http://bugzilla.suse.com/show_bug.cgi?id=1040381
http://bugzilla.suse.com/show_bug.cgi?id=1040381#c2
--- Comment #2 from Michal Hocko ---
Ohh, I can see it now
oom01 0 TINFO : start OOM testing for mlocked pages.
oom01 0 TINFO : expected victim is 1783.
oom01 0 TINFO : thread (7ff2c491e700), allocating 3221225472 bytes.
so the memory is mlocked which explains things. oom_reaper cannot handle
mlocked pages currently. Older implementations didn't have any upper bound on
how long we wait for the oom victim to die which has led to silent oom hangs.
I plan to address mlocked memory in oom_reaper but this had a lower priority
because this doesn't happen all that often. I will look into it but I cannot
give you any estimate right now, unfortunately.
--
You are receiving this mail because:
You are on the CC list for the bug.