Sock addr hook

[shankarseal]

Description

This is the second part of changes for #847 .

This change adds net_ebpf_ext_sock_addr.c file that holds the program info and hook NPI providers for sock_addr hook. It also has an empty implementation of the classify callback for WFP callouts added for this hook.

net_ebpf_extension_xdp_on_client_attach was refactored into a utility function net_ebpf_extension_hook_check_attach_parameter that is common to sock_addr hook type as well.

Testing

No new tests have been added. They will be added in the next iteration of the changes.

Documentation

Doxygen comments.

[codecov-commenter]

Codecov Report

Merging #871 (29b6e48) into main (68a7402) will increase coverage by 0.04%.
The diff coverage is n/a.

❗ Current head 29b6e48 differs from pull request most recent head 0805949. Consider uploading reports for the commit 0805949 to get more accurate results

@@            Coverage Diff             @@
##             main     #871      +/-   ##
==========================================
+ Coverage   75.28%   75.32%   +0.04%     
==========================================
  Files          83       83              
  Lines       14614    14606       -8     
  Branches      488      482       -6     
==========================================
  Hits        11002    11002              
+ Misses       3321     3313       -8     
  Partials      291      291              

Impacted Files	Coverage Δ
tools/bpf2c/bpf_code_generator.cpp	36.98% <0.00%> (+0.54%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 68a7402...0805949. Read the comment docs.

[dthaler]

I won't block this one, but in general it is bad practice to put tests in a separate PR. Instead, in the future look for other ways to factor the PR into smaller pieces where each smaller piece contains a test. For example, the function signature changes could have been in a separate PR from the sock addr additions.

[shankarseal]

I won't block this one, but in general it is bad practice to put tests in a separate PR. Instead, in the future look for other ways to factor the PR into smaller pieces where each smaller piece contains a test. For example, the function signature changes could have been in a separate PR from the sock addr additions.

Understood. This was indeed tested but the automation was not made part of this PR as otherwise I would have to pull in other code and would make it a long PR to review. There is still an unresolved comment where I have asked a question to you. Please answer and/or resolve.