Mathieu Poirier <mathieu.poir...@linaro.org> writes: > Function perf_event_mmap() is called by the MM subsystem each time > part of a binary is loaded in memory. There can be several mapping > for a binary, many times unrelated to the code section. > > Each time a section of a binary is mapped address filters are > updated, event when the map doesn't pertain to the code section. > The end result is that filters are configured based on the last map > event that was received rather than the last mapping of the code > segment.
Good catch! I'd like to fix it in 4.7-stable as well; I think it's too late for 4.7 already. > Signed-off-by: Mathieu Poirier <mathieu.poir...@linaro.org> > --- > kernel/events/core.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/kernel/events/core.c b/kernel/events/core.c > index df21611585d7..b9aa8f0ff070 100644 > --- a/kernel/events/core.c > +++ b/kernel/events/core.c > @@ -6604,7 +6604,8 @@ void perf_event_mmap(struct vm_area_struct *vma) > /* .flags (attr_mmap2 only) */ > }; > > - perf_addr_filters_adjust(vma); > + if ((vma->vm_flags & VM_EXEC) && (vma->vm_pgoff == 0)) > + perf_addr_filters_adjust(vma); You shouldn't need the vm_pgoff check; the range comparison logic in __perf_addr_filters_adjust() should already take it into account. Also, I'd put the check to perf_addr_filters_adjust() instead, with a comment that we don't do data-based filters yet or something along those lines. Thanks, -- Alex
