Messages in this thread | | | Subject | Re: [PATCH] Instrumenting high latency | From | Arjan van de Ven <> | Date | Mon, 12 Jul 2004 10:29:40 +0200 |
| |
On Mon, 2004-07-12 at 09:34, Andrew Morton wrote: > Con Kolivas <kernel@kolivas.org> wrote: > > > > He hacked > > together this simple patch which times periods according to the preempt > > count. > > OK, small problem. We have code which does, effectively, > > if (need_resched()) { > drop_the_lock(); > schedule(); > grab_the_lock(); > } > > so if need_resched() stays false then this will hold the lock for a long > time and bogus reports are generated:
... or reset the time(r) in need_resched() under the assumption that all need_resched() callers will yield when it returns true... [unhandled content-type:application/pgp-signature] | |