2019-01-10 00:52:42

by Sean Christopherson

[permalink] [raw]
Subject: [PATCH] mm/mmu_notifier: mm/rmap.c: Fix a mmu_notifier range bug in try_to_unmap_one

The conversion to use a structure for mmu_notifier_invalidate_range_*()
unintentionally changed the usage in try_to_unmap_one() to init the
'struct mmu_notifier_range' with vma->vm_start instead of @address,
i.e. it invalidates the wrong address range. Revert to the correct
address range.

Manifests as KVM use-after-free WARNINGs and subsequent "BUG: Bad page
state in process X" errors when reclaiming from a KVM guest due to KVM
removing the wrong pages from its own mappings.

Reported-by: [email protected]
Reported-by: Mike Galbraith <[email protected]>
Reported-by: Adam Borowski <[email protected]>
Cc: Jérôme Glisse <[email protected]>
Cc: Christian König <[email protected]>
Cc: Jan Kara <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Cc: Ross Zwisler <[email protected]>
Cc: Dan Williams <[email protected]>
Cc: Paolo Bonzini <[email protected]>
Cc: Radim Krčmář <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Felix Kuehling <[email protected]>
Cc: Ralph Campbell <[email protected]>
Cc: John Hubbard <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Linus Torvalds <[email protected]>
Fixes: ac46d4f3c432 ("mm/mmu_notifier: use structure for invalidate_range_start/end calls v2")
Signed-off-by: Sean Christopherson <[email protected]>
---

FWIW, I looked through all other calls to mmu_notifier_range_init() in
the patch and didn't spot any other unintentional functional changes.

mm/rmap.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mm/rmap.c b/mm/rmap.c
index 68a1a5b869a5..0454ecc29537 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -1371,8 +1371,8 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
* Note that the page can not be free in this function as call of
* try_to_unmap() must hold a reference on the page.
*/
- mmu_notifier_range_init(&range, vma->vm_mm, vma->vm_start,
- min(vma->vm_end, vma->vm_start +
+ mmu_notifier_range_init(&range, vma->vm_mm, address,
+ min(vma->vm_end, address +
(PAGE_SIZE << compound_order(page))));
if (PageHuge(page)) {
/*
--
2.19.2



2019-01-10 01:44:08

by Jerome Glisse

[permalink] [raw]
Subject: Re: [PATCH] mm/mmu_notifier: mm/rmap.c: Fix a mmu_notifier range bug in try_to_unmap_one

On Wed, Jan 09, 2019 at 04:51:17PM -0800, Sean Christopherson wrote:
> The conversion to use a structure for mmu_notifier_invalidate_range_*()
> unintentionally changed the usage in try_to_unmap_one() to init the
> 'struct mmu_notifier_range' with vma->vm_start instead of @address,
> i.e. it invalidates the wrong address range. Revert to the correct
> address range.
>
> Manifests as KVM use-after-free WARNINGs and subsequent "BUG: Bad page
> state in process X" errors when reclaiming from a KVM guest due to KVM
> removing the wrong pages from its own mappings.
>
> Reported-by: [email protected]
> Reported-by: Mike Galbraith <[email protected]>
> Reported-by: Adam Borowski <[email protected]>

Reviewed-by: Jérôme Glisse <[email protected]>

> Cc: Christian König <[email protected]>
> Cc: Jan Kara <[email protected]>
> Cc: Matthew Wilcox <[email protected]>
> Cc: Ross Zwisler <[email protected]>
> Cc: Dan Williams <[email protected]>
> Cc: Paolo Bonzini <[email protected]>
> Cc: Radim Krčmář <[email protected]>
> Cc: Michal Hocko <[email protected]>
> Cc: Felix Kuehling <[email protected]>
> Cc: Ralph Campbell <[email protected]>
> Cc: John Hubbard <[email protected]>
> Cc: Andrew Morton <[email protected]>
> Cc: Linus Torvalds <[email protected]>
> Fixes: ac46d4f3c432 ("mm/mmu_notifier: use structure for invalidate_range_start/end calls v2")
> Signed-off-by: Sean Christopherson <[email protected]>
> ---
>
> FWIW, I looked through all other calls to mmu_notifier_range_init() in
> the patch and didn't spot any other unintentional functional changes.
>
> mm/rmap.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/mm/rmap.c b/mm/rmap.c
> index 68a1a5b869a5..0454ecc29537 100644
> --- a/mm/rmap.c
> +++ b/mm/rmap.c
> @@ -1371,8 +1371,8 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
> * Note that the page can not be free in this function as call of
> * try_to_unmap() must hold a reference on the page.
> */
> - mmu_notifier_range_init(&range, vma->vm_mm, vma->vm_start,
> - min(vma->vm_end, vma->vm_start +
> + mmu_notifier_range_init(&range, vma->vm_mm, address,
> + min(vma->vm_end, address +
> (PAGE_SIZE << compound_order(page))));
> if (PageHuge(page)) {
> /*
> --
> 2.19.2
>

