Skip to content

Q
qemu

openeuler / qemu

通知 10
Star 0
Fork 0
Q
qemu

体验新版 GitCode，发现更多精彩内容 >>

切换分支/标签
  1. 13 6月, 2019 1 次提交
    • K
      qapi: Add feature flags to struct types · 6a8c0b51
      由 Kevin Wolf 提交于 
      Sometimes, the behaviour of QEMU changes without a change in the QMP
syntax (usually by allowing values or operations that previously
resulted in an error). QMP clients may still need to know whether
they can rely on the changed behavior.

Let's add feature flags to the QAPI schema language, so that we can make
such changes visible with schema introspection.

An example for a schema definition using feature flags looks like this:

    { 'struct': 'TestType',
      'data': { 'number': 'int' },
      'features': [ 'allow-negative-numbers' ] }

Introspection information then looks like this:

    { "name": "TestType", "meta-type": "object",
      "members": [
          { "name": "number", "type": "int" } ],
      "features": [ "allow-negative-numbers" ] }

This patch implements feature flags only for struct types. We'll
implement them more widely as needed.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Message-Id: <20190606153803.5278-2-armbru@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
      6a8c0b51
  3. 12 6月, 2019 2 次提交
    • M
      Include qemu-common.h exactly where needed · a8d25326
      由 Markus Armbruster 提交于 
      No header includes qemu-common.h after this commit, as prescribed by
qemu-common.h's file comment.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Message-Id: <20190523143508.25387-5-armbru@redhat.com>
[Rebased with conflicts resolved automatically, except for
include/hw/arm/xlnx-zynqmp.h hw/arm/nrf51_soc.c hw/arm/msf2-soc.c
block/qcow2-refcount.c block/qcow2-cluster.c block/qcow2-cache.c
target/arm/cpu.h target/lm32/cpu.h target/m68k/cpu.h target/mips/cpu.h
target/moxie/cpu.h target/nios2/cpu.h target/openrisc/cpu.h
target/riscv/cpu.h target/tilegx/cpu.h target/tricore/cpu.h
target/unicore32/cpu.h target/xtensa/cpu.h; bsd-user/main.c and
net/tap-bsd.c fixed up]
      a8d25326
    • M
      Include qemu/module.h where needed, drop it from qemu-common.h · 0b8fa32f
      由 Markus Armbruster 提交于 
      Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Message-Id: <20190523143508.25387-4-armbru@redhat.com>
[Rebased with conflicts resolved automatically, except for
hw/usb/dev-hub.c hw/misc/exynos4210_rng.c hw/misc/bcm2835_rng.c
hw/misc/aspeed_scu.c hw/display/virtio-vga.c hw/arm/stm32f205_soc.c;
ui/cocoa.m fixed up]
      0b8fa32f
  5. 05 3月, 2019 3 次提交
    • M
      qapi: Fix array first used in a different module · 56a46895
      由 Markus Armbruster 提交于 
      We generally put implicitly defined types in whatever module triggered
their definition.  This is wrong for array types, as the included test
case demonstrates.  Let's have a closer look at it.

Type 'Status' is defined sub-sub-module.json.  Array type ['Status']
occurs in main module qapi-schema-test.json and in
include/sub-module.json.  The main module's use is first, so the array
type gets put into the main module.

The generated C headers define StatusList in qapi-types.h.  But
include/qapi-types-sub-module.h uses it without including
qapi-types.h.  Oops.

To fix that, put the array type into its element type's module.

Now StatusList gets generated into qapi-types-sub-module.h, which all
its users include.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Message-Id: <20190301154051.23317-8-armbru@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
      56a46895
    • M
      qapi: Fix code generation for sub-modules in other directories · 709395f8
      由 Markus Armbruster 提交于 
      The #include directives to pull in sub-modules use file names relative
to the main module.  Works only when all modules are in the same
directory, or the main module's output directory is in the compiler's
include path.  Use relative file names instead.

The dummy variable we generate to avoid empty .o files has an invalid
name for sub-modules in other directories.  Fix that.

Both messed up in commit 252dc310 "qapi: Generate separate .h, .c
for each module".  Escaped testing because tests/qapi-schema-test.json
doesn't cover sub-modules in other directories, only
tests/qapi-schema/include-relpath.json does, and we generate and
compile C code only for the former, not the latter.  Fold the latter
into the former.  This would have caught the mistakes fixed in this
commit.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Message-Id: <20190301154051.23317-5-armbru@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
      709395f8
    • M
      qapi: Pass file name to QAPIGen constructor instead of methods · dddee4d7
      由 Markus Armbruster 提交于 
      Not much of an improvement now, but the next commit will profit.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Message-Id: <20190301154051.23317-4-armbru@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
      dddee4d7
  7. 18 2月, 2019 4 次提交
    • M
      Revert "qapi-events: add 'if' condition to implicit event enum" · 093e3679
      由 Markus Armbruster 提交于 
      This reverts commit 7bd26349.

