psi: clarify the Kconfig text for the default-disable option

Commit Message

Johannes Weiner Jan. 29, 2019, 11:36 p.m. UTC

The current help text caused some confusion in online forums about
whether or not to default-enable or default-disable psi in vendor
kernels. This is because it doesn't communicate the reason for why we
made this setting configurable in the first place: that the overhead
is non-zero in an artificial scheduler stress test.

Since this isn't representative of real workloads, and the effect was
not measurable in scheduler-heavy real world applications such as the
webservers and memcache installations at Facebook, it's fair to point
out that this is a pretty cautious option to select.

Signed-off-by: Johannes Weiner <[e-mail]>
---
 init/Kconfig | 11 +++++++++++
 1 file changed, 11 insertions(+)

Patch

diff --git a/init/Kconfig b/init/Kconfig
index 513fa544a134..ad3381e57402 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -512,6 +512,17 @@  config PSI_DEFAULT_DISABLED
 	  per default but can be enabled through passing psi=1 on the
 	  kernel commandline during boot.
 
+	  This feature adds some code to the task wakeup and sleep
+	  paths of the scheduler. The overhead is too low to affect
+	  common scheduling-intense workloads in practice (such as
+	  webservers, memcache), but it does show up in artificial
+	  scheduler stress tests, such as hackbench.
+
+	  If you are paranoid and not sure what the kernel will be
+	  used for, say Y.
+
+	  Say N if unsure.
+
 endmenu # "CPU/Task time and stats accounting"
 
 config CPU_ISOLATION