The patch titled
Subject: mm/vmscan: add tracepoints for node reclaim
has been added to the -mm tree. Its filename is
mm-vmscan-add-tracepoints-for-node-reclaim.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/mm-vmscan-add-tracepoints-for-node-reclaim.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/mm-vmscan-add-tracepoints-for-node-reclaim.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Yafang Shao <laoar.shao@xxxxxxxxx>
Subject: mm/vmscan: add tracepoints for node reclaim
The page alloc fast path it may perform node reclaim, which may cause a
latency spike. We should add tracepoint for this event, and also measure
the latency it causes.
So bellow two tracepoints are introduced,
mm_vmscan_node_reclaim_begin
mm_vmscan_node_reclaim_end
Link: http://lkml.kernel.org/r/1551421452-5385-1-git-send-email-laoar.shao@xxxxxxxxx
Signed-off-by: Yafang Shao <laoar.shao@xxxxxxxxx>
Cc: Vlastimil Babka <vbabka@xxxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxxx>
Cc: Souptick Joarder <jrdr.linux@xxxxxxxxx>
Cc: <shaoyafang@xxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/trace/events/vmscan.h | 32 ++++++++++++++++++++++++++++++++
mm/vmscan.c | 6 ++++++
2 files changed, 38 insertions(+)
--- a/include/trace/events/vmscan.h~mm-vmscan-add-tracepoints-for-node-reclaim
+++ a/include/trace/events/vmscan.h
@@ -469,6 +469,38 @@ TRACE_EVENT(mm_vmscan_inactive_list_is_l
__entry->ratio,
show_reclaim_flags(__entry->reclaim_flags))
);
+
+TRACE_EVENT(mm_vmscan_node_reclaim_begin,
+
+ TP_PROTO(int nid, int order, gfp_t gfp_flags),
+
+ TP_ARGS(nid, order, gfp_flags),
+
+ TP_STRUCT__entry(
+ __field(int, nid)
+ __field(int, order)
+ __field(gfp_t, gfp_flags)
+ ),
+
+ TP_fast_assign(
+ __entry->nid = nid;
+ __entry->order = order;
+ __entry->gfp_flags = gfp_flags;
+ ),
+
+ TP_printk("nid=%d order=%d gfp_flags=%s",
+ __entry->nid,
+ __entry->order,
+ show_gfp_flags(__entry->gfp_flags))
+);
+
+DEFINE_EVENT(mm_vmscan_direct_reclaim_end_template, mm_vmscan_node_reclaim_end,
+
+ TP_PROTO(unsigned long nr_reclaimed),
+
+ TP_ARGS(nr_reclaimed)
+);
+
#endif /* _TRACE_VMSCAN_H */
/* This part must be outside protection */
--- a/mm/vmscan.c~mm-vmscan-add-tracepoints-for-node-reclaim
+++ a/mm/vmscan.c
@@ -4122,6 +4122,9 @@ static int __node_reclaim(struct pglist_
.reclaim_idx = gfp_zone(gfp_mask),
};
+ trace_mm_vmscan_node_reclaim_begin(pgdat->node_id, order,
+ sc.gfp_mask);
+
cond_resched();
fs_reclaim_acquire(sc.gfp_mask);
/*
@@ -4148,6 +4151,9 @@ static int __node_reclaim(struct pglist_
current->flags &= ~PF_SWAPWRITE;
memalloc_noreclaim_restore(noreclaim_flag);
fs_reclaim_release(sc.gfp_mask);
+
+ trace_mm_vmscan_node_reclaim_end(sc.nr_reclaimed);
+
return sc.nr_reclaimed >= nr_pages;
}
_
Patches currently in -mm which might be from laoar.shao@xxxxxxxxx are
mm-vmscan-drop-zone-id-from-kswapd-tracepoints.patch
mm-compaction-show-gfp-flag-names-in-try_to_compact_pages-tracepoint.patch
mm-compaction-some-tracepoints-should-be-defined-only-when-config_compaction-is-set.patch
mm-vmscan-add-tracepoints-for-node-reclaim.patch
