Add a test for what happens when you call bdrv_replace_child_noperm()
for various drain situations ({old,new} child {drained,not drained}).
Most importantly, if both the old and the new child are drained, the
parent must not be undrained at any point.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Currently, bdrv_replace_child_noperm() undrains the parent until it is
completely undrained, then re-drains it after attaching the new child
node.
This is a problem with bdrv_drop_intermediate(): We want to keep the
whole subtree drained, including parents, while the operation is
under way. bdrv_replace_child_noperm() breaks this by allowing every
parent to become unquiesced briefly, and then redraining it.
In fact, there is no reason why the parent should become unquiesced and
be allowed to submit requests to the new child node if that new node is
supposed to be kept drained. So if anything, we have to drain the
parent before detaching the old child node. Conversely, we have to
undrain it only after attaching the new child node.
Thus, change the whole drain algorithm here: Calculate the number of
times we have to drain/undrain the parent before replacing the child
node then drain it (if necessary), replace the child node, and then
undrain it.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
bdrv_drop_intermediate() calls BdrvChildRole.update_filename(). That
may poll, thus changing the graph, which potentially breaks the
QLIST_FOREACH_SAFE() loop.
Just keep the whole subtree drained. This is probably the right thing
to do anyway (dropping nodes while the subtree is not drained seems
wrong).
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
The same change as commit 2b23f28639 ('block/copy-on-read: Fix
permissions for inactive node') made for the copy-on-read driver can be
made for bdrv_filter_default_perms(): Retaining the old permissions from
the BdrvChild if it is given complicates things unnecessarily when in
the end this only means that the options set in the c == NULL case (i.e.
during child creation) are retained.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
This test case is motivated by commit 2b23f28639 ('block/copy-on-read:
Fix permissions for inactive node'). Instead of just testing
copy-on-read on migration, let's stack all sorts of filter nodes on top
of each other and try if the resulting VM can still migrate
successfully. For good measure, put everything into an iothread, because
why not?
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
234 implements functions that are useful for doing migration between two
VMs. Move them to iotests.py so that other test cases can use them, too.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
The code path for -device drive=<node-name> or without a drive=...
option for empty drives, which is supposed to be used with -blockdev
differs enough from the -drive based path with a user-owned
BlockBackend, so we want to test both paths at least for the basic tests
implemented by TestInitiallyFilled and TestInitiallyEmpty.
This would have caught the bug recently fixed for inserting read-only
nodes into a scsi-cd created without a drive=... option.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
We're getting a ridiculous number of child classes of
TestInitiallyFilled and TestInitiallyEmpty that differ only in a few
attributes that we want to test in all combinations.
Instead of explicitly writing down every combination, let's use a loop
and create those classes dynamically.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
ppc patch queue 2019-08-13 (last minute qemu-4.1 fixes)
Here's a very, very last minute pull request for qemu-4.1. This fixes
two nasty bugs with the XIVE interrupt controller in "dual" mode
(where the guest decides which interrupt controller it wants to use).
One occurs when resetting the guest while I/O is active, and the other
with migration of hotplugged CPUs.
The timing here is very unfortunate. Alas, we only spotted these bugs
very late, and I was sick last week, delaying analysis and fix even
further.
This series hasn't had nearly as much testing as I'd really like, but
I'd still like to squeeze it into qemu-4.1 if possible, since
definitely fixing two bad bugs seems like an acceptable tradeoff for
the risk of introducing different bugs.
# gpg: Signature made Tue 13 Aug 2019 07:56:42 BST
# gpg: using RSA key 75F46586AE61A66CC44E87DC6C38CACA20D9B392
# gpg: Good signature from "David Gibson <david@gibson.dropbear.id.au>" [full]
# gpg: aka "David Gibson (Red Hat) <dgibson@redhat.com>" [full]
# gpg: aka "David Gibson (ozlabs.org) <dgibson@ozlabs.org>" [full]
# gpg: aka "David Gibson (kernel.org) <dwg@kernel.org>" [unknown]
# Primary key fingerprint: 75F4 6586 AE61 A66C C44E 87DC 6C38 CACA 20D9 B392
* remotes/dgibson/tags/ppc-for-4.1-20190813:
spapr/xive: Fix migration of hot-plugged CPUs
spapr: Reset CAS & IRQ subsystem after devices
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
The migration sequence of a guest using the XIVE exploitation mode
relies on the fact that the states of all devices are restored before
the machine is. This is not true for hot-plug devices such as CPUs
which state come after the machine. This breaks migration because the
thread interrupt context registers are not correctly set.
