lkml.org 
[lkml]   [2019]   [Jan]   [1]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
SubjectRe: [PATCH v3 2/2] mm: rid swapoff of quadratic complexity
On Tue, 1 Jan 2019, Vineeth Pillai wrote:

> Thanks a lot for the fixes and detailed explanation Hugh! I shall fold all
> the changes from you and Huang in the next iteration.
>
> Thanks for all the suggestions and comments as well. I am looking into all
> those and will include all the changes in the next version. Will discuss
> over mail in case of any clarifications.

One more fix on top of what I sent yesterday: once I delved into
the retries, I found that the major cause of exceeding MAX_RETRIES
was the way the retry code neatly avoided retrying the last part of
its work. With this fix in, I have not yet seen retries go above 1:
no doubt it could, but at present I have no actual evidence that
the MAX_RETRIES-or-livelock issue needs to be dealt with urgently.
Fix sent for completeness, but it reinforces the point that the
structure of try_to_unuse() should be reworked, and oldi gone.

Hugh

---

mm/swapfile.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)

--- mmotm/mm/swapfile.c 2018-12-31 12:30:55.822407154 -0800
+++ linux/mm/swapfile.c 2019-01-01 19:50:34.377277830 -0800
@@ -2107,8 +2107,8 @@ int try_to_unuse(unsigned int type, bool
struct swap_info_struct *si = swap_info[type];
struct page *page;
swp_entry_t entry;
- unsigned int i = 0;
- unsigned int oldi = 0;
+ unsigned int i;
+ unsigned int oldi;
int retries = 0;

if (!frontswap)
@@ -2154,6 +2154,7 @@ retry:
goto out;
}

+ i = oldi = 0;
while ((i = find_next_to_unuse(si, i, frontswap)) != 0) {
/*
* Under global memory pressure, swap entries
\
 
 \ /
  Last update: 2019-01-02 05:17    [W:0.076 / U:3.828 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site