From: Fabiano Rosas <farosas@suse.de>
To: "Daniel P. Berrangé" <berrange@redhat.com>,
"Philippe Mathieu-Daudé" <philmd@linaro.org>
Cc: qemu-devel@nongnu.org, "Juan Quintela" <quintela@redhat.com>,
"Peter Xu" <peterx@redhat.com>, "Thomas Huth" <thuth@redhat.com>,
"Laurent Vivier" <lvivier@redhat.com>,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Leonardo Bras" <leobras@redhat.com>,
"Alex Bennée" <alex.bennee@linaro.org>
Subject: Re: [RFC PATCH 1/1] qtest/migration: Support more than one QEMU binary
Date: Tue, 03 Oct 2023 13:24:30 -0300 [thread overview]
Message-ID: <874jj7u11d.fsf@suse.de> (raw)
In-Reply-To: <ZRw5Myc/joWb6why@redhat.com>
Daniel P. Berrangé <berrange@redhat.com> writes:
> On Tue, Oct 03, 2023 at 05:24:50PM +0200, Philippe Mathieu-Daudé wrote:
>> Hi Fabiano,
>>
>> [+Alex & Daniel]
>>
>> On 3/10/23 16:19, Fabiano Rosas wrote:
>> > We have strict rules around migration compatibility between different
>> > QEMU versions but not a single test that validates the migration state
>> > between different binaries.
>> >
>> > Add some infrastructure to allow running the migration tests with two
>> > different QEMU binaries as migration source and destination.
>> >
>> > The code now recognizes two new environment variables QTEST_QEMU_SRC
>> > and QTEST_QEMU_DST. Only one of the two is expected to be used along
>> > with the existing QTEST_QEMU_BINARY, which will automatically be used
>> > for the other side of migration.
>> >
>> > Usage:
>> > QTEST_QEMU_DST=./build-8.2.0/qemu-system-x86_64 \
>> > QTEST_QEMU_BINARY=../build-8.1.0/qemu-system-x86_64 \
>> > ./tests/qtest/migration-test
>>
>> I like it as a first step, but I'd rather run $QTEST_QEMU_SRC
>> directly from a docker image, i.e.:
>>
>> $ docker run -it opensuse/leap
>> # zypper update -y && zypper install -y qemu-x86
>> $ docker run opensuse/leap qemu-system-x86_64 ...
>
> In theory this does not need any support in libqtest at all
>
> $ cat myqemu.dkr
> FROM fedora:38
>
> RUN dnf -y install qemu-kvm
>
> $ podman build -f myqemu.dkr --tag myqemu .
>
> $ cat > myqemu <<EOF
> #!/bin/sh
> exec podman run --volume /tmp=/tmp --security-opt label=disable myqemu qemu-system-x86_64 "$@"
>
> $ chmod +x myqemu
>
> $ QTEST_QEMU_BINARY=./myqemu.sh ./build/tests/qtest/rtc-test
I'm favor of this. I usually set that variable to something like 'gdb
--args ...' and it works just fine.
> Except we fail on the last step, because bind mounts don't make UNIX domain
> sockets accessible. So we can see the /tmp/qtest-$PID.sock in the container,
> but it can't be used.
>
> UNIX domain sockets in the filesystem are tied to the mount namespace, and
> podman/docker inherantly creates a new mount namespace making the UNIX
> domani socket inaccessible.
>
> UNIX domain sockets in the abstract namespace, however, are tied to the
> network namespace, so if you used podman --network host, they should be
> accessible.
>
> libqtest could be changed to use abstract UNIX domain sockets on Linux
> only, and likely unlock the use of podman for QEMU.
>
>>
>> > This code also works for when debugging with GDB to pass the same
>> > binary, but different GDB options for src and dst.
