[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#817063: marked as done (Severe performance regression in Infiniband or IPoIB)

Your message dated Thu, 30 Mar 2017 01:44:56 +0100
with message-id <1490834696.2634.64.camel@decadent.org.uk>
and subject line Re: Severe performance regression in Infiniband or IPoIB
has caused the Debian Bug report #817063,
regarding Severe performance regression in Infiniband or IPoIB
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
817063: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=817063
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message --- 
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA224

Package: linux-image-4.3.0-1-powerpc64le
Version: 4.3.5-1

When upgrading to the latest 4.3 testing kernel on a Mellanox MT25208
controller, all IPoIB links showed a severe performance regression as
compared with kernel 3.16. The bandwidth of each link dropped from around
3Gbps to around 3Mbps (3 orders of magnitude worse!); reverting to the
3.16 kernel immediately fixed this problem.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)

iFYEARELAAYFAlbdyVkACgkQLaxZSoRZrGEfeADgrTgu8+cP87QUzG3dlt7WzFE1
sZqh6KD/6fBW/wDgreyn3o9yKw5Dd61zf1ehZc7GjzvCJyL7NnDMEQ==
=h8Kk
-----END PGP SIGNATURE-----

--- End Message ---
--- Begin Message --- 
Version: 4.9~rc5-1~exp1

On Mon, 7 Mar 2016 12:32:57 -0600 "Timothy Pearson"
<kb9vqf@pearsoncomputing.net> wrote:
> Package: linux-image-4.3.0-1-powerpc64le
> Version: 4.3.5-1
> 
> When upgrading to the latest 4.3 testing kernel on a Mellanox MT25208
> controller, all IPoIB links showed a severe performance regression as
> compared with kernel 3.16. The bandwidth of each link dropped from around
> 3Gbps to around 3Mbps (3 orders of magnitude worse!); reverting to the
> 3.16 kernel immediately fixed this problem.

This was apparently fixed upstream by:

commit fc791b6335152c5278dc4a4991bcb2d329f806f9
Author: Paolo Abeni <pabeni@redhat.com>
Date:   Thu Oct 13 18:26:56 2016 +0200

    IB/ipoib: move back IB LL address into the hard header

Ben.

-- 
Ben Hutchings
In a hierarchy, every employee tends to rise to his level of
incompetence.

Attachment: signature.asc
Description: This is a digitally signed message part


--- End Message ---
Reply to: