On Fri, 21 Apr 2023, Reinette Chatre wrote:
On 4/18/2023 4:44 AM, Ilpo Järvinen wrote:
fill_buf(), show_bw_info(), and resctrl_val_param.span define span as unsigned long.
There is no fill_buf() in the code and show_bw_info() does not define span as unsigned long (it is even the first function changed in this patch).
Shuffling a large number of patches around seems detrimental for the quality of the commit messages no matter how hard I try to maintain them up to date. Thanks for noticing this.
Consistently use unsigned long elsewhere too for span parameters.
Is unsigned long the right type to use? Tracing through all the indirections I do not see how making all usages unsigned long achieves consistency ... have you considered size_t?
I'll change to size_t as it refers to the size of the memory block.