>> >
>> > Signed-off-by: Fabiano Rosas <farosas@suse.de>
>> > ---
>> > tests/qtest/migration-helpers.c | 168 ++++++++++++++++++++++++++++++++
>> > tests/qtest/migration-helpers.h | 3 +
>> > tests/qtest/migration-test.c | 52 ++++++++--
>> > 3 files changed, 216 insertions(+), 7 deletions(-)
>> >
>> > diff --git a/tests/qtest/migration-helpers.c b/tests/qtest/migration-helpers.c
>> > index be00c52d00..e84360c3b3 100644
>> > --- a/tests/qtest/migration-helpers.c
>> > +++ b/tests/qtest/migration-helpers.c
>> > @@ -12,6 +12,8 @@
>> > #include "qemu/osdep.h"
>> > #include "qapi/qmp/qjson.h"
>> > +#include "qapi/qmp/qlist.h"
>> > +#include "qapi/qmp/qstring.h"
>> > #include "migration-helpers.h"
>> > @@ -180,3 +182,169 @@ void wait_for_migration_fail(QTestState *from, bool allow_active)
>> > g_assert(qdict_get_bool(rsp_return, "running"));
>> > qobject_unref(rsp_return);
>> > }
>> > +
>> > +static char *query_pkg_version(QTestState *who)
>> > +{
>> > + QDict *rsp;
>> > + char *pkg;
>> > +
>> > + rsp = qtest_qmp_assert_success_ref(who, "{ 'execute': 'query-version' }");
>> > + g_assert(rsp);
>> > +
>> > + pkg = g_strdup(qdict_get_str(rsp, "package"));
>> > + qobject_unref(rsp);
>> > +
>> > + return pkg;
>> > +}
>> > +
>> > +static QList *query_machines(void)
>> > +{
>> > + QDict *response;
>> > + QList *list;
>> > + QTestState *qts;
>> > +
>> > + qts = qtest_init("-machine none");
>> > + response = qtest_qmp(qts, "{ 'execute': 'query-machines' }");
>> > + g_assert(response);
>> > + list = qdict_get_qlist(response, "return");
>> > + g_assert(list);
>> > +
>> > + qtest_quit(qts);
>> > + return list;
>> > +}
>> > +
>> > +static char *get_default_machine(QList *list)
>> > +{
>> > + QDict *info;
>> > + QListEntry *entry;
>> > + QString *qstr;
>> > + char *name = NULL;
>> > +
>> > + QLIST_FOREACH_ENTRY(list, entry) {
>> > + info = qobject_to(QDict, qlist_entry_obj(entry));
>> > + g_assert(info);
>> > +
>> > + if (qdict_get(info, "is-default")) {
>> > + qstr = qobject_to(QString, qdict_get(info, "name"));
>> > + g_assert(qstr);
>> > + name = g_strdup(qstring_get_str(qstr));
>> > + break;
>> > + }
>> > + }
>> > +
>> > + g_assert(name);
>> > + return name;
>> > +}
>> > +
>> > +static bool search_default_machine(QList *list, const char *theirs)
>> > +{
>> > + QDict *info;
>> > + QListEntry *entry;
>> > + QString *qstr;
>> > +
>> > + if (!theirs) {
>> > + return false;
>> > + }
>> > +
>> > + QLIST_FOREACH_ENTRY(list, entry) {
>> > + info = qobject_to(QDict, qlist_entry_obj(entry));
>> > + g_assert(info);
>> > +
>> > + qstr = qobject_to(QString, qdict_get(info, "name"));
>> > + g_assert(qstr);
>> > +
>> > + if (g_str_equal(qstring_get_str(qstr), theirs)) {
>> > + return true;
>> > + }
>> > + }
>> > + return false;
>> > +}
>> > +
>> > +/*
>> > + * We need to ensure that both QEMU instances set via the QTEST_QEMU_*
>> > + * vars will use the same machine type. Use a custom query_machines
>> > + * function because the generic one in libqtest has a cache that would
>> > + * return the same machines for both binaries.
>> > + */
>> > +char *find_common_machine_type(const char *bin)
>> > +{
>> > + QList *m1, *m2;
>> > + g_autofree char *def1 = NULL;
>> > + g_autofree char *def2 = NULL;
>> > + const char *qemu_bin = getenv("QTEST_QEMU_BINARY");
>> > +
>> > + m1 = query_machines();
>> > +
>> > + g_setenv("QTEST_QEMU_BINARY", bin, true);
>> > + m2 = query_machines();
>> > + g_setenv("QTEST_QEMU_BINARY", qemu_bin, true);
>> > +
>> > + def1 = get_default_machine(m1);
>> > + def2 = get_default_machine(m2);
>> > +
>> > + if (g_str_equal(def1, def2)) {
>> > + /* either can be used */
>> > + return g_strdup(def1);
>> > + }
>> > +
>> > + if (search_default_machine(m1, def2)) {
>> > + return g_strdup(def2);
>> > + }
>> > +
>> > + if (search_default_machine(m2, def1)) {
>> > + return g_strdup(def1);
>> > + }
>> > +
>> > + g_assert_not_reached();
>> > +}
>> > +
>> > +/*
>> > + * Init a guest for migration tests using an alternate QEMU binary for
>> > + * either the source or destination, depending on @var. The other
>> > + * binary should be set as usual via QTEST_QEMU_BINARY.
