The visitor interface for mapping between QObject/QemuOpts/string
and QAPI is scandalously under-documented, making changes to visitor
core, individual visitors, and users of visitors difficult to
coordinate. Among other questions: when is it safe to pass NULL,
vs. when a string must be provided; which visitors implement which
callbacks; the difference between concrete and virtual visits.
Correct this by retrofitting proper contracts, and document where some
of the interface warts remain (for example, we may want to modify
visit_end_* to require the same 'obj' as the visit_start counterpart,
so the dealloc visitor can be simplified). Later patches in this
series will tackle some, but not all, of these warts.
Add assertions to (partially) enforce the contract. Some of these
were only made possible by recent cleanup commits.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <1461879932-9020-13-git-send-email-eblake@redhat.com>
[Doc fix from Eric squashed in]
Signed-off-by: Markus Armbruster <armbru@redhat.com>
In the QMP input visitor, visiting a list traverses two objects:
the QAPI GenericList of the caller (which gets advanced in
visit_next_list() regardless of this patch), and the QList input
that we are converting to QAPI. For consistency with QDict
visits, we want to consume elements from the input QList during
the visit_type_FOO() for the list element; that is, we want ALL
the code for consuming an input to live in qmp_input_get_object(),
rather than having it split according to whether we are visiting
a dict or a list. Making qmp_input_get_object() the common point
of consumption will make it easier for a later patch to refactor
visit_start_list() to cover the GenericList * head of a QAPI list,
and in turn will get rid of the 'first' flag (which lived in
qmp_input_next_list() pre-patch, and is hoisted to StackObject
by this patch).
This patch is therefore altering the post-condition use of 'entry',
while keeping what gets visited unchanged, from:
start_list next_list type_ELT ... next_list type_ELT next_list end_list
visits 1st elt last elt
entry NULL 1st elt 1st elt last elt last elt NULL gone
where type_ELT() returns (entry ? entry : 1st elt) and next_list() steps
entry
to this usage:
start_list next_list type_ELT ... next_list type_ELT next_list end_list
visits 1st elt last elt
entry 1st elt 1nd elt 2nd elt last elt NULL NULL gone
where type_ELT() steps entry and returns the old entry, and next_list()
leaves entry alone.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <1461879932-9020-12-git-send-email-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Don't embed the root of the visit into the stack of current
containers being visited. That way, we no longer get confused
on whether the first visit of a dictionary is to the dictionary
itself or to one of the members of the dictionary, based on
whether the caller passed name=NULL; and makes the QMP Input
visitor like other visitors where the value of 'name' is now
ignored on the root visit. (We may someday want to revisit
the rules on what 'name' should be on a top-level visit,
rather than just ignoring it; but that would be the topic of
another patch).
An audit of all qmp_input_visitor_new() call sites shows that
there were only two places where callers had previously been
visiting to a QDict with a non-NULL name to bypass a call to
visit_start_struct(), and those were fixed in prior patches.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <1461879932-9020-11-git-send-email-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
The qmp-input visitor was allowing callers to play rather fast
and loose: when visiting a QDict, you could grab members of the
root dictionary without first pushing into the dict; the final
such culprit was the QOM code for converting to and from object
properties. But we are about to tighten the input visitor, at
which point user_creatable_add_type() as called with a QMP input
visitor via qmp_object_add() MUST follow the same paradigms as
everyone else, of pushing into the struct before grabbing its
keys.
The use of 'err ? NULL : &err' is temporary; a later patch will
clean that up when it splits visit_end_struct().
Furthermore, note that both callers always pass qdict, so we can
convert the conditional into an assert and reduce indentation.
The change has no impact to the testsuite now, but is required to
avoid a failure in tests/test-netfilter once qmp-input is made
stricter to detect inconsistent 'name' arguments on the root visit.
Since user_creatable_add_type() is also called with OptsVisitor
through user_creatable_add_opts(), we must also check that there
is no negative impact there; both pre- and post-patch, we see:
$ ./x86_64-softmmu/qemu-system-x86_64 -nographic -nodefaults -qmp stdio -object secret,id=sec0,data=letmein,format=raw,foo=bar
qemu-system-x86_64: -object secret,id=sec0,data=letmein,format=raw,foo=bar: Property '.foo' not found
That is, the only new checking that the new visit_end_struct() can
perform is for excess input, but we already catch excess input
earlier in object_property_set().
