bosilca
bosilca
https://bitbucket.org/icldistcomp/parsec/pull-requests/337 Start less MPI requests. This ensure a smaller number of requests to check every time we need to progress the communication layer. Implement a cleaner Active Message layer
Recent Intel compilers (I checked starting from 2023.2) automatically enable fast-math mode for optimized builds. This mode allows optimizations for floating-point arithmetic that assume that arguments and results are not...
Provides support for both NEON and SVE ISA, with performance improvements in the order of 10%.
The decision to switch to a different algorithm based on count does not belong to the base algorithms, it should be done in the upper level decision functions. We need...
This PR mark 4 MCA params as deprecated. The others are either used indirectly via the param_index struct, or must be defined to be consistent with the rest of the...
Based on the discussions around #12384, it appears that we are not correctly handling the requests passed to the user in the ishrink case. After discussion with @abouteiller we have...
This is not a correctness issue but a user friendliness issue. Including some UCX headers while compiling OMPI with picky compiler options generate a ton of warnings. The warning in...
This is not a correctness issue but a user friendliness issue. Including some UCX headers while compiling OMPI with picky compiler options generate a ton of warnings. The warning in...
Provide a hint to UCC collective init to know it is dealing with a blocking collective.