Fix migration of hotplugged CPUs by restoring their context in the
'post_load' handler of the XiveTCTX model.
Fixes: 277dd3d771 ("spapr/xive: add migration support for KVM")
Signed-off-by: Cédric Le Goater <clg@kaod.org>
Message-Id: <20190813064853.29310-1-clg@kaod.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This fixes a nasty regression in qemu-4.1 for the 'pseries' machine,
caused by the new "dual" interrupt controller model. Specifically,
qemu can crash when used with KVM if a 'system_reset' is requested
while there's active I/O in the guest.
The problem is that in spapr_machine_reset() we:
1. Reset the CAS vector state
spapr_ovec_cleanup(spapr->ov5_cas);
2. Reset all devices
qemu_devices_reset()
3. Reset the irq subsystem
spapr_irq_reset();
However (1) implicitly changes the interrupt delivery mode, because
whether we're using XICS or XIVE depends on the CAS state. We don't
properly initialize the new irq mode until (3) though - in particular
setting up the KVM devices.
During (2), we can temporarily drop the BQL allowing some irqs to be
delivered which will go to an irq system that's not properly set up.
Specifically, if the previous guest was in (KVM) XIVE mode, the CAS
reset will put us back in XICS mode. kvm_kernel_irqchip() still
returns true, because XIVE was using KVM, however XICs doesn't have
its KVM components intialized and kernel_xics_fd == -1. When the irq
is delivered it goes via ics_kvm_set_irq() which assert()s that
kernel_xics_fd != -1.
This change addresses the problem by delaying the CAS reset until
after the devices reset. The device reset should quiesce all the
devices so we won't get irqs delivered while we mess around with the
IRQ. The CAS reset and irq re-initialize should also now be under the
same BQL critical section so nothing else should be able to interrupt
it either.
We also move the spapr_irq_msi_reset() used in one of the legacy irq
modes, since it logically makes sense at the same point as the
spapr_irq_reset() (it's essentially an equivalent operation for older
machine types). Since we don't need to switch between different
interrupt controllers for those old machine types it shouldn't
actually be broken in those cases though.
Cc: Cédric Le Goater <clg@kaod.org>
Fixes: b2e22477 "spapr: add a 'reset' method to the sPAPR IRQ backend"
Fixes: 13db0cd9 "spapr: introduce a new sPAPR IRQ backend supporting
XIVE and XICS"
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Set QEMU_PCI_CAP_EXPRESS unconditionally in init(), then clear it in
realize() in case the device is not connected to a PCIe bus.
This makes sure the pci config space allocation is big enough, so
accessing the PCIe extended config space doesn't overflow the pci
config space buffer.
PCI(e) config space is guest writable. Writes are limited by
write mask (which probably is also filled with random stuff),
so the guest can only flip enabled bits. But I suspect it
still might be exploitable, so rather serious because it might
be a host escape for the guest. On the other hand the device
is probably not yet in widespread use.
(For a QEMU version without this commit, a mitigation for the
bug is available: use "-device bochs-display" as a conventional pci
device only.)
