translate.c can not be compiled with --disable-tcg, but we need
the s390_cpu_dump_state() in KVM-only builds, too. So let's move
that function to helper.c instead, which will also be compiled
when --disable-tcg has been specified.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Message-Id: <1500886370-14572-2-git-send-email-thuth@redhat.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
There are certain features that we put into base models, but that are
not relevant for the actual search. The most famous example are
MSA subfunctions that might be disabled on certain real hardware out
there.
While the kvm host model detection will usually detect the correct model
on such machines (as it will in the common case not pass features to check
for into s390_find_cpu_def()), baselining will fall back to a quite old
model just because some MSA subfunctions are missing.
Let's improve that by ignoring lack of these features while performing
the search for a base model.
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20170720123721.12366-6-david@redhat.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
Using ordinary bitmap operations to set/test bits does not work properly
on architectures !s390x. Let's drop (test|set)_bit_inv and introduce
(test|set)_be_bit instead. These functions work on uint8_t array, not on
unsigned longs arrays and are for now only used in the context of
CPU features.
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20170720123721.12366-4-david@redhat.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
We'll have to do the same for TCG, so let's just move it in there.
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20170720123721.12366-3-david@redhat.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
Unused and broken, let's just get rid of it.
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20170720123721.12366-2-david@redhat.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
The SIE_KSS feature will allow a guest to use KSS for a nested guest.
To create a nested guest the SIE_F2 facility is still necessary.
Since SIE_F2 is not part of the default model it does not make
a lot of sense to provide the SIE_KSS feature in the default model.
Let's also create a dependency check.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: Jason J. Herne <jjherne@linux.vnet.ibm.com>
Reviewed-by: Janosch Frank <frankja@linux.vnet.ibm.com>
Message-Id: <1500550051-7821-2-git-send-email-borntraeger@de.ibm.com>
Acked-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
Commit 7da624e2 ("vfio: Test realized when using VFIOGroup.device_list
iterator") introduced a pointer to the Object DeviceState in the VFIO
common base-device and skipped non-realized devices as we iterate
VFIOGroup.device_list. While it missed to initialize the pointer for
the vfio-ccw case. Let's fix it.
Fixes: 7da624e2 ("vfio: Test realized when using VFIOGroup.device_list
iterator")
Cc: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Halil Pasic <pasic@linux.vnet.ibm.com>
Signed-off-by: Dong Jia Shi <bjsdjshi@linux.vnet.ibm.com>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Message-Id: <20170718014926.44781-3-bjsdjshi@linux.vnet.ibm.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
When allocating memory for the vfio_irq_info parameter of the
VFIO_DEVICE_GET_IRQ_INFO ioctl, we used the wrong size. Let's
fix it by using the right size.
Reviewed-by: Dong Jia Shi <bjsdjshi@linux.vnet.ibm.com>
Signed-off-by: Jing Zhang <bjzhjing@linux.vnet.ibm.com>
Signed-off-by: Dong Jia Shi <bjsdjshi@linux.vnet.ibm.com>
Message-Id: <20170718014926.44781-2-bjsdjshi@linux.vnet.ibm.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
The main changes are:
- fixes in PCI bridges code;
- LUN>255 are allowed not in virtio-scsi.
The full list is:
> pci-scan: Fix pci-bridge-set-mem-base and pci-bridge-set-mem-limit
> pci: Avoid 32-bit prefetchable memory area if possible
> Remove unused functions ishexdigit and $cat-comma
> pci: Translate PCI addresses to host addresses at the end of map-in
> Define 'open' and 'close' words of the /aliases nodes right from the start
> virtio-scsi: Allow LUNs bigger than 255
> paflof: Silence gcc's -Warray-bounds warning for stack pointers
> board_qemu: move code out of fdt-fix-node-phandle
> board_qemu: drop unused values early in fdt-fix-node-phandle
> pci: Improve the pci-var-out debug function
> libhvcall: drop unused KVMPPC_H_REPORT_MC_ERR and KVMPPC_H_NMI_MCE defines
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Commit 0cffce56 (hw/ppc/spapr.c: adding pending_dimm_unplugs to
sPAPRMachineState) introduced a new way to track pending LMBs of DIMM
device that is marked for removal. Since this commit we can hit the
assert in spapr_pending_dimm_unplugs_add() in the following situation:
- DIMM device removal fails as the guest doesn't allow the removal.
