aboutsummaryrefslogtreecommitdiff
path: root/bench.py
Commit message (Collapse)AuthorAgeFilesLines
* add option to exclude benchmarksFlorian Fischer2019-06-241-0/+4
|
* give each allocator a unique colorFlorian Fischer2019-06-241-0/+11
|
* rework allocator definitions #2Florian Fischer2019-06-241-15/+32
| | | | | | | | bench.py no evals only if argument to -a is file path. Otherwise it will check if "arg".py is found in src/allocators/ then it imports it. Collection definitions must export a iterable member called allocators. Allocator definitions must export a member named "arg".
* only delete during analyse created stats if they existFlorian Fischer2019-05-111-1/+1
|
* fix some hints from flake8Florian Fischer2019-05-081-8/+8
|
* move list of available benchmarks to src.globalvars.benchmarksFlorian Fischer2019-04-291-6/+4
| | | | fix analyse
* fix analyseFlorian Fischer2019-04-281-1/+1
|
* start moving allocator definitions to src/allocatorsFlorian Fischer2019-04-281-1/+1
|
* use special allocator "malt" during analyse to run the benchmark only onceFlorian Fischer2019-04-281-10/+31
|
* improve descriptive statistics, csv-export and add general barplot functionFlorian Fischer2019-04-131-1/+0
|
* add analyze and server_benchmark featureFlorian Fischer2019-04-021-16/+23
| | | | | | | | | | | | | | | | | | | | | | --analyze uses malt to trace the benchmarks behavior. It uses the run loop but the obtained results are not stored Benchmark.server_benchmark is used if only a server is started for each allocator and clients are used to measure its performance in the run loop. If server_benchmark is set to True the cmds are run with the system default allocator. Misc changes: * The global environment is no longer changed. Instead a custom env dict is passed to suprocesses containing LD_PRELOAD. * Failing cmds no longer skip the whole benchmark instead they now skip the malfunctioning allocator. * Fix default title in plot_single_arg an analyse run are not stored
* move benchmark definitions into src/benchmarksFlorian Fischer2019-03-251-2/+4
| | | | bench now loads all *.py files from src/benchmarks as benchmarks
* alwys reset LD_PRELOAD even when a exception occursFlorian Fischer2019-03-221-1/+10
|
* use exceptions in mysql benchmarkFlorian Fischer2019-03-211-6/+7
|
* use exceptions to indicate an errorFlorian Fischer2019-03-211-29/+34
|
* collect facts on demandFlorian Fischer2019-03-191-0/+3
| | | | | this fixes a exception because of the missing ccinfo file when src.facter is imported before the build of allocbench
* fix time recordingFlorian Fischer2019-03-101-1/+1
|
* remove second isoformat(timespec=) callFlorian Fischer2019-03-081-4/+6
|
* don't use isoformat(timespec=) for python3 < 3.6Florian Fischer2019-03-081-1/+4
|
* Don't print running if -r 0 is setFlorian Fischer2019-03-081-1/+0
|
* record start and end time in factsFlorian Fischer2019-03-081-1/+5
|
* save facts and rstore them on loadFlorian Fischer2019-03-061-3/+25
|
* add allocators to globalvars; kill mysql server on exit; save libc versions ↵Florian Fischer2019-03-061-6/+10
| | | | of requirements
* only import benchmarks if they are usedFlorian Fischer2019-03-061-2/+3
|
* rework rebuild mechanism to use timestamps instead of pickle stringsFlorian Fischer2019-03-051-3/+4
|
* use new verbosity system in allocator.pyFlorian Fischer2019-03-051-2/+2
|
* add new verbosity and colored outputFlorian Fischer2019-03-051-20/+53
|
* save if no summary flag is setFlorian Fischer2019-02-221-1/+1
|
* don't skip complete run if one prepare failsFlorian Fischer2019-02-221-1/+1
|
* Update Readme.mdFlorian Fischer2019-02-101-2/+2
|
* s/target/allocator/ and remove analyse featureFlorian Fischer2019-02-081-19/+18
| | | | Use systemwide installed allocators by default
* rework build system #2: call make before executing any benchmarkFlorian Fischer2019-02-021-1/+10
| | | | | | Make scans the project directory for Makefiles and executes them. The targets should be build by the new targets/Makefile. All included Makefiles are now "quiet" by default.
* add custom target optionFlorian Fischer2019-01-221-0/+11
| | | | | the custom target definitions must be a valid python script exporting a global dictionary "targets".
* add load path option and don't fail if result directories exists alreadyFlorian Fischer2019-01-221-4/+10
|
* save all results by default in results/hostname/timestampFlorian Fischer2019-01-221-8/+24
|
* add license switchFlorian Fischer2019-01-151-1/+7
|
* move source code to src/Florian Fischer2019-01-151-9/+3
|
* fix perf parsing and add new plot functions and use themFlorian Fischer2018-09-141-1/+1
|
* add usefull chattymalloc support and --nolibmemusage flagFlorian Fischer2018-09-141-3/+4
| | | | | | improve bench.py improve process_stdout hook -> process_output rename perf_cmd -> measure_cmd
* cleanup bench.pyFlorian Fischer2018-09-141-4/+2
|
* add larson's server benchmarkFlorian Fischer2018-09-101-1/+2
|
* add dj's workloadsFlorian Fischer2018-09-061-1/+2
|
* only prepare or cleanup if we have toFlorian Fischer2018-08-281-5/+6
|
* intruduce new way of handling analyse and cleanup mysqlFlorian Fischer2018-08-241-7/+8
| | | | chattymalloc writes now to chattymalloc.data
* pass summary directory to benchmarksFlorian Fischer2018-08-211-1/+1
|
* only analyse alloc sizes for mysqlFlorian Fischer2018-08-201-1/+1
|
* add summary directory optionFlorian Fischer2018-08-201-0/+5
|
* add basic analyse to mysqlFlorian Fischer2018-08-131-1/+8
|
* pass verbose to runFlorian Fischer2018-08-111-1/+1
|
* fix nosum optionFlorian Fischer2018-08-061-1/+1
|
generated by cgit v1.2.3 (git 2.25.1) at 2026-04-25 11:01:15 +0000