On 21 October 2016 at 11:00, Sitsofe Wheeler <sitsofe@xxxxxxxxx> wrote: > > Backtrace says this: > Program terminated with signal SIGSEGV, Segmentation fault. > #0 fio_libaio_getevents (td=0x7fb0711e0ad0, min=0, max=1, > t=<optimized out>) at engines/libaio.c:160 I've found an easy way to reproduce this: echo "0 137438953472 zero" | dmsetup create zero64t2 fio --runtime --allow_file_create=0 --direct=1 --iodepth=8 --ioengine=libaio --bsrange=512-32k --time_based=1 --runtime=12h --norandommap --name=go --filename /dev/mapper/zero64t2 --name=go2 --filename /dev/mapper/zero64t2 & sleep 5; dmsetup suspend zero64t2; sleep 5 && kill % && fg; dmsetup resume zero64t2 This happens even with fio-2.15. The cores don't appear until the dmsetup resume is performed and it speeds things up if #define FIO_REAP_TIMEOUT in fio.h is set to be a low number... -- Sitsofe | http://sucs.org/~sits/ -- To unsubscribe from this list: send the line "unsubscribe fio" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
