29 May
2013
29 May
'13
9:45 a.m.
On 29/05/13 10:59, David Bellot wrote:
Are you testing your code on a specific machine or a virtual one ? What's about things like Raspberry Pi ? I'd like to see benchmark on this little things. Maybe you can start benchmarking ublas on a tiny machine like that and/or an Android device and see how gcc is able to generate auto-vectorized code for this machine. Check the assembly code to see if Neon instructions have been correctly generated.
The Raspberry Pi does not support NEON. Also bear in mind that when it comes to benchmarking, difference is quite significant between Cortex-A8, Cortex-A9 and Cortex-A15.