Hi Tom,
On 30 April 2015 at 07:30, Tom Rini trini@konsulko.com wrote:
On Wed, Apr 29, 2015 at 06:33:41PM -0600, Simon Glass wrote:
Hi Jagan,
On 29 April 2015 at 08:58, Jagan Teki jagannadh.teki@gmail.com wrote:
Hi Simon,
On 29 April 2015 at 20:12, Simon Glass sjg@chromium.org wrote:
Hi Tom,
On 29 April 2015 at 07:08, Tom Rini trini@konsulko.com wrote:
On Wed, Apr 29, 2015 at 05:35:05PM +0530, Jagannadha Sutradharudu Teki wrote:
Replace (1 << nr) to BIT(nr) where nr = 0, 1, 2 .... 31
Signed-off-by: Jagannadha Sutradharudu Teki jagannadh.teki@gmail.com Cc: Masahiro Yamada yamada.m@jp.panasonic.com Cc: Tom Rini trini@konsulko.com Cc: Simon Glass sjg@chromium.org
I think that conceptually this is a good idea. We need to apply this fairly quickly however as it already throws out a few rejects with the PRs I've taken and doing some local testing on.
We have to date avoided this (I've been here before). But I think it is useful. One concern I have is misuse, when someone does BIT(0) | BIT(1))| BIT(2) | BIT(3) instead of 0xf, for example.
Probably we can achieve this with GENMASK(h, l), may be we can add that also?
I'm not sure about that...we might end up with this:
Welcome to the world of introducing an idea before the rest of the community is ready :( . GENMASK as Jagan is talking about came into the kernel (well, into <linux/bitops.h> in 2013. So yeah, we could pull that macro in too and clean up as needed / ensure people don't do funny looking things like BIT(1) | BIT(2) | ...
SGTM.
Regards, Simon