6:12 a.m.
On 2014/03/10, 6:01 AM, "E.S. Rosenberg"
<esr@cs.huji.ac.il<mailto:esr@cs.huji.ac.il>> wrote:
On Tue, Mar 4, 2014 at 4:04 PM, Simmons, James A.
>If yes, you most likely stumbled on
https://jira.hpdd.intel.com/browse/LU-4209
We are running the in-kernel client, kernel 3.13.x on Debian machines
The lfs utility in use is from the lustre 2.4.2 branch
I see that a patch was submitted to the linux-kernel team, does anyone have a link to that
patch? I'd like to or patch our kernel or see what version of the kernel we should be
building, if it made >it into 3.14-rcX then we'll build that...
http://www.spinics.net/lists/linux-fsdevel/msg72386.html
Correct me if I'm wrong but it looks like that patch hit a dead end...
Andreas did any fsdevs ever voice an opinion on generic/lustre flags?
The patch was resubmitted, and Al Viro had relatively little to say about it (i.e. he
didn't completely hate it, which is good). I've since sent an updated patch to
Greg again.
Any chance it'll be in 3.14?
The patch is currently in the staging tree, I'm not sure which release it will be in.
Cheers, Andreas
--
Andreas Dilger
Lustre Software Architect
Intel High Performance Data Division
10:29 p.m.
New subject: iozone and lustre
On Mon, Mar 10, 2014 at 9:12 PM, Dilger, Andreas
<andreas.dilger(a)intel.com>wrote:
On 2014/03/10, 6:01 AM, "E.S. Rosenberg"
<esr(a)cs.huji.ac.il<mailto:
esr(a)cs.huji.ac.il>> wrote:
On Tue, Mar 4, 2014 at 4:04 PM, Simmons, James A.
>>If yes, you most likely stumbled on
https://jira.hpdd.intel.com/browse/LU-4209
>We are running the in-kernel client, kernel 3.13.x on Debian machines
>The lfs utility in use is from the lustre 2.4.2 branch
>I see that a patch was submitted to the linux-kernel team, does anyone
have a link to that patch? I'd like to or patch our kernel or see what
version of the kernel we should be building, if it made >it into 3.14-rcX
then we'll build that...
http://www.spinics.net/lists/linux-fsdevel/msg72386.html
Correct me if I'm wrong but it looks like that patch hit a dead end...
Andreas did any fsdevs ever voice an opinion on generic/lustre flags?
The patch was resubmitted, and Al Viro had relatively little to say about
it (i.e. he didn't completely hate it, which is good). I've since sent an
updated patch to Greg again.
Any chance it'll be in 3.14?
The patch is currently in the staging tree, I'm not sure which release it
will be in.
Do I need an updated version of lustre once I have a patched kernel or is
this transparent to lustre 2.4.2?
Thanks,
Eli
Cheers, Andreas
--
Andreas Dilger
Lustre Software Architect
Intel High Performance Data Division
_______________________________________________
HPDD-discuss mailing list
HPDD-discuss(a)lists.01.org
https://lists.01.org/mailman/listinfo/hpdd-discuss
4:55 p.m.
New subject: iozone and lustre
On 2014/03/11, 5:29 AM, "E.S. Rosenberg"
<esr+hpdd-discuss@mail.hebrew.edu<mailto:esr+hpdd-discuss@mail.hebrew.edu>>
wrote:
On Mon, Mar 10, 2014 at 9:12 PM, Dilger, Andreas
<andreas.dilger@intel.com<mailto:andreas.dilger@intel.com>> wrote:
On 2014/03/10, 6:01 AM, "E.S. Rosenberg"
<esr@cs.huji.ac.il<mailto:esr@cs.huji.ac.il><mailto:esr@cs.huji.ac.il<mailto:esr@cs.huji.ac.il>>>
wrote:
On Tue, Mar 4, 2014 at 4:04 PM, Simmons, James A.
