Hi Veydan, my intent in this email is definitely not to sound abrasive, but the note on the top of this page it says: "This is a work in-progress. Everyone should feel free to extend this page with their experiences to help new users get started."
And then looking at the authors list from the history, I don't recognize any of the names who contributed this particular page. So it's not clear when you say "how did you run this?" that you are talking to any M5 developer in particular especially since spec2k6 workloads aren't in the M5 regression suite. As such, we wouldn't know if any particular benchmark did or didn't work without it being in the regression suite or someone reporting that they got it to work for their studies. The note on the bottom of that page seems to suggest the latter: "If you have any questions, please email to M5 Mailing-List or contact Meng-Ju Wu at [email protected]" If you aren't able to figure it out through that reference, then you'll have to go down the route of implementing any unimplemented behavior yourself. (Btw, if you found this system call in the source code, then how can it be unimplemented?) On Sat, Mar 26, 2011 at 11:24 PM, Veydan Wu <[email protected]> wrote: > I referred to this page as well. I used Alpha CPU. From this page, seems > that M5 SE already support this benchmark, I just wonder why I encountered > such an unimplemented system call error. If I need to implement this system > call myself, I will try it. But how did you run "povray" successfully, given > that I have found this system call in the M5 source code. Thanks. > > On Sat, Mar 26, 2011 at 1:30 PM, Korey Sewell <[email protected]> wrote: >> >> what M5 page are you referring to that says it works? >> >> This page shows some starter info on running spec2k6: >> http://m5sim.org/wiki/index.php/SPEC2006_benchmarks >> >> What CPU model are you running? >> >> Have you tried implementing the system call yourself? You may be able >> to look at how other system calls are defined and form your own >> definition. >> >> On Wed, Mar 23, 2011 at 3:46 PM, Veydan Wu <[email protected]> wrote: >> > Hi, all. I ran spec2006 on M5 Alpha FS mode. The benchmark "povray" >> > encounters an error: >> > fatal: syscall chdir (#12) unimplemented. >> > @ cycle 914691000 >> > [unimplementedFunc:build/ALPHA_SE/sim/syscall_emul.cc, line 82] >> > However, the M5 website says this program runs OK on M5. I can't find >> > the >> > implementation of this system call in the latest m5 source code. Can >> > anyone >> > tell me how to solve this problem? Thanks in advance. >> > -- >> > Regards, >> > >> > Veydan >> > >> > _______________________________________________ >> > m5-users mailing list >> > [email protected] >> > http://m5sim.org/cgi-bin/mailman/listinfo/m5-users >> > >> >> >> >> -- >> - Korey >> _______________________________________________ >> m5-users mailing list >> [email protected] >> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users > > > > -- > Regards, > > Veydan > > _______________________________________________ > m5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/m5-users > -- - Korey _______________________________________________ m5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
