[PATCH] mm: cma: print allocation failure reason and bitmap status

[PATCH] mm: cma: print allocation failure reason and bitmap status

[Jaewon Kim]

There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR.
This patch prints the error value and bitmap status to know available pages
regarding fragmentation.

This is an ENOMEM example with this patch.
[   11.616321]  [2:   Binder:711_1:  740] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12
[   11.616365]  [2:   Binder:711_1:  740] number of available pages: 4+7+7+8+38+166+127=>357 pages, total: 2048 pages

Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>
---
 mm/cma.c | 29 ++++++++++++++++++++++++++++-
 1 file changed, 28 insertions(+), 1 deletion(-)

diff --git a/mm/cma.c b/mm/cma.c
index c960459..535aa39 100644
--- a/mm/cma.c
+++ b/mm/cma.c
@@ -369,7 +369,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
 	unsigned long start = 0;
 	unsigned long bitmap_maxno, bitmap_no, bitmap_count;
 	struct page *page = NULL;
-	int ret;
+	int ret = -ENOMEM;
 
 	if (!cma || !cma->count)
 		return NULL;
@@ -427,6 +427,33 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
 	trace_cma_alloc(pfn, page, count, align);
 
 	pr_debug("%s(): returned %p\n", __func__, page);
+
+	if (ret != 0) {
+		unsigned int nr, nr_total = 0;
+		unsigned long next_set_bit;
+
+		pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
+			__func__, count, ret);
+		mutex_lock(&cma->lock);
+		printk("number of available pages: ");
+		start = 0;
+		for (;;) {
+			bitmap_no = find_next_zero_bit(cma->bitmap, cma->count, start);
+			next_set_bit = find_next_bit(cma->bitmap, cma->count, bitmap_no);
+			nr = next_set_bit - bitmap_no;
+			if (bitmap_no >= cma->count)
+				break;
+			if (nr_total == 0)
+				printk("%u", nr);
+			else
+				printk("+%u", nr);
+			nr_total += nr;
+			start = bitmap_no + nr;
+		}
+		printk("=>%u pages, total: %lu pages\n", nr_total, cma->count);
+		mutex_unlock(&cma->lock);
+	}
+
 	return page;
 }
 
-- 
1.9.1

Re: [PATCH] mm: cma: print allocation failure reason and bitmap status

[Michal Hocko]

On Thu 29-12-16 11:28:02, Jaewon Kim wrote:
> There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR.
> This patch prints the error value and bitmap status to know available pages
> regarding fragmentation.
> 
> This is an ENOMEM example with this patch.
> [   11.616321]  [2:   Binder:711_1:  740] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12

> [   11.616365]  [2:   Binder:711_1:  740] number of available pages: 4+7+7+8+38+166+127=>357 pages, total: 2048 pages

Could you be more specific why this part is useful?
 
> Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>
> ---
>  mm/cma.c | 29 ++++++++++++++++++++++++++++-
>  1 file changed, 28 insertions(+), 1 deletion(-)
> 
> diff --git a/mm/cma.c b/mm/cma.c
> index c960459..535aa39 100644
> --- a/mm/cma.c
> +++ b/mm/cma.c
> @@ -369,7 +369,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>  	unsigned long start = 0;
>  	unsigned long bitmap_maxno, bitmap_no, bitmap_count;
>  	struct page *page = NULL;
> -	int ret;
> +	int ret = -ENOMEM;
>  
>  	if (!cma || !cma->count)
>  		return NULL;
> @@ -427,6 +427,33 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>  	trace_cma_alloc(pfn, page, count, align);
>  
>  	pr_debug("%s(): returned %p\n", __func__, page);
> +
> +	if (ret != 0) {
> +		unsigned int nr, nr_total = 0;
> +		unsigned long next_set_bit;
> +
> +		pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
> +			__func__, count, ret);
> +		mutex_lock(&cma->lock);
> +		printk("number of available pages: ");
> +		start = 0;
> +		for (;;) {
> +			bitmap_no = find_next_zero_bit(cma->bitmap, cma->count, start);
> +			next_set_bit = find_next_bit(cma->bitmap, cma->count, bitmap_no);
> +			nr = next_set_bit - bitmap_no;
> +			if (bitmap_no >= cma->count)
> +				break;
> +			if (nr_total == 0)
> +				printk("%u", nr);
> +			else
> +				printk("+%u", nr);
> +			nr_total += nr;
> +			start = bitmap_no + nr;
> +		}
> +		printk("=>%u pages, total: %lu pages\n", nr_total, cma->count);
> +		mutex_unlock(&cma->lock);
> +	}
> +
>  	return page;
>  }
>  
> -- 
> 1.9.1
> 
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@kvack.org.  For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

-- 
Michal Hocko
SUSE Labs

Re: [PATCH] mm: cma: print allocation failure reason and bitmap status

[Jaewon Kim]

On 2016년 12월 29일 18:14, Michal Hocko wrote:
> On Thu 29-12-16 11:28:02, Jaewon Kim wrote:
>> There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR.
>> This patch prints the error value and bitmap status to know available pages
>> regarding fragmentation.
>>
>> This is an ENOMEM example with this patch.
>> [   11.616321]  [2:   Binder:711_1:  740] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12
>> [   11.616365]  [2:   Binder:711_1:  740] number of available pages: 4+7+7+8+38+166+127=>357 pages, total: 2048 pages
> Could you be more specific why this part is useful?
Hi
Without this patch we do not know why CMA allocation failed.
Additionally in case of ENOMEM, with bitmap status we can figure out that
if it is too small CMA region issue or if it is fragmentation issue.
>  
>> Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>
>> ---
>>  mm/cma.c | 29 ++++++++++++++++++++++++++++-
>>  1 file changed, 28 insertions(+), 1 deletion(-)
>>
>> diff --git a/mm/cma.c b/mm/cma.c
>> index c960459..535aa39 100644
>> --- a/mm/cma.c
>> +++ b/mm/cma.c
>> @@ -369,7 +369,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>>  	unsigned long start = 0;
>>  	unsigned long bitmap_maxno, bitmap_no, bitmap_count;
>>  	struct page *page = NULL;
>> -	int ret;
>> +	int ret = -ENOMEM;
>>  
>>  	if (!cma || !cma->count)
>>  		return NULL;
>> @@ -427,6 +427,33 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>>  	trace_cma_alloc(pfn, page, count, align);
>>  
>>  	pr_debug("%s(): returned %p\n", __func__, page);
>> +
>> +	if (ret != 0) {
>> +		unsigned int nr, nr_total = 0;
>> +		unsigned long next_set_bit;
>> +
>> +		pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
>> +			__func__, count, ret);
>> +		mutex_lock(&cma->lock);
>> +		printk("number of available pages: ");
>> +		start = 0;
>> +		for (;;) {
>> +			bitmap_no = find_next_zero_bit(cma->bitmap, cma->count, start);
>> +			next_set_bit = find_next_bit(cma->bitmap, cma->count, bitmap_no);
>> +			nr = next_set_bit - bitmap_no;
>> +			if (bitmap_no >= cma->count)
>> +				break;
>> +			if (nr_total == 0)
>> +				printk("%u", nr);
>> +			else
>> +				printk("+%u", nr);
>> +			nr_total += nr;
>> +			start = bitmap_no + nr;
>> +		}
>> +		printk("=>%u pages, total: %lu pages\n", nr_total, cma->count);
>> +		mutex_unlock(&cma->lock);
>> +	}
>> +
>>  	return page;
>>  }
>>  
>> -- 
>> 1.9.1
>>
>> --
>> To unsubscribe, send a message with 'unsubscribe linux-mm' in
>> the body to majordomo@kvack.org.  For more info on Linux MM,
>> see: http://www.linux-mm.org/ .
>> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] mm: cma: print allocation failure reason and bitmap status

