On 13/02/2018 16:05, Chris Wilson wrote:
It is a known issue that some "commodity" machines fail when asked to
hotplug CPU0, so don't.

Signed-off-by: Chris Wilson <ch...@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursu...@intel.com>
--
Tvrtko, is it just this simple? Or do we need to avoid the physical
package?

Problem is it makes the test pointless since it is never able to kick the PMU off CPU0.

Regards,

Tvrtko

-Chris
---
  tests/perf_pmu.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/perf_pmu.c b/tests/perf_pmu.c
index a7501ca5..e9ce2881 100644
--- a/tests/perf_pmu.c
+++ b/tests/perf_pmu.c
@@ -982,7 +982,7 @@ static void cpu_hotplug(int gem_fd)
         * this has not affected busyness stats in the parent.
         */
        igt_fork(child, 1) {
-               int cpu = 0;
+               int cpu = 1;
close(link[0]);
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Reply via email to