aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2023-09-15implement migration v2 in libvfio-userWilliam Henderson6-676/+263
Signed-off-by: William Henderson <william.henderson@nutanix.com>
2023-08-18initial spec changes based on qemu-devel review (#767)William Henderson1-16/+16
See https://lore.kernel.org/all/CAJSP0QVJZMuuPqyt3wA=NRLzKB-TLWjGWuYnfRvUd=LXraZUWg@mail.gmail.com/. Signed-off-by: William Henderson <william.henderson@nutanix.com>
2023-07-26add support for migration v2 to the spec (#745)William Henderson1-17/+396
* update to latest version of the spec This pulls in the version of the spec that was submitted to qemu-devel: https://lists.nongnu.org/archive/html/qemu-devel/2023-06/msg06567.html This version of the spec removes live migration support as it will be added later on following v2 of the VFIO migration protocol. However, we will not remove the code for live migration v1 until we implement v2. We do not implement all of the other new changes in this spec update but are otherwise compatible. * add support for migration v2 to the spec This adds the VFIO_USER_DEVICE_FEATURE, VFIO_USER_MIG_DATA_READ, and VFIO_USER_MIG_DATA_WRITE commands. Signed-off-by: William Henderson <william.henderson@nutanix.com> Reviewed-by: John Levon <john.levon@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2023-07-17update to latest version of the spec (#744)William Henderson1-381/+88
This pulls in the version of the spec that was submitted to qemu-devel: https://lists.nongnu.org/archive/html/qemu-devel/2023-06/msg06567.html This version of the spec removes live migration support as it will be added later on following v2 of the VFIO migration protocol. However, we will not remove the code for live migration v1 until we implement v2. We do not implement all of the other new changes in this spec update but are otherwise compatible. Signed-off-by: William Henderson <william.henderson@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2023-07-03Fix address calculation for message-based DMA (#740)Mattias Nissler2-25/+39
The correct DMA address is formed by adding base and offset - the latter was accidentally missing. Change the server example to read and write blocks at non-zero offsets, such that `test-client-server.sh` exercises offset handling. Signed-off-by: Mattias Nissler <mnissler@rivosinc.com>
2023-06-22fix build instructions for samples (#739)William Henderson1-5/+5
Signed-off-by: William Henderson <william.henderson@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2023-06-21Fix some python tests (#737)Axel PASCON3-6/+6
Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2023-06-08fix err/req irq fd issue (#731)limiao-intel2-17/+64
When handle_device_set_irqs set err irq/req irq, fd will be filled in vfu_ctx->irqs->efds[] rather than vfu_ctx->irqs->err_efd or vfu_ctx->irqs->req_efd. This patch adds irq index judgment before filling in fd to make sure fd is filled in the correct place. Signed-off-by: Miao Li <miao.li@intel.com> Reviewed-by: John Levon <john.levon@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2023-05-24test: don't leave global pointing to stack memory (#735)Jim Harris1-0/+1
test_device_is_stopped_and_copying points the global vfu_ctx structure to a local stack-allocated data structure. This is fine while the function is executing, but newer gcc complains that the pointer is left there after it returns. So clear the pointer to NULL before returning. Fixes issue #734. Reported-by: Kamil Godzwon <kamilx.godzwon@intel.com> Signed-off-by: Jim Harris <james.r.harris@intel.com>
2023-01-04allow -1 file descriptor for ioregionfd (#727)Thanos Makatos3-1/+111
Signed-off-by: Thanos Makatos <thanos.makatos@nutanix.com> Reviewed-by: John Levon <john.levon@nutanix.com>
2023-01-03fix FLR reset callback (#729)John Levon6-33/+64
A reset callback is allowed to call functions disallowed in quiescent state. However, the FLR reset path neglected to account for this properly, causing an incorrect assert to be triggered if, for example, vfu_sgl_put() is called. To fix this, make sure all reset paths go through call_reset_cb(). Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2023-01-01test: use `grep -E` instead of `egrep` (#728)Ville Skyttä1-1/+1
`egrep` has been deprecated in GNU grep since 2007, and since 3.8 it emits obsolescence warnings: https://git.savannah.gnu.org/cgit/grep.git/commit/?id=a9515624709865d480e3142fd959bccd1c9372d1 Signed-off-by: Ville Skyttä <ville.skytta@iki.fi>
2022-11-22shadow ioeventfd: add demo (#722)Thanos Makatos5-0/+329
Signed-off-by: Thanos Makatos <thanos.makatos@nutanix.com> Reviewed-by: John Levon <john.levon@nutanix.com>
2022-11-22vfu_pci_init: initialize PCI config space flags (#724)Thanos Makatos2-0/+74
vfu_pci_init() sets the size of the PCI config space but not the flags; vfu_realize_ctx() won't initialize the flags since the size if already set. vfu_pci_init() must initialize flags as well. Signed-off-by: Thanos Makatos <thanos.makatos@nutanix.com> Reviewed-by: John Levon <john.levon@nutanix.com>
2022-11-22add debugging to handle_device_get_region_io_fds (#723)Thanos Makatos1-0/+4
Signed-off-by: Thanos Makatos <thanos.makatos@nutanix.com> Reviewed-by: John Levon <john.levon@nutanix.com>
2022-11-22fix shadow ioeventfd unit test (#726)Thanos Makatos2-3/+3
Signed-off-by: Thanos Makatos <thanos.makatos@nutanix.com> Reviewed-by: John Levon <john.levon@nutanix.com>
2022-11-22allow shadow memory offset per shadow ioeventfd (#703)Thanos Makatos7-28/+34
Signed-off-by: Thanos Makatos <thanos.makatos@nutanix.com> Reviewed-by: John Levon <john.levon@nutanix.com>
2022-11-16check spelling (#720)John Levon6-3/+22
Use misspell-fixer if available, and correct the small number of errors it found. Rather than trying to install into the CI, run it directly from a github action. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Swapnil Ingle <swapnil.ingle@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2022-10-05add some unlikely (#717)Thanos Makatos1-13/+16
Signed-off-by: Thanos Makatos <thanos.makatos@nutanix.com> Reviewed-by: John Levon <john.levon@nutanix.com>
2022-10-05only call debug_region_access if in debug mode (#716)Thanos Makatos1-1/+9
Signed-off-by: Thanos Makatos <thanos.makatos@nutanix.com> Reviewed-by: John Levon <john.levon@nutanix.com>
2022-10-05don't duplicate FD in get region info (#715)Thanos Makatos2-22/+5
This is out of spec. Signed-off-by: Thanos Makatos <thanos.makatos@nutanix.com> Reviewed-by: John Levon <john.levon@nutanix.com>
2022-10-04fix compilation for i386 and ppc64 (#709)Thanos Makatos23-204/+319
Signed-off-by: Thanos Makatos <thanos.makatos@nutanix.com> Reviewed-by: John Levon <john.levon@nutanix.com> Reported-by: Eduardo Lima <eblima@gmail.com>
2022-10-03document how to run individual Python unit tests (#712)Thanos Makatos1-0/+7
Signed-off-by: Thanos Makatos <thanos.makatos@nutanix.com> Reviewed-by: John Levon <john.levon@nutanix.com>
2022-08-18make SGL error-checking DEBUG-only (#706)John Levon1-3/+11
As vfu_addr_to_sgl() and co are on the hot path, compile out these sanity checks for non-DEBUG builds. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Swapnil Ingle <swapnil.ingle@nutanix.com>
2022-08-18avoid vfu_log() in SGL hot path (#705)John Levon4-1/+18
Even though in non-debug, we don't actually log anything here, even assembling the arguments to vfu_log() has a performance impact. Hide them behind a DEBUG_SGL define - even in a DEBUG build, they are particularly noisy and low-value. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Swapnil Ingle <swapnil.ingle@nutanix.com>
2022-08-08delete socket on vfu_ctx_destroy (#702)Thanos Makatos4-5/+13
fixes #660 Signed-off-by: Thanos Makatos <thanos.makatos@nutanix.com> Reviewed-by: John Levon <john.levon@nutanix.com> Reviewed-by: Swapnil Ingle <swapnil.ingle@nutanix.com>
2022-07-21disable client-server test by default (#700)John Levon2-12/+16
This test is flaky: there is some kind of race that causes the test to hang. Now we are run as part of qemu CI, we need to disable this by default, until we can find time to fix the test. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Swapnil Ingle <swapnil.ingle@nutanix.com>
