Hello there. I'm a student and i'm trying some experiments with CilkPlus of icc 15. I'm using Ubuntu 12.04 with x64 Intel Processor.
I wonder how reducers work internally. So if a value is set into a reducer, does it block other threads each time a value is set?
I ask because normally I'm creating a local 'reducer', e.g. a local histogram on an image tile and on leaving the thread all the data is pushed at once into the global reducer. Just like local memory operations in OpenCL.
While optimizing a matrix manipulation code in C, I used CilkPlus to spawn a thread to execute in parallel two functions that are data independent and somewhat computationally intensive. Cilk_spawn is used in only one place in the code as follows:
I would like to understand run-time execution in Cilk a little better.
I have downloaded Intel Cilk run-time release (cilkplus-rtl-003365 - released 3-May-2013).
On 09/09/2013 I had asked a question seeking to figure out which is the last function executed before Cilk run-time ends assuming execution went without any problems.
Barry suggested to look at “__cilkrts_c_return_from_initial()” in scheduler.c and indeed that was what I needed at that time.
I was wondering whether there was some way to modify the makefiles for the GCC build of Cilk Plus to keep the labels for debugging (i.e. compile with -g). I am trying to get a fuller picture of how the runtime system works in order to possibly modify it
I would like to understand Cilk worker creation a litter better.
I am not sure how to phrase this question so I’ll give it my best.
I have downloaded Intel Cilk runtime release (cilkplus-rtl-003365 - released 3-May-2013).
I would like to create a new Cilk worker that does not cause cross-threading issues but this new worker would not be a part of the work collective.
I noticed in the Intel Cilk Plus runtime library change log (https://www.cilkplus.org/sites/default/files/runtime_source/changes-3453...), support for ARM with Intel C/C++ Compiler mainline-to-14.0 branch cutoff. Is there any possibility to use the beautiful cilk plus on the ARM architecture?
Is std::vector support concurrently read - write access for Cilk workers ? And which vector operator should I use ? Now I'm using like " data [i] = x ; " for assignment.
P1 and P2 data race problems are appear when I inspect program . I haven`t got experience about Cilk Plus. I guess error is occur in nested loop . What should I do ?