Messages in this thread | | | Date | Tue, 6 Nov 2018 21:42:55 +0100 | From | Jiri Olsa <> | Subject | Re: [PATCH RFC] hist lookups |
| |
On Mon, Nov 05, 2018 at 08:53:42PM -0800, David Miller wrote: > > Jiri, > > Because you now run queued_events__queue() lockless with that condvar > trick, it is possible for top->qe.in to be seen as one past the data[] > array, this is because the rotate_queues() code goes: > > if (++top->qe.in > &top->qe.data[1]) > top->qe.in = &top->qe.data[0]; > > So for a brief moment top->qe.in is out of range and thus > perf_top__mmap_read_idx() can try to enqueue to top->qe.data[2] > > We can just do: > > if (top->qe.in == &top->qe.data[1]) > top->qe.in = &top->qe.data[0]; > else > top->qe.in = &top->qe.data[1]; > > Or, make top->qe.in an index, and simply go: > > top->qe.in ^= 1; > > Either way will fix the bug.
I pushed that fix in perf/fixes branch, but I'm still occasionaly hitting the namespace crash.. working on it ;-)
[root@krava perf]# ./perf top -d 1 -g perf: Segmentation fault -------- backtrace -------- ./perf[0x5956aa] /lib64/libc.so.6(+0x36f2f)[0x7f99d05d3f2f] /lib64/libc.so.6(+0x15ce60)[0x7f99d06f9e60] ./perf(namespaces__new+0x5a)[0x4fdeaa] ./perf(thread__set_namespaces+0x55)[0x4ff335] ./perf(machine__process_namespaces_event+0xa5)[0x4f1c95] ./perf[0x44c78b] ./perf(queued_events__flush+0xcd)[0x4fd46d] ./perf[0x44c25b] /lib64/libpthread.so.0(+0x7593)[0x7f99d2928593] /lib64/libc.so.6(clone+0x3e)[0x7f99d0696e6e]
jirka
| |