On 12/7/05, jamal <[EMAIL PROTECTED]> wrote: > On the prefetch, i think would you agree now that it is problematic? Sorry, I don't agree.
> I just showed that if i changed the cycle of execution between the > moment the prefecth gets issued to the moment the data gets used we get > different performance results. Sure you can always add delays in places to break the use of getting data into cache correctly. In this case that time is set with the code in the driver. It does run differently on different systems but we (you) still have not shown that it's broke anywhere. Putting in delays to where the cache becomes stale will of course break the use of prefetch which you are showing. It's also not the case with the code that is in the driver between where the prefetch is done and the data is used. It's limited to what is in the driver. > > I think there is value for prefetch - just not the way the current patch > has it. Something less adventorous as suggested by Robert would make a > lot more sense. The different between the cases was not significant and the prefetching cases were better than no prefetching. Again, still no detriment to performance. -- Cheers, John - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
