2020-11-25 20:30:16

by Charan Teja Kalla

[permalink] [raw]
Subject: [PATCH] mm: cma: improve pr_debug log in cma_release()

It is required to print 'count' of pages, along with the pages, passed
to cma_release to debug the cases of mismatched count value passed
between cma_alloc() and cma_release() from a code path.

As an example, consider the below scenario:
1) CMA pool size is 4MB and
2) User doing the erroneous step of allocating 2 pages but freeing 1
page in a loop from this CMA pool.
The step 2 causes cma_alloc() to return NULL at one point of time
because of -ENOMEM condition.

And the current pr_debug logs is not giving the info about these types
of allocation patterns because of count value not being printed in
cma_release().

We are printing the count value in the trace logs, just extend the same
to pr_debug logs too.

Signed-off-by: Charan Teja Reddy <[email protected]>
---
mm/cma.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/cma.c b/mm/cma.c
index 7f415d7..07c904b 100644
--- a/mm/cma.c
+++ b/mm/cma.c
@@ -512,7 +512,7 @@ bool cma_release(struct cma *cma, const struct page *pages, unsigned int count)
if (!cma || !pages)
return false;

- pr_debug("%s(page %p)\n", __func__, (void *)pages);
+ pr_debug("%s(page %p, count %zu)\n", __func__, (void *)pages, count);

pfn = page_to_pfn(pages);

--
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a
member of the Code Aurora Forum, hosted by The Linux Foundation


2020-11-25 20:48:12

by Souptick Joarder

[permalink] [raw]
Subject: Re: [PATCH] mm: cma: improve pr_debug log in cma_release()

On Wed, Nov 25, 2020 at 9:02 PM Charan Teja Reddy
<[email protected]> wrote:
>
> It is required to print 'count' of pages, along with the pages, passed
> to cma_release to debug the cases of mismatched count value passed
> between cma_alloc() and cma_release() from a code path.
>
> As an example, consider the below scenario:
> 1) CMA pool size is 4MB and
> 2) User doing the erroneous step of allocating 2 pages but freeing 1
> page in a loop from this CMA pool.
> The step 2 causes cma_alloc() to return NULL at one point of time
> because of -ENOMEM condition.
>
> And the current pr_debug logs is not giving the info about these types
> of allocation patterns because of count value not being printed in
> cma_release().
>
> We are printing the count value in the trace logs, just extend the same
> to pr_debug logs too.
>
> Signed-off-by: Charan Teja Reddy <[email protected]>

Reviewed-by: Souptick Joarder <[email protected]>

> ---
> mm/cma.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/mm/cma.c b/mm/cma.c
> index 7f415d7..07c904b 100644
> --- a/mm/cma.c
> +++ b/mm/cma.c
> @@ -512,7 +512,7 @@ bool cma_release(struct cma *cma, const struct page *pages, unsigned int count)
> if (!cma || !pages)
> return false;
>
> - pr_debug("%s(page %p)\n", __func__, (void *)pages);
> + pr_debug("%s(page %p, count %zu)\n", __func__, (void *)pages, count);
>
> pfn = page_to_pfn(pages);
>
> --
> QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a
> member of the Code Aurora Forum, hosted by The Linux Foundation
>
>

2020-11-26 04:25:16

by David Hildenbrand

[permalink] [raw]
Subject: Re: [PATCH] mm: cma: improve pr_debug log in cma_release()

On 25.11.20 16:32, Charan Teja Reddy wrote:
> It is required to print 'count' of pages, along with the pages, passed
> to cma_release to debug the cases of mismatched count value passed
> between cma_alloc() and cma_release() from a code path.
>
> As an example, consider the below scenario:
> 1) CMA pool size is 4MB and
> 2) User doing the erroneous step of allocating 2 pages but freeing 1
> page in a loop from this CMA pool.
> The step 2 causes cma_alloc() to return NULL at one point of time
> because of -ENOMEM condition.
>
> And the current pr_debug logs is not giving the info about these types
> of allocation patterns because of count value not being printed in
> cma_release().
>
> We are printing the count value in the trace logs, just extend the same
> to pr_debug logs too.
>
> Signed-off-by: Charan Teja Reddy <[email protected]>
> ---
> mm/cma.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/mm/cma.c b/mm/cma.c
> index 7f415d7..07c904b 100644
> --- a/mm/cma.c
> +++ b/mm/cma.c
> @@ -512,7 +512,7 @@ bool cma_release(struct cma *cma, const struct page *pages, unsigned int count)
> if (!cma || !pages)
> return false;
>
> - pr_debug("%s(page %p)\n", __func__, (void *)pages);
> + pr_debug("%s(page %p, count %zu)\n", __func__, (void *)pages, count);
>
> pfn = page_to_pfn(pages);
>
>

Reviewed-by: David Hildenbrand <[email protected]>

--
Thanks,

David / dhildenb

2020-11-27 15:52:06

by Vlastimil Babka

[permalink] [raw]
Subject: Re: [PATCH] mm: cma: improve pr_debug log in cma_release()

On 11/25/20 4:32 PM, Charan Teja Reddy wrote:
> It is required to print 'count' of pages, along with the pages, passed
> to cma_release to debug the cases of mismatched count value passed
> between cma_alloc() and cma_release() from a code path.
>
> As an example, consider the below scenario:
> 1) CMA pool size is 4MB and
> 2) User doing the erroneous step of allocating 2 pages but freeing 1
> page in a loop from this CMA pool.
> The step 2 causes cma_alloc() to return NULL at one point of time
> because of -ENOMEM condition.
>
> And the current pr_debug logs is not giving the info about these types
> of allocation patterns because of count value not being printed in
> cma_release().
>
> We are printing the count value in the trace logs, just extend the same
> to pr_debug logs too.
>
> Signed-off-by: Charan Teja Reddy <[email protected]>

Acked-by: Vlastimil Babka <[email protected]>

> ---
> mm/cma.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/mm/cma.c b/mm/cma.c
> index 7f415d7..07c904b 100644
> --- a/mm/cma.c
> +++ b/mm/cma.c
> @@ -512,7 +512,7 @@ bool cma_release(struct cma *cma, const struct page *pages, unsigned int count)
> if (!cma || !pages)
> return false;
>
> - pr_debug("%s(page %p)\n", __func__, (void *)pages);
> + pr_debug("%s(page %p, count %zu)\n", __func__, (void *)pages, count);
>
> pfn = page_to_pfn(pages);
>
>