* [PATCH] writeback: trace global_dirty_state
@ 2011-06-10 14:48 Wu Fengguang
2011-06-13 14:33 ` Jan Kara
0 siblings, 1 reply; 3+ messages in thread
From: Wu Fengguang @ 2011-06-10 14:48 UTC (permalink / raw)
To: Andrew Morton
Cc: Jan Kara, Dave Chinner, Christoph Hellwig,
Linux Memory Management List, linux-fsdevel, LKML
[It seems beneficial to queue this simple trace event for
next/upstream after the review?]
Add trace event balance_dirty_state for showing the global dirty page
counts and thresholds at each global_dirty_limits() invocation. This
will cover the callers throttle_vm_writeout(), over_bground_thresh()
and each balance_dirty_pages() loop.
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
---
include/trace/events/writeback.h | 36 +++++++++++++++++++++++++++++
mm/page-writeback.c | 1
2 files changed, 37 insertions(+)
--- linux-next.orig/mm/page-writeback.c 2011-06-10 21:52:34.000000000 +0800
+++ linux-next/mm/page-writeback.c 2011-06-10 22:08:26.000000000 +0800
@@ -430,6 +430,7 @@ void global_dirty_limits(unsigned long *
}
*pbackground = background;
*pdirty = dirty;
+ trace_global_dirty_state(background, dirty);
}
/**
--- linux-next.orig/include/trace/events/writeback.h 2011-06-10 21:52:34.000000000 +0800
+++ linux-next/include/trace/events/writeback.h 2011-06-10 22:25:33.000000000 +0800
@@ -187,6 +187,42 @@ TRACE_EVENT(writeback_queue_io,
__entry->moved)
);
+TRACE_EVENT(global_dirty_state,
+
+ TP_PROTO(unsigned long background_thresh,
+ unsigned long dirty_thresh
+ ),
+
+ TP_ARGS(background_thresh,
+ dirty_thresh
+ ),
+
+ TP_STRUCT__entry(
+ __field(unsigned long, nr_dirty)
+ __field(unsigned long, nr_writeback)
+ __field(unsigned long, nr_unstable)
+ __field(unsigned long, background_thresh)
+ __field(unsigned long, dirty_thresh)
+ ),
+
+ TP_fast_assign(
+ __entry->nr_dirty = global_page_state(NR_FILE_DIRTY);
+ __entry->nr_writeback = global_page_state(NR_WRITEBACK);
+ __entry->nr_unstable = global_page_state(NR_UNSTABLE_NFS);
+ __entry->background_thresh = background_thresh;
+ __entry->dirty_thresh = dirty_thresh;
+ ),
+
+ TP_printk("dirty=%lu writeback=%lu unstable=%lu "
+ "bg_thresh=%lu thresh=%lu",
+ __entry->nr_dirty,
+ __entry->nr_writeback,
+ __entry->nr_unstable,
+ __entry->background_thresh,
+ __entry->dirty_thresh
+ )
+);
+
DECLARE_EVENT_CLASS(writeback_congest_waited_template,
TP_PROTO(unsigned int usec_timeout, unsigned int usec_delayed),
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] writeback: trace global_dirty_state
2011-06-10 14:48 [PATCH] writeback: trace global_dirty_state Wu Fengguang
@ 2011-06-13 14:33 ` Jan Kara
2011-06-14 13:46 ` Wu Fengguang
0 siblings, 1 reply; 3+ messages in thread
From: Jan Kara @ 2011-06-13 14:33 UTC (permalink / raw)
To: Wu Fengguang
Cc: Andrew Morton, Jan Kara, Dave Chinner, Christoph Hellwig,
Linux Memory Management List, linux-fsdevel, LKML
On Fri 10-06-11 22:48:05, Wu Fengguang wrote:
> [It seems beneficial to queue this simple trace event for
> next/upstream after the review?]
>
> Add trace event balance_dirty_state for showing the global dirty page
> counts and thresholds at each global_dirty_limits() invocation. This
> will cover the callers throttle_vm_writeout(), over_bground_thresh()
> and each balance_dirty_pages() loop.
OK, this might be useful. But shouldn't we also add similar trace point
for bdi limits? Otherwise the information is of limited use...
Honza
>
> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
> ---
> include/trace/events/writeback.h | 36 +++++++++++++++++++++++++++++
> mm/page-writeback.c | 1
> 2 files changed, 37 insertions(+)
>
> --- linux-next.orig/mm/page-writeback.c 2011-06-10 21:52:34.000000000 +0800
> +++ linux-next/mm/page-writeback.c 2011-06-10 22:08:26.000000000 +0800
> @@ -430,6 +430,7 @@ void global_dirty_limits(unsigned long *
> }
> *pbackground = background;
> *pdirty = dirty;
> + trace_global_dirty_state(background, dirty);
> }
>
> /**
> --- linux-next.orig/include/trace/events/writeback.h 2011-06-10 21:52:34.000000000 +0800
> +++ linux-next/include/trace/events/writeback.h 2011-06-10 22:25:33.000000000 +0800
> @@ -187,6 +187,42 @@ TRACE_EVENT(writeback_queue_io,
> __entry->moved)
> );
>
> +TRACE_EVENT(global_dirty_state,
> +
> + TP_PROTO(unsigned long background_thresh,
> + unsigned long dirty_thresh
> + ),
> +
> + TP_ARGS(background_thresh,
> + dirty_thresh
> + ),
> +
> + TP_STRUCT__entry(
> + __field(unsigned long, nr_dirty)
> + __field(unsigned long, nr_writeback)
> + __field(unsigned long, nr_unstable)
> + __field(unsigned long, background_thresh)
> + __field(unsigned long, dirty_thresh)
> + ),
> +
> + TP_fast_assign(
> + __entry->nr_dirty = global_page_state(NR_FILE_DIRTY);
> + __entry->nr_writeback = global_page_state(NR_WRITEBACK);
> + __entry->nr_unstable = global_page_state(NR_UNSTABLE_NFS);
> + __entry->background_thresh = background_thresh;
> + __entry->dirty_thresh = dirty_thresh;
> + ),
> +
> + TP_printk("dirty=%lu writeback=%lu unstable=%lu "
> + "bg_thresh=%lu thresh=%lu",
> + __entry->nr_dirty,
> + __entry->nr_writeback,
> + __entry->nr_unstable,
> + __entry->background_thresh,
> + __entry->dirty_thresh
> + )
> +);
> +
> DECLARE_EVENT_CLASS(writeback_congest_waited_template,
>
> TP_PROTO(unsigned int usec_timeout, unsigned int usec_delayed),
--
Jan Kara <jack@suse.cz>
SUSE Labs, CR
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] writeback: trace global_dirty_state
2011-06-13 14:33 ` Jan Kara
@ 2011-06-14 13:46 ` Wu Fengguang
0 siblings, 0 replies; 3+ messages in thread
From: Wu Fengguang @ 2011-06-14 13:46 UTC (permalink / raw)
To: Jan Kara
Cc: Andrew Morton, Dave Chinner, Christoph Hellwig,
Linux Memory Management List, linux-fsdevel@vger.kernel.org, LKML
On Mon, Jun 13, 2011 at 10:33:56PM +0800, Jan Kara wrote:
> On Fri 10-06-11 22:48:05, Wu Fengguang wrote:
> > [It seems beneficial to queue this simple trace event for
> > next/upstream after the review?]
> >
> > Add trace event balance_dirty_state for showing the global dirty page
> > counts and thresholds at each global_dirty_limits() invocation. This
> > will cover the callers throttle_vm_writeout(), over_bground_thresh()
> > and each balance_dirty_pages() loop.
> OK, this might be useful. But shouldn't we also add similar trace point
> for bdi limits? Otherwise the information is of limited use...
Good point. The bdi limits will be exported in another
balance_dirty_pages trace point.
Thanks,
Fengguang
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2011-06-14 13:47 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-06-10 14:48 [PATCH] writeback: trace global_dirty_state Wu Fengguang
2011-06-13 14:33 ` Jan Kara
2011-06-14 13:46 ` Wu Fengguang
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).