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


Groups > linux.kernel > #1684787 > unrolled thread

[PATCH] tracing/ring_buffer: Try harder to allocate

Started byJoel Fernandes <joelaf@google.com>
First post2017-07-11 08:10 +0200
Last post2017-07-11 19:30 +0200
Articles 4 — 4 participants

Back to article view | Back to linux.kernel


Contents

  [PATCH] tracing/ring_buffer: Try harder to allocate Joel Fernandes <joelaf@google.com> - 2017-07-11 08:10 +0200
    Re: [PATCH] tracing/ring_buffer: Try harder to allocate Michal Hocko <mhocko@kernel.org> - 2017-07-11 08:20 +0200
    Re: [PATCH] tracing/ring_buffer: Try harder to allocate Vlastimil Babka <vbabka@suse.cz> - 2017-07-11 08:20 +0200
    Re: [PATCH] tracing/ring_buffer: Try harder to allocate Johannes Weiner <hannes@cmpxchg.org> - 2017-07-11 19:30 +0200

#1684787 — [PATCH] tracing/ring_buffer: Try harder to allocate

FromJoel Fernandes <joelaf@google.com>
Date2017-07-11 08:10 +0200
Subject[PATCH] tracing/ring_buffer: Try harder to allocate
Message-ID<u1WwV-74G-3@gated-at.bofh.it>
ftrace can fail to allocate per-CPU ring buffer on systems with a large
number of CPUs coupled while large amounts of cache happening in the
page cache. Currently the ring buffer allocation doesn't retry in the VM
implementation even if direct-reclaim made some progress but still
wasn't able to find a free page. On retrying I see that the allocations
almost always succeed. The retry doesn't happen because __GFP_NORETRY is
used in the tracer to prevent the case where we might OOM, however if we
drop __GFP_NORETRY, we risk destabilizing the system if OOM killer is
triggered. To prevent this situation, use the __GFP_RETRY_MAYFAIL flag
introduced recently [1].

Tested the following still succeeds without destabilizing a system with
1GB memory.
echo 300000 > /sys/kernel/debug/tracing/buffer_size_kb

[1] https://marc.info/?l=linux-mm&m=149820805124906&w=2

Cc: Alexander Duyck <alexander.h.duyck@intel.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Hao Lee <haolee.swjtu@gmail.com>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Tim Murray <timmurray@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: stable@vger.kernel.org
Signed-off-by: Joel Fernandes <joelaf@google.com>
---
 kernel/trace/ring_buffer.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index 4ae268e687fe..529cc50d7243 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -1136,12 +1136,12 @@ static int __rb_allocate_pages(long nr_pages, struct list_head *pages, int cpu)
 	for (i = 0; i < nr_pages; i++) {
 		struct page *page;
 		/*
-		 * __GFP_NORETRY flag makes sure that the allocation fails
-		 * gracefully without invoking oom-killer and the system is
-		 * not destabilized.
+		 * __GFP_RETRY_MAYFAIL flag makes sure that the allocation fails
+		 * gracefully without invoking oom-killer and the system is not
+		 * destabilized.
 		 */
 		bpage = kzalloc_node(ALIGN(sizeof(*bpage), cache_line_size()),
-				    GFP_KERNEL | __GFP_NORETRY,
+				    GFP_KERNEL | __GFP_RETRY_MAYFAIL,
 				    cpu_to_node(cpu));
 		if (!bpage)
 			goto free_pages;
@@ -1149,7 +1149,7 @@ static int __rb_allocate_pages(long nr_pages, struct list_head *pages, int cpu)
 		list_add(&bpage->list, pages);
 
 		page = alloc_pages_node(cpu_to_node(cpu),
-					GFP_KERNEL | __GFP_NORETRY, 0);
+					GFP_KERNEL | __GFP_RETRY_MAYFAIL, 0);
 		if (!page)
 			goto free_pages;
 		bpage->page = page_address(page);
-- 
2.13.2.725.g09c95d1e9-goog

[toc] | [next] | [standalone]


