I used GEMS-2.1 and simics-2.2.19 to run deque-tm.
But when I use ruby+opal module to execute it , I miss the
transcational memory information. All "Transactional Memory Stats" are
zero. (Show below)
I use opal to get the WATTCH power information. In other words, I want
to execute the transactional memory program and get WATTCH power
information.
Are there any solutions?
Please help me! Thanks
Transactional Memory Stats:
------- xact --------
xact_size_dist: [binsize: 1 max: 0 count: 0 average: NaN |standard
deviation: NaN | 0 ]
xact_instr_count: [binsize: 1 max: 0 count: 0 average: NaN |standard
deviation: NaN | 0 ]
xact_time_dist: [binsize: 1 max: 0 count: 0 average: NaN |standard
deviation: NaN | 0 ]
xact_log_size_dist: [binsize: 1 max: 0 count: 0 average: NaN |standard
deviation: NaN | 0 ]
xact_read_set_size_dist: [binsize: 1 max: 0 count: 0 average: NaN
|standard deviation: NaN | 0 ]
xact_write_set_size_dist: [binsize: 1 max: 0 count: 0 average: NaN
|standard deviation: NaN | 0 ]
xact_overflow_read_lines_dist: [binsize: 1 max: 0 count: 0 average:
NaN |standard deviation: NaN | 0 ]
xact_overflow_write_lines_dist: [binsize: 1 max: 0 count: 0 average:
NaN |standard deviation: NaN | 0 ]
xact_overflow_read_set_size_dist: [binsize: 1 max: 0 count: 0 average:
NaN |standard deviation: NaN | 0 ]
xact_overflow_write_set_size_dist: [binsize: 1 max: 0 count: 0
average: NaN |standard deviation: NaN | 0 ]
xact_miss_load_dist: [binsize: 1 max: 0 count: 0 average: NaN
|standard deviation: NaN | 0 ]
xact_miss_store_dist: [binsize: 1 max: 0 count: 0 average: NaN
|standard deviation: NaN | 0 ]
xact_nacked: 0
xact_retries: [binsize: 1 max: 0 count: 0 average: NaN
|standard deviation: NaN | 0 ]
xact_abort_delays: [binsize: 1 max: 0 count: 0 average: NaN |standard
deviation: NaN | 0 ]
xact_aborts: 0
xact_writebacks: 0
xact_extra_wbs: 0
xact_handler_startup_delay: -1
xact_handler_per_block_delay: -1
xact_inferred_aborts: 0
2008/4/22 Luke Yen <lyen@xxxxxxxxxxx <mailto:lyen@xxxxxxxxxxx>>
The version of Opal for the GEMs release does not support running TM
programs.
However with alot of non-trivial work it can be made to run TM
programs.
Most of the complexity comes in how to track transactional
loads/stores
for correct conflict detection, and how to handle stalls & aborts.
Luke
On Mon, 21 Apr 2008, cge@xxxxxxxx <mailto:cge@xxxxxxxx> wrote:
> Hi,
> Is there anyway to run transactional momory workloads in
> MESI_CMP_filter_directory with Opal ?
>
> I know this was not supported before due to the conflicts
between magic
> calls (such as BEGIN_TRANSACTION and COMMIT_TRANSACTION)and Opal.
>
> Thanks,
> Ken
>
>
> _______________________________________________
> Gems-users mailing list
> Gems-users@xxxxxxxxxxx <mailto: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 <mailto: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.