On Tue, Jul 11, 2023 at 6:33 PM Eelco Chaudron <[email protected]> wrote:
>
> Maybe just define it here (and remove it above)..
>
> char *name = node->name;
>
Sure.
>
> Line feed.
>
Sure.
>
> Looks a bit odd in the middle of the definitions. Maybe move it down?
>
> struct ovsdb_error *error;
> struct ovsdb_condition cond = OVSDB_CONDITION_INITIALIZER(&cond);
> struct ovsdb_monitor_table_condition *mtc =
> shash_find_data(&condition->tables, table->schema->name);
>
> if (!mtc) {
> return NULL;
> }
>
> Or add surrounding new lines:
>
> struct ovsdb_monitor_table_condition *mtc =
> shash_find_data(&condition->tables, table->schema->name);
>
> if (!mtc) {
> return NULL;
> }
>
> struct ovsdb_error *error;
> struct ovsdb_condition cond = OVSDB_CONDITION_INITIALIZER(&cond);
>
>
Sure, will move it down in the next patch version. Looks a bit neater.
>
> Is this needed, as x2nrealloc will assert if no memory is available?
>
Yes this is still needed, as x2nrealloc does not always execute when
`ovsdb_row_set_add_row` is called. It is only executed if the
condition `set->n_rows >= set->allocated_rows` is satisfied.
Best regards,
James Raphael Tiovalen
_______________________________________________
dev mailing list
[email protected]
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
