aboutsummaryrefslogtreecommitdiff
path: root/modules/benchmark
AgeCommit message (Collapse)Author
2019-12-26bench results: add phys mem if knownBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2019-10-24sysbench memory tweaksBurt P
* Use a larger sample size for sysbench 1.0+ and 64-bit * Store pointer size with bench results * Re-enable single thread memory benchmark Use a larger sample size where available. The small size was chosen because the 32-bit ARM sysbench 0.4x in raspbian can't do more than about ~4G, but the problem is that much more powerful machines burn through that very quickly. The result is in MiB/s so it should still be comparable, but the results should be more stable. Noticed with Ryzen that multi-thread varies significantly based on what threads are used, but benchmark doesn't really have control over that. Signed-off-by: Burt P <pburt0@gmail.com>
2019-08-22various: entry hide flag, fix text format report, fix key_*() funcsBurt P
* fix key_is_highlighted() and key_wants_details() so that they don't look for flags in the label. * fix -r -f text, g_hash-related warnings from icon cache * add MODULE_FLAG_HIDE flag to hide module entries instead of using the '#' at the beginning of string hack, which didn't work everywhere, and screwed up translated strings. * hide GPU Drawing benchmark See: https://github.com/lpereira/hardinfo/issues/329 Signed-off-by: Burt P <pburt0@gmail.com>
2019-08-14sysbench: minor fixesBurt P
idk what. Signed-off-by: Burt P <pburt0@gmail.com>
2019-08-14benches: fewer, perhaps the more meaningful, sysbenches to reduce clutterBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2019-08-14benches: simplify bench definition a bit moreBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2019-08-14benches: fix titlesBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2019-08-14sysbench: try and work with version 0.xBurt P
formerly required 1.0.15. Signed-off-by: Burt P <pburt0@gmail.com>
2019-08-14bench_results: escape strings in conf linesBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2019-08-14bench_results: add machine_data_versionBurt P
For now 0, but planned changes. Signed-off-by: Burt P <pburt0@gmail.com>
2019-08-14benchmark: add sysbench cpuBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2019-08-14Add wrapper for Alexey Kopytov's sysbench, memory benchmarksBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2019-08-14Benchmark: fixes, user_note, verifiable test dataBurt P
* fix zlib display order * fix cryptohash MiB/s calculation * revision and params for other benchmarks * allow attaching user note to bench result with -u * don't inlcude the new result value bits if they are empty/invalid in bench_value_to_str(). bench_value_from_str() doesn't need to be modified. * bench_results: clean old result cpu name for x86 * use problem_marker() from dmi_memory to mark old version bench results * benchmark: verifiable test data size and content - The test data benchmark.data is stored in a file that could be edited to change the size or content. /* to guarantee size */ gchar *get_test_data(gsize min_size); /* to checksum content */ char *md5_digest_str(const char *data, unsigned int len); Signed-off-by: Burt P <pburt0@gmail.com>
2019-07-14Fix buffer overflow in fbenchLeandro Pereira
2019-07-10benchmark: add extra fields to result for revision and informationBurt P
* add an integer revision field * add an extra information string field * zlib benchmark: new revision[1] is 2, so that old results are obviously old, and now the zlib version string will be stored in extra information. [1] as of 6a8e19a14305079b03e45eeb0580a45104f300dd Signed-off-by: Burt P <pburt0@gmail.com>
2019-06-29benchmark/zlib: new version of zlib benchmarkBurt P
The new version uses a fixed time (7s) and the results are much more consistent than the previous version. A couple results for the new version, one "low" and one "high", replace the old results, which were quite strange. Signed-off-by: Burt P <pburt0@gmail.com>
2019-06-27GTK3/Benchmarks disable guibench onlyJuro Bystricky
GTK3 build inadvertably disabled all benchmarking. At present, the GTK3 build intentionally avoids building of "guibench" due to non-trivial changes from GTK2 to GTK3. However, all other benchmarks were disabled as well. This was not by design, as non-gui benchmarks have no GTK3 dependencies. The reason benchmarks were diasbled (did not show up in the menus) was that the code for "benchark_gui" still referenced the missing "guibench" routine. This routine was not present in the "benchmark.so", so the library would not load due to the missing entry point. This patch addresses the issue. All benchmarks (except guibench) show/work fine with GTK3. Signed-off-by: Juro Bystricky <juro.bystricky@intel.com>
2019-06-20move devices/memory to computer/memory usageBurt P