[Michal Hocko]

On Thu 29-12-16 18:26:38, Jaewon Kim wrote:
> 
> 
> On 2016년 12월 29일 18:14, Michal Hocko wrote:
> > On Thu 29-12-16 11:28:02, Jaewon Kim wrote:
> >> There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR.
> >> This patch prints the error value and bitmap status to know available pages
> >> regarding fragmentation.
> >>
> >> This is an ENOMEM example with this patch.
> >> [   11.616321]  [2:   Binder:711_1:  740] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12
> >> [   11.616365]  [2:   Binder:711_1:  740] number of available pages: 4+7+7+8+38+166+127=>357 pages, total: 2048 pages
> > Could you be more specific why this part is useful?
> Hi
> Without this patch we do not know why CMA allocation failed.

Yes, I understand the first part

> Additionally in case of ENOMEM, with bitmap status we can figure out that

The code doesn't seem to check for ENOMEM though

> if it is too small CMA region issue or if it is fragmentation issue.

then please describe that in the changelog. If I got it right the above
would tell us that the fragmentation is the problem, right?

> >  
> >> Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>
> >> ---
> >>  mm/cma.c | 29 ++++++++++++++++++++++++++++-
> >>  1 file changed, 28 insertions(+), 1 deletion(-)
> >>
> >> diff --git a/mm/cma.c b/mm/cma.c
> >> index c960459..535aa39 100644
> >> --- a/mm/cma.c
> >> +++ b/mm/cma.c
> >> @@ -369,7 +369,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
> >>  	unsigned long start = 0;
> >>  	unsigned long bitmap_maxno, bitmap_no, bitmap_count;
> >>  	struct page *page = NULL;
> >> -	int ret;
> >> +	int ret = -ENOMEM;
> >>  
> >>  	if (!cma || !cma->count)
> >>  		return NULL;
> >> @@ -427,6 +427,33 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
> >>  	trace_cma_alloc(pfn, page, count, align);
> >>  
> >>  	pr_debug("%s(): returned %p\n", __func__, page);
> >> +
> >> +	if (ret != 0) {
> >> +		unsigned int nr, nr_total = 0;
> >> +		unsigned long next_set_bit;
> >> +
> >> +		pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
> >> +			__func__, count, ret);
> >> +		mutex_lock(&cma->lock);
> >> +		printk("number of available pages: ");

I guess you want pr_info (or maybe pr_debug) here

> >> +		start = 0;
> >> +		for (;;) {
> >> +			bitmap_no = find_next_zero_bit(cma->bitmap, cma->count, start);
> >> +			next_set_bit = find_next_bit(cma->bitmap, cma->count, bitmap_no);
> >> +			nr = next_set_bit - bitmap_no;
> >> +			if (bitmap_no >= cma->count)
> >> +				break;
> >> +			if (nr_total == 0)
> >> +				printk("%u", nr);
> >> +			else
> >> +				printk("+%u", nr);

pr_cont

> >> +			nr_total += nr;
> >> +			start = bitmap_no + nr;
> >> +		}
> >> +		printk("=>%u pages, total: %lu pages\n", nr_total, cma->count);

pr_cont

> >> +		mutex_unlock(&cma->lock);
> >> +	}
> >> +
> >>  	return page;
> >>  }
> >>  
> >> -- 
> >> 1.9.1
> >>
> >> --
> >> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> >> the body to majordomo@kvack.org.  For more info on Linux MM,
> >> see: http://www.linux-mm.org/ .
> >> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
> 
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@kvack.org.  For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

-- 
Michal Hocko
SUSE Labs

Re: [PATCH] mm: cma: print allocation failure reason and bitmap status

[Jaewon Kim]

On 2016년 12월 29일 18:43, Michal Hocko wrote:
> On Thu 29-12-16 18:26:38, Jaewon Kim wrote:
>>
>> On 2016년 12월 29일 18:14, Michal Hocko wrote:
>>> On Thu 29-12-16 11:28:02, Jaewon Kim wrote:
>>>> There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR.
>>>> This patch prints the error value and bitmap status to know available pages
>>>> regarding fragmentation.
>>>>
>>>> This is an ENOMEM example with this patch.
>>>> [   11.616321]  [2:   Binder:711_1:  740] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12
>>>> [   11.616365]  [2:   Binder:711_1:  740] number of available pages: 4+7+7+8+38+166+127=>357 pages, total: 2048 pages
>>> Could you be more specific why this part is useful?
>> Hi
>> Without this patch we do not know why CMA allocation failed.
> Yes, I understand the first part
>
>> Additionally in case of ENOMEM, with bitmap status we can figure out that
> The code doesn't seem to check for ENOMEM though
Yes actually I wanted to look both ENOMEM case and EBUSY case.
Even in EBUSY case, we can look how much available pages existed, but all failed on those region because of EBUSY.
We may not need EINTR case, but I hope to look.
>
>> if it is too small CMA region issue or if it is fragmentation issue.
> then please describe that in the changelog. If I got it right the above
> would tell us that the fragmentation is the problem, right?
Yes fragmentation can be A problem, but bitmap status will explain EBUSY case too as I explained above.
>
>>>  
>>>> Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>
>>>> ---
>>>>  mm/cma.c | 29 ++++++++++++++++++++++++++++-
>>>>  1 file changed, 28 insertions(+), 1 deletion(-)
>>>>
>>>> diff --git a/mm/cma.c b/mm/cma.c
>>>> index c960459..535aa39 100644
>>>> --- a/mm/cma.c
>>>> +++ b/mm/cma.c
>>>> @@ -369,7 +369,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>>>>  	unsigned long start = 0;
>>>>  	unsigned long bitmap_maxno, bitmap_no, bitmap_count;
>>>>  	struct page *page = NULL;
>>>> -	int ret;
>>>> +	int ret = -ENOMEM;
>>>>  
>>>>  	if (!cma || !cma->count)
>>>>  		return NULL;
>>>> @@ -427,6 +427,33 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>>>>  	trace_cma_alloc(pfn, page, count, align);
>>>>  
>>>>  	pr_debug("%s(): returned %p\n", __func__, page);
>>>> +
>>>> +	if (ret != 0) {
>>>> +		unsigned int nr, nr_total = 0;
>>>> +		unsigned long next_set_bit;
>>>> +
>>>> +		pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
>>>> +			__func__, count, ret);
>>>> +		mutex_lock(&cma->lock);
>>>> +		printk("number of available pages: ");
> I guess you want pr_info (or maybe pr_debug) here
Thank you I will change as you and Michal Nazarewichz commented.
>
>>>> +		start = 0;
>>>> +		for (;;) {
>>>> +			bitmap_no = find_next_zero_bit(cma->bitmap, cma->count, start);
>>>> +			next_set_bit = find_next_bit(cma->bitmap, cma->count, bitmap_no);
>>>> +			nr = next_set_bit - bitmap_no;
>>>> +			if (bitmap_no >= cma->count)
>>>> +				break;
>>>> +			if (nr_total == 0)
>>>> +				printk("%u", nr);
>>>> +			else
>>>> +				printk("+%u", nr);
> pr_cont
>
>>>> +			nr_total += nr;
>>>> +			start = bitmap_no + nr;
>>>> +		}
>>>> +		printk("=>%u pages, total: %lu pages\n", nr_total, cma->count);
> pr_cont
>
>>>> +		mutex_unlock(&cma->lock);
>>>> +	}
>>>> +
>>>>  	return page;
>>>>  }
>>>>  
>>>> -- 
>>>> 1.9.1
>>>>
>>>> --
>>>> To unsubscribe, send a message with 'unsubscribe linux-mm' in
>>>> the body to majordomo@kvack.org.  For more info on Linux MM,
>>>> see: http://www.linux-mm.org/ .
>>>> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
>> --
>> To unsubscribe, send a message with 'unsubscribe linux-mm' in
>> the body to majordomo@kvack.org.  For more info on Linux MM,
>> see: http://www.linux-mm.org/ .
>> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] mm: cma: print allocation failure reason and bitmap status

