From: "wangyanan (Y)" <wangyanan55@huawei.com>
To: "Philippe Mathieu-Daudé" <philmd@redhat.com>, qemu-devel@nongnu.org
Cc: Andrew Jones <drjones@redhat.com>, Eduardo Habkost <ehabkost@redhat.com>
Subject: Re: [PATCH v4 2/2] tests/unit: Add an unit test for smp parsing
Date: Thu, 11 Nov 2021 17:31:33 +0800 [thread overview]
Message-ID: <14250bbd-c3fb-9afc-f08d-587326f0382c@huawei.com> (raw)
In-Reply-To: <fd32e519-f970-d039-8291-897363b20add@redhat.com>
On 2021/11/11 17:14, Philippe Mathieu-Daudé wrote:
> On 10/28/21 17:09, Philippe Mathieu-Daudé wrote:
>> From: Yanan Wang <wangyanan55@huawei.com>
>>
>> Now that we have a generic parser smp_parse(), let's add an unit
>> test for the code. All possible valid/invalid SMP configurations
>> that the user can specify are covered.
>>
>> Signed-off-by: Yanan Wang <wangyanan55@huawei.com>
>> Reviewed-by: Andrew Jones <drjones@redhat.com>
>> Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
>> Message-Id: <20211026034659.22040-3-wangyanan55@huawei.com>
>> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
>> ---
>> tests/unit/test-smp-parse.c | 594 ++++++++++++++++++++++++++++++++++++
>> MAINTAINERS | 1 +
>> tests/unit/meson.build | 1 +
>> 3 files changed, 596 insertions(+)
>> create mode 100644 tests/unit/test-smp-parse.c
>> +static struct SMPTestData data_generic_valid[] = {
>> + {
>> + /* config: no configuration provided
>> + * expect: cpus=1,sockets=1,cores=1,threads=1,maxcpus=1 */
> [1]
>
>> + .config = SMP_CONFIG_GENERIC(F, 0, F, 0, F, 0, F, 0, F, 0),
>> + .expect_prefer_sockets = CPU_TOPOLOGY_GENERIC(1, 1, 1, 1, 1),
>> + .expect_prefer_cores = CPU_TOPOLOGY_GENERIC(1, 1, 1, 1, 1),
>> + }, {
>> +static void test_generic(void)
>> +{
>> + Object *obj = object_new(TYPE_MACHINE);
>> + MachineState *ms = MACHINE(obj);
>> + MachineClass *mc = MACHINE_GET_CLASS(obj);
> Watch out, while you create a machine instance in each
> test, there is only one machine class registered (see
> type_register_static(&smp_machine_info) below in [2]),
> ...
Yes, I noticed this. So on the top of each sub-test function, the properties
of the single machine class is re-initialized by smp_machine_class_init(mc).
See [*] below.
>> + SMPTestData *data = &(SMPTestData){0};
>> + int i;
>> +
>> + smp_machine_class_init(mc);
[*]
>> +
>> + for (i = 0; i < ARRAY_SIZE(data_generic_valid); i++) {
>> + *data = data_generic_valid[i];
>> + unsupported_params_init(mc, data);
>> +
>> + smp_parse_test(ms, data, true);
>> +
>> + /* Unsupported parameters can be provided with their values as 1 */
>> + data->config.has_dies = true;
>> + data->config.dies = 1;
>> + smp_parse_test(ms, data, true);
>> + }
>> +
>> + /* Reset the supported min CPUs and max CPUs */
>> + mc->min_cpus = 2;
>> + mc->max_cpus = 511;
> ... and here you are modifying the single machine class state, ...
>
>> +
>> + for (i = 0; i < ARRAY_SIZE(data_generic_invalid); i++) {
>> + *data = data_generic_invalid[i];
>> + unsupported_params_init(mc, data);
>> +
>> + smp_parse_test(ms, data, false);
>> + }
>> +
>> + object_unref(obj);
>> +}
>> +
>> +static void test_with_dies(void)
>> +{
>> + Object *obj = object_new(TYPE_MACHINE);
>> + MachineState *ms = MACHINE(obj);
>> + MachineClass *mc = MACHINE_GET_CLASS(obj);
> ... so here the machine class state is inconsistent, ...
