Use VIR_AUTOFREE and get rid of the cleanup section.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

Require that all headers are guarded by a symbol named

  LIBVIRT_$FILENAME

where $FILENAME is the uppercased filename, with all characters
outside a-z changed into '_'.

Note we do not use a leading __ because that is technically a
namespace reserved for the toolchain.
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

This introduces a syntax-check script that validates header files use a
common layout:

  /*
   ...copyright header...
   */
  <one blank line>
  #ifndef SYMBOL
  # define SYMBOL
  ....content....
  #endif /* SYMBOL */

For any file ending priv.h, before the #ifndef, we will require a
guard to prevent bogus imports:

  #ifndef SYMBOL_ALLOW
  # error ....
  #endif /* SYMBOL_ALLOW */
  <one blank line>

The many mistakes this script identifies are then fixed.
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

All of the ones being removed are pulled in by internal.h. The only
exception is sanlock which expects the application to include <stdint.h>
before sanlock's headers, because sanlock prototypes use fixed width
int, but they don't include stdint.h themselves, so we have to leave
that one in place.
Signed-off-by: NErik Skultety <eskultet@redhat.com>
Acked-by: NMichal Privoznik <mprivozn@redhat.com>

It doesn't really make sense for us to have stdlib.h and string.h but
not stdio.h in the internal.h header.
Signed-off-by: NErik Skultety <eskultet@redhat.com>
Acked-by: NMichal Privoznik <mprivozn@redhat.com>

virQEMUQAPISchemaPathGet returns success when a given schema path was
not found but the returned object is set to NULL. This meant that we'd
call testQEMUSchemaValidate with the schemaroot being NULL which lead to
a crash if a mistyped monitor command was tested.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>

The test file can be broken up by newlines and is automatically
concatenated back. Fix the control flow so that the concatenation code
'continues' the loop rather than branching out.

Also add an anotation to the concatenation code.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>

On EOF, the loop can be terminated right away since most of it is
skipped anyways and the handling of the last command is repeated after
the loop.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>

Since libvirt called bind() and listen() on the UNIX socket, it is
guaranteed that connect() will immediately succeed, if QEMU is running
normally. It will only fail if QEMU has closed the monitor socket by
mistake or if QEMU has exited, letting the kernel close it.

With this in mind we can remove the retry loop and timeout when
connecting to the QEMU monitor if we are doing FD passing. Libvirt can
go straight to sending the QMP greeting and will simply block waiting
for a reply until QEMU is ready.
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

Extended the json monitor test program with support for query-cpus-fast
and added a sample file set for x86 data obtained using the it.
Signed-off-by: NViktor Mihajlovski <mihajlov@linux.vnet.ibm.com>
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>

Add infrastructure that will allow testing schema of the commands we
pass to the fake monitor object, so that we can make sure that it
actually does something.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>

Use the return value of virObjectRef directly. This way, it's easier
for another reader to identify the reason why the additional reference
is required.
Signed-off-by: NMarc Hartmayer <mhartmay@linux.vnet.ibm.com>
Reviewed-by: NBjoern Walk <bwalk@linux.vnet.ibm.com>

There were couple of reports on the list (e.g. [1]) that guests
with huge amounts of RAM are unable to start because libvirt
kills qemu in the initialization phase. The problem is that if
guest is configured to use hugepages kernel has to zero them all
out before handing over to qemu process. For instance, 402GiB
worth of 1GiB pages took around 105 seconds (~3.8GiB/s). Since we
do not want to make the timeout for connecting to monitor
configurable, we have to teach libvirt to count with this
fact. This commit implements "1s per each 1GiB of RAM" approach
as suggested here [2].

1: https://www.redhat.com/archives/libvir-list/2017-March/msg00373.html
2: https://www.redhat.com/archives/libvir-list/2017-March/msg00405.htmlSigned-off-by: NMichal Privoznik <mprivozn@redhat.com>

Similar to the existing qemuMonitorTestNewFromFile the *Full version
will allow to check both commands and supply responses for a better
monitor testing.

Add test monitor infrastructure that will test the commands verbatim
rather than trying to do any smart handling.

Add code that takes a string and matches it against the data passed as
arguments from qemu. This is a simpler version of
qemuMonitorTestAddItemParams.

The current version uses the first JSON reply from the file as monitor
greeting. With the new parameter the caller can now request a simple
test monitor to be created, which uses an artificial greeting and uses
all JSON strings from the file as regular replies.
Signed-off-by: NJiri Denemark <jdenemar@redhat.com>

Signed-off-by: NJiri Denemark <jdenemar@redhat.com>

It's a convenient wrapper around qemuMonitorTestNew which feeds the test
monitor with QMP replies from a specified file.
Signed-off-by: NJiri Denemark <jdenemar@redhat.com>

qemu HMP commands sent by libvirt are terminated just by a '\r'. The
fake monitor used in tests wasn't prepared to handle this and the
communication would hang on an attempt to do a HMP conversation.

Add a special case for handling commands separated by \r in case HMP is
used.

The qemu JSON monitor test allows to test also expected command
arguments. As the error from the monitor simulator is returned as a
simulated qemu error (in JSON) all other JSON contained in the error
message needs to be escaped. This will happen if the monitor command
under test receives a JSON array as an argument.

This will improve the error message from:
libvirt:  error : internal error: cannot parse json { "error":  { "desc":
"Invalid value of argument 'keys' of command 'send-key': expected 'ble'
got '[{"type":"number","data":43},{"type":"number","data":26},
{"type":"number","data":46},{"type":"number","data":32}]'",
"class": "UnexpectedCommand" } }: lexical error: invalid string in json text.

To:
libvirt: QEMU Driver error : internal error: unable to execute QEMU
command 'send-key': Invalid value of argument 'keys' of command
'send-key': expected 'ble' got '[{"type":"number","data":43},
{"type":"number","data":26},{"type":"number","data":46},
{"type":"number","data":32}]'

This improvement will not have any effect on tests executing as
expected, but it will help test development.

Any source file which calls the logging APIs now needs
to have a VIR_LOG_INIT("source.name") declaration at
the start of the file. This provides a static variable
of the virLogSource type.
Signed-off-by: NDaniel P. Berrange <berrange@redhat.com>

Most of the usage of getuid()/getgid() is in cases where we are
considering what privileges we have. As such the code should be
using the effective IDs, not real IDs.
Signed-off-by: NDaniel P. Berrange <berrange@redhat.com>

Join the worker thread no matter if it is running or zombie already.
With current implementation the thread is joined iff @running is true.
However, when worker executes the last line, @running is set to false.
Hence qemuMonitorTestFree() won't join it (and free resources) even
though we can clearly see worker has run (nobody else sets @running =
false).
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

In a few cases we might want to not care if monitor command executed on
the mocked monitor matches the one we have reply for. Sounds crazy, but
if we just want monitor to return certain values (e.g. read from a file)
there is no need to care about command match.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

Currently, when creating a new mocked monitor, the greeting can't be
chosen. This is crucial for next patches, because some info as qemu
version is obtained in the greeting message.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

The virDomainChrSourceDef variable should be memset to
0, so that the cleanup block does not free uninitialized
data on OOM.
Signed-off-by: NDaniel P. Berrange <berrange@redhat.com>

The qemuMonitorCommonTestInit method did not allocate the
test object, so it should not free it upon failure. Doing
so causes a double free with the caller.
Signed-off-by: NDaniel P. Berrange <berrange@redhat.com>

Don't leak the path string in qemuMonitorCommonTestNew if
an OOM occurs.
Signed-off-by: NDaniel P. Berrange <berrange@redhat.com>

The driver is then passed to monitor event handlers.

Some tests need the monitor to operate on an already existing VM object
rather than on a new mock-up the monitor test normally creates.

We will have to pass a mock-up of the driver when testing monitor
events.

When allocation of memory failed the error path didn't call va_end()
causing a coverity failure.

Reported by John Ferlan.

This patch adds helpers that allow to check for argument values in
commands sent to the monitor.

Use the JSON error messages to report errors back to the caller in
addition to erroring out. The error reported from the event loop from
the mock function of the monitor was later overwritten by the call to
the monitor/agent interaction API. This will also allow testing of error
reporting.

Add helper functions to open guest agent connections and a handler for
replying to the "guest-sync" command.