fs/proc/task_mmu.c:1042 pagemap_hugetlb() error: we previously assumed 'vma' could be null (see line 1037)
Subsystems (3)
A message that is not in the archive (sent off-list or not yet received): 52fdd350.dwn4aII31EyWlDq9%[email protected]
PATCH 97/220 fs/proc/task_mmu.c:1042 pagemap_hugetlb() error: we previously assumed 'vma' could be null (see line 1037)
Hi Naoya,
FYI, there are new smatch warnings show up in
tree: git://git.cmpxchg.org/linux-mmotm.git master head: 0363f94bc1c9b81f23ee7d2446331eb288568ea7 commit: 81272031cc2831a3d1abb3c681f1188aa36a1454 [97/220] pagewalk: remove argument hmask from hugetlb_entry()
fs/proc/task_mmu.c:1042 pagemap_hugetlb() error: we previously assumed 'vma' could be null (see line 1037)
vim +/vma +1042 fs/proc/task_mmu.c
d9104d1c Cyrill Gorcunov 2013-09-11 1031 int flags2;
16fbdce6 Konstantin Khlebnikov 2012-05-10 1032 pagemap_entry_t pme;
81272031 Naoya Horiguchi 2014-02-13 1033 unsigned long hmask;
5dc37642 Naoya Horiguchi 2009-12-14 1034
d9104d1c Cyrill Gorcunov 2013-09-11 1035 WARN_ON_ONCE(!vma);
d9104d1c Cyrill Gorcunov 2013-09-11 1036
d9104d1c Cyrill Gorcunov 2013-09-11 @1037 if (vma && (vma->vm_flags & VM_SOFTDIRTY))
d9104d1c Cyrill Gorcunov 2013-09-11 1038 flags2 = __PM_SOFT_DIRTY;
d9104d1c Cyrill Gorcunov 2013-09-11 1039 else
d9104d1c Cyrill Gorcunov 2013-09-11 1040 flags2 = 0;
d9104d1c Cyrill Gorcunov 2013-09-11 1041
21a2f342 Naoya Horiguchi 2014-02-13 @1042 hmask = huge_page_mask(hstate_vma(vma));
5dc37642 Naoya Horiguchi 2009-12-14 1043 for (; addr != end; addr += PAGE_SIZE) {
116354d1 Naoya Horiguchi 2010-04-06 1044 int offset = (addr & ~hmask) >> PAGE_SHIFT;
d9104d1c Cyrill Gorcunov 2013-09-11 1045 huge_pte_to_pagemap_entry(&pme, pm, *pte, offset, flags2);---
0-DAY kernel build testing backend Open Source Technology Center
http://lists.01.org/mailman/listinfo/kbuild Intel Corporation
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
PATCH 1/1 fs/proc/task_mmu.c: assume non-NULL vma in pagemap_hugetlb() (Re: [mmotm:master 97/220] fs/proc/task_mmu.c:1042 pagemap_hugetlb() error: we previously assumed 'vma' could be null (see line 1037))
Hi Fengguang,
On Fri, Feb 14, 2014 at 09:04:50PM +0800, Fengguang Wu wrote:
...
Quoted 10 lines
FYI, there are new smatch warnings show up in
tree: git://git.cmpxchg.org/linux-mmotm.git master head: 0363f94bc1c9b81f23ee7d2446331eb288568ea7 commit: 81272031cc2831a3d1abb3c681f1188aa36a1454 [97/220] pagewalk: remove argument hmask from hugetlb_entry()
fs/proc/task_mmu.c:1042pagemap_hugetlb()error: we previously assumed 'vma' could be null (see line 1037)vim +/vma +1042
fs/proc/task_mmu.c
Quoted 15 lines
d9104d1c Cyrill Gorcunov 2013-09-11 1031 int flags2;
16fbdce6 Konstantin Khlebnikov 2012-05-10 1032 pagemap_entry_t pme;
81272031 Naoya Horiguchi 2014-02-13 1033 unsigned long hmask;
5dc37642 Naoya Horiguchi 2009-12-14 1034
d9104d1c Cyrill Gorcunov 2013-09-11 1035 WARN_ON_ONCE(!vma);
d9104d1c Cyrill Gorcunov 2013-09-11 1036
d9104d1c Cyrill Gorcunov 2013-09-11 @1037 if (vma && (vma->vm_flags & VM_SOFTDIRTY))
d9104d1c Cyrill Gorcunov 2013-09-11 1038 flags2 = __PM_SOFT_DIRTY;
d9104d1c Cyrill Gorcunov 2013-09-11 1039 else
d9104d1c Cyrill Gorcunov 2013-09-11 1040 flags2 = 0;
d9104d1c Cyrill Gorcunov 2013-09-11 1041
21a2f342 Naoya Horiguchi 2014-02-13 @1042 hmask = huge_page_mask(hstate_vma(vma));
5dc37642 Naoya Horiguchi 2009-12-14 1043 for (; addr != end; addr += PAGE_SIZE) {
116354d1 Naoya Horiguchi 2010-04-06 1044 int offset = (addr & ~hmask) >> PAGE_SHIFT;
d9104d1c Cyrill Gorcunov 2013-09-11 1045 huge_pte_to_pagemap_entry(&pme, pm, *pte, offset, flags2);Thanks for reporting, here is a patch.
We never have NULL vma in pagemap_hugetlb(), I added the BUG_ON check.