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <1461879932-9020-10-git-send-email-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
The qmp-input visitor was allowing callers to play rather fast
and loose: when visiting a QDict, you could grab members of the
root dictionary without first pushing into the dict; among the
culprit callers was the generated marshal code on the 'arguments'
dictionary of a QMP command. But we are about to tighten the
input visitor, at which point the generated marshal code MUST
follow the same paradigms as everyone else, of pushing into the
struct before grabbing its keys.
Generated code grows as follows:
|@@ -515,7 +641,12 @@ void qmp_marshal_blockdev_backup(QDict *
| BlockdevBackup arg = {0};
|
| v = qmp_input_get_visitor(qiv);
|+ visit_start_struct(v, NULL, NULL, 0, &err);
|+ if (err) {
|+ goto out;
|+ }
| visit_type_BlockdevBackup_members(v, &arg, &err);
|+ visit_end_struct(v, err ? NULL : &err);
| if (err) {
| goto out;
| }
|@@ -527,7 +715,9 @@ out:
| qmp_input_visitor_cleanup(qiv);
| qdv = qapi_dealloc_visitor_new();
| v = qapi_dealloc_get_visitor(qdv);
|+ visit_start_struct(v, NULL, NULL, 0, NULL);
| visit_type_BlockdevBackup_members(v, &arg, NULL);
|+ visit_end_struct(v, NULL);
| qapi_dealloc_visitor_cleanup(qdv);
| }
The use of 'err ? NULL : &err' is temporary; a later patch will
clean that up when it splits visit_end_struct().
Prior to this patch, the fact that there was no final
visit_end_struct() meant that even though we are using a strict
input visit, the marshalling code was not detecting excess input
at the top level (only in nested levels). Fortunately, we have
code in monitor.c:qmp_check_client_args() that also checks for
no excess arguments at the top level. But as the generated code
is more compact than the manual check, a later patch will clean
up monitor.c to drop the redundancy added here.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <1461879932-9020-9-git-send-email-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Commit e8316d7 mistakenly passed consume=true within
qmp_input_optional() when checking if an optional member was
present, but the mistake was silently ignored since the code
happily let us extract a member more than once. Fix
qmp_input_optional() to not consume anything, then tighten up
the input visitor to ensure that a member is consumed exactly
once (all generated code follows this pattern; and the new
assert will catch any hand-written code that tries to visit
the same key more than once).
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <1461879932-9020-8-git-send-email-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
The following uses of a QMP input visitor should be strict
(that is, excess keys in QDict input should be flagged if not
converted to QAPI):
- Testsuite code unrelated to explicitly testing non-strict
mode (test-qmp-commands, test-visitor-serialization); since
we want more code to be strict by default, having more tests
of strict mode doesn't hurt
- Code used for cloning QAPI objects (replay-input.c,
qemu-sockets.c); we are reparsing a QObject just barely
produced by the qmp output visitor and which therefore should
not have any garbage, so while it is extra work to be strict,
it validates that our clone is correct [note that a later patch
series will simplify these two uses by creating an actual
clone visitor that is much more efficient than a
generate/reparse cycle]
- qmp_object_add(), which calls into user_creatable_add_type().
Since command line parsing for '-object' uses the same
user_creatable_add_type() through the OptsVisitor, and that is
always strict, we want to ensure that any nested dictionaries
would be treated the same in QMP and from the command line (I
don't actually know if such nested dictionaries exist). Note
that on this code change, strictness only matters for nested
dictionaries (if even possible), since we already flag excess
input at the top level during an earlier object_property_set()
on an unknown key, whether from QemuOpts:
$ ./x86_64-softmmu/qemu-system-x86_64 -nographic -nodefaults -qmp stdio -object secret,id=sec0,data=letmein,format=raw,foo=bar
qemu-system-x86_64: -object secret,id=sec0,data=letmein,format=raw,foo=bar: Property '.foo' not found
or from QMP:
$ ./x86_64-softmmu/qemu-system-x86_64 -nographic -nodefaults -qmp stdio
{"QMP": {"version": {"qemu": {"micro": 93, "minor": 5, "major": 2}, "package": ""}, "capabilities": []}}
{"execute":"qmp_capabilities"}
{"return": {}}
{"execute":"object-add","arguments":{"qom-type":"secret","id":"sec0","props":{"format":"raw","data":"letmein","foo":"bar"}}}
{"error": {"class": "GenericError", "desc": "Property '.foo' not found"}}
The only remaining uses of non-strict input visits are:
- QMP 'qom-set' (which eventually executes
object_property_set_qobject()) - mark it as something to revisit
in the future (I didn't want to spend any more time on this patch
auditing if we have any QOM dictionary properties that might be
impacted, and couldn't easily prove whether this code path is
shared with anything else).
