All of lore.kernel.org
 help / color / mirror / Atom feed
From: Benjamin Drung <benjamin.drung-EIkl63zCoXaH+58JC4qpiA@public.gmane.org>
To: Jason Gunthorpe
	<jgunthorpe-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>,
	Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	Ram Amrani <Ram.Amrani-YGCgFSpz5w/QT0dZR+AlfA@public.gmane.org>,
	Steve Wise
	<swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>,
	Ira Weiny <ira.weiny-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>,
	Jarod Wilson <jarod-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
Subject: Re: [PATCH v2 rdma-core 0/6] Common udev/systemd based module auto loading
Date: Fri, 04 Aug 2017 18:22:31 +0200	[thread overview]
Message-ID: <1501863751.3581.19.camel@profitbricks.com> (raw)
In-Reply-To: <20170803175242.GA31628-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>

Am Donnerstag, den 03.08.2017, 11:52 -0600 schrieb Jason Gunthorpe:
> On Thu, Aug 03, 2017 at 02:00:59PM +0200, Benjamin Drung wrote:
> 
> > Yes. That was the issue. I am using an initrd (from initramfs-
> > tools).
> > The mlx4_core module was included in the initrd, but mlx4_ib was
> > missing. I tweaked initramfs-tools to include mlx4_ib and then it
> > was
> > autoloaded. Thanks.
> 
> Do you know how to fix this generally for Debian? Is it possible?

Analysis for Debian jessie/stretch:

$ sudo update-initramfs -uv
[...]
Copying module directory kernel/drivers/net
(excluding appletalk arcnet bonding can dummy.ko hamradio hippi ifb.ko
irda macvlan.ko macvtap.ko pcmcia sb1000.ko team tokenring tun.ko usb
veth.ko wan wimax wireless xen-netback.ko)
[...]
Adding module /lib/modules/$version/kernel/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.ko
Adding module /lib/modules/$version/kernel/drivers/net/ethernet/mellanox/mlx4/mlx4_core.ko

update-initramfs calls auto_add_modules() in
/usr/share/initramfs-tools/hook-functions which includes 'net' as
default. The 'net' option calls 
"copy_modules_dir kernel/drivers/net [...]" which copies the
mlx5_core.ko and mlx4_core.ko modules.

but the mlx4_ib.ko module is in the infiniband directory:

$ find kernel/drivers/infiniband ! -type d
kernel/drivers/infiniband/hw/mlx5/mlx5_ib.ko
kernel/drivers/infiniband/hw/mthca/ib_mthca.ko
kernel/drivers/infiniband/hw/mlx4/mlx4_ib.ko
kernel/drivers/infiniband/core/ib_sa.ko
kernel/drivers/infiniband/core/ib_umad.ko
kernel/drivers/infiniband/core/ib_netlink.ko
kernel/drivers/infiniband/core/ib_core.ko
kernel/drivers/infiniband/core/rdma_cm.ko
kernel/drivers/infiniband/core/ib_ucm.ko
kernel/drivers/infiniband/core/ib_addr.ko
kernel/drivers/infiniband/core/rdma_ucm.ko
kernel/drivers/infiniband/core/ib_cm.ko
kernel/drivers/infiniband/core/ib_uverbs.ko
kernel/drivers/infiniband/core/iw_cm.ko
kernel/drivers/infiniband/core/ib_mad.ko
kernel/drivers/infiniband/ulp/ipoib/ib_ipoib.ko
kernel/drivers/infiniband/ulp/srp/ib_srp.ko

By default, nothing from the infiniband directory is included. The
question is: What is the right thing to do? Should everything from
kernel/drivers/infiniband/hw be included in the initrd by default?

> > Remaining problem: The IPoIB InfiniBand partition does not come up
> > correctly:
> 
> AFAIK, M-DOWN means 'master-down', it means that the interface 'ib1'
> is down. 'ip link set up dev ib1'
> 
> Something wrong with ordering of the legacy network setup script?

Thanks. The master-down is correct, since we just use only the
InifiniBand partition and not the master device. Even the old
configuration had M-DOWN.

My ping tests were incorrect. I compared the IP configuration with the
previous configuration and it was identical. Then I discovered that the
ping target was not reachable even with the old configuration. I took a
different host that was pingable with the old config and this one was
pingable with rdma-core too. So rdma-core works as designed.