[Michal Nazarewicz]

On Thu, Dec 29 2016, Michal Hocko wrote:
> On Thu 29-12-16 11:28:02, Jaewon Kim wrote:
>> There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR.
>> This patch prints the error value and bitmap status to know available pages
>> regarding fragmentation.
>> 
>> This is an ENOMEM example with this patch.
>> [   11.616321]  [2:   Binder:711_1:  740] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12
>
>> [   11.616365]  [2:   Binder:711_1:  740] number of available pages: 4+7+7+8+38+166+127=>357 pages, total: 2048 pages
>
> Could you be more specific why this part is useful?
>  
>> Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>
>> ---
>>  mm/cma.c | 29 ++++++++++++++++++++++++++++-
>>  1 file changed, 28 insertions(+), 1 deletion(-)
>> 
>> diff --git a/mm/cma.c b/mm/cma.c
>> index c960459..535aa39 100644
>> --- a/mm/cma.c
>> +++ b/mm/cma.c
>> @@ -369,7 +369,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>>  	unsigned long start = 0;
>>  	unsigned long bitmap_maxno, bitmap_no, bitmap_count;
>>  	struct page *page = NULL;
>> -	int ret;
>> +	int ret = -ENOMEM;
>>  
>>  	if (!cma || !cma->count)
>>  		return NULL;
>> @@ -427,6 +427,33 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>>  	trace_cma_alloc(pfn, page, count, align);
>>  
>>  	pr_debug("%s(): returned %p\n", __func__, page);
>> +
>> +	if (ret != 0) {
>> +		unsigned int nr, nr_total = 0;
>> +		unsigned long next_set_bit;
>> +
>> +		pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
>> +			__func__, count, ret);
>> +		mutex_lock(&cma->lock);
>> +		printk("number of available pages: ");
>> +		start = 0;
>> +		for (;;) {
>> +			bitmap_no = find_next_zero_bit(cma->bitmap, cma->count, start);
>> +			next_set_bit = find_next_bit(cma->bitmap, cma->count, bitmap_no);
>> +			nr = next_set_bit - bitmap_no;
>> +			if (bitmap_no >= cma->count)
>> +				break;

Put this just next to ‘bitmap_no = …’ line.  No need to call
find_next_bit if we’re gonna break anyway.

>> +			if (nr_total == 0)
>> +				printk("%u", nr);
>> +			else
>> +				printk("+%u", nr);

Perhaps also include location of the hole?  Something like:

		pr_cont("%s%u@%u", nr_total ? "+" : "", nr, bitmap_no);

>> +			nr_total += nr;
>> +			start = bitmap_no + nr;
>> +		}
>> +		printk("=>%u pages, total: %lu pages\n", nr_total, cma->count);
>> +		mutex_unlock(&cma->lock);
>> +	}
>> +

I wonder if this should be wrapped in

#ifdef CMA_DEBUG
…
#endif

On one hand it’s relatively expensive (even involving mutex locking) on
the other it’s in allocation failure path.

>>  	return page;
>>  }
>>  
>> -- 
>> 1.9.1
>> 
>> --
>> To unsubscribe, send a message with 'unsubscribe linux-mm' in
>> the body to majordomo@kvack.org.  For more info on Linux MM,
>> see: http://www.linux-mm.org/ .
>> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
>
> -- 
> Michal Hocko
> SUSE Labs

-- 
Best regards
ミハウ “𝓶𝓲𝓷𝓪86” ナザレヴイツ
«If at first you don’t succeed, give up skydiving»

Re: [PATCH] mm: cma: print allocation failure reason and bitmap status

[Jaewon Kim]

