Skip to contentSkip to conversation

/dev/zero: try to align PMD_SIZE for private mapping

PATCH zhangqilong 1 participant 1 message Jul 30, 2025 Memory management

lore
Share & tools
lore
#0 lore raw
Quoted 5 lines · Zhang Qilong
Quoted 1 line

By default, THP are usually enabled. Mapping /dev/zero with a size

Err... we can't rely on this.

OK, I will update this description in next version.

Quoted 75 lines

As per below comments on code, I'd update this to say something about fallback if it's not.

Quoted 7 lines

larger than 2MB could achieve performance gains by allocating aligned address. The mprot_tw4m in libMicro average execution time on arm64:
- Test case: mprot_tw4m
- Before the patch: 22 us
- After the patch: 17 us

This looks ok to me because there's a precedent for using thp_get_unmapped_area() directly as a file_operations-

Quoted 1 line

get_unmapped_area e.g. in ext4.

We also simply (amusingly, or perhaps not hugely amusingly, rather 'uniquely') establish an anonymous mapping on f_op->mmap via mmap_zero() using vma_set_anonymous(), so we can rely on the standard anon page memory faulting logic to sort out the actual allocation/mapping of the huge page via:

__handle_mm_fault() -> create_huge_pmd() ->
do_huge_pmd_anonymous_page() etc.

So everything should 'just work', and fallback if not permitted.

So in general seems fine.

Quoted 10 lines
---
 drivers/char/mem.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/char/mem.c index renamed from drivers/char/mem.c +0 -0

48839958b0b1..c57327ca9dd6 100644

drivers/char/mem.c +0 -0
@@ -515,10 +515,12 @@ static int mmap_zero(struct file *file, struct

vm_area_struct *vma) static unsigned long

get_unmapped_area_zero(struct file *file,

drivers/char/mem.c

Quoted 2 lines
drivers/char/mem.c +0 -0
 				unsigned long addr, unsigned long len,
 				unsigned long pgoff, unsigned long flags)

{ #ifdef CONFIG_MMU

drivers/char/mem.c

Quoted 6 lines
drivers/char/mem.c +2 -0
+	unsigned long ret;
+
 	if (flags & MAP_SHARED) {
 		/*
 		 * mmap_zero() will call shmem_zero_setup() to create a file,
 		 * so use shmem's get_unmapped_area in case it can be

huge;

Quoted 1 line

* and pass NULL for file as in mmap.c's

get_unmapped_area(), @@

Quoted 1 line

-526,10 +528,13 @@ static unsigned long get_unmapped_area_zero(struct

file *file,

drivers/char/mem.c

Quoted 2 lines
drivers/char/mem.c +0 -0
 		 */
 		return shmem_get_unmapped_area(NULL, addr, len, pgoff,

flags);

drivers/char/mem.c

Quoted 3 lines
drivers/char/mem.c +0 -0
 	}
 
 	/* Otherwise flags & MAP_PRIVATE: with no shmem object beneath

it */

Let's add a comment here like:

	/*
	 * Attempt to map aligned to huge page size if possible, otherwise

we
* fall back to system page size mappings. If THP is not enabled, this
* returns NULL and we always fallback.
*/

I think it'd be sensible to have an #ifdef CONFIG_TRANSPARENT_HUGEPAGE here, because thp_get_unmapped_area() does the fallback for you, and then otherwise we'd be trying it twice which is weird.

Quoted 7 lines

E.g.:

#ifdef CONFIG_TRANSPARENT_HUGEPAGE
	return thp_get_unmapped_area(file, addr, len, pgoff, flags); #else
	return mm_get_unmapped_area(current->mm, file, addr, len, pgoff,

flags); #endif

Trying it twice is realy unnecessary. This looks clearer and better, I will refer to your suggestion in patch V2. Thanks a lot for your and helpful advice.

Quoted 8 lines
Quoted 8 lines
drivers/char/mem.c +3 -0
+	ret = thp_get_unmapped_area(file, addr, len, pgoff, flags);
+	if (ret)
+		return ret;
 	return mm_get_unmapped_area(current->mm, file, addr, len, pgoff,

flags); #else
return -ENOSYS;
#endif
}

Quoted 8 lines
Quoted 3 lines

--
2.43.0

In _theory_ we should do the thing in mmap() where we check the size is PMD-aligned (see __get_unmapped_area()), but I don't think anybody's mapping a bunch of /dev/zero mappings next to each other or using them in any way where that'd matter... So yeah let's not :)

I agree with your thought. Do not make check here.

Thanks.
Zhang