-- 
Benjamin Drung
System Developer
Debian & Ubuntu Developer

ProfitBricks GmbH
Greifswalder Str. 207
D - 10405 Berlin

Email: benjamin.drung-EIkl63zCoXaH+58JC4qpiA@public.gmane.org
Web: https://www.profitbricks.com

Sitz der Gesellschaft: Berlin.
Registergericht: Amtsgericht Charlottenburg, HRB 125506B.
Geschäftsführer: Achim Weiss.

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

  parent reply	other threads:[~2017-08-04 16:22 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-07-27 22:31 [PATCH v2 rdma-core 0/6] Common udev/systemd based module auto loading Jason Gunthorpe
     [not found] ` <1501194685-9447-1-git-send-email-jgunthorpe-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2017-07-27 22:31   ` [PATCH v2 rdma-core 1/6] Common infrastructure for auto loading rdma modules Jason Gunthorpe
     [not found]     ` <1501194685-9447-2-git-send-email-jgunthorpe-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2017-07-28 21:48       ` Chien Tin Tung
     [not found]         ` <20170728214855.GA13840-TZeIlv3TuzOfrEmaQUPKxl95YUYmaKo1UNDiOz3kqAs@public.gmane.org>
2017-07-28 23:01           ` Jason Gunthorpe
2017-07-27 22:31   ` [PATCH v2 rdma-core 2/6] Move umad udev rules to rdma-core Jason Gunthorpe
2017-07-27 22:31   ` [PATCH v2 rdma-core 3/6] debian: Cause udev to evaluate the installed rules after install Jason Gunthorpe
2017-07-27 22:31   ` [PATCH v2 rdma-core 4/6] redhat: Remove cxgb3/4.modprobe files Jason Gunthorpe
2017-07-27 22:31   ` [PATCH v2 rdma-core 5/6] srp_daemon: Autoload the SRP kernel module if required Jason Gunthorpe
2017-07-27 22:31   ` [PATCH v2 rdma-core 6/6] iwpmd: Autoload iwpmd " Jason Gunthorpe
2017-07-28 13:54   ` [PATCH v2 rdma-core 0/6] Common udev/systemd based module auto loading Steve Wise
2017-07-28 16:11     ` Jason Gunthorpe
2017-07-28 15:42   ` Benjamin Drung
     [not found]     ` <1501256525.4572.44.camel-EIkl63zCoXaH+58JC4qpiA@public.gmane.org>
2017-07-28 15:47       ` Jason Gunthorpe
     [not found]         ` <20170728154749.GA9646-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2017-07-28 16:41           ` Benjamin Drung
     [not found]             ` <1501260065.4572.47.camel-EIkl63zCoXaH+58JC4qpiA@public.gmane.org>
2017-07-28 16:56               ` Jason Gunthorpe
     [not found]                 ` <20170728165614.GA14302-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2017-08-03 12:00                   ` Benjamin Drung
     [not found]                     ` <1501761659.3581.2.camel-EIkl63zCoXaH+58JC4qpiA@public.gmane.org>
2017-08-03 17:52                       ` Jason Gunthorpe
     [not found]                         ` <20170803175242.GA31628-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2017-08-04 16:22                           ` Benjamin Drung [this message]
     [not found]                             ` <1501863751.3581.19.camel-EIkl63zCoXaH+58JC4qpiA@public.gmane.org>
2017-08-08 22:38                               ` Jason Gunthorpe
     [not found]                                 ` <20170808223851.GE29372-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2017-08-09 17:43                                   ` Benjamin Drung
     [not found]                                     ` <1502300630.1259.4.camel-EIkl63zCoXaH+58JC4qpiA@public.gmane.org>
2017-08-09 19:54                                       ` Jason Gunthorpe
     [not found]                                         ` <20170809195430.GB7013-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2017-08-09 20:00                                           ` Jason Gunthorpe
     [not found]                                             ` <20170809200040.GA7167-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2017-08-10  7:55                                               ` Benjamin Drung
2017-08-09 17:54                               ` Doug Ledford

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=1501863751.3581.19.camel@profitbricks.com \
    --to=benjamin.drung-eikl63zcoxah+58jc4qpia@public.gmane.org \
    --cc=Ram.Amrani-YGCgFSpz5w/QT0dZR+AlfA@public.gmane.org \
    --cc=dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
    --cc=ira.weiny-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
    --cc=jarod-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
    --cc=jgunthorpe-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org \
    --cc=leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
    --cc=linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.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 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.