Hello Michal Hocko and and Michal Nazarewichz

On 2016년 12월 29일 23:20, Michal Nazarewicz wrote:
> On Thu, Dec 29 2016, Michal Hocko wrote:
>> On Thu 29-12-16 11:28:02, Jaewon Kim wrote:
>>> There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR.
>>> This patch prints the error value and bitmap status to know available pages
>>> regarding fragmentation.
>>>
>>> This is an ENOMEM example with this patch.
>>> [   11.616321]  [2:   Binder:711_1:  740] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12
>>> [   11.616365]  [2:   Binder:711_1:  740] number of available pages: 4+7+7+8+38+166+127=>357 pages, total: 2048 pages
>> Could you be more specific why this part is useful?
The first line is useful to know why the allocation failed.
Actually CMA internally try all available regions because some regions can be failed because of EBUSY.
The second showing bitmap status is useful to know in detail on both ENONEM and EBUSY;
 ENOMEM:  not tried at all because of no available region
 EBUSY:  tried some region but all failed
>>> Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>
>>> ---
>>>  mm/cma.c | 29 ++++++++++++++++++++++++++++-
>>>  1 file changed, 28 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/mm/cma.c b/mm/cma.c
>>> index c960459..535aa39 100644
>>> --- a/mm/cma.c
>>> +++ b/mm/cma.c
>>> @@ -369,7 +369,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>>>  	unsigned long start = 0;
>>>  	unsigned long bitmap_maxno, bitmap_no, bitmap_count;
>>>  	struct page *page = NULL;
>>> -	int ret;
>>> +	int ret = -ENOMEM;
>>>  
>>>  	if (!cma || !cma->count)
>>>  		return NULL;
>>> @@ -427,6 +427,33 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>>>  	trace_cma_alloc(pfn, page, count, align);
>>>  
>>>  	pr_debug("%s(): returned %p\n", __func__, page);
>>> +
>>> +	if (ret != 0) {
>>> +		unsigned int nr, nr_total = 0;
>>> +		unsigned long next_set_bit;
>>> +
>>> +		pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
>>> +			__func__, count, ret);
>>> +		mutex_lock(&cma->lock);
>>> +		printk("number of available pages: ");
>>> +		start = 0;
>>> +		for (;;) {
>>> +			bitmap_no = find_next_zero_bit(cma->bitmap, cma->count, start);
>>> +			next_set_bit = find_next_bit(cma->bitmap, cma->count, bitmap_no);
>>> +			nr = next_set_bit - bitmap_no;
>>> +			if (bitmap_no >= cma->count)
>>> +				break;
> Put this just next to ‘bitmap_no = …’ line.  No need to call
> find_next_bit if we’re gonna break anyway.
thank you I fixed
>>> +			if (nr_total == 0)
>>> +				printk("%u", nr);
>>> +			else
>>> +				printk("+%u", nr);
> Perhaps also include location of the hole?  Something like:
>
> 		pr_cont("%s%u@%u", nr_total ? "+" : "", nr, bitmap_no);
Thank you I fixed with @%lu
>
>>> +			nr_total += nr;
>>> +			start = bitmap_no + nr;
>>> +		}
>>> +		printk("=>%u pages, total: %lu pages\n", nr_total, cma->count);
>>> +		mutex_unlock(&cma->lock);
>>> +	}
>>> +
> I wonder if this should be wrapped in
>
> #ifdef CMA_DEBUG
> …
> #endif
>
> On one hand it’s relatively expensive (even involving mutex locking) on
> the other it’s in allocation failure path.
bitmap status, I think, could be in side of CMA_DEBUG with the mutex
but the first error log, I hope, to be out of CMA_DEBUG.
>
>>>  	return page;
>>>  }
>>>  
>>> -- 
>>> 1.9.1
>>>
>>> --
>>> To unsubscribe, send a message with 'unsubscribe linux-mm' in
>>> the body to majordomo@kvack.org.  For more info on Linux MM,
>>> see: http://www.linux-mm.org/ .
>>> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
>> -- 
>> Michal Hocko
>> SUSE Labs
This is fixed patch following your comment.
Please review again
If it is OK, let me know whether I need to resend this patch as a new mail thread.


>From 7577cc94da3af27907aa6eec590d2ef51e4b9d80 Mon Sep 17 00:00:00 2001
From: Jaewon Kim <jaewon31.kim@samsung.com>
Date: Thu, 29 Dec 2016 11:00:16 +0900
Subject: [PATCH] mm: cma: print allocation failure reason and bitmap status

There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR.
But we did not know error reason so far. This patch prints the error value.

Additionally if CONFIG_CMA_DEBUG is enabled, this patch shows bitmap status to
know available pages. Actually CMA internally try all available regions because
some regions can be failed because of EBUSY. Bitmap status is useful to know in
detail on both ENONEM and EBUSY;
 ENOMEM: not tried at all because of no available region
         it could be too small total region or could be fragmentation issue
 EBUSY:  tried some region but all failed

This is an ENOMEM example with this patch.
[   13.250961]  [1:   Binder:715_1:  846] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12
Avabile pages also will be shown if CONFIG_CMA_DEBUG is enabled
[   13.251052]  [1:   Binder:715_1:  846] cma: number of available pages: 4@572+7@585+7@601+8@632+38@730+166@1114+127@1921=>357 pages, total: 2048 pages

Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>
---
 mm/cma.c | 29 ++++++++++++++++++++++++++++-
 1 file changed, 28 insertions(+), 1 deletion(-)

diff --git a/mm/cma.c b/mm/cma.c
index c960459..1bcd9db 100644
--- a/mm/cma.c
+++ b/mm/cma.c
@@ -369,7 +369,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
     unsigned long start = 0;
     unsigned long bitmap_maxno, bitmap_no, bitmap_count;
     struct page *page = NULL;
-    int ret;
+    int ret = -ENOMEM;
 
     if (!cma || !cma->count)
         return NULL;
@@ -427,6 +427,33 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
     trace_cma_alloc(pfn, page, count, align);
 
     pr_debug("%s(): returned %p\n", __func__, page);
+
+    if (ret != 0)
+        pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
+            __func__, count, ret);
+#ifdef CONFIG_CMA_DEBUG
+    if (ret != 0) {
+        unsigned int nr, nr_total = 0;
+        unsigned long next_set_bit;
+
+        mutex_lock(&cma->lock);
+        pr_info("number of available pages: ");
+        start = 0;
+        for (;;) {
+            bitmap_no = find_next_zero_bit(cma->bitmap, cma->count, start);
+            if (bitmap_no >= cma->count)
+                break;
+            next_set_bit = find_next_bit(cma->bitmap, cma->count, bitmap_no);
+            nr = next_set_bit - bitmap_no;
+            pr_cont("%s%u@%lu", nr_total ? "+" : "", nr, bitmap_no);
+            nr_total += nr;
+            start = bitmap_no + nr;
+        }
+        pr_cont("=>%u pages, total: %lu pages\n", nr_total, cma->count);
+        mutex_unlock(&cma->lock);
+    }
+#endif
+
     return page;
 }
 
-- 
1.9.1

Re: [PATCH] mm: cma: print allocation failure reason and bitmap status

[Michal Hocko]

On Fri 30-12-16 16:24:46, Jaewon Kim wrote:
[...]
> >From 7577cc94da3af27907aa6eec590d2ef51e4b9d80 Mon Sep 17 00:00:00 2001
> From: Jaewon Kim <jaewon31.kim@samsung.com>
> Date: Thu, 29 Dec 2016 11:00:16 +0900
> Subject: [PATCH] mm: cma: print allocation failure reason and bitmap status
> 
> There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR.
> But we did not know error reason so far. This patch prints the error value.
> 
> Additionally if CONFIG_CMA_DEBUG is enabled, this patch shows bitmap status to
> know available pages. Actually CMA internally try all available regions because
> some regions can be failed because of EBUSY. Bitmap status is useful to know in
> detail on both ENONEM and EBUSY;
>  ENOMEM: not tried at all because of no available region
>          it could be too small total region or could be fragmentation issue
>  EBUSY:  tried some region but all failed
> 
> This is an ENOMEM example with this patch.
> [   13.250961]  [1:   Binder:715_1:  846] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12
> Avabile pages also will be shown if CONFIG_CMA_DEBUG is enabled
> [   13.251052]  [1:   Binder:715_1:  846] cma: number of available pages: 4@572+7@585+7@601+8@632+38@730+166@1114+127@1921=>357 pages, total: 2048 pages

please mention how to interpret this information.

