[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [patches] Possible PowerPC LIBC optimization



Mark Mitchell wrote:
> Steven Munroe wrote:
> 
>> My concern is that what works for 750 may not be best for 8540 or 970.
>> The powerpc-cpu add-on handles this by adding additional <cpu-type>
>> specific directory layer to the make system search order (i.e.
>> ./sysdeps/powerpc/powerpc32/750) that can override the base. This is
>> selected at configure time via --with-cpu=750 or your favorite <cpu-type>
> 
> As an aside, do you think that any of these kinds of optimizations are
> worth doing dynamically, based on the CPU type that we actually have?
> So that, for example, "strcpy" can be optimized for your CPU,
> independently of what CPU was used when configuring GLIBC?

I believe today at some level the memcpy already does this by paying
attention to the aux vector and the machine type to optimize by
cacheline size.

Some of this was discussed on the PowerPC IRC channel (#mklinux on
irc.freenode.net), and at least for the dcbt one, it's not clear if
there would be too much overhead to get an advantage.. but other cases
it's much clearer to dynamically change behavior based on the
cpu/cacheline/etc information available to glibc.

I think some of this stuff is going to take someone to just sitdown and
see if dynamically switching the optimizations is worth it or not..

--Mark