Examining Process Page Tables¶
pagemap is a new (as of 2.6.25) set of interfaces in the kernel that allowuserspace programs to examine the page tables and related information byreading files in/proc.
There are four components to pagemap:
/proc/pid/pagemap. This file lets a userspace process find out whichphysical frame each virtual page is mapped to. It contains one 64-bitvalue for each virtual page, containing the following data (fromfs/proc/task_mmu.c, above pagemap_read):
Bits 0-54 page frame number (PFN) if present
Bits 0-4 swap type if swapped
Bits 5-54 swap offset if swapped
Bit 55 pte is soft-dirty (seeSoft-Dirty PTEs)
Bit 56 page exclusively mapped (since 4.2)
Bit 57 pte is uffd-wp write-protected (since 5.13) (seeUserfaultfd)
Bit 58 pte is a guard region (since 6.15) (see madvise (2) man page)
Bits 59-60 zero
Bit 61 page is file-page or shared-anon (since 3.5)
Bit 62 page swapped
Bit 63 page present
Since Linux 4.0 only users with the CAP_SYS_ADMIN capability can get PFNs.In 4.0 and 4.1 opens by unprivileged fail with -EPERM. Starting from4.2 the PFN field is zeroed if the user does not have CAP_SYS_ADMIN.Reason: information about PFNs helps in exploiting Rowhammer vulnerability.
If the page is not present but in swap, then the PFN contains anencoding of the swap file number and the page’s offset into theswap. Unmapped pages return a null PFN. This allows determiningprecisely which pages are mapped (or in swap) and comparing mappedpages between processes.
Traditionally, bit 56 indicates that a page is mapped exactly once and bit56 is clear when a page is mapped multiple times, even when mapped in thesame process multiple times. In some kernel configurations, the semanticsfor pages part of a larger allocation (e.g., THP) can differ: bit 56 is setif all pages part of the corresponding large allocation arecertainlymapped in the same process, even if the page is mapped multiple times in thatprocess. Bit 56 is clear when any page page of the larger allocationismaybe mapped in a different process. In some cases, a large allocationmight be treated as “maybe mapped by multiple processes” even though thisis no longer the case.
Efficient users of this interface will use
/proc/pid/mapstodetermine which areas of memory are actually mapped and llseek toskip over unmapped regions.
/proc/kpagecount. This file contains a 64-bit count of the number oftimes each page is mapped, indexed by PFN. Some kernel configurations donot track the precise number of times a page part of a larger allocation(e.g., THP) is mapped. In these configurations, the average number ofmappings per page in this larger allocation is returned instead. However,if any page of the large allocation is mapped, the returned value willbe at least 1.
The page-types tool in the tools/mm directory can be used to query thenumber of times a page is mapped.
/proc/kpageflags. This file contains a 64-bit set of flags for eachpage, indexed by PFN.The flags are (from
fs/proc/page.c, above kpageflags_read):
LOCKED
ERROR
REFERENCED
UPTODATE
DIRTY
LRU
ACTIVE
SLAB
WRITEBACK
RECLAIM
BUDDY
MMAP
ANON
SWAPCACHE
SWAPBACKED
COMPOUND_HEAD
COMPOUND_TAIL
HUGE
UNEVICTABLE
HWPOISON
NOPAGE
KSM
THP
OFFLINE
ZERO_PAGE
IDLE
PGTABLE
/proc/kpagecgroup. This file contains a 64-bit inode number of thememory cgroup each page is charged to, indexed by PFN. Only available whenCONFIG_MEMCG is set.
Short descriptions to the page flags¶
- 0 - LOCKED
The page is being locked for exclusive access, e.g. by undergoing read/writeIO.
- 7 - SLAB
The page is managed by the SLAB/SLUB kernel memory allocator.When compound page is used, either will only set this flag on the headpage.
- 10 - BUDDY
A free memory block managed by the buddy system allocator.The buddy system organizes free memory in blocks of various orders.An order N block has 2^N physically contiguous pages, with the BUDDY flagset for all pages.Before 4.6 only the first page of the block had the flag set.
