* [PATCH] mm/sparse: Fix sparse_vmemmap_init_nid_early definition without CONFIG_SPARSEMEM
@ 2025-11-27 9:25 Jinhui Guo
2025-11-27 9:46 ` David Hildenbrand (Red Hat)
0 siblings, 1 reply; 4+ messages in thread
From: Jinhui Guo @ 2025-11-27 9:25 UTC (permalink / raw)
To: akpm, david, lorenzo.stoakes, Liam.Howlett, vbabka, rppt, surenb,
mhocko, fvdl
Cc: guojinhui.liam, linux-mm, linux-kernel
When CONFIG_SPARSEMEM is disabled, the macro
sparse_vmemmap_init_nid_early(_nid, _use) passes
two arguments, while the actual function accepts
only nid. Drop the extra argument _use.
Fixes: d65917c42373 ("mm/sparse: allow for alternate vmemmap section init at boot")
Signed-off-by: Jinhui Guo <guojinhui.liam@bytedance.com>
---
include/linux/mmzone.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 7fb7331c5725..f6dd181a4821 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -2293,7 +2293,7 @@ void sparse_init(void);
#else
#define sparse_init() do {} while (0)
#define sparse_index_init(_sec, _nid) do {} while (0)
-#define sparse_vmemmap_init_nid_early(_nid, _use) do {} while (0)
+#define sparse_vmemmap_init_nid_early(_nid) do {} while (0)
#define sparse_vmemmap_init_nid_late(_nid) do {} while (0)
#define pfn_in_present_section pfn_valid
#define subsection_map_init(_pfn, _nr_pages) do {} while (0)
--
2.20.1
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] mm/sparse: Fix sparse_vmemmap_init_nid_early definition without CONFIG_SPARSEMEM
2025-11-27 9:25 [PATCH] mm/sparse: Fix sparse_vmemmap_init_nid_early definition without CONFIG_SPARSEMEM Jinhui Guo
@ 2025-11-27 9:46 ` David Hildenbrand (Red Hat)
2025-11-27 10:29 ` Mike Rapoport
2025-11-27 12:14 ` Jinhui Guo
0 siblings, 2 replies; 4+ messages in thread
From: David Hildenbrand (Red Hat) @ 2025-11-27 9:46 UTC (permalink / raw)
To: Jinhui Guo, akpm, lorenzo.stoakes, Liam.Howlett, vbabka, rppt,
surenb, mhocko, fvdl
Cc: linux-mm, linux-kernel
On 11/27/25 10:25, Jinhui Guo wrote:
> When CONFIG_SPARSEMEM is disabled, the macro
> sparse_vmemmap_init_nid_early(_nid, _use) passes
> two arguments, while the actual function accepts
> only nid. Drop the extra argument _use.
>
> Fixes: d65917c42373 ("mm/sparse: allow for alternate vmemmap section init at boot")
> Signed-off-by: Jinhui Guo <guojinhui.liam@bytedance.com>
> ---
> include/linux/mmzone.h | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
> index 7fb7331c5725..f6dd181a4821 100644
> --- a/include/linux/mmzone.h
> +++ b/include/linux/mmzone.h
> @@ -2293,7 +2293,7 @@ void sparse_init(void);
> #else
> #define sparse_init() do {} while (0)
> #define sparse_index_init(_sec, _nid) do {} while (0)
> -#define sparse_vmemmap_init_nid_early(_nid, _use) do {} while (0)
> +#define sparse_vmemmap_init_nid_early(_nid) do {} while (0)
> #define sparse_vmemmap_init_nid_late(_nid) do {} while (0)
> #define pfn_in_present_section pfn_valid
> #define subsection_map_init(_pfn, _nr_pages) do {} while (0)
That commit is from early 2025.
So what are the symptoms we are fixing? I would have assumed a
compile-time problem, but no built bot caught that?
--
Cheers
David
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] mm/sparse: Fix sparse_vmemmap_init_nid_early definition without CONFIG_SPARSEMEM
2025-11-27 9:46 ` David Hildenbrand (Red Hat)
@ 2025-11-27 10:29 ` Mike Rapoport
2025-11-27 12:14 ` Jinhui Guo
1 sibling, 0 replies; 4+ messages in thread
From: Mike Rapoport @ 2025-11-27 10:29 UTC (permalink / raw)
To: David Hildenbrand (Red Hat)
Cc: Jinhui Guo, akpm, lorenzo.stoakes, Liam.Howlett, vbabka, surenb,
mhocko, fvdl, linux-mm, linux-kernel
On Thu, Nov 27, 2025 at 10:46:54AM +0100, David Hildenbrand (Red Hat) wrote:
> On 11/27/25 10:25, Jinhui Guo wrote:
> > When CONFIG_SPARSEMEM is disabled, the macro
> > sparse_vmemmap_init_nid_early(_nid, _use) passes
> > two arguments, while the actual function accepts
> > only nid. Drop the extra argument _use.
