qemu.git
2 years agovga: fix display update region calculation
Gerd Hoffmann [Tue, 9 May 2017 10:48:39 +0000 (12:48 +0200)] 
vga: fix display update region calculation

vga display update mis-calculated the region for the dirty bitmap
snapshot in case the scanlines are padded.  This can triggere an
assert in cpu_physical_memory_snapshot_get_dirty().

Fixes: fec5e8c92becad223df9d972770522f64aafdb72
Reported-by: Kevin Wolf <kwolf@redhat.com>
Reported-by: 李强 <liqiang6-s@360.cn>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 20170509104839.19415-1-kraxel@redhat.com

2 years agosm501: make display updates thread safe
Gerd Hoffmann [Tue, 9 May 2017 11:19:28 +0000 (13:19 +0200)] 
sm501: make display updates thread safe

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 20170509111928.30935-1-kraxel@redhat.com

2 years agotcx: make display updates thread safe
Mark Cave-Ayland [Wed, 10 May 2017 20:52:31 +0000 (21:52 +0100)] 
tcx: make display updates thread safe

Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Message-id: 1494449551-20227-3-git-send-email-mark.cave-ayland@ilande.co.uk
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2 years agocg3: make display updates thread safe
Mark Cave-Ayland [Wed, 10 May 2017 20:52:30 +0000 (21:52 +0100)] 
cg3: make display updates thread safe

Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Message-id: 1494449551-20227-2-git-send-email-mark.cave-ayland@ilande.co.uk
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2 years agoMerge remote-tracking branch 'armbru/tags/pull-qapi-2017-05-04-v3' into staging
Stefan Hajnoczi [Tue, 9 May 2017 19:49:08 +0000 (15:49 -0400)] 
Merge remote-tracking branch 'armbru/tags/pull-qapi-2017-05-04-v3' into staging

QAPI patches for 2017-05-04

# gpg: Signature made Tue 09 May 2017 03:16:12 AM EDT
# 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

* armbru/tags/pull-qapi-2017-05-04-v3: (28 commits)
  qmp-shell: improve help
  qmp-shell: don't show version greeting if unavailable
  qmp-shell: Cope with query-commands error
  qmp-shell: add -N option to skip negotiate
  qmp-shell: add persistent command history
  qobject-input-visitor: Catch misuse of end_struct vs. end_list
  qapi: Document intended use of @name within alternate visits
  qobject-input-visitor: Document full_name_nth()
  qmp: Improve QMP dispatch error messages
  sockets: Delete unused helper socket_address_crumple()
  sockets: Limit SocketAddressLegacy to external interfaces
  sockets: Rename SocketAddressFlat to SocketAddress
  sockets: Rename SocketAddress to SocketAddressLegacy
  qapi: New QAPI_CLONE_MEMBERS()
  sockets: Prepare inet_parse() for flattened SocketAddress
  sockets: Prepare vsock_parse() for flattened SocketAddress
  test-qga: Actually test 0xff sync bytes
  fdc-test: Avoid deprecated 'change' command
  QemuOpts: Simplify qemu_opts_to_qdict()
  block: Simplify bdrv_append_temp_snapshot() logic
  ...

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoqmp-shell: improve help
Marc-André Lureau [Thu, 4 May 2017 12:54:32 +0000 (16:54 +0400)] 
qmp-shell: improve help

Describe the arguments & fix the tool name.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20170504125432.21653-5-marcandre.lureau@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agoqmp-shell: don't show version greeting if unavailable
Marc-André Lureau [Thu, 4 May 2017 12:54:31 +0000 (16:54 +0400)] 
qmp-shell: don't show version greeting if unavailable

qemu-ga doesn't have greeting.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20170504125432.21653-4-marcandre.lureau@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agoqmp-shell: Cope with query-commands error
Marc-André Lureau [Thu, 4 May 2017 12:54:30 +0000 (16:54 +0400)] 
qmp-shell: Cope with query-commands error

qemu-ga doesn't implement it.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20170504125432.21653-3-marcandre.lureau@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
[Commit message tweaked]
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agoqmp-shell: add -N option to skip negotiate
Marc-André Lureau [Thu, 4 May 2017 12:54:29 +0000 (16:54 +0400)] 
qmp-shell: add -N option to skip negotiate

qemu-ga doesn't have negotiate phase.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20170504125432.21653-2-marcandre.lureau@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agoqmp-shell: add persistent command history
John Snow [Thu, 27 Apr 2017 22:36:28 +0000 (18:36 -0400)] 
qmp-shell: add persistent command history

Use the existing readline history function we are utilizing
to provide persistent command history across instances of qmp-shell.

This assists entering debug commands across sessions that may be
interrupted by QEMU sessions terminating, where the qmp-shell has
to be relaunched.

Signed-off-by: John Snow <jsnow@redhat.com>
Message-Id: <20170427223628.20893-1-jsnow@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Kashyap Chamarthy <kchamart@redhat.com>
Tested-by: Kashyap Chamarthy <kchamart@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agoqobject-input-visitor: Catch misuse of end_struct vs. end_list
Markus Armbruster [Thu, 27 Apr 2017 08:41:26 +0000 (10:41 +0200)] 
qobject-input-visitor: Catch misuse of end_struct vs. end_list

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1493282486-28338-5-git-send-email-armbru@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
[More elaborate assertions for clarity]

2 years agoqapi: Document intended use of @name within alternate visits
Markus Armbruster [Thu, 27 Apr 2017 08:41:25 +0000 (10:41 +0200)] 
qapi: Document intended use of @name within alternate visits

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1493282486-28338-4-git-send-email-armbru@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
2 years agoqobject-input-visitor: Document full_name_nth()
Markus Armbruster [Thu, 27 Apr 2017 08:41:24 +0000 (10:41 +0200)] 
qobject-input-visitor: Document full_name_nth()

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1493282486-28338-3-git-send-email-armbru@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
2 years agoqmp: Improve QMP dispatch error messages
Markus Armbruster [Thu, 27 Apr 2017 08:41:23 +0000 (10:41 +0200)] 
qmp: Improve QMP dispatch error messages

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <1493282486-28338-2-git-send-email-armbru@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
2 years agosockets: Delete unused helper socket_address_crumple()
Markus Armbruster [Wed, 26 Apr 2017 07:36:42 +0000 (09:36 +0200)] 
sockets: Delete unused helper socket_address_crumple()

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1493192202-3184-8-git-send-email-armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
[Commit message typo fixed]

2 years agosockets: Limit SocketAddressLegacy to external interfaces
Markus Armbruster [Wed, 26 Apr 2017 07:36:41 +0000 (09:36 +0200)] 
sockets: Limit SocketAddressLegacy to external interfaces

SocketAddressLegacy is a simple union, and simple unions are awkward:
they have their variant members wrapped in a "data" object on the
wire, and require additional indirections in C.  SocketAddress is the
equivalent flat union.  Convert all users of SocketAddressLegacy to
SocketAddress, except for existing external interfaces.

See also commit fce5d53..9445673 and 85a82e8..c5f1ae3.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1493192202-3184-7-git-send-email-armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
[Minor editing accident fixed, commit message and a comment tweaked]

Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agosockets: Rename SocketAddressFlat to SocketAddress
Markus Armbruster [Wed, 26 Apr 2017 07:36:40 +0000 (09:36 +0200)] 
sockets: Rename SocketAddressFlat to SocketAddress

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1493192202-3184-6-git-send-email-armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Prasanna Kumar Kalever <prasanna.kalever@redhat.com>
2 years agosockets: Rename SocketAddress to SocketAddressLegacy
Markus Armbruster [Wed, 26 Apr 2017 07:36:39 +0000 (09:36 +0200)] 
sockets: Rename SocketAddress to SocketAddressLegacy

The next commit will rename SocketAddressFlat to SocketAddress, and
the commit after that will replace most uses of SocketAddressLegacy by
SocketAddress, replacing most of this commit's renames right back.

