OOM may be being too nice to killed processes

From: Adam Pennington (adamp@andrew.cmu.edu)
Date: Sun Nov 04 2001 - 15:36:00 EST


I may be misunderstanding this, but looking at this portion of code from
the oom task killer... Isn't it dangerous to give a process PF_MEMALLOC and
then only pass it a SIGTERM? My take is that the high priority bump up
should only happen for the force_sig(SIGKILL,p).

        /*
         * We give our sacrificial lamb high priority and access to
         * all the memory it needs. That way it should be able to
         * exit() and clear out its resources quickly...
         */
        p->counter = 5 * HZ;
        p->flags |= PF_MEMALLOC;

        /* This process has hardware access, be more careful. */
        if (cap_t(p->cap_effective) & CAP_TO_MASK(CAP_SYS_RAWIO)) {
                force_sig(SIGTERM, p);
        } else {
                force_sig(SIGKILL, p);
        }

Adam Pennington
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Wed Nov 07 2001 - 21:00:24 EST