> >
> > Fixes: d65917c42373 ("mm/sparse: allow for alternate vmemmap section init at boot")
> > Signed-off-by: Jinhui Guo <guojinhui.liam@bytedance.com>
> > ---
> > include/linux/mmzone.h | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
> > index 7fb7331c5725..f6dd181a4821 100644
> > --- a/include/linux/mmzone.h
> > +++ b/include/linux/mmzone.h
> > @@ -2293,7 +2293,7 @@ void sparse_init(void);
> > #else
> > #define sparse_init() do {} while (0)
> > #define sparse_index_init(_sec, _nid) do {} while (0)
> > -#define sparse_vmemmap_init_nid_early(_nid, _use) do {} while (0)
> > +#define sparse_vmemmap_init_nid_early(_nid) do {} while (0)
> > #define sparse_vmemmap_init_nid_late(_nid) do {} while (0)
> > #define pfn_in_present_section pfn_valid
> > #define subsection_map_init(_pfn, _nr_pages) do {} while (0)
>
> That commit is from early 2025.
>
> So what are the symptoms we are fixing? I would have assumed a compile-time
> problem, but no built bot caught that?
Some of these are used only in sparse.c and sparse-vmmemap.c so with
CONFIG_SPARSE=n they are completely unused.
We probably should move them to mm/internal.h or even mm/sparse.h
> --
> Cheers
> David
--
Sincerely yours,
Mike.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] mm/sparse: Fix sparse_vmemmap_init_nid_early definition without CONFIG_SPARSEMEM
2025-11-27 9:46 ` David Hildenbrand (Red Hat)
2025-11-27 10:29 ` Mike Rapoport
@ 2025-11-27 12:14 ` Jinhui Guo
1 sibling, 0 replies; 4+ messages in thread
From: Jinhui Guo @ 2025-11-27 12:14 UTC (permalink / raw)
To: david
Cc: Liam.Howlett, akpm, fvdl, guojinhui.liam, linux-kernel, linux-mm,
lorenzo.stoakes, mhocko, rppt, surenb
Hi, David
On Thu, Nov 27, 2025 at 10:46:54AM +0100, David Hildenbrand (Red Hat) wrote:
> On 11/27/25 10:25, Jinhui Guo wrote:
> > When CONFIG_SPARSEMEM is disabled, the macro
> > sparse_vmemmap_init_nid_early(_nid, _use) passes
> > two arguments, while the actual function accepts
> > only nid. Drop the extra argument _use.
> >
> > Fixes: d65917c42373 ("mm/sparse: allow for alternate vmemmap section init at boot")
> > Signed-off-by: Jinhui Guo <guojinhui.liam@bytedance.com>
> > ---
> > include/linux/mmzone.h | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
> > index 7fb7331c5725..f6dd181a4821 100644
> > --- a/include/linux/mmzone.h
> > +++ b/include/linux/mmzone.h
> > @@ -2293,7 +2293,7 @@ void sparse_init(void);
> > #else
> > #define sparse_init() do {} while (0)
> > #define sparse_index_init(_sec, _nid) do {} while (0)
> > -#define sparse_vmemmap_init_nid_early(_nid, _use) do {} while (0)
> > +#define sparse_vmemmap_init_nid_early(_nid) do {} while (0)
> > #define sparse_vmemmap_init_nid_late(_nid) do {} while (0)
> > #define pfn_in_present_section pfn_valid
> > #define subsection_map_init(_pfn, _nr_pages) do {} while (0)
>
> That commit is from early 2025.
>
> So what are the symptoms we are fixing? I would have assumed a compile-time
> problem, but no built bot caught that?
I spotted the issue while porting the patches. It’s harmless. Both
sparse_index_init() and sparse_vmemmap_init_nid_early() are only
referenced in mm/sparse.c and mm/sparse-vmemmap.c. So they can simply
be deleted from include/linux/mmzone.h when CONFIG_SPARSEMEM is disabled.
> --
> Cheers
> David
--
Best regards,
Jinhui
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2025-11-27 12:14 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2025-11-27 9:25 [PATCH] mm/sparse: Fix sparse_vmemmap_init_nid_early definition without CONFIG_SPARSEMEM Jinhui Guo
2025-11-27 9:46 ` David Hildenbrand (Red Hat)
2025-11-27 10:29 ` Mike Rapoport
2025-11-27 12:14 ` Jinhui Guo
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox