Erik Tomusk
2012-10-29 11:07:48 UTC
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/1523/
-----------------------------------------------------------
Review request for Default.
Description
-------
Changeset 9355:eb19f07b2afc
---------------------------
TournamentBP: Fix some bugs with table sizes and counters
globalHistoryBits, globalPredictorSize, and choicePredictorSize are decoupled.
globalHistoryBits controls how much history is kept, global and choice
predictor sizes control how much of that history is used when accessing
predictor tables. This way, global and choice predictors can actually be
different sizes, and it is no longer possible to walk off the predictor arrays
and cause a seg fault.
There are now individual thresholds for choice, global, and local saturating
counters, so that taken/not taken decisions are correct even when the
predictors' counters' sizes are different.
Diffs
-----
src/cpu/pred/tournament.hh e71f71ce233a
src/cpu/pred/tournament.cc e71f71ce233a
Diff: http://reviews.gem5.org/r/1523/diff/
Testing
-------
Works with ARM SE quick regression (except for dram target, which fails
with and without this patch).
Thanks,
Erik Tomusk
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/1523/
-----------------------------------------------------------
Review request for Default.
Description
-------
Changeset 9355:eb19f07b2afc
---------------------------
TournamentBP: Fix some bugs with table sizes and counters
globalHistoryBits, globalPredictorSize, and choicePredictorSize are decoupled.
globalHistoryBits controls how much history is kept, global and choice
predictor sizes control how much of that history is used when accessing
predictor tables. This way, global and choice predictors can actually be
different sizes, and it is no longer possible to walk off the predictor arrays
and cause a seg fault.
There are now individual thresholds for choice, global, and local saturating
counters, so that taken/not taken decisions are correct even when the
predictors' counters' sizes are different.
Diffs
-----
src/cpu/pred/tournament.hh e71f71ce233a
src/cpu/pred/tournament.cc e71f71ce233a
Diff: http://reviews.gem5.org/r/1523/diff/
Testing
-------
Works with ARM SE quick regression (except for dram target, which fails
with and without this patch).
Thanks,
Erik Tomusk