Re: [Valgrind-developers] mtV : improve Valgrind to run multiple threads in parallel

[Josef W. <Jos...@gm...>]

Am 19.06.2012 18:38, schrieb Julian Seward:
>>> Yes, I agree.  They are going to share libc stuff, + supporting library
>>> code (X, Qt, etc).
>>
>> For sure, that is the case for any kind of "general" application using
>> threads.
>> For applications which are using threads to parallelise algorithms,
>> each thread will run exactly the same bunch of code.
>> I suspect this case is becoming more and more frequent.
>
> I completely agree.

Yes, it probably is mostly the same, but the question is how much code
is shared in absolute numbers, as otherwise, the time for redundant 
JITting probably mostly can be neglected.

I have the impression that
typically a programmer limits the ranges of code which really runs
in parallel, to not introduce bugs. With OpenMP,
you usually parallelize your code incrementally, loop after loop,
and each loop usually is restricted to some algorithm.
Perhaps this is just my HPC-centric view...

Josef