id,summary,reporter,owner,description,type,status,priority,milestone,component,resolution,keywords,cc,work_issues,upstream,reviewer,author,merged,dependencies,stopgaps
8666,Bug in cyclotomic linear algebra,craigcitro,AlexGhitza,"David Loeffler ran into this bug:

{{{
sage: K.<zeta4> = CyclotomicField(4) 
sage: m = matrix(K, [186]) 
sage: n = matrix(K, [125]) 
sage: m * n 
[-23087] 
}}}

(See http://groups.google.com/group/sage-devel/browse_thread/thread/4f8633d6acf1c4ef# for the full thread.)

The issue is that the bound for what modulus the entries can be computed modulo is off by a factor of 2, because it doesn't take the sign into consideration. (Amusingly, this was basically the same fix as in #4823.) ",defect,closed,blocker,sage-4.4,algebra,fixed,,robertwb,,N/A,Robert Bradshaw,"Craig Citro, Robert Bradshaw",sage-4.4.alpha0,,