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

Groups > linux.kernel > #1564940 > unrolled thread

[PATCH] mm: ensure alloc_flags in slow path are initialized

Back to article view | Back to linux.kernel

Contents

  [PATCH] mm: ensure alloc_flags in slow path are initialized Arnd Bergmann <arnd@arndb.de> - 2017-01-23 13:20 +0100
    Re: [PATCH] mm: ensure alloc_flags in slow path are initialized Vlastimil Babka <vbabka@suse.cz> - 2017-01-23 14:00 +0100
      Re: [PATCH] mm: ensure alloc_flags in slow path are initialized Arnd Bergmann <arnd@arndb.de> - 2017-01-23 17:10 +0100
    Re: [PATCH] mm: ensure alloc_flags in slow path are initialized David Rientjes <rientjes@google.com> - 2017-01-24 00:00 +0100
    Re: [PATCH] mm: ensure alloc_flags in slow path are initialized Andrew Morton <akpm@linux-foundation.org> - 2017-01-24 01:00 +0100
      Re: [PATCH] mm: ensure alloc_flags in slow path are initialized Vlastimil Babka <vbabka@suse.cz> - 2017-01-24 10:20 +0100

#1564940 — [PATCH] mm: ensure alloc_flags in slow path are initialized

The __alloc_pages_slowpath() has gotten rather complex and gcc
is no longer able to follow the gotos and prove that the
alloc_flags variable is initialized at the time it is used:

mm/page_alloc.c: In function '__alloc_pages_slowpath':
mm/page_alloc.c:3565:15: error: 'alloc_flags' may be used uninitialized in this function [-Werror=maybe-uninitialized]

To be honest, I can't figure that out either, maybe it is or
maybe not, but moving the existing initialization up a little
higher looks safe and makes it obvious to both me and gcc that
the initialization comes before the first use.

Fixes: 74eaa4a97e8e ("mm: consolidate GFP_NOFAIL checks in the allocator slowpath")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 mm/page_alloc.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index cf641932c015..d9fa4564524f 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3591,6 +3591,13 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
 				(__GFP_ATOMIC|__GFP_DIRECT_RECLAIM)))
 		gfp_mask &= ~__GFP_ATOMIC;
 
+	/*
+	 * The fast path uses conservative alloc_flags to succeed only until
+	 * kswapd needs to be woken up, and to avoid the cost of setting up
+	 * alloc_flags precisely. So we do that now.
+	 */
+	alloc_flags = gfp_to_alloc_flags(gfp_mask);
+
 retry_cpuset:
 	compaction_retries = 0;
 	no_progress_loops = 0;
@@ -3607,14 +3614,6 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
 	if (!ac->preferred_zoneref->zone)
 		goto nopage;
 
-
-	/*
-	 * The fast path uses conservative alloc_flags to succeed only until
-	 * kswapd needs to be woken up, and to avoid the cost of setting up
-	 * alloc_flags precisely. So we do that now.
-	 */
-	alloc_flags = gfp_to_alloc_flags(gfp_mask);
-
 	if (gfp_mask & __GFP_KSWAPD_RECLAIM)
 		wake_all_kswapds(order, ac);
 
-- 
2.9.0

[toc] | [next] | [standalone]

#1564963

On 01/23/2017 01:16 PM, Arnd Bergmann wrote:
> The __alloc_pages_slowpath() has gotten rather complex and gcc
> is no longer able to follow the gotos and prove that the
> alloc_flags variable is initialized at the time it is used:
>
> mm/page_alloc.c: In function '__alloc_pages_slowpath':
> mm/page_alloc.c:3565:15: error: 'alloc_flags' may be used uninitialized in this function [-Werror=maybe-uninitialized]
>
> To be honest, I can't figure that out either, maybe it is or
> maybe not,

Seems the report is correct and not false positive, in scenario when we goto 
nopage before the assignment, and then goto retry because of __GFP_NOFAIL.