Cc: qemu-stable@nongnu.org
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 20190812065221.20907-2-kraxel@redhat.com
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Block patches for 4.1.0-rc4:
- Fix the backup block job when using copy offloading
- Fix the mirror block job when using the write-blocking copy mode
- Fix incremental backups after the image has been grown with the
respective bitmap attached to it
# gpg: Signature made Tue 06 Aug 2019 12:57:07 BST
# gpg: using RSA key 91BEB60A30DB3E8857D11829F407DB0061D5CF40
# gpg: issuer "mreitz@redhat.com"
# gpg: Good signature from "Max Reitz <mreitz@redhat.com>" [full]
# Primary key fingerprint: 91BE B60A 30DB 3E88 57D1 1829 F407 DB00 61D5 CF40
* remotes/maxreitz/tags/pull-block-2019-08-06:
block/backup: disable copy_range for compressed backup
iotests: Test unaligned blocking mirror write
mirror: Only mirror granularity-aligned chunks
iotests: Test incremental backup after truncation
util/hbitmap: update orig_size on truncate
iotests: Test backup job with two guest writes
backup: Copy only dirty areas
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
In write-blocking mode, all writes to the top node directly go to the
target. We must only mirror chunks of data that are aligned to the
job's granularity, because that is how the dirty bitmap works.
Therefore, the request alignment for writes must be the job's
granularity (in write-blocking mode).
Unfortunately, this forces all reads and writes to have the same
granularity (we only need this alignment for writes to the target, not
the source), but that is something to be fixed another time.
Cc: qemu-stable@nongnu.org
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190805153308.2657-1-mreitz@redhat.com
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Fixes: d06107ade0
Signed-off-by: Max Reitz <mreitz@redhat.com>
Perform two guest writes to not yet backed up areas of an image, where
the former touches an inner area of the latter.
Before HEAD^, copy offloading broke this in two ways:
(1) The target image differs from the reference image (what the source
was when the backup started).
(2) But you will not see that in the failing output, because the job
offset is reported as being greater than the job length. This is
because one cluster is copied twice, and thus accounted for twice,
but of course the job length does not increase.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190801173900.23851-3-mreitz@redhat.com
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Tested-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
The backup job must only copy areas that the copy_bitmap reports as
dirty. This is always the case when using traditional non-offloading
backup, because it copies each cluster separately. When offloading the
copy operation, we sometimes copy more than one cluster at a time, but
we only check whether the first one is dirty.
Therefore, whenever copy offloading is possible, the backup job
currently produces wrong output when the guest writes to an area of
which an inner part has already been backed up, because that inner part
will be re-copied.
Fixes: 9ded4a0114
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-id: 20190801173900.23851-2-mreitz@redhat.com
Cc: qemu-stable@nongnu.org
Signed-off-by: Max Reitz <mreitz@redhat.com>
In Arm v8.0 M-profile CPUs without the Security Extension and also in
v7M CPUs, there is no NSACR register. However, the code we have to handle
the FPU does not always check whether the ARM_FEATURE_M_SECURITY bit
is set before testing whether env->v7m.nsacr permits access to the
FPU. This means that for a CPU with an FPU but without the Security
Extension we would always take a bogus fault when trying to stack
the FPU registers on an exception entry.
We could fix this by adding extra feature bit checks for all uses,
but it is simpler to just make the internal value of nsacr 0xcff
("all non-secure accesses allowed"), since this is not guest
visible when the Security Extension is not present. This allows
us to continue to follow the Arm ARM pseudocode which takes a
similar approach. (In particular, in the v8.1 Arm ARM the register
is documented as reading as 0xcff in this configuration.)
Fixes: https://bugs.launchpad.net/qemu/+bug/1838475
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Damien Hedde <damien.hedde@greensocs.com>
Message-id: 20190801105742.20036-1-peter.maydell@linaro.org
ACS got added in 4.0 unconditionally, that broke older<->4.0 migration
where there was a PCIe root port.
Fix this by turning it off for 3.1 and older machines; note this
fixes compatibility for older QEMUs but breaks compatibility with 4.0
for older machine types.
machine type source qemu dest qemu
3.1 3.1 4.0 broken
3.1 3.1 4.1rc2 broken
3.1 3.1 4.1+this OK ++
3.1 4.0 4.1rc2 OK
3.1 4.0 4.1+this broken --
4.0 4.0 4.1rc2 OK
4.0 4.0 4.1+this OK
So we gain and lose; the consensus seems to be treat this as a
fix for older machine types.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20190730093719.12958-3-dgilbert@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.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>
ACS was added in 4.0 unconditionally, this breaks migration
compatibility.
