This is the mail archive of the mailing list for the GCC project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: SPEC / testsuite results for disabling SFTs and the alias-oracle patches

On 03/05/08 06:48, Richard Guenther wrote:

you can see that in both cases the runs without SFTs are significantly
better(!)  Which hints at the fact that we do a poor job with parititoning
and/or that partitioning triggers earlier with SFTs enabled.

All these differences seem to be less than 1%. If I'm reading this table right, the alias oracle is not really having a significant effect on the scores

The oracle patches are able to slightly improve the results in the non-SFT
case, but overall there is less difference patched vs. unpatched compared
to the differences that result if you disable SFTs.

Without the alias oracle you get:

		SFTs (base)	No SFTs (peak)
SPECint		1914		1927 (+0.68%)
SPECfp		2029		2032 (+0.15%)

With the alias oracle you get:

		SFTs (base)	No SFTs (peak)
SPECint		1918		1923 (+0.26%)
SPECfp		2020		2039 (+0.94%)

The good news is that the oracle is not introducing any slowdowns. So I think this is very positive.

Thus, with the above results I propose we disable generating SFTs by
default on the mainline (--para max-fields-for-field-sensitive=100
is still available for comparision).  I will prepare a patch to adjust
the false negative testcases above to check for optimization outcome
as well.

Yes, good idea. Though I think we should just get rid of SFTs outright. They are only going to be a maintenance problem.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]