Hartmut Kaiser
Hartmut Kaiser
> Please note that the performance gains are actually not very significant. Reason for minimal performance gains is because std::generate_n has very similar performance when compiled with -O3 flag. Google...
@Johan511 could you please rebase this onto master, now that the release is out?
inspect was reporting: ``` /libs/core/algorithms/include/hpx/parallel/algorithms/detail/generate.hpp *I* missing #include (type_traits) for symbol std::true_type on line 57 ``` Please rebase one more time to pull in all changes from master.
retest lsu
> And for the performance, both **seq** and **unseq** are almost generating same assembly with **Release Mode** (uses -O3). Because compiler is able to vectorize the loops in **seq** mode...
@m-diers Marco could you please be more specific how to reproduce this? Would you have a small code reproducing it?
@Pansysk75 thanks. Doesn't that show trends opposite to the measurement we did before? This PR binds threads to cores in a way preventing for them to be stolen by other...
Here is another data point supporting the evidence that we have not entirely fixed the performance regression yet: The yellow line is what we have to strive for, the pink...
The Python `pass` construct could be translated to a simple `nil` (or nothing). The `continue` construct will require code transformations as we will not be able to implement something resembling...
@ipdemes: I just opened this PR for you to follow the progress. It will take more work, however. I'll remove the WIP tag as soon as it's ready to be...