On 15.12.2021 11:32, Jan Beulich wrote:
> (Re-sending an abridged version, as apparently spam filters didn't like
> the original message with more retained context; I'll have to see whether
> this one also isn't liked. Sorry.)
> 
> On 15.12.2021 10:48, Michal Orzel wrote:
>> This patch and the problem it solves is about clearing top 32bits of all gp 
>> registers so not only x0,x1.
> 
> That's well understood. Yet for everything still in registers simply
> using mov ahead of the respective push (as you had it) is still
> preferable imo.
> 
> Jan
> 

In that case let's wait for Julien's opinion to decide whether I should get 
back to the previous
solution with mov or to the stack solution.

Cheers,
Michal

Reply via email to