qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: Anthony Liguori <aliguori@us.ibm.com>
Cc: mst@redhat.com, qemu-devel@nongnu.org,
	Aurelien Jarno <aurelien@aurel32.net>,
	Andreas Faerber <afaerber@suse.de>
Subject: Re: [Qemu-devel] [PATCH for-1.5 0/9] Disable expensive QOM cast debugging for official releases
Date: Fri, 10 May 2013 15:08:00 +0200	[thread overview]
Message-ID: <518CF130.2030402@redhat.com> (raw)
In-Reply-To: <874nebdm4r.fsf@codemonkey.ws>

Il 10/05/2013 15:01, Anthony Liguori ha scritto:
> Paolo Bonzini <pbonzini@redhat.com> writes:
> 
>> Cast debugging can have a substantial cost (20% or more, measured by
>> Aurelien on qemu-system-ppc64).
> 
> [Needs citation]

Sure: http://permalink.gmane.org/gmane.comp.emulators.qemu/210830

 49,73%  perf-10672.map           [.] 0x7f7853ab4e0f
 13,23%  qemu-system-ppc64        [.] cpu_ppc_exec
 13,16%  libglib-2.0.so.0.3200.4  [.] g_hash_table_lookup
  8,18%  libglib-2.0.so.0.3200.4  [.] g_str_hash
  2,47%  qemu-system-ppc64        [.] object_class_dynamic_cast
  1,97%  qemu-system-ppc64        [.] type_is_ancestor
  1,05%  libglib-2.0.so.0.3200.4  [.] g_str_equal

That's ~27%.

>> Instead of adding special-cased "fast
>> casts" in the hot paths, we can just disable it in releases.  At the
>> same time, add tracing facilities that simplify the analysys of those
>> problems that cast debugging would reveal.
> 
> I'd prefer not to disable but instead focus on improving performance.

For 1.5?  This is a regression in 1.5 due to more and more usage of
foo_env_on_cpu.

Paolo

> I suspect any performance overhead is resolving the type string to
> typeimpl.  One work around is to have a dynamic cast that takes a
> typeimpl and then use a function that returns a static similar to how
> glib works.
> 
> If you've got a reproducible case where the overhead is high, it should
> be easy to check.
> 
> Regards,
> 
> Anthony Liguori
> 
>>
>> At least patches 1-7 are for 1.5.
>>
>> Paolo Bonzini (9):
>>   qom: improve documentation of cast functions
>>   qom: allow casting of a NULL class
>>   qom: add a fast path to object_class_dynamic_cast
>>   qom: pass file/line/function to asserting casts
>>   qom: trace asserting casts
>>   qom: allow turning cast debugging off
>>   build: disable QOM cast debugging for official releases
>>   qom: simplify object_class_dynamic_cast, part 1
>>   qom: simplify object_class_dynamic_cast, part 2
>>
>>  configure            | 20 ++++++++------
>>  include/qom/object.h | 40 ++++++++++++++++++++++-----
>>  qom/object.c         | 77 ++++++++++++++++++++++++++++++++++------------------
>>  trace-events         |  3 ++
>>  4 files changed, 99 insertions(+), 41 deletions(-)
>>
>> -- 
>> 1.8.1.4
> 

  reply	other threads:[~2013-05-10 13:08 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-05-10 12:16 [Qemu-devel] [PATCH for-1.5 0/9] Disable expensive QOM cast debugging for official releases Paolo Bonzini
2013-05-10 12:16 ` [Qemu-devel] [PATCH for-1.5 1/9] qom: improve documentation of cast functions Paolo Bonzini
2013-05-10 12:16 ` [Qemu-devel] [PATCH for-1.5 2/9] qom: allow casting of a NULL class Paolo Bonzini
2013-05-10 12:16 ` [Qemu-devel] [PATCH for-1.5 3/9] qom: add a fast path to object_class_dynamic_cast Paolo Bonzini
2013-05-10 12:16 ` [Qemu-devel] [PATCH for-1.5 4/9] qom: pass file/line/function to asserting casts Paolo Bonzini
2013-05-10 12:16 ` [Qemu-devel] [PATCH for-1.5 5/9] qom: trace " Paolo Bonzini
2013-05-10 12:16 ` [Qemu-devel] [PATCH for-1.5 6/9] qom: allow turning cast debugging off Paolo Bonzini
2013-05-10 12:16 ` [Qemu-devel] [PATCH for-1.5 7/9] build: disable QOM cast debugging for official releases Paolo Bonzini
2013-05-10 12:16 ` [Qemu-devel] [PATCH for-1.5 8/9] qom: simplify object_class_dynamic_cast, part 1 Paolo Bonzini
2013-05-10 17:52   ` Anthony Liguori
2013-05-10 12:16 ` [Qemu-devel] [PATCH for-1.5 9/9] qom: simplify object_class_dynamic_cast, part 2 Paolo Bonzini
2013-05-10 13:01 ` [Qemu-devel] [PATCH for-1.5 0/9] Disable expensive QOM cast debugging for official releases Anthony Liguori
2013-05-10 13:08   ` Paolo Bonzini [this message]
2013-05-10 13:23     ` Andreas Färber
2013-05-10 13:30       ` Paolo Bonzini
2013-05-10 14:22         ` Aurelien Jarno
2013-05-10 14:39         ` Anthony Liguori
2013-05-10 14:59           ` Paolo Bonzini
2013-05-10 17:41             ` Anthony Liguori
2013-05-10 19:00               ` Aurelien Jarno
2013-05-10 19:35                 ` Anthony Liguori
2013-05-10 20:59               ` Paolo Bonzini
2013-05-10 23:06                 ` Anthony Liguori
2013-05-11  7:59                   ` Paolo Bonzini
2013-05-11  7:23                 ` Aurelien Jarno
2013-05-10 14:27     ` Anthony Liguori
2013-05-10 14:46       ` Aurelien Jarno
2013-05-10 15:04         ` Andreas Färber
2013-05-10 15:56 ` Aurelien Jarno
2013-05-10 16:18   ` Andreas Färber
2013-05-10 16:40     ` Paolo Bonzini
2013-05-10 18:50   ` Anthony Liguori
2013-05-13 16:46 ` Anthony Liguori

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=518CF130.2030402@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=afaerber@suse.de \
    --cc=aliguori@us.ibm.com \
    --cc=aurelien@aurel32.net \
    --cc=mst@redhat.com \
    --cc=qemu-devel@nongnu.org \
    /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;
as well as URLs for NNTP newsgroup(s).