Currently we only enforce power-of-two mappings (required by the QEMU
notifier) for UNMAP requests. A MAP request not aligned on a
power-of-two may be successfully handled by VFIO, and then the
corresponding UNMAP notify will fail because it will attempt to split
that mapping. Ensure MAP and UNMAP notifications are consistent.
Fixes: dde3f08b5c ("virtio-iommu: Handle non power of 2 range invalidations")
Reported-by: Tina Zhang <tina.zhang@intel.com>
Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Message-Id: <20220718135636.338264-1-jean-philippe@linaro.org>
Tested-by: Tina Zhang <tina.zhang@intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The added enforcing is only relevant in the case of AMD where the
range right before the 1TB is restricted and cannot be DMA mapped
by the kernel consequently leading to IOMMU INVALID_DEVICE_REQUEST
or possibly other kinds of IOMMU events in the AMD IOMMU.
Although, there's a case where it may make sense to disable the
IOVA relocation/validation when migrating from a
non-amd-1tb-aware qemu to one that supports it.
Relocating RAM regions to after the 1Tb hole has consequences for
guest ABI because we are changing the memory mapping, so make
sure that only new machine enforce but not older ones.
Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <20220719170014.27028-12-joao.m.martins@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
It is assumed that the whole GPA space is available to be DMA
addressable, within a given address space limit, except for a
tiny region before the 4G. Since Linux v5.4, VFIO validates
whether the selected GPA is indeed valid i.e. not reserved by
IOMMU on behalf of some specific devices or platform-defined
restrictions, and thus failing the ioctl(VFIO_DMA_MAP) with
-EINVAL.
AMD systems with an IOMMU are examples of such platforms and
particularly may only have these ranges as allowed:
0000000000000000 - 00000000fedfffff (0 .. 3.982G)
00000000fef00000 - 000000fcffffffff (3.983G .. 1011.9G)
0000010000000000 - ffffffffffffffff (1Tb .. 16Pb[*])
We already account for the 4G hole, albeit if the guest is big
enough we will fail to allocate a guest with >1010G due to the
~12G hole at the 1Tb boundary, reserved for HyperTransport (HT).
[*] there is another reserved region unrelated to HT that exists
in the 256T boundary in Fam 17h according to Errata #1286,
documeted also in "Open-Source Register Reference for AMD Family
17h Processors (PUB)"
When creating the region above 4G, take into account that on AMD
platforms the HyperTransport range is reserved and hence it
cannot be used either as GPAs. On those cases rather than
establishing the start of ram-above-4g to be 4G, relocate instead
to 1Tb. See AMD IOMMU spec, section 2.1.2 "IOMMU Logical
Topology", for more information on the underlying restriction of
IOVAs.
After accounting for the 1Tb hole on AMD hosts, mtree should
look like:
0000000000000000-000000007fffffff (prio 0, i/o):
alias ram-below-4g @pc.ram 0000000000000000-000000007fffffff
0000010000000000-000001ff7fffffff (prio 0, i/o):
alias ram-above-4g @pc.ram 0000000080000000-000000ffffffffff
If the relocation is done or the address space covers it, we
also add the the reserved HT e820 range as reserved.
Default phys-bits on Qemu is TCG_PHYS_ADDR_BITS (40) which is enough
to address 1Tb (0xff ffff ffff). On AMD platforms, if a
ram-above-4g relocation is attempted and the CPU wasn't configured
with a big enough phys-bits, an error message will be printed
due to the maxphysaddr vs maxusedaddr check previously added.
Suggested-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <20220719170014.27028-11-joao.m.martins@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Calculate max *used* GPA against the CPU maximum possible address
and error out if the former surprasses the latter. This ensures
max used GPA is reacheable by configured phys-bits. Default phys-bits
on Qemu is TCG_PHYS_ADDR_BITS (40) which is enough for the CPU to
address 1Tb (0xff ffff ffff) or 1010G (0xfc ffff ffff) in AMD hosts
with IOMMU.
This is preparation for AMD guests with >1010G, where it will want relocate
ram-above-4g to be after 1Tb instead of 4G.
Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <20220719170014.27028-10-joao.m.martins@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Move obtaining hole64_start from device_memory memory region base/size
into an helper alongside correspondent getters in pc_memory_init() when
the hotplug range is unitialized. While doing that remove the memory
region based logic from this newly added helper.
This is the final step that allows pc_pci_hole64_start() to be callable
at the beginning of pc_memory_init() before any memory regions are
initialized.
Cc: Jonathan Cameron <jonathan.cameron@huawei.com>
Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <20220719170014.27028-9-joao.m.martins@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Remove pc_get_cxl_range_end() dependency on the CXL memory region,
and replace with one that does not require the CXL host_mr to determine
the start of CXL start.
This in preparation to allow pc_pci_hole64_start() to be called early
in pc_memory_init(), handle CXL memory region end when its underlying
memory region isn't yet initialized.
Cc: Jonathan Cameron <jonathan.cameron@huawei.com>
Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Message-Id: <20220719170014.27028-8-joao.m.martins@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
Factor out the calculation of the base address of the memory region.
It will be used later on for the cxl range end counterpart calculation
and as well in pc_memory_init() CXL memory region initialization, thus
avoiding duplication.
Cc: Jonathan Cameron <jonathan.cameron@huawei.com>
Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <20220719170014.27028-7-joao.m.martins@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Move calculation of CXL memory region end to separate helper.
This is in preparation to a future change that removes CXL range
dependency on the CXL memory region, with the goal of allowing
pc_pci_hole64_start() to be called before any memory region are
initialized.
Cc: Jonathan Cameron <jonathan.cameron@huawei.com>
Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <20220719170014.27028-6-joao.m.martins@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
There's a couple of places that seem to duplicate this calculation
of RAM size above the 4G boundary. Move all those to a helper function.
Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <20220719170014.27028-5-joao.m.martins@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Use the pre-initialized pci-host qdev and fetch the
pci-hole64-size into pc_memory_init() newly added argument.
Use PCI_HOST_PROP_PCI_HOLE64_SIZE pci-host property for
fetching pci-hole64-size.
This is in preparation to determine that host-phys-bits are
enough and for pci-hole64-size to be considered to relocate
ram-above-4g to be at 1T (on AMD platforms).
Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <20220719170014.27028-4-joao.m.martins@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
At the start of pc_memory_init() we usually pass a range of
0..UINT64_MAX as pci_memory, when really its 2G (i440fx) or
32G (q35). To get the real user value, we need to get pci-host
passed property for default pci_hole64_size. Thus to get that,
create the qdev prior to memory init to better make estimations
on max used/phys addr.
This is in preparation to determine that host-phys-bits are
enough and also for pci-hole64-size to be considered to relocate
ram-above-4g to be at 1T (on AMD platforms).
Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <20220719170014.27028-3-joao.m.martins@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Rather than hardcoding the 4G boundary everywhere, introduce a
X86MachineState field @above_4g_mem_start and use it
accordingly.
This is in preparation for relocating ram-above-4g to be
dynamically start at 1T on AMD platforms.
Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <20220719170014.27028-2-joao.m.martins@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Whilst the interleave granularity is always small enough that this isn't
a real problem (much less than 4GiB) let's change the constant
to ULL to fix the coverity warning.
Reported-by: Peter Maydell <peter.maydell@linaro.org>
Fixes: 829de299d1 ("hw/cxl/component: Add utils for interleave parameter encoding/decoding")
Fixes: Coverity CID 1488868
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Message-Id: <20220701132300.2264-4-Jonathan.Cameron@huawei.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Previously broken_reserved_end was taken into account, but Igor Mammedov
identified that this could lead to a clash between potential RAM being
mapped in the region and CXL usage. Hence always add the size of the
device_memory memory region. This only affects the case where the
broken_reserved_end flag was set.
Fixes: 6e4e3ae936 ("hw/cxl/component: Implement host bridge MMIO (8.2.5, table 142)")
Reported-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Message-Id: <20220701132300.2264-3-Jonathan.Cameron@huawei.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This got left behind in the move of the CXL setup code from core
files to the machines that support it.
