Re: [PATCH 3/3] sched/core / kcov: avoid kcov_area during task switch

2018-05-04 Thread Mark Rutland
On Fri, May 04, 2018 at 05:32:26PM +0300, Andrey Ryabinin wrote: > On 05/04/2018 04:55 PM, Mark Rutland wrote: > > > +#define kcov_prepare_switch(t) \ > > +do { \ > > + (t)->kcov_mode |= KCOV_IN_CTXSW;\ > > +} while (0) >

Re: [PATCH 3/3] sched/core / kcov: avoid kcov_area during task switch

2018-05-04 Thread Andrey Ryabinin
On 05/04/2018 04:55 PM, Mark Rutland wrote: > +#define kcov_prepare_switch(t) \ > +do { \ > + (t)->kcov_mode |= KCOV_IN_CTXSW;\ > +} while (0) > + > +#define kcov_finish_switch(t)\ > +do {

[PATCH 3/3] sched/core / kcov: avoid kcov_area during task switch

2018-05-04 Thread Mark Rutland
During a context switch, we first switch_mm() to the next task's mm, then switch_to() that new task. This means that vmalloc'd regions which had previously been faulted in can transiently disappear in the context of the prev task. Functions instrumented by KCOV may try to access a vmalloc'd kcov_a