Skip to contentSkip to conversation

mm/rmap: fix incorrect pte restoration for lazyfree folios

PATCH Applied Dev Jain 6 participants 10 messages Mar 3 – Mar 10

Subsystems (2)
Memory management MM rmap
lore
Share & tools
lore

Dev Jain

author
#0 lore raw

PATCH v3 1/1 mm/rmap: fix incorrect pte restoration for lazyfree folios

We batch unmap anonymous lazyfree folios by folio_unmap_pte_batch. If the batch has a mix of writable and non-writable bits, we may end up setting the entire batch writable. Fix this by respecting writable bit during batching.
Although on a successful unmap of a lazyfree folio, the soft-dirty bit is lost, preserve it on pte restoration by respecting the bit during batching, to make the fix consistent w.r.t both writable bit and soft-dirty bit.

I was able to write the below reproducer and crash the kernel. Explanation of reproducer (set 64K mTHP to always):

Fault in a 64K large folio. Split the VMA at mid-point with MADV_DONTFORK. fork() - parent points to the folio with 8 writable ptes and 8 non-writable ptes. Merge the VMAs with MADV_DOFORK so that folio_unmap_pte_batch() can determine all the 16 ptes as a batch. Do MADV_FREE on the range to mark the folio as lazyfree. Write to the memory to dirty the pte, eventually rmap will dirty the folio. Then trigger reclaim, we will hit the pte restoration path, and the kernel will crash with the following trace:

[   21.134473] kernel BUG at mm/page_table_check.c:118!
[   21.134497] Internal error: Oops - BUG: 00000000f2000800 [#1]  SMP
[   21.135917] Modules linked in:
[   21.136085] CPU: 1 UID: 0 PID: 1735 Comm: dup-lazyfree Not tainted 7.0.0-rc1-00116-g018018a17770 #1028 PREEMPT
[   21.136858] Hardware name: linux,dummy-virt (DT)
[   21.137019] pstate: 21400005 (nzCv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--)
[   21.137308] pc : page_table_check_set+0x28c/0x2a8
[   21.137607] lr : page_table_check_set+0x134/0x2a8
[   21.137885] sp : ffff80008a3b3340
[   21.138124] x29: ffff80008a3b3340 x28: fffffdffc3d14400 x27: ffffd1a55e03d000
[   21.138623] x26: 0040000000000040 x25: ffffd1a55f7dd000 x24: 0000000000000001
[   21.139045] x23: 0000000000000001 x22: 0000000000000001 x21: ffffd1a55f217f30
[   21.139629] x20: 0000000000134521 x19: 0000000000134519 x18: 005c43e000040000
[   21.140027] x17: 0001400000000000 x16: 0001700000000000 x15: 000000000000ffff
[   21.140578] x14: 000000000000000c x13: 005c006000000000 x12: 0000000000000020
[   21.140828] x11: 0000000000000000 x10: 005c000000000000 x9 : ffffd1a55c079ee0
[   21.141077] x8 : 0000000000000001 x7 : 005c03e000040000 x6 : 000000004000ffff
[   21.141490] x5 : ffff00017fffce00 x4 : 0000000000000001 x3 : 0000000000000002
[   21.141741] x2 : 0000000000134510 x1 : 0000000000000000 x0 : ffff0000c08228c0
[   21.141991] Call trace:
[   21.142093]  page_table_check_set+0x28c/0x2a8 (P)
[   21.142265]  __page_table_check_ptes_set+0x144/0x1e8
[   21.142441]  __set_ptes_anysz.constprop.0+0x160/0x1a8
[   21.142766]  contpte_set_ptes+0xe8/0x140
[   21.142907]  try_to_unmap_one+0x10c4/0x10d0
[   21.143177]  rmap_walk_anon+0x100/0x250
[   21.143315]  try_to_unmap+0xa0/0xc8
[   21.143441]  shrink_folio_list+0x59c/0x18a8
[   21.143759]  shrink_lruvec+0x664/0xbf0
[   21.144043]  shrink_node+0x218/0x878
[   21.144285]  __node_reclaim.constprop.0+0x98/0x338
[   21.144763]  user_proactive_reclaim+0x2a4/0x340
[   21.145056]  reclaim_store+0x3c/0x60
[   21.145216]  dev_attr_store+0x20/0x40
[   21.145585]  sysfs_kf_write+0x84/0xa8
[   21.145835]  kernfs_fop_write_iter+0x130/0x1c8
[   21.145994]  vfs_write+0x2b8/0x368
[   21.146119]  ksys_write+0x70/0x110
[   21.146240]  __arm64_sys_write+0x24/0x38
[   21.146380]  invoke_syscall+0x50/0x120
[   21.146513]  el0_svc_common.constprop.0+0x48/0xf8
[   21.146679]  do_el0_svc+0x28/0x40
[   21.146798]  el0_svc+0x34/0x110
[   21.146926]  el0t_64_sync_handler+0xa0/0xe8
[   21.147074]  el0t_64_sync+0x198/0x1a0
[   21.147225] Code: f9400441 b4fff241 17ffff94 d4210000 (d4210000)
[   21.147440] ---[ end trace 0000000000000000 ]---

#define _GNU_SOURCE
#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <sys/mman.h>
#include <string.h>
#include <sys/wait.h>
#include <sched.h>
#include <fcntl.h>

void write_to_reclaim() {
    const char *path = "/sys/devices/system/node/node0/reclaim";
    const char *value = "409600000000";
    int fd = open(path, O_WRONLY);
    if (fd == -1) {
        perror("open");
        exit(EXIT_FAILURE);
    }

    if (write(fd, value, sizeof("409600000000") - 1) == -1) {
        perror("write");
        close(fd);
        exit(EXIT_FAILURE);
    }

    printf("Successfully wrote %s to %s\n", value, path);
    close(fd);
}

int main()

{
	char *ptr = mmap((void *)(1UL << 30), 1UL << 16, PROT_READ | PROT_WRITE,
			 MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
	if ((unsigned long)ptr != (1UL << 30)) {
		perror("mmap");
		return 1;
	}
	
	/* a 64K folio gets faulted in */
	memset(ptr, 0, 1UL << 16);

	/* 32K half will not be shared into child */
	if (madvise(ptr, 1UL << 15, MADV_DONTFORK)) {
		perror("madvise madv dontfork");
		return 1;
	}

	pid_t pid = fork();

	if (pid < 0) {
		perror("fork");
		return 1;
	} else if (pid == 0) {
		sleep(15);
	} else {
		/* merge VMAs. now first half of the 16 ptes are writable, the other half not. */
		if (madvise(ptr, 1UL << 15, MADV_DOFORK)) {
			perror("madvise madv fork");
			return 1;
		}
		if (madvise(ptr, (1UL << 16), MADV_FREE)) {
			perror("madvise madv free");
			return 1;
		}

		/* dirty the large folio */
		(*ptr) += 10;

		write_to_reclaim();
		// sleep(10);
		waitpid(pid, NULL, 0);

	}
}
Fixes 354dffd29575 ("mm: support batched unmap for lazyfree large folios during reclamation")
Cc stable <[email protected]>
Signed-off-by Dev Jain <[email protected]>

---
Patch applies on mm-unstable (9af4957ef127).

v2->v3:
- Don't special case for anon folios

v1->v2:
- Just respect the writable bit instead of hacking in a pte_wrprotect() in failure path
- Also handle soft-dirty bit

 mm/rmap.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
mm/rmap.c +8 -1
index bff8f222004e4..5a3e408e3f179 100644
@@ -1955,7 +1955,14 @@ static inline unsigned int folio_unmap_pte_batch(struct folio *folio,
 	if (userfaultfd_wp(vma))
 		return 1;
 
-	return folio_pte_batch(folio, pvmw->pte, pte, max_nr);
+	/*
+	 * If unmap fails, we need to restore the ptes. To avoid accidentally
+	 * upgrading write permissions for ptes that were not originally
+	 * writable, and to avoid losing the soft-dirty bit, use the
+	 * appropriate FPB flags.
+	 */
+	return folio_pte_batch_flags(folio, vma, pvmw->pte, &pte, max_nr,
+				     FPB_RESPECT_WRITE | FPB_RESPECT_SOFT_DIRTY);
 }
 
 /*
Signature · 1 line
2.34.1
maintainer Acked-by on v3 1/1
#1 lore raw
Quoted 143 lines · Dev Jain

We batch unmap anonymous lazyfree folios by folio_unmap_pte_batch. If the batch has a mix of writable and non-writable bits, we may end up setting the entire batch writable. Fix this by respecting writable bit during batching.
Although on a successful unmap of a lazyfree folio, the soft-dirty bit is lost, preserve it on pte restoration by respecting the bit during batching, to make the fix consistent w.r.t both writable bit and soft-dirty bit.