- 15 - COMPOUND_HEAD
A compound page with order N consists of 2^N physically contiguous pages.A compound page with order 2 takes the form of “HTTT”, where H donates itshead page and T donates its tail page(s). The major consumers of compoundpages are hugeTLB pages (HugeTLB Pages),the SLUB etc. memory allocators and various device drivers.However in this interface, only huge/giga pages are made visibleto end users.
- 16 - COMPOUND_TAIL
A compound page tail (see description above).
- 17 - HUGE
This is an integral part of a HugeTLB page.
- 19 - HWPOISON
Hardware detected memory corruption on this page: don’t touch the data!
- 20 - NOPAGE
No page frame exists at the requested address.
- 21 - KSM
Identical memory pages dynamically shared between one or more processes.
- 22 - THP
Contiguous pages which construct THP of any size and mapped by any granularity.
- 23 - OFFLINE
The page is logically offline.
- 24 - ZERO_PAGE
Zero page for pfn_zero or huge_zero page.
- 25 - IDLE
The page has not been accessed since it was marked idle (seeIdle Page Tracking).Note that this flag may be stale in case the page was accessed viaa PTE. To make sure the flag is up-to-date one has to read
/sys/kernel/mm/page_idle/bitmapfirst.- 26 - PGTABLE
The page is in use as a page table.
IO related page flags¶
- 1 - ERROR
IO error occurred.
- 3 - UPTODATE
The page has up-to-date data.ie. for file backed page: (in-memory data revision >= on-disk one)
- 4 - DIRTY
The page has been written to, hence contains new data.i.e. for file backed page: (in-memory data revision > on-disk one)
- 8 - WRITEBACK
The page is being synced to disk.
LRU related page flags¶
- 5 - LRU
The page is in one of the LRU lists.
- 6 - ACTIVE
The page is in the active LRU list.
- 18 - UNEVICTABLE
The page is in the unevictable (non-)LRU list It is somehow pinned andnot a candidate for LRU page reclaims, e.g. ramfs pages,shmctl(SHM_LOCK) and
mlock()memory segments.- 2 - REFERENCED
The page has been referenced since last LRU list enqueue/requeue.
- 9 - RECLAIM
The page will be reclaimed soon after its pageout IO completed.
- 11 - MMAP
A memory mapped page.
- 12 - ANON
A memory mapped page that is not part of a file.
- 13 - SWAPCACHE
The page is mapped to swap space, i.e. has an associated swap entry.
- 14 - SWAPBACKED
The page is backed by swap/RAM.
The page-types tool in the tools/mm directory can be used to query theabove flags.
Exceptions for Shared Memory¶
Page table entries for shared pages are cleared when the pages are zapped orswapped out. This makes swapped out pages indistinguishable from never-allocatedones.
In kernel space, the swap location can still be retrieved from the page cache.However, values stored only on the normal PTE get lost irretrievably when thepage is swapped out (i.e. SOFT_DIRTY).
In user space, whether the page is present, swapped or none can be deduced withthe help of lseek and/or mincore system calls.
lseek() can differentiate between accessed pages (present or swapped out) andholes (none/non-allocated) by specifying the SEEK_DATA flag on the file wherethe pages are backed. For anonymous shared pages, the file can be found in/proc/pid/map_files/.
mincore() can differentiate between pages in memory (present, including swapcache) and out of memory (swapped out or none/non-allocated).
Other notes¶
Reading from any of the files will return -EINVAL if you are not startingthe read on an 8-byte boundary (e.g., if you sought an odd number of bytesinto the file), or if the size of the read is not a multiple of 8 bytes.
Before Linux 3.11 pagemap bits 55-60 were used for “page-shift” (which isalways 12 at most architectures). Since Linux 3.11 their meaning changesafter first clear of soft-dirty bits. Since Linux 4.2 they are used forflags unconditionally.
Pagemap Scan IOCTL¶
ThePAGEMAP_SCAN IOCTL on the pagemap file can be used to get or optionallyclear the info about page table entries. The following operations are supportedin this IOCTL:
Scan the address range and get the memory ranges matching the provided criteria.This is performed when the output buffer is specified.
Write-protect the pages. The
PM_SCAN_WP_MATCHINGis used to write-protectthe pages of interest. ThePM_SCAN_CHECK_WPASYNCaborts the operation ifnon-Async Write Protected pages are found. ThePM_SCAN_WP_MATCHINGcan beused with or withoutPM_SCAN_CHECK_WPASYNC.Both of those operations can be combined into one atomic operation where we canget and write protect the pages as well.
Following flags about pages are currently supported:
PAGE_IS_WPALLOWED- Page has async-write-protection enabledPAGE_IS_WRITTEN- Page has been written to from the time it was write protectedPAGE_IS_FILE- Page is file backedPAGE_IS_PRESENT- Page is present in the memoryPAGE_IS_SWAPPED- Page is in swappedPAGE_IS_PFNZERO- Page has zero PFNPAGE_IS_HUGE- Page is PMD-mapped THP or Hugetlb backedPAGE_IS_SOFT_DIRTY- Page is soft-dirtyPAGE_IS_GUARD- Page is a part of a guard region
Thestructpm_scan_arg is used as the argument of the IOCTL.
The size of the
structpm_scan_argmust be specified in thesizefield. This field will be helpful in recognizing the structure if extensionsare done later.The flags can be specified in the
flagsfield. ThePM_SCAN_WP_MATCHINGandPM_SCAN_CHECK_WPASYNCare the only added flags at this time. The getoperation is optionally performed depending upon if the output buffer isprovided or not.The range is specified through
startandend.The walk can abort before visiting the complete range such as the user buffercan get full etc. The walk ending address is specified in``end_walk``.
The output buffer of
structpage_regionarray and size is specified invecandvec_len.The optional maximum requested pages are specified in the
max_pages.The masks are specified in
category_mask,category_anyof_mask,category_invertedandreturn_mask.
Find pages which have been written and WP them as well:
struct pm_scan_arg arg = {.size = sizeof(arg),.flags = PM_SCAN_CHECK_WPASYNC | PM_SCAN_CHECK_WPASYNC,...category_mask = PAGE_IS_WRITTEN,.return_mask = PAGE_IS_WRITTEN,};Find pages which have been written, are file backed, not swapped and eitherpresent or huge:
struct pm_scan_arg arg = {.size = sizeof(arg),.flags = 0,...category_mask = PAGE_IS_WRITTEN | PAGE_IS_SWAPPED,.category_inverted = PAGE_IS_SWAPPED,.category_anyof_mask = PAGE_IS_PRESENT | PAGE_IS_HUGE,.return_mask = PAGE_IS_WRITTEN | PAGE_IS_SWAPPED | PAGE_IS_PRESENT | PAGE_IS_HUGE,};ThePAGE_IS_WRITTEN flag can be considered as a better-performing alternativeof soft-dirty flag. It doesn’t get affected by VMA merging of the kernel and hencethe user can find the true soft-dirty pages in case of normal pages. (There maystill be extra dirty pages reported for THP or Hugetlb pages.)
“PAGE_IS_WRITTEN” category is used with uffd write protect-enabled ranges toimplement memory dirty tracking in userspace:
The userfaultfd file descriptor is created with
userfaultfdsyscall.The
UFFD_FEATURE_WP_UNPOPULATEDandUFFD_FEATURE_WP_ASYNCfeaturesare set byUFFDIO_APIIOCTL.The memory range is registered with
UFFDIO_REGISTER_MODE_WPmodethroughUFFDIO_REGISTERIOCTL.Then any part of the registered memory or the whole memory region mustbe write protected using
PAGEMAP_SCANIOCTL with flagPM_SCAN_WP_MATCHINGor theUFFDIO_WRITEPROTECTIOCTL can be used. Both of these perform thesame operation. The former is better in terms of performance.Now the
PAGEMAP_SCANIOCTL can be used to either just find pages whichhave been written to since they were last marked and/or optionally write protectthe pages as well.