On Mon, 19 Jul 1999 17:28:51 +0100 (GMT), Matthew Kirkwood
<weejock@ferret.lmh.ox.ac.uk> said:
> What about another task flag to cope with this? Hunt down the places
> which do set_fs(KERNEL_DS) and set a flag in the dangerous case.
> Then the two-liner can examine that flag instead of get_ds().
There are all sorts of places in the VM where it is possible to end up
recursing back into the swapout code. I'd be very nervous about opening
up potential new races in that case, where a recursive IO in the pager
ended up mistakenly inheriting that flag.
--Stephen
-
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/