Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1508051 > unrolled thread

[PATCH stable 4.4 0/4] mm: workingset backports

Back to article view | Back to linux.kernel

Contents

  [PATCH stable 4.4 0/4] mm: workingset backports Michal Hocko <mhocko@kernel.org> - 2016-10-25 10:00 +0200
    [PATCH stable 4.4 2/4] mm: filemap: don't plant shadow entries without radix tree node Michal Hocko <mhocko@kernel.org> - 2016-10-25 10:00 +0200
      Re: [PATCH stable 4.4 2/4] mm: filemap: don't plant shadow entries  without radix tree node Michal Hocko <mhocko@kernel.org> - 2016-10-26 14:50 +0200
        Re: [PATCH stable 4.4 2/4] mm: filemap: don't plant shadow entries  without radix tree node Michal Hocko <mhocko@kernel.org> - 2016-10-26 14:50 +0200
          Re: [PATCH stable 4.4 2/4] mm: filemap: don't plant shadow entries  without radix tree node Greg KH <gregkh@linuxfoundation.org> - 2016-10-26 15:30 +0200
    Re: [PATCH stable 4.4 0/4] mm: workingset backports Michal Hocko <mhocko@kernel.org> - 2016-10-25 10:00 +0200
    Re: [PATCH stable 4.4 0/4] mm: workingset backports Johannes Weiner <hannes@cmpxchg.org> - 2016-10-25 16:20 +0200
    Re: [PATCH stable 4.4 0/4] mm: workingset backports Greg KH <greg@kroah.com> - 2016-10-26 11:40 +0200

#1508051 — [PATCH stable 4.4 0/4] mm: workingset backports

Hi,
here is the backport of (hopefully) all workingset related fixes for
4.4 kernel. The series has been reviewed by Johannes [1]. The main
motivation for the backport is 22f2ac51b6d6 ("mm: workingset: fix crash
in shadow node shrinker caused by replace_page_cache_page()") which is
a fix for a triggered BUG_ON. This is not sufficient because there are
follow up fixes which were introduced later.

[toc] | [next] | [standalone]

#1508053 — [PATCH stable 4.4 2/4] mm: filemap: don't plant shadow entries without radix tree node

From: Johannes Weiner <hannes@cmpxchg.org>

commit d3798ae8c6f3767c726403c2ca6ecc317752c9dd upstream.

When the underflow checks were added to workingset_node_shadow_dec(),
they triggered immediately:

  kernel BUG at ./include/linux/swap.h:276!
  invalid opcode: 0000 [#1] SMP
  Modules linked in: isofs usb_storage fuse xt_CHECKSUM ipt_MASQUERADE nf_nat_masquerade_ipv4 tun nf_conntrack_netbios_ns nf_conntrack_broadcast ip6t_REJECT nf_reject_ipv6
   soundcore wmi acpi_als pinctrl_sunrisepoint kfifo_buf tpm_tis industrialio acpi_pad pinctrl_intel tpm_tis_core tpm nfsd auth_rpcgss nfs_acl lockd grace sunrpc dm_crypt
  CPU: 0 PID: 20929 Comm: blkid Not tainted 4.8.0-rc8-00087-gbe67d60ba944 #1
  Hardware name: System manufacturer System Product Name/Z170-K, BIOS 1803 05/06/2016
  task: ffff8faa93ecd940 task.stack: ffff8faa7f478000
  RIP: page_cache_tree_insert+0xf1/0x100
  Call Trace:
    __add_to_page_cache_locked+0x12e/0x270
    add_to_page_cache_lru+0x4e/0xe0
    mpage_readpages+0x112/0x1d0
    blkdev_readpages+0x1d/0x20
    __do_page_cache_readahead+0x1ad/0x290
    force_page_cache_readahead+0xaa/0x100
    page_cache_sync_readahead+0x3f/0x50
    generic_file_read_iter+0x5af/0x740
    blkdev_read_iter+0x35/0x40
    __vfs_read+0xe1/0x130
    vfs_read+0x96/0x130
    SyS_read+0x55/0xc0
    entry_SYSCALL_64_fastpath+0x13/0x8f
  Code: 03 00 48 8b 5d d8 65 48 33 1c 25 28 00 00 00 44 89 e8 75 19 48 83 c4 18 5b 41 5c 41 5d 41 5e 5d c3 0f 0b 41 bd ef ff ff ff eb d7 <0f> 0b e8 88 68 ef ff 0f 1f 84 00
  RIP  page_cache_tree_insert+0xf1/0x100

This is a long-standing bug in the way shadow entries are accounted in
the radix tree nodes. The shrinker needs to know when radix tree nodes
contain only shadow entries, no pages, so node->count is split in half
to count shadows in the upper bits and pages in the lower bits.

