Re: PMC strawman

[Matt Thomas <matt%3am-software.com@localhost>]

At 06:46 PM 2/14/2003, Allen Briggs wrote:
> On Fri, Feb 14, 2003 at 10:51:52AM -0800, Matt Thomas wrote:
> > Note that to use perfctrs with individual pthreads we need
> > to add a PMC glob to struct mcontext.  Does anyone have a
> > problem with that?
>
> You mean a glob to treat the PMC registers as part of the context?
> If it's only conditionally restored, that seems reasonable to me.
> It's what?  Another 44 bytes of mcontext overhead?  It might be
> nice to perhaps have a way to avoid that for production system
> code, though.  Maybe a compile-time knob?

There would be a bit to note whether that blob has info.  I'd
make the blob explicitly opaque to userspace.

> > But what about the kernel?  I assume that doing performance
> > of user processes and the kernel concurrently will be desired.
> > If so, then on user traps and execptions, the user PM context
> > will need to be saved in the trapframe and then the kernel's
> > loaded from cpu_info (so it's per-cpu).
>
> This is something that I kind of ignored with the XScale
> implementation.  It would certainly be useful to have,
> though I would certainly want to avoid the overhead unless
> PERFCTRS is enabled.

Of course it would conditional.  I'd also prefer to have
separate contents for interrupt / non-interrupt code.

> > Though the current PMC framework doesn't allow for it, it
> > might be useful for multiple processes/threads to be able to
> > share one PMC context.
>
> What do you think this should look like?

Right now the pmc counters are per proc, not lwp, which needs to be
fixed.  I was thinking about creating a pmc context that a process
could establish and share among it's children (or all of its threads).
Make it refcnt'ed and on last reference, free it.
Then l_md.md_pmc would be a pointer to it.  To access it will be
fairly cheap, the adding of PMCs is not:

        mfmsr   r
        mtcr    r
        b       4,28,1f

        GET_INFO(r)
        lwz     r,CI_CURLWP(r)
        lwz     r,LWP_MD+MD_PMC(r)

        mfspr   s,SPR_PMC1
        lwz     hi,PMC_PMC1(r)
        lwz     lo,PMC_PMC1+4(r)
        add     lo,s,lo
        stw     lo,PMC_PMC1+4(r)
        addze   hi,hi
        stw     hi,PMC_PMC1(r)
1:

Note that it can't be in the PCB since PCB's are not shared.


--
Matt Thomas               Internet:   matt%3am-software.com@localhost
3am Software Foundry      WWW URL:    http://www.3am-software.com/bio/matt/
Cupertino, CA             Disclaimer: I avow all knowledge of this message