I was able to write the below reproducer and crash the kernel. Explanation of reproducer (set 64K mTHP to always):

Fault in a 64K large folio. Split the VMA at mid-point with MADV_DONTFORK. fork() - parent points to the folio with 8 writable ptes and 8 non-writable ptes. Merge the VMAs with MADV_DOFORK so that folio_unmap_pte_batch() can determine all the 16 ptes as a batch. Do MADV_FREE on the range to mark the folio as lazyfree. Write to the memory to dirty the pte, eventually rmap will dirty the folio. Then trigger reclaim, we will hit the pte restoration path, and the kernel will crash with the following trace:

[   21.134473] kernel BUG at mm/page_table_check.c:118!
[   21.134497] Internal error: Oops - BUG: 00000000f2000800 [#1]  SMP
[   21.135917] Modules linked in:
[   21.136085] CPU: 1 UID: 0 PID: 1735 Comm: dup-lazyfree Not tainted 7.0.0-rc1-00116-g018018a17770 #1028 PREEMPT
[   21.136858] Hardware name: linux,dummy-virt (DT)
[   21.137019] pstate: 21400005 (nzCv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--)
[   21.137308] pc : page_table_check_set+0x28c/0x2a8
[   21.137607] lr : page_table_check_set+0x134/0x2a8
[   21.137885] sp : ffff80008a3b3340
[   21.138124] x29: ffff80008a3b3340 x28: fffffdffc3d14400 x27: ffffd1a55e03d000
[   21.138623] x26: 0040000000000040 x25: ffffd1a55f7dd000 x24: 0000000000000001
[   21.139045] x23: 0000000000000001 x22: 0000000000000001 x21: ffffd1a55f217f30
[   21.139629] x20: 0000000000134521 x19: 0000000000134519 x18: 005c43e000040000
[   21.140027] x17: 0001400000000000 x16: 0001700000000000 x15: 000000000000ffff
[   21.140578] x14: 000000000000000c x13: 005c006000000000 x12: 0000000000000020
[   21.140828] x11: 0000000000000000 x10: 005c000000000000 x9 : ffffd1a55c079ee0
[   21.141077] x8 : 0000000000000001 x7 : 005c03e000040000 x6 : 000000004000ffff
[   21.141490] x5 : ffff00017fffce00 x4 : 0000000000000001 x3 : 0000000000000002
[   21.141741] x2 : 0000000000134510 x1 : 0000000000000000 x0 : ffff0000c08228c0
[   21.141991] Call trace:
[   21.142093]  page_table_check_set+0x28c/0x2a8 (P)
[   21.142265]  __page_table_check_ptes_set+0x144/0x1e8
[   21.142441]  __set_ptes_anysz.constprop.0+0x160/0x1a8
[   21.142766]  contpte_set_ptes+0xe8/0x140
[   21.142907]  try_to_unmap_one+0x10c4/0x10d0
[   21.143177]  rmap_walk_anon+0x100/0x250
[   21.143315]  try_to_unmap+0xa0/0xc8
[   21.143441]  shrink_folio_list+0x59c/0x18a8
[   21.143759]  shrink_lruvec+0x664/0xbf0
[   21.144043]  shrink_node+0x218/0x878
[   21.144285]  __node_reclaim.constprop.0+0x98/0x338
[   21.144763]  user_proactive_reclaim+0x2a4/0x340
[   21.145056]  reclaim_store+0x3c/0x60
[   21.145216]  dev_attr_store+0x20/0x40
[   21.145585]  sysfs_kf_write+0x84/0xa8
[   21.145835]  kernfs_fop_write_iter+0x130/0x1c8
[   21.145994]  vfs_write+0x2b8/0x368
[   21.146119]  ksys_write+0x70/0x110
[   21.146240]  __arm64_sys_write+0x24/0x38
[   21.146380]  invoke_syscall+0x50/0x120
[   21.146513]  el0_svc_common.constprop.0+0x48/0xf8
[   21.146679]  do_el0_svc+0x28/0x40
[   21.146798]  el0_svc+0x34/0x110
[   21.146926]  el0t_64_sync_handler+0xa0/0xe8
[   21.147074]  el0t_64_sync+0x198/0x1a0
[   21.147225] Code: f9400441 b4fff241 17ffff94 d4210000 (d4210000)
[   21.147440] ---[ end trace 0000000000000000 ]---

#define _GNU_SOURCE
#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <sys/mman.h>
#include <string.h>
#include <sys/wait.h>
#include <sched.h>
#include <fcntl.h>

void write_to_reclaim() {
    const char *path = "/sys/devices/system/node/node0/reclaim";
    const char *value = "409600000000";
    int fd = open(path, O_WRONLY);
    if (fd == -1) {
        perror("open");
        exit(EXIT_FAILURE);
    }

    if (write(fd, value, sizeof("409600000000") - 1) == -1) {
        perror("write");
        close(fd);
        exit(EXIT_FAILURE);
    }

    printf("Successfully wrote %s to %s\n", value, path);
    close(fd);
}

int main()

{
	char *ptr = mmap((void *)(1UL << 30), 1UL << 16, PROT_READ | PROT_WRITE,
			 MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
	if ((unsigned long)ptr != (1UL << 30)) {
		perror("mmap");
		return 1;
	}
	
	/* a 64K folio gets faulted in */
	memset(ptr, 0, 1UL << 16);

	/* 32K half will not be shared into child */
	if (madvise(ptr, 1UL << 15, MADV_DONTFORK)) {
		perror("madvise madv dontfork");
		return 1;
	}

	pid_t pid = fork();

	if (pid < 0) {
		perror("fork");
		return 1;
	} else if (pid == 0) {
		sleep(15);
	} else {
		/* merge VMAs. now first half of the 16 ptes are writable, the other half not. */
		if (madvise(ptr, 1UL << 15, MADV_DOFORK)) {
			perror("madvise madv fork");
			return 1;
		}
		if (madvise(ptr, (1UL << 16), MADV_FREE)) {
			perror("madvise madv free");
			return 1;
		}

		/* dirty the large folio */
		(*ptr) += 10;

		write_to_reclaim();
		// sleep(10);
		waitpid(pid, NULL, 0);

	}
}
Quoted 3 lines
Fixes 354dffd29575 ("mm: support batched unmap for lazyfree large folios during reclamation")
Cc stable <[email protected]>
Signed-off-by Dev Jain <[email protected]>

Thanks Dev!

Acked-by David Hildenbrand (Arm) <[email protected]>
Signature · 3 lines
Cheers,

David
Reviewed-by on v3 1/1
#2 lore raw
In reply to Dev Jain
Quoted 19 lines · Dev Jain

We batch unmap anonymous lazyfree folios by folio_unmap_pte_batch. If the batch has a mix of writable and non-writable bits, we may end up setting the entire batch writable. Fix this by respecting writable bit during batching.
Although on a successful unmap of a lazyfree folio, the soft-dirty bit is lost, preserve it on pte restoration by respecting the bit during batching, to make the fix consistent w.r.t both writable bit and soft-dirty bit.

I was able to write the below reproducer and crash the kernel. Explanation of reproducer (set 64K mTHP to always):

Fault in a 64K large folio. Split the VMA at mid-point with MADV_DONTFORK. fork() - parent points to the folio with 8 writable ptes and 8 non-writable ptes. Merge the VMAs with MADV_DOFORK so that folio_unmap_pte_batch() can determine all the 16 ptes as a batch. Do MADV_FREE on the range to mark the folio as lazyfree. Write to the memory to dirty the pte, eventually rmap will dirty the folio. Then trigger reclaim, we will hit the pte restoration path, and the kernel will crash with the following trace:

Quoted 1 line
[   21.134473] kernel BUG at mm/page_table_check.c:118!

Presumably:

			BUG_ON(atomic_inc_return(&ptc->anon_map_count) > 1 && rw);

It'd be useful to be explicit about this in the commit msg.

It's also probably worth saying explicitly that this is about a non-AnonExclusive mapping being mapping read/write as a result of this.

I'm not sure the stack trace is really that useful beyond that, as by the time somebody comes to read this later it's probably going to be fairly inaccurate :)

Maybe worth just spelling out e.g. try_to_unmap_one() -> ... rest of code path ...

I mean the stack already misses stuff due to inlining so that'd be more useful, otherwise I'd drop it.

