vector catenation way too slow
| Bugzilla Link | 79 |
| Created on | Oct 31, 2004 20:40 |
| Resolution | LATER |
| Resolved on | Nov 01, 2004 08:29 |
| Version | 1.00beta |
| OS | Linux |
| Architecture | PC |
| Attachments | bug79.sac.tar.gz |
Extended Description
This may be another case of bug#75... The benchmark operates by catenating vectors of shape 1,1 2, 1 2 3, 1 2 3 4...., under a FOR loop. In interpreted APL, a 1000-iteration loop takes 3.083 CPU seconds. Under SAC, it takes 6.3 seconds. If I can get the SISAL compiler to run again, I'll see how it does.
To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information