- Subsequent attempt to remove the same DIMM would hit the assert
as the corresponding sPAPRDIMMState is still part of the
pending_dimm_unplugs list.
Fix this by removing the assert and conditionally adding the
sPAPRDIMMState to pending_dimm_unplugs list only when it is not
already present.
Fixes: 0cffce56ae
Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
[dwg: Tweaked to avoid returning NULL when spapr_pending_dimm_unplugs_add()
does find an existing entry]
Reviewed-by: Daniel Barboza <danielhb@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Commit 3a38429 ("spapr: Add a "no HPT" encoding to HTAB migration stream")
allows to migrate an empty HPT, but doesn't mark correctly the
end of the migration stream.
The end condition (value returned by htab_save_iterate())
should be 1, whereas in 3a38429 it returns 0.
The problem can be reproduced with QEMU monitor command "savevm":
the command never stops and the disk image grows without limit.
Fixes: 3a38429748
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
f1c2dc7c86 "spapr-pci: rework MSI/MSIX" (07/2013) changed MSIX encoding
but forgot to change the comment so this changes it.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Instead of migrating the flash by creating the memory region
with memory_region_init_ram_nomigrate() and then calling
vmstate_register_ram_global(), just use memory_region_init_ram(),
which now handles migration registration automatically.
This is a migration compatibility break for the integratorcp
board, because the RAM region's migration name changes to
include the device path. This is OK because we don't guarantee
migration compatibility for this board.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1500310341-28931-1-git-send-email-peter.maydell@linaro.org
A cut-and-paste error meant that instead of setting the
qdev parent bus for the SCC device we were setting it
twice for the ARMv7M container device. Fix this bug.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1500634509-28011-1-git-send-email-peter.maydell@linaro.org
The fsl-imx* boards accidentally forgot to register the ROM memory
regions for migration. This used to require a manual step of calling
vmstate_register_ram(), but following commits
1cfe48c1ce21..b08199c6fbea194 we can use memory_region_init_rom() to
have it do the migration for us.
This is a migration break, but the migration code currently does not
handle the case of having two RAM regions which were not registered
for migration, and so prior to this commit a migration load would
always fail with:
"qemu-system-arm: Length mismatch: 0x4000 in != 0x18000: Invalid argument"
NB: migration appears at this point to be broken for this board
anyway -- it succeeds but the destination hangs; probably some
device in the system does not yet support migration.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1500309775-18361-1-git-send-email-peter.maydell@linaro.org
Fix a TCG temporary leak in the new aarch64 rev16 handling.
Signed-off-by: Emilio G. Cota <cota@braap.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
HACKING recommends listing system includes right after osdep.h,
and before any other in-project headers.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <20170721135047.25005-3-eblake@redhat.com>
qemu-io and qemu-img already mirror the qemu version string,
time to make qemu-nbd do the same.
Reported-by: 陳培泓 <pahome.chen@mirlab.org>
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <20170721135047.25005-2-eblake@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Test cases 030, 041 and 055 used to sleep for a second after calling
block-job-pause to make sure that the block job had time to actually
get into paused state. We can instead poll its status and use that one
second only as a timeout.
The tests also slept a second for checking that the block jobs don't
make progress while being paused. Half a second is more than enough for
this.
These changes reduce the total time for the three tests by 25 seconds on
my laptop (from 155 seconds to 130).
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Commits 0db832f and 6cdbceb introduced the automatic insertion of filter
nodes above the top layer of mirror and commit block jobs. The
assumption made there was that since libvirt doesn't do node-level
management of the block layer yet, it shouldn't be affected by added
nodes.
