Messages in this thread | | | Date | Thu, 03 Jan 2013 00:14:38 +0100 | From | Sylwester Nawrocki <> | Subject | Re: [PATCH RESEND 6/6] clk: s5p-g2d: Fix incorrect usage of IS_ERR_OR_NULL |
| |
On 01/02/2013 06:10 AM, Dan Carpenter wrote: > clk_get() returns NULL if CONFIG_HAVE_CLK is disabled.
It's not a problem for this driver, as it never dereferences what's returned from clk_get(). It would have to include <plat/clock.h>, which it doesn't and which would have clearly indicated abuse of the clock API.
Moreover, this driver now depends on architectures that select HAVE_CLK, so it couldn't be build when CONFIG_HAVE_CLK is disabled.
> I told Tony about this but everyone has been gone with end of year > holidays so it hasn't been addressed. > > Tony, please fix it so people don't apply these patches until > clk_get() is updated to not return NULL. It sucks to have to revert > patches.
As explained by Russell many times, the clock API users should not care whether the value returned from clk_get() is NULL or not. It should only be tested with IS_ERR(). The patches look fine to me, no need to do anything.
---
Regards, Sylwester
| |