On Thursday, October 16, 2008 5:24 pm Jesse Barnes wrote: > Ok this one has locking & fence register allocation. I'm still trying to > figure out why things get mapped cached though, clearly something is going > wrong with my vm_insert_pfn; it's acting like I'm mapping the wrong offset > since fencing doesn't seem to work either (though that could be due to > caching problems also), but afaict the pfn value is correct etc.
Additional data point on this: if you uncomment the cache flush in the fault handler, things are slow until the whole object in question is faulted in, but after that things seem to work ok in simple tests like x11perf -fitext (i.e. no cache-looking corruption) so it may be that I'm just missing a higher level flush somewhere. Anyway I'll try to fix it up when I get back home... Jesse ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ -- _______________________________________________ Dri-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/dri-devel
