mem: Adjust cache queue reserve to more conservative values
Review Request #3347 - Created Feb. 24, 2016 and submitted
Information | |
---|---|
Andreas Hansson | |
gem5 | |
default | |
Reviewers | |
Default | |
Changeset 11368:2a10e5484eae --------------------------- mem: Adjust cache queue reserve to more conservative values The cache queue reserve is there as an overflow to give us enough headroom based on when we block the cache, and how many transactions we may already have accepted before actually blocking. The previous values were probably chosen to be "big enough", when we actually know that we check the MSHRs after every single allocation, and for the write buffers we know that we implicitly may need one entry for every outstanding MSHR.
Seems fine, though:
1. It's weird that we're passing in a 1 to mean "no reserve entries" and "p->mshrs + 1" to mean "as may reserve entries as MSHRs". I realize that's an issue with the semantics of the parameter, and outside the scope of this patch, but if we're ever going to clean that up now would be a good time.
2. it would be nice to have a forward pointer comment by the initializers so people know there's a more detailed comment below
Description: |
|
||||||||||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Diff: |
Revision 2 (+12 -6) |
Description: |
|
||||||||||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Diff: |
Revision 3 (+12 -6) |
Ship It!