On 20.05.20 14:20, Boris Fiuczynski wrote:
> On 5/18/20 2:56 PM, Jiri Denemark wrote:
>> Signed-off-by: Jiri Denemark <jdenemar@xxxxxxxxxx>
>> ---
>> src/util/virhostcpu.c | 16 +++++++++++++++-
>> .../linux-s390x-with-frequency.signature | 1 +
>> 2 files changed, 16 insertions(+), 1 deletion(-)
>> create mode 100644 tests/virhostcpudata/linux-s390x-with-frequency.signature
>>
>> diff --git a/src/util/virhostcpu.c b/src/util/virhostcpu.c
>> index 0ce895cb39..0caf7959ef 100644
>> --- a/src/util/virhostcpu.c
>> +++ b/src/util/virhostcpu.c
>> @@ -1430,8 +1430,9 @@ virHostCPUReadSignature(virArch arch,
>> g_autofree char *model = NULL;
>> g_autofree char *stepping = NULL;
>> g_autofree char *revision = NULL;
>> + g_autofree char *proc = NULL;
>> - if (!ARCH_IS_X86(arch) && !ARCH_IS_PPC64(arch))
>> + if (!ARCH_IS_X86(arch) && !ARCH_IS_PPC64(arch) && !ARCH_IS_S390(arch))
>> return 0;
>> while (fgets(line, lineLen, cpuinfo)) {
>> @@ -1479,6 +1480,19 @@ virHostCPUReadSignature(virArch arch,
>> *signature = g_strdup_printf("%s, rev %s", name, revision);
>> return 0;
>> }
>> + } else if (ARCH_IS_S390(arch)) {
>> + if (STREQ(parts[0], "vendor_id")) {
>> + if (!vendor)
>> + vendor = g_steal_pointer(&parts[1]);
>> + } else if (STREQ(parts[0], "processor 0")) {
>> + if (!proc)
>> + proc = g_steal_pointer(&parts[1]);
>> + }
>
> To catch scenarios of moving from LPAR to nested kvm or the other way around facilties must be added here as well.
>
Yes. This would also cover the case when a firmware update adds features.