The commit applied the events' conditions to the members of enum
QAPIEvent.  Awkward, because it renders QAPIEvent unusable in
target-independent code as soon as we make an event target-dependent.
Reverting this has the following effects:

* ui/vnc.c can remain target independent.

* monitor_qapi_event_conf[] doesn't have to muck around with #ifdef.

* query-events again doesn't reflect conditionals.  I'm going to
  deprecate it in favor of query-qmp-schema.

Another option would be to split target-dependent parts off enum
QAPIEvent into a target-dependent enum.  Doesn't seem worthwhile right
now.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Reviewed-by: NMarc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20190214152251.2073-17-armbru@redhat.com>
      093e3679
    • M
      qapi: Generate QAPIEvent stuff into separate files · 5d75648b
      由 Markus Armbruster 提交于 
      Having to include qapi-events.h just for QAPIEvent is suboptimal, but
quite tolerable now.  It'll become problematic when we have events
conditional on the target, because then qapi-events.h won't be usable
from target-independent code anymore.  Avoid that by generating it
into separate files.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Reviewed-by: NMarc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20190214152251.2073-6-armbru@redhat.com>
      5d75648b
    • M
      qapi: Prepare for system modules other than 'builtin' · c2e196a9
      由 Markus Armbruster 提交于 
      The next commit wants to generate qapi-emit-events.{c.h}.  To enable
that, extend QAPISchemaModularCVisitor to support additional "system
modules", i.e. modules that don't correspond to a (user-defined) QAPI
schema module.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Reviewed-by: NMarc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20190214152251.2073-5-armbru@redhat.com>
      c2e196a9
    • M
      qapi: Clean up modular built-in code generation a bit · dcac6471
      由 Markus Armbruster 提交于 
      We neglect to call .visit_module() for the special module we use for
built-ins.  Harmless, but clean it up anyway.  The
tests/qapi-schema/*.out now show the built-in module as 'module None'.

Subclasses of QAPISchemaModularCVisitor need to ._add_module() this
special module to enable code generation for built-ins.  When this
hasn't been done, QAPISchemaModularCVisitor.visit_module() does
nothing for the special module.  That looks like built-ins could
accidentally be generated into the wrong module when a subclass
neglects to call ._add_module().  Can't happen, because built-ins are
all visited before any other module.  But that's non-obvious.  Switch
off code generation explicitly.

Rename QAPISchemaModularCVisitor._begin_module() to
._begin_user_module().

New QAPISchemaModularCVisitor._is_builtin_module(), for clarity.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Reviewed-by: NMarc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20190214152251.2073-4-armbru@redhat.com>
      dcac6471
  9. 24 1月, 2019 1 次提交
    • M
      qapi: Eliminate indirection through qmp_event_get_func_emit() · a9529100
      由 Markus Armbruster 提交于 
      The qapi_event_send_FOO() functions emit events like this:

    QMPEventFuncEmit emit;

    emit = qmp_event_get_func_emit();
    if (!emit) {
        return;
    }

    qmp = qmp_event_build_dict("FOO");
    [put event arguments into @qmp...]

    emit(QAPI_EVENT_FOO, qmp);

The value of qmp_event_get_func_emit() depends only on the program:

* In qemu-system-FOO, it's always monitor_qapi_event_queue.

* In tests/test-qmp-event, it's always event_test_emit.

* In all other programs, it's always null.

This is exactly the kind of dependence the linker is supposed to
resolve; we don't actually need an indirection.

Note that things would fall apart if we linked more than one QAPI
schema into a single program: each set of qapi_event_send_FOO() uses
its own event enumeration, yet they share a single emit function.
Which takes the event enumeration as an argument.  Which one if
there's more than one?

More seriously: how does this work even now?  qemu-system-FOO wants
QAPIEvent, and passes a function taking that to
qmp_event_set_func_emit().  test-qmp-event wants test_QAPIEvent, and
passes a function taking that to qmp_event_set_func_emit().

It works by type trickery, of course:

    typedef void (*QMPEventFuncEmit)(unsigned event, QDict *dict);

    void qmp_event_set_func_emit(QMPEventFuncEmit emit);

    QMPEventFuncEmit qmp_event_get_func_emit(void);

We use unsigned instead of the enumeration type.  Relies on both
enumerations boiling down to unsigned, which happens to be true for
the compilers we use.

