On 03/27/2015 12:32 PM, Ingo Molnar wrote:
> 
> * Denys Vlasenko <dvlasenk@redhat.com> wrote:
> 
>> This mimics the recent similar 64-bit change.
>> Saves ~110 bytes of code.
> 
> Would be nice to know whether you tested all the affected system entry 
> variants, on both AMD and Intel CPUs?

Yes, I did. I also looked at the diff of entry_64.o disassembly, to have
a different view of the changes.

I know how fragile this stuff is.