This is true as far as commands issued by libvirt are concerned. It only
uses BlockBackend names to address nodes, so any operations it performs
still operate on the root of the tree as intended.
However, the assumption breaks down when you consider query commands,
which return data for the wrong node now. These commands also return
information on some child nodes (bs->file and/or bs->backing), which
libvirt does make use of, and which refer to the wrong nodes, too.
One of the consequences is that oVirt gets wrong information about the
image size and stops the VM in response as long as a mirror or commit
job is running:
https://bugzilla.redhat.com/show_bug.cgi?id=1470634
This patch fixes the problem by hiding the implicit nodes created
automatically by the mirror and commit block jobs in the output of
query-block and BlockBackend-based query-blockstats as long as the user
doesn't indicate that they are aware of those nodes by providing a node
name for them in the QMP command to start the block job.
The node-based commands query-named-block-nodes and query-blockstats
with query-nodes=true still show all nodes, including implicit ones.
This ensures that users that are capable of node-level management can
still access the full information; users that only know BlockBackends
won't use these commands.
Cc: qemu-stable@nongnu.org
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Tested-by: Eric Blake <eblake@redhat.com>
We used MAX() instead of the intended MIN() when computing how many
sectors to view in the current loop iteration of qcow2_measure(),
and passed in a value of INT_MAX sectors instead of our more usual
limit of BDRV_REQUEST_MAX_SECTORS (the latter avoids 32-bit overflow
on conversion to bytes). For small files, the bug is harmless:
bdrv_get_block_status_above() clamps its *pnum answer to the BDS
size, regardless of any insanely larger input request. However, for
any file at least 2T in size, we can very easily end up going into an
infinite loop (the maximum of 0x100000000 sectors and INT_MAX is a
64-bit quantity, which becomes 0 when assigned to int; once nb_sectors
is 0, we never make progress).
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
We've been documenting the value in bytes since its introduction
in commit b9a9b3a4 (v1.3), where it was actually reported in bytes.
Commit e4654d2 (v2.0) then removed things from block/qapi.c, in
preparation for a rewrite to a list of dirty sectors in the next
commit 21b5683 in block.c, but the new code mistakenly started
reporting in sectors.
Fixes: https://bugzilla.redhat.com/1441460
CC: qemu-stable@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
A run of './check -qcow2 -g quick' on my machine produced only
two tests that took longer than 5 seconds; 178 took 18, and
189 took 7. Remove them from the quick group.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Reviewed-by: Jeff Cody <jcody@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
-----BEGIN PGP SIGNATURE-----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=DCKd
-----END PGP SIGNATURE-----
Merge remote-tracking branch 'remotes/armbru/tags/pull-qapi-2017-07-18-v2' into staging
QAPI patches for 2017-07-18
# gpg: Signature made Mon 24 Jul 2017 12:40:56 BST
# gpg: using RSA key 0x3870B400EB918653
# gpg: Good signature from "Markus Armbruster <armbru@redhat.com>"
# gpg: aka "Markus Armbruster <armbru@pond.sub.org>"
# Primary key fingerprint: 354B C8B3 D7EB 2A6B 6867 4E5F 3870 B400 EB91 8653
* remotes/armbru/tags/pull-qapi-2017-07-18-v2:
migration: Use JSON null instead of "" to reset parameter to default
migration: Unshare MigrationParameters struct for now
migration: Add TODO comments on duplication of QAPI_CLONE()
migration: Clean up around tls_creds, tls_hostname
hmp: Clean up and simplify hmp_migrate_set_parameter()
block: Use JSON null instead of "" to disable backing file
tests/test-qobject-input-visitor: Drop redundant test
qapi: Introduce a first class 'null' type
qapi: Use QNull for a more regular visit_type_null()
qapi: Separate type QNull from QObject
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Commit 97f4030 changed warning messages from
timestamp-if-enabled progname ":" location "warning: " message
to
"warning: " timestamp-if-enabled progname ":" location message
This regressed qemu-iotests 051. Put "warning: " right back where it
was, along with "info: ".
