Skip to contentSkip to conversation

mm: recheck PageCompound() before printing out info of head page

PATCH In review Naoya Horiguchi 3 participants 4 messages Apr 15, 2022 – Apr 16, 2022

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 mm: recheck PageCompound() before printing out info of head page

From: Naoya Horiguchi <[email protected]>

The following VM_BUG_ON_FOLIO() is triggered when memory error event happens on the (thp/folio) pages which are about to be freed:

  [ 1160.232771] page:00000000b36a8a0f refcount:1 mapcount:0 mapping:0000000000000000 index:0x1 pfn:0x16a000
  [ 1160.236916] page:00000000b36a8a0f refcount:0 mapcount:0 mapping:0000000000000000 index:0x1 pfn:0x16a000
  [ 1160.240684] flags: 0x57ffffc0800000(hwpoison|node=1|zone=2|lastcpupid=0x1fffff)
  [ 1160.243458] raw: 0057ffffc0800000 dead000000000100 dead000000000122 0000000000000000
  [ 1160.246268] raw: 0000000000000001 0000000000000000 00000000ffffffff 0000000000000000
  [ 1160.249197] page dumped because: VM_BUG_ON_FOLIO(!folio_test_large(folio))
  [ 1160.251815] ------------[ cut here ]------------
  [ 1160.253438] kernel BUG at include/linux/mm.h:788!
  [ 1160.256162] invalid opcode: 0000 [#1] PREEMPT SMP PTI
  [ 1160.258172] CPU: 2 PID: 115368 Comm: mceinj.sh Tainted: G            E     5.18.0-rc1-v5.18-rc1-220404-2353-005-g83111+ #3
  [ 1160.262049] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1.fc35 04/01/2014
  [ 1160.265103] RIP: 0010:dump_page.cold+0x27e/0x2bd
  [ 1160.266757] Code: fe ff ff 48 c7 c6 81 f1 5a 98 e9 4c fe ff ff 48 c7 c6 a1 95 59 98 e9 40 fe ff ff 48 c7 c6 50 bf 5a 98 48 89 ef e8 9d 04 6d ff <0f> 0b 41 f7 c4 ff 0f 00 00 0f 85 9f fd ff ff 49 8b 04 24 a9 00 00
  [ 1160.273180] RSP: 0018:ffffaa2c4d59fd18 EFLAGS: 00010292
  [ 1160.274969] RAX: 000000000000003e RBX: 0000000000000001 RCX: 0000000000000000
  [ 1160.277263] RDX: 0000000000000001 RSI: ffffffff985995a1 RDI: 00000000ffffffff
  [ 1160.279571] RBP: ffffdc9c45a80000 R08: 0000000000000000 R09: 00000000ffffdfff
  [ 1160.281794] R10: ffffaa2c4d59fb08 R11: ffffffff98940d08 R12: ffffdc9c45a80000
  [ 1160.283920] R13: ffffffff985b6f94 R14: 0000000000000000 R15: ffffdc9c45a80000
  [ 1160.286641] FS:  00007eff54ce1740(0000) GS:ffff99c67bd00000(0000) knlGS:0000000000000000
  [ 1160.289498] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  [ 1160.291106] CR2: 00005628381a5f68 CR3: 0000000104712003 CR4: 0000000000170ee0
  [ 1160.293031] Call Trace:
  [ 1160.293724]  <TASK>
  [ 1160.294334]  get_hwpoison_page+0x47d/0x570
  [ 1160.295474]  memory_failure+0x106/0xaa0
  [ 1160.296474]  ? security_capable+0x36/0x50
  [ 1160.297524]  hard_offline_page_store+0x43/0x80
  [ 1160.298684]  kernfs_fop_write_iter+0x11c/0x1b0
  [ 1160.299829]  new_sync_write+0xf9/0x160
  [ 1160.300810]  vfs_write+0x209/0x290
  [ 1160.301835]  ksys_write+0x4f/0xc0
  [ 1160.302718]  do_syscall_64+0x3b/0x90
  [ 1160.303664]  entry_SYSCALL_64_after_hwframe+0x44/0xae
  [ 1160.304981] RIP: 0033:0x7eff54b018b7

As shown in the RIP address, this VM_BUG_ON in folio_entire_mapcount() is called from dump_page("hwpoison: unhandlable page") in get_any_page(). The below explains the mechanism of the race:

CPU 0 CPU 1

memory_failure
get_hwpoison_page
get_any_page
dump_page
compound = PageCompound
free_pages_prepare
page->flags &= ~PAGE_FLAGS_CHECK_AT_PREP
folio_entire_mapcount
VM_BUG_ON_FOLIO(!folio_test_large(folio))

So in order to avoid the BUG_ON, recheck PageCompound() before __dump_page() prints out information about the head page. Precisely saying, this does not close up the race, but mitigates the impact to just inaccurate info, which should be acceptable.

