05/11/2021 11:24, Rongwei Liu:
> HI Tomas:
> Flow-perf application run-time options are growing quickly and some of
> them are a little complex.
> Under current architecture, I don't find a valid way to reduce the
> indention level unless maintainer agree to simply the logic.
Trust checkpatch, it is possible.
One basic solution is to have sub-functions.
Please do not top-post.
> For the fill_actions() function, I will rebase it once our github
> branch updated.
> It should be something like:
>
> > > + fill_actions(actions[i], all_actions[i], 0, 0, 0,
> > > + 0, 0, 0, unique_data, rx_queues_count,
> > > dst_port);
>
> BR
> Rongwei
>
> > -----Original Message-----
> > From: Thomas Monjalon <tho...@monjalon.net>
> > Sent: Friday, November 5, 2021 5:58 PM
> > To: Wisam Monther <wis...@nvidia.com>; Rongwei Liu
> > <rongw...@nvidia.com>
> > Cc: Matan Azrad <ma...@nvidia.com>; Slava Ovsiienko
> > <viachesl...@nvidia.com>; Ori Kam <or...@nvidia.com>; dev@dpdk.org;
> > Raslan Darawsheh <rasl...@nvidia.com>; Haifei Luo <haif...@nvidia.com>;
> > Jiawei(Jonny) Wang <jiaw...@nvidia.com>
> > Subject: Re: [dpdk-dev] [PATCH v3 1/3] app/flow-perf: support meter policy
> > API
> >
> > External email: Use caution opening links or attachments
> >
> >
> > 28/10/2021 05:25, Rongwei Liu:
> > > @@ -854,6 +862,46 @@ args_parse(int argc, char **argv)
> > > RTE_MAX_LCORE);
> > > }
> > > }
> > > + if (strcmp(lgopts[opt_idx].name, "policy-mtr") ==
> > > 0) {
> > > + j = 0;
> > > + k = 0;
> > > + arg = optarg;
> > > + policy_mtr = true;
> > > + token = strsep(&arg, ":\0");
> > > + while (token != NULL && j < RTE_COLORS) {
> > > + actions_str[j++] = token;
> > > + token = strsep(&arg, ":\0");
> > > + }
> > > + j = 0;
> > > + token = strtok(actions_str[0], ",\0");
> > > + while (token == NULL && j < RTE_COLORS - 1)
> > > + token = strtok(actions_str[++j],
> > > ",\0");
> > > + while (j < RTE_COLORS && token != NULL) {
> > > + for (i = 0;
> > > + i < RTE_DIM(flow_options);
> > > i++) {
> > > + if (!strcmp(token,
> > > + flow_options[i].str)) {
> > > + all_actions[j][k++]
> > > =
> > > +
> > > flow_options[i].mask;
> > > + break;
> > > + }
> > > + }
> > > + /* Reached last item with no match
> > > */
> > > + if (i >= RTE_DIM(flow_options)) {
> > > + fprintf(stderr,
> > > + "Invalid actions "
> > > + "item: %s\n",
> > > token);
> > > + usage(argv[0]);
> > > + rte_exit(EXIT_SUCCESS,
> > > "Invalid actions item\n");
> > > + }
> > > + token = strtok(NULL, ",\0");
> > > + while (!token && j < RTE_COLORS -
> > > 1) {
> > > + token =
> > > strtok(actions_str[++j],
> > > + ",\0");
> > > + k = 0;
> > > + }
> > > + }
> > > + }
> >
> > Could we avoid having so many indents?
> > Checkpatch suggest considering a refactoring.
> >
> > [...]
> > > + fill_actions(actions[i], all_actions[i], 0, 0, 0,
> > > + 0, 0, 0, unique_data, rx_queues_count);
> >
> > error: too few arguments to function ‘fill_actions’
> >
> > That's a rebase issue because I merged first the patch
> > "app/flow-perf: add destination ports parameter"
> >
> >
>
>