Quoted 114 lines
[   21.134497] Internal error: Oops - BUG: 00000000f2000800 [#1]  SMP
[   21.135917] Modules linked in:
[   21.136085] CPU: 1 UID: 0 PID: 1735 Comm: dup-lazyfree Not tainted 7.0.0-rc1-00116-g018018a17770 #1028 PREEMPT
[   21.136858] Hardware name: linux,dummy-virt (DT)
[   21.137019] pstate: 21400005 (nzCv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--)
[   21.137308] pc : page_table_check_set+0x28c/0x2a8
[   21.137607] lr : page_table_check_set+0x134/0x2a8
[   21.137885] sp : ffff80008a3b3340
[   21.138124] x29: ffff80008a3b3340 x28: fffffdffc3d14400 x27: ffffd1a55e03d000
[   21.138623] x26: 0040000000000040 x25: ffffd1a55f7dd000 x24: 0000000000000001
[   21.139045] x23: 0000000000000001 x22: 0000000000000001 x21: ffffd1a55f217f30
[   21.139629] x20: 0000000000134521 x19: 0000000000134519 x18: 005c43e000040000
[   21.140027] x17: 0001400000000000 x16: 0001700000000000 x15: 000000000000ffff
[   21.140578] x14: 000000000000000c x13: 005c006000000000 x12: 0000000000000020
[   21.140828] x11: 0000000000000000 x10: 005c000000000000 x9 : ffffd1a55c079ee0
[   21.141077] x8 : 0000000000000001 x7 : 005c03e000040000 x6 : 000000004000ffff
[   21.141490] x5 : ffff00017fffce00 x4 : 0000000000000001 x3 : 0000000000000002
[   21.141741] x2 : 0000000000134510 x1 : 0000000000000000 x0 : ffff0000c08228c0
[   21.141991] Call trace:
[   21.142093]  page_table_check_set+0x28c/0x2a8 (P)
[   21.142265]  __page_table_check_ptes_set+0x144/0x1e8
[   21.142441]  __set_ptes_anysz.constprop.0+0x160/0x1a8
[   21.142766]  contpte_set_ptes+0xe8/0x140
[   21.142907]  try_to_unmap_one+0x10c4/0x10d0
[   21.143177]  rmap_walk_anon+0x100/0x250
[   21.143315]  try_to_unmap+0xa0/0xc8
[   21.143441]  shrink_folio_list+0x59c/0x18a8
[   21.143759]  shrink_lruvec+0x664/0xbf0
[   21.144043]  shrink_node+0x218/0x878
[   21.144285]  __node_reclaim.constprop.0+0x98/0x338
[   21.144763]  user_proactive_reclaim+0x2a4/0x340
[   21.145056]  reclaim_store+0x3c/0x60
[   21.145216]  dev_attr_store+0x20/0x40
[   21.145585]  sysfs_kf_write+0x84/0xa8
[   21.145835]  kernfs_fop_write_iter+0x130/0x1c8
[   21.145994]  vfs_write+0x2b8/0x368
[   21.146119]  ksys_write+0x70/0x110
[   21.146240]  __arm64_sys_write+0x24/0x38
[   21.146380]  invoke_syscall+0x50/0x120
[   21.146513]  el0_svc_common.constprop.0+0x48/0xf8
[   21.146679]  do_el0_svc+0x28/0x40
[   21.146798]  el0_svc+0x34/0x110
[   21.146926]  el0t_64_sync_handler+0xa0/0xe8
[   21.147074]  el0t_64_sync+0x198/0x1a0
[   21.147225] Code: f9400441 b4fff241 17ffff94 d4210000 (d4210000)
[   21.147440] ---[ end trace 0000000000000000 ]---

#define _GNU_SOURCE
#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <sys/mman.h>
#include <string.h>
#include <sys/wait.h>
#include <sched.h>
#include <fcntl.h>

void write_to_reclaim() {
    const char *path = "/sys/devices/system/node/node0/reclaim";
    const char *value = "409600000000";
    int fd = open(path, O_WRONLY);
    if (fd == -1) {
        perror("open");
        exit(EXIT_FAILURE);
    }

    if (write(fd, value, sizeof("409600000000") - 1) == -1) {
        perror("write");
        close(fd);
        exit(EXIT_FAILURE);
    }

    printf("Successfully wrote %s to %s\n", value, path);
    close(fd);
}

int main()

{
	char *ptr = mmap((void *)(1UL << 30), 1UL << 16, PROT_READ | PROT_WRITE,
			 MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
	if ((unsigned long)ptr != (1UL << 30)) {
		perror("mmap");
		return 1;
	}

	/* a 64K folio gets faulted in */
	memset(ptr, 0, 1UL << 16);

	/* 32K half will not be shared into child */
	if (madvise(ptr, 1UL << 15, MADV_DONTFORK)) {
		perror("madvise madv dontfork");
		return 1;
	}

	pid_t pid = fork();

	if (pid < 0) {
		perror("fork");
		return 1;
	} else if (pid == 0) {
		sleep(15);
	} else {
		/* merge VMAs. now first half of the 16 ptes are writable, the other half not. */
		if (madvise(ptr, 1UL << 15, MADV_DOFORK)) {
			perror("madvise madv fork");
			return 1;
		}
		if (madvise(ptr, (1UL << 16), MADV_FREE)) {
			perror("madvise madv free");
			return 1;
		}

		/* dirty the large folio */
Quoted 8 lines
		(*ptr) += 10;

		write_to_reclaim();
		// sleep(10);
		waitpid(pid, NULL, 0);

	}
}

Thanks, having the repro code here is really useful!

Quoted 4 lines
Fixes 354dffd29575 ("mm: support batched unmap for lazyfree large folios during reclamation")
Cc stable <[email protected]>
Signed-off-by Dev Jain <[email protected]>

This looks correct to me, so other than nits around commit msg above:

Reviewed-by Lorenzo Stoakes <[email protected]>
Quoted 37 lines

---
Patch applies on mm-unstable (9af4957ef127).

v2->v3:
- Don't special case for anon folios

v1->v2:
- Just respect the writable bit instead of hacking in a pte_wrprotect() in failure path
- Also handle soft-dirty bit

 mm/rmap.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
mm/rmap.c +8 -2
index bff8f222004e4..5a3e408e3f179 100644
@@ -1955,7 +1955,14 @@ static inline unsigned int folio_unmap_pte_batch(struct folio *folio,
 	if (userfaultfd_wp(vma))
 		return 1;
 
-	return folio_pte_batch(folio, pvmw->pte, pte, max_nr);
+	/*
+	 * If unmap fails, we need to restore the ptes. To avoid accidentally
+	 * upgrading write permissions for ptes that were not originally
+	 * writable, and to avoid losing the soft-dirty bit, use the
+	 * appropriate FPB flags.
+	 */
+	return folio_pte_batch_flags(folio, vma, pvmw->pte, &pte, max_nr,
+				     FPB_RESPECT_WRITE | FPB_RESPECT_SOFT_DIRTY);
 }
 
 /*
--

2.34.1

author
#3 lore raw
Quoted 31 lines · Lorenzo Stoakes
Quoted 19 lines · Dev Jain

We batch unmap anonymous lazyfree folios by folio_unmap_pte_batch. If the batch has a mix of writable and non-writable bits, we may end up setting the entire batch writable. Fix this by respecting writable bit during batching.
Although on a successful unmap of a lazyfree folio, the soft-dirty bit is lost, preserve it on pte restoration by respecting the bit during batching, to make the fix consistent w.r.t both writable bit and soft-dirty bit.

I was able to write the below reproducer and crash the kernel. Explanation of reproducer (set 64K mTHP to always):

Fault in a 64K large folio. Split the VMA at mid-point with MADV_DONTFORK. fork() - parent points to the folio with 8 writable ptes and 8 non-writable ptes. Merge the VMAs with MADV_DOFORK so that folio_unmap_pte_batch() can determine all the 16 ptes as a batch. Do MADV_FREE on the range to mark the folio as lazyfree. Write to the memory to dirty the pte, eventually rmap will dirty the folio. Then trigger reclaim, we will hit the pte restoration path, and the kernel will crash with the following trace:

Quoted 1 line
[   21.134473] kernel BUG at mm/page_table_check.c:118!

Presumably:

			BUG_ON(atomic_inc_return(&ptc->anon_map_count) > 1 && rw);

It'd be useful to be explicit about this in the commit msg.

It's also probably worth saying explicitly that this is about a non-AnonExclusive mapping being mapping read/write as a result of this.

Quoted 7 lines

I'm not sure the stack trace is really that useful beyond that, as by the time somebody comes to read this later it's probably going to be fairly inaccurate :)

Maybe worth just spelling out e.g. try_to_unmap_one() -> ... rest of code path ...

I mean the stack already misses stuff due to inlining so that'd be more useful, otherwise I'd drop it.

Good point, thanks! Lemme reply to the patch.

Quoted 173 lines
Quoted 116 lines

[   21.134497] Internal error: Oops - BUG: 00000000f2000800 [#1]  SMP
[   21.135917] Modules linked in:
[   21.136085] CPU: 1 UID: 0 PID: 1735 Comm: dup-lazyfree Not tainted 7.0.0-rc1-00116-g018018a17770 #1028 PREEMPT
[   21.136858] Hardware name: linux,dummy-virt (DT)
[   21.137019] pstate: 21400005 (nzCv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--)
[   21.137308] pc : page_table_check_set+0x28c/0x2a8
[   21.137607] lr : page_table_check_set+0x134/0x2a8
[   21.137885] sp : ffff80008a3b3340
[   21.138124] x29: ffff80008a3b3340 x28: fffffdffc3d14400 x27: ffffd1a55e03d000
[   21.138623] x26: 0040000000000040 x25: ffffd1a55f7dd000 x24: 0000000000000001
[   21.139045] x23: 0000000000000001 x22: 0000000000000001 x21: ffffd1a55f217f30
[   21.139629] x20: 0000000000134521 x19: 0000000000134519 x18: 005c43e000040000
[   21.140027] x17: 0001400000000000 x16: 0001700000000000 x15: 000000000000ffff
[   21.140578] x14: 000000000000000c x13: 005c006000000000 x12: 0000000000000020
[   21.140828] x11: 0000000000000000 x10: 005c000000000000 x9 : ffffd1a55c079ee0
[   21.141077] x8 : 0000000000000001 x7 : 005c03e000040000 x6 : 000000004000ffff
[   21.141490] x5 : ffff00017fffce00 x4 : 0000000000000001 x3 : 0000000000000002
[   21.141741] x2 : 0000000000134510 x1 : 0000000000000000 x0 : ffff0000c08228c0
[   21.141991] Call trace:
[   21.142093]  page_table_check_set+0x28c/0x2a8 (P)
[   21.142265]  __page_table_check_ptes_set+0x144/0x1e8
[   21.142441]  __set_ptes_anysz.constprop.0+0x160/0x1a8
[   21.142766]  contpte_set_ptes+0xe8/0x140
[   21.142907]  try_to_unmap_one+0x10c4/0x10d0
[   21.143177]  rmap_walk_anon+0x100/0x250
[   21.143315]  try_to_unmap+0xa0/0xc8
[   21.143441]  shrink_folio_list+0x59c/0x18a8
[   21.143759]  shrink_lruvec+0x664/0xbf0
[   21.144043]  shrink_node+0x218/0x878
[   21.144285]  __node_reclaim.constprop.0+0x98/0x338
[   21.144763]  user_proactive_reclaim+0x2a4/0x340
[   21.145056]  reclaim_store+0x3c/0x60
[   21.145216]  dev_attr_store+0x20/0x40
[   21.145585]  sysfs_kf_write+0x84/0xa8
[   21.145835]  kernfs_fop_write_iter+0x130/0x1c8
[   21.145994]  vfs_write+0x2b8/0x368
[   21.146119]  ksys_write+0x70/0x110
[   21.146240]  __arm64_sys_write+0x24/0x38
[   21.146380]  invoke_syscall+0x50/0x120
[   21.146513]  el0_svc_common.constprop.0+0x48/0xf8
[   21.146679]  do_el0_svc+0x28/0x40
[   21.146798]  el0_svc+0x34/0x110
[   21.146926]  el0t_64_sync_handler+0xa0/0xe8
[   21.147074]  el0t_64_sync+0x198/0x1a0
[   21.147225] Code: f9400441 b4fff241 17ffff94 d4210000 (d4210000)
[   21.147440] ---[ end trace 0000000000000000 ]---