Fixes 74e8ee4708a8 ("mm: Turn head_compound_mapcount() into folio_entire_mapcount()")

---
I'm actually not sure how widely the race condition could affect. If we can consider this as a bug specific to memory_failure(), simply replacing the dump_page() with simple printk() might be a proper fix (I confirmed that that fix also works). But if any other dump_page()'s caller could race with freeing thp/folio, the more generic fix (I'm suggesting in this patch) would be desirable.

---
 mm/debug.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
mm/debug.c +1 -1
index bef329bf28f0..fce63b200bea 100644
@@ -93,7 +93,7 @@ static void __dump_page(struct page *page)
 	pr_warn("page:%p refcount:%d mapcount:%d mapping:%p index:%#lx pfn:%#lx\n",
 			page, page_ref_count(head), mapcount, mapping,
 			page_to_pgoff(page), page_to_pfn(page));
-	if (compound) {
+	if (compound && PageCompound(page)) {
 		pr_warn("head:%p order:%u compound_mapcount:%d compound_pincount:%d\n",
 				head, compound_order(head),
 				folio_entire_mapcount(folio),
Signature · 1 line
2.25.1
#1 lore raw
Quoted 57 lines · Naoya Horiguchi

From: Naoya Horiguchi <[email protected]>

The following VM_BUG_ON_FOLIO() is triggered when memory error event happens on the (thp/folio) pages which are about to be freed:

   [ 1160.232771] page:00000000b36a8a0f refcount:1 mapcount:0 mapping:0000000000000000 index:0x1 pfn:0x16a000
   [ 1160.236916] page:00000000b36a8a0f refcount:0 mapcount:0 mapping:0000000000000000 index:0x1 pfn:0x16a000
   [ 1160.240684] flags: 0x57ffffc0800000(hwpoison|node=1|zone=2|lastcpupid=0x1fffff)
   [ 1160.243458] raw: 0057ffffc0800000 dead000000000100 dead000000000122 0000000000000000
   [ 1160.246268] raw: 0000000000000001 0000000000000000 00000000ffffffff 0000000000000000
   [ 1160.249197] page dumped because: VM_BUG_ON_FOLIO(!folio_test_large(folio))
   [ 1160.251815] ------------[ cut here ]------------
   [ 1160.253438] kernel BUG at include/linux/mm.h:788!
   [ 1160.256162] invalid opcode: 0000 [#1] PREEMPT SMP PTI
   [ 1160.258172] CPU: 2 PID: 115368 Comm: mceinj.sh Tainted: G            E     5.18.0-rc1-v5.18-rc1-220404-2353-005-g83111+ #3
   [ 1160.262049] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1.fc35 04/01/2014
   [ 1160.265103] RIP: 0010:dump_page.cold+0x27e/0x2bd
   [ 1160.266757] Code: fe ff ff 48 c7 c6 81 f1 5a 98 e9 4c fe ff ff 48 c7 c6 a1 95 59 98 e9 40 fe ff ff 48 c7 c6 50 bf 5a 98 48 89 ef e8 9d 04 6d ff <0f> 0b 41 f7 c4 ff 0f 00 00 0f 85 9f fd ff ff 49 8b 04 24 a9 00 00
   [ 1160.273180] RSP: 0018:ffffaa2c4d59fd18 EFLAGS: 00010292
   [ 1160.274969] RAX: 000000000000003e RBX: 0000000000000001 RCX: 0000000000000000
   [ 1160.277263] RDX: 0000000000000001 RSI: ffffffff985995a1 RDI: 00000000ffffffff
   [ 1160.279571] RBP: ffffdc9c45a80000 R08: 0000000000000000 R09: 00000000ffffdfff
   [ 1160.281794] R10: ffffaa2c4d59fb08 R11: ffffffff98940d08 R12: ffffdc9c45a80000
   [ 1160.283920] R13: ffffffff985b6f94 R14: 0000000000000000 R15: ffffdc9c45a80000
   [ 1160.286641] FS:  00007eff54ce1740(0000) GS:ffff99c67bd00000(0000) knlGS:0000000000000000
   [ 1160.289498] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
   [ 1160.291106] CR2: 00005628381a5f68 CR3: 0000000104712003 CR4: 0000000000170ee0
   [ 1160.293031] Call Trace:
   [ 1160.293724]  <TASK>
   [ 1160.294334]  get_hwpoison_page+0x47d/0x570
   [ 1160.295474]  memory_failure+0x106/0xaa0
   [ 1160.296474]  ? security_capable+0x36/0x50
   [ 1160.297524]  hard_offline_page_store+0x43/0x80
   [ 1160.298684]  kernfs_fop_write_iter+0x11c/0x1b0
   [ 1160.299829]  new_sync_write+0xf9/0x160
   [ 1160.300810]  vfs_write+0x209/0x290
   [ 1160.301835]  ksys_write+0x4f/0xc0
   [ 1160.302718]  do_syscall_64+0x3b/0x90
   [ 1160.303664]  entry_SYSCALL_64_after_hwframe+0x44/0xae
   [ 1160.304981] RIP: 0033:0x7eff54b018b7