- test-qmp-input-visitor: explicit tests of non-strict mode. If
we later get rid of users that don't need strictness, then this
test should be merged with test-qmp-input-strict
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <1461879932-9020-7-git-send-email-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Rather than having two separate ways to create a QMP input
visitor, where the safer approach has the more verbose name,
it is better to consolidate things into a single function
where the caller must explicitly choose whether to be strict
or to ignore excess input. This patch is the strictly
mechanical conversion; the next patch will then audit which
uses can be made stricter.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <1461879932-9020-6-git-send-email-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Management of the top of stack was a bit verbose; creating a
temporary variable and adding some comments makes the existing
code more legible before the next few patches improve things.
No semantic changes other than asserting that we are always
visiting a QObject, and not a NULL value. In particular, the
check for 'name && qobject_type(qobj) == QTYPE_QDICT)' is a
bit overkill (a dict visit should always have a name); a later
patch revisits that, while this patch is only changing one
layer of indentation due to dropping 'if (qobj)'.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <1461879932-9020-5-git-send-email-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Ever since QMP was first added back in commit 43c20a43, we have
never had any QmpCommandType other than QCT_NORMAL. It's
pointless to carry around the cruft.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <1461879932-9020-4-git-send-email-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Our existing input visitors were not very consistent on errors in a
function taking 'TYPE **obj'. These are start_struct(),
start_alternate(), type_str(), and type_any(). next_list() is
similar, but can't fail (see commit 08f9541). While all of them set
'*obj' to allocated storage on success, it was not obvious whether
'*obj' was guaranteed safe on failure, or whether it was left
uninitialized. But a future patch wants to guarantee that
visit_type_FOO() does not leak a partially-constructed obj back to
the caller; it is easier to implement this if we can reliably state
that input visitors assign '*obj' regardless of success or failure,
and that on failure *obj is NULL. Add assertions to enforce
consistency in the final setting of err vs. *obj.
The opts-visitor start_struct() doesn't set an error, but it
also was doing a weird check for 0 size; all callers pass in
non-zero size if obj is non-NULL.
The testsuite has at least one spot where we no longer need
to pre-initialize a variable prior to a visit; valgrind confirms
that the test is still fine with the cleanup.
A later patch will document the design constraint implemented
here.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <1461879932-9020-3-git-send-email-eblake@redhat.com>
[visit_start_alternate()'s assertion tightened, commit message tweaked]
Signed-off-by: Markus Armbruster <armbru@redhat.com>
We have three classes of QAPI visitors: input, output, and dealloc.
Currently, all implementations of these visitors have one thing in
common based on their visitor type: the implementation used for the
visit_type_enum() callback. But since we plan to add more such
common behavior, in relation to documenting and further refining
the semantics, it makes more sense to have the visitor
implementations advertise which class they belong to, so the common
qapi-visit-core code can use that information in multiple places.
A later patch will better document the types of visitors directly
in visitor.h.
For this patch, knowing the class of a visitor implementation lets
us make input_type_enum() and output_type_enum() become static
functions, by replacing the callback function Visitor.type_enum()
with the simpler enum member Visitor.type. Share a common
assertion in qapi-visit-core as part of the refactoring.
Move comments in opts-visitor.c to match the refactored layout.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <1461879932-9020-2-git-send-email-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Call vbe_update_vgaregs() when the guest touches GFX, SEQ or CRT
registers, to make sure the vga registers will always have the
values needed by vbe mode. This makes sure the sanity checks
applied by vbe_fixup_regs() are effective.
Without this guests can muck with shift_control, can turn on planar
vga modes or text mode emulation while VBE is active, making qemu
take code paths meant for CGA compatibility, but with the very
large display widths and heigts settable using VBE registers.
Which is good for one or another buffer overflow. Not that
critical as they typically read overflows happening somewhere
in the display code. So guests can DoS by crashing qemu with a
segfault, but it is probably not possible to break out of the VM.
