On Tue, Oct 22, 2013 at 11:45:35AM +1100, Adam Carter wrote:
> If you havent already, I would first verify that its actually CPU bound,
> before changing CFLAGs and recompiling everything. So take a look at top,
> vmstat, mpstat etc when you're noticing slowness. If it is truely CPU bound
> and you're going to recompile everything, you could consider upgrading to
> the ~ version of gcc first, with the assumption that the optimizations
> maybe be better. However, my gut feeling is that you wont get much or any
> improvement over your current CFLAGs.
> 
> The i686 and -Os ideas are interesting. See if you can find any benchmarks.
> 
> Also - try diffing the kernel .configs - maybe you missed something
> important on the slow system.

Interestingly, I did carry out tests when I received my netbook in order
to decide between 32 and 64 bit. I did the same tests when I migrated my
big laptop from 32 to 64 bit, but I can't remember the results for the
netbook anymore except for LUKS performance:
the aforementioned hdparm -t on my encrypted /home amounts to 18 MB/s on
32 bit, but reaches almost 30 MB/s with 64 bit. In the end, I went for a
64 bit kernel to increase some computing performance, and 32 bit for all
the rest for memory reasons. The only additional "cost" is that I have
to maintain a 64 bit toolchain via crossdev.
-- 
Gruß | Greetings | Qapla’
Please do not share anything from, with or about me with any Facebook service.

Arrogance is the art of being proud of one’s own stupidity.