#define _GNU_SOURCE
#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <sys/mman.h>
#include <string.h>
#include <sys/wait.h>
#include <sched.h>
#include <fcntl.h>

void write_to_reclaim() {
    const char *path = "/sys/devices/system/node/node0/reclaim";
    const char *value = "409600000000";
    int fd = open(path, O_WRONLY);
    if (fd == -1) {
        perror("open");
        exit(EXIT_FAILURE);
    }

    if (write(fd, value, sizeof("409600000000") - 1) == -1) {
        perror("write");
        close(fd);
        exit(EXIT_FAILURE);
    }

    printf("Successfully wrote %s to %s\n", value, path);
    close(fd);
}

int main()

{
	char *ptr = mmap((void *)(1UL << 30), 1UL << 16, PROT_READ | PROT_WRITE,
			 MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
	if ((unsigned long)ptr != (1UL << 30)) {
		perror("mmap");
		return 1;
	}

	/* a 64K folio gets faulted in */
	memset(ptr, 0, 1UL << 16);

	/* 32K half will not be shared into child */
	if (madvise(ptr, 1UL << 15, MADV_DONTFORK)) {
		perror("madvise madv dontfork");
		return 1;
	}

	pid_t pid = fork();

	if (pid < 0) {
		perror("fork");
		return 1;
	} else if (pid == 0) {
		sleep(15);
	} else {
		/* merge VMAs. now first half of the 16 ptes are writable, the other half not. */
		if (madvise(ptr, 1UL << 15, MADV_DOFORK)) {
			perror("madvise madv fork");
			return 1;
		}
		if (madvise(ptr, (1UL << 16), MADV_FREE)) {
			perror("madvise madv free");
			return 1;
		}

		/* dirty the large folio */
Quoted 8 lines
		(*ptr) += 10;

		write_to_reclaim();
		// sleep(10);
		waitpid(pid, NULL, 0);

	}
}

Thanks, having the repro code here is really useful!

Quoted 4 lines
Fixes 354dffd29575 ("mm: support batched unmap for lazyfree large folios during reclamation")
Cc stable <[email protected]>
Signed-off-by Dev Jain <[email protected]>

This looks correct to me, so other than nits around commit msg above:

Reviewed-by Lorenzo Stoakes <[email protected]>
Quoted 37 lines

---
Patch applies on mm-unstable (9af4957ef127).

v2->v3:
- Don't special case for anon folios

v1->v2:
- Just respect the writable bit instead of hacking in a pte_wrprotect() in failure path
- Also handle soft-dirty bit

 mm/rmap.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
mm/rmap.c +8 -2
index bff8f222004e4..5a3e408e3f179 100644
@@ -1955,7 +1955,14 @@ static inline unsigned int folio_unmap_pte_batch(struct folio *folio,
 	if (userfaultfd_wp(vma))
 		return 1;
 
-	return folio_pte_batch(folio, pvmw->pte, pte, max_nr);
+	/*
+	 * If unmap fails, we need to restore the ptes. To avoid accidentally
+	 * upgrading write permissions for ptes that were not originally
+	 * writable, and to avoid losing the soft-dirty bit, use the
+	 * appropriate FPB flags.
+	 */
+	return folio_pte_batch_flags(folio, vma, pvmw->pte, &pte, max_nr,
+				     FPB_RESPECT_WRITE | FPB_RESPECT_SOFT_DIRTY);
 }
 
 /*
--

2.34.1

Reviewed-by on v3 1/1
#4 lore raw
In reply to Dev Jain
Quoted 144 lines · Dev Jain

We batch unmap anonymous lazyfree folios by folio_unmap_pte_batch. If the batch has a mix of writable and non-writable bits, we may end up setting the entire batch writable. Fix this by respecting writable bit during batching.
Although on a successful unmap of a lazyfree folio, the soft-dirty bit is lost, preserve it on pte restoration by respecting the bit during batching, to make the fix consistent w.r.t both writable bit and soft-dirty bit.

I was able to write the below reproducer and crash the kernel. Explanation of reproducer (set 64K mTHP to always):

Fault in a 64K large folio. Split the VMA at mid-point with MADV_DONTFORK. fork() - parent points to the folio with 8 writable ptes and 8 non-writable ptes. Merge the VMAs with MADV_DOFORK so that folio_unmap_pte_batch() can determine all the 16 ptes as a batch. Do MADV_FREE on the range to mark the folio as lazyfree. Write to the memory to dirty the pte, eventually rmap will dirty the folio. Then trigger reclaim, we will hit the pte restoration path, and the kernel will crash with the following trace:

[   21.134473] kernel BUG at mm/page_table_check.c:118!
[   21.134497] Internal error: Oops - BUG: 00000000f2000800 [#1]  SMP
[   21.135917] Modules linked in:
[   21.136085] CPU: 1 UID: 0 PID: 1735 Comm: dup-lazyfree Not tainted 7.0.0-rc1-00116-g018018a17770 #1028 PREEMPT
[   21.136858] Hardware name: linux,dummy-virt (DT)
[   21.137019] pstate: 21400005 (nzCv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--)
[   21.137308] pc : page_table_check_set+0x28c/0x2a8
[   21.137607] lr : page_table_check_set+0x134/0x2a8
[   21.137885] sp : ffff80008a3b3340
[   21.138124] x29: ffff80008a3b3340 x28: fffffdffc3d14400 x27: ffffd1a55e03d000
[   21.138623] x26: 0040000000000040 x25: ffffd1a55f7dd000 x24: 0000000000000001
[   21.139045] x23: 0000000000000001 x22: 0000000000000001 x21: ffffd1a55f217f30
[   21.139629] x20: 0000000000134521 x19: 0000000000134519 x18: 005c43e000040000
[   21.140027] x17: 0001400000000000 x16: 0001700000000000 x15: 000000000000ffff
[   21.140578] x14: 000000000000000c x13: 005c006000000000 x12: 0000000000000020
[   21.140828] x11: 0000000000000000 x10: 005c000000000000 x9 : ffffd1a55c079ee0
[   21.141077] x8 : 0000000000000001 x7 : 005c03e000040000 x6 : 000000004000ffff
[   21.141490] x5 : ffff00017fffce00 x4 : 0000000000000001 x3 : 0000000000000002
[   21.141741] x2 : 0000000000134510 x1 : 0000000000000000 x0 : ffff0000c08228c0
[   21.141991] Call trace:
[   21.142093]  page_table_check_set+0x28c/0x2a8 (P)
[   21.142265]  __page_table_check_ptes_set+0x144/0x1e8
[   21.142441]  __set_ptes_anysz.constprop.0+0x160/0x1a8
[   21.142766]  contpte_set_ptes+0xe8/0x140
[   21.142907]  try_to_unmap_one+0x10c4/0x10d0
[   21.143177]  rmap_walk_anon+0x100/0x250
[   21.143315]  try_to_unmap+0xa0/0xc8
[   21.143441]  shrink_folio_list+0x59c/0x18a8
[   21.143759]  shrink_lruvec+0x664/0xbf0
[   21.144043]  shrink_node+0x218/0x878
[   21.144285]  __node_reclaim.constprop.0+0x98/0x338
[   21.144763]  user_proactive_reclaim+0x2a4/0x340
[   21.145056]  reclaim_store+0x3c/0x60
[   21.145216]  dev_attr_store+0x20/0x40
[   21.145585]  sysfs_kf_write+0x84/0xa8
[   21.145835]  kernfs_fop_write_iter+0x130/0x1c8
[   21.145994]  vfs_write+0x2b8/0x368
[   21.146119]  ksys_write+0x70/0x110
[   21.146240]  __arm64_sys_write+0x24/0x38
[   21.146380]  invoke_syscall+0x50/0x120
[   21.146513]  el0_svc_common.constprop.0+0x48/0xf8
[   21.146679]  do_el0_svc+0x28/0x40
[   21.146798]  el0_svc+0x34/0x110
[   21.146926]  el0t_64_sync_handler+0xa0/0xe8
[   21.147074]  el0t_64_sync+0x198/0x1a0
[   21.147225] Code: f9400441 b4fff241 17ffff94 d4210000 (d4210000)
[   21.147440] ---[ end trace 0000000000000000 ]---

#define _GNU_SOURCE
#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <sys/mman.h>
#include <string.h>
#include <sys/wait.h>
#include <sched.h>
#include <fcntl.h>

void write_to_reclaim() {
    const char *path = "/sys/devices/system/node/node0/reclaim";
    const char *value = "409600000000";
    int fd = open(path, O_WRONLY);
    if (fd == -1) {
        perror("open");
        exit(EXIT_FAILURE);
    }

    if (write(fd, value, sizeof("409600000000") - 1) == -1) {
        perror("write");
        close(fd);
        exit(EXIT_FAILURE);
    }

    printf("Successfully wrote %s to %s\n", value, path);
    close(fd);
}

int main()

{
        char *ptr = mmap((void *)(1UL << 30), 1UL << 16, PROT_READ | PROT_WRITE,
                         MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
        if ((unsigned long)ptr != (1UL << 30)) {
                perror("mmap");
                return 1;
        }

        /* a 64K folio gets faulted in */
        memset(ptr, 0, 1UL << 16);

        /* 32K half will not be shared into child */
        if (madvise(ptr, 1UL << 15, MADV_DONTFORK)) {
                perror("madvise madv dontfork");
                return 1;
        }

        pid_t pid = fork();

        if (pid < 0) {
                perror("fork");
                return 1;
        } else if (pid == 0) {
                sleep(15);
        } else {
                /* merge VMAs. now first half of the 16 ptes are writable, the other half not. */
                if (madvise(ptr, 1UL << 15, MADV_DOFORK)) {
                        perror("madvise madv fork");
                        return 1;
                }
                if (madvise(ptr, (1UL << 16), MADV_FREE)) {
                        perror("madvise madv free");
                        return 1;
                }

                /* dirty the large folio */
                (*ptr) += 10;

                write_to_reclaim();
                // sleep(10);
                waitpid(pid, NULL, 0);

        }
}
Quoted 4 lines
Fixes 354dffd29575 ("mm: support batched unmap for lazyfree large folios during reclamation")
Cc stable <[email protected]>
Signed-off-by Dev Jain <[email protected]>

---

LGTM, thanks!

Reviewed-by Barry Song <[email protected]>
author
#5 lore raw
In reply to Dev Jain
Quoted 11 lines · Dev Jain

We batch unmap anonymous lazyfree folios by folio_unmap_pte_batch. If the batch has a mix of writable and non-writable bits, we may end up setting the entire batch writable. Fix this by respecting writable bit during batching.
Although on a successful unmap of a lazyfree folio, the soft-dirty bit is lost, preserve it on pte restoration by respecting the bit during batching, to make the fix consistent w.r.t both writable bit and soft-dirty bit.

I was able to write the below reproducer and crash the kernel. Explanation of reproducer (set 64K mTHP to always):

Quoted 7 lines

Fault in a 64K large folio. Split the VMA at mid-point with MADV_DONTFORK. fork() - parent points to the folio with 8 writable ptes and 8 non-writable ptes. Merge the VMAs with MADV_DOFORK so that folio_unmap_pte_batch() can determine all the 16 ptes as a batch. Do MADV_FREE on the range to mark the folio as lazyfree. Write to the memory to dirty the pte, eventually rmap will dirty the folio. Then trigger reclaim, we will hit the pte restoration path, and the kernel will crash with the following trace:

Andrew, could you convert the last line to the following:

..restoration path, and the kernel will crash with the trace given below. The BUG happens at:

BUG_ON(atomic_inc_return(&ptc->anon_map_count) > 1 && rw);

The code path is asking for anonymous page to be mapped writable into the pagetable. The BUG_ON() firing implies that such a writable page has been mapped into the pagetables of more than one process, which breaks anonymous memory/CoW semantics.

Quoted 162 lines
[   21.134473] kernel BUG at mm/page_table_check.c:118!
[   21.134497] Internal error: Oops - BUG: 00000000f2000800 [#1]  SMP
[   21.135917] Modules linked in:
[   21.136085] CPU: 1 UID: 0 PID: 1735 Comm: dup-lazyfree Not tainted 7.0.0-rc1-00116-g018018a17770 #1028 PREEMPT
[   21.136858] Hardware name: linux,dummy-virt (DT)
[   21.137019] pstate: 21400005 (nzCv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--)
[   21.137308] pc : page_table_check_set+0x28c/0x2a8
[   21.137607] lr : page_table_check_set+0x134/0x2a8
[   21.137885] sp : ffff80008a3b3340
[   21.138124] x29: ffff80008a3b3340 x28: fffffdffc3d14400 x27: ffffd1a55e03d000
[   21.138623] x26: 0040000000000040 x25: ffffd1a55f7dd000 x24: 0000000000000001
[   21.139045] x23: 0000000000000001 x22: 0000000000000001 x21: ffffd1a55f217f30
[   21.139629] x20: 0000000000134521 x19: 0000000000134519 x18: 005c43e000040000
[   21.140027] x17: 0001400000000000 x16: 0001700000000000 x15: 000000000000ffff
[   21.140578] x14: 000000000000000c x13: 005c006000000000 x12: 0000000000000020
[   21.140828] x11: 0000000000000000 x10: 005c000000000000 x9 : ffffd1a55c079ee0
[   21.141077] x8 : 0000000000000001 x7 : 005c03e000040000 x6 : 000000004000ffff
[   21.141490] x5 : ffff00017fffce00 x4 : 0000000000000001 x3 : 0000000000000002
[   21.141741] x2 : 0000000000134510 x1 : 0000000000000000 x0 : ffff0000c08228c0
[   21.141991] Call trace:
[   21.142093]  page_table_check_set+0x28c/0x2a8 (P)
[   21.142265]  __page_table_check_ptes_set+0x144/0x1e8
[   21.142441]  __set_ptes_anysz.constprop.0+0x160/0x1a8
[   21.142766]  contpte_set_ptes+0xe8/0x140
[   21.142907]  try_to_unmap_one+0x10c4/0x10d0
[   21.143177]  rmap_walk_anon+0x100/0x250
[   21.143315]  try_to_unmap+0xa0/0xc8
[   21.143441]  shrink_folio_list+0x59c/0x18a8
[   21.143759]  shrink_lruvec+0x664/0xbf0
[   21.144043]  shrink_node+0x218/0x878
[   21.144285]  __node_reclaim.constprop.0+0x98/0x338
[   21.144763]  user_proactive_reclaim+0x2a4/0x340
[   21.145056]  reclaim_store+0x3c/0x60
[   21.145216]  dev_attr_store+0x20/0x40
[   21.145585]  sysfs_kf_write+0x84/0xa8
[   21.145835]  kernfs_fop_write_iter+0x130/0x1c8
[   21.145994]  vfs_write+0x2b8/0x368
[   21.146119]  ksys_write+0x70/0x110
[   21.146240]  __arm64_sys_write+0x24/0x38
[   21.146380]  invoke_syscall+0x50/0x120
[   21.146513]  el0_svc_common.constprop.0+0x48/0xf8
[   21.146679]  do_el0_svc+0x28/0x40
[   21.146798]  el0_svc+0x34/0x110
[   21.146926]  el0t_64_sync_handler+0xa0/0xe8
[   21.147074]  el0t_64_sync+0x198/0x1a0
[   21.147225] Code: f9400441 b4fff241 17ffff94 d4210000 (d4210000)
[   21.147440] ---[ end trace 0000000000000000 ]---

#define _GNU_SOURCE
#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <sys/mman.h>
#include <string.h>
#include <sys/wait.h>
#include <sched.h>
#include <fcntl.h>

void write_to_reclaim() {
    const char *path = "/sys/devices/system/node/node0/reclaim";
    const char *value = "409600000000";
    int fd = open(path, O_WRONLY);
    if (fd == -1) {
        perror("open");
        exit(EXIT_FAILURE);
    }

    if (write(fd, value, sizeof("409600000000") - 1) == -1) {
        perror("write");
        close(fd);
        exit(EXIT_FAILURE);
    }

    printf("Successfully wrote %s to %s\n", value, path);
    close(fd);
}

int main()

{
	char *ptr = mmap((void *)(1UL << 30), 1UL << 16, PROT_READ | PROT_WRITE,
			 MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
	if ((unsigned long)ptr != (1UL << 30)) {
		perror("mmap");
		return 1;
	}
	
	/* a 64K folio gets faulted in */
	memset(ptr, 0, 1UL << 16);

	/* 32K half will not be shared into child */
	if (madvise(ptr, 1UL << 15, MADV_DONTFORK)) {
		perror("madvise madv dontfork");
		return 1;
	}

	pid_t pid = fork();

	if (pid < 0) {
		perror("fork");
		return 1;
	} else if (pid == 0) {
		sleep(15);
	} else {
		/* merge VMAs. now first half of the 16 ptes are writable, the other half not. */
		if (madvise(ptr, 1UL << 15, MADV_DOFORK)) {
			perror("madvise madv fork");
			return 1;
		}
		if (madvise(ptr, (1UL << 16), MADV_FREE)) {
			perror("madvise madv free");
			return 1;
		}

		/* dirty the large folio */
		(*ptr) += 10;

		write_to_reclaim();
		// sleep(10);
		waitpid(pid, NULL, 0);

