sync to latest stress-ng as it contains a wide range of small fixes
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
stress-ng (Ubuntu) |
Fix Released
|
Medium
|
Colin Ian King |
Bug Description
The latest stress-ng contains a range of minor fixes that should be pulled into the latest bionic release:
stress-fcntl: use | rather than |= to get new_flags
stress-tmpfs: clear mnts and check if mnts[i] is null
stress-sync-file: fix incorrect placement of break statement
stress-af-alkg: check for NULL cmsg pointers
stress-pipe: ensure no uninitialized data is written to pipe
test/test-
test/test-
stress-mmapfailed: check for null return from mremap
stress-resources: don't open on a null slavename
stress-sysinfo: zero the mnts array
stress-cpu: make loop index in jenkin func size_t rather than uint8_t
stress-vm: don't loop forever in child sleep loop
ignite-cpu: don't wait for parent to kill loop, add timeout check too
Double number of stressor processes up to 8192 (for larger systems)
Changed in stress-ng (Ubuntu): | |
status: | New → In Progress |
importance: | Undecided → Medium |
assignee: | nobody → Colin Ian King (colin-king) |
Changed in stress-ng (Ubuntu): | |
status: | In Progress → Fix Committed |
This bug was fixed in the package stress-ng - 0.09.24-1
---------------
stress-ng (0.09.24-1) unstable; urgency=medium
* Makefile: bump version posix-memalign. c: free allocated buffer fanotify. c: free buffer
* Misc small fixes detected by static analysis (LP: #1762848)
* stress-fcntl: use | rather than |= to get new_flags
* stress-tmpfs: clear mnts and check if mnts[i] is null
* stress-sync-file: fix incorrect placement of break statement
* stress-af-alkg: check for more NULL cmsg pointers
* stress-pipe: ensure no uninitialized data is written to pipe
* stress-fcntl: use | rather than |= to get new_flags
* test/test-
* test/test-
* stress-mmapfailed: check for null return from mremap
* stress-af-alkg: check for NULL cmsg
* stress-resources: don't open on a null slavename
* stress-sysinfo: ensure mnts array is zero'd
* stress-cpu: make loop index in jenkin func size_t to fix overflow
* stress-vm: don't loop forever in child sleep loop
* ignite-cpu: don't wait for parent to kill loop, add timeout
check too
* Double number of stressor processes up to 8192
* Add stream-index indexing option
* Make some settings global to fix args confustion
-- Colin King <email address hidden> Wed, 11 Apr 2018 09:58:44 +0100