Fixes: CVE-2016-3712
Reported-by: Zuozhi Fzz <zuozhi.fzz@alibaba-inc.com>
Reported-by: P J P <ppandit@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Call the new vbe_update_vgaregs() function on vbe configuration
changes, to make sure vga registers are up-to-date.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
When enabling vbe mode qemu will setup a bunch of vga registers to make
sure the vga emulation operates in correct mode for a linear
framebuffer. Move that code to a separate function so we can call it
from other places too.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
vga allows banked access to video memory using the window at 0xa00000
and it supports a different access modes with different address
calculations.
The VBE bochs extentions support banked access too, using the
VBE_DISPI_INDEX_BANK register. The code tries to take the different
address calculations into account and applies different limits to
VBE_DISPI_INDEX_BANK depending on the current access mode.
Which is probably effective in stopping misprogramming by accident.
But from a security point of view completely useless as an attacker
can easily change access modes after setting the bank register.
Drop the bogus check, add range checks to vga_mem_{readb,writeb}
instead.
Fixes: CVE-2016-3710
Reported-by: Qinghao Tang <luodalongde@gmail.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Fixes build failure with --enable-xfsctl and
new linux headers (>=4.5) and older xfsprogs(<4.5):
In file included from /usr/include/xfs/xfs.h:38:0,
from /var/tmp/portage/app-emulation/qemu-2.5.0-r1/work/qemu-2.5.0/block/raw-posix.c:97:
/usr/include/xfs/xfs_fs.h:42:8: error: redefinition of ‘struct fsxattr’
struct fsxattr {
^
In file included from /var/tmp/portage/app-emulation/qemu-2.5.0-r1/work/qemu-2.5.0/block/raw-posix.c:60:0:
/usr/include/linux/fs.h:155:8: note: originally defined here
struct fsxattr {
This is really a bug in the system headers, but we can work around it
by defining HAVE_FSXATTR in the QEMU headers if linux/fs.h provides
the struct, so that xfs_fs.h doesn't try to define it as well.
CC: qemu-trivial@nongnu.org
CC: Markus Armbruster <armbru@redhat.com>
CC: Peter Maydell <peter.maydell@linaro.org>
CC: Stefan Weil <sw@weilnetz.de>
Tested-by: Stefan Weil <sw@weilnetz.de>
Signed-off-by: Jan Vesely <jano.vesely@gmail.com>
[PMM: adjusted commit message, comments]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Minor, obvious fix only affecting BE hosts.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEcBAABAgAGBQJXJfnwAAoJECgfDbjSjVRpUqgH/jzhrMSe4r6JizzltOPoQ1ET
5kLv8tsnmJS6VbtEboz2e/4UJZgOKAAkqZ1LV6C9mzxiSmffJWe5gXwCmMrrxdcD
PFyAJwvnEY6chLF7B4/GxzdP8w6qxBhdSh6QGbGrSPodl5VkuceC4ks3E7Te2p+r
6SBSBFeVPusaGz/n0F7AktrPgjS9Yyr2fFVpydC52mamwQXIYrSVHmPR6Ch6GTNt
Z+NbaLq1cId+F8w3CDcDRox8we6UcTMnngg49UZ8XJbv9T19ea3+L1V3fyb7tJ4T
15M4lOxwNsgKvoa66VUNLAitbiar1bioKNP48i/kdG2GP8P30KKXyMazWAJYJ1o=
=Jsvf
-----END PGP SIGNATURE-----
Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging
acpi: last minute fix for 2.6
Minor, obvious fix only affecting BE hosts.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
# gpg: Signature made Sun 01 May 2016 13:43:28 BST using RSA key ID D28D5469
# gpg: Good signature from "Michael S. Tsirkin <mst@kernel.org>"
# gpg: aka "Michael S. Tsirkin <mst@redhat.com>"
* remotes/mst/tags/for_upstream:
acpi: fix bios linker loadder COMMAND_ALLOCATE on bigendian host
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
'make check' fails with:
ERROR:tests/bios-tables-test.c:493:load_expected_aml:
assertion failed: (g_file_test(aml_file, G_FILE_TEST_EXISTS))
since commit:
caf50c7166
tests: pc: acpi: drop not needed 'expected SSDT' blobs
Assert happens because qemu-system-x86_64 generates
SSDT table and test looks for a corresponding expected
table to compare with.
However there is no expected SSDT blob anymore, since
QEMU souldn't generate one. As it happens BIOS is not
able to read ACPI tables from QEMU and fallbacks to
embeded legacy ACPI codepath, which generates SSDT.
That happens due to wrongly sized endiannes conversion
which makes
uint8_t BiosLinkerLoaderEntry.alloc.zone
end up with 0 due to truncation of 32 bit integer
which on host is 1 or 2.
