>Once this is done, it works OK, but really slow.  The code has alot
>of excess baggage in counting instruction cycles and emulating pipelines
>which is terribly useful when your designing silicon, but just slows an
>emulator down when your trying to debug operating system code.

In a sense actually this stuff is good.  For writing OS code it's sometimes 
good to be able to count pipeline stalls, TLB misses and the like.  I agree 
that it would often also be good to be able to turn it off and make the 
emulator run quicker, though.

p.


unsubscribe: body of `unsubscribe linux-arm' to [EMAIL PROTECTED]

Reply via email to