On Mon, 29 Jan 2007 17:31:20 -0800
"Martin J. Bligh" <mbligh@xxxxxxxxxx> wrote:
Peter Zijlstra wrote:On Sun, 2007-01-28 at 14:29 -0800, Andrew Morton wrote:CONFIG_HIGHPTE is always horrid -we've known that for years.
As Christoph says, it's very much preferred that code be migrated over toCONFIG_HIGHPTE code in -rt was horrid. I'll do some measurements on
kmap_atomic(). Partly because kmap() is deadlockable in situations where a
large number of threads are trying to take two kmaps at the same time and
we run out. This happened in the past, but incidences have gone away,
probably because of kmap->kmap_atomic conversions.
From which callsite have you measured problems?
mainline.
We have? What's wrong with it? <looks around for bug reports>