lkml.org 
[lkml]   [2017]   [Apr]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
SubjectIs it safe for kthreadd to drain_all_pages?
Hi Mel,

I suspect that it's not safe for kthreadd to drain_all_pages();
but I haven't studied flush_work() etc, so don't really know what
I'm talking about: hoping that you will jump to a realization.

4.11-rc has been giving me hangs after hours of swapping load. At
first they looked like memory leaks ("fork: Cannot allocate memory");
but for no good reason I happened to do "cat /proc/sys/vm/stat_refresh"
before looking at /proc/meminfo one time, and the stat_refresh stuck
in D state, waiting for completion of flush_work like many kworkers.
kthreadd waiting for completion of flush_work in drain_all_pages().

But I only noticed that pattern later: originally tried to bisect
rc1 before rc2 came out, but underestimated how long to wait before
deciding a stage good - I thought 12 hours, but would now say 2 days.
Too late for bisection, I suspect your drain_all_pages() changes.

(I've also found order:0 page allocation stalls in /var/log/messages,
148804ms a nice example: which suggest that these hangs are perhaps a
condition it can sometimes get out of itself. None with the patch.)

Patch below has been running well for 36 hours now:
a bit too early to be sure, but I think it's time to turn to you.


[PATCH] mm: don't let kthreadd drain_all_pages

4.11-rc has been giving me hangs after many hours of swapping load: most
kworkers waiting for completion of a flush_work, kthreadd waiting for
completion of flush_work in drain_all_pages (while doing copy_process).
I suspect that kthreadd should not be allowed to drain_all_pages().

Signed-off-by: Hugh Dickins <hughd@google.com>
---

mm/page_alloc.c | 2 ++
1 file changed, 2 insertions(+)

--- 4.11-rc5/mm/page_alloc.c 2017-03-13 09:08:37.743209168 -0700
+++ linux/mm/page_alloc.c 2017-04-04 00:33:44.086867413 -0700
@@ -2376,6 +2376,8 @@ void drain_all_pages(struct zone *zone)
/* Workqueues cannot recurse */
if (current->flags & PF_WQ_WORKER)
return;
+ if (current == kthreadd_task)
+ return;

/*
* Do not drain if one is already in progress unless it's specific to
\
 
 \ /
  Last update: 2017-04-05 23:00    [W:0.449 / U:0.132 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site