Re: [PATCH] [0/6] kfifo fixes/improvements

[Dmitry Torokhov <dmitry.torokhov@gmail.com>]

On Wed, Dec 30, 2009 at 11:52:15AM +0100, Stefani Seibold wrote:
> Am Mittwoch, den 30.12.2009, 02:43 -0800 schrieb Dmitry Torokhov:
> > On Wed, Dec 30, 2009 at 10:29:50AM +0100, Stefani Seibold wrote:
> > >
> > I do not need to try the new behavior - you explained it quite well.
> > You changed the old API to allow processing multiple records at a time
> > and it does not quite work the way you want with Andi's patch. Now the
> 
> Wrong, i did not change the behavior of the old API. It is exactly the
> same at is was!!!!

You said:

"The kfifo_in() and kfifo_out() len parameter is than in the meaning
of elements not bytes."

This is the change from the existing API which works with _bytes_:

/**
 * kfifo_in - puts some data into the FIFO
 * @fifo: the fifo to be used.
 * @from: the data to be added.
 * @len: the length of the data to be added.
 *
 * This function copies at most @len bytes from the @from buffer into
                                ^^^^^^^^^^
 * the FIFO depending on the free space, and returns the number of
 * bytes copied.


> 
> > question is: when working with _records_ does anyone really want to
> > put/get more than 1 record at a time? My answer would be "no, most users
> 
> Your answer is wrong. All current user depend on it, because it
> (miss)use a byte stream to store values other than bytes to it.

However all of them that I know of deposit and fetch exactly one record
at a time (the fact that they are more than 1 byte is immaterial).

>  
> > work with 1 record at a time". Thus your changes to the old API are not
> > needed.
> > 
> 
> A lot of hot air...

*sigh* That's an iron-clad argument right there.

-- 
Dmitry