Thanks,
Naoya Horiguchi
---
From: Naoya Horiguchi <[email protected]>
Date: Fri, 14 Feb 2014 09:35:06 -0500
Subject: [PATCH] fs/proc/task_mmu.c: assume non-NULL vma in pagemap_hugetlb()
Fengguang reported smatch error about potential NULL pointer access.
In updated page table walker, we never run ->hugetlb_entry() callback on the address without vma. This is because __walk_page_range() checks it in advance. So we can assume non-NULL vma in pagemap_hugetlb().
--- fs/proc/task_mmu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
fs/proc/task_mmu.c +2 -2
@@ -1032,9 +1032,9 @@ static int pagemap_hugetlb(pte_t *pte, unsigned long addr, unsigned long end,pagemap_entry_t pme;unsigned long hmask; -WARN_ON_ONCE (!vma); +BUG_ON (!vma); -if (vma && (vma->vm_flags &VM_SOFTDIRTY )) +if (vma->vm_flags &VM_SOFTDIRTY ) flags2 = __PM_SOFT_DIRTY;else flags2 =0 ;
Signature · 7 lines
1.8.5.3 -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
A message that is not in the archive (sent off-list or not yet received): [email protected]
Andrew Morton
maintainerRe: [PATCH] fs/proc/task_mmu.c: assume non-NULL vma in pagemap_hugetlb() (Re: [mmotm:master 97/220] fs/proc/task_mmu.c:1042 pagemap_hugetlb() error: we previously assumed 'vma' could be null (see line 1037))
Quoted 11 lines · On Fri, 14 Feb 2014 10:09:58 -0500 Naoya Horiguchi <[email protected]>…
Fengguang reported smatch error about potential NULL pointer access.
In updated page table walker, we never run ->hugetlb_entry() callback on the address without vma. This is because
__walk_page_range()checks it in advance. So we can assume non-NULL vma inpagemap_hugetlb()....
fs/proc/task_mmu.c static int pagemap_hugetlb(pte_t *pte, unsigned long addr, unsigned long end, · line 1035
Quoted 5 lines
fs/proc/task_mmu.c +1
-1@@ -1032,9 +1032,9 @@ static int pagemap_hugetlb(pte_t *pte, unsigned long addr, unsigned long end,pagemap_entry_t pme;unsigned long hmask; -WARN_ON_ONCE (!vma); +BUG_ON (!vma);
Let's just remove it altogether.
fs/proc/task_mmu.c static int pagemap_hugetlb(pte_t *pte, unsigned long addr, unsigned long end, · line 1036
Quoted 2 lines
fs/proc/task_mmu.c +1
-1-if (vma && (vma->vm_flags &VM_SOFTDIRTY )) +if (vma->vm_flags &VM_SOFTDIRTY )
The null deref oops will provide us the same info as the BUG_ON. It will require a *little* more thinking to work out that `vma' was NULL, but it will be pretty obvious.
This requires knowing offsetof(vm_area_struct, vm_flags). I use a gdb macro:
define offsetof
set $off = &(((struct $arg0 *)0)->$arg1) printf "%d 0x%x\n", $off, $off
end
akpm3:/usr/src/25> gdb fs/proc/task_mmu.o
...
(gdb) offsetof vm_area_struct vm_flags
80 0x50
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Re: [PATCH] fs/proc/task_mmu.c: assume non-NULL vma in pagemap_hugetlb() (Re: [mmotm:master 97/220] fs/proc/task_mmu.c:1042 pagemap_hugetlb() error: we previously assumed 'vma' could be null (see line 1037))
Quoted 21 lines · Andrew Morton
Quoted 11 lines · On Fri, 14 Feb 2014 10:09:58 -0500 Naoya Horiguchi <[email protected]>…
Fengguang reported smatch error about potential NULL pointer access.
In updated page table walker, we never run ->hugetlb_entry() callback on the address without vma. This is because
__walk_page_range()checks it in advance. So we can assume non-NULL vma inpagemap_hugetlb()....
fs/proc/task_mmu.c static int pagemap_hugetlb(pte_t *pte, unsigned long addr, unsigned long end, · line 1035
Quoted 5 lines
fs/proc/task_mmu.c +1
-1@@ -1032,9 +1032,9 @@ static int pagemap_hugetlb(pte_t *pte, unsigned long addr, unsigned long end,pagemap_entry_t pme;unsigned long hmask; -WARN_ON_ONCE (!vma); +BUG_ON (!vma);Let's just remove it altogether.
Quoted 6 lines
fs/proc/task_mmu.c static int pagemap_hugetlb(pte_t *pte, unsigned long addr, unsigned long end, · line 1036
Quoted 2 lines
fs/proc/task_mmu.c +1
-1-if (vma && (vma->vm_flags &VM_SOFTDIRTY )) +if (vma->vm_flags &VM_SOFTDIRTY )The null deref oops will provide us the same info as the
BUG_ON. It will require a *little* more thinking to work out that `vma' was NULL, but it will be pretty obvious.
Yes, I agree. pagemap_hugetlb() is callback and never inlined, so we will have an info like 'NULL pointer access on pagemap_hugetlb()' with call trace, which is enough to pinpoint the problem.
Quoted 4 lines
This requires knowing
offsetof(vm_area_struct, vm_flags). I use a gdb macro:define offsetof
Quoted 8 lines
set $off = &(((struct $arg0 *)0)->$arg1) printf "%d 0x%x\n", $off, $offend
akpm3:/usr/src/25> gdb
fs/proc/task_mmu.o
...
(gdb) offsetofvm_area_structvm_flags
800x50
Nice note, thank you.
Naoya
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>