Reported-by: Kevin Wolf <kwolf@redhat.com>
Cc: Alistair Francis <alistair.francis@xilinx.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1500449614-16811-1-git-send-email-armbru@redhat.com>
Reviewed-by: Alistair Francis <alistair.francis@xilinx.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Clang 3.9 passes the CONFIG_AVX2_OPT configure test. However, the
supplied <cpuid.h> does not contain the bit_AVX2 define that we use
when detecting whether the routine can be enabled.
Introduce a qemu-specific header that uses the compiler's definition
of __cpuid et al, but supplies any missing bit_* definitions needed.
This avoids introducing any extra ifdefs to util/bufferiszero.c, and
allows quite a few to be removed from tcg/i386/tcg-target.inc.c.
Signed-off-by: Richard Henderson <rth@twiddle.net>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20170719044018.18063-1-rth@twiddle.net
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
migrate-set-parameters sets migration parameters according to is
arguments like this:
* Present means "set the parameter to this value"
* Absent means "leave the parameter unchanged"
* Except for parameters tls_creds and tls_hostname, "" means "reset
the parameter to its default value
The first two are perfectly normal: presence of the parameter makes
the command do something.
The third one overloads the parameter with a second meaning. The
overloading is *implicit*, i.e. it's not visible in the types. Works
here, because "" is neither a valid TLS credentials ID, nor a valid
host name.
Pressing argument values the schema accepts, but are semantically
invalid, into service to mean "reset to default" is not general, as
suitable invalid values need not exist. I also find it ugly.
To clean this up, we could add a separate flag argument to ask for
"reset to default", or add a distinct value to @tls_creds and
@tls_hostname. This commit implements the latter: add JSON null to
the values of @tls_creds and @tls_hostname, deprecate "".
Because we're so close to the 2.10 freeze, implement it in the
stupidest way possible: have qmp_migrate_set_parameters() rewrite null
to "" before anything else can see the null. The proper way to do it
would be rewriting "" to null, but that requires fixing up code to
work with null. Add TODO comments for that.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Commit de63ab6 "migrate: Share common MigrationParameters struct"
reused MigrationParameters for the arguments of
migrate-set-parameters, with the following rationale:
It is rather verbose, and slightly error-prone, to repeat
the same set of parameters for input (migrate-set-parameters)
as for output (query-migrate-parameters), where the only
difference is whether the members are optional. We can just
document that the optional members will always be present
on output, and then share a common struct between both
commands. The next patch can then reduce the amount of
code needed on input.
I need to unshare them to correct a design flaw in a stupid, but
minimally invasive way, in the next commit. We can restore the
sharing when we redo that patch in a less stupid way. Add a suitable
TODO comment.
Note that I revert only the sharing part of commit de63ab6, not the
part that made the members of query-migrate-parameters' result
optional. The schema (and thus introspection) remains inaccurate for
query-migrate-parameters. If we decide not to restore the sharing, we
should revert that part, too.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
qmp_query_migrate_parameters() and qmp_migrate_set_parameters()
effectively duplicate QAPI_CLONE() inline. Add suitable TODO
comments.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Optional MigrationParameters members tls_creds and tls_hostname can't
actually be absent outside qmp_migrate_set_parameters() since commit
4af245d (v2.9.0).
Note that commit 4af245d reverted the part of commit de63ab6 (v2.8.0)
that made tls_creds and tls_hostname absent instead of "" in the value
of query-migrate-parameters, even though commit de63ab6 called that a
mistake. What a mess.
Drop the redundant tests for presence, and update documentation.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
The bulk of hmp_migrate_set_parameter()'s code sets one member of
MigrationParameters according to the command's arguments. It uses a
string visitor for integer and boolean members, but not for string and
size members. It calls visit_type_bool() right away, but delays
visit_type_int() some. The delaying requires a flag variable and a
bit of trickery: we set all integer members instead of just the one we
want, and rely on the has_FOOs to mask the unwanted ones.