Clean this up as follows:

* Generate qapi_event_send_FOO() that call PREFIX_qapi_event_emit()
  instead of the value of qmp_event_set_func_emit().

* Generate a prototype for PREFIX_qapi_event_emit() into
  qapi-events.h.

* PREFIX_ is empty for qapi/qapi-schema.json, and test_ for
  tests/qapi-schema/qapi-schema-test.json.  It's qga_ for
  qga/qapi-schema.json, and doc-good- for
  tests/qapi-schema/doc-good.json, but those don't define any events.

* Rename monitor_qapi_event_queue() to qapi_event_emit() instead of
  passing it to qmp_event_set_func_emit().  This takes care of
  qemu-system-FOO.

* Rename event_test_emit() to test_qapi_event_emit() instead of
  passing it to qmp_event_set_func_emit().  This takes care of
  tests/test-qmp-event.

* Add a qapi_event_emit() that does nothing to stubs/monitor.c.  This
  takes care of all other programs that link code emitting QMP events.

* Drop qmp_event_set_func_emit(), qmp_event_get_func_emit().
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Message-Id: <20181218182234.28876-3-armbru@redhat.com>
Reviewed-by: NMarc-André Lureau <marcandre.lureau@redhat.com>
[Commit message typos fixed]
      a9529100
  11. 18 12月, 2018 1 次提交
  13. 14 12月, 2018 16 次提交
  15. 29 8月, 2018 5 次提交
    • E
      qapi: Add comments to aid debugging generated introspection · 8c643361
      由 Eric Blake 提交于 
      We consciously chose in commit 1a9a507b to hide QAPI type names
from the introspection output on the wire, but added a command
line option -u to unmask the type name when doing a debug build.
The unmask option still remains useful to some other forms of
automated analysis, so it will not be removed; however, when it
is not in use, the generated .c file can be hard to read.  At
the time when we first introduced masking, the generated file
consisted only of a monolithic C string, so there was no clean
way to inject any comments.

Later, in commit 7d0f982b, we switched the generation to output
a QLit object, in part to make it easier for future addition of
conditional compilation.  In fact, commit d626b6c1 took advantage
of this by passing a tuple instead of a bare object for encoding
the output of conditionals.  By extending that tuple, we can now
interject strategic comments.

For now, type name debug aid comments are only output once per
meta-type, rather than at all uses of the number used to encode
the type within the introspection data.  But this is still a lot
more convenient than having to regenerate the file with the
unmask operation temporarily turned on - merely search the
generated file for '"NNN" =' to learn the corresponding source
name and associated definition of type NNN.

The generated qapi-introspect.c changes only with the addition
of comments, such as:

| @@ -14755,6 +15240,7 @@
|          { "name", QLIT_QSTR("[485]"), },
|          {}
|      })),
| +    /* "485" = QCryptoBlockInfoLUKSSlot */
|      QLIT_QDICT(((QLitDictEntry[]) {
|          { "members", QLIT_QLIST(((QLitObject[]) {
|              QLIT_QDICT(((QLitDictEntry[]) {
Signed-off-by: NEric Blake <eblake@redhat.com>
Message-Id: <20180827213943.33524-3-eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
[Rebased, update to qapi-code-gen.txt corrected]
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
      8c643361
    • E
      qapi: Minor introspect.py cleanups · 1aa806cc
      由 Eric Blake 提交于 
      Commit 7d0f982b changed generated introspection output to no longer
produce long lines in the generated .c file, but failed to adjust
comments to match.  Add some clarity that the shorter length that
matters most is the overall QMP response on the wire.

Commit 25b1ef31 triggers a pep8 formatting nit.
Signed-off-by: NEric Blake <eblake@redhat.com>
Message-Id: <20180827213943.33524-2-eblake@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
      1aa806cc
    • M
      qapi: Emit a blank line before dummy declaration · 5f1450f5
      由 Markus Armbruster 提交于 
      We emit a dummy variable in each .c file "to shut up OSX toolchain
warnings about empty .o files" (commit 252dc310).  Separate it from
the code preceding it (if any) with a blank line.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Message-Id: <20180828120736.32323-2-armbru@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
      5f1450f5
    • P
      qapi: Drop qapi_event_send_FOO()'s Error ** argument · 3ab72385
      由 Peter Xu 提交于 
      The generated qapi_event_send_FOO() take an Error ** argument.  They
can't actually fail, because all they do with the argument is passing it
to functions that can't fail: the QObject output visitor, and the
@qmp_emit callback, which is either monitor_qapi_event_queue() or
event_test_emit().