As discussed in https://github.com/lpereira/hardinfo/issues/345 Signed-off-by: Burt P <pburt0@gmail.com>
2018-12-02bench/zlib: minor fix, changes results significantlyBurt P
(#298) W: hardinfo uninitialized-variable /home/abuild/rpmbuild/BUILD/hardinfo-20181022T194523/modules/benchmark/zlib.c:28 This change significantly impacts the bench result. All old results should be discarded, IMO. I think I will also attempt to remake this benchmark in a similar way to the blowfish benchmark. See PR #293. Signed-off-by: Burt P <pburt0@gmail.com>
2018-12-02bench_results: minor fixBurt P
(#298) E: hardinfo no-return-in-nonvoid-function /home/abuild/rpmbuild/BUILD/hardinfo-20181022T194523/modules/benchmark/bench_results.c:137 Signed-off-by: Burt P <pburt0@gmail.com>
2018-11-04fix blowfish bench sort orderBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2018-11-04benchmark/blowfish: new version of blowfish benchmarkBurt P
The new version uses a fixed time and provides variants for single-thread, multi-thread, and multi-core. A few results are included. Signed-off-by: Burt P <pburt0@gmail.com>
2018-10-02bench_results: bug fix for cpus with mixed configBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2018-05-27benchmark results: add gpu to resultsBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2018-03-12Do not use hardcoded path for iconLeandro Pereira
Use just the symbolic name "hardinfo", allowing for icon packs to provide alternative icons. Fixes #227
2018-02-27bench_results: store complete bench_valueBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2018-02-27benchmarks: split individual test funcs out of benchmark.c frameworkBurt P
The idea here is to make adding/removing benchmark tests easier by having benchmark.c be the framework and all the functioned needed for individual benchmarks live in benches.c. Signed-off-by: Burt P <pburt0@gmail.com>
2018-02-27Benchmark results: store threads usedBurt P
Benchmark results store actual number of threads used by benchmark when it was run. Previously, results assumed all available threads were used. Examples: * CPU Fib only uses one * FPU FFT uses 4, 2, or 1 * N-Queens uses 10, 5, 2, or 1 Signed-off-by: Burt P <pburt0@gmail.com>
2018-02-27struct simple_machine renamed to bench_machineBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2017-09-03bench results: Show unknown instead of 0 kB for old resultsBurt P
Other minor tweaks. Signed-off-by: Burt P <pburt0@gmail.com>
2017-08-21bench_results: store int processors/cores/threadsBurt P
In the future, won't have to rely on a translated string. Signed-off-by: Burt P <pburt0@gmail.com>
2017-08-21Add OpenGL Renderer to benchmark result detailsBurt P
For the GPU Drawing benchmark at least, it should give some idea of the graphics capabilities of the machine. This is the only data hardinfo currently has about that. Maybe in the future, more advanced GPU information could be given. Signed-off-by: Burt P <pburt0@gmail.com>
2017-08-21bench_results: fix language translation issues for stored stringsBurt P
* cpu_config string is translated as loaded * if board or cpu_desc is missing, it is left null, stored empty, and translated to (Unknown) only when displayed. Signed-off-by: Burt P <pburt0@gmail.com>
2017-08-21bench_result: changes suggested by LAFPBurt P
* hide bench/name and result * expand note for old results The old version is kept because it is useful for future development. Signed-off-by: Burt P <pburt0@gmail.com>
2017-08-21bench_results: fix cpu clock from id stringBurt P
Old versions of hardinfo used current frequency instead of max frequency. If the cpu id string had a clock rate in it, and it is significantly more than thre reported frequency in the benchmark result, use that. This feature exists in an earlier commit but it was broken. Signed-off-by: Burt P <pburt0@gmail.com>
2017-08-21Benchmark result detailsBurt P
* More information can be saved in benchmark.conf, while still being compatible with older versions. * Selecting a result give additional information about the result and the machine the produced it.
2017-07-02Add zlib benchmarkLeandro Pereira
2017-04-09Update fbench.cTotalCaesar659
2017-04-09Update fftbench.cTotalCaesar659
2016-12-30Free whole MD5 contextLeandro Pereira
2016-07-30Free result of lup_solve() when performing FFT benchmarkLeandro Pereira
2012-01-27Fix weird results in benchmark valuesLeandro Pereira
gcc was (correctly) assuming that benchmark_parallel_for() returned an integer, which was being to a double and messing up results. Added the missing prototype. Also, populate bench_results array with sane values on module init. Some other minor cleanups in benchmark code.
2010-05-03Remove outdated arch directoryLeandro A. F. Pereira
2010-05-03Move files from hardinfo2 to root.Leandro A. F. Pereira
generated by cgit v1.2.3 (git 2.39.1) at 2025-02-05 23:00:16 +0000