Fix it by dropping invalid cpu_to_le32() as uint8_t
doesn't require any conversion.
RHBZ: https://bugzilla.redhat.com/show_bug.cgi?id=1330174
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Tested-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
Commit d5941dd documented that it leaves the default volume name as it
was ("QEMU VVFAT"), but it doesn't actually implement this. You get an
empty name (eleven space characters) instead.
This fixes the implementation to apply the advertised default.
Cc: qemu-stable@nongnu.org
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Commit d5941dd made the volume name configurable, but it didn't consider
that the rw code compares the volume name string to assert that the
first directory entry is the volume name. This made vvfat crash in rw
mode.
This fixes the assertion to compare with the configured volume name
instead of a literal string.
Cc: qemu-stable@nongnu.org
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Make sure the error message for visit_type_uint64() gracefully
handles a NULL 'name' when called from the top level or a list
context, as not all the world behaves like glibc in allowing
NULL through a printf-family %s.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <1461879932-9020-21-git-send-email-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
On Darwin, connect, sendto and friends want the exact size of the sockaddr,
not more (and in particular, not sizeof(struct sockaddr_storaget))
This commit adds the sockaddr_size helper to be used when passing a sockaddr
size to such function, and makes use of it int sendto and connect calls.
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Reviewed-by: John Arbuckle <programmingkidx@gmail.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
This just has one, last-minute, fix for a serious regression of memory
hotplug.
Patch author's comment:
Really sorry for the way last-minute fix, but without this memory
hotplug is totally broken :( Hoping to get this in for Wednesday's
RC4, which I think will be the final before release.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQIcBAABAgAGBQJXHtfkAAoJEGw4ysog2bOSt/AP/jfZJvt4zjGAZ4BqVpyISLhM
n/hPZNglmcvg+YmjOrH/5G7PeX+r6Bf2gyrBjL06sWDSff1FdiBySiJQec6RlEwo
hq+EBaUWMeWS7Fg/C3evNvg2WGvSDs8KjRKWA76pm8pu+110cxnrRUybIRbbHqaI
3vxR5Gdewmbvp5VYQgr5e/BSMac7wTT+Us6/Cjr2n5q56e2jjLryM8zMiyqMtP+S
7XdCqcowXv4dQsaqwLxS0Ata8f2ykrjtp9qADErbULZtEf8cAf8zvTR3ElXtnywq
pEgxbKJeYQwGScudWb9lzwGRRGTMpM2zLbIuBKjQE2E77yQK0+hsq3FkH6OdipgT
T8K5gU3p3AFuksZ5/+COVACC364Hry4OrYjTvGDXvsXXfWablrlETqoNtaj5A7Uz
cvjcNEpja88B+dDxT/bt8Ab8p5pTuo58uUUgT4gLIduzSJlgaJ2qzMvwF4cfIGcB
5wsKr36GY4jyw3UD0MbXAwDhwvKP9t1FUmhST4qLF0WvNjfM62jGgh1E2DaXKuwF
+ndVkoaus9yxYgMKyuuI/hXH/M4qk6yuUeNj+E1qM1823XpXQgFW1hnvnPZkbf0f
nqT1as59kltGhFrv7DVLd3ovYLSjK5XtLAq8NfhTjjEe8nQlzHhql9RRkXedPpLO
GVwp6DA608M2krlS5TxK
=zSL+
-----END PGP SIGNATURE-----
Merge remote-tracking branch 'remotes/dgibson/tags/ppc-for-2.6-20160426' into staging
ppc patch queue for 2016-04-26 (last minute qemu-2.6 fix)
This just has one, last-minute, fix for a serious regression of memory
hotplug.
Patch author's comment:
Really sorry for the way last-minute fix, but without this memory
hotplug is totally broken :( Hoping to get this in for Wednesday's
RC4, which I think will be the final before release.
# gpg: Signature made Tue 26 Apr 2016 03:52:20 BST using RSA key ID 20D9B392
# gpg: Good signature from "David Gibson <david@gibson.dropbear.id.au>"
# gpg: aka "David Gibson (Red Hat) <dgibson@redhat.com>"
# gpg: aka "David Gibson (ozlabs.org) <dgibson@ozlabs.org>"
# 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: 75F4 6586 AE61 A66C C44E 87DC 6C38 CACA 20D9 B392
* remotes/dgibson/tags/ppc-for-2.6-20160426:
spapr_drc: fix aborts during DRC-count based hotplug
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Commit b00c72180c ("target-mips: add PC, XNP reg numbers to RDHWR")
changed the rdhwr helpers to use check_hwrena() to check the register
being accessed is enabled in CP0_HWREna when used from user mode. If
that check fails an EXCP_RI exception is raised at the host PC
calculated with GETPC().
