Merge #424665 from ~alanec/plainbox-provider-checkbox:sngt_threads

3 files changed, 26 insertions, 9 deletions

diff --git a/bin/stress_ng_test.py b/bin/stress_ng_test.py
index 35aad2d..5f3b9a5 100755
--- a/bin/stress_ng_test.py
+++ b/bin/stress_ng_test.py
@@ -58,12 +58,14 @@ class StressNg():
 stressors,
 wrapper_timeout,
 sng_timeout,
+ thread_count,
 extra_options=""):
 
 self.stressors = stressors
 self.wrapper_timeout = wrapper_timeout
 self.sng_timeout = sng_timeout
 self.extra_options = extra_options
+ self.thread_count = thread_count
 self.results = ""
 self.returncode = 0
 
@@ -71,10 +73,12 @@ class StressNg():
 """Run a stress-ng test, storing results in self.results."""
 
 stressor_list = "--" + " 0 --".join(self.stressors)
- command = "stress-ng --aggressive --verify --timeout {} {} {} 0". \
+
+ command = "stress-ng --aggressive --verify --timeout {} {} {} {}". \
 format(self.sng_timeout,
 self.extra_options,
- stressor_list)
+ stressor_list,
+ self.thread_count)
 time_str = time.strftime("%d %b %H:%M", time.gmtime())
 if len(self.stressors) == 1:
 print("{}: Running stress-ng {} stressor for {:.0f} seconds...".
@@ -217,15 +221,18 @@ def stress_memory(args):
 # Constant-run-time stressors -- run them for the same length of time on
 # all systems....
 crt_stressors = ['bsearch', 'context', 'hsearch', 'lsearch', 'matrix',
- 'memcpy', 'null', 'pipe', 'qsort', 'stack', 'str',
+ 'memcpy', 'null', 'pipe', 'qsort', 'str',
 'stream', 'tsearch', 'vm-rw', 'wcs', 'zero', 'mlock',
 'mmapfork', 'mmapmany', 'mremap', 'shm-sysv',
 'vm-splice']
+
 if num_numa_nodes() > 1:
 crt_stressors.append('numa')
 
 # Variable-run-time stressors -- run longer on systems with more RAM....
- vrt_stressors = ['malloc', 'mincore', 'vm', 'bigheap', 'brk', 'mmap']
+ vrt_stressors = ['malloc', 'mincore', 'vm', 'mmap']
+ # Low-thread-count stressors -- throttle to >8 threads...
+ ltc_stressors = ['stack', 'bigheap', 'brk']
 
 est_runtime = len(crt_stressors) * args.base_time + \
 len(vrt_stressors) * vrt
@@ -234,12 +241,22 @@ def stress_memory(args):
 for stressor in crt_stressors:
 test_object = StressNg(stressors=stressor.split(),
 sng_timeout=args.base_time,
- wrapper_timeout=args.base_time*2)
+ wrapper_timeout=args.base_time * 2,
+ thread_count=0)
 retval = retval | test_object.run()
 print(test_object.results)
 for stressor in vrt_stressors:
- test_object = StressNg(stressors=stressor.split(), sng_timeout=vrt,
- wrapper_timeout=vrt*2)
+ test_object = StressNg(stressors=stressor.split(),
+ sng_timeout=vrt,
+ wrapper_timeout=vrt * 2,
+ thread_count=0)
+ retval = retval | test_object.run()
+ print(test_object.results)
+ for stressor in ltc_stressors:
+ test_object = StressNg(stressors=stressor.split(),
+ sng_timeout=vrt,
+ wrapper_timeout=vrt * 2,
+ thread_count=8) # throttle to 8 threads
 retval = retval | test_object.run()
 print(test_object.results)
 if my_swap is not None and args.keep_swap is False:
@@ -271,7 +288,7 @@ def stress_disk(args):
 est_runtime = len(disk_stressors) * args.base_time
 print("Using test directory: '{}'".format(test_disk.test_dir))
 print("Estimated total run time is {:.0f} minutes\n".
- format(est_runtime/60))
+ format(est_runtime / 60))
 retval = 0
 if not args.simulate:
 for stressor in disk_stressors:
@@ -279,7 +296,7 @@ def stress_disk(args):
 "--hdd-opts dsync --readahead-bytes 16M -k"
 test_object = StressNg(stressors=stressor.split(),
 sng_timeout=args.base_time,
- wrapper_timeout=args.base_time*5,
+ wrapper_timeout=args.base_time * 5,
 extra_options=disk_options)
 retval = retval | test_object.run()
 print(test_object.results)
diff --git a/data/.DS_Store b/data/.DS_Store
new file mode 100644
index 0000000..f457547
--- /dev/null
+++ b/data/.DS_Store
diff --git a/units/.DS_Store b/units/.DS_Store
new file mode 100644
index 0000000..34cdc7c
--- /dev/null
+++ b/units/.DS_Store