aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2018-02-27x86,arm: cpu clocks list in package informationBurt P
Show actual clocks where cores or threads share a clock. Ex: x86 SMT each core has one clock shared by both threads. Ex: BCM2837 has one clock for all four cores. Signed-off-by: Burt P <pburt0@gmail.com>
2018-02-27Fix benchmark_parallel_for() to use correct number of threadsBurt P
If (end - start) / threads left a remainder, benchmark_parallel_for() would start an extra thread for the leftover elements. Now, it has the last thread process the few extra elements. Added a note: benchmark_parallel_for(): element [start] included, but [end] is excluded. callback(): expected to processes elements [start] through [end] inclusive. Signed-off-by: Burt P <pburt0@gmail.com>
2018-02-27hardinfo: fix typoBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
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.conf: remove "new" results that may be wrongBurt P
Results assumed all threads were used for each benchmark. Signed-off-by: Burt P <pburt0@gmail.com>
2018-02-27benchmark: New benchmark typesBurt P
New benchmark types: * benchmark_crunch_for(): call function repeatedly for a number of seconds; result is number of completions. * benchmark_parallel(): one call for each available thread up to n_threads; result is sum of return values. Tweaks: * Store return values from callback via benchmark_parallel_for_dispatcher() Signed-off-by: Burt P <pburt0@gmail.com>
2018-02-27Add cli param to format result for -bBurt P
For running benchmarks from the command line, add a new param to specify the output format: -g "conf" gives a line suitable for benchmark.conf -g "shell" gives the complete "moreinfo" shell data -g "short" (or nothing) gives the bench_value string as before 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>
2018-02-22cpu_util: bug fixBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2018-02-22report.c: don't try to use column values that weren't provided (html version)Burt P
Like 96980fbf494d9c453c6b558967c90c55e41abe2d, but the html version of that function. Fixes https://github.com/lpereira/hardinfo/issues/217 Signed-off-by: Burt P <pburt0@gmail.com>
2018-02-09x86: metldown and spectre bug flagsBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2018-02-09x86: increase the cpuinfo read buffer sizeBurt P
The flags list is often more than 512 bytes long these days. It was being truncated. Signed-off-by: Burt P <pburt0@gmail.com>
2018-02-09Add Intel cpu_insecure bug workaround flagBurt P
cpu_insecure = page table isolation is being used to work around a security vulnerability in Intel x86 CPUs. Signed-off-by: Burt P <pburt0@gmail.com>
2017-12-09x86: add summary of caches to package informationBurt P
Show actual physical caches by counting only unique references from each "cpu" (hardware thread). Signed-off-by: Burt P <pburt0@gmail.com>
2017-12-09report.c: don't try to use colum values that weren't providedBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2017-12-09cpubits_max() fixBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2017-12-09Build debug builds with address sanitizerLeandro Pereira
2017-11-29x86 CPU flags fix; show socket:core and thread in proc listBurt P
* Bug fix: Some flags are just an index, for example power management might have "[13] [14]" as flags. This looks like a new section to hardinfo shell and it truncates the CPU information there. * Show the Socket:Core and thread in the processor list, for x86 only right now. In the future, the idea is to show only one line for each core, and list the threads on that core, where currently, there is one line for each thread. Signed-off-by: Burt P <pburt0@gmail.com>
2017-11-06Updated Russian translationTotalCaesar659
2017-11-02Computer/Display: Add X Display NameBurt P
A field exists in struct _DisplayInfo, but it wasn't being filled or shown. Perhaps fix: https://github.com/lpereira/hardinfo/issues/181 Signed-off-by: Burt P <pburt0@gmail.com>
2017-10-29Fixed possible memory leak (#201)Roman Kalashnikov
Warnings, found using PVS-Studio: hardinfo/modules/devices/usb.c 397 err V773 The function was exited without releasing the 'temp' pointer. A memory leak is possible. hardinfo/modules/computer/uptime.c 33 err V773 The function was exited without releasing the 'ui' pointer. A memory leak is possible.
2017-10-09Fixed compilation warning when building on GTK3TotalCaesar659
2017-09-28Fix more compilation warnings when building with GTK+3TotalCaesar659
2017-09-26shell: select Computer Summary at startBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2017-09-21Russian translation updateTotalCaesar659
2017-09-21Ensure report saving dialog buttons are translatableTotalCaesar659
2017-09-21Do not use deprecated GtkDialog method when building for GTK+ 3TotalCaesar659
2017-09-10Remove the appended "Module" from about module dialog titleBurt P
The string is constructed one word at a time, "About", X, "Module" and it ends up being very strange in languages other than English. Cutting out the module makes it a little better, and not much is lost. A module could add "Module" to it's title, and then at least that string can be translated as a complete unit. Signed-off-by: Burt P <pburt0@gmail.com>
2017-09-10Fix compilation warning when benchmarkingTotalCaesar659
2017-09-06Fix compilation warningTotalCaesar659
2017-09-04Build fix after 8913af3eLeandro Pereira
Fixes #188.
2017-09-04Add prototype for processor_frequency_desc()Leandro Pereira
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>
2017-09-03Do not use deprecated GtkStockTotalCaesar659
GtkStock has been deprecated in GTK+ 3.0, so use strings that work in both GTK+ 2 and GTK+ 3.
2017-09-03Conditionally compile PowerPC and ARM code while summarizing DTLeandro Pereira
Also, use properly-named variables and plug a memory leak.
2017-09-03Clean up device tree codeLeandro Pereira
- No need for a global dtr variable: just pass it around as parameters - Internal functions marked static
2017-09-02Updated Russian translationTotalCaesar659
2017-08-28ARM: Add some mediatek SOCsBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2017-08-27Show motherboard version again after DMI refactoringLeandro Pereira
2017-08-27dmi: check for null from dmi_get_str() in dmi_chassis_type_str()Burt P
Fix a crash when there is no DMI available. Signed-off-by: Burt P <pburt0@gmail.com>
2017-08-27Fix 802.11 signal to bar conversionLeandro Pereira
Just steal the code from GNOME shell: https://github.com/GNOME/gnome-shell/blob/9cc143aaee8aa7c33d7815f744f8dc9e958bddf8/js/ui/status/network.js#L89-L98
2017-08-27dmi: requested changesBurt P
Change requested by lpereira in https://github.com/lpereira/hardinfo/pull/160 Signed-off-by: Burt P <pburt0@gmail.com>
2017-08-27devices/dmi: keys only need to be unique in the same groupBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2017-08-27DMI: allow specifying type for dmi_chassis_type_str()Burt P
Signed-off-by: Burt P <pburt0@gmail.com>
2017-08-27Create dmi_util.c and move *_util.c to hardinfo/Burt P
These stand alone interfaces and are used across modules, so just make them available to any module by default. Signed-off-by: Burt P <pburt0@gmail.com>
2017-08-27DMI: Add more string names, display more informationBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
2017-08-27DMI: dmi_chassis_type_str() maps chassis-type to stringBurt P
dmi_chassis_type_str() is moved from code in computer.c:detect_machine_type(). It makes sense to keep it with dmi, as it maps a dmi-defined value to a string. This way it can be used in both dmi.c and computer.c. Signed-off-by: Burt P <pburt0@gmail.com>
2017-08-27DMI: add dmi_sysfs_root() to find path for dmi in sysfsBurt P
Signed-off-by: Burt P <pburt0@gmail.com>
generated by cgit v1.2.3 (git 2.39.1) at 2025-02-06 11:39:59 +0000