Patch [1] replaced vmalloc() function with a new definition but it did
not adjust the forward declaration used in UML architecture. Change it
to act as before.
Note that this prevents the vmalloc() allocations in __wrap_malloc()
from being accounted. If accounting here is critical, we will have
to remove this forward declaration and include vmalloc.h, however
that would pull in more dependencies and would require introducing more
architecture-specific headers, like asm/bug.h, asm/rwonce.h, etc.
This is likely the reason why this forward declaration was introduced
in the first place.
[1] https://lore.kernel.org/all/[email protected]/
Fixes: 576477564ede ("mm: vmalloc: enable memory allocation profiling")
Reported-by: SeongJae Park <[email protected]>
Signed-off-by: Suren Baghdasaryan <[email protected]>
---
arch/um/include/shared/um_malloc.h | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/arch/um/include/shared/um_malloc.h b/arch/um/include/shared/um_malloc.h
index 13da93284c2c..bf503658f08e 100644
--- a/arch/um/include/shared/um_malloc.h
+++ b/arch/um/include/shared/um_malloc.h
@@ -11,7 +11,8 @@
extern void *uml_kmalloc(int size, int flags);
extern void kfree(const void *ptr);
-extern void *vmalloc(unsigned long size);
+extern void *vmalloc_noprof(unsigned long size);
+#define vmalloc(...) vmalloc_noprof(__VA_ARGS__)
extern void vfree(void *ptr);
#endif /* __UM_MALLOC_H__ */
--
2.44.0.396.g6e790dbe36-goog
On Tue, 26 Mar 2024 00:37:50 -0700 Suren Baghdasaryan <[email protected]> wrote:
> Patch [1] replaced vmalloc() function with a new definition but it did
> not adjust the forward declaration used in UML architecture. Change it
> to act as before.
> Note that this prevents the vmalloc() allocations in __wrap_malloc()
> from being accounted. If accounting here is critical, we will have
> to remove this forward declaration and include vmalloc.h, however
> that would pull in more dependencies and would require introducing more
> architecture-specific headers, like asm/bug.h, asm/rwonce.h, etc.
> This is likely the reason why this forward declaration was introduced
> in the first place.
>
> [1] https://lore.kernel.org/all/[email protected]/
>
> Fixes: 576477564ede ("mm: vmalloc: enable memory allocation profiling")
> Reported-by: SeongJae Park <[email protected]>
> Signed-off-by: Suren Baghdasaryan <[email protected]>
Thank you for this fix, Suren. I confirmed that this patch fixes the issue I
reported.
Closes: https://lore.kernel.org/all/[email protected]/
Tested-by: SeongJae Park <[email protected]>
Thanks,
SJ
> ---
> arch/um/include/shared/um_malloc.h | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/arch/um/include/shared/um_malloc.h b/arch/um/include/shared/um_malloc.h
> index 13da93284c2c..bf503658f08e 100644
> --- a/arch/um/include/shared/um_malloc.h
> +++ b/arch/um/include/shared/um_malloc.h
> @@ -11,7 +11,8 @@
> extern void *uml_kmalloc(int size, int flags);
> extern void kfree(const void *ptr);
>
> -extern void *vmalloc(unsigned long size);
> +extern void *vmalloc_noprof(unsigned long size);
> +#define vmalloc(...) vmalloc_noprof(__VA_ARGS__)
> extern void vfree(void *ptr);
>
> #endif /* __UM_MALLOC_H__ */
> --
> 2.44.0.396.g6e790dbe36-goog
On Tue, 2024-03-26 at 00:37 -0700, Suren Baghdasaryan wrote:
>
> -extern void *vmalloc(unsigned long size);
> +extern void *vmalloc_noprof(unsigned long size);
> +#define vmalloc(...) vmalloc_noprof(__VA_ARGS__)
>
I was confused a bit by the define at first, but that's because this is
a user-side header file.
Reviewed-by: Johannes Berg <[email protected]
johannes
----- Ursprüngliche Mail -----
> Von: "Suren Baghdasaryan" <[email protected]>
> Betreff: [PATCH 1/1] arch/um: fix forward declaration for vmalloc
> Patch [1] replaced vmalloc() function with a new definition but it did
> not adjust the forward declaration used in UML architecture. Change it
> to act as before.
> Note that this prevents the vmalloc() allocations in __wrap_malloc()
> from being accounted. If accounting here is critical, we will have
> to remove this forward declaration and include vmalloc.h, however
> that would pull in more dependencies and would require introducing more
> architecture-specific headers, like asm/bug.h, asm/rwonce.h, etc.
> This is likely the reason why this forward declaration was introduced
> in the first place.
>
> [1] https://lore.kernel.org/all/[email protected]/
>
> Fixes: 576477564ede ("mm: vmalloc: enable memory allocation profiling")
This commit id is not in Linus tree.
Do I miss something?
Thanks,
//richard
On Mon, Apr 22, 2024 at 1:31 PM Richard Weinberger <[email protected]> wrote:
>
> ----- Ursprüngliche Mail -----
> > Von: "Suren Baghdasaryan" <[email protected]>
> >> > Fixes: 576477564ede ("mm: vmalloc: enable memory allocation profiling")
> >>
> >> This commit id is not in Linus tree.
> >> Do I miss something?
> >
> > It's in mm-unstable under dc26c7e79daf2fc11169b23c150862f0e878ee5a. I
> > think it just didn't reach Linus' tree yet.
>
> Hmm, so we better postpone this path until said commit hits Linus tree,
> or you carry it together with the commit in mm-unstable.
Oh, sorry, I didn't realize you were talking about the `Fixes:
576477564ede` part... Yeah, unfortunately SHAs in mm-unstable are
unstable, so the change being fixed is under
edf0a25633bda1e5b7844478dd13b4326a3d5d09 now. I think Andrew placed
this fix right after the change it fixes with intention to merge them
together later on.
>
> Thanks,
> //richard