	}
}
Fixes 354dffd29575 ("mm: support batched unmap for lazyfree large folios during reclamation")
Cc stable <[email protected]>
Signed-off-by Dev Jain <[email protected]>

---
Patch applies on mm-unstable (9af4957ef127).

v2->v3:
- Don't special case for anon folios

v1->v2:
- Just respect the writable bit instead of hacking in a pte_wrprotect() in failure path
- Also handle soft-dirty bit

 mm/rmap.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
mm/rmap.c +8 -1
index bff8f222004e4..5a3e408e3f179 100644
@@ -1955,7 +1955,14 @@ static inline unsigned int folio_unmap_pte_batch(struct folio *folio,
 	if (userfaultfd_wp(vma))
 		return 1;
 
-	return folio_pte_batch(folio, pvmw->pte, pte, max_nr);
+	/*
+	 * If unmap fails, we need to restore the ptes. To avoid accidentally
+	 * upgrading write permissions for ptes that were not originally
+	 * writable, and to avoid losing the soft-dirty bit, use the
+	 * appropriate FPB flags.
+	 */
+	return folio_pte_batch_flags(folio, vma, pvmw->pte, &pte, max_nr,
+				     FPB_RESPECT_WRITE | FPB_RESPECT_SOFT_DIRTY);
 }
 
 /*
#6 lore raw
In reply to Dev Jain
Quoted 169 lines · Dev Jain

We batch unmap anonymous lazyfree folios by folio_unmap_pte_batch. If the batch has a mix of writable and non-writable bits, we may end up setting the entire batch writable. Fix this by respecting writable bit during batching.
Although on a successful unmap of a lazyfree folio, the soft-dirty bit is lost, preserve it on pte restoration by respecting the bit during batching, to make the fix consistent w.r.t both writable bit and soft-dirty bit.

I was able to write the below reproducer and crash the kernel. Explanation of reproducer (set 64K mTHP to always):

Fault in a 64K large folio. Split the VMA at mid-point with MADV_DONTFORK. fork() - parent points to the folio with 8 writable ptes and 8 non-writable ptes. Merge the VMAs with MADV_DOFORK so that folio_unmap_pte_batch() can determine all the 16 ptes as a batch. Do MADV_FREE on the range to mark the folio as lazyfree. Write to the memory to dirty the pte, eventually rmap will dirty the folio. Then trigger reclaim, we will hit the pte restoration path, and the kernel will crash with the following trace:

[   21.134473] kernel BUG at mm/page_table_check.c:118!
[   21.134497] Internal error: Oops - BUG: 00000000f2000800 [#1]  SMP
[   21.135917] Modules linked in:
[   21.136085] CPU: 1 UID: 0 PID: 1735 Comm: dup-lazyfree Not tainted 7.0.0-rc1-00116-g018018a17770 #1028 PREEMPT
[   21.136858] Hardware name: linux,dummy-virt (DT)
[   21.137019] pstate: 21400005 (nzCv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--)
[   21.137308] pc : page_table_check_set+0x28c/0x2a8
[   21.137607] lr : page_table_check_set+0x134/0x2a8
[   21.137885] sp : ffff80008a3b3340
[   21.138124] x29: ffff80008a3b3340 x28: fffffdffc3d14400 x27: ffffd1a55e03d000
[   21.138623] x26: 0040000000000040 x25: ffffd1a55f7dd000 x24: 0000000000000001
[   21.139045] x23: 0000000000000001 x22: 0000000000000001 x21: ffffd1a55f217f30
[   21.139629] x20: 0000000000134521 x19: 0000000000134519 x18: 005c43e000040000
[   21.140027] x17: 0001400000000000 x16: 0001700000000000 x15: 000000000000ffff
[   21.140578] x14: 000000000000000c x13: 005c006000000000 x12: 0000000000000020
[   21.140828] x11: 0000000000000000 x10: 005c000000000000 x9 : ffffd1a55c079ee0
[   21.141077] x8 : 0000000000000001 x7 : 005c03e000040000 x6 : 000000004000ffff
[   21.141490] x5 : ffff00017fffce00 x4 : 0000000000000001 x3 : 0000000000000002
[   21.141741] x2 : 0000000000134510 x1 : 0000000000000000 x0 : ffff0000c08228c0
[   21.141991] Call trace:
[   21.142093]  page_table_check_set+0x28c/0x2a8 (P)
[   21.142265]  __page_table_check_ptes_set+0x144/0x1e8
[   21.142441]  __set_ptes_anysz.constprop.0+0x160/0x1a8
[   21.142766]  contpte_set_ptes+0xe8/0x140
[   21.142907]  try_to_unmap_one+0x10c4/0x10d0
[   21.143177]  rmap_walk_anon+0x100/0x250
[   21.143315]  try_to_unmap+0xa0/0xc8
[   21.143441]  shrink_folio_list+0x59c/0x18a8
[   21.143759]  shrink_lruvec+0x664/0xbf0
[   21.144043]  shrink_node+0x218/0x878
[   21.144285]  __node_reclaim.constprop.0+0x98/0x338
[   21.144763]  user_proactive_reclaim+0x2a4/0x340
[   21.145056]  reclaim_store+0x3c/0x60
[   21.145216]  dev_attr_store+0x20/0x40
[   21.145585]  sysfs_kf_write+0x84/0xa8
[   21.145835]  kernfs_fop_write_iter+0x130/0x1c8
[   21.145994]  vfs_write+0x2b8/0x368
[   21.146119]  ksys_write+0x70/0x110
[   21.146240]  __arm64_sys_write+0x24/0x38
[   21.146380]  invoke_syscall+0x50/0x120
[   21.146513]  el0_svc_common.constprop.0+0x48/0xf8
[   21.146679]  do_el0_svc+0x28/0x40
[   21.146798]  el0_svc+0x34/0x110
[   21.146926]  el0t_64_sync_handler+0xa0/0xe8
[   21.147074]  el0t_64_sync+0x198/0x1a0
[   21.147225] Code: f9400441 b4fff241 17ffff94 d4210000 (d4210000)
[   21.147440] ---[ end trace 0000000000000000 ]---

#define _GNU_SOURCE
#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <sys/mman.h>
#include <string.h>
#include <sys/wait.h>
#include <sched.h>
#include <fcntl.h>

void write_to_reclaim() {
   const char *path = "/sys/devices/system/node/node0/reclaim";
   const char *value = "409600000000";
   int fd = open(path, O_WRONLY);
   if (fd == -1) {
       perror("open");
       exit(EXIT_FAILURE);
   }

   if (write(fd, value, sizeof("409600000000") - 1) == -1) {
       perror("write");
       close(fd);
       exit(EXIT_FAILURE);
   }

   printf("Successfully wrote %s to %s\n", value, path);
   close(fd);
}

int main()

{
	char *ptr = mmap((void *)(1UL << 30), 1UL << 16, PROT_READ | PROT_WRITE,
			 MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
	if ((unsigned long)ptr != (1UL << 30)) {
		perror("mmap");
		return 1;
	}
	
	/* a 64K folio gets faulted in */
	memset(ptr, 0, 1UL << 16);

	/* 32K half will not be shared into child */
	if (madvise(ptr, 1UL << 15, MADV_DONTFORK)) {
		perror("madvise madv dontfork");
		return 1;
	}

	pid_t pid = fork();

	if (pid < 0) {
		perror("fork");
		return 1;
	} else if (pid == 0) {
		sleep(15);
	} else {
		/* merge VMAs. now first half of the 16 ptes are writable, the other half not. */
		if (madvise(ptr, 1UL << 15, MADV_DOFORK)) {
			perror("madvise madv fork");
			return 1;
		}
		if (madvise(ptr, (1UL << 16), MADV_FREE)) {
			perror("madvise madv free");
			return 1;
		}

		/* dirty the large folio */
		(*ptr) += 10;

		write_to_reclaim();
		// sleep(10);
		waitpid(pid, NULL, 0);

	}
}
Fixes 354dffd29575 ("mm: support batched unmap for lazyfree large folios during reclamation")
Cc stable <[email protected]>
Signed-off-by Dev Jain <[email protected]>

---
Patch applies on mm-unstable (9af4957ef127).

v2->v3:
- Don't special case for anon folios

v1->v2:
- Just respect the writable bit instead of hacking in a pte_wrprotect() in failure path
- Also handle soft-dirty bit

mm/rmap.c | 9 ++++++++-

1 file changed, 8 insertions(+), 1 deletion(-)
mm/rmap.c +0 -1
index bff8f222004e4..5a3e408e3f179 100644
@@ -1955,7 +1955,14 @@ static inline unsigned int folio_unmap_pte_batch(struct folio *folio,
 	if (userfaultfd_wp(vma))
 		return 1;
 
-	return folio_pte_batch(folio, pvmw->pte, pte, max_nr);

mm/rmap.c static inline unsigned int folio_unmap_pte_batch(struct folio *folio, · line 1967

Quoted 10 lines
mm/rmap.c +8 -0
+	/*
+	 * If unmap fails, we need to restore the ptes. To avoid accidentally
+	 * upgrading write permissions for ptes that were not originally
+	 * writable, and to avoid losing the soft-dirty bit, use the
+	 * appropriate FPB flags.
+	 */
+	return folio_pte_batch_flags(folio, vma, pvmw->pte, &pte, max_nr,
+				     FPB_RESPECT_WRITE | FPB_RESPECT_SOFT_DIRTY);
 }

Hi, Dev

When reading the code, I got one confusion. Current call flow is like below:

    try_to_unmap_one();
        nr_pages = folio_unmap_pte_batch(folio, &pvmw, flags, pteval);
	..
	pteval = get_and_clear_ptes(mm, address, pvmw.pte, nr_pages);
	..
	set_ptes(mm, address, pvmw.pte, pteval, nr_pages);

We get pteval by folio_unmap_pte_batch() but it is set again by get_and_clear_ptes(), which maybe a different value. Then we use this pteval to restore ptes.