Unfortunately, the radix tree implementation doesn't know of this and
assumes all entries are in node->count. When there is a shadow entry
directly in root->rnode and the tree is later extended, the radix tree
implementation will copy that entry into the new node and and bump its
node->count, i.e. increases the page count bits. Once the shadow gets
removed and we subtract from the upper counter, node->count underflows
and triggers the warning. Afterwards, without node->count reaching 0
again, the radix tree node is leaked.

Limit shadow entries to when we have actual radix tree nodes and can
count them properly. That means we lose the ability to detect refaults
from files that had only the first page faulted in at eviction time.

[hannes@cmpxchg.org: backport for 4.4 stable]
Fixes: 449dd6984d0e ("mm: keep page cache radix tree nodes in check")
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reported-and-tested-by: Linus Torvalds <torvalds@linux-foundation.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Michal Hocko <mhocko@suse.com>
---
 mm/filemap.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/mm/filemap.c b/mm/filemap.c
index 4cfe423d3e8a..7ad648c9780c 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -164,6 +164,14 @@ static void page_cache_tree_delete(struct address_space *mapping,
 
 	__radix_tree_lookup(&mapping->page_tree, page->index, &node, &slot);
 
+	if (!node) {
+		/*
+		 * We need a node to properly account shadow
+		 * entries. Don't plant any without. XXX
+		 */
+		shadow = NULL;
+	}
+
 	if (shadow) {
 		mapping->nrshadows++;
 		/*
-- 
2.9.3

[toc] | [prev] | [next] | [standalone]

#1509266 — Re: [PATCH stable 4.4 2/4] mm: filemap: don't plant shadow entries without radix tree node

Greg,
I do not see this one in the 4.4 queue you have just sent today.

On Tue 25-10-16 09:51:46, Michal Hocko wrote:
> From: Johannes Weiner <hannes@cmpxchg.org>
> 
> commit d3798ae8c6f3767c726403c2ca6ecc317752c9dd upstream.
> 
> When the underflow checks were added to workingset_node_shadow_dec(),
> they triggered immediately:
> 
>   kernel BUG at ./include/linux/swap.h:276!
>   invalid opcode: 0000 [#1] SMP
>   Modules linked in: isofs usb_storage fuse xt_CHECKSUM ipt_MASQUERADE nf_nat_masquerade_ipv4 tun nf_conntrack_netbios_ns nf_conntrack_broadcast ip6t_REJECT nf_reject_ipv6
>    soundcore wmi acpi_als pinctrl_sunrisepoint kfifo_buf tpm_tis industrialio acpi_pad pinctrl_intel tpm_tis_core tpm nfsd auth_rpcgss nfs_acl lockd grace sunrpc dm_crypt
>   CPU: 0 PID: 20929 Comm: blkid Not tainted 4.8.0-rc8-00087-gbe67d60ba944 #1
>   Hardware name: System manufacturer System Product Name/Z170-K, BIOS 1803 05/06/2016
>   task: ffff8faa93ecd940 task.stack: ffff8faa7f478000
>   RIP: page_cache_tree_insert+0xf1/0x100
>   Call Trace:
>     __add_to_page_cache_locked+0x12e/0x270
>     add_to_page_cache_lru+0x4e/0xe0
>     mpage_readpages+0x112/0x1d0
>     blkdev_readpages+0x1d/0x20
>     __do_page_cache_readahead+0x1ad/0x290
>     force_page_cache_readahead+0xaa/0x100
>     page_cache_sync_readahead+0x3f/0x50
>     generic_file_read_iter+0x5af/0x740
>     blkdev_read_iter+0x35/0x40
>     __vfs_read+0xe1/0x130
>     vfs_read+0x96/0x130
>     SyS_read+0x55/0xc0
>     entry_SYSCALL_64_fastpath+0x13/0x8f
>   Code: 03 00 48 8b 5d d8 65 48 33 1c 25 28 00 00 00 44 89 e8 75 19 48 83 c4 18 5b 41 5c 41 5d 41 5e 5d c3 0f 0b 41 bd ef ff ff ff eb d7 <0f> 0b e8 88 68 ef ff 0f 1f 84 00
>   RIP  page_cache_tree_insert+0xf1/0x100
> 
> This is a long-standing bug in the way shadow entries are accounted in
> the radix tree nodes. The shrinker needs to know when radix tree nodes
> contain only shadow entries, no pages, so node->count is split in half
> to count shadows in the upper bits and pages in the lower bits.
> 
> Unfortunately, the radix tree implementation doesn't know of this and
> assumes all entries are in node->count. When there is a shadow entry
> directly in root->rnode and the tree is later extended, the radix tree
> implementation will copy that entry into the new node and and bump its
> node->count, i.e. increases the page count bits. Once the shadow gets
> removed and we subtract from the upper counter, node->count underflows
> and triggers the warning. Afterwards, without node->count reaching 0
> again, the radix tree node is leaked.
> 
> Limit shadow entries to when we have actual radix tree nodes and can
> count them properly. That means we lose the ability to detect refaults
> from files that had only the first page faulted in at eviction time.
> 
> [hannes@cmpxchg.org: backport for 4.4 stable]
> Fixes: 449dd6984d0e ("mm: keep page cache radix tree nodes in check")
> Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
> Reported-and-tested-by: Linus Torvalds <torvalds@linux-foundation.org>
> Reviewed-by: Jan Kara <jack@suse.cz>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
> Signed-off-by: Michal Hocko <mhocko@suse.com>
> ---
>  mm/filemap.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/mm/filemap.c b/mm/filemap.c
> index 4cfe423d3e8a..7ad648c9780c 100644
> --- a/mm/filemap.c
> +++ b/mm/filemap.c
> @@ -164,6 +164,14 @@ static void page_cache_tree_delete(struct address_space *mapping,
>  
>  	__radix_tree_lookup(&mapping->page_tree, page->index, &node, &slot);
>  
> +	if (!node) {
> +		/*
> +		 * We need a node to properly account shadow
> +		 * entries. Don't plant any without. XXX
> +		 */
> +		shadow = NULL;
> +	}
> +
>  	if (shadow) {
>  		mapping->nrshadows++;
>  		/*
> -- 
> 2.9.3
> 

