diff mbox series

[v2,1/1] alloc_tag: mark pages reserved during CMA activation as not tagged

Message ID 20240812192428.151825-1-surenb@google.com (mailing list archive)
State New
Headers show
Series [v2,1/1] alloc_tag: mark pages reserved during CMA activation as not tagged | expand

Commit Message

Suren Baghdasaryan Aug. 12, 2024, 7:24 p.m. UTC 
During CMA activation, pages in CMA area are prepared and then freed
without being allocated. This triggers warnings when memory allocation
debug config (CONFIG_MEM_ALLOC_PROFILING_DEBUG) is enabled. Fix this
by marking these pages not tagged before freeing them.

Fixes: d224eb0287fb ("codetag: debug: mark codetags for reserved pages as empty")
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Cc: stable@vger.kernel.org # v6.10
---
changes since v1 [1]
- Added Fixes tag
- CC'ed stable

[1] https://lore.kernel.org/all/20240812184455.86580-1-surenb@google.com/

 mm/mm_init.c | 10 ++++++++++
 1 file changed, 10 insertions(+)


base-commit: d74da846046aeec9333e802f5918bd3261fb5509

Comments

David Hildenbrand Aug. 13, 2024, 9:25 a.m. UTC | #1 
On 12.08.24 21:24, Suren Baghdasaryan wrote:
> During CMA activation, pages in CMA area are prepared and then freed
> without being allocated. This triggers warnings when memory allocation
> debug config (CONFIG_MEM_ALLOC_PROFILING_DEBUG) is enabled. Fix this
> by marking these pages not tagged before freeing them.
> 
> Fixes: d224eb0287fb ("codetag: debug: mark codetags for reserved pages as empty")
> Signed-off-by: Suren Baghdasaryan <surenb@google.com>
> Cc: stable@vger.kernel.org # v6.10
> ---
> changes since v1 [1]
> - Added Fixes tag
> - CC'ed stable
> 
> [1] https://lore.kernel.org/all/20240812184455.86580-1-surenb@google.com/
> 
>   mm/mm_init.c | 10 ++++++++++
>   1 file changed, 10 insertions(+)
> 
> diff --git a/mm/mm_init.c b/mm/mm_init.c
> index 75c3bd42799b..ec9324653ad9 100644
> --- a/mm/mm_init.c
> +++ b/mm/mm_init.c
> @@ -2245,6 +2245,16 @@ void __init init_cma_reserved_pageblock(struct page *page)
>   
>   	set_pageblock_migratetype(page, MIGRATE_CMA);
>   	set_page_refcounted(page);
> +
> +	/* pages were reserved and not allocated */
> +	if (mem_alloc_profiling_enabled()) {
> +		union codetag_ref *ref = get_page_tag_ref(page);
> +
> +		if (ref) {
> +			set_codetag_empty(ref);
> +			put_page_tag_ref(ref);
> +		}
> +	}

Should we have a helper like clear_page_tag_ref() that wraps this?
Suren Baghdasaryan Aug. 13, 2024, 2:27 p.m. UTC | #2 
On Tue, Aug 13, 2024 at 2:25 AM David Hildenbrand <david@redhat.com> wrote:
>
> On 12.08.24 21:24, Suren Baghdasaryan wrote:
> > During CMA activation, pages in CMA area are prepared and then freed
> > without being allocated. This triggers warnings when memory allocation
> > debug config (CONFIG_MEM_ALLOC_PROFILING_DEBUG) is enabled. Fix this
> > by marking these pages not tagged before freeing them.
> >
> > Fixes: d224eb0287fb ("codetag: debug: mark codetags for reserved pages as empty")
> > Signed-off-by: Suren Baghdasaryan <surenb@google.com>
> > Cc: stable@vger.kernel.org # v6.10
> > ---
> > changes since v1 [1]
> > - Added Fixes tag
> > - CC'ed stable
> >
> > [1] https://lore.kernel.org/all/20240812184455.86580-1-surenb@google.com/
> >
> >   mm/mm_init.c | 10 ++++++++++
> >   1 file changed, 10 insertions(+)
> >
> > diff --git a/mm/mm_init.c b/mm/mm_init.c
> > index 75c3bd42799b..ec9324653ad9 100644
> > --- a/mm/mm_init.c
> > +++ b/mm/mm_init.c
> > @@ -2245,6 +2245,16 @@ void __init init_cma_reserved_pageblock(struct page *page)
> >
> >       set_pageblock_migratetype(page, MIGRATE_CMA);
> >       set_page_refcounted(page);
> > +
> > +     /* pages were reserved and not allocated */
> > +     if (mem_alloc_profiling_enabled()) {
> > +             union codetag_ref *ref = get_page_tag_ref(page);
> > +
> > +             if (ref) {
> > +                     set_codetag_empty(ref);
> > +                     put_page_tag_ref(ref);
> > +             }
> > +     }
>
> Should we have a helper like clear_page_tag_ref() that wraps this?

