Demonstrate that the qapi generator silently parses confusing
types, which may cause other errors later on. Later patches
will update the expected results as the generator is made stricter.

Most of the new tests focus on blatant errors.  But
returns-whitelist is a case where we have historically allowed
returning something other than a JSON object from particular
commands; we have to keep that behavior to avoid breaking clients,
but it would be nicer to avoid adding such commands in the future,
because any return that is not an (array of) object cannot be
easily extended if future qemu wants to return additional
information.  The QMP protocol already documents that clients
should ignore unknown dictionary keys, but does not require
clients to have to handle more than one type of JSON object.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

For a few QMP commands, we are forced to pass an arbitrary type
without tracking it properly in QAPI.  Among the existing clients,
this unnamed type was spelled 'dict', 'visitor', and '**'; this
patch standardizes on '**', matching the documentation changes
earlier in the series.

Meanwhile, for the 'gen' key, we have been ignoring the value,
although the schema consistently used "'no'" ('success-response'
was hard-coded to checking for 'no').  But now that we can support
a literal "false" in the schema, we might as well use that rather
than ignoring the value or special-casing a random string.  Note
that these are one-way switches (use of 'gen':true is not the same
as omitting 'gen'). Also, the use of '**' requires 'gen':false,
but the use of 'gen':false does not mandate the use of '**'.

There is no difference to the generated code.  Add some tests on
what we'd like to guarantee, although it will take later patches
to clean up test results and actually enforce the use of a bool
parameter.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

In the near term, we will use it for a sensible-looking
'gen':false inside command declarations, instead of the
current ugly 'gen':'no'.

In the long term, it will allow conversion from shorthand
with defaults mentioned only in side-band documentation:
 'data':{'*flag':'bool', '*string':'str'}
into an explicit default value documentation, as in:
 'data':{'flag':{'type':'bool', 'optional':true, 'default':true},
         'string':{'type':'str', 'optional':true, 'default':null}}

We still don't parse integer values (also necessary before
we can allow explicit defaults), but that can come in a later
series.

Update the testsuite to match an improved error message.
Signed-off-by: NFam Zheng <famz@redhat.com>
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

The previous commit demonstrated that the generator overlooked
duplicate expressions:
- a complex type or command reusing a built-in type name
- redeclaration of a type name, whether by the same or different
metatype
- redeclaration of a command or event
- collision of a type with implicit 'Kind' enum for a union
- collision with an implicit MAX enum constant

Since the c_type() function in the generator treats all names
as being in the same namespace, this patch adds a global array
to track all known names and their source, to prevent collisions
before it can cause further problems.  While valid .json files
won't trigger any of these cases, we might as well be nicer to
developers that make a typo while trying to add new QAPI code.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

Demonstrate that the qapi generator doesn't deal very well with
redefined expressions.  At the parse level, they are silently
accepted; and while the testsuite just stops at parsing, I've
further tested that many of them cause generator crashes or
invalid C code if they were appended to qapi-schema-test.json.
A later patch will tighten things up and adjust the testsuite
to match.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

The previous commit demonstrated that the generator overlooked some
fairly basic broken expressions:
- missing metataype
- metatype key has a non-string value
- unknown key in relation to the metatype
- conflicting metatype (this patch treats the second metatype as an
unknown key of the first key visited, which is not necessarily the
first key the user typed)

Add check_keys to cover these situations, and update testcases to
match.  A couple other tests (enum-missing-data, indented-expr) had
to change since the validation added here occurs so early.
Conversely, changes to ident-with-escape results show that we still
have problems where our handling of escape sequences differs from
true JSON, which will matter down the road if we allow arbitrary
default string values for optional parameters (but for now is not
too bad, as we currently can avoid unicode escaping as we don't
need to represent anything beyond C identifier material).

While valid .json files won't trigger any of these cases, we might
as well be nicer to developers that make a typo while trying to add
new QAPI code.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

Demonstrate that the qapi generator doesn't deal well with
expressions that aren't up to par. Later patches will improve
the expected results as the generator is made stricter.  Only
a few of the the added tests actually behave sanely at
rejecting obvious problems or demonstrating success.

