24 Oct
2010
24 Oct
'10
3:18 p.m.
On Sat, Oct 23, 2010 at 1:24 PM, Dimitris Papastamos dp@opensource.wolfsonmicro.com wrote:
The only problem I see with the above code, is when codec_drv->reg_word_size > sizeof (unsigned int) but that can't really happen in practice.
I'm going to have to agree with Mark that this code is suspect. I understand everything you said, but it makes me nervous. Unless this code is in some kind of fast-path, I would prefer to see it rewritten to avoid any assumption about the sizes of the types involved.
--
Timur Tabi
Linux kernel developer at Freescale