Kevin,
Thanks a lot for this information.
- Avadh
On 4/20/07, Kevin Moore <kmoore@xxxxxxxxxxx> wrote:
Avadh, Yes, 1024 will trigger TM behavior. You can turn that off in ruby/simics/commands.C in magic_instruction_callback.
--Kevin
On Apr 19, 2007, at 11:58 PM, avadh patel wrote:
> Thanks for quick reply. > I am using MOSI_SMP_bcast (unmodified) without any transactional > memory system. > > I am unaware of any magic calls that inadvertently invoke
> transactions. But i think this must be the case because when I run the > tester it doesn't give any such errors. If this is the case then how > should I deal with it? In my Ocean benchmark compiled i call the
> magic-breakpoint with value of 1024. Do you think this should be > different? > > Thanks in advance. > > - Avadh > > On 4/19/07, Kevin Moore <
kmoore@xxxxxxxxxxx> wrote: >> The transaction manager is a part of LogTM, our transactional >> memory >> system.Are you trying to simulate transactional memory?Are you >> using magic calls that might inadvertently invoke transactions?If
>> you >> are trying to use transactional memory, please use >> MESI/MOESI_SMP_LogTM_directory instead. >> >> --Kevin >> >> >> On Apr 19, 2007, at 10:31 PM, avadh patel wrote:
>> >> > Hi all, >> > >> > I am using MOSI_SMP_bcast protocol and using Crossbar switch network >> > to simulate Ruby and Opal. >> > When I give 'opal0.sim-step
100000000', it gives me the following >> > error: >> > >> > failed assertion 'm_size != 0' at fn const TYPE& >> > Vector<TYPE>::ref(int) const [with TYPE = TransactionManager*] in
>> > ../common/Vector.h:156 >> > >> > As per my knowledge this is because there is some call to use >> > TransactionManager and as I am using MOSI_SMP_bcast it should have
>> any >> > support for TransactionManager. >> > So does anyone has idea why is this happening? >> > I am trying to run Ocean benchmark. >> > >> > Thanks in advance.
>> > >> > - Avadh >> > _______________________________________________ >> > 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. >> > >> >> _______________________________________________
>> 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.
>> > _______________________________________________ > 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. >
_______________________________________________ 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.
|