On 27.03.20 19:01, Christian Borntraeger wrote: > > > On 27.03.20 17:55, David Hildenbrand wrote: >> On 27.03.20 17:33, Christian Borntraeger wrote: >>> On s390x hosts with a single CPU, the smp test case hangs (loops). >>> The check is our restart has finished is wrong. >>> Sigp sense running status checks if the CPU is currently backed by a >>> real CPU. This means that on single CPU hosts a sigp sense running >>> will never claim that a target is running. We need to check for not >>> being stopped instead. >>> >>> Signed-off-by: Christian Borntraeger <borntraeger@xxxxxxxxxx> >>> --- >>> lib/s390x/smp.c | 2 +- >>> 1 file changed, 1 insertion(+), 1 deletion(-) >>> >>> diff --git a/lib/s390x/smp.c b/lib/s390x/smp.c >>> index 2555bf4..5ed8b7b 100644 >>> --- a/lib/s390x/smp.c >>> +++ b/lib/s390x/smp.c >>> @@ -128,7 +128,7 @@ static int smp_cpu_restart_nolock(uint16_t addr, struct psw *psw) >>> * The order has been accepted, but the actual restart may not >>> * have been performed yet, so wait until the cpu is running. >>> */ >>> - while (!smp_cpu_running(addr)) >>> + while (smp_cpu_stopped(addr)) >>> mb(); >>> cpu->active = true; >>> return 0; >>> >> >> Yeah, same as the other issue we fixed before. There is no trusting on >> SIGP SENSE RUNNING STATUS. >> >> Can you please get rid of smp_cpu_running()? (looks like this was the >> last user) > > I think we should keep it and rename it to smp_cpu_running_status. This bug > actually showed that we should be able to test this feature (which is used > for spinlocks in Linux) somehow. AFAIK, there is no trusting on SIGP SENSE RUNNING STATUS at all (I discussed this with Janosch back then). And I don't see a way for a reasonable test either. But if you have plans to add a test, then yes, we can keep it. -- Thanks, David / dhildenb
