I was fiddling with vg sift, using the option -1 whereby
-1 / --calc-insert Calculate and print the insert size mean / sd every 1000 reads.
It looks like this works by default or if -t 1, but if users try more than one thread, there's a Segmentation Fault. That's somewhat reasonable, as the options state:
-t / --threads <MTHRDS> number of OMP threads (not all algorithms are parallelized).
Perhaps this is something to polish at some point (I could consider a PR if there's time).
Here's an example, whereby file1.gam is approx. 45GB:
$ vg sift -t 2 -1 /path/file1.gam
Filtering /path/file1.gam
ERROR: Signal 11 occurred. VG has crashed. Run 'vg bugs --new' to report a bug.
Stack trace path: /tmp/vg_crash_SN6StQ/stacktrace.txt
Segmentation fault
and then
$ cat /tmp/vg_crash_SN6StQ/stacktrace.txt
Crash report for vg v1.14.0-206-gc9ba0364 "Quadrelle"
Stack trace (most recent call last):
#9 Object "", at 0xffffffffffffffff, in
#8 Object "", at 0x7efe808bcd0e, in
#7 Object "", at 0x7efe8585e4a3, in
#6 Object "", at 0x7efe80da06b5, in
#5 Object "", at 0x56303e9b6260, in
#4 Object "", at 0x7efe80da2f1b, in
#3 Object "", at 0x7efe80d9a4ae, in
#2 Object "", at 0x56303e9b5cae, in
#1 Object "", at 0x56303e9b5a07, in
#0 Object "", at 0x56303e911c4d, in
Similar behavior if -t N s.t. N>1, but never if N==1.
I was fiddling with
vg sift
, using the option-1
wherebyIt looks like this works by default or if
-t 1
, but if users try more than one thread, there's aSegmentation Fault
. That's somewhat reasonable, as the options state:Perhaps this is something to polish at some point (I could consider a PR if there's time).
Here's an example, whereby
file1.gam
is approx. 45GB:and then
Similar behavior if
-t N
s.t.N>1
, but never ifN==1
.