2022-07-04support for shadow ioeventfd (#698)Thanos Makatos12-13/+184
When an ioeventfd is written to, KVM discards the value since it has no memory to write it to, and simply kicks the eventfd. This a problem for devices such a NVMe controllers that need the value (e.g. doorbells on BAR0). This patch allows the vfio-user server to pass a file descriptor that can be mmap'ed and KVM can write the ioeventfd value to this _shadow_ memory instead of discarding it. This shadow memory is not exposed to the guest. Signed-off-by: Thanos Makatos <thanos.makatos@nutanix.com> Reviewed-by: John Levon <john.levon@nutanix.com> Change-Id: Iad849c94076ffa5988e034c8bf7ec312d01f095f
2022-06-16improve README.md (#696)John Levon1-114/+100
Re-organize the README so it flows better, and make several fixes/improvements. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2022-06-09test/py: fix the arguments for vfu_dev_irq_state_cb_t (#695)Jag Raman2-4/+7
There is a typo in the arguments for vfu_dev_irq_state_cb_t - fix it in this patch. Signed-off-by: Jagannathan Raman <jag.raman@oracle.com> Reviewed-by: John Levon <john.levon@nutanix.com>
2022-06-09report function in quiesce_check_allowed() (#693)John Levon1-7/+10
Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Swapnil Ingle <swapnil.ingle@nutanix.com>
2022-06-07irq: inform device of IRQ mask & unmask via callback (#694)Jag Raman6-2/+107
Client masks or unmasks a device IRQ using the VFIO_USER_DEVICE_SET_IRQS message. Inform the device of such changes to the IRQ state. Signed-off-by: Jagannathan Raman <jag.raman@oracle.com> Reviewed-by: John Levon <john.levon@nutanix.com>
2022-05-30allow all LOG_* levels (#691)John Levon1-2/+1
While libvfio-user doesn't use them all, at least SPDK was expecting to be able to set LOG_NOTICE level, and silently failing. There's no reason we can't support any valid syslog level. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Swapnil Ingle <swapnil.ingle@nutanix.com>
2022-05-30allow concurrent dirty bitmap get (#677)John Levon4-45/+161
Use atomic operations to allow concurrent bitmap updates with VFIO_IOMMU_DIRTY_PAGES_FLAG_GET_BITMAP operations. Dirtying clients can race against each other, so we must use atomic or when marking dirty: we do this byte-by-byte. When reading the dirty bitmap, we must be careful to not race and lose any set bits within the same byte. If we miss an update, we'll catch it the next time around, presuming that before the final pass we'll have quiesced all I/O. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Raphael Norwitz <raphael.norwitz@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2022-05-27add some documentation on memory handling (#676)John Levon2-1/+141
Cover briefly how to access client memory. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Raphael Norwitz <raphael.norwitz@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2022-05-27re-work SGL API (#675)John Levon13-168/+175
Harmonize and rename the vfu_*sg() APIs to better reflect their functionality: in our case, there is no mapping happening as part of these calls, they are merely housekeeping for range splitting, dirty tracking, and so on. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2022-05-27remove maps list from DMA controller (#674)John Levon4-45/+8
->maps existed so that if a consumer does vfu_map_sg() and then we are asked to enable dirty page tracking, we won't mark those pages as dirty, and will hence potentially lose data. Now that we require quiesce and the use of either vfu_unmap_sg() or vfu_sg_mark_dirty(), there's no need to have this list any more. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2022-05-27remove refcnt from region (#673)John Levon3-6/+0
The reference count is unused, and not atomically handled, remove it. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2022-05-27re-work SG dirty tracking (#672)John Levon4-37/+91
