Date: | Tue, 24 Jul 2007 11:56:30 -0500 (CDT) |
---|---|
From: | Mike Marty <mikem@xxxxxxxxxxx> |
Subject: | Re: [Gems-users] sequencer request at time 1 |
That is a good approach, but you probably didn't correctly modify the
MOESI_CMP_directory-L1cache.sm to account for two different CacheMemory
objects. See how the SMP protocols do it (i.e.,
MOESI_SMP_directory-cache.sm)
--Mike Hi, I have renamed the L2 cache controller of MOESI_CMP_directory as L3 cache and created a combined L1/L2 cache controller to have private L1 and L2 caches. I'm not sure if I have successfully done all the necessary changes though. I get a sequencer deadlock error at time 50001, which means the request was made at time 1 (deadlock threashold=50000). What is the first request made by the sequencer? What could the problem be related to? The network configuration? The message buffers? Thank you. Dave ____________________________________________________________________________________ Choose the right car based on your needs. Check out Yahoo! Autos new Car Finder tool. http://autos.yahoo.com/carfinder/ _______________________________________________ Gems-users mailing list Gems-users@xxxxxxxxxxx https://lists.cs.wisc.edu/mailman/listinfo/gems-users Use Google to search the GEMS Users mailing list by adding "site:https://lists.cs.wisc.edu/archive/gems-users/" to your search. |
[← Prev in Thread] | Current Thread | [Next in Thread→] |
---|---|---|
|
Previous by Date: | Re: [Gems-users] LxCACHE_TRANSITIONS_PER_RUBY, Mike Marty |
---|---|
Next by Date: | Re: [Gems-users] sequencer request at time 1, Dave Z. |
Previous by Thread: | [Gems-users] sequencer request at time 1, Dave Z. |
Next by Thread: | Re: [Gems-users] sequencer request at time 1, Dave Z. |
Indexes: | [Date] [Thread] |