#36 - credit2 only uses one runqueue instead of one runq per socket

Owner: Dario Faggioli <dario.faggioli@citrix.com>

Date: Tue Jan 21 16:00:02 2014

Last Update: Tue Jan 21 16:00:02 2014

Severity: normal

Affects:

State: Open

[ Retrieve as mbox ]


Missing Control message: <1390320437.32519.18.camel@kazak.uk.xensource.com>; (Archives: marc.info, gmane)


From: Dario Faggioli <dario.faggioli@citrix.com>
To: Justin Weaver <jtweaver@hawaii.edu>
Cc: xen-devel <xen-devel@lists.xen.org>, George Dunlap <George.Dunlap@eu.citrix.com>
Subject: [Xen-devel] multiple runqueues in credit2
Date: Sat, 14 Dec 2013 02:33:05 +0100
Message-ID: <1386984785.3980.96.camel@Solace>

[ Reply to this message; Retrieve Raw Message; Archives: marc.info, gmane ]

[Part 1 (text/plain, inline)]
Hi George,

Bot Justin and I were able to reproduce a situation where, on a 2 socket
system (see below), credit2 was activating only 1 runqueue.

That seemed in line with some comment in the sched_credit2.c source
file, such as this one:

 /*
  * Design:
  *
  * VMs "burn" credits based on their weight; higher weight means
  * credits burn more slowly.  The highest weight vcpu burns credits at
  * a rate of 1 credit per nanosecond.  Others burn proportionally
  * more.
  *
  * vcpus are inserted into the runqueue by credit order.
  *
  * Credits are "reset" when the next vcpu in the runqueue is less than
  * or equal to zero.  At that point, everyone's credits are "clipped"
  * to a small value, and a fixed credit is added to everyone.
  *
  * The plan is for all cores that share an L2 will share the same
  * runqueue.  At the moment, there is one global runqueue for all
  * cores.
  */

However, I remembered it different, and looking at init_pcpu() I spotted
this:

    /* Figure out which runqueue to put it in */
    /* NB: cpu 0 doesn't get a STARTING callback, so we hard-code it to runqueue 0. */
    if ( cpu == 0 )
        rqi = 0;
    else
        rqi = cpu_to_socket(cpu);

which looks to me like the code for having one runqueue per socket _is_
there already! That means two things: (1) that comment above is
wrong :-) but, at the same time, (2) this code right here is not
working!

Justin also noticed that init_pcpu() was actually being called twice,
for all pcpus except #0, triggering the following warning:

    printk("%s: Strange, cpu %d already initialized!\n", __func__, cpu);

I did some investigation, in the following system:

cpu_topology           :
cpu:    core    socket     node
  0:       0        0        0
  1:       1        0        0
  2:       2        0        0
  3:       3        0        0
  4:       0        1        1
  5:       1        1        1
  6:       2        1        1
  7:       3        1        1

So, what I expect is, for instance, cpu 1 to be on runqueue 0, and cpu 5
on runqueue 1.

The problem is here:

  static void *
  csched_alloc_pdata(const struct scheduler *ops, int cpu)
  {
      /* Check to see if the cpu is online yet */
      /* Note: cpu 0 doesn't get a STARTING callback */
      if ( cpu == 0 || cpu_to_socket(cpu) >= 0 )
          init_pcpu(ops, cpu);
      else
          printk("%s: cpu %d not online yet, deferring initializatgion\n",
                 __func__, cpu);

      return (void *)1;
  }

In fact, this is meant to actually call init_pcpu() *only* on pcpu 0
(which don't get the STARTING notification) and on those pcpus that are
already onlined. Unfortunately, "cpu_to_socket(cpu) >= 0" is not (any
longer?) a valid way to check the latter, and in fact init_pcpus() is
always called, even for pcpus that are not identified and inited yet.
That, with cpu_to_socket() returning constantly 0, means all the pcpus
end up in the sole and only runqueue 0.