As shown in the RIP address, this VM_BUG_ON in folio_entire_mapcount() is called from dump_page("hwpoison: unhandlable page") in get_any_page(). The below explains the mechanism of the race:

CPU 0 CPU 1

memory_failure
get_hwpoison_page
get_any_page
dump_page
compound = PageCompound
free_pages_prepare
page->flags &= ~PAGE_FLAGS_CHECK_AT_PREP
folio_entire_mapcount
VM_BUG_ON_FOLIO(!folio_test_large(folio))

Quoted 4 lines

So in order to avoid the BUG_ON, recheck PageCompound() before __dump_page() prints out information about the head page. Precisely saying, this does not close up the race, but mitigates the impact to just inaccurate info, which should be acceptable.

Actually, I don't think it really closes the race even to that extent, because a machine check (MCE, on x86) could come in at any point, so rechecking (without locks) won't really completely fix such a bug, right?

Quoted 7 lines
Fixes 74e8ee4708a8 ("mm: Turn head_compound_mapcount() into folio_entire_mapcount()")

---
I'm actually not sure how widely the race condition could affect. If we can consider this as a bug specific to memory_failure(), simply replacing the dump_page() with simple printk() might be a proper fix

I think we should have separate, limited, special case functions that are safe for use by exotic things like the MCE handler. And so just doing a printk there seems better to me.

Quoted 3 lines

(I confirmed that that fix also works). But if any other dump_page()'s caller could race with freeing thp/folio, the more generic fix (I'm suggesting in this patch) would be desirable.

*If* there are other races, they should be addressed separately, IMHO. Because MCE should be a special case.

Quoted 9 lines
---
  mm/debug.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)
mm/debug.c +0 -0
index bef329bf28f0..fce63b200bea 100644

mm/debug.c static void __dump_page(struct page *page) · line 96

Quoted 5 lines
mm/debug.c +1 -1
@@ -93,7 +93,7 @@ static void __dump_page(struct page *page)
  	pr_warn("page:%p refcount:%d mapcount:%d mapping:%p index:%#lx pfn:%#lx\n",
  			page, page_ref_count(head), mapcount, mapping,
  			page_to_pgoff(page), page_to_pfn(page));
-	if (compound) {
+	if (compound && PageCompound(page)) {

If people decide to keep this approach, then I'd go with something that makes it clear, like:

	/* Re-check, in a semi-doomed attempt to avoid racing with MCE: */
	compound = PageCompound(page);

:)

thanks,

Signature · 2 lines
John Hubbard
NVIDIA
Quoted 3 lines
mm/debug.c +0 -0
  		pr_warn("head:%p order:%u compound_mapcount:%d compound_pincount:%d\n",
  				head, compound_order(head),
  				folio_entire_mapcount(folio),
#2 lore raw
Quoted 58 lines · John Hubbard
Quoted 57 lines · Naoya Horiguchi

From: Naoya Horiguchi <[email protected]>

The following VM_BUG_ON_FOLIO() is triggered when memory error event happens on the (thp/folio) pages which are about to be freed:

