The Fedora CI and coverity runs are using a slightly different set of
packages.  Copy most of the content over from tests/docker while
keeping the commands at the end that unpack the tools.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20230331174844.376300-1-pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Coroutine commands have to be declared as coroutine_fn, but the
marker does not show up in the qapi-comands-* headers; likewise, the
marshaling function calls the command and therefore must be coroutine_fn.
Static analysis would want coroutine_fn to match between prototype and
declaration, because in principle coroutines might be compiled to a
completely different calling convention.  So we would like to add the
marker to the header.

Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Hexagon is split into two components because it has hundreds of false positives
in the generated files.

capstone and slirp have been removed.

hw/nvme is added to block.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

A subproject test may be simply in the "PROJECT" suite (such as
"qemu-common" with the following patches)

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20230302131848.1527460-2-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

This is a very rough and ready first pass at comparing gcovr's json
output between two different runs. At the moment it will give you a
file level diff between two runs but hopefully it wont be too hard to
extend to give better insight.

After generating the coverage results you run with something like:

  ./scripts/coverage/compare_gcov_json.py \
    -a ./builds/gcov.config1/coverage.json \
    -b ./builds/gcov.config2/coverage.json

My hope is we can use this to remove some redundancy from testing as
well as evaluate if new tests are actually providing additional
coverage or just burning our precious CI time.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Cc: Kautuk Consul <kconsul@linux.vnet.ibm.com>
Acked-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230403134920.2132362-2-alex.bennee@linaro.org>

This was broken when we moved to using the pre-built packages as we
didn't take care to ensure we used RPMs where required.

NB: I could never get this to complete on my test setup but I suspect
this was down to network connectivity and timeouts while downloading.

Fixes: 69c4befb (scripts/ci: update gitlab-runner playbook to use latest runner)
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-5-alex.bennee@linaro.org>

Without libslip enabled we won't have user networking which means the
KVM tests won't run.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-4-alex.bennee@linaro.org>

This automates ethtool tests for igb registers, interrupts, etc.

Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>

Currently when we encounter a gdb that is old or not built with
multiarch in mind we fail rather messily. Try and improve the
situation by probing ahead of time and setting
HOST_GDB_SUPPORTS_ARCH=y in the relevant tcg configs. We can then skip
and give a more meaningful message if we don't run the test.

Cc: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230302190846.2593720-24-alex.bennee@linaro.org>
Message-Id: <20230303025805.625589-24-richard.henderson@linaro.org>

This file, and a couple of uses, got left behind when the
tcg stuff was removed from tracetool.

Fixes: 126d4123 ("tracing: excise the tcg related from tracetool")
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

Since commit 262a69f4 ("osdep.h: Prohibit disabling assert()
in supported builds") we can not build QEMU with NDEBUG (or
G_DISABLE_ASSERT) defined, thus 'assert(0)' always aborts QEMU.

However some static analyzers / compilers doesn't notice NDEBUG
can't be defined and emit warnings if code is used after an
'assert(0)' call.

Apparently such compiler isn't as clever with G_DISABLE_ASSERT,
so we can silent these warnings by using g_assert_not_reached()
which is easier to read anyway.

In order to avoid these annoying warnings, add a checkpatch rule
to prohibit 'assert(0)'. Suggest using g_assert_not_reached()
instead. For example when reverting the previous patch we get:

  ERROR: use g_assert_not_reached() instead of assert(0)
  #21: FILE: target/ppc/dfp_helper.c:124:
  +            assert(0); /* cannot get here */

  ERROR: use g_assert_not_reached() instead of assert(0)
  #30: FILE: target/ppc/dfp_helper.c:141:
  +            assert(0); /* cannot get here */

  total: 2 errors, 0 warnings, 16 lines checked

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230221232520.14480-3-philmd@linaro.org>

Once upon a time, "sphinx-build" on certain RPM platforms invoked
specifically a Python 2.x version, while "sphinx-build-3" was a distro
shim for the Python 3.x version.

These days, none of our supported platforms utilize a 2.x version, and
those that still have 'sphinx-build-3' make it a symbolic link to
'sphinx-build'.  Not searching for 'sphinx-build-3' will prefer
pip/venv installed versions of sphinx if they're available.

This adds an extremely convenient ability to test document building
ability in QEMU across multiple versions of Sphinx for the purposes of
compatibility testing.

Signed-off-by: John Snow <jsnow@redhat.com>
Message-Id: <20230221012456.2607692-6-jsnow@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Commit eed56e9a "configure, meson: move config-poison.h to meson"
lost a "| sort -u".  Restore it.  config-poison shrinks from ~4500 to
~700 lines when all targets are enabled.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230206102037.3621709-1-armbru@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

With the two major JSON-ish type hierarchies clarified for distinct
purposes; QAPIExpression for parsed expressions and JSONValue for
introspection data, remove this FIXME as no longer an action item.

A third JSON-y data type, _ExprValue, is not meant to represent JSON in
the abstract but rather only the possible legal return values from a
single function, get_expr(). It isn't appropriate to attempt to merge it
with either of the above two types.

In theory, it may be possible to define a completely agnostic
one-size-fits-all JSON type hierarchy that any other user could borrow -
in practice, it's tough to wrangle the differences between invariant,
covariant and contravariant types: input and output parameters demand
different properties of such a structure.

However, QAPIExpression serves to authoritatively type user input to the
QAPI parser, while JSONValue serves to authoritatively type qapi
generator *output* to be served back to client users at runtime via
QMP. The AST for these two types are different and cannot be wholly
merged into a unified syntax.

They could, in theory, share some JSON primitive definitions. In
practice, this is currently more trouble than it's worth with mypy's
current expressive power. As such, declare this "done enough for now".

Signed-off-by: John Snow <jsnow@redhat.com>
Message-Id: <20230215000011.1725012-7-jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>

We can remove this alias as it only has two usages now, and no longer
pays for the confusion of "yet another type".

Signed-off-by: John Snow <jsnow@redhat.com>
Message-Id: <20230215000011.1725012-6-jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>

This patch creates a new type, QAPIExpression, which represents a parsed
expression complete with QAPIDoc and QAPISourceInfo.

This patch turns parser.exprs into a list of QAPIExpression instead,
and adjusts expr.py to match.

This allows the types we specify in parser.py to be "remembered" all the
way through expr.py and into schema.py. Several assertions around
packing and unpacking this data can be removed as a result.

It also corrects a harmless typing error.  Before the patch,
check_exprs() allegedly takes a List[_JSONObject].  It actually takes
a list of dicts of the form

    {'expr': E, 'info': I, 'doc': D}

where E is of type _ExprValue, I is of type QAPISourceInfo, and D is
of type QAPIDoc.  Key 'doc' is optional.  This is not a _JSONObject!
Passes type checking anyway, because _JSONObject is Dict[str, object].

Signed-off-by: John Snow <jsnow@redhat.com>
Message-Id: <20230215000011.1725012-5-jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
[Commit message amended to point out the typing fix]

Pylint under 3.6 does not believe that Collection is subscriptable at
runtime. It is, making this a Pylint
bug. https://github.com/PyCQA/pylint/issues/2377



They closed it as fixed, but that doesn't seem to be true as of Pylint
2.13.9, the latest version you can install under Python 3.6. 2.13.9 was
released 2022-05-13, about seven months after the bug was closed.

The least-annoying fix here is to just use the concret type.

Signed-off-by: John Snow <jsnow@redhat.com>
Message-Id: <20230215000011.1725012-4-jsnow@redhat.com>
[Dumbed down from Sequence[str] to List[str], commit message adjusted]
Reviewed-by: Markus Armbruster <armbru@redhat.com>

Newer versions of pylint disable the "no-self-use" message by
default. Older versions don't, though. If we leave the suppressions in,
pylint yelps about useless options. Just tell pylint to shush.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20230215000011.1725012-3-jsnow@redhat.com>

New versions of flake8 don't like same-line comments. (It's a version
newer than what fc37 ships, but it still makes my life easier to fix it
now.)

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20230215000011.1725012-2-jsnow@redhat.com>

Some functions must not be called from coroutine context. The common
pattern to use them anyway from a coroutine is running them in a BH and
letting the calling coroutine yield to be woken up when the BH is
completed.

Instead of manually writing such wrappers, add support for generating
them to block-coroutine-wrapper.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20230126172432.436111-2-kwolf@redhat.com>
Reviewed-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Reviewed-by: Hanna Czenczek <hreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

Remove all the virtiofsd build and docs infrastructure.

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>

Rmove the avocado test for virtiofsd, since we're about to remove
the C implementation.

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>

Add the missing meson infrastructure bits for the new libdw
dependency. Model them after the existing capstone knobs.

Fixes: 7c10cb38 ("accel/tcg: Add debuginfo support")
Reported-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230210005208.438142-1-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

This commit is the same with [PATCH v6 1/2], and provides avx512 support for xbzrle_encode_buffer
function to accelerate xbzrle encoding speed. Runtime check of avx512
support and benchmark for this feature are added. Compared with C
version of xbzrle_encode_buffer function, avx512 version can achieve
50%-70% performance improvement on benchmarking. In addition, if dirty
data is randomly located in 4K page, the avx512 version can achieve
almost 140% performance gain.

