Re: [PATCH next-20260615] mm/slab: Fix build error due to extra argument

[Vlastimil Babka (SUSE)]

On 6/15/26 20:10, Calvin Owens wrote:
> On Monday 06/15 at 10:12 -0700, Calvin Owens wrote:
>> On Monday 06/15 at 09:32 -0700, Calvin Owens wrote:
>> >     mm/slub.c:5553:30: error: too many arguments to function call, expected 5, have 6
>> >      5552 |                 return __do_kmalloc_node(size, NULL, flags, node,
>> >           |                        ~~~~~~~~~~~~~~~~~
>> > 
>> > Fix by removing the nonexistent size argument.
>> > 
>> > Closes: https://lore.kernel.org/all/178153914528.15282.16770125845520023465@330cfa3079ca/
>> > Signed-off-by: Calvin Owens <calvin@xxxxxxxxxx>
>> > ---
>> > I needed to test -next this morning, so passing this along, but please
>> > don't go to any extra trouble to use it or credit me if you already
>> > fixed it :)
>> 
>> Ugh, nevermind, a different kconfig now fails the opposite way:
>> 
>>     mm/slub.c:5551:24: error: too few arguments to function ‘__do_kmalloc_node’; expected 6, have 5
>>      5551 |                 return __do_kmalloc_node(NULL, flags, node,
>> 
>> >  mm/slub.c | 2 +-
>> >  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> I guess the line number should have been a clue... I was building on
> different machines that fetched at different times, and there were
> actually two different SHAs for next-20260615 this morning:
> 
>     700f8f17ac2c
>     Author:  Mark Brown <broonie@xxxxxxxxxx>
>     Date:    Mon Jun 15 15:43:39 2026 +0100 (3 hours ago)
>     Commit:  Mark Brown <broonie@xxxxxxxxxx>
>     Date:    Mon Jun 15 15:43:39 2026 +0100 (3 hours ago)
>     Subject: Add linux-next specific files for 20260615
> 
>     8d6dbbbe3ba6
>     Author:  Mark Brown <broonie@xxxxxxxxxx>
>     Date:    Mon Jun 15 17:26:20 2026 +0100 (2 hours ago)
>     Commit:  Mark Brown <broonie@xxxxxxxxxx>
>     Date:    Mon Jun 15 17:26:20 2026 +0100 (2 hours ago)
>     Subject: Add linux-next specific files for 20260615
> 
> My patch was for the first one. But the second one has a fix for this
> applied, so just drop this patch :)

Not a fix, but the newer -next snapshot used older version of the slab tree.
The current version was actually fixed a while ago but I forgot to git push
:/ Sorry about the trouble.

>> > diff --git a/mm/slub.c b/mm/slub.c
>> > index b1de95046b0b..7660f7fa0b2c 100644
>> > --- a/mm/slub.c
>> > +++ b/mm/slub.c
>> > @@ -5549,7 +5549,7 @@ void *__kmalloc_flags_noprof(DECL_TOKEN_PARAMS(size, token), gfp_t flags,
>> >  	};
>> > 
>> >  	if (alloc_flags_allow_spinning(alloc_flags)) {
>> > -		return __do_kmalloc_node(size, NULL, flags, node,
>> > +		return __do_kmalloc_node(NULL, flags, node,
>> >  				PASS_TOKEN_PARAM(token), &ac);
>> >  	} else {
>> >  		return __kmalloc_nolock_noprof(PASS_TOKEN_PARAMS(size, token),
>> > --
>> > 2.53.0
>> >