Note that in some cases, we reject bad QAPI merely because our
pseudo-JSON parser does not yet know how to parse numbers.  This
series does not address that, but when a later series adds support
for numeric defaults of integer fields, the testsuite will ensure
that we don't lose the error (and hopefully that the error
message quality is improved).
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

Previous patches have led up to the point where I create the
new meta-type "'alternate':'Foo'".  See the previous patches
for documentation; I intentionally split as much work into
earlier patches to minimize the size of this patch, but a lot
of it is churn due to testsuite fallout after updating to the
new type.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

The next patch will quit special-casing "'union':'Foo',
'discriminator':{}" and instead use "'alternate':'Foo'".

Separating docs from implementation makes it easier to focus
on wording without holding up code.  In particular, making
alternate a separate type makes for a nice type hierarchy:

          /-------- meta-type ------\
         /              |            \
    simple types    alternate     complex types
    |         |                   |           |
 built-in   enum             type(struct)   union
 |       \    /                            /    \
numeric  string                         simple  flat

A later patch will then clean up 'type' vs. 'struct'
confusion.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

Reduce churn in the future patch that replaces anonymous unions
with a new metatype 'alternate' by changing 'AnonUnion' to
'Alternate'.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

Special-casing 'discriminator == {}' for handling anonymous unions
is getting awkward; since this particular type is not always a
dictionary on the wire, it is easier to treat it as a completely
different class of type, "alternate", so that if a type is listed
in the union_types array, we know it is not an anonymous union.

This patch just further segregates union handling, to make sure that
anonymous unions are not stored in union_types, and splitting up
check_union() into separate functions.  A future patch will change
the qapi grammar, and having the segregation already in place will
make it easier to deal with the distinct meta-type.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

This patch widens the scope of a try block (with the attending
reindentation required by Python) in preparation for a future
patch adding more instances of QAPIExprError inside the block.
It's easier to separate indentation from semantic changes, so
this patch has no real behavior change.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

Previous commits demonstrated that the generator had several
flaws with less-than-perfect unions:
- a simple union that listed the same branch twice (or two variant
names that map to the same C enumerator, including the implicit
MAX sentinel) ended up generating invalid C code
- an anonymous union that listed two branches with the same qtype
ended up generating invalid C code
- the generator crashed on anonymous union attempts to use an
array type
- the generator was silently ignoring a base type for anonymous
unions
- the generator allowed unknown types or nested anonymous unions
as a branch in an anonymous union
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

None of the existing QMP or QGA interfaces uses a union with a
base type but no discriminator; it is easier to avoid this in the
generator to save room for other future extensions more likely to
be useful.  An earlier commit added a union-base-no-discriminator
test to ensure that we eventually give a decent error message;
likewise, removing UserDefUnion outright is okay, because we moved
all the tests we wish to keep into the tests of the simple union
UserDefNativeListUnion in the previous commit.  Now is the time to
actually forbid simple union with base, and remove the last
vestiges from the testsuite.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

The tests of UserDefNativeListUnion serve to validate code
generation of simple unions without a base type, except that it
did not have full coverage in the strict test.  The next commits
will remove tests and support for simple unions with a base type,
so there is no real loss at repurposing that test here as
opposed to churn of adding a new test then deleting the old one.

Fix some indentation and long lines while at it.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

Demonstrate that the qapi generator doesn't deal well with unions
that aren't up to par. Later patches will update the expected
reseults as the generator is made stricter.  A few tests work
as planned, but most show poor or missing error messages.

Of particular note, qapi-code-gen.txt documents 'base' only for
flat unions, but the tests here demonstrate that we currently allow
a 'base' to a simple union, although it is exercised only in the
testsuite.  Later patches will remove this undocumented feature, to
give us more flexibility in adding other future extensions to union
types.  For example, one possible extension is the idea of a
type-safe simple enum, where added fields tie the discriminator to
a user-defined enum type rather than creating an implicit enum from
the names in 'data'.  But adding such safety on top of a simple
enum with a base type could look ambiguous with a flat enum;
besides, the documentation also mentions how any simple union can
be represented by an equivalent flat union.  So it will be simpler
to just outlaw support for something we aren't using.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

The previous commit demonstrated that the generator had several
flaws with less-than-perfect enums:
- an enum that listed the same string twice (or two variant
strings that map to the same C enumerator) ended up generating
an invalid C enum
- because the generator adds a _MAX terminator to each enum,
the use of an enum member 'max' can also cause this clash
- if an enum omits 'data', the generator left a python stack
trace rather than a graceful message
- an enum that used a non-array 'data' was silently accepted by
the parser
- an enum that used non-string members in the 'data' member
was silently accepted by the parser