> but moving the existing initialization up a little
> higher looks safe and makes it obvious to both me and gcc that
> the initialization comes before the first use.
>
> Fixes: 74eaa4a97e8e ("mm: consolidate GFP_NOFAIL checks in the allocator slowpath")

That's a non-stable -next commit ID for mmotm patch:
mm-consolidate-gfp_nofail-checks-in-the-allocator-slowpath.patch

The patch itself was OK, the problem only comes from integration with another 
mmotm patch (also independently OK):
mm-page_alloc-fix-premature-oom-when-racing-with-cpuset-mems-update.patch

By their ordering in mmotm, it would work to treat this as a fix for the 
GFP_NOFAIL patch, possibly merged into it.

> Signed-off-by: Arnd Bergmann <arnd@arndb.de>

Acked-by: Vlastimil Babka <vbabka@suse.cz>

> ---
>  mm/page_alloc.c | 15 +++++++--------
>  1 file changed, 7 insertions(+), 8 deletions(-)
>
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index cf641932c015..d9fa4564524f 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -3591,6 +3591,13 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
>  				(__GFP_ATOMIC|__GFP_DIRECT_RECLAIM)))
>  		gfp_mask &= ~__GFP_ATOMIC;
>
> +	/*
> +	 * The fast path uses conservative alloc_flags to succeed only until
> +	 * kswapd needs to be woken up, and to avoid the cost of setting up
> +	 * alloc_flags precisely. So we do that now.
> +	 */
> +	alloc_flags = gfp_to_alloc_flags(gfp_mask);
> +
>  retry_cpuset:
>  	compaction_retries = 0;
>  	no_progress_loops = 0;
> @@ -3607,14 +3614,6 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
>  	if (!ac->preferred_zoneref->zone)
>  		goto nopage;
>
> -
> -	/*
> -	 * The fast path uses conservative alloc_flags to succeed only until
> -	 * kswapd needs to be woken up, and to avoid the cost of setting up
> -	 * alloc_flags precisely. So we do that now.
> -	 */
> -	alloc_flags = gfp_to_alloc_flags(gfp_mask);
> -
>  	if (gfp_mask & __GFP_KSWAPD_RECLAIM)
>  		wake_all_kswapds(order, ac);
>
>

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

#1565088

On Mon, Jan 23, 2017 at 1:55 PM, Vlastimil Babka <vbabka@suse.cz> wrote:
> On 01/23/2017 01:16 PM, Arnd Bergmann wrote:

>> To be honest, I can't figure that out either, maybe it is or
>> maybe not,
>
>
> Seems the report is correct and not false positive, in scenario when we goto
> nopage before the assignment, and then goto retry because of __GFP_NOFAIL.

Ok, thanks for checking!

>> but moving the existing initialization up a little
>> higher looks safe and makes it obvious to both me and gcc that
>> the initialization comes before the first use.
>>
>> Fixes: 74eaa4a97e8e ("mm: consolidate GFP_NOFAIL checks in the allocator
>> slowpath")
>
>
> That's a non-stable -next commit ID for mmotm patch:
> mm-consolidate-gfp_nofail-checks-in-the-allocator-slowpath.patch
>
> The patch itself was OK, the problem only comes from integration with
> another mmotm patch (also independently OK):
> mm-page_alloc-fix-premature-oom-when-racing-with-cpuset-mems-update.patch
>
> By their ordering in mmotm, it would work to treat this as a fix for the
> GFP_NOFAIL patch, possibly merged into it.

Ok. I only tracked down which commit introduced the warning, which was
the one above.

    Arnd

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

#1565350

On Mon, 23 Jan 2017, Arnd Bergmann wrote:

> The __alloc_pages_slowpath() has gotten rather complex and gcc
> is no longer able to follow the gotos and prove that the
> alloc_flags variable is initialized at the time it is used:
> 
> mm/page_alloc.c: In function '__alloc_pages_slowpath':
> mm/page_alloc.c:3565:15: error: 'alloc_flags' may be used uninitialized in this function [-Werror=maybe-uninitialized]
> 
> To be honest, I can't figure that out either, maybe it is or
> maybe not, but moving the existing initialization up a little
> higher looks safe and makes it obvious to both me and gcc that
> the initialization comes before the first use.
> 
> Fixes: 74eaa4a97e8e ("mm: consolidate GFP_NOFAIL checks in the allocator slowpath")
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>

