On Fri, 17 Nov 2017, Alan Altmark wrote:
> So what you're saying, Sebastian, is that (a) fine-grained control over
> what devices interrupt what CPUs is not in the Z architecture, (b) it's
> not in there for a good reason, (c) providing such a capability would
> adversely affect overall system per
On Fri, 17 Nov 2017, Timothy Sipples wrote:
> Sebastian Ott wrote:
> >Setting irq affinity is currently not supported on s390.
>
> This platform's I/O architecture is unique. What are the real-world issues
> associated with not being able to set this parameter on this platf
James,
On Thu, 16 Nov 2017, Jingmin Zhai wrote:
> Recently we're testing Mellanox 10GbE performance with Ubuntu 17.04 s390x
> on z13. During the test, we found that interrupt affinity cannot be set
> like other platforms.
>
> We stopped the 'irqbalance' service first, then
>
> echo > /proc/irq//
On Fri, 19 Aug 2016, Mark Pace wrote:
> I'm not sure what Sebastian means by grouping them.
Channel IO based network devices need to be grouped in order to use them.
# echo "0.0.3d00,0.0.3d01,0.0.3d02" > /sys/bus/ccwgroup/drivers/qeth/group
After that you can enable the group device:
# echo 1 >
Hello Mark,
On Fri, 19 Aug 2016, Mark Pace wrote:
> Trying to dynamically add a network port.
> Granted linux access to the vswitch
>
> set vswitch vswbpic grant sles001
> Command complete
>
> Then defined the virtual nic to linux
> sles001:~ # vmcp define nic 3d00 type qdio dev 3
> NIC 3D00 is cr
On Fri, 14 Aug 2015, Agblad Tore wrote:
> Anyone knows if this is possible yet ?
zEDC can be used from within Linux with the genwqe device driver.
genwqe is available in RHEL7.1 and SLES12.
Details can be found here:
https://www.ibm.com/developerworks/community/files/form/anonymous/api/library/f
On Fri, 7 Aug 2015, Agblad Tore wrote:
> Hi, is it possible to use the Flashdisks inside the z-box for LVM:s in the
> Linux server ?
> We have z/VM 6.3, RHEL7 is the version we want to use it for.
> The box is (two) z12.
> HDS disk.
Flash Express can be used in Linux via the scm_block driver. But
On Thu, 26 Mar 2015, Offer Baruch wrote:
> Did you remove the devices from the cio ignore list? Try this before the
> group command:
> cio_ignore -r 2e60-26e2
> chccwdev -e 2e60-26e2
In such cases I would do the following:
* check if these devices are known to Linux (lscss -d 2e60-26e2)
* if not t
On Wed, 25 Mar 2015, Neale Ferguson wrote:
> I have come across some strange behavior with yum under RHEL 7. The
> simplest test case I have created is as follows (in reality I was doing
> something more complex with a yum install):
>
> yum list bash
>
> Vs
>
> yum --installroot=/tmp list bash
>
>
On Tue, 16 Dec 2014, Michael MacIsaac wrote:
> Gerhard,
>
> Thanks for the info.
>
> I see with interest in the s390-tools package:
> Add -s/--safeoffline feature
>
> Hmm, what's this? (does this imply the default is an unsafe offline? :))
If you made sure that you are no longer using the devic
On Thu, 4 Sep 2014, Neale Ferguson wrote:
> Sebastian, at what point and by what bit of software does the change of
> condev to an actual device address?
This all happens in the Linux Kernel early during IPL. The translation
from "condev" to the actual device number takes place when we parse the
0.00] cio: |condev is not a valid device for the cio_ignore
> > > kernel parameter
> > it is definitely a wrong character typed in the cio_ignore option, |
> > (pipe) instead of ! (exclamation mark)
> >
> >
> > Dan
> >
> > >
> > > /Tore
> > &
clamation mark)
>
>
> Dan
>
> >
> >
> > /Tore
> >
> > -Original Message-
> > From: Linux on 390 Port [mailto:LINUX-390@VM.MARIST.EDU] On Behalf Of
> > Sebastian Ott Sent: den 3 september 2014 3:58
> > To: LINUX-390@V
n
> Well, RedHat is working on it and they have been able to reproduce the error.