Drop the argument, and pass &error_abort to the QObject output visitor
and @qmp_emit instead.
Suggested-by: NEric Blake <eblake@redhat.com>
Suggested-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NPeter Xu <peterx@redhat.com>
Message-Id: <20180815133747.25032-4-peterx@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
[Commit message rewritten, update to qapi-code-gen.txt corrected]
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
      3ab72385
    • M
      qapi: Fix build_params() for empty parameter list · bdd2d42b
      由 Markus Armbruster 提交于 
      build_params() returns '' instead of 'void' when there are no
parameters.  Can't happen now, but the next commit will change that.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
[peterx: compose the patch from email replies]
Signed-off-by: NPeter Xu <peterx@redhat.com>
Message-Id: <20180815133747.25032-3-peterx@redhat.com>
      bdd2d42b
  17. 15 8月, 2018 1 次提交
    • M
      qapi: Fix some pycodestyle-3 complaints · b736e25a
      由 Markus Armbruster 提交于 
      Fix the following issues:

    common.py:873:13: E129 visually indented line with same indent as next logical line
    common.py:1766:5: E741 ambiguous variable name 'l'
    common.py:1784:1: E305 expected 2 blank lines after class or function definition, found 1
    common.py:1833:1: E305 expected 2 blank lines after class or function definition, found 1
    common.py:1843:1: E305 expected 2 blank lines after class or function definition, found 1
    visit.py:181:18: E127 continuation line over-indented for visual indent
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Reviewed-by: NPhilippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20180621083551.775-1-armbru@redhat.com>
[Fixup squashed in:]
Message-ID: <871sd0nzw9.fsf@dusky.pond.sub.org>
Reviewed-by: NDaniel P. Berrangé <berrange@redhat.com>
      b736e25a
  19. 23 7月, 2018 1 次提交
  21. 16 7月, 2018 1 次提交
    • M
      qapi: Do not expose "allow-preconfig" in query-qmp-schema · 1f214ee1
      由 Markus Armbruster 提交于 
      According to commit 047f7038, option --preconfig

    [...] allows pausing QEMU in the new RUN_STATE_PRECONFIG state,
    allowing the configuration of QEMU from QMP before the machine
    jumps into board initialization code of machine_run_board_init()

    The intent is to allow management to query machine state and
    additionally configure it using previous query results within one
    QEMU instance (i.e. eliminate the need to start QEMU twice, 1st to
    query board specific parameters and 2nd for actual VM start using
    query results for additional parameters).

The implementation is a bit of a hack: it splices in an additional
main loop before machine creation, in special runstate preconfig.  New
command exit-preconfig exits that main loop.  QEMU continues
initializing, creates the machine, and runs the good old main loop.
The replacement of the main loop is transparent to monitors.

Sadly, some commands expect initialization to be complete.  Running
them in --preconfig's main loop violates their preconditions.  Since
we don't really know which commands are safe, we use a whitelist.
This drags the concept of run state into the QMP core.

The whitelist is done as a command flag in the QAPI schema (commit
d6fe3d02).  Drags the concept of run state further into the QAPI
language.

The command flag is exposed in query-qmp-schema (also commit
d6fe3d02).  This makes it ABI.

I consider the whole thing an offensively ugly hack, but sometimes an
ugly hack is the best we can do to solve a problem people have.

The need described by the commit message quote above is genuine.  The
proper solution would be a main loop that permits complete
configuration via QMP.  This is out of reach, thus the hack.

However, even though the need is genuine, it isn't urgent: libvirt is
not going to use this anytime soon.  Baking a hack into ABI before it
has any users is a bad idea.

This commit reverts the parts of commit d6fe3d02 that affect ABI
via query-qmp-schema.  The commit did the following:

(1) Add command flag 'allow-preconfig' to the QAPI schema language

(2) Pass it to code generators

(3) Have the commands.py code generator pass it to the command
    registry (so commit 047f7038 can use it as whitelist)

(4) Add 'allow-preconfig' to SchemaInfoCommand (neglecting to update
    qapi-code-gen.txt section "Client JSON Protocol introspection")

(5) Set 'allow-preconfig': true for commands qmp_capabilities,
    query-commands, query-command-line-options, query-status

Revert exactly (4), plus a bit of documentation added to
qemu-tech.info in commit 047f7038.

Shrinks query-qmp-schema's output from 126.5KiB to 121.8KiB for me.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Message-Id: <20180705091402.26244-2-armbru@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Acked-by: NEduardo Habkost <ehabkost@redhat.com>
Acked-by: NIgor Mammedov <imammedo@redhat.com>
[Straightforward conflict with commit d626b6c1 resolved]
      1f214ee1
  23. 04 7月, 2018 4 次提交