Displaying 5 results from an estimated 5 matches for "syntehsizer".
2015 Jul 23
3
[LLVMdev] some superoptimizer results
...ces where the
> result uses a "subset" of the input instruction sequence.
Yeah, I had been noticing those subsets too. It sounds like it's worth a
try doing a run that looks only for those.
One nice thing is that if we're just looking for subsets, we will not even
give the syntehsizer the option to use instructions that aren't in the
subset -- this should give a nice speedup, and probably will even make
synthesis more likely to succeed (assuming that an optimization exists
within the subset).
John
2015 Jul 23
0
[LLVMdev] some superoptimizer results
...es a "subset" of the input instruction sequence.
>
> Yeah, I had been noticing those subsets too. It sounds like it's
> worth a try doing a run that looks only for those.
>
> One nice thing is that if we're just looking for subsets, we will not
> even give the syntehsizer the option to use instructions that aren't
> in the subset -- this should give a nice speedup, and probably will
> even make synthesis more likely to succeed (assuming that an
> optimization exists within the subset).
+1 - This sounds like a very interesting approach. I suspect it...
2015 Jul 23
0
[LLVMdev] some superoptimizer results
On Wed, Jul 22, 2015 at 11:33 PM, John Regehr <regehr at cs.utah.edu> wrote:
> Interesting. Do you happen to have some examples laying around?
>>
>
> Sorry, no, I'll have to re-run. I felt that the select-heavy results were
> sort of humorous and clever at first, but then just annoying.
>
> Except for decode-limited situations, in general decreasing the critical
2015 Jul 24
2
[LLVMdev] some superoptimizer results
...; of the input instruction sequence.
>>
>> Yeah, I had been noticing those subsets too. It sounds like it's worth a try
>> doing a run that looks only for those.
>>
>> One nice thing is that if we're just looking for subsets, we will not even
>> give the syntehsizer the option to use instructions that aren't in the subset
>> -- this should give a nice speedup, and probably will even make synthesis more
>> likely to succeed (assuming that an optimization exists within the subset).
> +1 - This sounds like a very interesting approach. I suspec...
2015 Jul 23
3
[LLVMdev] some superoptimizer results
> Interesting. Do you happen to have some examples laying around?
Sorry, no, I'll have to re-run. I felt that the select-heavy results were
sort of humorous and clever at first, but then just annoying.
> Except for decode-limited situations, in general decreasing the critical path length is more important
> than eliminating instructions. The critical path length is a