aboutsummaryrefslogtreecommitdiffstats
path: root/TODO
diff options
context:
space:
mode:
authorMattias Andrée <maandree@kth.se>2016-03-16 14:30:29 +0100
committerMattias Andrée <maandree@kth.se>2016-03-16 14:30:29 +0100
commit3e3b44d087ab616089402129b2bc4c4831c6b33a (patch)
tree2c0a6e9550dded9e336906514c6ad7343dc8257d /TODO
parentFix bug in libzahl_msb_nz_* and optimise and simplify libzahl_realloc (diff)
downloadlibzahl-3e3b44d087ab616089402129b2bc4c4831c6b33a.tar.gz
libzahl-3e3b44d087ab616089402129b2bc4c4831c6b33a.tar.bz2
libzahl-3e3b44d087ab616089402129b2bc4c4831c6b33a.tar.xz
Optimise zsqr, zmul, zstr, zdivmod, zpow, and zpowu
Signed-off-by: Mattias Andrée <maandree@kth.se>
Diffstat (limited to 'TODO')
-rw-r--r--TODO4
1 files changed, 4 insertions, 0 deletions
diff --git a/TODO b/TODO
index 603fabb..5c69340 100644
--- a/TODO
+++ b/TODO
@@ -15,3 +15,7 @@ Test optimisation of zmul:
- [(Hb - Lb)(Hc - Lc) << m2]
+ [(Lb * Lc) << m2]
+ (Lb * Lc)
+
+Would zmul be faster if we split only one of the
+factors until they are both approximately the same
+size?