>
> /Tore Agblad
>
> -Original Message-
> From: Linux on 390 Port [mailto:LINUX-390@VM.MARIST.EDU] On Behalf Of
> Sebastian Ott
> Sent: den 21 augusti 2014 4:14
> To: LINUX-390@VM
On Thu, 21 Aug 2014, Sebastian Ott wrote:
> Hi,
>
> On Thu, 21 Aug 2014, Agblad Tore wrote:
> > Actually it's supposed to be a bug, and is reported to RedHat.
> >
> > You did get some complaints:
> >
> > ap='se' inst.sshd
st of my knowledge this is not a
rhel7 kernel.
Regards,
Sebastian
>
> Also more than 4 disks crashed the installation at that disksetup :-(
>
> BR /Tore
>
> -Original Message-
> From: Linux on 390 Port [mailto:LINUX-390@VM.MARIST.EDU] On Behalf Of
> Sebastian Ott
> Se
On Fri, 15 Aug 2014, Sebastian Ott wrote:
> On Fri, 8 Aug 2014, Agblad Tore wrote:
> > We just solved that installation problem.
> >
> > There is a recommendation to use cio_ignore=all,!condev in the GENERIC PRM
> > file used at booting the install.
> > But t
Hello,
On Fri, 8 Aug 2014, Agblad Tore wrote:
> We just solved that installation problem.
>
> There is a recommendation to use cio_ignore=all,!condev in the GENERIC PRM
> file used at booting the install.
> But this also rules out the console on 0009 that is needed at the end of the
> installat
NNEL =
> Ready; T=0.01/0.01 09:01:46
Nothing special here, no reason why specifying !condev for cio_ignore=
shouldn't work.
Regards,
Sebastian
>
> BR /Tore
>
>
> -Original Message-
> From: Linux on 390 Port [mailto:LINUX-390@VM.MARIST.EDU] On Behalf Of
> Sebas
Hello Dan,
On Fri, 8 Aug 2014, Dan HorĂ¡k wrote:
> On Fri, 8 Aug 2014 11:51:48 +
> Agblad Tore wrote:
>
> > We just solved that installation problem.
> >
> > There is a recommendation to use cio_ignore=all,!condev in the GENERIC PRM
> > file used at booting the install.
> > But this also ru
On Fri, 8 Aug 2014, Agblad Tore wrote:
> There is a recommendation to use cio_ignore=all,!condev in the GENERIC PRM
> file used at booting the install.
> But this also rules out the console on 0009 that is needed at the end of the
> installation.
> It should be: "cio_ignore=all,!0.0009"
>
> Then
On Tue, 9 Jul 2013, Rick Barlow wrote:
> 2) Why doesn't Linux receive and handle pathing issues when they occur?
It does (after receiving a notification from the hardware/firmware or
hypervisor, or after beeing triggered by the admin). But it looks like it
didn't receive a notification that someth
On Thu, 25 Oct 2012, Marcy Cortes wrote:
> Yes! We are using LVM extensively.
> So cfq is being used on LVM by default then?
Nope. As explained the default IO scheduler on sles11 on z for all block
devices is deadline (unless you change the default; DASD being the
exception using deadline regardl
Hi,
On Wed, 24 Oct 2012, Marcy Cortes wrote:
> Hello listers,
>
> We've hit a pretty nasty bug in SLES 11 SP2 kernel at 3.0.34 and higher.
> Basically the I/O goes crazy - like 4000-6000 i/o per sec - and the server
> becomes unresponsive. It's highly creatable by running a WebSphere deploy o
On Tue, 14 Aug 2012, Alan Altmark wrote:
> On Tuesday, 08/14/2012 at 05:35 EDT, Neale Ferguson
> > [root@serv-srv1 ~]# ls /sys/devices/css0/
> > 0.0. 0.0.0002 0.0.0004 0.0.0006 0.0.0008 chp0.01 chp0.a0
> chp0.f0 defunct uevent
> > 0.0.0001 0.0.0003 0.0.0005 0.0.0007 chp0.00 chp0.02
On Tue, 14 Aug 2012, Neale Ferguson wrote:
> Happened again, this time with device 0x1001:
>
> vmcp link USER0101 0316 1001 W
> [root@serv1-srv1 ~]# ls /sys/devices/css0/
> 0.0./ 0.0.0002/ 0.0.0004/ 0.0.0006/ chp0.00/ chp0.02/ chp0.ef/
> chp0.f1/ power/
> 0.0.0001/ 0.0.0003/ 0.0.0005/ 0.0.00
On Tue, 14 Aug 2012, Mauro Souza wrote:
> Looks like udevadm settle would solve this, like the long story of errors
> dasdfmt->fdasd->mkfs we saw one of those days on the list...