   [ 1160.232771] page:00000000b36a8a0f refcount:1 mapcount:0 mapping:0000000000000000 index:0x1 pfn:0x16a000
   [ 1160.236916] page:00000000b36a8a0f refcount:0 mapcount:0 mapping:0000000000000000 index:0x1 pfn:0x16a000
   [ 1160.240684] flags: 0x57ffffc0800000(hwpoison|node=1|zone=2|lastcpupid=0x1fffff)
   [ 1160.243458] raw: 0057ffffc0800000 dead000000000100 dead000000000122 0000000000000000
   [ 1160.246268] raw: 0000000000000001 0000000000000000 00000000ffffffff 0000000000000000
   [ 1160.249197] page dumped because: VM_BUG_ON_FOLIO(!folio_test_large(folio))
   [ 1160.251815] ------------[ cut here ]------------
   [ 1160.253438] kernel BUG at include/linux/mm.h:788!
   [ 1160.256162] invalid opcode: 0000 [#1] PREEMPT SMP PTI
   [ 1160.258172] CPU: 2 PID: 115368 Comm: mceinj.sh Tainted: G            E     5.18.0-rc1-v5.18-rc1-220404-2353-005-g83111+ #3
   [ 1160.262049] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1.fc35 04/01/2014
   [ 1160.265103] RIP: 0010:dump_page.cold+0x27e/0x2bd
   [ 1160.266757] Code: fe ff ff 48 c7 c6 81 f1 5a 98 e9 4c fe ff ff 48 c7 c6 a1 95 59 98 e9 40 fe ff ff 48 c7 c6 50 bf 5a 98 48 89 ef e8 9d 04 6d ff <0f> 0b 41 f7 c4 ff 0f 00 00 0f 85 9f fd ff ff 49 8b 04 24 a9 00 00
   [ 1160.273180] RSP: 0018:ffffaa2c4d59fd18 EFLAGS: 00010292
   [ 1160.274969] RAX: 000000000000003e RBX: 0000000000000001 RCX: 0000000000000000
   [ 1160.277263] RDX: 0000000000000001 RSI: ffffffff985995a1 RDI: 00000000ffffffff
   [ 1160.279571] RBP: ffffdc9c45a80000 R08: 0000000000000000 R09: 00000000ffffdfff
   [ 1160.281794] R10: ffffaa2c4d59fb08 R11: ffffffff98940d08 R12: ffffdc9c45a80000
   [ 1160.283920] R13: ffffffff985b6f94 R14: 0000000000000000 R15: ffffdc9c45a80000
   [ 1160.286641] FS:  00007eff54ce1740(0000) GS:ffff99c67bd00000(0000) knlGS:0000000000000000
   [ 1160.289498] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
   [ 1160.291106] CR2: 00005628381a5f68 CR3: 0000000104712003 CR4: 0000000000170ee0
   [ 1160.293031] Call Trace:
   [ 1160.293724]  <TASK>
   [ 1160.294334]  get_hwpoison_page+0x47d/0x570
   [ 1160.295474]  memory_failure+0x106/0xaa0
   [ 1160.296474]  ? security_capable+0x36/0x50
   [ 1160.297524]  hard_offline_page_store+0x43/0x80
   [ 1160.298684]  kernfs_fop_write_iter+0x11c/0x1b0
   [ 1160.299829]  new_sync_write+0xf9/0x160
   [ 1160.300810]  vfs_write+0x209/0x290
   [ 1160.301835]  ksys_write+0x4f/0xc0
   [ 1160.302718]  do_syscall_64+0x3b/0x90
   [ 1160.303664]  entry_SYSCALL_64_after_hwframe+0x44/0xae
   [ 1160.304981] RIP: 0033:0x7eff54b018b7

As shown in the RIP address, this VM_BUG_ON in folio_entire_mapcount() is called from dump_page("hwpoison: unhandlable page") in get_any_page(). The below explains the mechanism of the race:

   CPU 0                                       CPU 1

     memory_failure
       get_hwpoison_page
         get_any_page
           dump_page
             compound = PageCompound
                                                 free_pages_prepare page->flags &= ~PAGE_FLAGS_CHECK_AT_PREP folio_entire_mapcount
               VM_BUG_ON_FOLIO(!folio_test_large(folio))

Quoted 8 lines
Quoted 4 lines

So in order to avoid the BUG_ON, recheck PageCompound() before __dump_page() prints out information about the head page.  Precisely saying, this does not close up the race, but mitigates the impact to just inaccurate info, which should be acceptable.

Actually, I don't think it really closes the race even to that extent, because a machine check (MCE, on x86) could come in at any point, so rechecking (without locks) won't really completely fix such a bug, right?

It might also be unsafe to do dump_mapping or dump_page_owner when page could be freed at any time or we might come across some use-after-free issues?