some more style suggestions below
> 
> Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>
> ---
>  mm/cma.c | 29 ++++++++++++++++++++++++++++-
>  1 file changed, 28 insertions(+), 1 deletion(-)
> 
> diff --git a/mm/cma.c b/mm/cma.c
> index c960459..1bcd9db 100644
> --- a/mm/cma.c
> +++ b/mm/cma.c
> @@ -369,7 +369,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>      unsigned long start = 0;
>      unsigned long bitmap_maxno, bitmap_no, bitmap_count;
>      struct page *page = NULL;
> -    int ret;
> +    int ret = -ENOMEM;
>  
>      if (!cma || !cma->count)
>          return NULL;
> @@ -427,6 +427,33 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>      trace_cma_alloc(pfn, page, count, align);
>  
>      pr_debug("%s(): returned %p\n", __func__, page);
> +
> +    if (ret != 0)

you can simply do
	if (!ret) {

		pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
			__func__, count, ret);
		debug_show_cma_areas();
	}

	return page;

static void debug_show_cma_areas(void)
{
#ifdef CONFIG_CMA_DEBUG
	unsigned int nr, nr_total = 0;
	unsigned long next_set_bit;

	mutex_lock(&cma->lock);
	pr_info("number of available pages: ");
	start = 0;
	for (;;) {
		bitmap_no = find_next_zero_bit(cma->bitmap, cma->count, start);
		if (bitmap_no >= cma->count)
		break;
		next_set_bit = find_next_bit(cma->bitmap, cma->count, bitmap_no);
		nr = next_set_bit - bitmap_no;
		pr_cont("%s%u@%lu", nr_total ? "+" : "", nr, bitmap_no);
		nr_total += nr;
		start = bitmap_no + nr;
	}
	pr_cont("=>%u pages, total: %lu pages\n", nr_total, cma->count);
	mutex_unlock(&cma->lock);
#endif
}

-- 
Michal Hocko
SUSE Labs

Re: [PATCH] mm: cma: print allocation failure reason and bitmap status

[Michal Nazarewicz]

On Fri, Dec 30 2016, Michal Hocko wrote:
> On Fri 30-12-16 16:24:46, Jaewon Kim wrote:
> [...]
>> >From 7577cc94da3af27907aa6eec590d2ef51e4b9d80 Mon Sep 17 00:00:00 2001
>> From: Jaewon Kim <jaewon31.kim@samsung.com>
>> Date: Thu, 29 Dec 2016 11:00:16 +0900
>> Subject: [PATCH] mm: cma: print allocation failure reason and bitmap status
>> 
>> There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR.
>> But we did not know error reason so far. This patch prints the error value.
>> 
>> Additionally if CONFIG_CMA_DEBUG is enabled, this patch shows bitmap status to
>> know available pages. Actually CMA internally try all available regions because
>> some regions can be failed because of EBUSY. Bitmap status is useful to know in
>> detail on both ENONEM and EBUSY;
>>  ENOMEM: not tried at all because of no available region
>>          it could be too small total region or could be fragmentation issue
>>  EBUSY:  tried some region but all failed
>> 
>> This is an ENOMEM example with this patch.
>> [   13.250961]  [1:   Binder:715_1:  846] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12
>> Avabile pages also will be shown if CONFIG_CMA_DEBUG is enabled
>> [   13.251052]  [1:   Binder:715_1:  846] cma: number of available pages: 4@572+7@585+7@601+8@632+38@730+166@1114+127@1921=>357 pages, total: 2048 pages
>
> please mention how to interpret this information.
>
> some more style suggestions below
>> 
>> Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>

Acked-by: Michal Nazarewicz <mina86@mina86.com>

>> ---
>>  mm/cma.c | 29 ++++++++++++++++++++++++++++-
>>  1 file changed, 28 insertions(+), 1 deletion(-)
>> 
>> diff --git a/mm/cma.c b/mm/cma.c
>> index c960459..1bcd9db 100644
>> --- a/mm/cma.c
>> +++ b/mm/cma.c
>> @@ -369,7 +369,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>>      unsigned long start = 0;
>>      unsigned long bitmap_maxno, bitmap_no, bitmap_count;
>>      struct page *page = NULL;
>> -    int ret;
>> +    int ret = -ENOMEM;
>>  
>>      if (!cma || !cma->count)
>>          return NULL;
>> @@ -427,6 +427,33 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>>      trace_cma_alloc(pfn, page, count, align);
>>  
>>      pr_debug("%s(): returned %p\n", __func__, page);

This line should be moved after the ‘if (ret != 0)’ block, i.e. just
before return.

>> +
>> +    if (ret != 0)
>
> you can simply do
> 	if (!ret) {
>
> 		pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
> 			__func__, count, ret);
> 		debug_show_cma_areas();
> 	}
>
> 	return page;
>
> static void debug_show_cma_areas(void)
> {
> #ifdef CONFIG_CMA_DEBUG
> 	unsigned int nr, nr_total = 0;
> 	unsigned long next_set_bit;
>
> 	mutex_lock(&cma->lock);
> 	pr_info("number of available pages: ");
> 	start = 0;
> 	for (;;) {
> 		bitmap_no = find_next_zero_bit(cma->bitmap, cma->count, start);
> 		if (bitmap_no >= cma->count)
> 		break;
> 		next_set_bit = find_next_bit(cma->bitmap, cma->count, bitmap_no);
> 		nr = next_set_bit - bitmap_no;
> 		pr_cont("%s%u@%lu", nr_total ? "+" : "", nr, bitmap_no);
> 		nr_total += nr;
> 		start = bitmap_no + nr;
> 	}
> 	pr_cont("=>%u pages, total: %lu pages\n", nr_total, cma->count);

Perhaps:
	pr_cont("=> %u free of %lu total pages\n", nr_total, cma->count);
or shorter (but more cryptic):
	pr_cont("=> %u/%lu pages\n", nr_total, cma->count);

> 	mutex_unlock(&cma->lock);
> #endif
> }

Actually, Linux style is more like:

#ifdef CONFIG_CMA_DEBUG
static void cma_debug_show_areas()
{
	…
}
#else
static inline void cma_debug_show_areas() { }
#endif

>
> -- 
> Michal Hocko
> SUSE Labs

-- 
Best regards
ミハウ “𝓶𝓲𝓷𝓪86” ナザレヴイツ
«If at first you don’t succeed, give up skydiving»

Re: [PATCH] mm: cma: print allocation failure reason and bitmap status

[Jaewon Kim]

