Quoting Mika Kuoppala (2019-02-15 15:21:11) > Chris Wilson <ch...@chris-wilson.co.uk> writes: > > > Since we no longer need to hold struct_mutex to perform a global device > > reset, don't do so for igt_reset_wedge(). > > > > Oh but the interesting question is not about need, but should =) > Do it both ways?
Just doing it without should be enough for lockdep, and specifically remember to guard everything with lockdep_assert_held() when required. The reverse case of testing with it held can only prove a deadlock by hitting it! (At which point CI reboots the machine.) -Chris _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx