runstat has a carry bug
Bug #1218129 reported by
Andrew Pollock
This bug affects 1 person
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
cronutils (Ubuntu) |
Fix Released
|
Undecided
|
Unassigned | ||
Precise |
Fix Released
|
Undecided
|
Unassigned | ||
Quantal |
Won't Fix
|
Undecided
|
Unassigned | ||
Raring |
Won't Fix
|
Undecided
|
Unassigned | ||
Saucy |
Fix Released
|
Undecided
|
Unassigned |
Bug Description
runstat has a carry bug where it will sometimes report the elapsed time as something like 10.-814381950
It was fixed upstream in https:/
Related branches
To post a comment you must log in.
It's been some time since I've prepared an SRU, so please be gentle.
[Impact]
* runstats produces output that is not a valid number (e.g. 10.-814381950) when the nanosecond of the start time is less than the nanosecond of the end time.
* Justification: Allow runstats to provide correctly formatted output
* This upload fixes the bug by backporting the fix from upstream. (The fix simply handles the case where the value is negative)
[Test Case]
* Due to the precise timing issues that trigger this bug, it is difficult to come up with a trivial test case to reproduce the bug.
[Regression Potential]
* I do not see any potential for regression based on the patch applied.