Move SG dirtying to vfu_unmap_sg(): as we don't want to track SGs ourselves, doing this in vfu_map_sg() is no longer the right place. Note that the lack of tracking implies that any SGs must be unmapped before the final stop and copy phase. To avoid the need for this, add vfu_mark_sg_dirty(): this allows a consumer to mark a region as dirty explicitly without needing to unmap it. Currently it's the same as vfu_unmap_sg(), but that's an implementation detail. Note this still marks current maps after a get operation; that will change subsequently. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2022-05-27require quiesce for VFIO_USER_DIRTY_PAGES (#671)John Levon2-6/+90
If we require a quiesce for these calls, we can be sure that it will not race with any usage of vfu_*_sg() calls, as a first step towards concurrency. This is not ideal for VFIO_IOMMU_DIRTY_PAGES_FLAG_GET_BITMAP, which can potentially be called multiple times during pre-copy phase, but that's something we can fix later. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2022-05-23libvfio-user.h: sync VFIO_DEVICE_STATE_XXXX definitions with upstream (#690)Jag Raman10-92/+110
Rename VFIO_DEVICE_STATE_XXXX defines as VFIO_DEVICE_STATE_V1_XXXX. Upstream renamed these variable to be of the XXXX_V1_XXXX format and switched an enum for VFIO_DEVICE_STATE_XXXX. Signed-off-by: Jagannathan Raman <jag.raman@oracle.com> Reviewed-by: John Levon <john.levon@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2022-05-17tests: add checks to confirm test environment (#688)Jag Raman2-3/+8
test-lspci.sh: some test platforms don't include the lspci command, as such skip this test if lspci is not found test-linkage.sh: specify the source and build root paths of the subproject instead of the root paths of the master project Signed-off-by: Jagannathan Raman <jag.raman@oracle.com> Reviewed-by: John Levon <john.levon@nutanix.com>
2022-05-16add a slack invite link (#685)John Levon1-1/+2
Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Swapnil Ingle <swapnil.ingle@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2022-05-16add a diagram to README.md (#686)John Levon3-0/+3
Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2022-05-13samples/client.c: fix implicit casting of const char pointers (#682)Jag Raman2-2/+3
samples/client.c implicitly casts const char * to char * in a couple of places - as such discards the const qualifier. QEMU complains about this as it builds with -Werror=discarded-qualifiers This patch declares irq_to_str as an array of const char pointers. It also casts a "migrate_to() -> _argv" member explicitly Also adds '-Wwrite-strings' build flag to catch similar issues in the future Signed-off-by: Jagannathan Raman <jag.raman@oracle.com> Reviewed-by: John Levon <john.levon@nutanix.com>
2022-05-13fix fedora CI (#683)John Levon1-1/+1
Presumably Fedora 36 came out, as our docker label moved, and we can no longer install. Change the docker label to fedora:35 to fix. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Swapnil Ingle <swapnil.ingle@nutanix.com>
2022-05-12re-instate -Werror (#681)John Levon1-0/+4
This got lost in the translation to meson; add -Werror for debug builds. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Swapnil Ingle <swapnil.ingle@nutanix.com>
2022-05-12run scan-build in CI (#680)John Levon6-27/+40
Yet another static analyzer pass, this one is used by SPDK, and as it did detect some minor issues, it's worth running. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
2022-05-12build PR runs in release mode (#669)John Levon2-12/+12
Instead of "plain", which doesn't specify any optimization, use "release", which both turns off -DDEBUG and enables -O3. Switch the default build to "debug" - we can't use optimization if we want to run valgrind, as there are apparent CPython bugs that trip up valgrind in that case. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Swapnil Ingle <swapnil.ingle@nutanix.com>
2022-05-12add an ubuntu-latest PR run (#668)John Levon3-0/+15
Without this, we didn't detect the openssl-3.0 build breakage. Signed-off-by: John Levon <john.levon@nutanix.com> Reviewed-by: Thanos Makatos <thanos.makatos@nutanix.com>
generated by cgit v1.1 at 2024-09-07 05:50:19 +0000