Re: [PATCH net-next] net/sock: Introduce trace_sk_data_ready()

On Sat, Oct 15, 2022 at 1:07 PM Cong Wang [off-list ref] wrote:

On Thu, Sep 29, 2022 at 09:19:34AM -0700, Eric Dumazet wrote:

quoted

Second patch adding the tracing point once in the helper ?

Alternatively, why not add the tracepoint directly in the called
functions (we have few of them),
instead of all call points ?

Why do we want to give implementations of sk_data_ready() freedom
to not to call this trace_sk_data_ready()?

Thanks.

I proposed an alternative. Choose one, but not the one you do not like :/

The first proposition was to split the patch in a more logical way.

I think we have less sk_data_ready() functions than call sites, so
code size would be smaller,
but this is probably something we care about enough.