Re: [PATCH] mm: vmscan: Avoid signedness error for GCC 5.4

[Andrew Morton <akpm@linux-foundation.org>]

On Wed, 7 May 2025 12:06:11 +0800 WangYuli <wangyuli@uniontech.com> wrote:

> Perhaps a possible approach for now could be to change this line to for 
> (i = tier % MAX_NR_TIERS; i <= min_t(int, tier, MAX_NR_TIERS - 1); i++) 
> {, which would allow us to keep the signed int type for the tier 
> variable itself.

Sure.

> Regarding the potential for a more comprehensive change in the future to 
> redefine all these "tier" variables and related ones as unsigned int, I 
> would be very grateful for your guidance on whether that's a direction 
> we should consider.

`int' is a curse.  Yes, we do this very frequently - we unthinkingly
use a signed type for a naturally unsigned concept.  Ths signed type
spreads and spreads and the incorrect signage causes (small) problems
in various places.  By then it's a big mess trying to switch to an
unsigned type.

Oh well, we just battle on.  We should at least be more vigilant about
this when adding new things.


hp2:/usr/src/25> grep "int nid" mm/*.c | wc -l
316

argh.