> Could you try a udevadm settle after the LINK?
Nope. Looks more like linux would not recognize this device at all.
If
On Tue, 14 Aug 2012, Neale Ferguson wrote:
> An app of ours will link to a disk, bring it online, check its contents,
> take it offline, and then detach it. This works pretty well all the time but
> on occasion the device gets linked but nothing appears in the /sys tree nor
> a udev event generated
On Tue, 31 Jul 2012, David Boyes wrote:
> Florian's original post.
> Corroborating posts from other users (Mark Post, etc)
> My data (on average 3 out of 100 tests fail)
>
> I'd be happy to send you more examples. Are you looking for something
> specific? The script recently posted here (by you, I
On Sun, 29 Jul 2012, David Boyes wrote:
> Since we have ample data from multiple sources that this DOES NOT operate
> reliably, the original question still stands.
Would you mind sharing some of this ample data? Are this all cases where
dasdfmt complains about other users after "udevadm settle" r
On Sat, 28 Jul 2012, Michael MacIsaac wrote:
> function enableDevice { chccwdev -e $1; udevadm settle; }
>
> and always call that function instead chccwdev -e. So my question is
> still: "If a udevadm settle is always required after a chccwdev -e, then
> why is it not just built into the command?
On Fri, 27 Jul 2012, Florian Bilek wrote:
> I can confirm that the udev settle is returning always with zero. I have
> the timeout set to even to 60 sec and an exit if the device node is
> available. And that is stlll not enough because udev exits. Without exit I
> had set the timeout to 30 secs. O
On Fri, 27 Jul 2012, David Boyes wrote:
> > On which distro do you have problems with chccwdev?
> > [snip]
> > ..which appears to work fine.
>
> It's not a distribution issue; it's a timing-dependent issue that has to do
> with how quickly your hardware responds. Your script will work *most* of th
On Thu, 26 Jul 2012, David Boyes wrote:
> A week or two back, someone (I think it was Florian Bilek) asked why there
> was a delay between invoking chccwdev and the device becoming available, and
> whether there was an option or command that would exit only when the device
> was actually availab
Lee,
On Fri, 30 Mar 2012, Lee Stewart wrote:
> Hi all,
> I've been trying to think of any reason to ever have cio_ignore in a VM
> guest. I can see real use for it in an LPAR where you may have
> thousands of devices that have nothing to do with the Linux instance.
> But in a virtual machine I o
On Tue, 14 Dec 2010, Mark Post wrote:
> >>> On 12/14/2010 at 04:13 PM, Marcy Cortes
> >>> wrote:
>
> > Don't you have to bring online the c01 and c02 devices too
> > chccwdev -e 0.0.0c00-0.0.0c02
>
> No, you can only bring the lowest address of the three devices that have been
> grouped toge
Hi Florian,
On Mon, 22 Nov 2010, Florian Bilek wrote:
> Dear all
>
> Since some weeks I am facing some weird problem with SLES 11 SP1. I am on
> the current patch level with that image (I hope), the kernel is:
>
> # uname -a
> Linux sles11 2.6.32.24-0.2-default #1 SMP 2010-10-29 16:39:49 +0200 s39
Hi,
On Wed, 1 Sep 2010, Vlacil, Radek wrote:
> Is there any way how to detect in an application that 34xx tape was
> attached through ccw. In other words, detect that new directory for that
> tape appeared in /sys/bus/ccw/devices/ and that it is ready for
> "chccwdev -e". I would like to have a pr
Hi Michael,
On Fri, 23 Apr 2010, Michael MacIsaac wrote:
> It seemed that the chccwdev command finished asynchronously.
chccwdev does not finish asynchronously.
If chccwdev returns the device driver completed its online/offline
processing. The problem you try to solve is, that some things are han
39 matches
Mail list logo