Quoted 38 lines
Quoted 8 lines
Fixes 74e8ee4708a8 ("mm: Turn head_compound_mapcount() into folio_entire_mapcount()")

---
I'm actually not sure how widely the race condition could affect. If we can consider this as a bug specific to memory_failure(), simply replacing the dump_page() with simple printk() might be a proper fix

I think we should have separate, limited, special case functions that are safe for use by exotic things like the MCE handler. And so just doing a printk there seems better to me.

Quoted 3 lines

(I confirmed that that fix also works).  But if any other dump_page()'s caller could race with freeing thp/folio, the more generic fix (I'm suggesting in this patch) would be desirable.

*If* there are other races, they should be addressed separately, IMHO. Because MCE should be a special case.

Quoted 9 lines

---
  mm/debug.c | 2 +-

  1 file changed, 1 insertion(+), 1 deletion(-)
mm/debug.c +0 -0
index bef329bf28f0..fce63b200bea 100644

mm/debug.c static void __dump_page(struct page *page) · line 96

Quoted 5 lines
mm/debug.c +1 -1
@@ -93,7 +93,7 @@ static void __dump_page(struct page *page)
       pr_warn("page:%p refcount:%d mapcount:%d mapping:%p index:%#lx pfn:%#lx\n",
               page, page_ref_count(head), mapcount, mapping,
               page_to_pgoff(page), page_to_pfn(page));
-    if (compound) {
+    if (compound && PageCompound(page)) {

If people decide to keep this approach, then I'd go with something that makes it clear, like:

Quoted 2 lines

    /* Re-check, in a semi-doomed attempt to avoid racing with MCE: */ compound = PageCompound(page);

Agree. A comment will be needed or the code might look obscure.

Thanks!

Quoted 4 lines

:)

thanks,

#3 lore raw
Quoted 64 lines · Miaohe Lin
Quoted 63 lines · John Hubbard
Quoted 61 lines · Naoya Horiguchi

From: Naoya Horiguchi <[email protected]>

The following VM_BUG_ON_FOLIO() is triggered when memory error event happens on the (thp/folio) pages which are about to be freed:

   [ 1160.232771] page:00000000b36a8a0f refcount:1 mapcount:0 mapping:0000000000000000 index:0x1 pfn:0x16a000
   [ 1160.236916] page:00000000b36a8a0f refcount:0 mapcount:0 mapping:0000000000000000 index:0x1 pfn:0x16a000
   [ 1160.240684] flags: 0x57ffffc0800000(hwpoison|node=1|zone=2|lastcpupid=0x1fffff)
   [ 1160.243458] raw: 0057ffffc0800000 dead000000000100 dead000000000122 0000000000000000
   [ 1160.246268] raw: 0000000000000001 0000000000000000 00000000ffffffff 0000000000000000
   [ 1160.249197] page dumped because: VM_BUG_ON_FOLIO(!folio_test_large(folio))
   [ 1160.251815] ------------[ cut here ]------------
   [ 1160.253438] kernel BUG at include/linux/mm.h:788!
   [ 1160.256162] invalid opcode: 0000 [#1] PREEMPT SMP PTI
   [ 1160.258172] CPU: 2 PID: 115368 Comm: mceinj.sh Tainted: G            E     5.18.0-rc1-v5.18-rc1-220404-2353-005-g83111+ #3
   [ 1160.262049] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1.fc35 04/01/2014
   [ 1160.265103] RIP: 0010:dump_page.cold+0x27e/0x2bd
   [ 1160.266757] Code: fe ff ff 48 c7 c6 81 f1 5a 98 e9 4c fe ff ff 48 c7 c6 a1 95 59 98 e9 40 fe ff ff 48 c7 c6 50 bf 5a 98 48 89 ef e8 9d 04 6d ff <0f> 0b 41 f7 c4 ff 0f 00 00 0f 85 9f fd ff ff 49 8b 04 24 a9 00 00
   [ 1160.273180] RSP: 0018:ffffaa2c4d59fd18 EFLAGS: 00010292
   [ 1160.274969] RAX: 000000000000003e RBX: 0000000000000001 RCX: 0000000000000000
   [ 1160.277263] RDX: 0000000000000001 RSI: ffffffff985995a1 RDI: 00000000ffffffff
   [ 1160.279571] RBP: ffffdc9c45a80000 R08: 0000000000000000 R09: 00000000ffffdfff
   [ 1160.281794] R10: ffffaa2c4d59fb08 R11: ffffffff98940d08 R12: ffffdc9c45a80000
   [ 1160.283920] R13: ffffffff985b6f94 R14: 0000000000000000 R15: ffffdc9c45a80000
   [ 1160.286641] FS:  00007eff54ce1740(0000) GS:ffff99c67bd00000(0000) knlGS:0000000000000000
   [ 1160.289498] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
   [ 1160.291106] CR2: 00005628381a5f68 CR3: 0000000104712003 CR4: 0000000000170ee0
   [ 1160.293031] Call Trace:
   [ 1160.293724]  <TASK>
   [ 1160.294334]  get_hwpoison_page+0x47d/0x570
   [ 1160.295474]  memory_failure+0x106/0xaa0
   [ 1160.296474]  ? security_capable+0x36/0x50
   [ 1160.297524]  hard_offline_page_store+0x43/0x80
   [ 1160.298684]  kernfs_fop_write_iter+0x11c/0x1b0
   [ 1160.299829]  new_sync_write+0xf9/0x160
   [ 1160.300810]  vfs_write+0x209/0x290
   [ 1160.301835]  ksys_write+0x4f/0xc0
   [ 1160.302718]  do_syscall_64+0x3b/0x90
   [ 1160.303664]  entry_SYSCALL_64_after_hwframe+0x44/0xae
   [ 1160.304981] RIP: 0033:0x7eff54b018b7

