Merge lp://staging/~spiv/meliae/linux-support into lp://staging/meliae
Proposed by
Andrew Bennetts
| Status: | Merged |
|---|---|
| Approved by: | John A Meinel |
| Approved revision: | 143 |
| Merged at revision: | 150 |
| Proposed branch: | lp://staging/~spiv/meliae/linux-support |
| Merge into: | lp://staging/meliae |
| Diff against target: |
50 lines (+21/-1) (has conflicts) 2 files modified
meliae/perf_counter.py (+16/-0) track_memory.py (+5/-1) Text conflict in meliae/perf_counter.py |
| To merge this branch: | bzr merge lp://staging/~spiv/meliae/linux-support |
| Related bugs: |
| Reviewer | Review Type | Date Requested | Status |
|---|---|---|---|
| John A Meinel | Approve | ||
|
Review via email:
|
|||
Description of the change
I wrote this before the trunk had a linux implementation of get_memory, but I think my implementation is a bit better: I found that in practice the Vm* fields can be absent sometimes (I think between process exit but before it is reaped?), and so meliae should take to care to avoid those late None results from overriding actual values from when the process had Vm* values available.
To post a comment you must log in.
Preview Diff
[H/L] Next/Prev Comment, [J/K] Next/Prev File, [N/P] Next/Prev Hunk
The new code returns (None, None) instead of just None, so that it can handle the process not having a value. I think after discussing it, we also prefer VmSize to VmHWM. We can certainly revisit it later if we change our mind.