Link: 1ebf9001fb
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Message-Id: <20220701132300.2264-2-Jonathan.Cameron@huawei.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Robert Hoo <robert.hu@linux.intel.com>
Reviewed-by: Jingqi Liu <jingqi.liu@intel.com>
Message-Id: <20220704085852.330005-1-robert.hu@linux.intel.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Version: GnuPG v1
iQEcBAABAgAGBQJi36ocAAoJEO8Ells5jWIRuOYH/jtaDNGTBs/h8A041gQaCMmw
jufUXHCdKGgmZMpJ/AoCUWx4USdx8hEGSt/j4kvSmIPX+VLuCfLefHDlTxndiWAv
fnUr4NB7LAz2b5D3d5QX1Np+zHG5mHx95KfDIaWdcz9N1HUHlEOakxTDc2EvR1hF
yh8g2n5xdvzK5kWvPcNgJpU/ezDumOFo04JndBb4fIqDmZfW3hvJQ3IKiS3P1J9C
Kbb/usoXGrdoZ9T1R2cqtn1CxrgfMlF2pKJFWzs3nU+ewD9C6oKS4rDQCZxx+JEx
ZvfnSTUPgBBlT4zqZTTjyFQMQdtis5qK5iAKDEENkqVC1iULPhnM9DN0qxcIoQs=
=SpWG
-----END PGP SIGNATURE-----
Merge tag 'net-pull-request' of https://github.com/jasowang/qemu into staging
# gpg: Signature made Tue 26 Jul 2022 09:47:24 BST
# gpg: using RSA key EF04965B398D6211
# gpg: Good signature from "Jason Wang (Jason Wang on RedHat) <jasowang@redhat.com>" [marginal]
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg: It is not certain that the signature belongs to the owner.
# Primary key fingerprint: 215D 46F4 8246 689E C77F 3562 EF04 965B 398D 6211
* tag 'net-pull-request' of https://github.com/jasowang/qemu:
vdpa: Fix memory listener deletions of iova tree
vhost: Get vring base from vq, not svq
e1000e: Fix possible interrupt loss when using MSI
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
vhost_vdpa_listener_region_del is always deleting the first iova entry
of the tree, since it's using the needle iova instead of the result's
one.
This was detected using a vga virtual device in the VM using vdpa SVQ.
It makes some extra memory adding and deleting, so the wrong one was
mapped / unmapped. This was undetected before since all the memory was
mappend and unmapped totally without that device, but other conditions
could trigger it too:
* mem_region was with .iova = 0, .translated_addr = (correct GPA).
* iova_tree_find_iova returned right result, but does not update
mem_region.
* iova_tree_remove always removed region with .iova = 0. Right iova were
sent to the device.
* Next map will fill the first region with .iova = 0, causing a mapping
with the same iova and device complains, if the next action is a map.
* Next unmap will cause to try to unmap again iova = 0, causing the
device to complain that no region was mapped at iova = 0.
Fixes: 34e3c94eda ("vdpa: Add custom IOTLB translations to SVQ")
Reported-by: Lei Yang <leiyang@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
The SVQ vring used idx usually match with the guest visible one, as long
as all the guest buffers (GPA) maps to exactly one buffer within qemu's
VA. However, as we can see in virtqueue_map_desc, a single guest buffer
could map to many buffers in SVQ vring.
Also, its also a mistake to rewind them at the source of migration.
Since VirtQueue is able to migrate the inflight descriptors, its
responsability of the destination to perform the rewind just in case it
cannot report the inflight descriptors to the device.
This makes easier to migrate between backends or to recover them in
vhost devices that support set in flight descriptors.
Fixes: 6d0b222666 ("vdpa: Adapt vhost_vdpa_get_vring_base to SVQ")
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Commit "e1000e: Prevent MSI/MSI-X storms" introduced msi_causes_pending
to prevent interrupt storms problem. It was tested with MSI-X.
In case of MSI, the guest can rely solely on interrupts to clear ICR.
Upon clearing all pending interrupts, msi_causes_pending gets cleared.
However, when e1000e_itr_should_postpone() in e1000e_send_msi() returns
true, MSI never gets fired by e1000e_intrmgr_on_throttling_timer()
because msi_causes_pending is still set. This results in interrupt loss.