On 2017년 01월 02일 06:59, Michal Nazarewicz wrote:
> On Fri, Dec 30 2016, Michal Hocko wrote:
>> On Fri 30-12-16 16:24:46, Jaewon Kim wrote:
>> [...]
>>> >From 7577cc94da3af27907aa6eec590d2ef51e4b9d80 Mon Sep 17 00:00:00 2001
>>> From: Jaewon Kim <jaewon31.kim@samsung.com>
>>> Date: Thu, 29 Dec 2016 11:00:16 +0900
>>> Subject: [PATCH] mm: cma: print allocation failure reason and bitmap status
>>>
>>> There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR.
>>> But we did not know error reason so far. This patch prints the error value.
>>>
>>> Additionally if CONFIG_CMA_DEBUG is enabled, this patch shows bitmap status to
>>> know available pages. Actually CMA internally try all available regions because
>>> some regions can be failed because of EBUSY. Bitmap status is useful to know in
>>> detail on both ENONEM and EBUSY;
>>>  ENOMEM: not tried at all because of no available region
>>>          it could be too small total region or could be fragmentation issue
>>>  EBUSY:  tried some region but all failed
>>>
>>> This is an ENOMEM example with this patch.
>>> [   13.250961]  [1:   Binder:715_1:  846] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12
>>> Avabile pages also will be shown if CONFIG_CMA_DEBUG is enabled
>>> [   13.251052]  [1:   Binder:715_1:  846] cma: number of available pages: 4@572+7@585+7@601+8@632+38@730+166@1114+127@1921=>357 pages, total: 2048 pages
>> please mention how to interpret this information.
Thank you Michal Hocko. I added like this
If CONFIG_CMA_DEBUG is enabled, avabile pages also will be shown as concatenated
size@position format. So 4@572 means that there are 4 available pages at 572
position starting from 0 position.
>>
>> some more style suggestions below
>>> Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>
> Acked-by: Michal Nazarewicz <mina86@mina86.com>
Thank you I added your Ack
Acked-by: Michal Nazarewicz <mina86@mina86.com>
>>> ---
>>>  mm/cma.c | 29 ++++++++++++++++++++++++++++-
>>>  1 file changed, 28 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/mm/cma.c b/mm/cma.c
>>> index c960459..1bcd9db 100644
>>> --- a/mm/cma.c
>>> +++ b/mm/cma.c
>>> @@ -369,7 +369,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>>>      unsigned long start = 0;
>>>      unsigned long bitmap_maxno, bitmap_no, bitmap_count;
>>>      struct page *page = NULL;
>>> -    int ret;
>>> +    int ret = -ENOMEM;
>>>  
>>>      if (!cma || !cma->count)
>>>          return NULL;
>>> @@ -427,6 +427,33 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>>>      trace_cma_alloc(pfn, page, count, align);
>>>  
>>>      pr_debug("%s(): returned %p\n", __func__, page);
> This line should be moved after the ‘if (ret != 0)’ block, i.e. just
> before return.
Thank you Michal Nazarewicz
I moved the pr_debug right before return
     pr_debug("%s(): returned %p\n", __func__, page);
     return page;
>>> +
>>> +    if (ret != 0)
>> you can simply do
>> 	if (!ret) {
Thank you
I changed like this, it should be if(ret) rather than if(!ret)
+    if (ret) {
+        pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
+            __func__, count, ret);
+        debug_show_cma_areas(cma);
+    }
>>
>> 		pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
>> 			__func__, count, ret);
>> 		debug_show_cma_areas();
>> 	}
>>
>> 	return page;
>>
>> static void debug_show_cma_areas(void)
>> {
>> #ifdef CONFIG_CMA_DEBUG
>> 	unsigned int nr, nr_total = 0;
>> 	unsigned long next_set_bit;
>>
>> 	mutex_lock(&cma->lock);
>> 	pr_info("number of available pages: ");
>> 	start = 0;
>> 	for (;;) {
>> 		bitmap_no = find_next_zero_bit(cma->bitmap, cma->count, start);
>> 		if (bitmap_no >= cma->count)
>> 		break;
>> 		next_set_bit = find_next_bit(cma->bitmap, cma->count, bitmap_no);
>> 		nr = next_set_bit - bitmap_no;
>> 		pr_cont("%s%u@%lu", nr_total ? "+" : "", nr, bitmap_no);
>> 		nr_total += nr;
>> 		start = bitmap_no + nr;
>> 	}
>> 	pr_cont("=>%u pages, total: %lu pages\n", nr_total, cma->count);
> Perhaps:
> 	pr_cont("=> %u free of %lu total pages\n", nr_total, cma->count);
Thank you I will take this way.
+    pr_cont("=> %u free of %lu total pages\n", nr_total, cma->count);
> or shorter (but more cryptic):
> 	pr_cont("=> %u/%lu pages\n", nr_total, cma->count);
>
>> 	mutex_unlock(&cma->lock);
>> #endif
>> }
> Actually, Linux style is more like:
>
> #ifdef CONFIG_CMA_DEBUG
> static void cma_debug_show_areas()
> {
> 	…
> }
> #else
> static inline void cma_debug_show_areas() { }
> #endif
Thank you I will take this way. FYI struct cma address should be passed as a argument.
+#ifdef CONFIG_CMA_DEBUG
+static void debug_show_cma_areas(struct cma *cma)
+{
...
+#else
+static inline void debug_show_cma_areas(struct cma *cma) { }
+#endif
>> -- 
>> Michal Hocko
>> SUSE Labs

Thank you all of you.
Let me reattach my full patch to be clear.
Let me know if I resend this as a new mail thread.

>From 03bf46bac222587f66fe9935a7879da30ff42cff Mon Sep 17 00:00:00 2001
From: Jaewon Kim <jaewon31.kim@samsung.com>
Date: Thu, 29 Dec 2016 11:00:16 +0900
Subject: [PATCH] mm: cma: print allocation failure reason and bitmap status

There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR.
But we did not know error reason so far. This patch prints the error value.

Additionally if CONFIG_CMA_DEBUG is enabled, this patch shows bitmap status to
know available pages. Actually CMA internally tries on all available regions
because some regions can be failed because of EBUSY. Bitmap status is useful to
know in detail on both ENONEM and EBUSY;
 ENOMEM: not tried at all because of no available region
         it could be too small total region or could be fragmentation issue
 EBUSY:  tried some region but all failed

This is an ENOMEM example with this patch.
[   12.415458]  [2:   Binder:714_1:  744] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12
If CONFIG_CMA_DEBUG is enabled, avabile pages also will be shown as concatenated
size@position format. So 4@572 means that there are 4 available pages at 572
position starting from 0 position.
[   12.415503]  [2:   Binder:714_1:  744] cma: number of available pages: 4@572+7@585+7@601+8@632+38@730+166@1114+127@1921=> 357 free of 2048 total pages

Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>
Acked-by: Michal Nazarewicz <mina86@mina86.com>
---
 mm/cma.c | 34 +++++++++++++++++++++++++++++++++-
 1 file changed, 33 insertions(+), 1 deletion(-)

diff --git a/mm/cma.c b/mm/cma.c
index c960459..9e037541 100644
--- a/mm/cma.c
+++ b/mm/cma.c
@@ -353,6 +353,32 @@ int __init cma_declare_contiguous(phys_addr_t base,
     return ret;
 }
 
