[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH v3 1/6] x86: NOP out XPTI entry/exit code when it's not in use



>>> On 16.03.18 at 11:37, <jgross@xxxxxxxx> wrote:
> I think we should do some performance testing to decide whether it makes
> sense to take the patch or not. After all I don't see a reason to punish
> AMD processors for INTEL's bugs. And I've already found out that some
> branches in interrupt handling can really make a difference.

Well, I had done that before first posting, and found a marginal
improvement. But of course I didn't do this with parts of the patch
used and parts left out, and to be honest I don't think I'm going
to, as the differences will likely disappear in the noise. Hence I
stand to what I've said earlier to Andrew - I could accept to split
the patch (to move out the STI patching), and I could also accept
to drop the patching of the xen_cr3 restores. But I think the big
chunks being patched should at least be taken.

Jan


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.