On Tue, Oct 10, 2006 at 11:37:04AM -0700, Nishanth Aravamudan wrote:
> On 10.10.2006 [17:24:07 +1000], David Gibson wrote:
> > This time commit c3f4d10a80d6ad8bb37ef6f80e70c3c7d778117a.  The goal
> > noted in the comment of avoiding normal page mappings in the same
> > segment as the hugepage bss is all very laudable, but this has the
> > side effect of requiring up to 15 more hugepages for the program than
> > might otherwise be needed.  Ouch.
> 
> Yes, this is sub-optimal. But the alternative was to have a library
> which we knew would not work for certain applications. We have a
> post-1.0 solution in the works, fwiw, but it will require a bit of time
> to get done.

Ok.. tell me more about this proposed approach.  At present, with the
amount of memory in my test machine, this patch means that a great
many of the testcases fail.

-- 
David Gibson                    | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au  | minimalist, thank you.  NOT _the_ _other_
                                | _way_ _around_!
http://www.ozlabs.org/~dgibson

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Libhugetlbfs-devel mailing list
Libhugetlbfs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/libhugetlbfs-devel

Reply via email to