To prevent this, we need to clear msi_causes_pending when MSI is going
to get fired by the throttling timer. The guest can then receive
interrupts eventually.
Signed-off-by: Ake Koomsin <ake@igel.co.jp>
Signed-off-by: Jason Wang <jasowang@redhat.com>
* Pass random seed to x86 and other FDT platforms
-----BEGIN PGP SIGNATURE-----
iQFIBAABCAAyFiEE8TM4V0tmI4mGbHaCv/vSX3jHroMFAmLa3dUUHHBib256aW5p
QHJlZGhhdC5jb20ACgkQv/vSX3jHroObpwf/ceqT05kDypSbPmSZPspzfGZimoQL
9egbI27siFGUYhmZ/odiv5YU82Y44FHaElsmGsKZQAFvJ4JhROR8ZrDIejI/mWhk
9yCTW5y+DlFHwZbeAfqMQeK1sfI4TvZ70SnBtpFKsA0bkHmYNAtPJZOSL8SEtZJS
HA0+jOQdk1+ddjQjgy1AOg5R51nHQGELNz29aF2Z3elKN8ZM9BGY2TQzJ+SMfyRW
+iU2r5teqRzHDK005WFZgaH5OtG5f2t/fgRycG9WDQYiYmna9wZQICyCiwEEgFu+
G7lqtPR0YRuVgFwqhhHW7i0wg0GvpEjCRyzc3Gets2j4FjYKn66xy2EPSA==
=OYEp
-----END PGP SIGNATURE-----
Merge tag 'for-upstream2' of https://gitlab.com/bonzini/qemu into staging
* Bug fixes
* Pass random seed to x86 and other FDT platforms
# gpg: Signature made Fri 22 Jul 2022 18:26:45 BST
# gpg: using RSA key F13338574B662389866C7682BFFBD25F78C7AE83
# gpg: issuer "pbonzini@redhat.com"
# gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>" [full]
# gpg: aka "Paolo Bonzini <pbonzini@redhat.com>" [full]
# Primary key fingerprint: 46F5 9FBD 57D6 12E7 BFD4 E2F7 7E15 100C CD36 69B1
# Subkey fingerprint: F133 3857 4B66 2389 866C 7682 BFFB D25F 78C7 AE83
* tag 'for-upstream2' of https://gitlab.com/bonzini/qemu:
hw/i386: pass RNG seed via setup_data entry
hw/rx: pass random seed to fdt
hw/mips: boston: pass random seed to fdt
hw/nios2: virt: pass random seed to fdt
oss-fuzz: ensure base_copy is a generic-fuzzer
oss-fuzz: remove binaries from qemu-bundle tree
accel/kvm: Avoid Coverity warning in query_stats()
docs: Add caveats for Windows as the build platform
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
When writing back the fd[1] pipe file handle to emulated userspace
memory, use sizeof(abi_int) as offset insted of the hosts's int type.
There is no functional change in this patch.
Signed-off-by: Helge Deller <deller@gmx.de>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <YtQ3Id6z8slpVr7r@p100>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
The pipe2() syscall is available on all Linux platforms since kernel
2.6.27, so use it unconditionally to emulate pipe() and pipe2().
Signed-off-by: Helge Deller <deller@gmx.de>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <YtbZ2ojisTnzxN9Y@p100>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
This program:
int main(void) { asm("bv %r0(%r0)"); return 0; }
produces on real hppa hardware the expected segfault:
SIGSEGV {si_signo=SIGSEGV, si_code=SEGV_MAPERR, si_addr=0x3} ---
killed by SIGSEGV +++
Segmentation fault
But when run on linux-user you get instead internal qemu errors:
ERROR: linux-user/hppa/cpu_loop.c:172:cpu_loop: code should not be reached
Bail out! ERROR: linux-user/hppa/cpu_loop.c:172:cpu_loop: code should not be reached
ERROR: accel/tcg/cpu-exec.c:933:cpu_exec: assertion failed: (cpu == current_cpu)
Bail out! ERROR: accel/tcg/cpu-exec.c:933:cpu_exec: assertion failed: (cpu == current_cpu)
Fix it by adding the missing case for the EXCP_IMP trap in
cpu_loop() and raise a segfault.