As shown in the RIP address, this VM_BUG_ON in folio_entire_mapcount() is called from dump_page("hwpoison: unhandlable page") in get_any_page(). The below explains the mechanism of the race:

   CPU 0                                       CPU 1

     memory_failure
       get_hwpoison_page
         get_any_page
           dump_page
             compound = PageCompound
                                                 free_pages_prepare page->flags &= ~PAGE_FLAGS_CHECK_AT_PREP folio_entire_mapcount
               VM_BUG_ON_FOLIO(!folio_test_large(folio))

So in order to avoid the BUG_ON, recheck PageCompound() before __dump_page() prints out information about the head page.  Precisely saying, this does not close up the race, but mitigates the impact to just inaccurate info, which should be acceptable.

Quoted 6 lines
Quoted 3 lines

Actually, I don't think it really closes the race even to that extent, because a machine check (MCE, on x86) could come in at any point, so rechecking (without locks) won't really completely fix such a bug, right?

It might also be unsafe to do dump_mapping or dump_page_owner when page could be freed at any time or we might come across some use-after-free issues?

OK, so memory_failure() simply shouldn't call dump_page() before taking page refcnt.

Quoted 10 lines
Quoted 10 lines
Quoted 9 lines
Fixes 74e8ee4708a8 ("mm: Turn head_compound_mapcount() into folio_entire_mapcount()")

---
I'm actually not sure how widely the race condition could affect. If we can consider this as a bug specific to memory_failure(), simply replacing the dump_page() with simple printk() might be a proper fix

Quoted 3 lines
Quoted 3 lines

I think we should have separate, limited, special case functions that are safe for use by exotic things like the MCE handler. And so just doing a printk there seems better to me.

Thanks, so now I lean to the printk-approach.

Quoted 7 lines
Quoted 7 lines
Quoted 4 lines

(I confirmed that that fix also works).  But if any other dump_page()'s caller could race with freeing thp/folio, the more generic fix (I'm suggesting in this patch) would be desirable.

*If* there are other races, they should be addressed separately, IMHO. Because MCE should be a special case.

I agree.

Thanks,
Naoya Horiguchi

Quoted 31 lines
Quoted 19 lines
Quoted 10 lines

---
  mm/debug.c | 2 +-

  1 file changed, 1 insertion(+), 1 deletion(-)
mm/debug.c +0 -0
index bef329bf28f0..fce63b200bea 100644

mm/debug.c static void __dump_page(struct page *page) · line 96

Quoted 5 lines
mm/debug.c +1 -1
@@ -93,7 +93,7 @@ static void __dump_page(struct page *page)
       pr_warn("page:%p refcount:%d mapcount:%d mapping:%p index:%#lx pfn:%#lx\n",
               page, page_ref_count(head), mapcount, mapping,
               page_to_pgoff(page), page_to_pfn(page));
-    if (compound) {
+    if (compound && PageCompound(page)) {

If people decide to keep this approach, then I'd go with something that makes it clear, like:

Quoted 2 lines

    /* Re-check, in a semi-doomed attempt to avoid racing with MCE: */ compound = PageCompound(page);

Agree. A comment will be needed or the code might look obscure.

Thanks!

Quoted 4 lines

:)

thanks,