Allow ACS to be disabled by adding a property that's
checked by pcie_root_port.
Unfortunately pcie-root-port doesn't have any instance data,
so there's no where for that flag to live, so stuff it into
PCIESlot.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20190730093719.12958-2-dgilbert@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Most Arm architectural debug exceptions (eg watchpoints) are ignored
if the configured "debug exception level" is below the current
exception level (so for example EL1 can't arrange to get debug exceptions
for EL2 execution). Exceptions generated by the BRK or BPKT instructions
are a special case -- they must always cause an exception, so if
we're executing above the debug exception level then we
must take them to the current exception level.
This fixes a bug where executing BRK at EL2 could result in an
exception being taken at EL1 (which is strictly forbidden by the
architecture).
Fixes: https://bugs.launchpad.net/qemu/+bug/1838277
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20190730132522.27086-1-peter.maydell@linaro.org
In order to insert a read-only medium (i.e. a read-only block node) to
the BlockBackend of a floppy drive, we must not have taken write
permissions on that BlockBackend, or the operation will fail with the
error message "Block node is read-only".
The device already takes care to remove all permissions when the medium
is ejected, but the state isn't correct if the drive is initially empty:
It uses blk_is_read_only() to check whether write permissions should be
taken, but this function returns false for empty BlockBackends in the
common case.
Fix floppy_drive_realize() to avoid taking write permissions if the
drive is empty.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
scsi-disks decides whether it has a read-only device by looking at
whether the BlockBackend specified as drive=... is read-only. In the
case of an anonymous BlockBackend (with a node name specified in
drive=...), this is the read-only flag of the attached node. In the case
of an empty anonymous BlockBackend, it's always read-write because
nothing prevented it from being read-write.
This is a problem because scsi-cd would take write permissions on the
anonymous BlockBackend of an empty drive created without a drive=...
option. Using blockdev-insert-medium with a read-only node fails then
with the error message "Block node is read-only".
Fix scsi_realize() so that scsi-cd devices always take read-only
permissions on their BlockBackend instead.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1733920
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
The copy-on-read drive must not request the WRITE_UNCHANGED permission
for its child if the node is inactive, otherwise starting a migration
destination with -incoming will fail because the child cannot provide
write access yet:
qemu-system-x86_64: -blockdev copy-on-read,file=img,node-name=cor: Block node is read-only
Earlier QEMU versions additionally ran into an abort() on the migration
source side: bdrv_inactivate_recurse() failed to update permissions.
This is silently ignored today because it was only supposed to loosen
restrictions. This is the symptom that was originally reported here:
https://bugzilla.redhat.com/show_bug.cgi?id=1733022
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
While older toolchains produced binaries where the physical load address
of ELF segments was the same as the virtual address, newer versions seem
to choose a different physical address if it isn't specified explicitly.
The means that the test kernel doesn't use the right addresses to access
e.g. format strings any more and the whole output disappears, causing
all test cases to fail.
Fix this by specifying the physical load address of sections explicitly.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
QEMU will crash with:
Segmentation fault (core dumped)
when negative slot number is used, ex:
qemu-system-x86_64 -m 1G,maxmem=20G,slots=256 \
-object memory-backend-ram,id=mem1,size=1G \
-device pc-dimm,id=dimm1,memdev=mem1,slot=-2
fix it by checking that slot number is within valid range.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <20190723160859.27250-1-imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Li Qiang <liq3ea@gmail.com>
Signed-off-by: Igor Mammedov <<a href="mailto:imammedo@redhat.com" target="_blank">imammedo@redhat.com</a>><br></blockquote><div><br></div><div>Reviewed-by: Li Qiang <<a href="mailto:liq3ea@gmail.com">liq3ea@gmail.com</a>><br></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
Reviewed-by: Pankaj Gupta <pagupta@redhat.com>