Clean this up as follows. Don't delay calling visit_type_int(). Use
the string visitor for strings, too. This involves extra allocations
and cleanup, but doing them is simpler and cleaner than avoiding them.
Sadly, using the string visitor for sizes isn't possible, because it
defaults to Bytes rather than Mebibytes. Add a comment explaining
that.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
BlockdevRef is an alternate of BlockdevOptions (inline definition) and
str (reference to an existing block device by name). BlockdevRef
value "" is special: "no block device should be referenced." It's
actually interpreted that way in just one place: optional member
@backing of COW formats. Semantics:
* Present means "use this block device" as backing storage
* Absent means "default to the one stored in the image"
* Except "" means "don't use backing storage at all"
The first two are perfectly normal: when the parameter is absent, it
defaults to an implied value, but the value's meaning is the same.
The third one overloads the parameter with a second meaning. The
overloading is *implicit*, i.e. it's not visible in the types. Works
here, because "" is not a value block device ID.
Pressing argument values the schema accepts, but are semantically
invalid, into service to mean "do something else entirely" is not
general, as suitable invalid values need not exist. I also find it
ugly.
To clean this up, we could add a separate flag argument to suppress
@backing, or add a distinct value to @backing. This commit implements
the latter: add JSON null to the values of @backing, deprecate "".
Because we're so close to the 2.10 freeze, implement it in the
stupidest way possible: have qmp_blockdev_add() rewrite null to ""
before anything else can see the null. Works, because BlockdevRef
occurs only within arguments of blockdev-add. The proper way to do it
would be rewriting "" to null, preferably in a cleaner way, but that
requires fixing up code to work with null. Add a TODO comment for
that.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Acked-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
test_visitor_in_alternate() tests UserDefAlternate with inadmissible
input. test_visitor_in_fail_alternate() does basically the same.
Drop the former, keep the latter.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
I expect the 'null' type to be useful mostly for members of alternate
types.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Make visit_type_null() take an @obj argument like its buddies. This
helps keep the next commit simple.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Under certain circumstances normal xen-mapcache functioning may be broken
by guest's actions. This may lead to either QEMU performing exit() due to
a caught bad pointer (and with QEMU process gone the guest domain simply
appears hung afterwards) or actual use of the incorrect pointer inside
QEMU address space -- a write to unmapped memory is possible. The bug is
hard to reproduce on a i440 machine as multiple DMA sources are required
(though it's possible in theory, using multiple emulated devices), but can
be reproduced somewhat easily on a Q35 machine using an emulated AHCI
controller -- each NCQ queue command slot may be used as an independent
DMA source ex. using READ FPDMA QUEUED command, so a single storage
device on the AHCI controller port will be enough to produce multiple DMAs
(up to 32). The detailed description of the issue follows.
Xen-mapcache provides an ability to map parts of a guest memory into
QEMU's own address space to work with.
There are two types of cache lookups:
- translating a guest physical address into a pointer in QEMU's address
space, mapping a part of guest domain memory if necessary (while trying
to reduce a number of such (re)mappings to a minimum)
- translating a QEMU's pointer back to its physical address in guest RAM
These lookups are managed via two linked-lists of structures.
MapCacheEntry is used for forward cache lookups, while MapCacheRev -- for
reverse lookups.
Every guest physical address is broken down into 2 parts:
address_index = phys_addr >> MCACHE_BUCKET_SHIFT;
address_offset = phys_addr & (MCACHE_BUCKET_SIZE - 1);
MCACHE_BUCKET_SHIFT depends on a system (32/64) and is equal to 20 for
a 64-bit system (which assumed for the further description). Basically,
this means that we deal with 1 MB chunks and offsets within those 1 MB
chunks. All mappings are created with 1MB-granularity, i.e. 1MB/2MB/3MB
etc. Most DMA transfers typically are less than 1MB, however, if the
transfer crosses any 1MB border(s) - than a nearest larger mapping size
will be used, so ex. a 512-byte DMA transfer with the start address
700FFF80h will actually require a 2MB range.