2019-01-10 05:55:22

by Pankaj Gupta

[permalink] [raw]
Subject: Re: [PATCH] mm/mmu_notifier: mm/rmap.c: Fix a mmu_notifier range bug in try_to_unmap_one


> The conversion to use a structure for mmu_notifier_invalidate_range_*()
> unintentionally changed the usage in try_to_unmap_one() to init the
> 'struct mmu_notifier_range' with vma->vm_start instead of @address,
> i.e. it invalidates the wrong address range. Revert to the correct
> address range.
>
> Manifests as KVM use-after-free WARNINGs and subsequent "BUG: Bad page
> state in process X" errors when reclaiming from a KVM guest due to KVM
> removing the wrong pages from its own mappings.
>
> Reported-by: [email protected]
> Reported-by: Mike Galbraith <[email protected]>
> Reported-by: Adam Borowski <[email protected]>
> Cc: Jérôme Glisse <[email protected]>
> Cc: Christian König <[email protected]>
> Cc: Jan Kara <[email protected]>
> Cc: Matthew Wilcox <[email protected]>
> Cc: Ross Zwisler <[email protected]>
> Cc: Dan Williams <[email protected]>
> Cc: Paolo Bonzini <[email protected]>
> Cc: Radim Krčmář <[email protected]>
> Cc: Michal Hocko <[email protected]>
> Cc: Felix Kuehling <[email protected]>
> Cc: Ralph Campbell <[email protected]>
> Cc: John Hubbard <[email protected]>
> Cc: Andrew Morton <[email protected]>
> Cc: Linus Torvalds <[email protected]>
> Fixes: ac46d4f3c432 ("mm/mmu_notifier: use structure for
> invalidate_range_start/end calls v2")
> Signed-off-by: Sean Christopherson <[email protected]>
> ---
>
> FWIW, I looked through all other calls to mmu_notifier_range_init() in
> the patch and didn't spot any other unintentional functional changes.
>
> mm/rmap.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/mm/rmap.c b/mm/rmap.c
> index 68a1a5b869a5..0454ecc29537 100644
> --- a/mm/rmap.c
> +++ b/mm/rmap.c
> @@ -1371,8 +1371,8 @@ static bool try_to_unmap_one(struct page *page, struct
> vm_area_struct *vma,
> * Note that the page can not be free in this function as call of
> * try_to_unmap() must hold a reference on the page.
> */
> - mmu_notifier_range_init(&range, vma->vm_mm, vma->vm_start,
> - min(vma->vm_end, vma->vm_start +
> + mmu_notifier_range_init(&range, vma->vm_mm, address,
> + min(vma->vm_end, address +
> (PAGE_SIZE << compound_order(page))));
> if (PageHuge(page)) {
> /*
> --

I was suspecting this patch for some other issue. But could not spot this after
in depth analyzing the changed "invalidate_range_start/end calls".

Its indeed a good catch.

Reviewed-by: Pankaj gupta <[email protected]>

> 2.19.2
>
>

2019-01-10 10:48:44

by Adam Borowski

[permalink] [raw]
Subject: Re: [PATCH] mm/mmu_notifier: mm/rmap.c: Fix a mmu_notifier range bug in try_to_unmap_one

On Wed, Jan 09, 2019 at 04:51:17PM -0800, Sean Christopherson wrote:
> Manifests as KVM use-after-free WARNINGs and subsequent "BUG: Bad page
> state in process X" errors when reclaiming from a KVM guest due to KVM
> removing the wrong pages from its own mappings.

With your patch, no badness happened so far. Thanks!

> Reported-by: Adam Borowski <[email protected]>
> Fixes: ac46d4f3c432 ("mm/mmu_notifier: use structure for invalidate_range_start/end calls v2")

> --- a/mm/rmap.c
> +++ b/mm/rmap.c
> - mmu_notifier_range_init(&range, vma->vm_mm, vma->vm_start,
> - min(vma->vm_end, vma->vm_start +
> + mmu_notifier_range_init(&range, vma->vm_mm, address,
> + min(vma->vm_end, address +


Meow.
--
⢀⣴⠾⠻⢶⣦⠀ Hans 1 was born and raised in Johannesburg, then moved to Boston,
⣾⠁⢠⠒⠀⣿⡁ and has just became a naturalized citizen. Hans 2's grandparents
⢿⡄⠘⠷⠚⠋⠀ came from Melanesia to Düsseldorf, and he hasn't ever been outside
⠈⠳⣄⠀⠀⠀⠀ Germany until yesterday. Which one is an African-American?