From: Costin Iancu (cciancu_at_lbl_dot_gov)
Date: Fri Dec 10 2004 - 10:36:49 PST
Marc, there's a bug when translating quake. The code that triggers is double ds[][4]; ds[1][0] = .. ds[2][0] = ... (index value 0 in the last dimension) Same code appears in a couple of other places. There will be a fix by next week. For the time being, as a workaround i=0; d[1][i] = ... should work If you have other problems, please submit a bug report using our bugzilla page http://mantis.lbl.gov/bugzilla This will make sure we are aware of the problem and we can keep track of the test files. Thanks. Costin On Fri, 2004-12-10 at 09:12, Marc Gonzalez-Sigler wrote: > Hello again, > > 183.equake is a single-file SPECfp2000 benchmark written in C. > http://www-rocq.inria.fr/~gonzalez/quake.c > > Given my previous message, this might not be relevant, but that program > does not compile with upcc on my system. Could someone tell me if it > does on a functional system? > > For reference, I ran: > $ upcc -vv quake.c -lm 1>upcc.out 2>upcc.err > > http://www-rocq.inria.fr/~gonzalez/upcc.out > http://www-rocq.inria.fr/~gonzalez/upcc.err > http://www-rocq.inria.fr/~gonzalez/upcc.both > > (I apologize for my spamming the list.) -- Costin C. Iancu cciancu_at_lbl_dot_gov Future Technologies Group Phone: 510-495-2122 Lawrence Berkeley National Laboratory Fax: 510-486-6900