Note that checkpatch emits a few "line over 80 characters" warnings.
The long lines are all temporary; the SocketAddressLegacy replacement
will shorten them again.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1493192202-3184-5-git-send-email-armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agoqapi: New QAPI_CLONE_MEMBERS()
Markus Armbruster [Wed, 26 Apr 2017 07:36:38 +0000 (09:36 +0200)] 
qapi: New QAPI_CLONE_MEMBERS()

QAPI_CLONE() returns a newly allocated QAPI object.  Inconvenient when
we want to clone into an existing object.  QAPI_CLONE_MEMBERS() does
exactly that.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1493192202-3184-4-git-send-email-armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
2 years agosockets: Prepare inet_parse() for flattened SocketAddress
Markus Armbruster [Wed, 26 Apr 2017 07:36:37 +0000 (09:36 +0200)] 
sockets: Prepare inet_parse() for flattened SocketAddress

I'm going to flatten SocketAddress: rename SocketAddress to
SocketAddressLegacy, SocketAddressFlat to SocketAddress, eliminate
SocketAddressLegacy except in external interfaces.

inet_parse() returns a newly allocated InetSocketAddress.  Lift the
allocation from inet_parse() into its caller socket_parse() to prepare
for flattening SocketAddress.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1493192202-3184-3-git-send-email-armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
[Straightforward rebase]

2 years agosockets: Prepare vsock_parse() for flattened SocketAddress
Markus Armbruster [Wed, 26 Apr 2017 07:36:36 +0000 (09:36 +0200)] 
sockets: Prepare vsock_parse() for flattened SocketAddress

I'm going to flatten SocketAddress: rename SocketAddress to
SocketAddressLegacy, SocketAddressFlat to SocketAddress, eliminate
SocketAddressLegacy except in external interfaces.

vsock_parse() returns a newly allocated VsockSocketAddress.  Lift the
allocation from vsock_parse() into its caller socket_parse() to
prepare for flattening SocketAddress.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1493192202-3184-2-git-send-email-armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
2 years agotest-qga: Actually test 0xff sync bytes
Eric Blake [Thu, 27 Apr 2017 21:58:21 +0000 (16:58 -0500)] 
test-qga: Actually test 0xff sync bytes

Commit 62c39b3 introduced test-qga, and at face value, appears
to be testing the 'guest-sync' behavior that is recommended for
guests in sending 0xff to QGA to force the parser to reset.  But
this aspect of the test has never actually done anything: the
qmp_fd() call chain converts its string argument into QObject,
then converts that QObject back to the actual string that is
sent over the wire - and the conversion process silently drops
the 0xff byte from the string sent to QGA, thus never resetting
the QGA parser.

An upcoming patch will get rid of the wasteful round trip
through QObject, at which point the string in test-qga will be
directly sent over the wire.

But fixing qmp_fd() to actually send 0xff over the wire is not
all we have to do - the actual QMP parser loudly complains that
0xff is not valid JSON, and sends an error message _prior_ to
actually parsing the 'guest-sync' or 'guest-sync-delimited'
command.  With 'guest-sync', we cannot easily tell if this error
message is a result of our command - which is WHY we invented
the 'guest-sync-delimited' command.  So for the testsuite, fix
things to only check 0xff behavior on 'guest-sync-delimited',
and to loop until we've consumed all garbage prior to the
requested delimiter, which is compatible with the documented actions
that a real QGA client is supposed to do.

Ideally, we'd fix the QGA JSON parser to silently ignore 0xff
rather than sending an error message back, at which point we
could enhance this test for 'guest-sync' as well as for
'guest-sync-delimited'.  But for the sake of this patch, our
testing of 'guest-sync' is no worse than it was pre-patch,
because we have never been sending 0xff over the wire in the
first place.

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20170427215821.19397-11-eblake@redhat.com>
Reviewed-by: Michael Roth <mdroth@linux.vnet.ibm.com>
[Additional comment squashed in, along with matching commit message
update]
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agofdc-test: Avoid deprecated 'change' command
Eric Blake [Thu, 27 Apr 2017 21:58:20 +0000 (16:58 -0500)] 
fdc-test: Avoid deprecated 'change' command

Use the preferred blockdev-change-medium command instead.

Also, use of 'device' is deprecated; adding an explicit id on
the command line lets us use 'id' for both blockdev-change-medium
and eject.

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20170427215821.19397-10-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agoQemuOpts: Simplify qemu_opts_to_qdict()
Eric Blake [Thu, 27 Apr 2017 21:58:19 +0000 (16:58 -0500)] 
QemuOpts: Simplify qemu_opts_to_qdict()

Noticed while investigating Coccinelle cleanups. There is no need
for a temporary variable when we can use the new macro to do the
same thing with less typing.

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20170427215821.19397-9-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agoblock: Simplify bdrv_append_temp_snapshot() logic
Eric Blake [Thu, 27 Apr 2017 21:58:18 +0000 (16:58 -0500)] 
block: Simplify bdrv_append_temp_snapshot() logic

Noticed while checking Coccinelle results. Naming a label 'out:'
when it is only used on error paths is weird.  Also, we had some
dead stores to 'ret'.  Meanwhile we know that snapshot_options
is NULL on success and that QDECREF(NULL) is safe.  So merge the
two exit paths into one by careful control over bs_snapshot.

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20170427215821.19397-8-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agoqobject: Use simpler QDict/QList scalar insertion macros
Eric Blake [Thu, 27 Apr 2017 21:58:17 +0000 (16:58 -0500)] 
qobject: Use simpler QDict/QList scalar insertion macros

We now have macros in place to make it less verbose to add a scalar
to QDict and QList, so use them.

Patch created mechanically via:
  spatch --sp-file scripts/coccinelle/qobject.cocci \
    --macro-file scripts/cocci-macro-file.h --dir . --in-place
then touched up manually to fix a couple of '?:' back to original
spacing, as well as avoiding a long line in monitor.c.

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20170427215821.19397-7-eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agoqobject: Add helper macros for common scalar insertions
Eric Blake [Thu, 27 Apr 2017 21:58:16 +0000 (16:58 -0500)] 
qobject: Add helper macros for common scalar insertions

Rather than making lots of callers wrap a scalar in a QInt, QString,
or QBool, provide helper macros that do the wrapping automatically.

Update the Coccinelle script to make mass conversions easy, although
the conversion itself will be done as a separate patches to ease
review and backport efforts.

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20170427215821.19397-6-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agoqobject: Drop useless QObject casts
Eric Blake [Thu, 27 Apr 2017 21:58:15 +0000 (16:58 -0500)] 
qobject: Drop useless QObject casts

We have macros in place to make it less verbose to add a subtype
of QObject to both QDict and QList. While we have made cleanups
like this in the past (see commit fcfcd8ffc, for example), having
it be automated by Coccinelle makes it easier to maintain.

Patch created mechanically via:
  spatch --sp-file scripts/coccinelle/qobject.cocci \
    --macro-file scripts/cocci-macro-file.h --dir . --in-place
then I verified that no manual touchups were required.

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20170427215821.19397-5-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agococcinelle: Add script to remove useless QObject casts
Eric Blake [Thu, 27 Apr 2017 21:58:14 +0000 (16:58 -0500)] 
coccinelle: Add script to remove useless QObject casts

We have macros in place to make it less verbose to add a subtype
of QObject to both QDict and QList. While we have made cleanups
like this in the past (see commit fcfcd8ffc, for example), having
it be automated by Coccinelle makes it easier to maintain.

The script is separate from the cleanups, for ease of review and
backporting.  A later patch will then add further possible cleanups.

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20170427215821.19397-4-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agopci: Reduce scope of error injection
Eric Blake [Thu, 27 Apr 2017 21:58:13 +0000 (16:58 -0500)] 
pci: Reduce scope of error injection

No one outside of pcie_aer.h was using error injection; mark them
static for internal use.

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20170427215821.19397-3-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agopci: Use struct instead of QDict to pass back parameters
Eric Blake [Thu, 27 Apr 2017 21:58:12 +0000 (16:58 -0500)] 
pci: Use struct instead of QDict to pass back parameters

It's simpler to just use a C struct than it is to bundle things
into a QDict in one function just to pull them back out in the
caller.  Plus, doing this gets rid of one more user of dynamic
JSON through qobject_from_jsonf(), as well as a memory leak of
the QDict.

While cleaning the code, fix things to report all errors (the
code was previously silently ignoring a failure of
pcie_aer_inject_error(), at a distance).

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20170427215821.19397-2-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agotest-keyval: fix leaks
Marc-André Lureau [Wed, 3 May 2017 22:38:42 +0000 (02:38 +0400)] 
test-keyval: fix leaks

Spotted by ASAN.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20170503223846.6559-2-marcandre.lureau@redhat.com>
Reviewed-by: Eric blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agotests/check-qdict: Fix missing brackets
Dr. David Alan Gilbert [Thu, 6 Apr 2017 15:41:07 +0000 (16:41 +0100)] 
tests/check-qdict: Fix missing brackets

Gcc 7 (on Fedora 26) spotted odd use of integers instead of a
boolean; it's got a point.