#1684794

FromMichal Hocko <mhocko@kernel.org>
Date2017-07-11 08:20 +0200
Message-ID<u1WGB-77T-1@gated-at.bofh.it>
In reply to#1684787
On Mon 10-07-17 23:05:00, Joel Fernandes wrote:
> ftrace can fail to allocate per-CPU ring buffer on systems with a large
> number of CPUs coupled while large amounts of cache happening in the
> page cache. Currently the ring buffer allocation doesn't retry in the VM
> implementation even if direct-reclaim made some progress but still
> wasn't able to find a free page. On retrying I see that the allocations
> almost always succeed. The retry doesn't happen because __GFP_NORETRY is
> used in the tracer to prevent the case where we might OOM, however if we
> drop __GFP_NORETRY, we risk destabilizing the system if OOM killer is
> triggered. To prevent this situation, use the __GFP_RETRY_MAYFAIL flag
> introduced recently [1].
> 
> Tested the following still succeeds without destabilizing a system with
> 1GB memory.
> echo 300000 > /sys/kernel/debug/tracing/buffer_size_kb
> 
> [1] https://marc.info/?l=linux-mm&m=149820805124906&w=2

Yes this is the correct usage of the new flag.

> Cc: Alexander Duyck <alexander.h.duyck@intel.com>
> Cc: Mel Gorman <mgorman@suse.de>
> Cc: Hao Lee <haolee.swjtu@gmail.com>
> Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
> Cc: Johannes Weiner <hannes@cmpxchg.org>
> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
> Cc: Michal Hocko <mhocko@kernel.org>
> Cc: Tim Murray <timmurray@google.com>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: Steven Rostedt <rostedt@goodmis.org>
> Cc: stable@vger.kernel.org

I do not think stable tag is appropriate. The new flag hasn't been
merged yet and it is not a stable material.

> Signed-off-by: Joel Fernandes <joelaf@google.com>

Feel free to add
Acked-by: Michal Hocko <mhocko@suse.com>

> ---
>  kernel/trace/ring_buffer.c | 10 +++++-----
>  1 file changed, 5 insertions(+), 5 deletions(-)
> 
> diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
> index 4ae268e687fe..529cc50d7243 100644
> --- a/kernel/trace/ring_buffer.c
> +++ b/kernel/trace/ring_buffer.c
> @@ -1136,12 +1136,12 @@ static int __rb_allocate_pages(long nr_pages, struct list_head *pages, int cpu)
>  	for (i = 0; i < nr_pages; i++) {
>  		struct page *page;
>  		/*
> -		 * __GFP_NORETRY flag makes sure that the allocation fails
> -		 * gracefully without invoking oom-killer and the system is
> -		 * not destabilized.
> +		 * __GFP_RETRY_MAYFAIL flag makes sure that the allocation fails
> +		 * gracefully without invoking oom-killer and the system is not
> +		 * destabilized.
>  		 */
>  		bpage = kzalloc_node(ALIGN(sizeof(*bpage), cache_line_size()),
> -				    GFP_KERNEL | __GFP_NORETRY,
> +				    GFP_KERNEL | __GFP_RETRY_MAYFAIL,
>  				    cpu_to_node(cpu));
>  		if (!bpage)
>  			goto free_pages;
> @@ -1149,7 +1149,7 @@ static int __rb_allocate_pages(long nr_pages, struct list_head *pages, int cpu)
>  		list_add(&bpage->list, pages);
>  
>  		page = alloc_pages_node(cpu_to_node(cpu),
> -					GFP_KERNEL | __GFP_NORETRY, 0);
> +					GFP_KERNEL | __GFP_RETRY_MAYFAIL, 0);
>  		if (!page)
>  			goto free_pages;
>  		bpage->page = page_address(page);
> -- 
> 2.13.2.725.g09c95d1e9-goog

-- 
Michal Hocko
SUSE Labs

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


#1684796

FromVlastimil Babka <vbabka@suse.cz>
Date2017-07-11 08:20 +0200
Message-ID<u1WGB-77T-5@gated-at.bofh.it>
In reply to#1684787
On 07/11/2017 08:05 AM, Joel Fernandes wrote:
> ftrace can fail to allocate per-CPU ring buffer on systems with a large
> number of CPUs coupled while large amounts of cache happening in the
> page cache. Currently the ring buffer allocation doesn't retry in the VM
> implementation even if direct-reclaim made some progress but still
> wasn't able to find a free page. On retrying I see that the allocations
> almost always succeed. The retry doesn't happen because __GFP_NORETRY is
> used in the tracer to prevent the case where we might OOM, however if we
> drop __GFP_NORETRY, we risk destabilizing the system if OOM killer is
> triggered. To prevent this situation, use the __GFP_RETRY_MAYFAIL flag
> introduced recently [1].
> 
> Tested the following still succeeds without destabilizing a system with
> 1GB memory.
> echo 300000 > /sys/kernel/debug/tracing/buffer_size_kb
> 
> [1] https://marc.info/?l=linux-mm&m=149820805124906&w=2
> 
> Cc: Alexander Duyck <alexander.h.duyck@intel.com>
> Cc: Mel Gorman <mgorman@suse.de>
> Cc: Hao Lee <haolee.swjtu@gmail.com>
> Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
> Cc: Johannes Weiner <hannes@cmpxchg.org>
> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
> Cc: Michal Hocko <mhocko@kernel.org>
> Cc: Tim Murray <timmurray@google.com>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: Steven Rostedt <rostedt@goodmis.org>
> Cc: stable@vger.kernel.org

