mlx5hws_mat...e_set_target(9)
๐ mlx5hws_matcher_resize_set_target โ Kernel API Manual, bestie edition โจ
Okay so this is a page from the Kernel API Manual and it's about one very specific little function. Buckle up, it's short but it matters ๐
๐ธ NAME ๐ธ
mlx5hws_matcher_resize_set_target โ the function that links two matchers together and turns on moving rules. That's it, that's the whole job ๐ฆ
โจ SYNOPSIS โจ
int mlx5hws_matcher_resize_set_target (struct mlx5hws_matcher
*src_matcher , struct mlx5hws_matcher *dst_matcher );
Like, literally type it exactly like that, no cap ๐
๐ซถ ARGUMENTS ๐ซถ
src_matcherโ the source matcher, aka where you're moving rules from. She's the one packing her bags ๐ผdst_matcherโ the destination matcher, aka where you're moving rules to. She's the one getting the new roommates ๐
๐ DESCRIPTION ๐
Both matchers have to be in the same table type, and they must be created with the match templates and action templates).
Yes bestie, that sentence is a little incomplete in the original and I am not about to invent words that aren't there ๐ญ Just know: same table type, created with the match templates and action templates. Periodt.
Also, it is the user's responsibility to make sure the destination matcher is allocated with the appropriate size. Not the kernel's problem, not my problem โ your problem ๐ So like, size her correctly, okay?
Once the function is done doing its thing, the user is:
- โ Allowed to move rules from the source into the destination matcher.
- โ No longer allowed to insert rules into the source matcher. She's closed for business, doors locked, lights off ๐
And the stuff you can always do:
- The user is always allowed to insert rules into the destination matcher.
- ...and to delete rules from any matcher. Any of them. Go off ๐ฅ
โญ RETURN โญ
Zero on successful move, non-zero otherwise. Classic zero-means-slay situation ๐
๐ฆ SEE ALSO ๐ฆ
First up, the kernel file itself, exactly where it lives:
./drivers/net/ethernet/mellanox/mlx5/core/steering/hws/mlx5hws.h
And now the whole friend group โ every single one of these, no skipping, I don't make the rules (I do, actually, but still) ๐
mlx5hws_is_supported(9)mlx5hws_context_open(9)mlx5hws_context_close(9)mlx5hws_context_set_peer(9)mlx5hws_table_create(9)mlx5hws_table_destroy(9)mlx5hws_table_get_id(9)mlx5hws_table_set_default_miss(9)mlx5hws_match_template_create(9)mlx5hws_match_template_destroy(9)mlx5hws_action_template_create(9)mlx5hws_action_template_destroy(9)mlx5hws_matcher_create(9)mlx5hws_matcher_destroy(9)mlx5hws_matcher_attach_at(9)mlx5hws_matcher_resize_rule_move(9)mlx5hws_rule_create(9)mlx5hws_rule_destroy(9)mlx5hws_rule_action_update(9)mlx5hws_action_get_type(9)mlx5hws_action_get_dev(9)mlx5hws_action_create_dest_drop(9)mlx5hws_action_create_default_miss(9)mlx5hws_action_create_dest_table(9)mlx5hws_action_create_dest_table_num(9)mlx5hws_action_create_dest_match_range(9)mlx5hws_action_create_flow_sampler(9)mlx5hws_action_create_dest_vport(9)mlx5hws_action_create_tag(9)mlx5hws_action_create_counter(9)mlx5hws_action_create_reformat(9)mlx5hws_action_create_modify_header(9)mlx5hws_action_create_aso_meter(9)mlx5hws_action_create_pop_vlan(9)mlx5hws_action_create_push_vlan(9)mlx5hws_action_create_dest_array(9)mlx5hws_action_create_insert_header(9)mlx5hws_action_create_remove_header(9)mlx5hws_action_create_last(9)mlx5hws_action_destroy(9)mlx5hws_send_queue_poll(9)mlx5hws_send_queue_action(9)mlx5hws_debug_dump(9)mlx5hws_bwc_matcher_create(9)mlx5hws_bwc_matcher_destroy(9)mlx5hws_bwc_rule_create(9)mlx5hws_bwc_rule_destroy(9)mlx5hws_bwc_rule_action_update(9)
That's the whole page, bestie! ๐ It came from ./drivers/net/ethernet/mellano... and it's dated October 2026 (yes, the future, we're iconic like that ๐ฎโจ). Mwah, go resize responsibly ๐๐
๐ฆ where this page comes from
- nixpkgs unstable: linux-manual 7.3.1
- ๐ girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms