Re: [PATCH] Fix the data type inconsistency issue of min (tier, MAX_CR_TIERS-1) in read_ctrl_pos

[David Hildenbrand <david@redhat.com>]

On 08.08.25 11:26, fffsqian wrote:
> Hi David,
> 
> 
> Thanks a lot for your feedback!
> 
> 
> Regarding the question of why only I encountered this error: it’s 
> because my compilation environment has CONFIG_WERROR enabled (treating 
> warnings as errors). The original code triggers a static assertion 
> warning (due to the signed/unsigned mismatch in min(tier, 4U - 1)), 
> which would normally be a non-fatal warning. However, with 
> CONFIG_WERROR, this gets elevated to a compile error, making it visible 
> in my builds.

Does it?

$ make mm/vmscan.o
...
   CC      mm/vmscan.o
$ gcc --version
gcc (GCC) 15.1.1 20250521 (Red Hat 15.1.1-2)

What am I missing?

> 
> 
> The error message is as follows:
> 
> from mm/vmscan.c:15:
> 
> mm/vmscan.c: In function ‘read_ctrl_pos’:
> 
> ./include/linux/build_bug.h:78:41: error: static assertion failed: 
> "min(tier, 4U - 1) signedness error, fix types or consider umin() before 
> min_t()"
> 
> 
> I really appreciate your suggestions on the patch format issues—I’ve 
> addressed those as well, including adding fix information and 
> streamlining patch descriptions, etc.
> 
> 
> V2 of the patch will be sent shortly.

Please wait with v2 until the discussion is over. Also, flag your 
patches as v2 like [PATCH v2] etc.

-- 
Cheers,

David / dhildenb