However check_hwrena() may not be fully inlined as the
do_raise_exception() part of it is common regardless of the arguments.
This causes GETPC() to calculate the address in the call in the helper
instead of the generated code calling the helper. No TB will be found
and the EPC reported with the resulting guest RI exception points to the
beginning of the TB instead of the RDHWR instruction.
We can't reliably force check_hwrena() to be inlined, and converting it
to a macro would be ugly, so instead pass the host PC in as an argument,
with each rdhwr helper passing GETPC(). This should avoid any dependence
on compiler behaviour, and in practice seems to ensure the full inlining
of check_hwrena() on x86_64.
This issue causes failures when running a MIPS KVM (trap & emulate)
guest in a MIPS QEMU TCG guest, as the inner guest kernel will do a
RDHWR of counter, which is disabled in the outer guest's CP0_HWREna by
KVM so it can emulate the inner guest's counter. The emulation fails and
the RI exception is passed to the inner guest.
Fixes: b00c72180c ("target-mips: add PC, XNP reg numbers to RDHWR")
Signed-off-by: James Hogan <james.hogan@imgtec.com>
Cc: Leon Alrae <leon.alrae@imgtec.com>
Cc: Yongbok Kim <yongbok.kim@imgtec.com>
Cc: Aurelien Jarno <aurelien@aurel32.net>
Reviewed-by: Aurelien Jarno <aurelien@aurel32.net>
Reviewed-by: Leon Alrae <leon.alrae@imgtec.com>
Signed-off-by: Leon Alrae <leon.alrae@imgtec.com>
qemu_opts_foreach() runs its callback with the error location set to
the option's location. Any errors the callback reports use the
option's location automatically.
Commit 90998d5 moved the actual error reporting from "inside"
qemu_opts_foreach() to after it. Here's a typical hunk:
if (qemu_opts_foreach(qemu_find_opts("object"),
- object_create,
- object_create_initial, NULL)) {
+ user_creatable_add_opts_foreach,
+ object_create_initial, &err)) {
+ error_report_err(err);
exit(1);
}
Before, object_create() reports from within qemu_opts_foreach(), using
the option's location. Afterwards, we do it after
qemu_opts_foreach(), using whatever location happens to be current
there. Commonly a "none" location.
This is because Error objects don't have location information.
Problematic.
Reproducer:
$ qemu-system-x86_64 -nodefaults -display none -object secret,id=foo,foo=bar
qemu-system-x86_64: Property '.foo' not found
Note no location. This commit restores it:
qemu-system-x86_64: -object secret,id=foo,foo=bar: Property '.foo' not found
Note that the qemu_opts_foreach() bug just fixed could mask the bug
here: if the location it leaves dangling hasn't been clobbered, yet,
it's the correct one.
Reported-by: Eric Blake <eblake@redhat.com>
Cc: Daniel P. Berrange <berrange@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1461767349-15329-4-git-send-email-armbru@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
[Paragraph on Error added to commit message]
replay_configure() pushes and pops a Location with automatic storage
duration. Except it fails to pop when -icount parameter "rr" isn't
given. cur_loc then points to unused stack space, and will most
likely get clobbered in short order.
Clobbered cur_loc can make loc_pop() and error_print_loc() crash or
report bogus locations.
Broken in commit 890ad55.
I didn't take the time to find a reproducer.
Cc: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1461767349-15329-3-git-send-email-armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
qemu_opts_foreach() pushes and pops a Location with automatic storage
duration. Except it fails to pop when @func() returns non-zero.
cur_loc then points to unused stack space, and will most likely get
clobbered in short order.
Clobbered cur_loc can make loc_pop() and error_print_loc() crash or
report bogus locations.
Affects several qemu command line options as well as qemu-img,
qemu-io, qemu-nbd -object, and blkdebug's configuration file.
Broken in commit a4c7367, v2.4.0.