Signed-off-by: ling xu <ling1.xu@intel.com>
Co-authored-by: Zhou Zhao <zhou.zhao@intel.com>
Co-authored-by: Jun Jin <jun.i.jin@intel.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>

The library directory can change depending on the multilib setup of the host.
It would be even better to detect it in configure with the same algorithm
that Meson uses, but the important thing to avoid confusing developers is
to have identical contents of scripts/meson-buildoptions.sh, independent
of the distro and architecture on which it was created.

So, for now just give a custom default value to libdir.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

The script drops #include "qemu/osdep.h" from headers.  Mention it in
the commit message it uses for --git.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20230202133830.2152150-5-armbru@redhat.com>

When a symbolic link points to a file that needs cleaning, the script
replaces the link with a cleaned regular file.  Not wanted; skip them.

We have a few symbolic links under subprojects/libvduse/ and
subprojects/libvhost-user/.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20230202133830.2152150-4-armbru@redhat.com>

When running with --check-dup-head, the script always claims it "Found
duplicate header file includes."  Fix to do it only when it actually
found some.

Fixes: d66253e4 ("scripts/clean-includes: added duplicate #include check")
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20230202133830.2152150-3-armbru@redhat.com>

When clean-includes claims to skip or ignore a file, only the part
that sanitizes use of qemu/osdep.h skips the file.  The part that
looks for duplicate #include does not, and neither does committing to
Git.

The latter can get unrelated stuff included in the commit, but only if
you run clean-includes in a dirty tree, which is unwise.  Messed up
when we added skipping in commit fd3e39a4 "scripts/clean-includes:
Enhance to handle header files".

The former can cause bogus reports for --check-dup-head.  Added in
commit d66253e4 "scripts/clean-includes: added duplicate #include
check", duplicating the prior mistake.

Fix the script to fully skip files.

Fixes: fd3e39a4 ("scripts/clean-includes: Enhance to handle header files")
Fixes: d66253e4 ("scripts/clean-includes: added duplicate #include check")
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20230202133830.2152150-2-armbru@redhat.com>

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

scripts/ci/org.centos/stream/8/build-environment.yml has a slightly different
list of packages compared to scripts/ci/setup/build-environment.yaml.  Make
them the same.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Update the CI playbook so that it is able to prepare a system with a
fresh CentOS Stream 8 install, rather than just support RHEL.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Since this playbook is meant for a CentOS 8 install, no need to check
the facts.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230110132700.833690-8-marcandre.lureau@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230124180127.1881110-12-alex.bennee@linaro.org>

The up-coming upgrade to Fedora 37 will bring in libtcmalloc as a
dependency of libglusterfs which confuses our fuzz run. Rather than
disable the build lets use LSAN's suppression mechanism to prevent the
job from failing.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Cc: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230124180127.1881110-10-alex.bennee@linaro.org>

On msys2, the shader-to-C script produces bad C:
./ui/shader/texture-blit-vert.h:2:5: error: missing terminating " character [-Werror]

Fix it by changing the line ending from crlf to lf, and convert the
script to Python (qemu build seems perl-free after that).

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Acked-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230110132700.833690-2-marcandre.lureau@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230124180127.1881110-6-alex.bennee@linaro.org>

We were using quite and old runner on our machines and running into
issues with stalling jobs. Gitlab in the meantime now reliably provide
the latest packaged versions of the runner under a stable URL. This
update:

  - creates a per-arch subdir for builds
  - switches from binary tarballs to deb packages
  - re-uses the same binary for the secondary runner
  - updates distro check for second to 22.04

Note this script isn't fully idempotent as we end up accumulating
runners especially during testing. However we also want to be able to
run twice with different GitLab keys (e.g. project and personal) so I
think we just have to be mindful of that during testing.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Acked-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230124180127.1881110-2-alex.bennee@linaro.org>

Just omit the various 'return' when the return type is void.

Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20230113204212.359076-2-kwolf@redhat.com>
Reviewed-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

We have two inclusion loops:

       block/block.h
    -> block/block-global-state.h
    -> block/block-common.h
    -> block/blockjob.h
    -> block/block.h

       block/block.h
    -> block/block-io.h
    -> block/block-common.h
    -> block/blockjob.h
    -> block/block.h

I believe these go back to Emanuele's reorganization of the block API,
merged a few months ago in commit d7e2fe4a.

Fortunately, breaking them is merely a matter of deleting unnecessary
includes from headers, and adding them back in places where they are
now missing.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20221221133551.3967339-2-armbru@redhat.com>