This is a straightforward extension of the existing multi-column CPU meter
code, which now allows for up CPU meters to be displayed in up to 16 columns.
This also adds the meter declarations to all the platform-specific code.
Instead of scanning the meter name to determine the number of columns in a
CPU meter, move the common code behind some wrapper functions, and specify the
number of columns as an explicit parameter when called from the wrappers.
While this does add a bit of code for all the necessary wrapper functions, this
should be less brittle in case of future changes to the CPU meter code.
Use the more portable sysfs node /sys/devices/system/cpu/cpuX/cpufreq/scaling_cur_freq
to get the CPU frequency.
In case of an error fall back to /proc/cpuinfo .
Also use a fixed width of 4 for the frequency to avoid position jumps
in case the frequency moves in the range 900-1100 MHz.
The MIN, MAX, CLAMP, MINIMUM, and MAXIMUM macros appear
throughout the codebase with many re-definitions. Make
a single copy of each in a common header file, and use
the BSD variants of MINIMUM/MAXIMUM due to conflicts in
the system <sys/param.h> headers.
Reasoning:
- implementation was unsound -- broke down when I added a fairly
basic macro definition expanding to a struct initializer in a *.c
file.
- made it way too easy (e.g. via otherwise totally innocuous git
commands) to end up with timestamps such that it always ran
MakeHeader.py but never used its output, leading to overbuild noise
when running what should be a null 'make'.
- but mostly: it's just an awkward way of dealing with C code.
The option is only implemented on Linux. On other platforms, and on Linuxes
that do not expose the relevant sysfs file, the frequency will be 0.
The "CPU average" meter does not show a frequency, only
the individual per-CPU meters.
In all the cases where sprintf was being used within htop, snprintf
could have been used. This patch replaces all uses of sprintf with
snprintf which makes sure that if a buffer is too small to hold the
resulting string, the string is simply cut short instead of causing
a buffer overflow which leads to undefined behaviour.
`sizeof(variable)` was used in these cases, as opposed to `sizeof
variable` which is my personal preference because `sizeof(variable)`
was already used in one way or another in other parts of the code.
Rationale (copied from htop issue #471):
The function name "setValues" is misleading. For most OOP (object-
oriented programming) contexts, setXXX functions mean they will change
some member variables of an object into something specified in
function arguments. But in the *Meter_setValues() case, the new values
are not from the arguments, but from a hard-coded source. The caller
is not supposed to change the values[] to anything it likes, but
rather to "update" the values from the source. Hence, updateValues is
a better name for this family of functions.
Two changes in this commit:
- All meters now explicitly specify "maxItems" property, even for just
1 item. (Exception is "container" CPU meter classes, which use
CUSTOM_METERMODE.)
- "maxItems" being 0 is now allowed. This will let bar meters and graph
meters render an empty meter.
* Performance improvements
* Support for splitting CPU meters into two or four columns
(thanks to Wim Heirman)
* Switch from PLPA, which is now deprecated, to HWLOC.
* Bring back support for native Linux sched_setaffinity,
so we don't have to use HWLOC where we don't need to.
* Support for typing in user names and column fields in selection panels.
Christian Göttsche