[OLPC-devel] Re: compiler optimizations for the geode
vmakarov at redhat.com
Tue Aug 15 11:17:29 EDT 2006
Jordan Crouse wrote:
>On 14/08/06 17:43 -0400, Christopher Blizzard wrote:
>>"If somebody is interesting in GEODE processor, you can find
>>documentation on AMD website (unfortunately, it is not
>>enough detailed to do tunning)."
>>Anyone from AMD want to give him a helping hand?
>I'll turn it back around - what does he need from us?
Jim Gettys wrote me that Geode
would only issue FP instructions every other cycle. Is it true? I can
not find it in the documentation. If it is true I need to change DFA
description and run benchmarks again. Currently my Geode description
simulates possibility of issuing one insn per cycle (int or fp).
I found that there are a few functional units for execution (fp, int,
load/store). I've just need to make sure that Geode can issue at most
one insn per cycle. Is this true?
Last thing not clear to me is int multiply and division. Will these
insns block execution all other int insns until the division or multiply
is finished? That is what I simulate for now. Is execution of all
other int/fp insns pipelined or their execution blocks execution of
other insns in the corresponidng functional unit (fp, int)?
If you are interesting there are two problems with gcc insn scheduler
for now. Insn scheduling before the register allocation is off for
x86. There is a bug which results in compiler crash when the 1st insn
scheduling is on and it is not fixed yet. But there is even bigger
problem. The current gcc insn scheduler can not estimate register
pressure. Therefore code generated with the 1st insn scheduling is
usually worse for x86 (because of numerous register spills). If the two
problems are solved, the generated code would be better. But I don't
know when it will be fixed.
More information about the Devel