lkml.org 
[lkml]   [2014]   [Jun]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectInteractivity regression since v3.11 in mm/vmscan.c
Hi,

For a while I've noticed that my machine bogs down in certain
situations, usually while doing heavy I/O operations, it is not just the
I/O operations, but everything, including the graphical interface, even
the mouse pointer.

As far as I can recall this did not happen in the past.

I noticed this specially on certain operations, for example updating a
a game on Steam (to an exteranl USB 3.0 device), or copying TV episodes
to a USB memory stick (probably flash-based).

Today I decided to finally hunt down the problem, so I created a
synthetic test that basically consists on copying a bunch of files from
one drive to another (from an SSD to an external USB 3.0). This is
pretty similar to what I noticed; the graphical interface slows down.

Then I bisected the issue and it turns out that indeed it wasn't
happening in the past, it started happening in v3.11, and it was
triggered by this commit:

e2be15f (mm: vmscan: stall page reclaim and writeback pages based on
dirty/writepage pages encountered)

Then I went back to the latest stable version (v3.14.5), and commented
out the line I think is causing the slow down:

if (nr_unqueued_dirty == nr_taken || nr_immediate)
congestion_wait(BLK_RW_ASYNC, HZ/10);

After that I don't notice the slow down any more.

Anybody has any ideas how to fix the issue properly?

--
Felipe Contreras


\
 
 \ /
  Last update: 2014-06-05 14:21    [W:0.160 / U:0.036 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site