From 63893d99ae0950bd13b6abf20f02d4eadac60b28 Mon Sep 17 00:00:00 2001 From: Tim Warburton <timwar@caam.rice.edu> Date: Tue, 25 Oct 2011 22:45:00 -0500 Subject: [PATCH] More MEMO hacking. --- MEMO | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-) diff --git a/MEMO b/MEMO index 64e3043fc..b87b37d0a 100644 --- a/MEMO +++ b/MEMO @@ -46,25 +46,19 @@ To-do - user interface for dim length prescription -- Sharing of checks across ILP instances - - Give a good error message if a parameter assignment in get_problems() is missing. - Deal with equality constraints. (These arise, e.g., when partitioning a loop of length 16 into 16s.) -- Slab decomposition for ILP - -> I don't think that's possible. +Future ideas +^^^^^^^^^^^^ - Better for loop bound generation -> Try a triangular loop -- Nested slab decomposition (in conjunction with conditional hoisting) could - generate nested conditional code. - -Future ideas -^^^^^^^^^^^^ +- Sharing of checks across ILP instances - Eliminate the first (pre-)barrier in a loop. @@ -87,12 +81,20 @@ Future ideas - Use gists (why do disjoint sets arise?) +- Nested slab decomposition (in conjunction with conditional hoisting) could + generate nested conditional code. + Dealt with ^^^^^^^^^^ +- Slab decomposition for ILP + -> I don't think that's possible. + - It is hard to understand error messages that referred to instructions that are generated during preprocessing. - -> Expose preprocessing to the user so she can introspect. + + -> Expose preprocessing to the user so she can inspect the preprocessed + kernel. - Which variables need to be duplicated for ILP? -> Only reduction -- GitLab