Add check_enum to cover these situations, and update testcases
to match.  While valid .json files won't trigger any of these
cases, we might as well be nicer to developers that make a typo
while trying to add new QAPI code.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

Demonstrate that the qapi generator doesn't deal well with enums
that aren't up to par. Later patches will update the expected
results as the generator is made stricter.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

Python 2 and Python 3 have a wild history of whether strings
default to ascii or unicode, where Python 3 requires checking
isinstance(foo, basestr) to cover all strings, but where that
code is not portable to Python 2.  It's simpler to just state
that we don't care about Unicode strings, and to just always
use the simpler isinstance(foo, str) everywhere.

I'm no python expert, so I'm basing it on this conversation:
https://lists.gnu.org/archive/html/qemu-devel/2014-09/msg05278.htmlSigned-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

We were missing the 'size' builtin type (which means that QAPI using
[ 'size' ] would fail to compile).
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

There was some redundancy between builtin_types[] and
builtin_type_qtypes{}.  Merge them into one.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

Go into more details about the various types of valid expressions
in a qapi schema, including tweaks to document fixes being done
later in the current patch series.  Also fix some stale and missing
documentation in the QMP specification.
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

While our top-level COPYING with its GPLv2+ license applies to
any documentation file that omits explicit instructions, these
days it's better to be a good example of calling out our
intentions.  Correct use of GPL requires the use of a copyright
statement, so I'm adding notice to two QAPI documents, by
attributing these files to the initial authors and major
contributors.  I used:

$ git blame --line-porcelain $file \
  | sed -n 's/^author //p' | sort | uniq -c | sort -rn

to determine authorship of these two files.  qmp-spec.txt blames
entirely to Red Hat (easy, since my contribution falls in that
category); while qapi-code-gen.txt has multiple contributors
representing multiple entities.  But since it was originally
supplied by Michael Roth, the notice I added there copies the
notice he has used in other files.  As there is no intended
change in license from the implicit one previously present from
the top level, I have not bothered to CC other contributors;
if we want to weaken things to something looser (such as LGPL)
so that there is no question that someone re-implementing the
spec is not forced to use GPL, that would be a different commit.

CC: Michael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

sdl2: add opengl support

# gpg: Signature made Tue May  5 10:36:25 2015 BST using RSA key ID D3E87138
# 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>"

* remotes/kraxel/tags/pull-sdl-20150505-1:
  sdl2: Fix RGB555
  sdl2: add support for display rendering using opengl.
  sdl2: move SDL_* includes to sdl2.h
  console-gl: add opengl rendering helper functions
  opengl: add shader helper functions.
  opengl: add shader build infrastructure
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

coverity: fix address_space_rw model

# gpg: Signature made Tue May  5 09:44:26 2015 BST using RSA key ID EB918653
# gpg: Good signature from "Markus Armbruster <armbru@redhat.com>"
# gpg:                 aka "Markus Armbruster <armbru@pond.sub.org>"

* remotes/armbru/tags/pull-cov-model-2015-05-05:
  coverity: fix address_space_rw model
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

Reproducable with:

$ x86_64-softmmu/qemu-system-x86_64 \
    -kernel $vmlinuz_of_your_choice \
    -append vga=0x313 -sdl
Signed-off-by: NMax Reitz <mreitz@redhat.com>
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Add new sdl2-gl.c file, with display
rendering functions using opengl.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NMax Reitz <mreitz@redhat.com>

Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NMax Reitz <mreitz@redhat.com>

Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NMax Reitz <mreitz@redhat.com>

Helper functions to compile, link and run opengl shader programs.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

If the is_write argument is true, address_space_rw writes to memory
and thus reads from the buffer.  The opposite holds if is_write is
false.  Fix the model.

