Re: [PATCH v2 1/2] ring-buffer: Introducing ring-buffer mapping functions

[Vincent Donnefort <vdonnefort@google.com>]

On Wed, Mar 29, 2023 at 08:47:35AM -0400, Steven Rostedt wrote:
> On Wed, 29 Mar 2023 13:23:01 +0100
> Vincent Donnefort <vdonnefort@google.com> wrote:
> 
> > On Wed, Mar 29, 2023 at 07:03:53AM -0400, Steven Rostedt wrote:
> > > On Wed, 29 Mar 2023 10:19:44 +0100
> > > Vincent Donnefort <vdonnefort@google.com> wrote:
> > >   
> > > > > I've been playing with this a bit, and I'm thinking, do we need the
> > > > > data_pages[] array on the meta page?
> > > > > 
> > > > > I noticed that I'm not even using it.
> > > > > 
> > > > > Currently, we need to do a ioctl every time we finish with the reader page,
> > > > > and that updates the reader_page in the meta data to point to the next page
> > > > > to read. When do we need to look at the data_start section?    
> > > > 
> > > > This is for non-consuming read, to get all the pages in order.  
> > > 
> > > Yeah, I was trying to see how a non consuming read would work, and was
> > > having issues figuring that out without the tail page being updated.  
> > 
> > Would the userspace really need to know where is the tail page? It can just stop
> > whenever it finds out a page doesn't have any events, and make sure it does not
> > loop once back to the head?
> 
> I'm trying to come up with a possible algorithm that doesn't need
> ioctls. It would need to know if the writer moved or not. Probably need
> a counter that gets incremented every time the writer goes to a new page.

The v2 of this series only updates the head page in the update ioctl
(ring_buffer_update_meta_page()) Couldn't find a nice way around that as it can
be either updated by the reader or the writer. So the best solution seemed a
call to set_head_page().

[...]