I verified that removing the right side of the || makes things work (I
enabled some debug output and added some more myself):

(XEN) csched_alloc_pdata for cpu 0 on socket 0
(XEN) Adding cpu 0 to runqueue 0
(XEN)  First cpu on runqueue, activating
...
(XEN) CPU 1 APIC 1 -> Node 0
(XEN) csched_vcpu_insert: Inserting d32767v1
(XEN) csched_alloc_pdata for cpu 1 on socket 0
(XEN) csched_alloc_pdata: cpu 1 not online yet, deferring initializatgion
(XEN) Booting processor 1/1 eip 8e000
(XEN) Initializing CPU#1
(XEN) CPU: L1 I cache 64K (64 bytes/line), D cache 64K (64 bytes/line)
(XEN) CPU: L2 Cache: 512K (64 bytes/line)
(XEN) CPU 1(4) -> Processor 0, Core 1
(XEN) CPU1: AMD Quad-Core AMD Opteron(tm) Processor 2376 stepping 02
(XEN) csched_cpu_starting on cpu 1
(XEN) Adding cpu 1 to runqueue 0
...
(XEN) CPU 5 APIC 5 -> Node 1
(XEN) microcode: CPU4 collect_cpu_info: patch_id=0x1000086
(XEN) csched_vcpu_insert: Inserting d32767v5
(XEN) csched_alloc_pdata for cpu 5 on socket 0
(XEN) csched_alloc_pdata: cpu 5 not online yet, deferring initializatgion
(XEN) Booting processor 5/5 eip 8e000
(XEN) Initializing CPU#5
(XEN) CPU: L1 I cache 64K (64 bytes/line), D cache 64K (64 bytes/line)
(XEN) CPU: L2 Cache: 512K (64 bytes/line)
(XEN) CPU 5(4) -> Processor 1, Core 1
(XEN) CPU5: AMD Quad-Core AMD Opteron(tm) Processor 2376 stepping 02
(XEN) csched_cpu_starting on cpu 5
(XEN) Adding cpu 5 to runqueue 1
...

Now the question is, for fixing this, would it be preferable to do
something along this line (i.e., removing the right side of the || and,
in general, make csched_alloc_pdata() a pcpu 0 only thing)? Or, perhaps,
should I look into a way to properly initialize the cpu_data array, so
that cpu_to_socket() actually returns something '< 0' for pcpus not yet
onlined and identified?

The former is surely quicker, but I think I like the latter better
(provided it's doable). What do you think?

Thanks and Regards,
Dario

-- 
<<This happens because I choose it to happen!>> (Raistlin Majere)
-----------------------------------------------------------------
Dario Faggioli, Ph.D, http://about.me/dario.faggioli
Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)
[signature.asc (application/pgp-signature, inline)]
[Part 3 (text/plain, inline)]

From: Dario Faggioli <dario.faggioli@citrix.com>
To: Justin Weaver <jtweaver@hawaii.edu>
Cc: xen-devel <xen-devel@lists.xen.org>, George Dunlap <George.Dunlap@eu.citrix.com>
Subject: Re: [Xen-devel] multiple runqueues in credit2
Date: Mon, 20 Jan 2014 19:16:39 +0100
Message-ID: <1390241799.23576.42.camel@Solace>

[ Reply to this message; Retrieve Raw Message; Archives: marc.info, gmane ]

[Part 1 (text/plain, inline)]
create ^
title it credit2 only uses one runqueue instead of one runq per socket
thanks

On sab, 2013-12-14 at 02:33 +0100, Dario Faggioli wrote:
> Hi George,
> 
BTW, creating a tracking bug entry for this issue.

> Now the question is, for fixing this, would it be preferable to do
> something along this line (i.e., removing the right side of the || and,
> in general, make csched_alloc_pdata() a pcpu 0 only thing)? Or, perhaps,
> should I look into a way to properly initialize the cpu_data array, so
> that cpu_to_socket() actually returns something '< 0' for pcpus not yet
> onlined and identified?
> 
I prepared and gave it a quick try to the attached patch... Only to
figure out that it won't work.

Well, it does for certain configurations (so, perhaps, Justin, if that
is your case you may be able to at least do some development on top of
it), but it's not the correct approach... Or at least it's not enough.

In fact, what it does is initializing the pCPU info field used by
cpu_to_socket() to -1, which means now all pCPUs --apart from pCPU 0--
are associated with the proper runqueue.

pCPU 0, OTOH, is always associated with runqueue 0, and that is
necessary and intended, as it does not get the notifier call, and hence
it needs to be initialized when the correct cpu_to_socket() information
is still not available. And that's where the problem is. In fact, this
is fine if pCPU 0 is actually on socket 0, but what if it is, say, on
socket 1? :-O

That happens to be the case on one of my test boxes, and here's what I
get on it:

root@Zhaman:~# xl dmesg |grep runqueue
(XEN) Adding cpu 0 to runqueue 0
(XEN)  First cpu on runqueue, activating
(XEN) Adding cpu 1 to runqueue 1
(XEN)  First cpu on runqueue, activating
(XEN) Adding cpu 2 to runqueue 1
(XEN) Adding cpu 3 to runqueue 1
(XEN) Adding cpu 4 to runqueue 1
(XEN) Adding cpu 5 to runqueue 1
(XEN) Adding cpu 6 to runqueue 1
(XEN) Adding cpu 7 to runqueue 1
(XEN) Adding cpu 8 to runqueue 0
(XEN) Adding cpu 9 to runqueue 0
(XEN) Adding cpu 10 to runqueue 0
(XEN) Adding cpu 11 to runqueue 0
(XEN) Adding cpu 12 to runqueue 0
(XEN) Adding cpu 13 to runqueue 0
(XEN) Adding cpu 14 to runqueue 0
(XEN) Adding cpu 15 to runqueue 0

root@Zhaman:~# xl dmesg |grep 'runqueue 0'|cat -n
     1	(XEN) Adding cpu 0 to runqueue 0
     2	(XEN) Adding cpu 8 to runqueue 0
     3	(XEN) Adding cpu 9 to runqueue 0
     4	(XEN) Adding cpu 10 to runqueue 0
     5	(XEN) Adding cpu 11 to runqueue 0
     6	(XEN) Adding cpu 12 to runqueue 0
     7	(XEN) Adding cpu 13 to runqueue 0
     8	(XEN) Adding cpu 14 to runqueue 0
     9	(XEN) Adding cpu 15 to runqueue 0
root@Zhaman:~# xl dmesg |grep 'runqueue 1'|cat -n
     1	(XEN) Adding cpu 1 to runqueue 1
     2	(XEN) Adding cpu 2 to runqueue 1
     3	(XEN) Adding cpu 3 to runqueue 1
     4	(XEN) Adding cpu 4 to runqueue 1
     5	(XEN) Adding cpu 5 to runqueue 1
     6	(XEN) Adding cpu 6 to runqueue 1
     7	(XEN) Adding cpu 7 to runqueue 1

:-(

I'll keep looking into this, although I can't promise it will be my top
priority for the coming weeks. :-/

If, in the meantime, someone (George?) has an idea on how to solve this,
I gladly accept suggestions. :-)

Regards,
Dario

-- 
<<This happens because I choose it to happen!>> (Raistlin Majere)
-----------------------------------------------------------------
Dario Faggioli, Ph.D, http://about.me/dario.faggioli
Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)
[phys_proc_id-init.patch (text/x-patch, attachment)]
[Part 3 (text/plain, inline)]

From: Dario Faggioli <dario.faggioli@citrix.com>
To: Justin Weaver <jtweaver@hawaii.edu>
Cc: George Dunlap <George.Dunlap@eu.citrix.com>, xen-devel <xen-devel@lists.xen.org>
Subject: Re: [Xen-devel] multiple runqueues in credit2
Date: Tue, 21 Jan 2014 16:55:12 +0100
Message-ID: <1390319712.23576.202.camel@Solace>

[ Reply to this message; Retrieve Raw Message; Archives: marc.info, gmane ]

create ^
title it credit2 only uses one runqueue instead of one runq per socket
thanks

-- 
<<This happens because I choose it to happen!>> (Raistlin Majere)
-----------------------------------------------------------------
Dario Faggioli, Ph.D, http://about.me/dario.faggioli
Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel