS", @@ -1011,10 +1013,19 @@ static const struct cxl_cmd cxl_cmd_set_sw[256][256] = { cmd_get_physical_port_state, ~0, 0 }, }; +/* + * While the command is executing in the background, the device should + * update the percentage complete in the Background Command Status Register + * at least once per second. + */ + +#define CXL_MBOX_BG_UPDATE_FREQ 1000UL + int cxl_process_cci_message(CXLCCI *cci, uint8_t set, uint8_t cmd, size_t len_in, uint8_t *pl_in, size_t *len_out, uint8_t *pl_out, bool *bg_started) { + int ret; const struct cxl_cmd *cxl_cmd; opcode_handler h; @@ -1031,7 +1042,81 @@ int cxl_process_cci_message(CXLCCI *cci, uint8_t set, uint8_t cmd, return CXL_MBOX_INVALID_PAYLOAD_LENGTH; } - return (*h)(cxl_cmd, pl_in, len_in, pl_out, len_out, cci); + /* Only one bg command at a time */ + if ((cxl_cmd->effect & BACKGROUND_OPERATION) && + cci->bg.runtime > 0) { + return CXL_MBOX_BUSY; + } + + ret = (*h)(cxl_cmd, pl_in, len_in, pl_out, len_out, cci); + if ((cxl_cmd->effect & BACKGROUND_OPERATION) && + ret == CXL_MBOX_BG_STARTED) { + *bg_started = true; + } else { + *bg_started = false; + } + + /* Set bg and the return code */ + if (*bg_started) { + uint64_t now; + + cci->bg.opcode = (set << 8) | cmd; + + cci->bg.complete_pct = 0; + cci->bg.ret_code = 0; + + now = qemu_clock_get_ms(QEMU_CLOCK_VIRTUAL); + cci->bg.starttime = now; + timer_mod(cci->bg.timer, now + CXL_MBOX_BG_UPDATE_FREQ); + } + + return ret; +} + +static void bg_timercb(void *opaque) +{ + CXLCCI *cci = opaque; + CXLDeviceState *cxl_dstate = &CXL_TYPE3(cci->d)->cxl_dstate; + uint64_t bg_status_reg = 0; + uint64_t now = qemu_clock_get_ms(QEMU_CLOCK_VIRTUAL); + uint64_t total_time = cci->bg.starttime + cci->bg.runtime; + + assert(cci->bg.runtime > 0); + bg_status_reg = FIELD_DP64(bg_status_reg, CXL_DEV_BG_CMD_STS, + OP, cci->bg.opcode); + + if (now >= total_time) { /* we are done */ + uint64_t status_reg; + uint16_t ret = CXL_MBOX_SUCCESS; + + cci->bg.complete_pct = 100; + /* Clear bg */ + status_reg = FIELD_DP64(0, CXL_DEV_MAILBOX_STS, BG_OP, 0); + cxl_dstate->mbox_reg_state64[R_CXL_DEV_MAILBOX_STS] = status_reg; + + bg_status_reg = FIELD_DP64(bg_status_reg, CXL_DEV_BG_CMD_STS, + RET_CODE, ret); + + /* TODO add ad-hoc cmd succesful completion handling */ + + qemu_log("Background command %04xh finished: %s\n", + cci->bg.opcode, + ret == CXL_MBOX_SUCCESS ? "success" : "aborted"); + } else { + /* estimate only */ + cci->bg.complete_pct = 100 * now / total_time; + timer_mod(cci->bg.timer, now + CXL_MBOX_BG_UPDATE_FREQ); + } + + bg_status_reg = FIELD_DP64(bg_status_reg, CXL_DEV_BG_CMD_STS, + PERCENTAGE_COMP, cci->bg.complete_pct); + cxl_dstate->mbox_reg_state64[R_CXL_DEV_BG_CMD_STS] = bg_status_reg; + + if (cci->bg.complete_pct == 100) { + cci->bg.starttime = 0; + /* registers are updated, allow new bg-capable cmds */ + cci->bg.runtime = 0; + } } void cxl_init_cci(CXLCCI *cci, size_t payload_max) @@ -1050,6 +1135,11 @@ void cxl_init_cci(CXLCCI *cci, size_t payload_max) } } } + cci->bg.complete_pct = 0; + cci->bg.starttime = 0; + cci->bg.runtime = 0; + cci->bg.timer = timer_new_ms(QEMU_CLOCK_VIRTUAL, + bg_timercb, cci); } void cxl_initialize_mailbox_swcci(CXLCCI *cci, DeviceState *intf, diff --git a/include/hw/cxl/cxl_device.h b/include/hw/cxl/cxl_device.h index fa73ed03e5..124ff969ec 100644 --- a/include/hw/cxl/cxl_device.h +++ b/include/hw/cxl/cxl_device.h @@ -149,6 +149,16 @@ typedef struct CXLCCI { } cel_log[1 << 16]; size_t cel_size; + /* background command handling (times in ms) */ + struct { + uint16_t opcode; + uint16_t complete_pct; + uint16_t ret_code; /* Current value of retcode */ + uint64_t starttime; + /* set by each bg cmd, cleared by the bg_timer when complete */ + uint64_t runtime; + QEMUTimer *timer; + } bg; size_t payload_max; /* Pointer to device hosting the CCI */ DeviceState *d; -- cgit v1.2.3 From 43efb0bfad2b81b32fc19da442cbea2835cf38d4 Mon Sep 17 00:00:00 2001 From: Davidlohr Bueso Date: Mon, 23 Oct 2023 17:08:01 +0100 Subject: hw/cxl/mbox: Wire up interrupts for background completion Notify when the background operation is done. Note that for now background commands are only supported on the main Type 3 mailbox. Signed-off-by: Davidlohr Bueso Signed-off-by: Jonathan Cameron Message-Id: <20231023160806.13206-13-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-device-utils.c | 10 +++++++++- hw/cxl/cxl-mailbox-utils.c | 31 ++++++++++++++----------------- include/hw/cxl/cxl_device.h | 1 + 3 files changed, 24 insertions(+), 18 deletions(-) diff --git a/hw/cxl/cxl-device-utils.c b/hw/cxl/cxl-device-utils.c index 51466a626b..61a3c4dc2e 100644 --- a/hw/cxl/cxl-device-utils.c +++ b/hw/cxl/cxl-device-utils.c @@ -357,10 +357,18 @@ static void device_reg_init_common(CXLDeviceState *cxl_dstate) static void mailbox_reg_init_common(CXLDeviceState *cxl_dstate) { - /* 2048 payload size, with no interrupt */ + const uint8_t msi_n = 9; + + /* 2048 payload size */ ARRAY_FIELD_DP32(cxl_dstate->mbox_reg_state32, CXL_DEV_MAILBOX_CAP, PAYLOAD_SIZE, CXL_MAILBOX_PAYLOAD_SHIFT); cxl_dstate->payload_size = CXL_MAILBOX_MAX_PAYLOAD_SIZE; + /* irq support */ + ARRAY_FIELD_DP32(cxl_dstate->mbox_reg_state32, CXL_DEV_MAILBOX_CAP, + BG_INT_CAP, 1); + ARRAY_FIELD_DP32(cxl_dstate->mbox_reg_state32, CXL_DEV_MAILBOX_CAP, + MSI_N, msi_n); + cxl_dstate->mbox_msi_n = msi_n; } static void memdev_reg_init_common(CXLDeviceState *cxl_dstate) { } diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c index f1226f8f39..f3fd97deb5 100644 --- a/hw/cxl/cxl-mailbox-utils.c +++ b/hw/cxl/cxl-mailbox-utils.c @@ -8,6 +8,8 @@ */ #include "qemu/osdep.h" +#include "hw/pci/msi.h" +#include "hw/pci/msix.h" #include "hw/cxl/cxl.h" #include "hw/cxl/cxl_events.h" #include "hw/pci/pci.h" @@ -1076,28 +1078,16 @@ int cxl_process_cci_message(CXLCCI *cci, uint8_t set, uint8_t cmd, static void bg_timercb(void *opaque) { CXLCCI *cci = opaque; - CXLDeviceState *cxl_dstate = &CXL_TYPE3(cci->d)->cxl_dstate; - uint64_t bg_status_reg = 0; uint64_t now = qemu_clock_get_ms(QEMU_CLOCK_VIRTUAL); uint64_t total_time = cci->bg.starttime + cci->bg.runtime; assert(cci->bg.runtime > 0); - bg_status_reg = FIELD_DP64(bg_status_reg, CXL_DEV_BG_CMD_STS, - OP, cci->bg.opcode); if (now >= total_time) { /* we are done */ - uint64_t status_reg; uint16_t ret = CXL_MBOX_SUCCESS; cci->bg.complete_pct = 100; - /* Clear bg */ - status_reg = FIELD_DP64(0, CXL_DEV_MAILBOX_STS, BG_OP, 0); - cxl_dstate->mbox_reg_state64[R_CXL_DEV_MAILBOX_STS] = status_reg; - - bg_status_reg = FIELD_DP64(bg_status_reg, CXL_DEV_BG_CMD_STS, - RET_CODE, ret); - - /* TODO add ad-hoc cmd succesful completion handling */ + cci->bg.ret_code = ret; qemu_log("Background command %04xh finished: %s\n", cci->bg.opcode, @@ -1108,14 +1098,21 @@ static void bg_timercb(void *opaque) timer_mod(cci->bg.timer, now + CXL_MBOX_BG_UPDATE_FREQ); } - bg_status_reg = FIELD_DP64(bg_status_reg, CXL_DEV_BG_CMD_STS, - PERCENTAGE_COMP, cci->bg.complete_pct); - cxl_dstate->mbox_reg_state64[R_CXL_DEV_BG_CMD_STS] = bg_status_reg; - if (cci->bg.complete_pct == 100) { + /* TODO: generalize to switch CCI */ + CXLType3Dev *ct3d = CXL_TYPE3(cci->d); + CXLDeviceState *cxl_dstate = &ct3d->cxl_dstate; + PCIDevice *pdev = PCI_DEVICE(cci->d); + cci->bg.starttime = 0; /* registers are updated, allow new bg-capable cmds */ cci->bg.runtime = 0; + + if (msix_enabled(pdev)) { + msix_notify(pdev, cxl_dstate->mbox_msi_n); + } else if (msi_enabled(pdev)) { + msi_notify(pdev, cxl_dstate->mbox_msi_n); + } } } diff --git a/include/hw/cxl/cxl_device.h b/include/hw/cxl/cxl_device.h index 124ff969ec..2a813cdddd 100644 --- a/include/hw/cxl/cxl_device.h +++ b/include/hw/cxl/cxl_device.h @@ -193,6 +193,7 @@ typedef struct cxl_device_state { struct { MemoryRegion mailbox; uint16_t payload_size; + uint8_t mbox_msi_n; union { uint8_t mbox_reg_state[CXL_MAILBOX_REGISTERS_LENGTH]; uint16_t mbox_reg_state16[CXL_MAILBOX_REGISTERS_LENGTH / 2]; -- cgit v1.2.3 From 25a52959f99d6860a186175bda898e3bdb605f91 Mon Sep 17 00:00:00 2001 From: Davidlohr Bueso Date: Mon, 23 Oct 2023 17:08:02 +0100 Subject: hw/cxl: Add support for device sanitation Make use of the background operations through the sanitize command, per CXL 3.0 specs. Traditionally run times can be rather long, depending on the size of the media. Estimate times based on: https://pmem.io/documents/NVDIMM_DSM_Interface-V1.8.pdf Signed-off-by: Davidlohr Bueso Signed-off-by: Jonathan Cameron Message-Id: <20231023160806.13206-14-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-mailbox-utils.c | 140 ++++++++++++++++++++++++++++++++++++++++++++ hw/mem/cxl_type3.c | 10 ++++ include/hw/cxl/cxl_device.h | 17 ++++++ 3 files changed, 167 insertions(+) diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c index f3fd97deb5..2463f239af 100644 --- a/hw/cxl/cxl-mailbox-utils.c +++ b/hw/cxl/cxl-mailbox-utils.c @@ -18,6 +18,7 @@ #include "qemu/log.h" #include "qemu/units.h" #include "qemu/uuid.h" +#include "sysemu/hostmem.h" #define CXL_CAPACITY_MULTIPLIER (256 * MiB) @@ -68,6 +69,9 @@ enum { #define GET_PARTITION_INFO 0x0 #define GET_LSA 0x2 #define SET_LSA 0x3 + SANITIZE = 0x44, + #define OVERWRITE 0x0 + #define SECURE_ERASE 0x1 MEDIA_AND_POISON = 0x43, #define GET_POISON_LIST 0x0 #define INJECT_POISON 0x1 @@ -749,6 +753,108 @@ static CXLRetCode cmd_ccls_set_lsa(const struct cxl_cmd *cmd, return CXL_MBOX_SUCCESS; } +/* Perform the actual device zeroing */ +static void __do_sanitization(CXLType3Dev *ct3d) +{ + MemoryRegion *mr; + + if (ct3d->hostvmem) { + mr = host_memory_backend_get_memory(ct3d->hostvmem); + if (mr) { + void *hostmem = memory_region_get_ram_ptr(mr); + memset(hostmem, 0, memory_region_size(mr)); + } + } + + if (ct3d->hostpmem) { + mr = host_memory_backend_get_memory(ct3d->hostpmem); + if (mr) { + void *hostmem = memory_region_get_ram_ptr(mr); + memset(hostmem, 0, memory_region_size(mr)); + } + } + if (ct3d->lsa) { + mr = host_memory_backend_get_memory(ct3d->lsa); + if (mr) { + void *lsa = memory_region_get_ram_ptr(mr); + memset(lsa, 0, memory_region_size(mr)); + } + } +} + +/* + * CXL 3.0 spec section 8.2.9.8.5.1 - Sanitize. + * + * Once the Sanitize command has started successfully, the device shall be + * placed in the media disabled state. If the command fails or is interrupted + * by a reset or power failure, it shall remain in the media disabled state + * until a successful Sanitize command has been completed. During this state: + * + * 1. Memory writes to the device will have no effect, and all memory reads + * will return random values (no user data returned, even for locations that + * the failed Sanitize operation didn’t sanitize yet). + * + * 2. Mailbox commands shall still be processed in the disabled state, except + * that commands that access Sanitized areas shall fail with the Media Disabled + * error code. + */ +static CXLRetCode cmd_sanitize_overwrite(const struct cxl_cmd *cmd, + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLCCI *cci) +{ + CXLType3Dev *ct3d = CXL_TYPE3(cci->d); + uint64_t total_mem; /* in Mb */ + int secs; + + total_mem = (ct3d->cxl_dstate.vmem_size + ct3d->cxl_dstate.pmem_size) >> 20; + if (total_mem <= 512) { + secs = 4; + } else if (total_mem <= 1024) { + secs = 8; + } else if (total_mem <= 2 * 1024) { + secs = 15; + } else if (total_mem <= 4 * 1024) { + secs = 30; + } else if (total_mem <= 8 * 1024) { + secs = 60; + } else if (total_mem <= 16 * 1024) { + secs = 2 * 60; + } else if (total_mem <= 32 * 1024) { + secs = 4 * 60; + } else if (total_mem <= 64 * 1024) { + secs = 8 * 60; + } else if (total_mem <= 128 * 1024) { + secs = 15 * 60; + } else if (total_mem <= 256 * 1024) { + secs = 30 * 60; + } else if (total_mem <= 512 * 1024) { + secs = 60 * 60; + } else if (total_mem <= 1024 * 1024) { + secs = 120 * 60; + } else { + secs = 240 * 60; /* max 4 hrs */ + } + + /* EBUSY other bg cmds as of now */ + cci->bg.runtime = secs * 1000UL; + *len_out = 0; + + cxl_dev_disable_media(&ct3d->cxl_dstate); + + if (secs > 2) { + /* sanitize when done */ + return CXL_MBOX_BG_STARTED; + } else { + __do_sanitization(ct3d); + cxl_dev_enable_media(&ct3d->cxl_dstate); + + return CXL_MBOX_SUCCESS; + } +} + /* * This is very inefficient, but good enough for now! * Also the payload will always fit, so no need to handle the MORE flag and @@ -993,6 +1099,8 @@ static const struct cxl_cmd cxl_cmd_set[256][256] = { [CCLS][GET_LSA] = { "CCLS_GET_LSA", cmd_ccls_get_lsa, 8, 0 }, [CCLS][SET_LSA] = { "CCLS_SET_LSA", cmd_ccls_set_lsa, ~0, IMMEDIATE_CONFIG_CHANGE | IMMEDIATE_DATA_CHANGE }, + [SANITIZE][OVERWRITE] = { "SANITIZE_OVERWRITE", cmd_sanitize_overwrite, 0, + IMMEDIATE_DATA_CHANGE | SECURITY_STATE_CHANGE | BACKGROUND_OPERATION }, [MEDIA_AND_POISON][GET_POISON_LIST] = { "MEDIA_AND_POISON_GET_POISON_LIST", cmd_media_get_poison_list, 16, 0 }, [MEDIA_AND_POISON][INJECT_POISON] = { "MEDIA_AND_POISON_INJECT_POISON", @@ -1050,6 +1158,21 @@ int cxl_process_cci_message(CXLCCI *cci, uint8_t set, uint8_t cmd, return CXL_MBOX_BUSY; } + /* forbid any selected commands while overwriting */ + if (sanitize_running(cci)) { + if (h == cmd_events_get_records || + h == cmd_ccls_get_partition_info || + h == cmd_ccls_set_lsa || + h == cmd_ccls_get_lsa || + h == cmd_logs_get_log || + h == cmd_media_get_poison_list || + h == cmd_media_inject_poison || + h == cmd_media_clear_poison || + h == cmd_sanitize_overwrite) { + return CXL_MBOX_MEDIA_DISABLED; + } + } + ret = (*h)(cxl_cmd, pl_in, len_in, pl_out, len_out, cci); if ((cxl_cmd->effect & BACKGROUND_OPERATION) && ret == CXL_MBOX_BG_STARTED) { @@ -1088,6 +1211,23 @@ static void bg_timercb(void *opaque) cci->bg.complete_pct = 100; cci->bg.ret_code = ret; + if (ret == CXL_MBOX_SUCCESS) { + switch (cci->bg.opcode) { + case 0x4400: /* sanitize */ + { + CXLType3Dev *ct3d = CXL_TYPE3(cci->d); + + __do_sanitization(ct3d); + cxl_dev_enable_media(&ct3d->cxl_dstate); + } + break; + case 0x4304: /* TODO: scan media */ + break; + default: + __builtin_unreachable(); + break; + } + } qemu_log("Background command %04xh finished: %s\n", cci->bg.opcode, diff --git a/hw/mem/cxl_type3.c b/hw/mem/cxl_type3.c index 0529745786..cc8220592f 100644 --- a/hw/mem/cxl_type3.c +++ b/hw/mem/cxl_type3.c @@ -23,6 +23,7 @@ #include "qemu/pmem.h" #include "qemu/range.h" #include "qemu/rcu.h" +#include "qemu/guest-random.h" #include "sysemu/hostmem.h" #include "sysemu/numa.h" #include "hw/cxl/cxl.h" @@ -897,6 +898,11 @@ MemTxResult cxl_type3_read(PCIDevice *d, hwaddr host_addr, uint64_t *data, return MEMTX_ERROR; } + if (sanitize_running(&CXL_TYPE3(d)->cci)) { + qemu_guest_getrandom_nofail(data, size); + return MEMTX_OK; + } + return address_space_read(as, dpa_offset, attrs, data, size); } @@ -913,6 +919,10 @@ MemTxResult cxl_type3_write(PCIDevice *d, hwaddr host_addr, uint64_t data, return MEMTX_ERROR; } + if (sanitize_running(&CXL_TYPE3(d)->cci)) { + return MEMTX_OK; + } + return address_space_write(as, dpa_offset, attrs, &data, size); } diff --git a/include/hw/cxl/cxl_device.h b/include/hw/cxl/cxl_device.h index 2a813cdddd..70aca9024c 100644 --- a/include/hw/cxl/cxl_device.h +++ b/include/hw/cxl/cxl_device.h @@ -343,6 +343,23 @@ REG64(CXL_MEM_DEV_STS, 0) FIELD(CXL_MEM_DEV_STS, MBOX_READY, 4, 1) FIELD(CXL_MEM_DEV_STS, RESET_NEEDED, 5, 3) +static inline void __toggle_media(CXLDeviceState *cxl_dstate, int val) +{ + uint64_t dev_status_reg; + + dev_status_reg = FIELD_DP64(0, CXL_MEM_DEV_STS, MEDIA_STATUS, val); + cxl_dstate->mbox_reg_state64[R_CXL_MEM_DEV_STS] = dev_status_reg; +} +#define cxl_dev_disable_media(cxlds) \ + do { __toggle_media((cxlds), 0x3); } while (0) +#define cxl_dev_enable_media(cxlds) \ + do { __toggle_media((cxlds), 0x1); } while (0) + +static inline bool sanitize_running(CXLCCI *cci) +{ + return !!cci->bg.runtime && cci->bg.opcode == 0x4400; +} + typedef struct CXLError { QTAILQ_ENTRY(CXLError) node; int type; /* Error code as per FE definition */ -- cgit v1.2.3 From 9dd15ab6e6b35db6ff166e5016b0c2042c5439c6 Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 17:08:03 +0100 Subject: hw/cxl/mbox: Add Get Background Operation Status Command For now, provide this command on type 3 main mailbox only. Signed-off-by: Jonathan Cameron Message-Id: <20231023160806.13206-15-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-mailbox-utils.c | 33 +++++++++++++++++++++++++++++++++ 1 file changed, 33 insertions(+) diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c index 2463f239af..2b78136588 100644 --- a/hw/cxl/cxl-mailbox-utils.c +++ b/hw/cxl/cxl-mailbox-utils.c @@ -50,6 +50,7 @@ enum { INFOSTAT = 0x00, #define IS_IDENTIFY 0x1 + #define BACKGROUND_OPERATION_STATUS 0x2 EVENTS = 0x01, #define GET_RECORDS 0x0 #define CLEAR_RECORDS 0x1 @@ -452,6 +453,36 @@ static CXLRetCode cmd_get_physical_port_state(const struct cxl_cmd *cmd, return CXL_MBOX_SUCCESS; } +/* CXL r3.0 8.2.9.1.2 */ +static CXLRetCode cmd_infostat_bg_op_sts(const struct cxl_cmd *cmd, + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLCCI *cci) +{ + struct { + uint8_t status; + uint8_t rsvd; + uint16_t opcode; + uint16_t returncode; + uint16_t vendor_ext_status; + } QEMU_PACKED *bg_op_status; + QEMU_BUILD_BUG_ON(sizeof(*bg_op_status) != 8); + + bg_op_status = (void *)payload_out; + memset(bg_op_status, 0, sizeof(*bg_op_status)); + bg_op_status->status = cci->bg.complete_pct << 1; + if (cci->bg.runtime > 0) { + bg_op_status->status |= 1U << 0; + } + bg_op_status->opcode = cci->bg.opcode; + bg_op_status->returncode = cci->bg.ret_code; + *len_out = sizeof(*bg_op_status); + + return CXL_MBOX_SUCCESS; +} + /* 8.2.9.2.1 */ static CXLRetCode cmd_firmware_update_get_info(const struct cxl_cmd *cmd, uint8_t *payload_in, @@ -1111,6 +1142,8 @@ static const struct cxl_cmd cxl_cmd_set[256][256] = { static const struct cxl_cmd cxl_cmd_set_sw[256][256] = { [INFOSTAT][IS_IDENTIFY] = { "IDENTIFY", cmd_infostat_identify, 0, 0 }, + [INFOSTAT][BACKGROUND_OPERATION_STATUS] = { "BACKGROUND_OPERATION_STATUS", + cmd_infostat_bg_op_sts, 0, 0 }, [TIMESTAMP][GET] = { "TIMESTAMP_GET", cmd_timestamp_get, 0, 0 }, [TIMESTAMP][SET] = { "TIMESTAMP_SET", cmd_timestamp_set, 0, IMMEDIATE_POLICY_CHANGE }, -- cgit v1.2.3 From ede604d505d711e45abcc37dd4389967fda410d7 Mon Sep 17 00:00:00 2001 From: Gregory Price Date: Mon, 23 Oct 2023 17:08:04 +0100 Subject: hw/cxl/type3: Cleanup multiple CXL_TYPE3() calls in read/write functions Call CXL_TYPE3 once at top of function to avoid multiple invocations. Signed-off-by: Gregory Price Signed-off-by: Jonathan Cameron Message-Id: <20231023160806.13206-16-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/mem/cxl_type3.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/hw/mem/cxl_type3.c b/hw/mem/cxl_type3.c index cc8220592f..a766c64575 100644 --- a/hw/mem/cxl_type3.c +++ b/hw/mem/cxl_type3.c @@ -888,17 +888,18 @@ static int cxl_type3_hpa_to_as_and_dpa(CXLType3Dev *ct3d, MemTxResult cxl_type3_read(PCIDevice *d, hwaddr host_addr, uint64_t *data, unsigned size, MemTxAttrs attrs) { + CXLType3Dev *ct3d = CXL_TYPE3(d); uint64_t dpa_offset = 0; AddressSpace *as = NULL; int res; - res = cxl_type3_hpa_to_as_and_dpa(CXL_TYPE3(d), host_addr, size, + res = cxl_type3_hpa_to_as_and_dpa(ct3d, host_addr, size, &as, &dpa_offset); if (res) { return MEMTX_ERROR; } - if (sanitize_running(&CXL_TYPE3(d)->cci)) { + if (sanitize_running(&ct3d->cci)) { qemu_guest_getrandom_nofail(data, size); return MEMTX_OK; } @@ -909,17 +910,18 @@ MemTxResult cxl_type3_read(PCIDevice *d, hwaddr host_addr, uint64_t *data, MemTxResult cxl_type3_write(PCIDevice *d, hwaddr host_addr, uint64_t data, unsigned size, MemTxAttrs attrs) { + CXLType3Dev *ct3d = CXL_TYPE3(d); uint64_t dpa_offset = 0; AddressSpace *as = NULL; int res; - res = cxl_type3_hpa_to_as_and_dpa(CXL_TYPE3(d), host_addr, size, + res = cxl_type3_hpa_to_as_and_dpa(ct3d, host_addr, size, &as, &dpa_offset); if (res) { return MEMTX_ERROR; } - if (sanitize_running(&CXL_TYPE3(d)->cci)) { + if (sanitize_running(&ct3d->cci)) { return MEMTX_OK; } -- cgit v1.2.3 From 44e4b316e4bf8f7327f3917c25aae38172695680 Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 17:08:05 +0100 Subject: hw/cxl: Add dummy security state get Needed to allow the santize comamnds to be tested with proposed Linux Kernel support. Default value + no control of the security state will work for now. Signed-off-by: Jonathan Cameron Message-Id: <20231023160806.13206-17-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-mailbox-utils.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c index 2b78136588..693c2cbdcd 100644 --- a/hw/cxl/cxl-mailbox-utils.c +++ b/hw/cxl/cxl-mailbox-utils.c @@ -73,6 +73,8 @@ enum { SANITIZE = 0x44, #define OVERWRITE 0x0 #define SECURE_ERASE 0x1 + PERSISTENT_MEM = 0x45, + #define GET_SECURITY_STATE 0x0 MEDIA_AND_POISON = 0x43, #define GET_POISON_LIST 0x0 #define INJECT_POISON 0x1 @@ -886,6 +888,19 @@ static CXLRetCode cmd_sanitize_overwrite(const struct cxl_cmd *cmd, } } +static CXLRetCode cmd_get_security_state(const struct cxl_cmd *cmd, + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLCCI *cci) +{ + uint32_t *state = (uint32_t *)payload_out; + + *state = 0; + *len_out = 4; + return CXL_MBOX_SUCCESS; +} /* * This is very inefficient, but good enough for now! * Also the payload will always fit, so no need to handle the MORE flag and @@ -1132,6 +1147,8 @@ static const struct cxl_cmd cxl_cmd_set[256][256] = { ~0, IMMEDIATE_CONFIG_CHANGE | IMMEDIATE_DATA_CHANGE }, [SANITIZE][OVERWRITE] = { "SANITIZE_OVERWRITE", cmd_sanitize_overwrite, 0, IMMEDIATE_DATA_CHANGE | SECURITY_STATE_CHANGE | BACKGROUND_OPERATION }, + [PERSISTENT_MEM][GET_SECURITY_STATE] = { "GET_SECURITY_STATE", + cmd_get_security_state, 0, 0 }, [MEDIA_AND_POISON][GET_POISON_LIST] = { "MEDIA_AND_POISON_GET_POISON_LIST", cmd_media_get_poison_list, 16, 0 }, [MEDIA_AND_POISON][INJECT_POISON] = { "MEDIA_AND_POISON_INJECT_POISON", -- cgit v1.2.3 From 004e3a93b814ca2d13ee2feb1f9ebacef6c83b9e Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Mon, 23 Oct 2023 17:08:06 +0100 Subject: hw/cxl: Add tunneled command support to mailbox for switch cci. This implementation of tunneling makes the choice that our Type 3 device is a Logical Device (LD) of a Multi-Logical Device (MLD) that just happens to only have one LD for now. Tunneling is supported from a Switch Mailbox CCI (and shortly via MCTP over I2C connected to the switch MCTP CCI) via an outer level to the FM owned LD in the MLD Type 3 device. From there an inner tunnel may be used to access particular LDs. Protocol wise, the following is what happens in a real system but we don't emulate the transports - just the destinations and the payloads. ( Host -> Switch Mailbox CCI - in band FM-API mailbox command or Host -> Switch MCTP CCI - MCTP over I2C using the CXL FM-API MCTP Binding. ) then (if a tunnel command) Switch -> Type 3 FM Owned LD - MCTP over PCI VDM using the CXL FM-API binding (addressed by switch port) then (if unwrapped command also a tunnel command) Type 3 FM Owned LD to LD0 via internal transport (addressed by LD number) or (added shortly) Host to Type 3 FM Owned MCTP CCI - MCTP over I2C using the CXL FM-API MCTP Binding. then (if unwrapped comand is a tunnel comamnd) Type 3 FM Owned LD to LD0 via internal transport. (addressed by LD number) It is worth noting that the tunneling commands over PCI VDM presumably use the appropriate MCTP binding depending on opcode. This may be the CXL FMAPI binding or the CXL Memory Device Binding. Additional commands will need to be added to make this useful beyond testing the tunneling works. Signed-off-by: Jonathan Cameron Message-Id: <20231023160806.13206-18-Jonathan.Cameron@huawei.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/cxl/cxl-mailbox-utils.c | 162 ++++++++++++++++++++++++++++++++++++++++++++ hw/mem/cxl_type3.c | 11 +++ include/hw/cxl/cxl_device.h | 9 +++ 3 files changed, 182 insertions(+) diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c index 693c2cbdcd..b365575097 100644 --- a/hw/cxl/cxl-mailbox-utils.c +++ b/hw/cxl/cxl-mailbox-utils.c @@ -82,8 +82,132 @@ enum { PHYSICAL_SWITCH = 0x51, #define IDENTIFY_SWITCH_DEVICE 0x0 #define GET_PHYSICAL_PORT_STATE 0x1 + TUNNEL = 0x53, + #define MANAGEMENT_COMMAND 0x0 }; +/* CCI Message Format CXL r3.0 Figure 7-19 */ +typedef struct CXLCCIMessage { + uint8_t category; +#define CXL_CCI_CAT_REQ 0 +#define CXL_CCI_CAT_RSP 1 + uint8_t tag; + uint8_t resv1; + uint8_t command; + uint8_t command_set; + uint8_t pl_length[3]; + uint16_t rc; + uint16_t vendor_specific; + uint8_t payload[]; +} QEMU_PACKED CXLCCIMessage; + +/* This command is only defined to an MLD FM Owned LD or an MHD */ +static CXLRetCode cmd_tunnel_management_cmd(const struct cxl_cmd *cmd, + uint8_t *payload_in, + size_t len_in, + uint8_t *payload_out, + size_t *len_out, + CXLCCI *cci) +{ + PCIDevice *tunnel_target; + CXLCCI *target_cci; + struct { + uint8_t port_or_ld_id; + uint8_t target_type; + uint16_t size; + CXLCCIMessage ccimessage; + } QEMU_PACKED *in; + struct { + uint16_t resp_len; + uint8_t resv[2]; + CXLCCIMessage ccimessage; + } QEMU_PACKED *out; + size_t pl_length, length_out; + bool bg_started; + int rc; + + if (cmd->in < sizeof(*in)) { + return CXL_MBOX_INVALID_INPUT; + } + in = (void *)payload_in; + out = (void *)payload_out; + + /* Enough room for minimum sized message - no payload */ + if (in->size < sizeof(in->ccimessage)) { + return CXL_MBOX_INVALID_PAYLOAD_LENGTH; + } + /* Length of input payload should be in->size + a wrapping tunnel header */ + if (in->size != len_in - offsetof(typeof(*out), ccimessage)) { + return CXL_MBOX_INVALID_PAYLOAD_LENGTH; + } + if (in->ccimessage.category != CXL_CCI_CAT_REQ) { + return CXL_MBOX_INVALID_INPUT; + } + + if (in->target_type != 0) { + qemu_log_mask(LOG_UNIMP, + "Tunneled Command sent to non existent FM-LD"); + return CXL_MBOX_INVALID_INPUT; + } + + /* + * Target of a tunnel unfortunately depends on type of CCI readint + * the message. + * If in a switch, then it's the port number. + * If in an MLD it is the ld number. + * If in an MHD target type indicate where we are going. + */ + if (object_dynamic_cast(OBJECT(cci->d), TYPE_CXL_TYPE3)) { + CXLType3Dev *ct3d = CXL_TYPE3(cci->d); + if (in->port_or_ld_id != 0) { + /* Only pretending to have one for now! */ + return CXL_MBOX_INVALID_INPUT; + } + target_cci = &ct3d->ld0_cci; + } else if (object_dynamic_cast(OBJECT(cci->d), TYPE_CXL_USP)) { + CXLUpstreamPort *usp = CXL_USP(cci->d); + + tunnel_target = pcie_find_port_by_pn(&PCI_BRIDGE(usp)->sec_bus, + in->port_or_ld_id); + if (!tunnel_target) { + return CXL_MBOX_INVALID_INPUT; + } + tunnel_target = + pci_bridge_get_sec_bus(PCI_BRIDGE(tunnel_target))->devices[0]; + if (!tunnel_target) { + return CXL_MBOX_INVALID_INPUT; + } + if (object_dynamic_cast(OBJECT(tunnel_target), TYPE_CXL_TYPE3)) { + CXLType3Dev *ct3d = CXL_TYPE3(tunnel_target); + /* Tunneled VDMs always land on FM Owned LD */ + target_cci = &ct3d->vdm_fm_owned_ld_mctp_cci; + } else { + return CXL_MBOX_INVALID_INPUT; + } + } else { + return CXL_MBOX_INVALID_INPUT; + } + + pl_length = in->ccimessage.pl_length[2] << 16 | + in->ccimessage.pl_length[1] << 8 | in->ccimessage.pl_length[0]; + rc = cxl_process_cci_message(target_cci, + in->ccimessage.command_set, + in->ccimessage.command, + pl_length, in->ccimessage.payload, + &length_out, out->ccimessage.payload, + &bg_started); + /* Payload should be in place. Rest of CCI header and needs filling */ + out->resp_len = length_out + sizeof(CXLCCIMessage); + st24_le_p(out->ccimessage.pl_length, length_out); + out->ccimessage.rc = rc; + out->ccimessage.category = CXL_CCI_CAT_RSP; + out->ccimessage.command = in->ccimessage.command; + out->ccimessage.command_set = in->ccimessage.command_set; + out->ccimessage.tag = in->ccimessage.tag; + *len_out = length_out + sizeof(*out); + + return CXL_MBOX_SUCCESS; +} static CXLRetCode cmd_events_get_records(const struct cxl_cmd *cmd, uint8_t *payload_in, size_t len_in, @@ -1171,6 +1295,8 @@ static const struct cxl_cmd cxl_cmd_set_sw[256][256] = { cmd_identify_switch_device, 0, 0 }, [PHYSICAL_SWITCH][GET_PHYSICAL_PORT_STATE] = { "SWITCH_PHYSICAL_PORT_STATS", cmd_get_physical_port_state, ~0, 0 }, + [TUNNEL][MANAGEMENT_COMMAND] = { "TUNNEL_MANAGEMENT_COMMAND", + cmd_tunnel_management_cmd, ~0, 0 }, }; /* @@ -1347,3 +1473,39 @@ void cxl_initialize_mailbox_t3(CXLCCI *cci, DeviceState *d, size_t payload_max) cci->intf = d; cxl_init_cci(cci, payload_max); } + +static const struct cxl_cmd cxl_cmd_set_t3_ld[256][256] = { + [INFOSTAT][IS_IDENTIFY] = { "IDENTIFY", cmd_infostat_identify, 0, 0 }, + [LOGS][GET_SUPPORTED] = { "LOGS_GET_SUPPORTED", cmd_logs_get_supported, 0, + 0 }, + [LOGS][GET_LOG] = { "LOGS_GET_LOG", cmd_logs_get_log, 0x18, 0 }, +}; + +void cxl_initialize_t3_ld_cci(CXLCCI *cci, DeviceState *d, DeviceState *intf, + size_t payload_max) +{ + cci->cxl_cmd_set = cxl_cmd_set_t3_ld; + cci->d = d; + cci->intf = intf; + cxl_init_cci(cci, payload_max); +} + +static const struct cxl_cmd cxl_cmd_set_t3_fm_owned_ld_mctp[256][256] = { + [INFOSTAT][IS_IDENTIFY] = { "IDENTIFY", cmd_infostat_identify, 0, 0}, + [LOGS][GET_SUPPORTED] = { "LOGS_GET_SUPPORTED", cmd_logs_get_supported, 0, + 0 }, + [LOGS][GET_LOG] = { "LOGS_GET_LOG", cmd_logs_get_log, 0x18, 0 }, + [TIMESTAMP][GET] = { "TIMESTAMP_GET", cmd_timestamp_get, 0, 0 }, + [TUNNEL][MANAGEMENT_COMMAND] = { "TUNNEL_MANAGEMENT_COMMAND", + cmd_tunnel_management_cmd, ~0, 0 }, +}; + +void cxl_initialize_t3_fm_owned_ld_mctpcci(CXLCCI *cci, DeviceState *d, + DeviceState *intf, + size_t payload_max) +{ + cci->cxl_cmd_set = cxl_cmd_set_t3_fm_owned_ld_mctp; + cci->d = d; + cci->intf = intf; + cxl_init_cci(cci, payload_max); +} diff --git a/hw/mem/cxl_type3.c b/hw/mem/cxl_type3.c index a766c64575..52647b4ac7 100644 --- a/hw/mem/cxl_type3.c +++ b/hw/mem/cxl_type3.c @@ -936,6 +936,17 @@ static void ct3d_reset(DeviceState *dev) cxl_component_register_init_common(reg_state, write_msk, CXL2_TYPE3_DEVICE); cxl_device_register_init_t3(ct3d); + + /* + * Bring up an endpoint to target with MCTP over VDM. + * This device is emulating an MLD with single LD for now. + */ + cxl_initialize_t3_fm_owned_ld_mctpcci(&ct3d->vdm_fm_owned_ld_mctp_cci, + DEVICE(ct3d), DEVICE(ct3d), + 512); /* Max payload made up */ + cxl_initialize_t3_ld_cci(&ct3d->ld0_cci, DEVICE(ct3d), DEVICE(ct3d), + 512); /* Max payload made up */ + } static Property ct3_props[] = { diff --git a/include/hw/cxl/cxl_device.h b/include/hw/cxl/cxl_device.h index 70aca9024c..61b7f897f7 100644 --- a/include/hw/cxl/cxl_device.h +++ b/include/hw/cxl/cxl_device.h @@ -279,6 +279,12 @@ int cxl_process_cci_message(CXLCCI *cci, uint8_t set, uint8_t cmd, size_t len_in, uint8_t *pl_in, size_t *len_out, uint8_t *pl_out, bool *bg_started); +void cxl_initialize_t3_fm_owned_ld_mctpcci(CXLCCI *cci, DeviceState *d, + DeviceState *intf, + size_t payload_max); + +void cxl_initialize_t3_ld_cci(CXLCCI *cci, DeviceState *d, + DeviceState *intf, size_t payload_max); #define cxl_device_cap_init(dstate, reg, cap_id, ver) \ do { \ @@ -397,6 +403,9 @@ struct CXLType3Dev { CXLComponentState cxl_cstate; CXLDeviceState cxl_dstate; CXLCCI cci; /* Primary PCI mailbox CCI */ + /* Always intialized as no way to know if a VDM might show up */ + CXLCCI vdm_fm_owned_ld_mctp_cci; + CXLCCI ld0_cci; /* DOE */ DOECap doe_cdat; -- cgit v1.2.3 From a874ddc95a2b8b4b6303366c6ac6a2be53c41be4 Mon Sep 17 00:00:00 2001 From: Ani Sinha Date: Fri, 27 Oct 2023 08:51:18 +0530 Subject: acpi/tests/avocado/bits: enforce 32-bit SMBIOS entry point QEMU defaults to 64-bit entry point since the following commit bf376f3020 ("hw/i386/pc: Default to use SMBIOS 3.0 for newer machine models") The above change is applicable for all newer machine versions from version 8.1 and newer. i440fx and q35 machine versions 8.0 and older still use 32-bit entry points. Unfortunately, bits currently does not recognize 64-bit entry points and hence is not able to parse SMBIOS tables. Therefore, we need to enforce 32-bit SMBIOS entry point in QEMU command line so that bits is able to parse the SMBIOS tables. Once we implement the support in bits to parse 64-bit entry points, we can remove the extra command line that is passed to enforce a 32-bit entry point. The support can be added to the following smbios test script: tests/avocado/acpi-bits/bits-tests/smbios.py2 in QEMU repository. CC: jusual@redhat.com CC: imammedo@redhat.com Signed-off-by: Ani Sinha Message-Id: <20231027032120.6012-2-anisinha@redhat.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/avocado/acpi-bits.py | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/tests/avocado/acpi-bits.py b/tests/avocado/acpi-bits.py index eca13dc518..042007b0b8 100644 --- a/tests/avocado/acpi-bits.py +++ b/tests/avocado/acpi-bits.py @@ -380,6 +380,11 @@ class AcpiBitsTest(QemuBaseTest): #pylint: disable=too-many-instance-attributes # consistent in terms of timing. smilatency tests have consistent # timing requirements. self._vm.add_args('-icount', 'auto') + # currently there is no support in bits for recognizing 64-bit SMBIOS + # entry points. QEMU defaults to 64-bit entry points since the + # upstream commit bf376f3020 ("hw/i386/pc: Default to use SMBIOS 3.0 + # for newer machine models"). Therefore, enforce 32-bit entry point. + self._vm.add_args('-machine', 'smbios-entry-point-type=32') args = " ".join(str(arg) for arg in self._vm.base_args()) + \ " " + " ".join(str(arg) for arg in self._vm.args) -- cgit v1.2.3 From 94cd94f1c0137b56000c01208e03d0907ad34910 Mon Sep 17 00:00:00 2001 From: Ani Sinha Date: Fri, 27 Oct 2023 08:51:19 +0530 Subject: acpi/tests/avocado/bits: enable console logging from bits VM Console logs from the VM can be useful for debugging when things go wrong. Other avocado tests enables them. This change enables console logging with the following changes: - point to the newer bios bits image that actually enabled VM console. - change the bits test to drain the console logs from the VM and write the logs. - wait for SHUTDOWN event from QEMU so that console logs can be drained out of the socket before it is closed as a part of vm.wait(). Additionally, following two cosmetic changes have been made: - Removed VM QEMU command line logging as avocado framework already logs it. This is a minor cleanup along the way. - Update my email to my work email in the avocado acpi bios bits test. CC: jsnow@redhat.com Signed-off-by: Ani Sinha Message-Id: <20231027032120.6012-3-anisinha@redhat.com> Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- tests/avocado/acpi-bits.py | 26 ++++++++++++++++---------- 1 file changed, 16 insertions(+), 10 deletions(-) diff --git a/tests/avocado/acpi-bits.py b/tests/avocado/acpi-bits.py index 042007b0b8..68b9e98d4e 100644 --- a/tests/avocado/acpi-bits.py +++ b/tests/avocado/acpi-bits.py @@ -18,7 +18,7 @@ # # # Author: -# Ani Sinha +# Ani Sinha # pylint: disable=invalid-name # pylint: disable=consider-using-f-string @@ -48,6 +48,7 @@ from typing import ( ) from qemu.machine import QEMUMachine from avocado import skipIf +from avocado.utils import datadrainer as drainer from avocado_qemu import QemuBaseTest deps = ["xorriso", "mformat"] # dependent tools needed in the test setup/box. @@ -141,12 +142,12 @@ class AcpiBitsTest(QemuBaseTest): #pylint: disable=too-many-instance-attributes self._baseDir = None # following are some standard configuration constants - self._bitsInternalVer = 2020 - self._bitsCommitHash = 'b48b88ff' # commit hash must match + self._bitsInternalVer = 2020 # gitlab CI does shallow clones of depth 20 + self._bitsCommitHash = 'c7920d2b' # commit hash must match # the artifact tag below - self._bitsTag = "qemu-bits-10182022" # this is the latest bits + self._bitsTag = "qemu-bits-10262023" # this is the latest bits # release as of today. - self._bitsArtSHA1Hash = 'b04790ac9b99b5662d0416392c73b97580641fe5' + self._bitsArtSHA1Hash = 'b22cdfcfc7453875297d06d626f5474ee36a343f' self._bitsArtURL = ("https://gitlab.com/qemu-project/" "biosbits-bits/-/jobs/artifacts/%s/" "download?job=qemu-bits-build" %self._bitsTag) @@ -386,15 +387,20 @@ class AcpiBitsTest(QemuBaseTest): #pylint: disable=too-many-instance-attributes # for newer machine models"). Therefore, enforce 32-bit entry point. self._vm.add_args('-machine', 'smbios-entry-point-type=32') - args = " ".join(str(arg) for arg in self._vm.base_args()) + \ - " " + " ".join(str(arg) for arg in self._vm.args) + # enable console logging + self._vm.set_console() + self._vm.launch() - self.logger.info("launching QEMU vm with the following arguments: %s", - args) + self.logger.debug("Console output from bits VM follows ...") + c_drainer = drainer.LineLogger(self._vm.console_socket.fileno(), + logger=self.logger.getChild("console"), + stop_check=(lambda : + not self._vm.is_running())) + c_drainer.start() - self._vm.launch() # biosbits has been configured to run all the specified test suites # in batch mode and then automatically initiate a vm shutdown. # Rely on avocado's unit test timeout. + self._vm.event_wait('SHUTDOWN') self._vm.wait(timeout=None) self.parse_log() -- cgit v1.2.3