All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Anders K. Pedersen | Cohaesio" <akp@cohaesio.com>
To: "pstaszewski@itcare.pl" <pstaszewski@itcare.pl>,
	"alexander.duyck@gmail.com" <alexander.duyck@gmail.com>
Cc: "netdev@vger.kernel.org" <netdev@vger.kernel.org>,
	"pavlos.parissis@gmail.com" <pavlos.parissis@gmail.com>,
	"intel-wired-lan@lists.osuosl.org"
	<intel-wired-lan@lists.osuosl.org>,
	"alexander.h.duyck@intel.com" <alexander.h.duyck@intel.com>
Subject: Re: Linux 4.12+ memory leak on router with i40e NICs
Date: Thu, 19 Oct 2017 12:19:42 +0000	[thread overview]
Message-ID: <1508415582.1186.49.camel@cohaesio.com> (raw)
In-Reply-To: <CAKgT0Ud=GRrsNtoM0gOgzsh1uPJ7nuT9S-xbigmbUMi+iz-ufw@mail.gmail.com>

Hi Alex,

On ons, 2017-10-18 at 16:37 -0700, Alexander Duyck wrote:
> When we last talked I had asked if you could do a git bisect to find
> the memory leak and you said you would look into it. The most useful
> way to solve this would be to do a git bisect between your current
> kernel and the 4.11 kernel to find the point at which this started.
> If
> we can do that then fixing this becomes much simpler as we just have
> to fix the patch that introduced the issue.

We're also seeing a smaller memory leak (about 1 GB per day) than the
original one even with the "Fix memory leak related filter programming
status" fix applied. So far I've determined that the leak is present on
4.13.7 and was introduced between 4.11 and 4.12, so I'll do another
round of bisection to identify the patch that introduced this.

Since the router must run for a couple of hours before I can be sure
whether a kernel is good or bad, and I can't reboot it during working
hours, it'll probably be about a week before I have a result.

-- 
Venlig hilsen / Best Regards

Anders K. Pedersen
Senior Technical Manager

Cohaesio A/S                          email:  akp@cohaesio.com
Parken | Per Henrik Lings Allé 4, 4   main:   +45 45 88 08 88
DK-2100 Copenhagen | Denmark          direct: +45 45 20 78 11

Visit us at www.cohaesio.com

Cohaesio is a leading IT solutions provider in Scandinavia, providing
certified IT outsourcing, service desk and managed hosting solutions to
professional customers.

WARNING: multiple messages have this Message-ID (diff)
From: Anders K. Pedersen | Cohaesio <akp@cohaesio.com>
To: intel-wired-lan@osuosl.org
Subject: [Intel-wired-lan] Linux 4.12+ memory leak on router with i40e NICs
Date: Thu, 19 Oct 2017 12:19:42 +0000	[thread overview]
Message-ID: <1508415582.1186.49.camel@cohaesio.com> (raw)
In-Reply-To: <CAKgT0Ud=GRrsNtoM0gOgzsh1uPJ7nuT9S-xbigmbUMi+iz-ufw@mail.gmail.com>

Hi Alex,

On ons, 2017-10-18 at 16:37 -0700, Alexander Duyck wrote:
> When we last talked I had asked if you could do a git bisect to find
> the memory leak and you said you would look into it. The most useful
> way to solve this would be to do a git bisect between your current
> kernel and the 4.11 kernel to find the point at which this started.
> If
> we can do that then fixing this becomes much simpler as we just have
> to fix the patch that introduced the issue.

We're also seeing a smaller memory leak (about 1 GB per day) than the
original one even with the "Fix memory leak related filter programming
status" fix applied. So far I've determined that the leak is present on
4.13.7 and was introduced between 4.11 and 4.12, so I'll do another
round of bisection to identify the patch that introduced this.

Since the router must run for a couple of hours before I can be sure
whether a kernel is good or bad, and I can't reboot it during working
hours, it'll probably be about a week before I have a result.

-- 
Venlig hilsen / Best Regards

Anders K. Pedersen
Senior Technical Manager

Cohaesio A/S                          email:  akp at cohaesio.com
Parken | Per Henrik Lings All? 4, 4   main:   +45 45 88 08 88
DK-2100 Copenhagen | Denmark          direct: +45 45 20 78 11

Visit us at www.cohaesio.com

Cohaesio is a leading IT solutions provider in Scandinavia, providing
certified IT outsourcing, service desk and managed hosting solutions to
professional customers.

  parent reply	other threads:[~2017-10-19 12:19 UTC|newest]