Not stable, as Michal mentioned.

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

> Signed-off-by: Joel Fernandes <joelaf@google.com>
> ---
>  kernel/trace/ring_buffer.c | 10 +++++-----
>  1 file changed, 5 insertions(+), 5 deletions(-)
> 
> diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
> index 4ae268e687fe..529cc50d7243 100644
> --- a/kernel/trace/ring_buffer.c
> +++ b/kernel/trace/ring_buffer.c
> @@ -1136,12 +1136,12 @@ static int __rb_allocate_pages(long nr_pages, struct list_head *pages, int cpu)
>  	for (i = 0; i < nr_pages; i++) {
>  		struct page *page;
>  		/*
> -		 * __GFP_NORETRY flag makes sure that the allocation fails
> -		 * gracefully without invoking oom-killer and the system is
> -		 * not destabilized.
> +		 * __GFP_RETRY_MAYFAIL flag makes sure that the allocation fails
> +		 * gracefully without invoking oom-killer and the system is not
> +		 * destabilized.
>  		 */
>  		bpage = kzalloc_node(ALIGN(sizeof(*bpage), cache_line_size()),
> -				    GFP_KERNEL | __GFP_NORETRY,
> +				    GFP_KERNEL | __GFP_RETRY_MAYFAIL,
>  				    cpu_to_node(cpu));
>  		if (!bpage)
>  			goto free_pages;
> @@ -1149,7 +1149,7 @@ static int __rb_allocate_pages(long nr_pages, struct list_head *pages, int cpu)
>  		list_add(&bpage->list, pages);
>  
>  		page = alloc_pages_node(cpu_to_node(cpu),
> -					GFP_KERNEL | __GFP_NORETRY, 0);
> +					GFP_KERNEL | __GFP_RETRY_MAYFAIL, 0);
>  		if (!page)
>  			goto free_pages;
>  		bpage->page = page_address(page);
> 

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


#1685232

FromJohannes Weiner <hannes@cmpxchg.org>
Date2017-07-11 19:30 +0200
Message-ID<u278Z-5cu-7@gated-at.bofh.it>
In reply to#1684787
On Mon, Jul 10, 2017 at 11:05:00PM -0700, Joel Fernandes wrote:
> ftrace can fail to allocate per-CPU ring buffer on systems with a large
> number of CPUs coupled while large amounts of cache happening in the
> page cache. Currently the ring buffer allocation doesn't retry in the VM
> implementation even if direct-reclaim made some progress but still
> wasn't able to find a free page. On retrying I see that the allocations
> almost always succeed. The retry doesn't happen because __GFP_NORETRY is
> used in the tracer to prevent the case where we might OOM, however if we
> drop __GFP_NORETRY, we risk destabilizing the system if OOM killer is
> triggered. To prevent this situation, use the __GFP_RETRY_MAYFAIL flag
> introduced recently [1].
> 
> Tested the following still succeeds without destabilizing a system with
> 1GB memory.
> echo 300000 > /sys/kernel/debug/tracing/buffer_size_kb
> 
> [1] https://marc.info/?l=linux-mm&m=149820805124906&w=2
> 
> Cc: Alexander Duyck <alexander.h.duyck@intel.com>
> Cc: Mel Gorman <mgorman@suse.de>
> Cc: Hao Lee <haolee.swjtu@gmail.com>
> Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
> Cc: Johannes Weiner <hannes@cmpxchg.org>
> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
> Cc: Michal Hocko <mhocko@kernel.org>
> Cc: Tim Murray <timmurray@google.com>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: Steven Rostedt <rostedt@goodmis.org>
> Cc: stable@vger.kernel.org
> Signed-off-by: Joel Fernandes <joelaf@google.com>

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

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web