Signed-off-by: Helge Deller <deller@gmx.de>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <YtWNC56seiV6VenA@p100>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Tiny machines optimized for fast boot time generally don't use EFI,
which means a random seed has to be supplied some other way. For this
purpose, Linux (≥5.20) supports passing a seed in the setup_data table
with SETUP_RNG_SEED, specially intended for hypervisors, kexec, and
specialized bootloaders. The linked commit shows the upstream kernel
implementation.
At Paolo's request, we don't pass these to versioned machine types ≤7.0.
Link: https://git.kernel.org/tip/tip/c/68b8e9713c8
Cc: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Richard Henderson <richard.henderson@linaro.org>
Cc: Eduardo Habkost <eduardo@habkost.net>
Cc: Peter Maydell <peter.maydell@linaro.org>
Cc: Philippe Mathieu-Daudé <f4bug@amsat.org>
Cc: Laurent Vivier <laurent@vivier.eu>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Message-Id: <20220721125636.446842-1-Jason@zx2c4.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If the FDT contains /chosen/rng-seed, then the Linux RNG will use it to
initialize early. Set this using the usual guest random number
generation function. This FDT node is part of the DT specification.
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Message-Id: <20220719122033.135902-1-Jason@zx2c4.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If the FDT contains /chosen/rng-seed, then the Linux RNG will use it to
initialize early. Set this using the usual guest random number
generation function. This FDT node is part of the DT specification.
I'd do the same for other MIPS platforms but boston is the only one that
seems to use FDT.
Cc: Paul Burton <paulburton@kernel.org>
Cc: Aleksandar Rikalo <aleksandar.rikalo@syrmia.com>
Cc: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Message-Id: <20220719120843.134392-1-Jason@zx2c4.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If the FDT contains /chosen/rng-seed, then the Linux RNG will use it to
initialize early. Set this using the usual guest random number
generation function. This FDT node is part of the DT specification.
Cc: Chris Wulff <crwulff@gmail.com>
Cc: Marek Vasut <marex@denx.de>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Message-Id: <20220719120113.118034-1-Jason@zx2c4.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Depending on how the target list is sorted in by qemu, the first target
(used as the base copy of the fuzzer, to which all others are linked)
might not be a generic-fuzzer. Since we are trying to only use
generic-fuzz, on oss-fuzz, fix that, to ensure the base copy is a
generic-fuzzer.
Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
Message-Id: <20220720180946.2264253-1-alxndr@bu.edu>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
oss-fuzz is finding possible fuzzing targets even under qemu-bundle/.../bin, but they
cannot be used because the required shared libraries are missing. Since the
fuzzing targets are already placed manually in $OUT, the bindir and libexecdir
subtrees are not needed; remove them.
Cc: Alexander Bulekov <alxndr@bu.edu>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Coverity complains that there is a codepath in the query_stats()
function where it can leak the memory pointed to by stats_list. This
can only happen if the caller passes something other than
STATS_TARGET_VM or STATS_TARGET_VCPU as the 'target', which no
callsite does. Enforce this assumption using g_assert_not_reached(),
so that if we have a future bug we hit the assert rather than
silently leaking memory.
Resolves: Coverity CID 1490140
Fixes: cc01a3f4ca ("kvm: Support for querying fd-based stats")
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20220719134853.327059-1-peter.maydell@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Commit cf60ccc330 ("cutils: Introduce bundle mechanism") introduced
a Python script to populate a bundle directory using os.symlink() to
point to the binaries in the pc-bios directory of the source tree.
Commit 882084a04a ("datadir: Use bundle mechanism") removed previous
logic in pc-bios/meson.build to create a link/copy of pc-bios binaries
in the build tree so os.symlink() is the way to go.
However os.symlink() may fail [1] on Windows if an unprivileged Windows
user started the QEMU build process, which results in QEMU executables
generated in the build tree not able to load the default BIOS/firmware
images due to symbolic links not present in the bundle directory.