Cc: Markus Armbruster <armbru@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

perl script to transform shader programs into c include files with
static string constands containing the shader programs, so we can
easily embed them into qemu.  Also some Makefile logic for them.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NMax Reitz <mreitz@redhat.com>

Block patches

# gpg: Signature made Thu Apr 30 19:51:16 2015 BST using RSA key ID C88F2FD6
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>"

* remotes/kevin/tags/for-upstream:
  Enable NVMe start controller for Windows guest.
  MAINTAINERS: Add qemu-block list where missing
  MAINTAINERS: make block layer core Kevin Wolf's responsibility
  MAINTAINERS: make image fuzzer Stefan Hajnoczi's responsibility
  MAINTAINERS: make block I/O path Stefan Hajnoczi's responsibility
  MAINTAINERS: split out image formats
  MAINTAINERS: make virtio-blk Stefan Hajnoczi's responsibility
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

trivial patches for 2015-04-30

# gpg: Signature made Thu Apr 30 14:07:50 2015 BST using RSA key ID A4C3D7DB
# gpg: Good signature from "Michael Tokarev <mjt@tls.msk.ru>"
# gpg:                 aka "Michael Tokarev <mjt@corpit.ru>"
# gpg:                 aka "Michael Tokarev <mjt@debian.org>"

* remotes/mjt/tags/pull-trivial-patches-2015-04-30: (42 commits)
  openrisc: cpu: Remove unused cpu_get_pc
  microblaze: fix memory leak
  tcg: Delete unused cpu_pc_from_tb()
  kvm: Silence warning from valgrind
  vhost-user: remove superfluous '\n' around error_report()
  target-mips: fix memory leak
  qmp-commands: Fix typo
  linux-user/elfload: use QTAILQ_FOREACH instead of open-coding it
  coroutine: remove unnecessary parentheses in qemu_co_queue_empty
  qemu-char: remove unused list node from FDCharDriver
  input: remove unused mouse_handlers list
  cpus: use first_cpu macro instead of QTAILQ_FIRST(&cpus)
  microblaze: cpu: delete unused cpu_interrupts_enabled
  microblaze: cpu: Renumber EXCP_* constants to close gap
  microblaze: cpu: Delete EXCP_NMI
  microblaze: cpu: Remove unused CC_OP enum
  microblaze: cpu: Remote unused cpu_get_pc
  microblaze: mmu: Delete flip_um fn prototype
  defconfigs: Piggyback microblazeel on microblaze
  libcacard: do not use full paths for include files in the same dir
  ...
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

Windows seems to send two separate calls to NVMe controller configuration. The
first sends configuration info and the second the enable bit. I couldn't
enable the Windows 8.1 in-box NVMe driver with base Qemu. I made the
following change to store the configuration data and then handle enable and
NVMe driver works on Windows 8.1.

I am not a Windows expert and I'm not entirely sure this is the correct
approach. I'm offering it for anyone who wishes to use NVMe on Windows 8.1
using Qemu.

I have tested this change with Linux and Windows guests with NVMe devices.
Signed-off-by: NDaniel Stekloff <dan@wendan.org>
Acked-by: NKeith Busch <keith.busch@intel.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

First pile of s390x patches for 2.4, including:
- some cleanup patches
- sort most of the s390x devices into categories
- support for the new STSI post handler, used to insert vm name and
  friends
- support for the new MEM_OP ioctl (including access register mode)
  for accessing guest memory

# gpg: Signature made Thu Apr 30 12:56:58 2015 BST using RSA key ID C6F02FAF
# gpg: Good signature from "Cornelia Huck <huckc@linux.vnet.ibm.com>"
# gpg:                 aka "Cornelia Huck <cornelia.huck@de.ibm.com>"

* remotes/cohuck/tags/s390x-20150430:
  kvm: better advice for failed s390x startup
  s390x/kvm: Support access register mode for KVM_S390_MEM_OP ioctl
  s390x/mmu: Use ioctl for reading and writing from/to guest memory
  s390x/kvm: Put vm name, extended name and UUID into STSI322 SYSIB
  linux-headers: update
  s390x/mmu: Use access type definitions instead of magic values
  s390x/ipl: sort into categories
  sclp: sort into categories
  s390-virtio: sort into categories
  virtio-ccw: sort into categories
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

Signed-off-by: NKevin Wolf <kwolf@redhat.com>

Kevin is now sole maintainer of the core block layer, including
BlockDriverState graphs and monitor commands.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

The block I/O path includes the asynchronous I/O machinery and
read/write/flush/discard processing.  It somewhat arbitrarily also
includes block migration, which I've found myself reviewing patches for
over the years.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>