Re: [PATCH v5 03/15] ring-buffer: Add interface for configuring trace sub buffer size

[Google]

On Wed, 20 Dec 2023 09:40:30 -0500
Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:

> On Wed, 20 Dec 2023 23:26:19 +0900
> Masami Hiramatsu (Google) <mhiramat@xxxxxxxxxx> wrote:
> 
> > On Tue, 19 Dec 2023 13:54:17 -0500
> > Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:
> > 
> > > +/**
> > > + * ring_buffer_subbuf_order_set - set the size of ring buffer sub page.
> > > + * @buffer: The ring_buffer to set the new page size.
> > > + * @order: Order of the system pages in one sub buffer page
> > > + *
> > > + * By default, one ring buffer pages equals to one system page. This API can be
> > > + * used to set new size of the ring buffer page. The size must be order of
> > > + * system page size, that's why the input parameter @order is the order of
> > > + * system pages that are allocated for one ring buffer page:
> > > + *  0 - 1 system page
> > > + *  1 - 2 system pages
> > > + *  3 - 4 system pages
> > > + *  ...  
> > 
> > Don't we have the max order of the pages?
> 
> Actually there is. I think it's 7?
> 
> Honestly, anything over 5 is probably too much. But hey.

Ah, I see. It is checked in subbuf_order_write() method (and it is embedded
directly). I think that 7 should be replaced with a macro, something like
RB_SUBBUF_ORDER_MAX and check it in this exposed function instead of write
method.

Thank you,

-- 
Masami Hiramatsu (Google) <mhiramat@xxxxxxxxxx>