On 2024-03-08 at 16:07:05 +0200, Ilpo Järvinen wrote:
On Wed, 6 Mar 2024, Maciej Wieczor-Retman wrote:
@@ -697,12 +700,16 @@ int resctrl_val(const struct resctrl_test *test, struct resctrl_val_param *param) { char *resctrl_val = param->resctrl_val;
- unsigned long bw_resc_start = 0; int res_id, ret = 0, pipefd[2];
- unsigned long *bw_resc_start; struct sigaction sigact; char pipe_message = 0; union sigval value;
- bw_resc_start = calloc(snc_ways(), sizeof(unsigned long));
While correct, this seems a bit overkill given is MAX_SNC = 4, not something large or unbounded.
This patch would be be much simpler on top of my resctrl_val() cleanup patches because bw_resc_start is only local to the measurement function.
You're right, the series will get a lot simpler rebased onto yours. I'll try out some different approaches and comment any relevant points under your thread [1].
[1] https://lore.kernel.org/all/20240311135230.7007-1-ilpo.jarvinen@linux.intel....
-- i.