Revert "perf: Remove the extra validity check on nr_pages"

From: Kan Liang <[email protected]>

This reverts commit 74390aa55678 ("perf: Remove the extra validity check on nr_pages")

nr_pages equals to number of pages - 1 in perf_mmap. So nr_pages = 0 is valide. So the nr_pages != 0 && !is_power_of_2(nr_pages) are all needed for checking. Otherwise, for example, perf test 6 failed. perf test 6
6: x86 rdpmc test :Error:
mmap() syscall returned with (Invalid argument)
FAILED!

Signed-off-by Kan Liang <[email protected]>

---
 kernel/events/core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

kernel/events/core.c +1 -1

index 89f0f16..a83581c 100644
@@ -4423,7 +4423,7 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma)
 	 * If we have rb pages ensure they're a power-of-two number, so we
 	 * can do bitmasks instead of modulo.
 	 */
-	if (!is_power_of_2(nr_pages))
+	if (nr_pages != 0 && !is_power_of_2(nr_pages))
 		return -EINVAL;
 
 	if (vma_size != PAGE_SIZE * (1 + nr_pages))

Signature · 1 line

1.8.3.1

kernel/events/core.c static int perf_mmap(struct file *file, struct

Quoted 4 lines

kernel/events/core.c +0 -0

@@ -4446,7 +4446,7 @@ static int perf_mmap(struct file *file, struct

vm_area_struct *vma)

kernel/events/core.c static int perf_mmap(struct file *file, struct · line 4446

Quoted 1 line

kernel/events/core.c +0 -0

 	 * If we have rb pages ensure they're a power-of-two number, so

we

kernel/events/core.c static int perf_mmap(struct file *file, struct vm_area_struct *vma) · line 4450

Quoted 5 lines

kernel/events/core.c +1 -1

 	 * can do bitmasks instead of modulo.
 	 */
-	if (!is_power_of_2(nr_pages))
+	if (nr_pages != 0 && !is_power_of_2(nr_pages))
 		return -EINVAL;

Hm, what does is_power_of_2(0) return? It should return 0, because 0 is not a power of 2!

and if it's fixed to return 0, then the check should properly be something like:

	if (!nr_pages || !is_power_of_2(nr_pages))

or so?

Thanks,

Ingo

Quoted 22 lines · tip-bot for Kan Liang

kernel/events/core.c static int perf_mmap(struct file *file, struct

Quoted 4 lines

kernel/events/core.c +0 -0

@@ -4446,7 +4446,7 @@ static int perf_mmap(struct file *file, struct

vm_area_struct *vma)

kernel/events/core.c static int perf_mmap(struct file *file, struct · line 4446

Quoted 1 line

kernel/events/core.c +0 -0

 	 * If we have rb pages ensure they're a power-of-two number, so

we

kernel/events/core.c static int perf_mmap(struct file *file, struct vm_area_struct *vma) · line 4450

Quoted 5 lines

kernel/events/core.c +1 -1

 	 * can do bitmasks instead of modulo.
 	 */
-	if (!is_power_of_2(nr_pages))
+	if (nr_pages != 0 && !is_power_of_2(nr_pages))
 		return -EINVAL;

Hm, what does is_power_of_2(0) return? It should return 0, because 0 is not a power of 2!

and if it's fixed to return 0, then the check should properly be something like:

	if (!nr_pages || !is_power_of_2(nr_pages))

Quoted 1 line

or so?

nr_pages == 0 is valid, we cannot return -EINVAL. Because

	vma_size = vma->vm_end - vma->vm_start;
	nr_pages = (vma_size / PAGE_SIZE) - 1;

So here we only want to check is_power_of_2
when the pages > 1 (or nr_pages > 0).

Quoted 2 lines

Hm, what does is_power_of_2(0) return? It should return 0, because 0 is not a power of 2!

Quoted 8 lines

Quoted 8 lines · tip-bot for Kan Liang

kernel/events/core.c static int perf_mmap(struct file *file, struct

Quoted 4 lines

kernel/events/core.c +0 -0

@@ -4446,7 +4446,7 @@ static int perf_mmap(struct file *file, struct

vm_area_struct *vma)

Quoted 1 line

kernel/events/core.c +0 -0

 	 * If we have rb pages ensure they're a power-of-two number, so

Quoted 8 lines

Quoted 8 lines

we

kernel/events/core.c static int perf_mmap(struct file *file, struct vm_area_struct *vma) · line 4450

Quoted 5 lines

kernel/events/core.c +1 -1

 	 * can do bitmasks instead of modulo.
 	 */
-	if (!is_power_of_2(nr_pages))
+	if (nr_pages != 0 && !is_power_of_2(nr_pages))
 		return -EINVAL;

Hm, what does is_power_of_2(0) return? It should return 0, because 0

is

Quoted 3 lines

Quoted 3 lines

not a power of 2!

and if it's fixed to return 0, then the check should properly be

something

Quoted 10 lines

Quoted 5 lines

like:

	if (!nr_pages || !is_power_of_2(nr_pages))

or so?

nr_pages == 0 is valid, we cannot return -EINVAL. Because

	vma_size = vma->vm_end - vma->vm_start;
	nr_pages = (vma_size / PAGE_SIZE) - 1;

Quoted 2 lines

So here we only want to check is_power_of_2
when the pages > 1 (or nr_pages > 0).

I don't think you answered my first question, on which my later comments rely:

Quoted 1 line

Quoted 1 line

Hm, what does is_power_of_2(0) return? It should return 0, because 0

is

Quoted 1 line

Quoted 1 line

not a power of 2!

Thanks,

Ingo

Quoted 34 lines · Liang, Kan

Quoted 32 lines

Quoted 18 lines · tip-bot for Kan Liang

kernel/events/core.c static int perf_mmap(struct file *file, struct

Quoted 4 lines

kernel/events/core.c +0 -0

@@ -4446,7 +4446,7 @@ static int perf_mmap(struct file *file, struct

vm_area_struct *vma)

kernel/events/core.c static int perf_mmap(struct file *file, struct · line 4446

Quoted 1 line

kernel/events/core.c +0 -0

 	 * If we have rb pages ensure they're a power-of-two number, so

we

kernel/events/core.c static int perf_mmap(struct file *file, struct vm_area_struct *vma) · line 4450

Quoted 5 lines

kernel/events/core.c +1 -1

 	 * can do bitmasks instead of modulo.
 	 */
-	if (!is_power_of_2(nr_pages))
+	if (nr_pages != 0 && !is_power_of_2(nr_pages))
 		return -EINVAL;

Hm, what does is_power_of_2(0) return? It should return 0, because 0 is not a power of 2!

Quoted 7 lines

and if it's fixed to return 0, then the check should properly be something like:

	if (!nr_pages || !is_power_of_2(nr_pages))

return -EINVAL;

or so?

nr_pages == 0 is valid, we cannot return -EINVAL. Because

	vma_size = vma->vm_end - vma->vm_start;
	nr_pages = (vma_size / PAGE_SIZE) - 1;

So here we only want to check is_power_of_2
when the pages > 1 (or nr_pages > 0).

Quoted 6 lines

I don't think you answered my first question, on which my later comments rely:

Quoted 1 line

Quoted 1 line

Hm, what does is_power_of_2(0) return? It should return 0, because 0

is

Quoted 1 line

Quoted 1 line

not a power of 2!

Yes, is_power_of_2(0) returns 0.