On 04/22/12 20:15, Nilay Vaish wrote:
> On Sun, 22 Apr 2012, Steve Reinhardt wrote:
>
>> On Sun, Apr 22, 2012 at 7:20 PM, Nilay Vaish <[email protected]> wrote:
>>
>>>
>>> The way we currently recognise the CC register needs to be read, is
>>> that
>>> it appears on the RHS of some assignment statement. It seems to me that
>>> this would remain true even after the register is split. The expression
>>> will just change to something like
>>>              zaps = genZaps(zaps, <something>);
>>>
>>> The isa parser would then mark zaps as both source and destination.
>>> And the
>>> two add instructions would still not execute in parallel, as the
>>> second one
>>> would be dependent on the first for the value of the zaps register.
>>>
>>> We can either  --
>>> a. drop the default assumption that we need to make partial updates,
>>> and
>>> handle
>>>   partial update as a special case.
>>> b. keep the default assumption that we need to make partial updates,
>>> and
>>> handle
>>>   full update as a special case.
>>>
>>> Current patches are along the lines of the second option.
>>>
>>
>> I'm jumping in partially informed, but can we just have two
>> functions, like:
>>
>>    zaps = setZaps(<something>);
>>
>> and
>>
>>    zaps = modifyZaps(zaps, <something>);
>>
>> and then let the isa parser do its stuff naturally?
>>
>> Steve
>>
>
> I don't think that is possible. This code will appear in the .isa
> file. In the .isa file, we cannot decide which version to use as the
> CC bits to be written vary with the context in which the microop is
> used. So, we need a run time condition that figures out tries to
> evaluate if the register needs to be read.
>
> -- 
> Nilay
> _______________________________________________
> gem5-dev mailing list
> [email protected]
> http://m5sim.org/mailman/listinfo/gem5-dev

One of the main challenges of this effort is going to be figuring out a
better way for this mechanism to work. I don't have a concrete plan, but
we should figure out how to make the parser smarter/give it better tools
rather than try to hack around its limitations. I was thinking there
would somehow still be a single condition codes register, and it, the
python operand class representing it, or something, would figure out
what flags to read and stuff into it and what flags to pick out to write
back. When I said before that I took some steps towards composite
operands, this is one of the things I had in mind.

Gabe
_______________________________________________
gem5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/gem5-dev

Reply via email to