This commits updates the documentation by adding such caveats for users
who want to build QEMU on the Windows platform.
[1] https://docs.python.org/3/library/os.html#os.symlink
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Stefan Weil <sw@weilnetz.de>
Reviewed-by: Akihiko Odaki <akihiko.odaki@gmail.com>
Message-Id: <20220719135014.764981-1-bmeng.cn@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This replaces yesterdays pull and:
a) Fixes some test build errors without TLS
b) Reenabled the zlib acceleration on s390
now that we have Ilya's fix
Hyman's dirty page rate limit set
Ilya's fix for zlib vs migration
Peter's postcopy-preempt
Cleanup from Dan
zero-copy tidy ups from Leo
multifd doc fix from Juan
Revert disable of zlib acceleration on s390x
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
-----BEGIN PGP SIGNATURE-----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=wiSA
-----END PGP SIGNATURE-----
Merge tag 'pull-migration-20220720c' of https://gitlab.com/dagrh/qemu into staging
Migration pull 2022-07-20
This replaces yesterdays pull and:
a) Fixes some test build errors without TLS
b) Reenabled the zlib acceleration on s390
now that we have Ilya's fix
Hyman's dirty page rate limit set
Ilya's fix for zlib vs migration
Peter's postcopy-preempt
Cleanup from Dan
zero-copy tidy ups from Leo
multifd doc fix from Juan
Revert disable of zlib acceleration on s390x
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
# gpg: Signature made Wed 20 Jul 2022 12:18:56 BST
# gpg: using RSA key 45F5C71B4A0CB7FB977A9FA90516331EBC5BFDE7
# gpg: Good signature from "Dr. David Alan Gilbert (RH2) <dgilbert@redhat.com>" [full]
# Primary key fingerprint: 45F5 C71B 4A0C B7FB 977A 9FA9 0516 331E BC5B FDE7
* tag 'pull-migration-20220720c' of https://gitlab.com/dagrh/qemu: (30 commits)
Revert "gitlab: disable accelerated zlib for s390x"
migration: Avoid false-positive on non-supported scenarios for zero-copy-send
multifd: Document the locking of MultiFD{Send/Recv}Params
migration/multifd: Report to user when zerocopy not working
Add dirty-sync-missed-zero-copy migration stat
QIOChannelSocket: Fix zero-copy flush returning code 1 when nothing sent
migration: remove unreachable code after reading data
tests: Add postcopy preempt tests
tests: Add postcopy tls recovery migration test
tests: Add postcopy tls migration test
tests: Move MigrateCommon upper
migration: Respect postcopy request order in preemption mode
migration: Enable TLS for preempt channel
migration: Export tls-[creds|hostname|authz] params to cmdline too
migration: Add helpers to detect TLS capability
migration: Add property x-postcopy-preempt-break-huge
migration: Create the postcopy preempt channel asynchronously
migration: Postcopy recover with preempt enabled
migration: Postcopy preemption enablement
migration: Postcopy preemption preparation on channel creation
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Version: GnuPG v1
iQEcBAABAgAGBQJi18PHAAoJEO8Ells5jWIRCEQH+wepXDoT6Q56xmUgxVs+hlAD
CXGy71/cNV08Yu3PTTXo8SYaw+KXxsA9ECgIr2hsfPXarAdoOpJFpZR0HoqIzaXd
kpD6bvwN8bEEOlAHxKcb6/VM+VYntZBfkH9m1WLGx3fHILazLblyL8w2Hkp7NK9J
IBpQQ63uU8Xt0+js96Z/sPOKRjrtbKXFT1bhY2CI8MKZpuqNyED0jZYwbNdnRwZN
fuKbpsaaT4Wxx+mQMg7H7a0e/xx3DNi2F6cAtGLH98WYzbLFgExSSK8G8jnwEVfM
EKWfU7N4zmokq7jN99yvGzjIzLrnLX6yn/ifSs+lQOzdtCA9zEbotI+CDCVdPs4=
=9zus
-----END PGP SIGNATURE-----
Merge tag 'net-pull-request' of https://github.com/jasowang/qemu into staging
# gpg: Signature made Wed 20 Jul 2022 09:58:47 BST
# gpg: using RSA key EF04965B398D6211
# gpg: Good signature from "Jason Wang (Jason Wang on RedHat) <jasowang@redhat.com>" [marginal]
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg: It is not certain that the signature belongs to the owner.
