Author: Bo Persson
Date: 00:54:28 08/06/03
Go up one level in this thread
On August 05, 2003 at 14:32:56, Frank Phillips wrote: >On August 05, 2003 at 14:27:03, Eugene Nalimov wrote: > >>On August 05, 2003 at 13:35:18, Frank Phillips wrote: >> >>>... >>> >>>Eugene has previously stated that gcc is philosophically slow because of the >>>developers intention to not concentrate on specific cpu architecture – I think. >> >>No, I did bot stated that. I stated that >> >>(1) Some ancient decisions were made in gcc design when VAX and mc68k were >>primary gcc targets; the worst one is poor memory disambiguation. Now that >>decisions hurt optimizations a lot, especially on CPUs with large number of >>registers, but fixing those decisions require *highly coordinated* rewrite of >>lot of code for lot of targets. >> >>(2) Some optimizations that would benefit *all* target CPUs are rejected on the >>strictly ideological principles. >> >>Thanks, >>Eugene >> >>>Frank > >I had in mind (2), but clearly misunderstood – apologies. > >(1) does perhaps not bode well for the AMD64 depending on what a large number of >registers is .....? I think he refers to the Itanium with 128 registers. Compared to that, 16 registers is not extremely large... Bo Persson bop2@telia.com > >Frank
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.