X-Git-Url: https://bilbo.iut-bm.univ-fcomte.fr/and/gitweb/loba.git/blobdiff_plain/d43e446400ac7b65d7d2dd3acd57f8f669484449..23529d55ccbfd0a851fdc86708320811ddda8ec4:/TODO?ds=inline diff --git a/TODO b/TODO index 07ec631..3152ad9 100644 --- a/TODO +++ b/TODO @@ -1,17 +1,31 @@ -* implement loba_* algorithms (start with some trivial one) -* add loba algorithm selection (-a number ?) +* segfault with ./loba cluster1000.xml -N64 -a fairstrategy + + this is a deadlock occuring when: + - a process is in the finalize stage; + - all processes but one are blocked on receive; + - the process that is still running owns all the remaining load, + and sends it all to the finalizing process, and then goes in + blocking receive. + The finalizing process receives the load, and blocks again, + waiting for a close message. + All processes are then blocked, and non-one is able to see that + there is no more load in the system! + +* verify bookkeeping version. + +* add several metrics + - flops (total, avg, stddev) + - message exchanges (number, volume) -* fix process::run when load is 0 - -> wait for a message... - -> how does it work with opt::bookkeeping ??? +* add options -j/-J : minimum number of iterations ? + +* add a variant to (not) change neighbor load information at send. + +* implement loba_* algorithms (start with some trivial one) -* implement automatic process topology - (line, ring, star, btree, clique, hypercube, etc..) +* for automatic process topology, + -> implement some random initial distribution of load - use MSG_process_create_with_arguments - name function name (is duped...) - code use MSG_get_registered_function - data NULL - argc - argv argv[0] = process_name +* add synchronized mode +* translate README file?