+#ifdef CONFIG_CMA_DEBUG
+static void debug_show_cma_areas(struct cma *cma)
+{
+    unsigned long next_zero_bit, next_set_bit;
+    unsigned long start = 0;
+    unsigned int nr_zero, nr_total = 0;
+
+    mutex_lock(&cma->lock);
+    pr_info("number of available pages: ");
+    for (;;) {
+        next_zero_bit = find_next_zero_bit(cma->bitmap, cma->count, start);
+        if (next_zero_bit >= cma->count)
+            break;
+        next_set_bit = find_next_bit(cma->bitmap, cma->count, next_zero_bit);
+        nr_zero = next_set_bit - next_zero_bit;
+        pr_cont("%s%u@%lu", nr_total ? "+" : "", nr_zero, next_zero_bit);
+        nr_total += nr_zero;
+        start = next_zero_bit + nr_zero;
+    }
+    pr_cont("=> %u free of %lu total pages\n", nr_total, cma->count);
+    mutex_unlock(&cma->lock);
+}
+#else
+static inline void debug_show_cma_areas(struct cma *cma) { }
+#endif
+
 /**
  * cma_alloc() - allocate pages from contiguous area
  * @cma:   Contiguous memory region for which the allocation is performed.
@@ -369,7 +395,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
     unsigned long start = 0;
     unsigned long bitmap_maxno, bitmap_no, bitmap_count;
     struct page *page = NULL;
-    int ret;
+    int ret = -ENOMEM;
 
     if (!cma || !cma->count)
         return NULL;
@@ -426,6 +452,12 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
 
     trace_cma_alloc(pfn, page, count, align);
 
+    if (ret) {
+        pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
+            __func__, count, ret);
+        debug_show_cma_areas(cma);
+    }
+
     pr_debug("%s(): returned %p\n", __func__, page);
     return page;
 }
-- 

Re: [PATCH] mm: cma: print allocation failure reason and bitmap status

[Michal Nazarewicz]

On Mon, Jan 02 2017, Jaewon Kim wrote:
> There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR.
> But we did not know error reason so far. This patch prints the error value.
>
> Additionally if CONFIG_CMA_DEBUG is enabled, this patch shows bitmap status to
> know available pages. Actually CMA internally tries on all available regions
> because some regions can be failed because of EBUSY. Bitmap status is useful to
> know in detail on both ENONEM and EBUSY;
>  ENOMEM: not tried at all because of no available region
>          it could be too small total region or could be fragmentation issue
>  EBUSY:  tried some region but all failed
>
> This is an ENOMEM example with this patch.
> [   12.415458]  [2:   Binder:714_1:  744] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12
> If CONFIG_CMA_DEBUG is enabled, avabile pages also will be shown as concatenated
> size@position format. So 4@572 means that there are 4 available pages at 572
> position starting from 0 position.
> [   12.415503]  [2:   Binder:714_1:  744] cma: number of available pages: 4@572+7@585+7@601+8@632+38@730+166@1114+127@1921=> 357 free of 2048 total pages
>
> Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>
> Acked-by: Michal Nazarewicz <mina86@mina86.com>
> ---
>  mm/cma.c | 34 +++++++++++++++++++++++++++++++++-
>  1 file changed, 33 insertions(+), 1 deletion(-)
>
> diff --git a/mm/cma.c b/mm/cma.c
> index c960459..9e037541 100644
> --- a/mm/cma.c
> +++ b/mm/cma.c
> @@ -353,6 +353,32 @@ int __init cma_declare_contiguous(phys_addr_t base,
>      return ret;
>  }
>  
> +#ifdef CONFIG_CMA_DEBUG
> +static void debug_show_cma_areas(struct cma *cma)

Make it ‘cma_debug_show_areas’.  All other functions have ‘cma’ as
prefix so that’s more consistent.

> +{
> +    unsigned long next_zero_bit, next_set_bit;
> +    unsigned long start = 0;
> +    unsigned int nr_zero, nr_total = 0;
> +
> +    mutex_lock(&cma->lock);
> +    pr_info("number of available pages: ");
> +    for (;;) {
> +        next_zero_bit = find_next_zero_bit(cma->bitmap, cma->count, start);
> +        if (next_zero_bit >= cma->count)
> +            break;
> +        next_set_bit = find_next_bit(cma->bitmap, cma->count, next_zero_bit);
> +        nr_zero = next_set_bit - next_zero_bit;
> +        pr_cont("%s%u@%lu", nr_total ? "+" : "", nr_zero, next_zero_bit);
> +        nr_total += nr_zero;
> +        start = next_zero_bit + nr_zero;
> +    }
> +    pr_cont("=> %u free of %lu total pages\n", nr_total, cma->count);
> +    mutex_unlock(&cma->lock);
> +}
> +#else
> +static inline void debug_show_cma_areas(struct cma *cma) { }
> +#endif
> +
>  /**
>   * cma_alloc() - allocate pages from contiguous area
>   * @cma:   Contiguous memory region for which the allocation is performed.
> @@ -369,7 +395,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>      unsigned long start = 0;
>      unsigned long bitmap_maxno, bitmap_no, bitmap_count;
>      struct page *page = NULL;
> -    int ret;
> +    int ret = -ENOMEM;
>  
>      if (!cma || !cma->count)
>          return NULL;
> @@ -426,6 +452,12 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>  
>      trace_cma_alloc(pfn, page, count, align);
>  
> +    if (ret) {
> +        pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
> +            __func__, count, ret);
> +        debug_show_cma_areas(cma);
> +    }
> +
>      pr_debug("%s(): returned %p\n", __func__, page);
>      return page;
>  }
> -- 
>

-- 
Best regards
ミハウ “𝓶𝓲𝓷𝓪86” ナザレヴイツ
«If at first you don’t succeed, give up skydiving»

Re: [PATCH] mm: cma: print allocation failure reason and bitmap status

[Jaewon Kim]

On 2017년 01월 02일 15:46, Michal Nazarewicz wrote:
> On Mon, Jan 02 2017, Jaewon Kim wrote:
>> There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR.
>> But we did not know error reason so far. This patch prints the error value.
>>
>> Additionally if CONFIG_CMA_DEBUG is enabled, this patch shows bitmap status to
>> know available pages. Actually CMA internally tries on all available regions
>> because some regions can be failed because of EBUSY. Bitmap status is useful to
>> know in detail on both ENONEM and EBUSY;
>>  ENOMEM: not tried at all because of no available region
>>          it could be too small total region or could be fragmentation issue
>>  EBUSY:  tried some region but all failed
>>
>> This is an ENOMEM example with this patch.
>> [   12.415458]  [2:   Binder:714_1:  744] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12
>> If CONFIG_CMA_DEBUG is enabled, avabile pages also will be shown as concatenated
>> size@position format. So 4@572 means that there are 4 available pages at 572
>> position starting from 0 position.
>> [   12.415503]  [2:   Binder:714_1:  744] cma: number of available pages: 4@572+7@585+7@601+8@632+38@730+166@1114+127@1921=> 357 free of 2048 total pages
>>
>> Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>
>> Acked-by: Michal Nazarewicz <mina86@mina86.com>
>> ---
>>  mm/cma.c | 34 +++++++++++++++++++++++++++++++++-
>>  1 file changed, 33 insertions(+), 1 deletion(-)
>>
>> diff --git a/mm/cma.c b/mm/cma.c
>> index c960459..9e037541 100644
>> --- a/mm/cma.c
>> +++ b/mm/cma.c
>> @@ -353,6 +353,32 @@ int __init cma_declare_contiguous(phys_addr_t base,
>>      return ret;
>>  }
>>  
>> +#ifdef CONFIG_CMA_DEBUG
>> +static void debug_show_cma_areas(struct cma *cma)
> Make it ‘cma_debug_show_areas’.  All other functions have ‘cma’ as
> prefix so that’s more consistent.
OK no problem.
>
>> +{
>> +    unsigned long next_zero_bit, next_set_bit;
>> +    unsigned long start = 0;
>> +    unsigned int nr_zero, nr_total = 0;
>> +
>> +    mutex_lock(&cma->lock);
>> +    pr_info("number of available pages: ");
>> +    for (;;) {
>> +        next_zero_bit = find_next_zero_bit(cma->bitmap, cma->count, start);
>> +        if (next_zero_bit >= cma->count)
>> +            break;
>> +        next_set_bit = find_next_bit(cma->bitmap, cma->count, next_zero_bit);
>> +        nr_zero = next_set_bit - next_zero_bit;
>> +        pr_cont("%s%u@%lu", nr_total ? "+" : "", nr_zero, next_zero_bit);
>> +        nr_total += nr_zero;
>> +        start = next_zero_bit + nr_zero;
>> +    }
>> +    pr_cont("=> %u free of %lu total pages\n", nr_total, cma->count);
>> +    mutex_unlock(&cma->lock);
>> +}
>> +#else
>> +static inline void debug_show_cma_areas(struct cma *cma) { }
>> +#endif
>> +
>>  /**
>>   * cma_alloc() - allocate pages from contiguous area
>>   * @cma:   Contiguous memory region for which the allocation is performed.
>> @@ -369,7 +395,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>>      unsigned long start = 0;
>>      unsigned long bitmap_maxno, bitmap_no, bitmap_count;
>>      struct page *page = NULL;
>> -    int ret;
>> +    int ret = -ENOMEM;
>>  
>>      if (!cma || !cma->count)
>>          return NULL;
>> @@ -426,6 +452,12 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
>>  
>>      trace_cma_alloc(pfn, page, count, align);
>>  
>> +    if (ret) {
>> +        pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
>> +            __func__, count, ret);
>> +        debug_show_cma_areas(cma);
>> +    }
>> +
>>      pr_debug("%s(): returned %p\n", __func__, page);
>>      return page;
>>  }
>> -- 
>>

Added the latest.

>From ec724faf7ace5e690715e23e82f17f27c697b6b1 Mon Sep 17 00:00:00 2001
From: Jaewon Kim <jaewon31.kim@samsung.com>
Date: Thu, 29 Dec 2016 11:00:16 +0900
Subject: [PATCH] mm: cma: print allocation failure reason and bitmap status

There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR.
But we did not know error reason so far. This patch prints the error value.

Additionally if CONFIG_CMA_DEBUG is enabled, this patch shows bitmap status to
know available pages. Actually CMA internally tries on all available regions
because some regions can be failed because of EBUSY. Bitmap status is useful to
know in detail on both ENONEM and EBUSY;
 ENOMEM: not tried at all because of no available region
         it could be too small total region or could be fragmentation issue
 EBUSY:  tried some region but all failed

This is an ENOMEM example with this patch.
[   12.415458]  [2:   Binder:714_1:  744] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12
If CONFIG_CMA_DEBUG is enabled, avabile pages also will be shown as concatenated
size@position format. So 4@572 means that there are 4 available pages at 572
position starting from 0 position.
[   12.415503]  [2:   Binder:714_1:  744] cma: number of available pages: 4@572+7@585+7@601+8@632+38@730+166@1114+127@1921=> 357 free of 2048 total pages

Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>
Acked-by: Michal Nazarewicz <mina86@mina86.com>
---
 mm/cma.c | 34 +++++++++++++++++++++++++++++++++-
 1 file changed, 33 insertions(+), 1 deletion(-)

diff --git a/mm/cma.c b/mm/cma.c
index c960459..c393229 100644
--- a/mm/cma.c
+++ b/mm/cma.c
@@ -353,6 +353,32 @@ int __init cma_declare_contiguous(phys_addr_t base,
     return ret;
 }
 
+#ifdef CONFIG_CMA_DEBUG
+static void cma_debug_show_areas(struct cma *cma)
+{
+    unsigned long next_zero_bit, next_set_bit;
+    unsigned long start = 0;
+    unsigned int nr_zero, nr_total = 0;
+
+    mutex_lock(&cma->lock);
+    pr_info("number of available pages: ");
+    for (;;) {
+        next_zero_bit = find_next_zero_bit(cma->bitmap, cma->count, start);
+        if (next_zero_bit >= cma->count)
+            break;
+        next_set_bit = find_next_bit(cma->bitmap, cma->count, next_zero_bit);
+        nr_zero = next_set_bit - next_zero_bit;
+        pr_cont("%s%u@%lu", nr_total ? "+" : "", nr_zero, next_zero_bit);
+        nr_total += nr_zero;
+        start = next_zero_bit + nr_zero;
+    }
+    pr_cont("=> %u free of %lu total pages\n", nr_total, cma->count);
+    mutex_unlock(&cma->lock);
+}
+#else
+static inline void cma_debug_show_areas(struct cma *cma) { }
+#endif
+
 /**
  * cma_alloc() - allocate pages from contiguous area
  * @cma:   Contiguous memory region for which the allocation is performed.
@@ -369,7 +395,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
     unsigned long start = 0;
     unsigned long bitmap_maxno, bitmap_no, bitmap_count;
     struct page *page = NULL;
-    int ret;
+    int ret = -ENOMEM;
 
     if (!cma || !cma->count)
         return NULL;
@@ -426,6 +452,12 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
 
     trace_cma_alloc(pfn, page, count, align);
 
+    if (ret) {
+        pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
+            __func__, count, ret);
+        cma_debug_show_areas(cma);
+    }
+
     pr_debug("%s(): returned %p\n", __func__, page);
     return page;
 }
-- 
1.9.1

Re: [PATCH] mm: cma: print allocation failure reason and bitmap status

[Michal Hocko]

On Sun 01-01-17 22:59:40, Michal Nazarewicz wrote:
[...]
> Actually, Linux style is more like:
> 
> #ifdef CONFIG_CMA_DEBUG
> static void cma_debug_show_areas()
> {
> 	…
> }
> #else
> static inline void cma_debug_show_areas() { }
> #endif

yes, we usually do that when the function is externally visible. Inline
ifdef for static functions saves few lines. Not that it would matter
much though.
-- 
Michal Hocko
SUSE Labs