ehoogeveen suggested this. It only makes sense when used with --sample-below=1, so maybe it should give a warning if you use it without that.
Technically I guess if a small allocation shows up so often that even with the sampling it's one of the top entries, that's more reason to look into it :) I was thinking that it could use cmpByUsableSize  on a tie, though for large block counts a tie might not be very likely.  http://dxr.mozilla.org/mozilla-central/source/memory/replace/dmd/dmd.py?from=dmd.py#134
Created attachment 8536285 [details] [diff] [review] Add --sort-by=num-blocks option to dmd.py This required tweaking the "sort-by" tests a little.
Attachment #8536285 - Flags: review?(continuation)
Attachment #8536285 - Flags: review?(continuation) → review+
Status: ASSIGNED → RESOLVED
Last Resolved: 3 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla37
You need to log in before you can comment on or make changes to this bug.