Reproducer:
$ qemu-system-x86_64 -nodefaults -display none -object secret,id=foo,foo=bar
main() reports "Property '.foo' not found" like this:
if (qemu_opts_foreach(qemu_find_opts("object"),
user_creatable_add_opts_foreach,
object_create_delayed, &err)) {
error_report_err(err);
exit(1);
}
cur_loc then points to where qemu_opts_foreach()'s Location used to
be, i.e. unused stack space. With optimization, this Location doesn't
get clobbered for me, and also happens to be the correct location.
Without optimization, it does get clobbered in a way that makes
error_report_err() report no location.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1461767349-15329-2-git-send-email-armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
CPU/memory resources can be signalled en-masse via
spapr_hotplug_req_add_by_count(), and when doing so, actually change
the meaning of the 'drc' parameter passed to
spapr_hotplug_req_event() to be a count rather than an index.
f40eb92 added a hook in spapr_hotplug_req_event() to record when a
device had been 'signalled' to the guest, but that code assumes that
drc is always an index. In cases where it's a count, such as memory
hotplug, the DRC lookup will fail, leading to an assert.
Fix this by only explicitly setting the signalled state for cases where
we are doing PCI hotplug.
For other resources types, since we cannot selectively track whether a
resource has been signalled in cases where we signal attach as a count,
set the 'signalled' state to true immediately upon making the
resource available via drck->attach().
Reported-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
Cc: david@gibson.dropbear.id.au
Cc: qemu-ppc@nongnu.org
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
commit "5f77e06 usb: add pid check at the first of uhci_handle_td()"
moved the pid verification to the start of the uhci_handle_td function,
to simplify the error handling (we don't have to free stuff which we
didn't allocate in the first place ...).
Problem is now the check fires too often, it raises error IRQs even for
TDs which we are not going to process because they are not set active.
So, lets move down the check a bit, so it is done only for active TDs,
but still before we are going to allocate stuff to process the requested
transfer.
Reported-by: Joe Clifford <joe@thunderbug.co.uk>
Tested-by: Joe Clifford <joe@thunderbug.co.uk>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 1461321893-15811-1-git-send-email-kraxel@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
A single fix for a bug in parameter handling for the spapr PCI host
bridge.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=MxYq
-----END PGP SIGNATURE-----
Merge remote-tracking branch 'remotes/dgibson/tags/ppc-for-2.6-20160423' into staging
ppc patch queue for 2016-03-23
A single fix for a bug in parameter handling for the spapr PCI host
bridge.
# gpg: Signature made Sat 23 Apr 2016 07:55:29 BST using RSA key ID 20D9B392
# gpg: Good signature from "David Gibson <david@gibson.dropbear.id.au>"
# gpg: aka "David Gibson (Red Hat) <dgibson@redhat.com>"
# gpg: aka "David Gibson (ozlabs.org) <dgibson@ozlabs.org>"
# 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: 75F4 6586 AE61 A66C C44E 87DC 6C38 CACA 20D9 B392
* remotes/dgibson/tags/ppc-for-2.6-20160423:
hw/ppc/spapr: Fix crash when specifying bad parameters to spapr-pci-host-bridge
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
QEMU currently crashes when using bad parameters for the
spapr-pci-host-bridge device:
$ qemu-system-ppc64 -device spapr-pci-host-bridge,buid=0x123,liobn=0x321,mem_win_addr=0x1,io_win_addr=0x10
Segmentation fault
The problem is that spapr_tce_find_by_liobn() might return NULL, but
the code in spapr_populate_pci_dt() does not check for this condition
and then tries to dereference this NULL pointer.
Apart from that, the return value of spapr_populate_pci_dt() also
has to be checked for all PCI buses, not only for the last one, to
make sure we catch all errors.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Commit 5a7e7a0ba moved mirror_exit to a BH handler but didn't add any
protection against new requests that could sneak in just before the
BH is dispatched. For example (assuming a code base at that commit):
main_loop_wait # 1
os_host_main_loop_wait
g_main_context_dispatch
aio_ctx_dispatch
aio_dispatch
...
mirror_run
bdrv_drain
(a) block_job_defer_to_main_loop
qemu_iohandler_poll
virtio_queue_host_notifier_read
...
virtio_submit_multiwrite
(b) blk_aio_multiwrite
main_loop_wait # 2
<snip>
aio_dispatch
aio_bh_poll
(c) mirror_exit
At (a) we know the BDS has no pending request. However, the same
main_loop_wait call is going to dispatch iohandlers (EventNotifier
events), which may lead to a new I/O from guest. So the invariant is
already broken at (c). Data loss.