>
>> + SMPTestData *data = &(SMPTestData){0};
>> + unsigned int num_dies = 2;
>> + int i;
>> +
>> + smp_machine_class_init(mc);
And here [*].
>> + mc->smp_props.dies_supported = true;
>> +
>> + for (i = 0; i < ARRAY_SIZE(data_generic_valid); i++) {
>> + *data = data_generic_valid[i];
>> + unsupported_params_init(mc, data);
>> +
>> + /* when dies parameter is omitted, it will be set as 1 */
>> + data->expect_prefer_sockets.dies = 1;
>> + data->expect_prefer_cores.dies = 1;
>> +
>> + smp_parse_test(ms, data, true);
> ... in particular the first test [1] is tested with mc->min_cpus = 2.
>
> I wonder why you are not getting:
>
> Output error report: Invalid SMP CPUs 1. The min CPUs supported by
> machine '(null)' is 2
>
> for [1].
So as I have explained above, we won't get an output error report like
this here. :)
Thanks,
Yanan
>> +
>> + /* when dies parameter is specified */
>> + data->config.has_dies = true;
>> + data->config.dies = num_dies;
>> + if (data->config.has_cpus) {
>> + data->config.cpus *= num_dies;
>> + }
>> + if (data->config.has_maxcpus) {
>> + data->config.maxcpus *= num_dies;
>> + }
>> +
>> + data->expect_prefer_sockets.dies = num_dies;
>> + data->expect_prefer_sockets.cpus *= num_dies;
>> + data->expect_prefer_sockets.max_cpus *= num_dies;
>> + data->expect_prefer_cores.dies = num_dies;
>> + data->expect_prefer_cores.cpus *= num_dies;
>> + data->expect_prefer_cores.max_cpus *= num_dies;
>> +
>> + smp_parse_test(ms, data, true);
>> + }
>> +
>> + for (i = 0; i < ARRAY_SIZE(data_with_dies_invalid); i++) {
>> + *data = data_with_dies_invalid[i];
>> + unsupported_params_init(mc, data);
>> +
>> + smp_parse_test(ms, data, false);
>> + }
>> +
>> + object_unref(obj);
>> +}
>> +
>> +int main(int argc, char *argv[])
>> +{
>> + g_test_init(&argc, &argv, NULL);
>> +
>> + module_call_init(MODULE_INIT_QOM);
>> + type_register_static(&smp_machine_info);
> [2]
>
> .
next prev parent reply other threads:[~2021-11-11 9:33 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-28 15:09 [PATCH v4 0/2] hw/core/machine: Add an unit test for smp_parse Philippe Mathieu-Daudé
2021-10-28 15:09 ` [PATCH v4 1/2] hw/core/machine: Split out the smp parsing code Philippe Mathieu-Daudé
2021-10-28 15:09 ` [PATCH v4 2/2] tests/unit: Add an unit test for smp parsing Philippe Mathieu-Daudé
2021-10-31 12:02 ` wangyanan (Y)
2021-11-01 10:43 ` Philippe Mathieu-Daudé
2021-11-09 9:36 ` Philippe Mathieu-Daudé
2021-11-09 12:10 ` wangyanan (Y)
2021-11-09 12:18 ` wangyanan (Y)
2021-11-10 8:10 ` Thomas Huth
2021-11-11 9:14 ` Philippe Mathieu-Daudé
2021-11-11 9:31 ` wangyanan (Y) [this message]
2021-11-11 9:37 ` Philippe Mathieu-Daudé
2021-11-11 13:00 ` wangyanan (Y)
2021-10-29 17:10 ` [PATCH v4 0/2] hw/core/machine: Add an unit test for smp_parse Eduardo Habkost
2021-11-01 18:28 ` Philippe Mathieu-Daudé
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=14250bbd-c3fb-9afc-f08d-587326f0382c@huawei.com \
--to=wangyanan55@huawei.com \
--cc=drjones@redhat.com \
--cc=ehabkost@redhat.com \
--cc=philmd@redhat.com \
--cc=qemu-devel@nongnu.org \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).