So even we fix folio_unmap_pte_batch(), how this impact on the final restored value?

Hope I don't miss something.

Quoted 4 lines
mm/rmap.c +0 -0
 /*
Signature · 1 line
2.34.1
Signature · 2 lines
Wei Yang
Help you, Help me
author
#7 lore raw
Quoted 185 lines · Wei Yang
Quoted 169 lines · Dev Jain

We batch unmap anonymous lazyfree folios by folio_unmap_pte_batch. If the batch has a mix of writable and non-writable bits, we may end up setting the entire batch writable. Fix this by respecting writable bit during batching.
Although on a successful unmap of a lazyfree folio, the soft-dirty bit is lost, preserve it on pte restoration by respecting the bit during batching, to make the fix consistent w.r.t both writable bit and soft-dirty bit.

I was able to write the below reproducer and crash the kernel. Explanation of reproducer (set 64K mTHP to always):

Fault in a 64K large folio. Split the VMA at mid-point with MADV_DONTFORK. fork() - parent points to the folio with 8 writable ptes and 8 non-writable ptes. Merge the VMAs with MADV_DOFORK so that folio_unmap_pte_batch() can determine all the 16 ptes as a batch. Do MADV_FREE on the range to mark the folio as lazyfree. Write to the memory to dirty the pte, eventually rmap will dirty the folio. Then trigger reclaim, we will hit the pte restoration path, and the kernel will crash with the following trace:

[   21.134473] kernel BUG at mm/page_table_check.c:118!
[   21.134497] Internal error: Oops - BUG: 00000000f2000800 [#1]  SMP
[   21.135917] Modules linked in:
[   21.136085] CPU: 1 UID: 0 PID: 1735 Comm: dup-lazyfree Not tainted 7.0.0-rc1-00116-g018018a17770 #1028 PREEMPT
[   21.136858] Hardware name: linux,dummy-virt (DT)
[   21.137019] pstate: 21400005 (nzCv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--)
[   21.137308] pc : page_table_check_set+0x28c/0x2a8
[   21.137607] lr : page_table_check_set+0x134/0x2a8
[   21.137885] sp : ffff80008a3b3340
[   21.138124] x29: ffff80008a3b3340 x28: fffffdffc3d14400 x27: ffffd1a55e03d000
[   21.138623] x26: 0040000000000040 x25: ffffd1a55f7dd000 x24: 0000000000000001
[   21.139045] x23: 0000000000000001 x22: 0000000000000001 x21: ffffd1a55f217f30
[   21.139629] x20: 0000000000134521 x19: 0000000000134519 x18: 005c43e000040000
[   21.140027] x17: 0001400000000000 x16: 0001700000000000 x15: 000000000000ffff
[   21.140578] x14: 000000000000000c x13: 005c006000000000 x12: 0000000000000020
[   21.140828] x11: 0000000000000000 x10: 005c000000000000 x9 : ffffd1a55c079ee0
[   21.141077] x8 : 0000000000000001 x7 : 005c03e000040000 x6 : 000000004000ffff
[   21.141490] x5 : ffff00017fffce00 x4 : 0000000000000001 x3 : 0000000000000002
[   21.141741] x2 : 0000000000134510 x1 : 0000000000000000 x0 : ffff0000c08228c0
[   21.141991] Call trace:
[   21.142093]  page_table_check_set+0x28c/0x2a8 (P)
[   21.142265]  __page_table_check_ptes_set+0x144/0x1e8
[   21.142441]  __set_ptes_anysz.constprop.0+0x160/0x1a8
[   21.142766]  contpte_set_ptes+0xe8/0x140
[   21.142907]  try_to_unmap_one+0x10c4/0x10d0
[   21.143177]  rmap_walk_anon+0x100/0x250
[   21.143315]  try_to_unmap+0xa0/0xc8
[   21.143441]  shrink_folio_list+0x59c/0x18a8
[   21.143759]  shrink_lruvec+0x664/0xbf0
[   21.144043]  shrink_node+0x218/0x878
[   21.144285]  __node_reclaim.constprop.0+0x98/0x338
[   21.144763]  user_proactive_reclaim+0x2a4/0x340
[   21.145056]  reclaim_store+0x3c/0x60
[   21.145216]  dev_attr_store+0x20/0x40
[   21.145585]  sysfs_kf_write+0x84/0xa8
[   21.145835]  kernfs_fop_write_iter+0x130/0x1c8
[   21.145994]  vfs_write+0x2b8/0x368
[   21.146119]  ksys_write+0x70/0x110
[   21.146240]  __arm64_sys_write+0x24/0x38
[   21.146380]  invoke_syscall+0x50/0x120
[   21.146513]  el0_svc_common.constprop.0+0x48/0xf8
[   21.146679]  do_el0_svc+0x28/0x40
[   21.146798]  el0_svc+0x34/0x110
[   21.146926]  el0t_64_sync_handler+0xa0/0xe8
[   21.147074]  el0t_64_sync+0x198/0x1a0
[   21.147225] Code: f9400441 b4fff241 17ffff94 d4210000 (d4210000)
[   21.147440] ---[ end trace 0000000000000000 ]---

#define _GNU_SOURCE
#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <sys/mman.h>
#include <string.h>
#include <sys/wait.h>
#include <sched.h>
#include <fcntl.h>

void write_to_reclaim() {
   const char *path = "/sys/devices/system/node/node0/reclaim";
   const char *value = "409600000000";
   int fd = open(path, O_WRONLY);
   if (fd == -1) {
       perror("open");
       exit(EXIT_FAILURE);
   }

   if (write(fd, value, sizeof("409600000000") - 1) == -1) {
       perror("write");
       close(fd);
       exit(EXIT_FAILURE);
   }

   printf("Successfully wrote %s to %s\n", value, path);
   close(fd);
}

int main()

{
	char *ptr = mmap((void *)(1UL << 30), 1UL << 16, PROT_READ | PROT_WRITE,
			 MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
	if ((unsigned long)ptr != (1UL << 30)) {
		perror("mmap");
		return 1;
	}
	
	/* a 64K folio gets faulted in */
	memset(ptr, 0, 1UL << 16);

	/* 32K half will not be shared into child */
	if (madvise(ptr, 1UL << 15, MADV_DONTFORK)) {
		perror("madvise madv dontfork");
		return 1;
	}

	pid_t pid = fork();

	if (pid < 0) {
		perror("fork");
		return 1;
	} else if (pid == 0) {
		sleep(15);
	} else {
		/* merge VMAs. now first half of the 16 ptes are writable, the other half not. */
		if (madvise(ptr, 1UL << 15, MADV_DOFORK)) {
			perror("madvise madv fork");
			return 1;
		}
		if (madvise(ptr, (1UL << 16), MADV_FREE)) {
			perror("madvise madv free");
			return 1;
		}

		/* dirty the large folio */
		(*ptr) += 10;

		write_to_reclaim();
		// sleep(10);
		waitpid(pid, NULL, 0);

	}
}
Fixes 354dffd29575 ("mm: support batched unmap for lazyfree large folios during reclamation")
Cc stable <[email protected]>
Signed-off-by Dev Jain <[email protected]>

---
Patch applies on mm-unstable (9af4957ef127).

v2->v3:
- Don't special case for anon folios

v1->v2:
- Just respect the writable bit instead of hacking in a pte_wrprotect() in failure path
- Also handle soft-dirty bit

mm/rmap.c | 9 ++++++++-

1 file changed, 8 insertions(+), 1 deletion(-)
mm/rmap.c +0 -1
index bff8f222004e4..5a3e408e3f179 100644
@@ -1955,7 +1955,14 @@ static inline unsigned int folio_unmap_pte_batch(struct folio *folio,
 	if (userfaultfd_wp(vma))
 		return 1;
 
-	return folio_pte_batch(folio, pvmw->pte, pte, max_nr);

mm/rmap.c static inline unsigned int folio_unmap_pte_batch(struct folio *folio, · line 1967

Quoted 10 lines
mm/rmap.c +8 -0
+	/*
+	 * If unmap fails, we need to restore the ptes. To avoid accidentally
+	 * upgrading write permissions for ptes that were not originally
+	 * writable, and to avoid losing the soft-dirty bit, use the
+	 * appropriate FPB flags.
+	 */
+	return folio_pte_batch_flags(folio, vma, pvmw->pte, &pte, max_nr,
+				     FPB_RESPECT_WRITE | FPB_RESPECT_SOFT_DIRTY);
 }

Hi, Dev

When reading the code, I got one confusion. Current call flow is like below:

Quoted 8 lines
    try_to_unmap_one();
        nr_pages = folio_unmap_pte_batch(folio, &pvmw, flags, pteval);
	..
	pteval = get_and_clear_ptes(mm, address, pvmw.pte, nr_pages);
	..
	set_ptes(mm, address, pvmw.pte, pteval, nr_pages);

We get pteval by folio_unmap_pte_batch() but it is set again by

folio_unmap_pte_batch() gives the batch size, not pteval. pteval is given by get_and_clear_ptes() after accumulating a/d bits.

