On Tue, 8 Aug 2023 11:56:36 +0800 Lin Ma wrote:
> Provide some suggestions that who deal with Netlink code could follow
> (of course using the word "best-practices" may sound somewhat
> exaggerate).
>
> According to my recent practices, the parsing of the Netlink attributes
> lacks documents for kernel developers. Since recently the relevant docs
> for Netlink user space get replenished, I guess is a good chance for
> kernel space part to catch with.
>
> First time to write a document and any reviews are appreciated.
Oh, you put quite some effort into this!
I have a hard time coming up with reasonable review comments, because
while the doc is informative and pretty correct, it focuses on what's
legacy use of netlink at this point. People writing new code will only
get confuse with this information :(
Modern families are all generic netlink, and validation is taken care
of in the core :(
If you post a v2 please make sure to CC netdev@
Hello Jakub,
> Oh, you put quite some effort into this!
>
> I have a hard time coming up with reasonable review comments, because
> while the doc is informative and pretty correct, it focuses on what's
> legacy use of netlink at this point. People writing new code will only
> get confuse with this information :(
Oops, my bad, I will somewhat adjust the content structure, or just claim
to just concern legacy use for simplicity and avoid confusion.
>
> Modern families are all generic netlink, and validation is taken care
> of in the core :(
>
OK, you are right.
> If you post a v2 please make sure to CC netdev@
I see. It appears that the issue is due to an SMTP error., will watch
out this.
Regards
Lin