Re: [PATCH] update fix X86_64 procfs provide stack information for threads

[Stefani Seibold <stefani@seibold.net>]

Am Mittwoch, den 04.11.2009, 13:00 +0100 schrieb Andi Kleen:
> > This is true, but i think it is better to get an outdated value than a
> > complete wrong value like -1.
> 
> -1 means "I don't know".  I don't think "completely wrong" 
> is the correct term to describe that.
>  
> > The truth is that KSTK_ESP always return an outdated value on a multi
> > core system if the process never do a system call.
> 
> I think not supporting updates on interrupts at least is very poor.
> Unfortunately there's no good way fast path way to detect this I know of
> (that is why I originally added -1 here)
> 

I am sorry, i did not know that was your code. But anyway.

> 
> > Question: is task_pt_regs(task)->sp set in 64 bit mode when the process
> > is blocked in an interrupt? If true, we can add two additional assembly
> > instruction to the system call handler and store the stack pointer into
> > this. Than KSTK_ESP wil be again a simple macro like
> 
> You want to add instructions to one of the hottest kernel paths
> for this hyper-obscure application?  Bad idea.
> 

You complain that the the value is outdated and i told you how you can
get a more accuracy value. I agree that this is bad idea.
 

> My recommendation would be to just deprecate this proc field
> and if anyone really wants that information they can use
> a trivial ptrace() based user program.
> 

I spend a lot of time doing this, it would be nice to give it a change a
fix the KSTK_ESP macro. It will be not only used by my code. It would be
great if we can do this together. 

You have the knowledge, so i will ask my question again:
Is task_pt_regs(task)->sp set in 64 bit mode when the process is block
in an interrupt? 
Is there a way to detected if a process is blocked by an interrupt?

If you answer both with true than i can fix KSTK_ESP without performance
penalty for the rest of the system.

Stefani