# Primary key fingerprint: 215D 46F4 8246 689E C77F 3562 EF04 965B 398D 6211
* tag 'net-pull-request' of https://github.com/jasowang/qemu: (25 commits)
net/colo.c: fix segmentation fault when packet is not parsed correctly
net/colo.c: No need to track conn_list for filter-rewriter
net/colo: Fix a "double free" crash to clear the conn_list
softmmu/runstate.c: add RunStateTransition support form COLO to PRELAUNCH
vdpa: Add x-svq to NetdevVhostVDPAOptions
vdpa: Add device migration blocker
vdpa: Extract get features part from vhost_vdpa_get_max_queue_pairs
vdpa: Buffer CVQ support on shadow virtqueue
vdpa: manual forward CVQ buffers
vhost-net-vdpa: add stubs for when no virtio-net device is present
vdpa: Export vhost_vdpa_dma_map and unmap calls
vhost: Add svq avail_handler callback
vhost: add vhost_svq_poll
vhost: Expose vhost_svq_add
vhost: add vhost_svq_push_elem
vhost: Track number of descs in SVQDescState
vhost: Add SVQDescState
vhost: Decouple vhost_svq_add from VirtQueueElement
vhost: Check for queue full at vhost_svq_add
vhost: Move vhost_svq_kick call to vhost_svq_add
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
This reverts commit 309df6acb2.
With Ilya's 'multifd: Copy pages before compressing them with zlib'
in the latest migration series, this shouldn't be a problem any more.
Suggested-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Migration with zero-copy-send currently has it's limitations, as it can't
be used with TLS nor any kind of compression. In such scenarios, it should
output errors during parameter / capability setting.
But currently there are some ways of setting this not-supported scenarios
without printing the error message:
!) For 'compression' capability, it works by enabling it together with
zero-copy-send. This happens because the validity test for zero-copy uses
the helper unction migrate_use_compression(), which check for compression
presence in s->enabled_capabilities[MIGRATION_CAPABILITY_COMPRESS].
The point here is: the validity test happens before the capability gets
enabled. If all of them get enabled together, this test will not return
error.
In order to fix that, replace migrate_use_compression() by directly testing
the cap_list parameter migrate_caps_check().
2) For features enabled by parameters such as TLS & 'multifd_compression',
there was also a possibility of setting non-supported scenarios: setting
zero-copy-send first, then setting the unsupported parameter.
In order to fix that, also add a check for parameters conflicting with
zero-copy-send on migrate_params_check().
3) XBZRLE is also a compression capability, so it makes sense to also add
it to the list of capabilities which are not supported with zero-copy-send.
Fixes: 1abaec9a1b ("migration: Change zero_copy_send from migration parameter to migration capability")
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Message-Id: <20220719122345.253713-1-leobras@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reorder the structures so we can know if the fields are:
- Read only
- Their own locking (i.e. sems)
- Protected by 'mutex'
- Only for the multifd channel
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20220531104318.7494-2-quintela@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
dgilbert: Typo fixes from Chen Zhang
Some errors, like the lack of Scatter-Gather support by the network
interface(NETIF_F_SG) may cause sendmsg(...,MSG_ZEROCOPY) to fail on using
zero-copy, which causes it to fall back to the default copying mechanism.
After each full dirty-bitmap scan there should be a zero-copy flush
happening, which checks for errors each of the previous calls to
sendmsg(...,MSG_ZEROCOPY). If all of them failed to use zero-copy, then
increment dirty_sync_missed_zero_copy migration stat to let the user know
about it.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Acked-by: Peter Xu <peterx@redhat.com>
Message-Id: <20220711211112.18951-4-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Acked-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20220711211112.18951-3-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
If flush is called when no buffer was sent with MSG_ZEROCOPY, it currently
returns 1. This return code should be used only when Linux fails to use
MSG_ZEROCOPY on a lot of sendmsg().
Fix this by returning early from flush if no sendmsg(...,MSG_ZEROCOPY)
was attempted.
Fixes: 2bc58ffc29 ("QIOChannelSocket: Implement io_writev zero copy flag & io_flush for CONFIG_LINUX")
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Acked-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Message-Id: <20220711211112.18951-2-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
The code calls qio_channel_read() in a loop when it reports
QIO_CHANNEL_ERR_BLOCK. This code is reported when errno==EAGAIN.
As such the later block of code will always hit the 'errno != EAGAIN'
condition, making the final 'else' unreachable.
Fixes: Coverity CID 1490203
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20220627135318.156121-1-berrange@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Four tests are added for preempt mode:
- Postcopy plain
- Postcopy recovery
- Postcopy tls
- Postcopy tls+recovery
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20220707185530.27801-1-peterx@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
dgilbert: Manual merge
It's easy to build this upon the postcopy tls test. Rename the old
postcopy recovery test to postcopy/recovery/plain.
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20220707185527.27747-1-peterx@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
dgilbert: Manual merge
We just added TLS tests for precopy but not postcopy. Add the
corresponding test for vanilla postcopy.
Rename the vanilla postcopy to "postcopy/plain" because all postcopy tests
will only use unix sockets as channel.
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20220707185525.27692-1-peterx@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
dgilbert: Manual merge
So that it can be used in postcopy tests too soon.
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20220707185522.27638-1-peterx@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
With preemption mode on, when we see a postcopy request that was requesting
for exactly the page that we have preempted before (so we've partially sent
the page already via PRECOPY channel and it got preempted by another
postcopy request), currently we drop the request so that after all the
other postcopy requests are serviced then we'll go back to precopy stream
and start to handle that.
We dropped the request because we can't send it via postcopy channel since
the precopy channel already contains partial of the data, and we can only
send a huge page via one channel as a whole. We can't split a huge page
into two channels.
That's a very corner case and that works, but there's a change on the order
of postcopy requests that we handle since we're postponing this (unlucky)
postcopy request to be later than the other queued postcopy requests. The
problem is there's a possibility that when the guest was very busy, the
postcopy queue can be always non-empty, it means this dropped request will
never be handled until the end of postcopy migration. So, there's a chance
that there's one dest QEMU vcpu thread waiting for a page fault for an
extremely long time just because it's unluckily accessing the specific page
that was preempted before.
The worst case time it needs can be as long as the whole postcopy migration
procedure. It's extremely unlikely to happen, but when it happens it's not
good.
The root cause of this problem is because we treat pss->postcopy_requested
variable as with two meanings bound together, as the variable shows:
1. Whether this page request is urgent, and,
2. Which channel we should use for this page request.
With the old code, when we set postcopy_requested it means either both (1)
and (2) are true, or both (1) and (2) are false. We can never have (1)
and (2) to have different values.
However it doesn't necessarily need to be like that. It's very legal that
there's one request that has (1) very high urgency, but (2) we'd like to
use the precopy channel. Just like the corner case we were discussing
above.
To differenciate the two meanings better, introduce a new field called
postcopy_target_channel, showing which channel we should use for this page
request, so as to cover the old meaning (2) only. Then we leave the
postcopy_requested variable to stand only for meaning (1), which is the
urgency of this page request.
With this change, we can easily boost priority of a preempted precopy page
as long as we know that page is also requested as a postcopy page. So with
the new approach in get_queued_page() instead of dropping that request, we
send it right away with the precopy channel so we get back the ordering of
the page faults just like how they're requested on dest.
Reported-by: Manish Mishra <manish.mishra@nutanix.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Manish Mishra <manish.mishra@nutanix.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20220707185520.27583-1-peterx@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
This patch is based on the async preempt channel creation. It continues
wiring up the new channel with TLS handshake to destionation when enabled.
Note that only the src QEMU needs such operation; the dest QEMU does not
need any change for TLS support due to the fact that all channels are
established synchronously there, so all the TLS magic is already properly
handled by migration_tls_channel_process_incoming().
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20220707185518.27529-1-peterx@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>