With this one we have 3 instances of this sequence, so it makes sense
to have a helper. I'm going to send a v3 with 2 patches - one
introducing clear_page_tag_ref() and the next one adding this
instance.
Thanks for the suggestion, David!

>
> --
> Cheers,
>
> David / dhildenb
>
Suren Baghdasaryan Aug. 13, 2024, 3:18 p.m. UTC | #3 
On Tue, Aug 13, 2024 at 7:27 AM Suren Baghdasaryan <surenb@google.com> wrote:
>
> On Tue, Aug 13, 2024 at 2:25 AM David Hildenbrand <david@redhat.com> wrote:
> >
> > On 12.08.24 21:24, Suren Baghdasaryan wrote:
> > > During CMA activation, pages in CMA area are prepared and then freed
> > > without being allocated. This triggers warnings when memory allocation
> > > debug config (CONFIG_MEM_ALLOC_PROFILING_DEBUG) is enabled. Fix this
> > > by marking these pages not tagged before freeing them.
> > >
> > > Fixes: d224eb0287fb ("codetag: debug: mark codetags for reserved pages as empty")
> > > Signed-off-by: Suren Baghdasaryan <surenb@google.com>
> > > Cc: stable@vger.kernel.org # v6.10
> > > ---
> > > changes since v1 [1]
> > > - Added Fixes tag
> > > - CC'ed stable
> > >
> > > [1] https://lore.kernel.org/all/20240812184455.86580-1-surenb@google.com/
> > >
> > >   mm/mm_init.c | 10 ++++++++++
> > >   1 file changed, 10 insertions(+)
> > >
> > > diff --git a/mm/mm_init.c b/mm/mm_init.c
> > > index 75c3bd42799b..ec9324653ad9 100644
> > > --- a/mm/mm_init.c
> > > +++ b/mm/mm_init.c
> > > @@ -2245,6 +2245,16 @@ void __init init_cma_reserved_pageblock(struct page *page)
> > >
> > >       set_pageblock_migratetype(page, MIGRATE_CMA);
> > >       set_page_refcounted(page);
> > > +
> > > +     /* pages were reserved and not allocated */
> > > +     if (mem_alloc_profiling_enabled()) {
> > > +             union codetag_ref *ref = get_page_tag_ref(page);
> > > +
> > > +             if (ref) {
> > > +                     set_codetag_empty(ref);
> > > +                     put_page_tag_ref(ref);
> > > +             }
> > > +     }
> >
> > Should we have a helper like clear_page_tag_ref() that wraps this?
>
> With this one we have 3 instances of this sequence, so it makes sense
> to have a helper. I'm going to send a v3 with 2 patches - one
> introducing clear_page_tag_ref() and the next one adding this
> instance.
> Thanks for the suggestion, David!

v3 posted at https://lore.kernel.org/all/20240813150758.855881-1-surenb@google.com/

>
> >
> > --
> > Cheers,
> >
> > David / dhildenb
> >
diff mbox series

Patch

diff --git a/mm/mm_init.c b/mm/mm_init.c
index 75c3bd42799b..ec9324653ad9 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -2245,6 +2245,16 @@  void __init init_cma_reserved_pageblock(struct page *page)
 
 	set_pageblock_migratetype(page, MIGRATE_CMA);
 	set_page_refcounted(page);
+
+	/* pages were reserved and not allocated */
+	if (mem_alloc_profiling_enabled()) {
+		union codetag_ref *ref = get_page_tag_ref(page);
+
+		if (ref) {
+			set_codetag_empty(ref);
+			put_page_tag_ref(ref);
+		}
+	}
 	__free_pages(page, pageblock_order);
 
 	adjust_managed_page_count(page, pageblock_nr_pages);