Current implementation assumes that MapCacheEntries are unique for a given
address_index and size pair and that a single MapCacheEntry may be reused
by multiple requests -- in this case the 'lock' field will be larger than
1. On other hand, each requested guest physical address (with 'lock' flag)
is described by each own MapCacheRev. So there may be multiple MapCacheRev
entries corresponding to a single MapCacheEntry. The xen-mapcache code
uses MapCacheRev entries to retrieve the address_index & size pair which
in turn used to find a related MapCacheEntry. The 'lock' field within
a MapCacheEntry structure is actually a reference counter which shows
a number of corresponding MapCacheRev entries.
The bug lies in ability for the guest to indirectly manipulate with the
xen-mapcache MapCacheEntries list via a special sequence of DMA
operations, typically for storage devices. In order to trigger the bug,
guest needs to issue DMA operations in specific order and timing.
Although xen-mapcache is protected by the mutex lock -- this doesn't help
in this case, as the bug is not due to a race condition.
Suppose we have 3 DMA transfers, namely A, B and C, where
- transfer A crosses 1MB border and thus uses a 2MB mapping
- transfers B and C are normal transfers within 1MB range
- and all 3 transfers belong to the same address_index
In this case, if all these transfers are to be executed one-by-one
(without overlaps), no special treatment necessary -- each transfer's
mapping lock will be set and then cleared on unmap before starting
the next transfer.
The situation changes when DMA transfers overlap in time, ex. like this:
|===== transfer A (2MB) =====|
|===== transfer B (1MB) =====|
|===== transfer C (1MB) =====|
time --->
In this situation the following sequence of actions happens:
1. transfer A creates a mapping to 2MB area (lock=1)
2. transfer B (1MB) tries to find available mapping but cannot find one
because transfer A is still in progress, and it has 2MB size + non-zero
lock. So transfer B creates another mapping -- same address_index,
but 1MB size.
3. transfer A completes, making 1st mapping entry available by setting its
lock to 0
4. transfer C starts and tries to find available mapping entry and sees
that 1st entry has lock=0, so it uses this entry but remaps the mapping
to a 1MB size
5. transfer B completes and by this time
- there are two locked entries in the MapCacheEntry list with the SAME
values for both address_index and size
- the entry for transfer B actually resides farther in list while
transfer C's entry is first
6. xen_ram_addr_from_mapcache() for transfer B gets correct address_index
and size pair from corresponding MapCacheRev entry, but then it starts
looking for MapCacheEntry with these values and finds the first entry
-- which belongs to transfer C.
At this point there may be following possible (bad) consequences:
1. xen_ram_addr_from_mapcache() will use a wrong entry->vaddr_base value
in this statement:
raddr = (reventry->paddr_index << MCACHE_BUCKET_SHIFT) +
((unsigned long) ptr - (unsigned long) entry->vaddr_base);
resulting in an incorrent raddr value returned from the function. The
(ptr - entry->vaddr_base) expression may produce both positive and negative
numbers and its actual value may differ greatly as there are many
map/unmap operations take place. If the value will be beyond guest RAM
limits then a "Bad RAM offset" error will be triggered and logged,
followed by exit() in QEMU.
2. If raddr value won't exceed guest RAM boundaries, the same sequence
of actions will be performed for xen_invalidate_map_cache_entry() on DMA
unmap, resulting in a wrong MapCacheEntry being unmapped while DMA
operation which uses it is still active. The above example must
be extended by one more DMA transfer in order to allow unmapping as the
first mapping in the list is sort of resident.
The patch modifies the behavior in which MapCacheEntry's are added to the
list, avoiding duplicates.
Signed-off-by: Alexey Gerasimenko <x1917x@gmail.com>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
Solaris 9 was released in 2002, its successor Solaris 10 was
released in 2005, and Solaris 9 was end-of-lifed in 2014.
Nobody has stepped forward to express interest in supporting
Solaris of any flavour, so removing support for the ancient
versions seems uncontroversial.
In particular, this allows us to remove a use of 'uname'
in configure that won't work if you're cross-compiling.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Message-id: 1499955697-28045-1-git-send-email-peter.maydell@linaro.org
For a very long time we have used 'uname -s' as our fallback if
we don't identify the target OS using a compiler #define. This
obviously doesn't work for cross-compilation, and we've had
a comment suggesting we fix this in configure for a long time.
Since we now have an exhaustive list of which OSes we can run
on (thanks to commit 898be3e041 making an unrecognized OS
be a fatal error), we know which ones we're missing.
Add check_define tests for the remaining OSes we support. The
defines checked are based on ones we already use in the codebase for
identifying the host OS (with the exception of GNU/kFreeBSD).
We can now set bogus_os immediately rather than doing it later.
We leave the comment about uname being bad untouched, since
there is still a use of it for the fallback for unrecognized
host CPU type.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-id: 1499958932-23839-1-git-send-email-peter.maydell@linaro.org
On OpenBSD the compiler warns:
bsd-user/main.c:622:21: warning: variable 'sig' set but not used [-Wunused-but-set-variable]
This is because a lot of the signal delivery code is #if-0'd
out as unused. Reshuffle #ifdefs a bit to silence the warning.
(We make the minimum change here rather than removing all the
bsd-user patchset which should make this all work correctly and
there's no point giving them an awkward rebase task.)
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-id: 1500395194-21455-5-git-send-email-peter.maydell@linaro.org
On OpenBSD the compiler complains:
bsd-user/bsdload.c:54:17: warning: variable 'id_change' set but not used [-Wunused-but-set-variable]
This is dead code that was originally copied from linux-user.
We fixed this in linux-user in commit 331c23b5ca in 2011;
delete the useless code from bsd-user too.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-id: 1500395194-21455-4-git-send-email-peter.maydell@linaro.org
Since commit cfc87e00 "block/vpc.c: Handle write failures in
get_image_offset()" older versions of gcc (in this case 4.7) incorrectly
warn that "ret" can be used uninitialised in vpc_co_pwritev().
Setting ret to 0 at the start of vpc_co_pwritev() prevents the warning
in gcc 4.7 and enables compilation with -Werror to succeed.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-id: 1500625265-23844-1-git-send-email-mark.cave-ayland@ilande.co.uk
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Changes:
* Add Enhanced Virtual Addressing (EVA) support
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)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=qlJH
-----END PGP SIGNATURE-----
Merge remote-tracking branch 'remotes/yongbok/tags/mips-20170721' into staging
MIPS patches 2017-07-21
Changes:
* Add Enhanced Virtual Addressing (EVA) support
# gpg: Signature made Fri 21 Jul 2017 03:25:15 BST
# gpg: using RSA key 0x2238EB86D5F797C2
# gpg: Good signature from "Yongbok Kim <yongbok.kim@imgtec.com>"
# gpg: WARNING: This key is not certified with a trusted signature!
# gpg: There is no indication that the signature belongs to the owner.
# Primary key fingerprint: 8600 4CF5 3415 A5D9 4CFA 2B5C 2238 EB86 D5F7 97C2
* remotes/yongbok/tags/mips-20170721:
target/mips: Enable CP0_EBase.WG on MIPS64 CPUs
target/mips: Add EVA support to P5600
target/mips: Implement segmentation control
target/mips: Add segmentation control registers
target/mips: Add an MMU mode for ERL
target/mips: Abstract mmu_idx from hflags
target/mips: Check memory permissions with mem_idx
target/mips: Decode microMIPS EVA load & store instructions
target/mips: Decode MIPS32 EVA load & store instructions
target/mips: Prepare loads/stores for EVA
target/mips: Add CP0_Ebase.WG (write gate) support
target/mips: Weaken TLB flush on UX,SX,KX,ASID changes
target/mips: Fix TLBWI shadow flush for EHINV,XI,RI
target/mips: Fix MIPS64 MFC0 UserLocal on BE host
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>