>> > + *
>> > + * Expected values:
>> > + * QTEST_QEMU_SRC
>> > + * QTEST_QEMU_DST
>> > + *
>> > + * Warning: The generic parts of qtest could be using
>> > + * QTEST_QEMU_BINARY to query for properties before we reach the
>> > + * migration code. If the alternate binary is too dissimilar that
>> > + * could cause issues.
>> > + */
>> > +static QTestState *init_vm(const char *extra_args, const char *var)
>> > +{
>> > + const char *alt_bin = getenv(var);
>> > + const char *qemu_bin = getenv("QTEST_QEMU_BINARY");
>> > + g_autofree char *pkg = NULL;
>> > + bool src = !!strstr(var, "SRC");
>> > + QTestState *qts;
>> > +
>> > + if (alt_bin) {
>> > + g_setenv("QTEST_QEMU_BINARY", alt_bin, true);
>> > + }
>> > +
>> > + qts = qtest_init(extra_args);
>> > + pkg = query_pkg_version(qts);
>> > +
>> > + g_test_message("Using %s (%s) as migration %s",
>> > + alt_bin ? alt_bin : qemu_bin,
>> > + pkg,
>> > + src ? "source" : "destination");
>> > +
>> > + if (alt_bin) {
>> > + /* restore the original */
>> > + g_setenv("QTEST_QEMU_BINARY", qemu_bin, true);
>> > + }
>> > + return qts;
>> > +}
>
> IMHO, we should just create a new qtest_init_env variant, that is the
> same as qtest_init, but accepts an environment variable name to use as
> an override.
>
> eg
>
> qtest_init_env("QTEST_QEMU_BINARY_SRC", extra_args)
>
> it would look for $QTEST_QEMU_BINARY_SRC and if not found automatically
> fallback to $QTEST_QEMU_BINARY.
>
This was initially intended to be an off-tree patch that I could use to
test migration compatibility, so I avoided touching libqtest. Now that I
learned this might be of interest we can make it less hackish.
> I don't think there's any need to explicitly forbid setting both
> QTEST_QEMU_BINARY_SRC and QTEST_QEMU_BINARY_DST at the same time.
>
This is a little biased on my usage of migration-test from the command
line. Adding one of SRC|DST is easier when coming/going from a test that
already uses QTEST_QEMU_BINARY. No big deal, of course.
We'll just have to do something about qtest_get_machines(),
qtest_has_device() and qtest_get_arch(), which expect QTEST_QEMU_BINARY
to be present.
next prev parent reply other threads:[~2023-10-03 16:25 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-03 14:19 [RFC PATCH 0/1] tests/migration-test: Allow testing older machine types Fabiano Rosas
2023-10-03 14:19 ` [RFC PATCH 1/1] qtest/migration: Support more than one QEMU binary Fabiano Rosas
2023-10-03 15:24 ` Philippe Mathieu-Daudé
2023-10-03 15:54 ` Daniel P. Berrangé
2023-10-03 16:24 ` Fabiano Rosas [this message]
2023-10-04 8:45 ` Juan Quintela
2023-10-04 15:59 ` Fabiano Rosas
2023-10-04 17:56 ` Daniel P. Berrangé
2023-10-04 18:09 ` Juan Quintela
2023-10-04 18:21 ` Thomas Huth
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=874jj7u11d.fsf@suse.de \
--to=farosas@suse.de \
--cc=alex.bennee@linaro.org \
--cc=berrange@redhat.com \
--cc=leobras@redhat.com \
--cc=lvivier@redhat.com \
--cc=pbonzini@redhat.com \
--cc=peterx@redhat.com \
--cc=philmd@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=quintela@redhat.com \
--cc=thuth@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.