>But even if the bogus check will be deleted, I think it would perhaps be
> handy to specify a "saefty margin" of non mlock-able mem, or the kernel
> will in some cases not have any phys pages left for swapping in and out
> mem.
>But the same applies to the memory overcommit problem.
>
>What would the optimal tradeoff look like ?
>BTW: any idea how other UNICES solve this problem ?
IRIX 6.5.x has a kernel parameter limiting the amount of memory an ordinary
user can lock. In case a user goes over this maximum the mlockall call will
fail. In case the user has specified MCL_FUTURE, a SIGSEGV will be raised
with a special code. (At least that is what they say they do, but in reality
this doesn't work).
IMHO, an ordinary user may only lock a limited amount of memory. Maybe it is
wise to specify a certain amount which may *not* be locked. This memory is
reserved for the rest of the processes.
Robert
-- Robert de Vries rhdv@rhdv.cistron.nl- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/