Thread overview: 74+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-10-04 12:56 Linux 4.12+ memory leak on router with i40e NICs Anders K. Pedersen | Cohaesio
2017-10-04 12:56 ` [Intel-wired-lan] " Anders K. Pedersen | Cohaesio
2017-10-04 15:32 ` Alexander Duyck
2017-10-04 15:32   ` [Intel-wired-lan] " Alexander Duyck
2017-10-05  5:19   ` Anders K. Pedersen | Cohaesio
2017-10-05  5:19     ` [Intel-wired-lan] " Anders K. Pedersen | Cohaesio
2017-10-14 22:00     ` =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-14 22:03       ` =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-15  0:58         ` Alexander Duyck
2017-10-15  0:58           ` [Intel-wired-lan] " Alexander Duyck
2017-10-15 15:03           ` Paweł Staszewski
2017-10-15 15:03             ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-16 11:20           ` Pavlos Parissis
2017-10-16 11:20             ` [Intel-wired-lan] " Pavlos Parissis
2017-10-16 14:11             ` Alexander Duyck
2017-10-16 14:11               ` [Intel-wired-lan] " Alexander Duyck
2017-10-16 16:26             ` Paweł Staszewski
2017-10-16 16:26               ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-16 23:34               ` Paweł Staszewski
2017-10-16 23:34                 ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-16 23:56                 ` Alexander Duyck
2017-10-16 23:56                   ` [Intel-wired-lan] " Alexander Duyck
2017-10-17  0:44                   ` Paweł Staszewski
2017-10-17  0:44                     ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-17  9:48                     ` Paweł Staszewski
2017-10-17  9:48                       ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-17 10:20                       ` Paweł Staszewski
2017-10-17 10:20                         ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-17 10:51                         ` Paweł Staszewski
2017-10-17 10:51                           ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-17 10:59                           ` Paweł Staszewski
2017-10-17 10:59                             ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-17 11:05                             ` Paweł Staszewski
2017-10-17 11:05                               ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-17 11:52                               ` Paweł Staszewski
2017-10-17 11:52                                 ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-17 14:08                                 ` Paweł Staszewski
2017-10-17 14:08                                   ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-18 15:44                                   ` Paweł Staszewski
2017-10-18 15:44                                     ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-18 22:20                                     ` Paweł Staszewski
2017-10-18 22:20                                       ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-18 22:50                                       ` Paweł Staszewski
2017-10-18 22:50                                         ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-18 22:58                                         ` Paweł Staszewski
2017-10-18 22:58                                           ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-18 23:22                                           ` Paweł Staszewski
2017-10-18 23:22                                             ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-18 23:37                                             ` Alexander Duyck
2017-10-18 23:37                                               ` [Intel-wired-lan] " Alexander Duyck
2017-10-18 23:51                                               ` Paweł Staszewski
2017-10-18 23:51                                                 ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-18 23:56                                                 ` Paweł Staszewski
2017-10-18 23:56                                                   ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-18 23:59                                                   ` Paweł Staszewski
2017-10-18 23:59                                                     ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-19 17:10                                                 ` Alexander Duyck
2017-10-19 17:10                                                   ` [Intel-wired-lan] " Alexander Duyck
2017-10-19 12:19                                               ` Anders K. Pedersen | Cohaesio [this message]
2017-10-19 12:19                                                 ` Anders K. Pedersen | Cohaesio
2017-10-19 15:40                                                 ` Alexander Duyck
2017-10-19 15:40                                                   ` [Intel-wired-lan] " Alexander Duyck
2017-10-22 13:56                                                   ` Anders K. Pedersen | Cohaesio
2017-10-22 13:56                                                     ` [Intel-wired-lan] " Anders K. Pedersen | Cohaesio
2017-10-17  5:51                 ` Vitezslav Samel
2017-10-17  5:51                   ` [Intel-wired-lan] " Vitezslav Samel
2017-10-18 23:29                   ` Alexander Duyck
2017-10-18 23:29                     ` [Intel-wired-lan] " Alexander Duyck
2017-10-18 23:40                     ` Paweł Staszewski
2017-10-18 23:40                       ` [Intel-wired-lan] " =?unknown-8bit?q?Pawe=C5=82?= Staszewski
2017-10-19 11:41                       ` Pavlos Parissis
2017-10-19 11:41                         ` [Intel-wired-lan] " Pavlos Parissis
2017-10-19 15:53                         ` Alexander Duyck
2017-10-19 15:53                           ` [Intel-wired-lan] " Alexander Duyck

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=1508415582.1186.49.camel@cohaesio.com \
    --to=akp@cohaesio.com \
    --cc=alexander.duyck@gmail.com \
    --cc=alexander.h.duyck@intel.com \
    --cc=intel-wired-lan@lists.osuosl.org \
    --cc=netdev@vger.kernel.org \
    --cc=pavlos.parissis@gmail.com \
    --cc=pstaszewski@itcare.pl \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.