Acked-by: David Rientjes <rientjes@google.com>

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

#1565374

On Mon, 23 Jan 2017 13:16:12 +0100 Arnd Bergmann <arnd@arndb.de> wrote:

> The __alloc_pages_slowpath() has gotten rather complex and gcc
> is no longer able to follow the gotos and prove that the
> alloc_flags variable is initialized at the time it is used:
> 
> mm/page_alloc.c: In function '__alloc_pages_slowpath':
> mm/page_alloc.c:3565:15: error: 'alloc_flags' may be used uninitialized in this function [-Werror=maybe-uninitialized]
> 
> To be honest, I can't figure that out either, maybe it is or
> maybe not, but moving the existing initialization up a little
> higher looks safe and makes it obvious to both me and gcc that
> the initialization comes before the first use.
> 
> ...
>
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -3591,6 +3591,13 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
>  				(__GFP_ATOMIC|__GFP_DIRECT_RECLAIM)))
>  		gfp_mask &= ~__GFP_ATOMIC;
>  
> +	/*
> +	 * The fast path uses conservative alloc_flags to succeed only until
> +	 * kswapd needs to be woken up, and to avoid the cost of setting up
> +	 * alloc_flags precisely. So we do that now.
> +	 */
> +	alloc_flags = gfp_to_alloc_flags(gfp_mask);
> +
>  retry_cpuset:
>  	compaction_retries = 0;
>  	no_progress_loops = 0;
> @@ -3607,14 +3614,6 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
>  	if (!ac->preferred_zoneref->zone)
>  		goto nopage;
>  
> -
> -	/*
> -	 * The fast path uses conservative alloc_flags to succeed only until
> -	 * kswapd needs to be woken up, and to avoid the cost of setting up
> -	 * alloc_flags precisely. So we do that now.
> -	 */
> -	alloc_flags = gfp_to_alloc_flags(gfp_mask);
> -
>  	if (gfp_mask & __GFP_KSWAPD_RECLAIM)
>  		wake_all_kswapds(order, ac);

hm.  But we later do

	if (gfp_pfmemalloc_allowed(gfp_mask))
		alloc_flags = ALLOC_NO_WATERMARKS;

	...
	if (read_mems_allowed_retry(cpuset_mems_cookie))
		goto retry_cpuset;

so with your patch there's a path where we can rerun everything with
alloc_flags == ALLOC_NO_WATERMARKS.  That's changed behaviour.

When I saw the test robot warning I did this, which I think preserves
behaviour?

--- a/mm/page_alloc.c~mm-consolidate-gfp_nofail-checks-in-the-allocator-slowpath-fix
+++ a/mm/page_alloc.c
@@ -3577,6 +3577,14 @@ retry_cpuset:
 	no_progress_loops = 0;
 	compact_priority = DEF_COMPACT_PRIORITY;
 	cpuset_mems_cookie = read_mems_allowed_begin();
+
+	/*
+	 * The fast path uses conservative alloc_flags to succeed only until
+	 * kswapd needs to be woken up, and to avoid the cost of setting up
+	 * alloc_flags precisely. So we do that now.
+	 */
+	alloc_flags = gfp_to_alloc_flags(gfp_mask);
+
 	/*
 	 * We need to recalculate the starting point for the zonelist iterator
 	 * because we might have used different nodemask in the fast path, or
@@ -3588,14 +3596,6 @@ retry_cpuset:
 	if (!ac->preferred_zoneref->zone)
 		goto nopage;
 
-
-	/*
-	 * The fast path uses conservative alloc_flags to succeed only until
-	 * kswapd needs to be woken up, and to avoid the cost of setting up
-	 * alloc_flags precisely. So we do that now.
-	 */
-	alloc_flags = gfp_to_alloc_flags(gfp_mask);
-
 	if (gfp_mask & __GFP_KSWAPD_RECLAIM)
 		wake_all_kswapds(order, ac);
 
_

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

#1565737

On 01/24/2017 12:56 AM, Andrew Morton wrote:
> On Mon, 23 Jan 2017 13:16:12 +0100 Arnd Bergmann <arnd@arndb.de> wrote:
> 
>> The __alloc_pages_slowpath() has gotten rather complex and gcc
>> is no longer able to follow the gotos and prove that the
>> alloc_flags variable is initialized at the time it is used:
>>
>> mm/page_alloc.c: In function '__alloc_pages_slowpath':
>> mm/page_alloc.c:3565:15: error: 'alloc_flags' may be used uninitialized in this function [-Werror=maybe-uninitialized]
>>
>> To be honest, I can't figure that out either, maybe it is or
>> maybe not, but moving the existing initialization up a little
>> higher looks safe and makes it obvious to both me and gcc that
>> the initialization comes before the first use.
>>
>> ...
>>
>> --- a/mm/page_alloc.c
>> +++ b/mm/page_alloc.c
>> @@ -3591,6 +3591,13 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
>>  				(__GFP_ATOMIC|__GFP_DIRECT_RECLAIM)))
>>  		gfp_mask &= ~__GFP_ATOMIC;
>>  
>> +	/*
>> +	 * The fast path uses conservative alloc_flags to succeed only until
>> +	 * kswapd needs to be woken up, and to avoid the cost of setting up
>> +	 * alloc_flags precisely. So we do that now.
>> +	 */
>> +	alloc_flags = gfp_to_alloc_flags(gfp_mask);
>> +
>>  retry_cpuset:
>>  	compaction_retries = 0;
>>  	no_progress_loops = 0;
>> @@ -3607,14 +3614,6 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
>>  	if (!ac->preferred_zoneref->zone)
>>  		goto nopage;
>>  
>> -
>> -	/*
>> -	 * The fast path uses conservative alloc_flags to succeed only until
>> -	 * kswapd needs to be woken up, and to avoid the cost of setting up
>> -	 * alloc_flags precisely. So we do that now.
>> -	 */
>> -	alloc_flags = gfp_to_alloc_flags(gfp_mask);
>> -
>>  	if (gfp_mask & __GFP_KSWAPD_RECLAIM)
>>  		wake_all_kswapds(order, ac);
> 
> hm.  But we later do
> 
> 	if (gfp_pfmemalloc_allowed(gfp_mask))
> 		alloc_flags = ALLOC_NO_WATERMARKS;
> 
> 	...
> 	if (read_mems_allowed_retry(cpuset_mems_cookie))
> 		goto retry_cpuset;
> 
> so with your patch there's a path where we can rerun everything with
> alloc_flags == ALLOC_NO_WATERMARKS.  That's changed behaviour.

Right.

> When I saw the test robot warning I did this, which I think preserves
> behaviour?

Yes, that's cleaner. Thanks.

> --- a/mm/page_alloc.c~mm-consolidate-gfp_nofail-checks-in-the-allocator-slowpath-fix
> +++ a/mm/page_alloc.c
> @@ -3577,6 +3577,14 @@ retry_cpuset:
>  	no_progress_loops = 0;
>  	compact_priority = DEF_COMPACT_PRIORITY;
>  	cpuset_mems_cookie = read_mems_allowed_begin();
> +
> +	/*
> +	 * The fast path uses conservative alloc_flags to succeed only until
> +	 * kswapd needs to be woken up, and to avoid the cost of setting up
> +	 * alloc_flags precisely. So we do that now.
> +	 */
> +	alloc_flags = gfp_to_alloc_flags(gfp_mask);
> +
>  	/*
>  	 * We need to recalculate the starting point for the zonelist iterator
>  	 * because we might have used different nodemask in the fast path, or
> @@ -3588,14 +3596,6 @@ retry_cpuset:
>  	if (!ac->preferred_zoneref->zone)
>  		goto nopage;
>  
> -
> -	/*
> -	 * The fast path uses conservative alloc_flags to succeed only until
> -	 * kswapd needs to be woken up, and to avoid the cost of setting up
> -	 * alloc_flags precisely. So we do that now.
> -	 */
> -	alloc_flags = gfp_to_alloc_flags(gfp_mask);
> -
>  	if (gfp_mask & __GFP_KSWAPD_RECLAIM)
>  		wake_all_kswapds(order, ac);
>  
> _
> 

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web