On 3/2/21 6:59 AM, Thomas Huth wrote:
> On 01/03/2021 19.28, Claudio Imbrenda wrote:
>> TCG is known to fail these tests, so add an explicit exception to skip them.
>>
>> Once TCG has been fixed, it will be enough to revert this patch.
>>
>> Signed-off-by: Claudio Imbrenda <imbrenda@xxxxxxxxxxxxx>
>> ---
>> s390x/mvpg.c | 31 +++++++++++++++++++------------
>> 1 file changed, 19 insertions(+), 12 deletions(-)
>>
>> diff --git a/s390x/mvpg.c b/s390x/mvpg.c
>> index 792052ad..148095e0 100644
>> --- a/s390x/mvpg.c
>> +++ b/s390x/mvpg.c
>> @@ -20,6 +20,7 @@
>> #include <smp.h>
>> #include <alloc_page.h>
>> #include <bitops.h>
>> +#include <vm.h>
>>
>> /* Used to build the appropriate test values for register 0 */
>> #define KFC(x) ((x) << 10)
>> @@ -224,20 +225,26 @@ static void test_mmu_prot(void)
>> report(clear_pgm_int() == PGM_INT_CODE_PROTECTION, "destination read only");
>> fresh += PAGE_SIZE;
>>
>> - protect_page(fresh, PAGE_ENTRY_I);
>> - cc = mvpg(CCO, fresh, source);
>> - report(cc == 1, "destination invalid");
>> - fresh += PAGE_SIZE;
>> + if (vm_is_tcg()) {
>> + report_skip("destination invalid");
>> + report_skip("source invalid");
>> + report_skip("source and destination invalid");
>
> You could also use report_xfail(vm_is_tcg(), ...) instead. That shows that
> there are still problems without failing CI runs.
If I remember correctly we fail with a PGM so we would also need to add
a expect_pgm_int() call before each test when running under tcg
therefore it's not just a 1:1 replacement in this case.
But yes, I'd also like an indication why we're skipping. A comment and a
TCG prefix for skips should be enough.
>
> Anyway:
> Reviewed-by: Thomas Huth <thuth@xxxxxxxxxx>
>
