32bit PAE v.s. non-PAE performance

Andy Whitcroft apw at canonical.com
Tue Jan 5 10:58:44 UTC 2010


On Sun, Jan 03, 2010 at 07:42:13AM -0700, Tim Gardner wrote:
> Gleaned from a discussion on LKML entitled 'Ubuntu 32-bit, 32-bit PAE,
> 64-bit Kernel Benchmarks'
> 
> http://www.phoronix.com/scan.php?page=article&item=ubuntu_32_pae&num=1
> 
> It appears that the performance impact of PAE is not as great as was
> originally feared (given a normal workload).

Yeah that is an interesting benchmark.  PAE is clearly not something to
be worrying about heavily in the common case.  If all CPUs supported PAE
we could consider dropping the generic flavour completely.  Cirtainly
migrating people with >3GB of ram seems pretty safe.

The most intresting thing though for me, was that 64bit rocks
comparitivly.  That is amazing.

-apw




More information about the kernel-team mailing list