On Wed, Jun 02, 2010 at 11:23:37AM -0400, David Michael wrote: > On Wed, Jun 2, 2010 at 2:39 AM, Peter Robinson <pbrobinson@xxxxxxxxx> wrote: > > It does work in F-12, the response for the lack of support in F-13 was > > 'deal with it'. There is suppose to be a patch to emulate it in the > > kernel but apparently it won't go upstream until its a generic infra > > patch that can allow support of other emulated bits in other cpus in a > > generic way. So its possible it will come back, but I don't hold up > > hope of a quick resolution. Which leaves us in a big predicament as to > > how we're going to support the 1.5 million odd XO-1s out there moving > > forward. > > I believe this was the latest post of the NOPL emulation patch: > http://lkml.org/lkml/2010/3/1/430 This is not the general instruction > emulator mentioned, but a fix intended just for getting the Geode LX > classed as i686. > > I haven't used this patch myself yet; my Geode LX machine runs an > older Fedora, so it still works. I suppose I'll need to try the patch > during the next upgrade until things are settled. I wonder what the performance impact is. NOPL appears to be a variable length NOP (no-op). Obviously a very useful instruction for things like alignment, and gcc seems to stuff lots of them into the code: $ objdump -d /bin/ls | wc -l 16867 $ objdump -d /bin/ls | grep nopl | wc -l 369 369/16867 ~ 2% This is not a very fair comparison because we'd want to know how frequently NOPL is executed, but I hope it shows that these instructions are not infrequent. Having said that, AMD Geodes are slowwwwww anyway ... Rich. -- Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones virt-df lists disk usage of guests without needing to install any software inside the virtual machine. Supports Linux and Windows. http://et.redhat.com/~rjones/virt-df/ -- devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/devel