-- 
Michal Hocko
SUSE Labs

[toc] | [prev] | [next] | [standalone]

#1509272 — Re: [PATCH stable 4.4 2/4] mm: filemap: don't plant shadow entries without radix tree node

On Wed 26-10-16 14:45:53, Michal Hocko wrote:
> Greg,
> I do not see this one in the 4.4 queue you have just sent today.

Scratch that. I can see it now on lkml. I just wasn't on the CC so it
hasn't shown up in my inbox.
-- 
Michal Hocko
SUSE Labs

[toc] | [prev] | [next] | [standalone]

#1509400 — Re: [PATCH stable 4.4 2/4] mm: filemap: don't plant shadow entries without radix tree node

On Wed, Oct 26, 2016 at 02:47:53PM +0200, Michal Hocko wrote:
> On Wed 26-10-16 14:45:53, Michal Hocko wrote:
> > Greg,
> > I do not see this one in the 4.4 queue you have just sent today.
> 
> Scratch that. I can see it now on lkml. I just wasn't on the CC so it
> hasn't shown up in my inbox.

Sorry about that, I had applied it earlier in the sequence due to it
being part of the "normal" stable request process.

thanks,

greg k-h

[toc] | [prev] | [next] | [standalone]

#1508055

On Tue 25-10-16 09:51:44, Michal Hocko wrote:
> Hi,
> here is the backport of (hopefully) all workingset related fixes for
> 4.4 kernel. The series has been reviewed by Johannes [1]. The main
> motivation for the backport is 22f2ac51b6d6 ("mm: workingset: fix crash
> in shadow node shrinker caused by replace_page_cache_page()") which is
> a fix for a triggered BUG_ON. This is not sufficient because there are
> follow up fixes which were introduced later.
 
 Ups, forgot to add
[1] http://lkml.kernel.org/r/20161024152605.11707-1-mhocko@kernel.org

-- 
Michal Hocko
SUSE Labs

[toc] | [prev] | [next] | [standalone]

#1508322

All 4 backport patches:

Acked-by: Johannes Weiner <hannes@cmpxchg.org>

Thank you Michal.

[toc] | [prev] | [next] | [standalone]

#1509002

On Tue, Oct 25, 2016 at 09:51:44AM +0200, Michal Hocko wrote:
> Hi,
> here is the backport of (hopefully) all workingset related fixes for
> 4.4 kernel. The series has been reviewed by Johannes [1]. The main
> motivation for the backport is 22f2ac51b6d6 ("mm: workingset: fix crash
> in shadow node shrinker caused by replace_page_cache_page()") which is
> a fix for a triggered BUG_ON. This is not sufficient because there are
> follow up fixes which were introduced later.

Thanks for these, all now queued up.

greg k-h

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web