Commit f3926945c8 made iohandler to use aio API. The order of
virtio_queue_host_notifier_read and block_job_defer_to_main_loop within
a main_loop_wait becomes unpredictable, and even worse, if the host
notifier event arrives at the next main_loop_wait call, the
unpredictable order between mirror_exit and
virtio_queue_host_notifier_read is also a trouble. As shown below, this
commit made the bug easier to trigger:
- Bug case 1:
main_loop_wait # 1
os_host_main_loop_wait
g_main_context_dispatch
aio_ctx_dispatch (qemu_aio_context)
...
mirror_run
bdrv_drain
(a) block_job_defer_to_main_loop
aio_ctx_dispatch (iohandler_ctx)
virtio_queue_host_notifier_read
...
virtio_submit_multiwrite
(b) blk_aio_multiwrite
main_loop_wait # 2
...
aio_dispatch
aio_bh_poll
(c) mirror_exit
- Bug case 2:
main_loop_wait # 1
os_host_main_loop_wait
g_main_context_dispatch
aio_ctx_dispatch (qemu_aio_context)
...
mirror_run
bdrv_drain
(a) block_job_defer_to_main_loop
main_loop_wait # 2
...
aio_ctx_dispatch (iohandler_ctx)
virtio_queue_host_notifier_read
...
virtio_submit_multiwrite
(b) blk_aio_multiwrite
aio_dispatch
aio_bh_poll
(c) mirror_exit
In both cases, (b) breaks the invariant wanted by (a) and (c).
Until then, the request loss has been silent. Later, 3f09bfbc7b added
asserts at (c) to check the invariant (in
bdrv_replace_in_backing_chain), and Max reported an assertion failure
first visible there, by doing active committing while the guest is
running bonnie++.
2.5 added bdrv_drained_begin at (a) to protect the dataplane case from
similar problems, but we never realize the main loop bug until now.
As a bandage, this patch disables iohandler's external events
temporarily together with bs->ctx.
Launchpad Bug: 1570134
Cc: qemu-stable@nongnu.org
Signed-off-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Jeff Cody <jcody@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
aio_poll doesn't poll the external nodes so this should never be true,
but aio_ctx_dispatch may get notified by the events from GSource. To
make bdrv_drained_begin effective in main loop, we should check the
is_external flag here too.
Also do the check in aio_pending so aio_dispatch is not called
superfluously, when there is no events other than external ones.
Signed-off-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Jeff Cody <jcody@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
The effect of this change is the block layer drained section can work,
for example when mirror job is being completed.
Signed-off-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
All callers pass "false" keeping the old semantics. The windows
implementation doesn't distinguish the flag yet. On posix, it is passed
down to the underlying aio context.
Signed-off-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
For KVM to use Transparent Huge Pages (THP) we have to ensure that the
alignment of the userspace address of the KVM memory slot and the IPA
that the guest sees for a memory region have the same offset from the 2M
huge page size boundary.
One way to achieve this is to always align the IPA region at a 2M
boundary and ensure that the mmap alignment is also at 2M.
Unfortunately, we were only doing this for __arm__, not for __aarch64__,
so add this simple condition.
This fixes a performance regression using KVM/ARM on AArch64 platforms
that showed a performance penalty of more than 50%, introduced by the
following commit:
9fac18f (oslib: allocate PROT_NONE pages on top of RAM, 2015-09-10)
We were only lucky before the above commit, because we were allocating
large regions and naturally getting a 2M alignment on those allocations
then.
Cc: qemu-stable@nongnu.org
Reported-by: Shih-Wei Li <shihwei@cs.columbia.edu>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
[PMM: wrapped long line]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
The NBD protocol does not (yet) force any alignment constraints
on clients. Even though qemu NBD clients always send requests
that are aligned to 512 bytes, we must be prepared for non-qemu
clients that don't care about alignment (even if it means they
are less efficient). Our use of blk_read() and blk_write() was
silently operating on the wrong file offsets when the client
made an unaligned request, corrupting the client's data (but
as the client already has control over the file we are serving,
I don't think it is a security hole, per se, just a data
corruption bug).
Note that in the case of NBD_CMD_READ, an unaligned length could
cause us to return up to 511 bytes of uninitialized trailing
garbage from blk_try_blockalign() - hopefully nothing sensitive
from the heap's prior usage is ever leaked in that manner.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Tested-by: Kevin Wolf <kwolf@redhat.com>
Message-id: 1461249750-31928-1-git-send-email-eblake@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
