mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: dimm <dmitry.adamushko@gmail.com>
To: linux-kernel@vger.kernel.org
Cc: Ingo Molnar <mingo@elte.hu>,
	Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Subject: Re: [git] CFS-devel, group scheduler, fixes
Date: Tue, 18 Sep 2007 21:36:30 +0200	[thread overview]
Message-ID: <1190144190.5204.24.camel@earth> (raw)


[ well, don't expect to find here anything like RDCFS
(no, 'D' does not stand for 'dumb'!). I was focused
on more prosaic things in the mean time so just
didn't have time for writing it.. ]

here is a few cleanup/simplification/optimization(s)
based on the recent modifications in the sched-dev tree.

(1) optimize task_new_fair()
(2) simplify yield_task()
(3) rework enqueue/dequeue_entity() to get rid of
sched_class::set_curr_task()

additionally, the changes somewhat decrease code size:

   text    data     bss     dec     hex filename
  43538    5398      48   48984    bf58 build/kernel/sched.o.before
  43250    5390      48   48688    be30 build/kernel/sched.o

(SMP + lots of debugging options but, I guess, in this case the diff
should remain visible for any combination).

---

(1)

due to the fact that we no longer keep the 'current' within the tree,
dequeue/enqueue_entity() is useless for the 'current' in task_new_fair().
We are about to reschedule and sched_class->put_prev_task() will put
the 'current' back into the tree, based on its new key.

Signed-off-by: Dmitry Adamushko <dmitry.adamushko@gmail.com>

---
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 6e52d5a..5a244e2 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -969,10 +969,11 @@ static void task_new_fair(struct rq *rq, struct task_struct *p)
 
 	if (sysctl_sched_child_runs_first &&
 			curr->vruntime < se->vruntime) {
-
-		dequeue_entity(cfs_rq, curr, 0);
+		/*
+ 		 * Upon rescheduling, sched_class::put_prev_task() will place
+ 		 * 'current' within the tree based on its new key value.
+ 		 */
 		swap(curr->vruntime, se->vruntime);
-		enqueue_entity(cfs_rq, curr, 0);
 	}
 
 	update_stats_enqueue(cfs_rq, se);

---

Dmitry




             reply	other threads:[~2007-09-18 19:36 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-09-18 19:36 dimm [this message]
2007-09-18 20:16 ` Ingo Molnar
2007-09-19  6:03   ` Tong Li
2007-09-19  6:28     ` Mike Galbraith
2007-09-19  7:51       ` Mike Galbraith
2007-09-19  8:42         ` Mike Galbraith
2007-09-19 17:06           ` Tong Li
2007-09-20  4:55             ` Mike Galbraith
2007-09-20  7:15               ` Mike Galbraith
2007-09-20  7:51                 ` Ingo Molnar
2007-09-20  8:11                   ` Mike Galbraith
2007-09-22  3:27                     ` Tong Li
2007-09-22 10:01                       ` Mike Galbraith
2007-09-23  7:14                         ` Mike Galbraith
2007-09-23 11:37                           ` Mike Galbraith
     [not found]                           ` <20070923115847.GA13061@elte.hu>
2007-09-23 15:53                             ` [git] CFS-devel, updates Mike Galbraith
2007-09-24  6:21                           ` [git] CFS-devel, group scheduler, fixes Tong Li
2007-09-24 10:10                             ` Mike Galbraith
2007-09-24 10:24                               ` Peter Zijlstra
2007-09-24 10:42                                 ` Mike Galbraith
2007-09-24 11:08                                   ` Peter Zijlstra
2007-09-24 11:43                                     ` Mike Galbraith
2007-09-24 11:22                                   ` Mike Galbraith
2007-09-24 11:51                                     ` Peter Zijlstra
2007-09-24 16:43                                       ` Tong Li
2007-09-20 19:48                 ` Willy Tarreau
2007-09-21  2:40                   ` Mike Galbraith
2007-09-21  3:11                     ` Willy Tarreau
2007-09-19 19:35     ` Siddha, Suresh B
2007-09-19 20:58       ` Tong Li
2007-09-18 20:22 ` Ingo Molnar
2007-09-19  3:55   ` Srivatsa Vaddagiri
  -- strict thread matches above, loose matches on Subject: below --
2007-09-18 19:56 Dmitry Adamushko
2007-09-18 20:18 ` Ingo Molnar
2007-09-18 19:46 Dmitry Adamushko
2007-09-18 20:17 ` Ingo Molnar
2007-09-15 13:06 Ingo Molnar

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1190144190.5204.24.camel@earth \
    --to=dmitry.adamushko@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=vatsa@linux.vnet.ibm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

Powered by JetHome