On Fri, Feb 27, 2026 at 12:05:13PM -0800, Mukesh R wrote: [...] > > > > So? But does it means to 'be in asmlinkage' in your interpretation? Did you > > check what 'asmlinkage' actually evaluates to? > > > > I am not asking you to justify why this broken code works in practice, I am > > asking you to fix it. > > > STOP bossing me! I am not your servant nor your slave. And you are not the > only genius around here. > > Now, many people looked at this code before it was merged and no one really > thought any self respecting compiler in modern times would create an issue > here. Still, I see the remote possibility of that happening. All you had > to do was to show your concern and suggest using __naked here (which looks > like we all missed, or maybe it came after the code was written), and it > would have been addressed. This is x64 specific code for very special case > of hyperv or kernel-on-hyperv crashing. > > In future if you choose to correspond, watch your tone!
Mukesh, there is no need to be so emotional and defensive. I don't think anyone, no matter how good he or she is, knows all the intricacies in the kernel. We're lucky to have other people look at our code and point out potential issues. Regardless of your opinion on the discussion, we should be thankful for the time and effort people put into even sending an email, let alone a patch. Let's keep the discussion civil and constructive, and focus on the technical aspects of the code. Ard, I want to let you know that I appreciate you raising this issue with us. Thanks, Wei