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20170406154107.9178-1-dgilbert@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2 years agoMerge remote-tracking branch 'aurel32/tags/pull-tcg-mips-20170506' into staging
Stefan Hajnoczi [Mon, 8 May 2017 17:29:50 +0000 (13:29 -0400)] 
Merge remote-tracking branch 'aurel32/tags/pull-tcg-mips-20170506' into staging

Fix MIPS R2 hosts support

# gpg: Signature made Sat 06 May 2017 06:56:28 AM EDT
# gpg:                using RSA key 0xBA9C78061DDD8C9B
# gpg: Good signature from "Aurelien Jarno <aurelien@aurel32.net>"
# gpg:                 aka "Aurelien Jarno <aurelien@jarno.fr>"
# gpg:                 aka "Aurelien Jarno <aurel32@debian.org>"
# Primary key fingerprint: 7746 2642 A9EF 94FD 0F77  196D BA9C 7806 1DDD 8C9B

* aurel32/tags/pull-tcg-mips-20170506:
  tcg/mips: fix field extraction opcode

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoMerge remote-tracking branch 'bonzini/tags/for-upstream' into staging
Stefan Hajnoczi [Mon, 8 May 2017 17:29:40 +0000 (13:29 -0400)] 
Merge remote-tracking branch 'bonzini/tags/for-upstream' into staging

A large set of small patches.  I have not included yet vhost-user-scsi,
but it'll come in the next pull request.

* use GDB XML register description for x86
* use _Static_assert in QEMU_BUILD_BUG_ON
* add "R:" to MAINTAINERS and get_maintainers
* checkpatch improvements
* dump threading fixes
* first part of vhost-user-scsi support
* QemuMutex tracing
* vmw_pvscsi and megasas fixes
* sgabios module update
* use Rev3 (ACPI 2.0) FADT
* deprecate -hdachs
* improve -accel documentation
* hax fix
* qemu-char GSource bugfix

# gpg: Signature made Fri 05 May 2017 06:10:40 AM EDT
# gpg:                using RSA key 0xBFFBD25F78C7AE83
# gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>"
# gpg:                 aka "Paolo Bonzini <pbonzini@redhat.com>"
# Primary key fingerprint: 46F5 9FBD 57D6 12E7 BFD4  E2F7 7E15 100C CD36 69B1
#      Subkey fingerprint: F133 3857 4B66 2389 866C  7682 BFFB D25F 78C7 AE83

* bonzini/tags/for-upstream: (21 commits)
  vhost-scsi: create a vhost-scsi-common abstraction
  libvhost-user: replace vasprintf() to fix build
  get_maintainer: add subsystem to reviewer output
  get_maintainer: --r (list reviewer) is on by default
  get_maintainer: it's '--pattern-depth', not '-pattern-depth'
  get_maintainer: Teach get_maintainer.pl about the new "R:" tag
  MAINTAINERS: Add "R:" tag for self-appointed reviewers
  Fix the -accel parameter and the documentation for 'hax'
  dump: Acquire BQL around vm_start() in dump thread
  hax: Fix memory mapping de-duplication logic
  checkpatch: Disallow glib asserts in main code
  trace: add qemu mutex lock and unlock trace events
  vmw_pvscsi: check message ring page count at initialisation
  sgabios: update for "fix wrong video attrs for int 10h,ah==13h"
  scsi: avoid an off-by-one error in megasas_mmio_write
  vl: deprecate the "-hdachs" option
  use _Static_assert in QEMU_BUILD_BUG_ON
  target/i386: Add GDB XML register description support
  char: Fix removing wrong GSource that be found by fd_in_tag
  hw/i386: Build-time assertion on pc/q35 reset register being identical.
  ...

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoMerge remote-tracking branch 'mcayland/tags/qemu-sparc-signed' into staging
Stefan Hajnoczi [Mon, 8 May 2017 16:43:15 +0000 (12:43 -0400)] 
Merge remote-tracking branch 'mcayland/tags/qemu-sparc-signed' into staging

qemu-sparc update

# gpg: Signature made Fri 05 May 2017 04:51:46 AM EDT
# gpg:                using RSA key 0x5BC2C56FAE0F321F
# gpg: Good signature from "Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>"
# Primary key fingerprint: CC62 1AB9 8E82 200D 915C  C9C4 5BC2 C56F AE0F 321F

* mcayland/tags/qemu-sparc-signed:
  cg3: add explicit ram_addr_t cast to scanline page variable
  tcx: fix cut/paste error in update_palette_entries()

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agomaintainers: add maintainer for replay* files
Pavel Dovgalyuk [Wed, 3 May 2017 11:33:04 +0000 (14:33 +0300)] 
maintainers: add maintainer for replay* files

Updating MAINTAINERS to set Pavel Dovgalyuk as record/replay maintainer
and Paolo Bonzini as a reviewer.

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
Message-id: 20170503113304.8704.13997.stgit@PASHA-ISP
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoMerge tag 'tracing-pull-request' into staging
Stefan Hajnoczi [Mon, 8 May 2017 13:39:48 +0000 (09:39 -0400)] 
Merge tag 'tracing-pull-request' into staging

# gpg: Signature made Mon 08 May 2017 09:39:00 AM EDT
# gpg:                using RSA key 0x9CA4ABB381AB73C8
# gpg: Good signature from "Stefan Hajnoczi <stefanha@redhat.com>"
# gpg:                 aka "Stefan Hajnoczi <stefanha@gmail.com>"
# Primary key fingerprint: 8695 A8BF D3F9 7CDA AC35  775A 9CA4 ABB3 81AB 73C8

* tag 'tracing-pull-request':
  trace: disallow more than 10 arguments per trace event

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agotrace: disallow more than 10 arguments per trace event
Daniel P. Berrange [Wed, 26 Apr 2017 15:39:00 +0000 (16:39 +0100)] 
trace: disallow more than 10 arguments per trace event

The UST trace backend can only cope with upto 10 arguments. To ensure we
don't exceed the limit when UST is not compiled in, disallow more than
10 arguments upfront.

This prevents the case where:

  commit 0fc8aec7de64f2bf83a274a2a38b938ce03425d2
  Author: Zhang Chen <zhangchen.fnst@cn.fujitsu.com>
  Date:   Tue Apr 18 10:20:20 2017 +0800

    COLO-compare: Optimize tcp compare trace event

    Optimize two trace events as one, adjust print format make
    it easy to read. rename trace_colo_compare_pkt_info_src/dst
    to trace_colo_compare_tcp_info.

regressed the fix done in

  commit 2dfe5113b11ce0ddb08176ebb54ab7ac4104b413
  Author: Alex Bennée <alex.bennee@linaro.org>
  Date:   Fri Oct 28 14:25:59 2016 +0100

    net: split colo_compare_pkt_info into two trace events

    It seems there is a limit to the number of arguments a UST trace event
    can take and at 11 the previous trace command broke the build. Split the
    trace into a src pkt and dst pkt trace to fix this.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
    Message-id: 20161028132559.8324-1-alex.bennee@linaro.org
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Now we get an immediate fail even when UST is disabled:

  GEN     net/trace.h
Traceback (most recent call last):
  File "/home/berrange/src/virt/qemu/scripts/tracetool.py", line 154, in <module>
    main(sys.argv)
  File "/home/berrange/src/virt/qemu/scripts/tracetool.py", line 145, in main
    events.extend(tracetool.read_events(fh))
  File "/home/berrange/src/virt/qemu/scripts/tracetool/__init__.py", line 307, in read_events
    event = Event.build(line)
  File "/home/berrange/src/virt/qemu/scripts/tracetool/__init__.py", line 244, in build
    event = Event(name, props, fmt, args)
  File "/home/berrange/src/virt/qemu/scripts/tracetool/__init__.py", line 196, in __init__
    "argument count" % name)
ValueError: Event 'colo_compare_tcp_info' has more than maximum permitted argument count
Makefile:96: recipe for target 'net/trace.h-timestamp' failed

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-id: 20170426153900.21066-1-berrange@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agogdbstub: implement remote debugging protocol escapes for command receive
Doug Gale [Mon, 1 May 2017 16:22:10 +0000 (12:22 -0400)] 
gdbstub: implement remote debugging protocol escapes for command receive

- decode escape sequences
- decompress run-length encoding escape sequences
- report command parsing problems to output when debug output is enabled
- reject packet checksums that are not valid hex digits
- compute the checksum based on the packet stream, not based on the
  decoded packet

Tested with GDB and QtCreator integrated debugger on SMP QEMU instance.
Works for me.

Signed-off-by: Doug Gale <doug16k@gmail.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agotcg/mips: fix field extraction opcode
Aurelien Jarno [Sun, 30 Apr 2017 14:46:20 +0000 (16:46 +0200)] 
tcg/mips: fix field extraction opcode

The "msb" argument should correspond to (len - 1).

Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
2 years agoMerge remote-tracking branch 'elmarco/tags/chr-tests-pull-request' into staging
Stefan Hajnoczi [Fri, 5 May 2017 16:04:07 +0000 (17:04 +0100)] 
Merge remote-tracking branch 'elmarco/tags/chr-tests-pull-request' into staging

# gpg: Signature made Thu 04 May 2017 12:42:10 PM BST
# gpg:                using RSA key 0xDAE8E10975969CE5
# gpg: Good signature from "Marc-André Lureau <marcandre.lureau@redhat.com>"
# gpg:                 aka "Marc-André Lureau <marcandre.lureau@gmail.com>"
# Primary key fingerprint: 87A9 BD93 3F87 C606 D276  F62D DAE8 E109 7596 9CE5

* elmarco/tags/chr-tests-pull-request: (21 commits)
  tests: add /char/console test
  tests: add /char/udp test
  tests: add /char/socket test
  tests: add /char/file test
  tests: add /char/pipe test
  tests: add alias check in /char/ringbuf
  char-udp: flush as much buffer as possible
  char-socket: add 'connected' property
  char-socket: add 'addr' property
  char-socket: update local address after listen
  char-socket: introduce update_disconnected_filename()
  char: useless NULL check
  char: remove chardevs list
  char: remove qemu_chardev_add
  char: use /chardevs container instead of chardevs list
  vl: add todo note about root container cleanup
  char: add a /chardevs container
  container: don't leak container reference
  xen: use a better chardev type check
  mux: simplfy muxes_realize_done
  ...

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoMerge remote-tracking branch 'cohuck/tags/s390x-3270-20170504' into staging
Stefan Hajnoczi [Fri, 5 May 2017 15:56:23 +0000 (16:56 +0100)] 
Merge remote-tracking branch 'cohuck/tags/s390x-3270-20170504' into staging

Basic support for using channel-attached 3270 'green-screen'
devices via tn3270. Actual handling of the data stream is
delegated to x3270; more info at http://wiki.qemu.org/Features/3270

# gpg: Signature made Thu 04 May 2017 11:36:51 AM BST
# gpg:                using RSA key 0xDECF6B93C6F02FAF
# gpg: Good signature from "Cornelia Huck <conny@cornelia-huck.de>"
# gpg:                 aka "Cornelia Huck <cohuck@kernel.org>"
# gpg:                 aka "Cornelia Huck <cornelia.huck@de.ibm.com>"
# gpg:                 aka "Cornelia Huck <huckc@linux.vnet.ibm.com>"
# Primary key fingerprint: C3D0 D66D C362 4FF6 A8C0  18CE DECF 6B93 C6F0 2FAF

* cohuck/tags/s390x-3270-20170504:
  s390x/3270: Mark non-migratable and enable the device
  s390x/3270: Detect for continued presence of a 3270 client
  s390x/3270: Add the TCP socket events handler for 3270
  s390x/3270: 3270 data stream handling
  s390x/3270: Add emulated terminal3270 device
  s390x/3270: Add abstract emulated ccw-attached 3270 device
  s390x/css: Add an algorithm to find a free chpid
  chardev: Basic support for TN3270

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoMerge remote-tracking branch 'quintela/tags/migration/20170504' into staging
Stefan Hajnoczi [Fri, 5 May 2017 15:52:12 +0000 (16:52 +0100)] 
Merge remote-tracking branch 'quintela/tags/migration/20170504' into staging

migration/next for 20170504

# gpg: Signature made Thu 04 May 2017 10:35:41 AM BST
# gpg:                using RSA key 0xF487EF185872D723
# gpg: Good signature from "Juan Quintela <quintela@redhat.com>"
# gpg:                 aka "Juan Quintela <quintela@trasno.org>"
# Primary key fingerprint: 1899 FF8E DEBF 58CC EE03  4B82 F487 EF18 5872 D723

* quintela/tags/migration/20170504:
  migration: Extra tracing
  migration: Move postcopy-ram.h to migration/
  monitor: Move hmp_info_snapshots from savevm.c to hmp.c
  monitor: Move hmp_delvm from savevm.c to hmp.c
  monitor: Move hmp_savevm from savevm.c to hmp.c
  monitor: Move hmp_loadvm from monitor.c to hmp.c
  monitor: Remove monitor parameter from save_vmstate
  migration: to_dst_file at that point is NULL
  migration: setup bi-directional I/O channel for exec: protocol
  ram: Split dirty bitmap by RAMBlock

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoMerge remote-tracking branch 'kraxel/tags/pull-audio-20170504-1' into staging
Stefan Hajnoczi [Fri, 5 May 2017 15:46:51 +0000 (16:46 +0100)] 
Merge remote-tracking branch 'kraxel/tags/pull-audio-20170504-1' into staging

audio: cleanups, bugfixes (memory leaks).

# gpg: Signature made Thu 04 May 2017 08:16:50 AM BST
# gpg:                using RSA key 0x4CB6D8EED3E87138
# gpg: Good signature from "Gerd Hoffmann (work) <kraxel@redhat.com>"
# gpg:                 aka "Gerd Hoffmann <gerd@kraxel.org>"
# gpg:                 aka "Gerd Hoffmann (private) <kraxel@gmail.com>"
# Primary key fingerprint: A032 8CFF B93A 17A7 9901  FE7D 4CB6 D8EE D3E8 7138

* kraxel/tags/pull-audio-20170504-1: (30 commits)
  audio: Use ARRAY_SIZE from qemu/osdep.h
  audio: un-export OPLResetChip
  audio: Remove unused typedefs
  audio: UpdateHandler is not used anymore
  audio: IRQHandler is not used anymore
  audio: OPLSetUpdateHandler is not used anywhere
  audio: OPLSetIRQHandler is not used anywhere
  audio: GUSsample is int16_t
  audio: GUSword is uint16_t
  audio: GUSword is uint16_t
  audio: remove GUSchar
  audio: GUSbyte is uint8_t
  audio: Remove unused fields
  audio: Remove type field
  audio: Remove Unused OPL_TYPE_*
  audio: Unfold OPLSAMPLE
  audio: Remove INT32
  audio: remove INT16
  audio: Remove INT8
  audio: remove UINT32
  ...

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoMerge remote-tracking branch 'kraxel/tags/pull-input-20170504-1' into staging
Stefan Hajnoczi [Fri, 5 May 2017 15:30:59 +0000 (16:30 +0100)] 
Merge remote-tracking branch 'kraxel/tags/pull-input-20170504-1' into staging

input: limit kbd queue depth
input: don't queue delay if paused
input: Add trace event for empty keyboard queue

# gpg: Signature made Thu 04 May 2017 06:48:37 AM BST
# gpg:                using RSA key 0x4CB6D8EED3E87138
# gpg: Good signature from "Gerd Hoffmann (work) <kraxel@redhat.com>"
# gpg:                 aka "Gerd Hoffmann <gerd@kraxel.org>"
# gpg:                 aka "Gerd Hoffmann (private) <kraxel@gmail.com>"
# Primary key fingerprint: A032 8CFF B93A 17A7 9901  FE7D 4CB6 D8EE D3E8 7138

* kraxel/tags/pull-input-20170504-1:
  input: Add trace event for empty keyboard queue
  input: don't queue delay if paused
  input: limit kbd queue depth

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoMerge remote-tracking branch 'shorne/tags/pull-or-20170504' into staging
Stefan Hajnoczi [Fri, 5 May 2017 15:21:00 +0000 (16:21 +0100)] 
Merge remote-tracking branch 'shorne/tags/pull-or-20170504' into staging

Openrisc Features and Fixes for qemu 2.10

# gpg: Signature made Thu 04 May 2017 01:41:45 AM BST
# gpg:                using RSA key 0xC3B31C2D5E6627E4
# gpg: Good signature from "Stafford Horne <shorne@gmail.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: D9C4 7354 AEF8 6C10 3A25  EFF1 C3B3 1C2D 5E66 27E4

* shorne/tags/pull-or-20170504:
  target/openrisc: Support non-busy idle state using PMR SPR
  target/openrisc: Remove duplicate features property
  target/openrisc: Implement full vmstate serialization
  migration: Add VMSTATE_STRUCT_2DARRAY()
  target/openrisc: implement shadow registers
  migration: Add VMSTATE_UINTTL_2DARRAY()
  target/openrisc: add numcores and coreid support
  target/openrisc: Fixes for memory debugging
  target/openrisc: Implement EPH bit
  target/openrisc: Implement EVBAR register
  MAINTAINERS: Add myself as openrisc maintainer

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoMerge remote-tracking branch 'awilliam/tags/vfio-updates-20170503.0' into staging
Stefan Hajnoczi [Fri, 5 May 2017 15:14:08 +0000 (16:14 +0100)] 
Merge remote-tracking branch 'awilliam/tags/vfio-updates-20170503.0' into staging

VFIO fixes 2017-05-03

 - Enable 8-byte memory region accesses (Jose Ricardo Ziviani)
 - Fix vfio-pci error message (Dong Jia Shi)

# gpg: Signature made Wed 03 May 2017 10:28:55 PM BST
# gpg:                using RSA key 0x239B9B6E3BB08B22
# gpg: Good signature from "Alex Williamson <alex.williamson@redhat.com>"
# gpg:                 aka "Alex Williamson <alex@shazbot.org>"
# gpg:                 aka "Alex Williamson <alwillia@redhat.com>"
# gpg:                 aka "Alex Williamson <alex.l.williamson@gmail.com>"
# Primary key fingerprint: 42F6 C04E 540B D1A9 9E7B  8A90 239B 9B6E 3BB0 8B22

* awilliam/tags/vfio-updates-20170503.0:
  vfio/pci: Fix incorrect error message
  vfio: enable 8-byte reads/writes to vfio
  vfio: Set MemoryRegionOps:max_access_size and min_access_size

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoMerge remote-tracking branch 'cohuck/tags/s390x-20170502' into staging
Stefan Hajnoczi [Fri, 5 May 2017 14:59:56 +0000 (15:59 +0100)] 
Merge remote-tracking branch 'cohuck/tags/s390x-20170502' into staging

More s390x patches, this time boot related:
- LOADPARM machine property, exposed to the guest via SCLP and
  diagnose 308
- Use LOADPARM in the s390-ccw bios to select a boot entry
- Fix a crash in the ipl device code when a virtio-scsi-pci device
  has been specified

# gpg: Signature made Tue 02 May 2017 02:29:26 PM BST
# gpg:                using RSA key 0xDECF6B93C6F02FAF
# gpg: Good signature from "Cornelia Huck <conny@cornelia-huck.de>"
# gpg:                 aka "Cornelia Huck <cohuck@kernel.org>"
# gpg:                 aka "Cornelia Huck <cornelia.huck@de.ibm.com>"
# gpg:                 aka "Cornelia Huck <huckc@linux.vnet.ibm.com>"
# Primary key fingerprint: C3D0 D66D C362 4FF6 A8C0  18CE DECF 6B93 C6F0 2FAF

* cohuck/tags/s390x-20170502:
  hw/s390x/ipl: Fix crash with virtio-scsi-pci device
  pc-bios/s390-ccw.img: update image
  pc-bios/s390-ccw: add boot entry selection to El Torito routine
  pc-bios/s390-ccw: add boot entry selection for ECKD DASD
  pc-bios/s390-ccw: provide entry selection on LOADPARM for SCSI disk
  pc-bios/s390-ccw: provide a function to interpret LOADPARM value
  pc-bios/s390-ccw: get LOADPARM stored in SCP Read Info
  pc-bios/s390-ccw: Make ebcdic/ascii conversion public
  util/qemu-config: Add loadparm to qemu machine_opts
  hw/s390x/sclp: update LOADPARM in SCP Info
  hw/s390x/ipl: enable LOADPARM in IPIB for a boot device
  hw/s390x: provide loadparm property for the machine

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agovhost-scsi: create a vhost-scsi-common abstraction
Felipe Franciosi [Thu, 2 Mar 2017 18:25:51 +0000 (10:25 -0800)] 
vhost-scsi: create a vhost-scsi-common abstraction

In order to introduce a new vhost-user-scsi host device type, it makes
sense to abstract part of vhost-scsi into a common parent class. This
commit does exactly that.

Signed-off-by: Felipe Franciosi <felipe@nutanix.com>
Message-Id: <1488479153-21203-3-git-send-email-felipe@nutanix.com>

2 years agolibvhost-user: replace vasprintf() to fix build
Felipe Franciosi [Thu, 2 Mar 2017 18:25:50 +0000 (10:25 -0800)] 
libvhost-user: replace vasprintf() to fix build

On gcc 3.4 and newer, simply using (void) in front of WUR functions is
not sufficient to ignore the return value. That prevents a build when
handling warnings as errors.

libvhost-user had a usage of (void)vasprintf() which triggered such a
condition. This fixes it by replacing this call with g_strdup_vprintf()
which aborts on OOM.

Signed-off-by: Felipe Franciosi <felipe@nutanix.com>
Message-Id: <1488479153-21203-2-git-send-email-felipe@nutanix.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
2 years agoget_maintainer: add subsystem to reviewer output
Joe Perches [Sat, 7 Nov 2015 00:30:52 +0000 (16:30 -0800)] 
get_maintainer: add subsystem to reviewer output

Reviewer output currently does not include the subsystem
that matched.  Add it.

Miscellanea:

o Add a get_subsystem_name routine to centralize this

Cherry picked from Linux commit 2a7cb1dc82fc2a52e747b4c496c13f6575fb1790.

Signed-off-by: Joe Perches <joe@perches.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoget_maintainer: --r (list reviewer) is on by default
Brian Norris [Sat, 7 Nov 2015 00:30:49 +0000 (16:30 -0800)] 
get_maintainer: --r (list reviewer) is on by default

We don't consistenly document the default value next to the option
listing, but we do have a list of defaults here, so let's keep it up to
date.

Cherry picked from Linux commit 4f07510df2e8c47fd65b8ffaaf6c5d334d59d598.

Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoget_maintainer: it's '--pattern-depth', not '-pattern-depth'
Brian Norris [Wed, 3 May 2017 09:16:44 +0000 (11:16 +0200)] 
get_maintainer: it's '--pattern-depth', not '-pattern-depth'

Though it appears that Perl's GetOptions will take either, the latter is
not documented in the options listing.

Cherry picked from Linux commit cc7ff0ef6eca3deeea4a424ca47a67c8450d5424.

Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoget_maintainer: Teach get_maintainer.pl about the new "R:" tag
Joe Perches [Mon, 2 Jun 2014 19:05:17 +0000 (12:05 -0700)] 
get_maintainer: Teach get_maintainer.pl about the new "R:" tag

We can now designate reviewers in the MAINTAINERS file with the new
"R:" tag, so this commit teaches get_maintainers.pl to add their
email addresses.

Cherry picked from Linux commit c1c3f2c906e35bcb6e4cdf5b8e077660fead14fe,
with fixes to avoid \C as in QEMU commit ba10f729f1 ("get_maintainer.pl:
\C is deprecated", 2015-09-25).

Signed-off-by: Joe Perches <joe@perches.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoMAINTAINERS: Add "R:" tag for self-appointed reviewers
Paul E. McKenney [Mon, 2 Jun 2014 16:06:37 +0000 (09:06 -0700)] 
MAINTAINERS: Add "R:" tag for self-appointed reviewers

Some people are not content with the amount of mail they get, and would
like to be CCed on patches for areas they do not maintain.  Let them
satisfy their own appetite for qemu-devel messages.

Seriously: the purpose here is a bit different from the Linux kernel.
While Linux uses "R" to designate non-maintainers for reviewing patches
in a given area, in QEMU I would also like to use "R" so that people can
delegate sending pull requests while keeping some degree of oversight.

Based on Linux commit eafbaac3093760d1fd3b2a5b9f016362dd68af36.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoFix the -accel parameter and the documentation for 'hax'
Thomas Huth [Thu, 4 May 2017 05:24:41 +0000 (07:24 +0200)] 
Fix the -accel parameter and the documentation for 'hax'

Since 'hax' is a possible accelerator nowadays, too, the '-accel'
option should support it and we should mention this accelerator
in the documentation, too.

Signed-off-by: Thomas Huth <thuth@redhat.com>
Message-Id: <1493875481-16388-1-git-send-email-thuth@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agodump: Acquire BQL around vm_start() in dump thread
Fam Zheng [Wed, 3 May 2017 07:28:19 +0000 (15:28 +0800)] 
dump: Acquire BQL around vm_start() in dump thread

This fixes an assertion failure in the following backtrace:

    __GI___assert_fail
    memory_region_transaction_commit
    memory_region_add_eventfd
    virtio_pci_ioeventfd_assign
    virtio_bus_set_host_notifier
    virtio_blk_data_plane_start
    virtio_bus_start_ioeventfd
    virtio_vmstate_change
    vm_state_notify
    vm_prepare_start
    vm_start
    dump_cleanup
    dump_process
    dump_thread
    start_thread
    clone

vm_start need BQL, acquire it if doing cleaning up from main thread.

Signed-off-by: Fam Zheng <famz@redhat.com>
Message-Id: <20170503072819.14462-1-famz@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agohax: Fix memory mapping de-duplication logic
Yu Ning [Fri, 28 Apr 2017 07:27:23 +0000 (15:27 +0800)] 
hax: Fix memory mapping de-duplication logic

hax_update_mapping() avoids unnecessary and potentially expensive
calls to HAX_VM_IOCTL_SET_RAM by computing the net result (i.e.
effective mapping changes) of each MemoryRegion transaction, with
the help of a linked list of HAXMapping objects.

However, when processing a new mapping that overlaps with an
existing mapping in the list, it fails to handle the case where the
start address of the new mapping is above that of the existing
mapping in the guest physical address space. This happens when QEMU
is launched with "-machine q35 -enable-hax", which involves the
following MemoryRegion transaction for digging the VGA hole:

 region_del: 0x00000000->0x08000000 VA 05fa0000 ('pc.ram')
 region_add: 0x00000000->0x000a0000 VA 05fa0000 ('pc.ram')
 region_add: 0x000a0000->0x000c0000 VA 00000000 ('vga-lowmem')
 region_add: 0x000c0000->0x08000000 VA 06060000 ('pc.ram')

where the third MemoryRegion is MMIO and is ignored. The current
de-duplication logic handles the last MemoryRegion incorrectly and
produces the following result:

 hax_mapping_dump_list updates:
         + 0x000c0000->0x08000000 VA 0x06060000
         - 0x07fe0000->0x08000000 VA 0x0df80000

which is why VGA emulation does not work for Q35.

With this patch, one can see VGA output as Q35 boots up. Note that
Q35 support also requires a change to HAXM kernel module, which is
not available in the current HAXM release (6.1.2).

+ Add a warning if the input MemoryRegion is a ROM device, which is
  not supported by HAXM kernel module at this time.

Signed-off-by: Yu Ning <yu.ning@linux.intel.com>
Message-Id: <20170428072723.7036-1-yu.ning@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agocheckpatch: Disallow glib asserts in main code
Dr. David Alan Gilbert [Thu, 27 Apr 2017 16:55:26 +0000 (17:55 +0100)] 
checkpatch: Disallow glib asserts in main code

Glib commit a6a875068779 (from 2013) made many of the glib assert
macros non-fatal if a flag is set.
This causes two problems:
  a) Compilers moan that your code is unsafe even though you've
     put an assert in before the point of use.
  b) Someone evil could, in a library, call
     g_test_set_nonfatal_assertions() and cause our assertions in
     important places not to fail and potentially allow memory overruns.

Ban most of the glib assertion functions (basically everything except
g_assert and g_assert_not_reached) except in tests/

This makes checkpatch gives an error such as:

  ERROR: Use g_assert or g_assert_not_reached
  #77: FILE: vl.c:4725:
  +    g_assert_cmpstr("Chocolate", >, "Cheese");

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20170427165526.19836-1-dgilbert@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agotrace: add qemu mutex lock and unlock trace events
Jose Ricardo Ziviani [Mon, 24 Apr 2017 17:19:58 +0000 (14:19 -0300)] 
trace: add qemu mutex lock and unlock trace events

These trace events were very useful to help me to understand and find a
reordering issue in vfio, for example:

qemu_mutex_lock locked mutex 0x10905ad8
  vfio_region_write  (0001:03:00.0:region1+0xc0, 0x2020c, 4)
qemu_mutex_unlock unlocked mutex 0x10905ad8
qemu_mutex_lock locked mutex 0x10905ad8
  vfio_region_write  (0001:03:00.0:region1+0xc4, 0xa0000, 4)
qemu_mutex_unlock unlocked mutex 0x10905ad8

that also helped me to see the desired result after the fix:

qemu_mutex_lock locked mutex 0x10905ad8
  vfio_region_write  (0001:03:00.0:region1+0xc0, 0x2000c, 4)
  vfio_region_write  (0001:03:00.0:region1+0xc4, 0xb0000, 4)
qemu_mutex_unlock unlocked mutex 0x10905ad8

So it could be a good idea to have these traces implemented. It's worth
mentioning that they should be surgically enabled during the debugging,
otherwise it can flood the trace logs with lock/unlock messages.

How to use it:
trace-event qemu_mutex_lock on|off
trace-event qemu_mutex_unlock on|off
or
trace-event qemu_mutex* on|off

Signed-off-by: Jose Ricardo Ziviani <joserz@linux.vnet.ibm.com>
Message-Id: <1493054398-26013-1-git-send-email-joserz@linux.vnet.ibm.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
[Also handle trylock, cond_wait and win32; trace "unlocked" while still
 in the critical section, so that "unlocked" always comes before the
 next "locked" tracepoint. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agovmw_pvscsi: check message ring page count at initialisation
P J P [Tue, 25 Apr 2017 13:06:23 +0000 (18:36 +0530)] 
vmw_pvscsi: check message ring page count at initialisation

A guest could set the message ring page count to zero, resulting in
infinite loop. Add check to avoid it.

Reported-by: YY Z <bigbird475958471@gmail.com>
Signed-off-by: P J P <ppandit@redhat.com>
Message-Id: <20170425130623.3649-1-ppandit@redhat.com>
Reviewed-by: Dmitry Fleytman <dmitry@daynix.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agosgabios: update for "fix wrong video attrs for int 10h,ah==13h"
Paolo Bonzini [Thu, 27 Apr 2017 08:39:14 +0000 (10:39 +0200)] 
sgabios: update for "fix wrong video attrs for int 10h,ah==13h"

Update the submodule and rebuild the binary.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoscsi: avoid an off-by-one error in megasas_mmio_write
Prasad J Pandit [Mon, 24 Apr 2017 12:06:34 +0000 (17:36 +0530)] 
scsi: avoid an off-by-one error in megasas_mmio_write

While reading magic sequence(MFI_SEQ) in megasas_mmio_write,
an off-by-one error could occur as 's->adp_reset' index is not
reset after reading the last sequence.

Reported-by: YY Z <bigbird475958471@gmail.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Message-Id: <20170424120634.12268-1-ppandit@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agovl: deprecate the "-hdachs" option
Thomas Huth [Thu, 27 Apr 2017 05:20:54 +0000 (07:20 +0200)] 
vl: deprecate the "-hdachs" option

If the user needs to specify the disk geometry, the corresponding
parameters of the "-device ide-hd" option should be used instead.
"-hdachs" is considered as deprecated and might be removed soon.

Signed-off-by: Thomas Huth <thuth@redhat.com>
Message-Id: <1493270454-1448-1-git-send-email-thuth@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agouse _Static_assert in QEMU_BUILD_BUG_ON
Andreas Grapentin [Tue, 14 Mar 2017 16:59:53 +0000 (17:59 +0100)] 
use _Static_assert in QEMU_BUILD_BUG_ON

QEMU_BUILD_BUG_ON should use C11's _Static_assert, if the compiler supports it,
to provide more readable messages on failure.

We check for _Static_assert in configure, and set CONFIG_STATIC_ASSERT
accordingly. QEMU_BUILD_BUG_ON invokes _Static_assert if CONFIG_STATIC_ASSERT
is defined, and reverts to the old way otherwise.

That way, systems without C11 conforming compiler will still have the old
messages, as verified by intentionally breaking the configure check.

the following example output was generated by inverting the condition in
QEMU_BUILD_BUG_ON:

without _Static_assert:

> In file included from /qemu/include/qemu/osdep.h:36:0,
>                  from /qemu/qga/commands.c:13:
> /qemu/qga/commands.c: In function â€˜qmp_guest_exec_status’:
> /qemu/include/qemu/compiler.h:89:12: error: negative width in bit-field â€˜<anonymous>’
>      struct { \
>             ^
> /qemu/include/qemu/compiler.h:96:38: note: in expansion of macro  QEMU_BUILD_BUG_ON_STRUCT’
>  #define QEMU_BUILD_BUG_ON(x) typedef QEMU_BUILD_BUG_ON_STRUCT(x) \
>                                       ^~~~~~~~~~~~~~~~~~~~~~~~
> /qemu/include/qemu/atomic.h:146:5: note: in expansion of macro â€˜QEMU_BUILD_BUG_ON’
>      QEMU_BUILD_BUG_ON(sizeof(*ptr) > sizeof(void *));   \
>      ^~~~~~~~~~~~~~~~~
> /qemu/include/qemu/atomic.h:417:5: note: in expansion of macro â€˜atomic_load_acquire’
>      atomic_load_acquire(ptr)
>      ^~~~~~~~~~~~~~~~~~~
> /qemu/qga/commands.c:160:21: note: in expansion of macro â€˜atomic_mb_read’
>      bool finished = atomic_mb_read(&gei->finished);
>                      ^~~~~~~~~~~~~~

with _Static_assert:

> In file included from /qemu/include/qemu/osdep.h:36:0,
>                  from /qemu/qga/commands.c:13:
> /qemu/qga/commands.c: In function â€˜qmp_guest_exec_status’:
> /qemu/include/qemu/compiler.h:94:30: error: static assertion failed: "not expecting: sizeof(*&gei->finished) > sizeof(void *)"
>  #define QEMU_BUILD_BUG_ON(x) _Static_assert((x), #x)
>                               ^
> /qemu/include/qemu/atomic.h:146:5: note: in expansion of macro â€˜QEMU_BUILD_BUG_ON’
>      QEMU_BUILD_BUG_ON(sizeof(*ptr) > sizeof(void *));   \
>      ^~~~~~~~~~~~~~~~~
> /qemu/include/qemu/atomic.h:417:5: note: in expansion of macro â€˜atomic_load_acquire’
>      atomic_load_acquire(ptr)
>      ^~~~~~~~~~~~~~~~~~~
> /qemu/qga/commands.c:160:21: note: in expansion of macro â€˜atomic_mb_read’
>      bool finished = atomic_mb_read(&gei->finished);
>                      ^~~~~~~~~~~~~~

Signed-off-by: Andreas Grapentin <andreas@grapentin.org>
Message-Id: <20170314165953.18506-1-andreas@grapentin.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agotarget/i386: Add GDB XML register description support
Abdallah Bouassida [Wed, 28 Dec 2016 16:34:02 +0000 (17:34 +0100)] 
target/i386: Add GDB XML register description support

This patch implements XML target description support for X86 and X86-64
architectures in the GDB stub, as the way with ARM and PowerPC:
- gdb-xml/32bit-core.xml & gdb-xml/64bit-core.xml: Adding the XML target
  description files, these files are picked from GDB source code.
- configure: Define gdb_xml_files for X86 targets.
- target/i386/cpu.c: Define gdb_core_xml_file and gdb_arch_name to add
  XML awareness for this architecture, modify the gdb_num_core_regs to
  fit the registers number defined in each XML file.

Signed-off-by: Abdallah Bouassida <abdallah.bouassida@lauterbach.com>
Message-Id: <2b3c8119-1602-28c7-eab4-296593877103@lauterbach.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agocg3: add explicit ram_addr_t cast to scanline page variable
Mark Cave-Ayland [Mon, 1 May 2017 07:33:47 +0000 (08:33 +0100)] 
cg3: add explicit ram_addr_t cast to scanline page variable

Coverity warns that multiplying two 32-bit values gives a 32-bit result which
is assigned to a 64-bit variable. Add an explicit ram_addr_t cast to silence
the warning.

Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
2 years agotcx: fix cut/paste error in update_palette_entries()
Mark Cave-Ayland [Mon, 1 May 2017 07:33:47 +0000 (08:33 +0100)] 
tcx: fix cut/paste error in update_palette_entries()

Commit ee72bed0 "tcx: remove primitives for non-32-bit surfaces" accidentally
left a trailing break in update_palette_entries() causing the palette update
routine to exit after just one iteration. Remove it.

Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
2 years agoMerge remote-tracking branch 'kwolf/tags/for-upstream' into staging
Stefan Hajnoczi [Thu, 4 May 2017 12:44:32 +0000 (13:44 +0100)] 
Merge remote-tracking branch 'kwolf/tags/for-upstream' into staging

Block layer patches

# gpg: Signature made Fri 28 Apr 2017 09:20:17 PM BST
# gpg:                using RSA key 0x7F09B272C88F2FD6
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>"
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74  56FE 7F09 B272 C88F 2FD6

* kwolf/tags/for-upstream: (34 commits)
  progress: Show current progress on SIGINFO
  iotests: fix exclusion option
  iotests: clarify help text
  qemu-img: use blk_co_pwrite_zeroes for zero sectors when compressed
  qemu-img: improve convert_iteration_sectors()
  block: assert no image modification under BDRV_O_INACTIVE
  block: fix obvious coding style mistakes in block_int.h
  qcow2: Allow discard of final unaligned cluster
  block: Add .bdrv_truncate() error messages
  block: Add errp to BD.bdrv_truncate()
  block: Add errp to b{lk,drv}_truncate()
  block/vhdx: Make vhdx_create() always set errp
  qemu-img: Document backing options
  qemu-img/convert: Move bs_n > 1 && -B check down
  qemu-img/convert: Use @opts for one thing only
  block: fix alignment calculations in bdrv_co_do_zero_pwritev
  block: Do not unref bs->file on error in BD's open
  iotests: 109: Filter out "len" of failed jobs
  iotests: Fix typo in 026
  Issue a deprecation warning if the user specifies the "-hdachs" option.
  ...

Message-id: 1493411622-5343-1-git-send-email-kwolf@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agotests: add /char/console test
Marc-André Lureau [Tue, 3 Jan 2017 18:55:55 +0000 (19:55 +0100)] 
tests: add /char/console test

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
2 years agotests: add /char/udp test
Marc-André Lureau [Tue, 3 Jan 2017 18:22:19 +0000 (19:22 +0100)] 
tests: add /char/udp test

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
2 years agotests: add /char/socket test
Marc-André Lureau [Tue, 3 Jan 2017 18:22:03 +0000 (19:22 +0100)] 
tests: add /char/socket test

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
2 years agotests: add /char/file test
Marc-André Lureau [Tue, 3 Jan 2017 18:21:39 +0000 (19:21 +0100)] 
tests: add /char/file test

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
2 years agotests: add /char/pipe test
Marc-André Lureau [Tue, 3 Jan 2017 18:20:05 +0000 (19:20 +0100)] 
tests: add /char/pipe test

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
2 years agotests: add alias check in /char/ringbuf
Marc-André Lureau [Tue, 3 Jan 2017 18:21:09 +0000 (19:21 +0100)] 
tests: add alias check in /char/ringbuf

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
2 years agochar-udp: flush as much buffer as possible
Marc-André Lureau [Fri, 23 Dec 2016 13:55:54 +0000 (14:55 +0100)] 
char-udp: flush as much buffer as possible

Instead of flushing the buffer byte by byte, call qemu_chr_be_write()
with as much byte possible accepted by the front-end.

Factor out buffer flushing in a common function udp_chr_flush_buffer().

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
2 years agochar-socket: add 'connected' property
Marc-André Lureau [Wed, 21 Dec 2016 14:43:46 +0000 (15:43 +0100)] 
char-socket: add 'connected' property

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
2 years agochar-socket: add 'addr' property
Marc-André Lureau [Wed, 21 Dec 2016 11:26:38 +0000 (12:26 +0100)] 
char-socket: add 'addr' property

Add a property to lookup the connection details.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
2 years agochar-socket: update local address after listen
Marc-André Lureau [Tue, 20 Dec 2016 14:36:57 +0000 (15:36 +0100)] 
char-socket: update local address after listen

This is mainly useful to know the actual bound port when using port 0.

