ARM11 MPCore with Linux is slower than ARM9 with OSE

[Bhattiprolu RaviKumar]

Hi All,
  I am working on porting some code from ARM9 (without MMU) with OSE to
ARM11 MPCore with Linux. This particular code reads and writes data from a
peripheral registers two bytes at a time. This register is memory mapped to
ARM address space. Even after moving the code to Kernel on ARM11, it runs
very slow compared to similar code on ARM9 with OSE. All the branch
predictions are enabled in ARM11. Any pointers on what is going wrong here?

regards,
Ravi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.infradead.org/pipermail/linux-arm/attachments/20110317/95072d68/attachment.html>