Commit e6a9a2cb authored by Andrei Vagin's avatar Andrei Vagin Committed by Andrew Morton
Browse files

fs/proc/task_mmu: report SOFT_DIRTY bits through the PAGEMAP_SCAN ioctl 

The PAGEMAP_SCAN ioctl returns information regarding page table entries. 
It is more efficient compared to reading pagemap files.  CRIU can start to
utilize this ioctl, but it needs info about soft-dirty bits to track
memory changes.

We are aware of a new method for tracking memory changes implemented in
the PAGEMAP_SCAN ioctl.  For CRIU, the primary advantage of this method is
its usability by unprivileged users.  However, it is not feasible to
transparently replace the soft-dirty tracker with the new one.  The main
problem here is userfault descriptors that have to be preserved between
pre-dump iterations.  It means criu continues supporting the soft-dirty
method to avoid breakage for current users.  The new method will be
implemented as a separate feature.

[avagin@google.com: update tools/include/uapi/linux/fs.h]
  Link: https://lkml.kernel.org/r/20231107164139.576046-1-avagin@google.com
Link: https://lkml.kernel.org/r/20231106220959.296568-1-avagin@google.com


Signed-off-by: default avatarAndrei Vagin <avagin@google.com>
Reviewed-by: default avatarMuhammad Usama Anjum <usama.anjum@collabora.com>
Cc: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
parent 8ff25266

Documentation/admin-guide/mm/pagemap.rst

+1 −0
Original line number Diff line number Diff line
@@ -253,6 +253,7 @@ Following flags about pages are currently supported: 
- ``PAGE_IS_SWAPPED`` - Page is in swapped

- ``PAGE_IS_PFNZERO`` - Page has zero PFN

- ``PAGE_IS_HUGE`` - Page is THP or Hugetlb backed

- ``PAGE_IS_SOFT_DIRTY`` - Page is soft-dirty



The ``struct pm_scan_arg`` is used as the argument of the IOCTL.

fs/proc/task_mmu.c

+16 −1
Original line number Diff line number Diff line
@@ -1761,7 +1761,7 @@ static int pagemap_release(struct inode *inode, struct file *file) 
#define PM_SCAN_CATEGORIES	(PAGE_IS_WPALLOWED | PAGE_IS_WRITTEN |	\

				 PAGE_IS_FILE |	PAGE_IS_PRESENT |	\

				 PAGE_IS_SWAPPED | PAGE_IS_PFNZERO |	\

				 PAGE_IS_HUGE)

				 PAGE_IS_HUGE | PAGE_IS_SOFT_DIRTY)

#define PM_SCAN_FLAGS		(PM_SCAN_WP_MATCHING | PM_SCAN_CHECK_WPASYNC)



struct pagemap_scan_private {
@@ -1793,6 +1793,8 @@ static unsigned long pagemap_page_category(struct pagemap_scan_private *p, 


		if (is_zero_pfn(pte_pfn(pte)))

			categories |= PAGE_IS_PFNZERO;

		if (pte_soft_dirty(pte))

			categories |= PAGE_IS_SOFT_DIRTY;

	} else if (is_swap_pte(pte)) {

		swp_entry_t swp;
@@ -1806,6 +1808,8 @@ static unsigned long pagemap_page_category(struct pagemap_scan_private *p, 
			    !PageAnon(pfn_swap_entry_to_page(swp)))

				categories |= PAGE_IS_FILE;

		}

		if (pte_swp_soft_dirty(pte))

			categories |= PAGE_IS_SOFT_DIRTY;

	}



	return categories;
@@ -1853,12 +1857,16 @@ static unsigned long pagemap_thp_category(struct pagemap_scan_private *p, 


		if (is_zero_pfn(pmd_pfn(pmd)))

			categories |= PAGE_IS_PFNZERO;

		if (pmd_soft_dirty(pmd))

			categories |= PAGE_IS_SOFT_DIRTY;

	} else if (is_swap_pmd(pmd)) {

		swp_entry_t swp;



		categories |= PAGE_IS_SWAPPED;

		if (!pmd_swp_uffd_wp(pmd))

			categories |= PAGE_IS_WRITTEN;

		if (pmd_swp_soft_dirty(pmd))

			categories |= PAGE_IS_SOFT_DIRTY;



		if (p->masks_of_interest & PAGE_IS_FILE) {

			swp = pmd_to_swp_entry(pmd);
@@ -1905,10 +1913,14 @@ static unsigned long pagemap_hugetlb_category(pte_t pte) 
			categories |= PAGE_IS_FILE;

		if (is_zero_pfn(pte_pfn(pte)))

			categories |= PAGE_IS_PFNZERO;

		if (pte_soft_dirty(pte))

			categories |= PAGE_IS_SOFT_DIRTY;

	} else if (is_swap_pte(pte)) {

		categories |= PAGE_IS_SWAPPED;

		if (!pte_swp_uffd_wp_any(pte))

			categories |= PAGE_IS_WRITTEN;

		if (pte_swp_soft_dirty(pte))

			categories |= PAGE_IS_SOFT_DIRTY;

	}



	return categories;
@@ -2007,6 +2019,9 @@ static int pagemap_scan_test_walk(unsigned long start, unsigned long end, 
	if (wp_allowed)

		vma_category |= PAGE_IS_WPALLOWED;



	if (vma->vm_flags & VM_SOFTDIRTY)

		vma_category |= PAGE_IS_SOFT_DIRTY;



	if (!pagemap_scan_is_interesting_vma(vma_category, p))

		return 1;

include/uapi/linux/fs.h

+1 −0
Original line number Diff line number Diff line
@@ -316,6 +316,7 @@ typedef int __bitwise __kernel_rwf_t; 
#define PAGE_IS_SWAPPED		(1 << 4)

#define PAGE_IS_PFNZERO		(1 << 5)

#define PAGE_IS_HUGE		(1 << 6)

#define PAGE_IS_SOFT_DIRTY	(1 << 7)



/*

 * struct page_region - Page region with flags

tools/include/uapi/linux/fs.h

+1 −0
Original line number Diff line number Diff line
@@ -316,6 +316,7 @@ typedef int __bitwise __kernel_rwf_t; 
#define PAGE_IS_SWAPPED		(1 << 4)

#define PAGE_IS_PFNZERO		(1 << 5)

#define PAGE_IS_HUGE		(1 << 6)

#define PAGE_IS_SOFT_DIRTY	(1 << 7)



/*

 * struct page_region - Page region with flags