For example, when starting qemu with socket on port 0, before:
QEMU waiting for connection on: disconnected:tcp:localhost:0,server
After:
QEMU waiting for connection on: disconnected:tcp:localhost:32454,server

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
2 years agochar-socket: introduce update_disconnected_filename()
Marc-André Lureau [Tue, 20 Dec 2016 14:32:31 +0000 (15:32 +0100)] 
char-socket: introduce update_disconnected_filename()

This helper will be used in yet another place in the following patch.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
2 years agochar: useless NULL check
Marc-André Lureau [Wed, 14 Dec 2016 20:15:44 +0000 (23:15 +0300)] 
char: useless NULL check

g_strdup(NULL) returns NULL already.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
2 years agochar: remove chardevs list
Marc-André Lureau [Wed, 14 Dec 2016 19:47:29 +0000 (22:47 +0300)] 
char: remove chardevs list

The list is now empty, the chardev cleanup is taken care of by the unref
of the root container.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
2 years agochar: remove qemu_chardev_add
Marc-André Lureau [Wed, 14 Dec 2016 17:58:50 +0000 (20:58 +0300)] 
char: remove qemu_chardev_add

qemu_chardev_new() now uses object_new_with_props() with /chardevs
parent container. It will fail to insert the object if the same "id"
already exists. "chardevs" list usage has been removed in previous
commits.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
2 years agochar: use /chardevs container instead of chardevs list
Marc-André Lureau [Wed, 14 Dec 2016 13:44:19 +0000 (16:44 +0300)] 
char: use /chardevs container instead of chardevs list

Use object_resolve_path_component() and object_child_foreach() on
/chardevs container instead of iterating over chardevs list.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
2 years agovl: add todo note about root container cleanup
Marc-André Lureau [Fri, 10 Feb 2017 10:27:28 +0000 (14:27 +0400)] 
vl: add todo note about root container cleanup

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
2 years agochar: add a /chardevs container
Marc-André Lureau [Wed, 14 Dec 2016 12:23:36 +0000 (15:23 +0300)] 
char: add a /chardevs container

Add a /chardevs container object to hold the list of chardevs.
(Note: QTAILQ chardevs is going away in the following commits)

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
2 years agocontainer: don't leak container reference
Marc-André Lureau [Wed, 14 Dec 2016 19:42:55 +0000 (22:42 +0300)] 
container: don't leak container reference

object_property_add_child() references the child, unref it after to
avoid ref leaks.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
2 years agoxen: use a better chardev type check
Marc-André Lureau [Wed, 14 Dec 2016 12:23:13 +0000 (15:23 +0300)] 
xen: use a better chardev type check

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
2 years agomux: simplfy muxes_realize_done
Marc-André Lureau [Wed, 14 Dec 2016 11:40:45 +0000 (14:40 +0300)] 
mux: simplfy muxes_realize_done

mux_chr_event() already send events to all backends, rename it,
export it, and use it from muxes_realize_done. This should help abstract
away mux implementation.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
2 years agochar: remove qemu_chr_be_generic_open
Marc-André Lureau [Wed, 14 Dec 2016 11:23:02 +0000 (14:23 +0300)] 
char: remove qemu_chr_be_generic_open

The function simply alias and hides the real event function.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
2 years agomigration: Extra tracing
Dr. David Alan Gilbert [Wed, 26 Apr 2017 18:37:21 +0000 (19:37 +0100)] 
migration: Extra tracing

A couple more traces that would have made fixing that postcopy
bug a bit easier.

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
2 years agomigration: Move postcopy-ram.h to migration/
Juan Quintela [Thu, 20 Apr 2017 11:12:24 +0000 (13:12 +0200)] 
migration: Move postcopy-ram.h to migration/

It is internal to migration, not intended for other users.

Signed-off-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2 years agos390x/3270: Mark non-migratable and enable the device
Jing Liu [Thu, 12 Jan 2017 09:38:25 +0000 (10:38 +0100)] 
s390x/3270: Mark non-migratable and enable the device

Mark 3270 as non-migratable for the experimental stage. Enable
the 3270 device so that we can use x3270 client to operate the guest.

Run qemu with the arguments:
    -chardev socket,id=char3270_0,host=0.0.0.0,port=23,nowait,server,tn3270 \
    -device x-terminal3270,chardev=char3270_0,devno=fe.0.000a,id=terminal3270_0 \

There are some restrictions for the first stage: We don't support SSL
connections, multiple client connections and client resizing. Only
tested with the x3270 client.

Signed-off-by: Jing Liu <liujbjl@linux.vnet.ibm.com>
Signed-off-by: Yang Chen <bjcyang@linux.vnet.ibm.com>
Reviewed-by: QingFeng Hao <haoqf@linux.vnet.ibm.com>
Reviewed-by: Dong Jia Shi <bjsdjshi@linux.vnet.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
2 years agos390x/3270: Detect for continued presence of a 3270 client
Jing Liu [Fri, 14 Oct 2016 09:16:23 +0000 (11:16 +0200)] 
s390x/3270: Detect for continued presence of a 3270 client

To ensure that we do not keep any 3270 sockets where the client is not
connected anymore, we send a packet with the timing mark option after
ten minutes of client inactivity. If the client does not answer it,
then the socket will be closed automatically.

This helps to ensure that there is no half-open situation on the 3270
socket.

Signed-off-by: Jing Liu <liujbjl@linux.vnet.ibm.com>
Reviewed-by: QingFeng Hao <haoqf@linux.vnet.ibm.com>
Reviewed-by: Dong Jia Shi <bjsdjshi@linux.vnet.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
2 years agos390x/3270: Add the TCP socket events handler for 3270
Jing Liu [Thu, 21 Jul 2016 06:00:58 +0000 (08:00 +0200)] 
s390x/3270: Add the TCP socket events handler for 3270

This introduces a chr_event handler to handle the 3270 connection
and disconnection events.

Signed-off-by: Jing Liu <liujbjl@linux.vnet.ibm.com>
Reviewed-by: QingFeng Hao <haoqf@linux.vnet.ibm.com>
Reviewed-by: Dong Jia Shi <bjsdjshi@linux.vnet.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
2 years agos390x/3270: 3270 data stream handling
Jing Liu [Fri, 1 Apr 2016 06:32:58 +0000 (08:32 +0200)] 
s390x/3270: 3270 data stream handling

This introduces the input and output handlers for 3270 device, setting
up the data tunnel among guest kernel, qemu and the 3270 client.

After the client connected and TN3270 handshake done, signal the not-ready
to ready status by an unsolicited device-end interrupt, and then the 3270
data stream could be handled correctly between the channel and socket.
Multiple commands generated by "Reset" key on x3270 are not supported now,
just simply terminate the connection.

Signed-off-by: Jing Liu <liujbjl@linux.vnet.ibm.com>
Signed-off-by: Yang Chen <bjcyang@linux.vnet.ibm.com>
Reviewed-by: QingFeng Hao <haoqf@linux.vnet.ibm.com>
Reviewed-by: Dong Jia Shi <bjsdjshi@linux.vnet.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
2 years agos390x/3270: Add emulated terminal3270 device
Yang Chen [Fri, 19 Feb 2016 09:45:26 +0000 (10:45 +0100)] 
s390x/3270: Add emulated terminal3270 device

This is a basic implementation of the emulated ccw-attached 3270
called x-terminal3270, which provides visibility of the device in
the qemu monitor and guest. The x prefix indicates that this is
just an experimental implementation for the current stage. This
device will not be compiled until the basic functions are available.

Signed-off-by: Yang Chen <bjcyang@linux.vnet.ibm.com>
Signed-off-by: Jing Liu <liujbjl@linux.vnet.ibm.com>
Reviewed-by: QingFeng Hao <haoqf@linux.vnet.ibm.com>
Reviewed-by: Dong Jia Shi <bjsdjshi@linux.vnet.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
2 years agos390x/3270: Add abstract emulated ccw-attached 3270 device
Yang Chen [Mon, 19 Sep 2016 07:46:33 +0000 (09:46 +0200)] 
s390x/3270: Add abstract emulated ccw-attached 3270 device

This introduces the infrastructure for the emulated 3270
devices, which will be attached to the virtual-css-bus.

Signed-off-by: Yang Chen <bjcyang@linux.vnet.ibm.com>
Signed-off-by: Jing Liu <liujbjl@linux.vnet.ibm.com>
Reviewed-by: QingFeng Hao <haoqf@linux.vnet.ibm.com>
Reviewed-by: Dong Jia Shi <bjsdjshi@linux.vnet.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
2 years agos390x/css: Add an algorithm to find a free chpid
Jing Liu [Mon, 19 Sep 2016 07:10:43 +0000 (09:10 +0200)] 
s390x/css: Add an algorithm to find a free chpid

This introduces a function named css_find_free_chpid() to find a
free channel path. Because virtio-ccw device used zero as its
channel path number, it would be sensible to skip the reserved one
and search upwards.

Signed-off-by: Jing Liu <liujbjl@linux.vnet.ibm.com>
Reviewed-by: QingFeng Hao <haoqf@linux.vnet.ibm.com>
Reviewed-by: Dong Jia Shi <bjsdjshi@linux.vnet.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>