Re: [PATCH printk 3/5] printk: use buffer pool for sprint buffers

[Petr Mladek]

On Wed 2020-09-23 15:21:41, David Laight wrote:
> From: Petr Mladek
> > Sent: 23 September 2020 16:11
> > 
> > On Tue 2020-09-22 17:44:14, John Ogness wrote:
> > > vprintk_store() is using a single static buffer as a temporary
> > > sprint buffer for the message text. This will not work once
> > > @logbuf_lock is removed. Replace the single static buffer with a
> > > pool of buffers.
> > 
> > The buffer is used because we do not know the length of the
> > formatted message to reserve the right space in the ring buffer
> > in advance.
> > 
> > There was the idea to call vsprintf(NULL, fmt, args) to count
> > the length in advance.
> > 
> > AFAIK, there is one catch. We need to use va_copy() around
> > the 1st call because va_format can be proceed only once.
> > See, va_format() in lib/vsprintf.c as an example.
> > 
> > Is there any other problem, please?
> 
> Potentially the data can change after the vsprintf(NULL, ...)
> call so that the buffer isn't guaranteed to be the right length.

Yup. Well, in the worst case, we would shrink the text when the length
increased in the meantime. The content of such a message is racy
anyway.

> Never mind the extra cost of doing all the work twice.

IMHO, neither of this is worth the extra complexity with a pool
of buffers.

I would prefer to increase the complexity only when there are real life
problems.

Best Regards,
Petr