Re: [Qemu-devel] [PATCH v2] tests: numa-test: use -numa memdev option instead of legacy 'mem' option

[Eduardo Habkost]

On Wed, Jul 03, 2019 at 11:47:23AM -0400, Igor Mammedov wrote:
> it drops testing of deprecated 'mem' option and will test preferred memdev
> option instead.
> 
> Signed-off-by: Igor Mammedov <address@hidden>
> ---
> v2:
>   - make backend size depend on initial RAM size and a number of backends
>     fixes test faulure with spapr machine (Eduardo Habkost <address@hidden>)
> 
>  tests/numa-test.c | 85 +++++++++++++++++++++++++++++------------------
>  1 file changed, 52 insertions(+), 33 deletions(-)
> 
> diff --git a/tests/numa-test.c b/tests/numa-test.c
> index 8de8581231..e102b3b70f 100644
> --- a/tests/numa-test.c
> +++ b/tests/numa-test.c
> @@ -14,9 +14,28 @@
>  #include "qapi/qmp/qdict.h"
>  #include "qapi/qmp/qlist.h"
>  
> -static char *make_cli(const char *generic_cli, const char *test_cli)
> +typedef struct {
> +    const char *args;
> +    int ram_size; /* size in MB */
> +} TestData;
> +
> +static char *make_cli(const TestData *data, int memdev_count,
> +                      const char *test_cli)
>  {
> -    return g_strdup_printf("%s %s", generic_cli ? generic_cli : "", 
> test_cli);
> +    int i;
> +    char *tmp = NULL;
> +    char *cli = g_strdup_printf("%s -m %d", data->args ? : "", 
> data->ram_size);
> +
> +    for (i = 0; memdev_count && (i < memdev_count); i++) {
> +        tmp = g_strdup_printf("%s -object 
> memory-backend-ram,id=r%d,size=%dM",
> +                              cli, i, data->ram_size / memdev_count);
> +        g_free(cli);
> +        cli = tmp;
> +    }
> +    tmp = g_strdup_printf("%s %s", cli, test_cli);
> +    g_free(cli);
> +    cli = tmp;
> +    return cli;
>  }
[...]
> +    TestData data = { .ram_size = 128 };
>  
> -    if (strcmp(arch, "aarch64") == 0) {
> -        args = "-machine virt";
> +    if (!strcmp(arch, "aarch64")) {
> +        data.args = "-machine virt";
> +    } else if (!strcmp(arch, "ppc64")) {
> +        data.ram_size = 512;

Wouldn't it be simpler to always use "-m 512M" and the same node
sizes?

-- 
Eduardo