Quoted 5 lines

get_and_clear_ptes(), which maybe a different value. Then we use this pteval to restore ptes.

So even we fix folio_unmap_pte_batch(), how this impact on the final restored value?

By respecting writable bit, we ensure that the ptes in the batch do not have a mix of writable and non writable ptes.

So, if pteval returned by get_and_clear_ptes() is writable, then it is guaranteed via folio_unmap_pte_batch() that the all pte values of these nr_pages consecutive ptes, are writable. And vice versa.

Quoted 8 lines

Hope I don't miss something.

Quoted 4 lines
mm/rmap.c +0 -0
 /*
Signature · 1 line
2.34.1
Reviewed-by
#8 lore raw

On Tue, Mar 03, 2026 at 05:55:58PM +0530, Dev Jain wrote:
[...]

Quoted 54 lines
Quoted 42 lines
Quoted 27 lines
Fixes 354dffd29575 ("mm: support batched unmap for lazyfree large folios during reclamation")
Cc stable <[email protected]>
Signed-off-by Dev Jain <[email protected]>

---
Patch applies on mm-unstable (9af4957ef127).

v2->v3:
- Don't special case for anon folios

v1->v2:
- Just respect the writable bit instead of hacking in a pte_wrprotect() in failure path
- Also handle soft-dirty bit

mm/rmap.c | 9 ++++++++-

1 file changed, 8 insertions(+), 1 deletion(-)
mm/rmap.c +0 -1
index bff8f222004e4..5a3e408e3f179 100644
@@ -1955,7 +1955,14 @@ static inline unsigned int folio_unmap_pte_batch(struct folio *folio,
 	if (userfaultfd_wp(vma))
 		return 1;
 
-	return folio_pte_batch(folio, pvmw->pte, pte, max_nr);

mm/rmap.c static inline unsigned int folio_unmap_pte_batch(struct folio *folio, · line 1967

Quoted 10 lines
mm/rmap.c +8 -0
+	/*
+	 * If unmap fails, we need to restore the ptes. To avoid accidentally
+	 * upgrading write permissions for ptes that were not originally
+	 * writable, and to avoid losing the soft-dirty bit, use the
+	 * appropriate FPB flags.
+	 */
+	return folio_pte_batch_flags(folio, vma, pvmw->pte, &pte, max_nr,
+				     FPB_RESPECT_WRITE | FPB_RESPECT_SOFT_DIRTY);
 }

Hi, Dev

When reading the code, I got one confusion. Current call flow is like below:

Quoted 8 lines
    try_to_unmap_one();
        nr_pages = folio_unmap_pte_batch(folio, &pvmw, flags, pteval);
	..
	pteval = get_and_clear_ptes(mm, address, pvmw.pte, nr_pages);
	..
	set_ptes(mm, address, pvmw.pte, pteval, nr_pages);

We get pteval by folio_unmap_pte_batch() but it is set again by

folio_unmap_pte_batch() gives the batch size, not pteval. pteval is given by get_and_clear_ptes() after accumulating a/d bits.

Quoted 9 lines
Quoted 5 lines

get_and_clear_ptes(), which maybe a different value. Then we use this pteval to restore ptes.

So even we fix folio_unmap_pte_batch(), how this impact on the final restored value?

By respecting writable bit, we ensure that the ptes in the batch do not have a mix of writable and non writable ptes.

Thanks for your explanation. I got it.

Then the fix LGTM.

Reviewed-by Wei Yang <[email protected]>
Quoted 12 lines

So, if pteval returned by get_and_clear_ptes() is writable, then it is guaranteed via folio_unmap_pte_batch() that the all pte values of these nr_pages consecutive ptes, are writable. And vice versa.

Quoted 8 lines

Hope I don't miss something.

Quoted 4 lines
mm/rmap.c +0 -0
 /*
Signature · 1 line
2.34.1
Signature · 2 lines
Wei Yang
Help you, Help me
reviewer Tested-by on v3 1/1
#9 lore raw
In reply to Dev Jain
Quoted 78 lines · Dev Jain

We batch unmap anonymous lazyfree folios by folio_unmap_pte_batch. If the batch has a mix of writable and non-writable bits, we may end up setting the entire batch writable. Fix this by respecting writable bit during batching.
Although on a successful unmap of a lazyfree folio, the soft-dirty bit is lost, preserve it on pte restoration by respecting the bit during batching, to make the fix consistent w.r.t both writable bit and soft-dirty bit.

I was able to write the below reproducer and crash the kernel. Explanation of reproducer (set 64K mTHP to always):

Fault in a 64K large folio. Split the VMA at mid-point with MADV_DONTFORK. fork() - parent points to the folio with 8 writable ptes and 8 non-writable ptes. Merge the VMAs with MADV_DOFORK so that folio_unmap_pte_batch() can determine all the 16 ptes as a batch. Do MADV_FREE on the range to mark the folio as lazyfree. Write to the memory to dirty the pte, eventually rmap will dirty the folio. Then trigger reclaim, we will hit the pte restoration path, and the kernel will crash with the following trace:

[   21.134473] kernel BUG at mm/page_table_check.c:118!
[   21.134497] Internal error: Oops - BUG: 00000000f2000800 [#1]  SMP
[   21.135917] Modules linked in:
[   21.136085] CPU: 1 UID: 0 PID: 1735 Comm: dup-lazyfree Not tainted 7.0.0-rc1-00116-g018018a17770 #1028 PREEMPT
[   21.136858] Hardware name: linux,dummy-virt (DT)
[   21.137019] pstate: 21400005 (nzCv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--)
[   21.137308] pc : page_table_check_set+0x28c/0x2a8
[   21.137607] lr : page_table_check_set+0x134/0x2a8
[   21.137885] sp : ffff80008a3b3340
[   21.138124] x29: ffff80008a3b3340 x28: fffffdffc3d14400 x27: ffffd1a55e03d000
[   21.138623] x26: 0040000000000040 x25: ffffd1a55f7dd000 x24: 0000000000000001
[   21.139045] x23: 0000000000000001 x22: 0000000000000001 x21: ffffd1a55f217f30
[   21.139629] x20: 0000000000134521 x19: 0000000000134519 x18: 005c43e000040000
[   21.140027] x17: 0001400000000000 x16: 0001700000000000 x15: 000000000000ffff
[   21.140578] x14: 000000000000000c x13: 005c006000000000 x12: 0000000000000020
[   21.140828] x11: 0000000000000000 x10: 005c000000000000 x9 : ffffd1a55c079ee0
[   21.141077] x8 : 0000000000000001 x7 : 005c03e000040000 x6 : 000000004000ffff
[   21.141490] x5 : ffff00017fffce00 x4 : 0000000000000001 x3 : 0000000000000002
[   21.141741] x2 : 0000000000134510 x1 : 0000000000000000 x0 : ffff0000c08228c0
[   21.141991] Call trace:
[   21.142093]  page_table_check_set+0x28c/0x2a8 (P)
[   21.142265]  __page_table_check_ptes_set+0x144/0x1e8
[   21.142441]  __set_ptes_anysz.constprop.0+0x160/0x1a8
[   21.142766]  contpte_set_ptes+0xe8/0x140
[   21.142907]  try_to_unmap_one+0x10c4/0x10d0
[   21.143177]  rmap_walk_anon+0x100/0x250
[   21.143315]  try_to_unmap+0xa0/0xc8
[   21.143441]  shrink_folio_list+0x59c/0x18a8
[   21.143759]  shrink_lruvec+0x664/0xbf0
[   21.144043]  shrink_node+0x218/0x878
[   21.144285]  __node_reclaim.constprop.0+0x98/0x338
[   21.144763]  user_proactive_reclaim+0x2a4/0x340
[   21.145056]  reclaim_store+0x3c/0x60
[   21.145216]  dev_attr_store+0x20/0x40
[   21.145585]  sysfs_kf_write+0x84/0xa8
[   21.145835]  kernfs_fop_write_iter+0x130/0x1c8
[   21.145994]  vfs_write+0x2b8/0x368
[   21.146119]  ksys_write+0x70/0x110
[   21.146240]  __arm64_sys_write+0x24/0x38
[   21.146380]  invoke_syscall+0x50/0x120
[   21.146513]  el0_svc_common.constprop.0+0x48/0xf8
[   21.146679]  do_el0_svc+0x28/0x40
[   21.146798]  el0_svc+0x34/0x110
[   21.146926]  el0t_64_sync_handler+0xa0/0xe8
[   21.147074]  el0t_64_sync+0x198/0x1a0
[   21.147225] Code: f9400441 b4fff241 17ffff94 d4210000 (d4210000)
[   21.147440] ---[ end trace 0000000000000000 ]---

#define _GNU_SOURCE
#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <sys/mman.h>
#include <string.h>
#include <sys/wait.h>
#include <sched.h>
#include <fcntl.h>

Quoted 2 lines
void write_to_reclaim() {
   const char *path = "/sys/devices/system/node/node0/reclaim";

I wasn't able to get this reproducer working with node reclaim, but using memcg v1 memory.force_empty worked fine for me.

[...]

Quoted 5 lines
Fixes 354dffd29575 ("mm: support batched unmap for lazyfree large folios during reclamation")
Cc stable <[email protected]>
Signed-off-by Dev Jain <[email protected]>

---

Thanks!

Tested-by Lance Yang <[email protected]>