>If yes, you most likely stumbled on
https://jira.hpdd.intel.com/browse/LU-4209
We are running the in-kernel client, kernel 3.13.x on Debian machines
The lfs utility in use is from the lustre 2.4.2 branch
The patch was resubmitted, and Al Viro had relatively little to say about it (i.e. he
didn't completely hate it, which is good). I've since sent an updated patch to
Greg again. Any chance it'll be in 3.14?
The patch is currently in the staging tree, I'm not sure which release it will be in.
Do I need an updated version of lustre once I have a patched kernel or is this transparent
to lustre 2.4.2?
The Lustre userspace patch needed to work with the newest kernel (when it arrives, looks
like it will be 3.15) has landed for Lustre 2.5.3 and 2.6.0. It will also be in 2.4.4
when that is released. Once you have a newer lfs from one of those branches, it will work
on both older (e.g. 2.6.32) and newer kernels (3.15+), excluding the 3.11-3.14 range
unless the fix is also landed into the -stable kernels.
Cheers, Andreas
--
Andreas Dilger
Lustre Software Architect
Intel High Performance Data Division
7:48 p.m.
New subject: iozone and lustre
On Fri, Mar 21, 2014 at 7:55 AM, Dilger, Andreas
<andreas.dilger(a)intel.com>wrote:
On 2014/03/11, 5:29 AM, "E.S. Rosenberg"
<esr+hpdd-discuss(a)mail.hebrew.edu
<mailto:esr+hpdd-discuss@mail.hebrew.edu>> wrote:
On Mon, Mar 10, 2014 at 9:12 PM, Dilger, Andreas <andreas.dilger(a)intel.com
<mailto:andreas.dilger@intel.com>> wrote:
On 2014/03/10, 6:01 AM, "E.S. Rosenberg" <esr(a)cs.huji.ac.il<mailto:
esr@cs.huji.ac.il><mailto:esr@cs.huji.ac.il<mailto:esr@cs.huji.ac.il>>>
wrote:
On Tue, Mar 4, 2014 at 4:04 PM, Simmons, James A.
>>If yes, you most likely stumbled on
https://jira.hpdd.intel.com/browse/LU-4209
>We are running the in-kernel client, kernel 3.13.x on Debian machines
>The lfs utility in use is from the lustre 2.4.2 branch
The patch was resubmitted, and Al Viro had relatively little to say about
it (i.e. he didn't completely hate it, which is good). I've since sent an
updated patch to Greg again. Any chance it'll be in 3.14?
The patch is currently in the staging tree, I'm not sure which release it
will be in.
Do I need an updated version of lustre once I have a patched kernel or is
this transparent to lustre 2.4.2?
The Lustre userspace patch needed to work with the newest kernel (when it
arrives, looks like it will be 3.15) has landed for Lustre 2.5.3 and 2.6.0.
It will also be in 2.4.4 when that is released. Once you have a newer lfs
from one of those branches, it will work on both older (e.g. 2.6.32) and
newer kernels (3.15+), excluding the 3.11-3.14 range unless the fix is also
landed into the -stable kernels.
Did the kernel patch make it in this merge-window?
Thanks,
Eli
Cheers, Andreas
--
Andreas Dilger
Lustre Software Architect
Intel High Performance Data Division
8:07 p.m.
New subject: iozone and lustre
FYI,
the O_LOV_DELAY_CREATE fix made it into the upstream 3.14 kernel. I just verified commit
38585ccc4 was in the 3.14-rc1 tag. The Lustre userspace tools for 2.5.1 and later should
work correctly with this kernel. There is also a patch pending for the Lustre 2.4.4
userspace tools.
Cheers, Andreas
On 2014/04/23, 3:48 AM, "E.S. Rosenberg"
<esr+hpdd-discuss@mail.hebrew.edu<mailto:esr+hpdd-discuss@mail.hebrew.edu>>
wrote:
On Fri, Mar 21, 2014 at 7:55 AM, Dilger, Andreas
<andreas.dilger@intel.com<mailto:andreas.dilger@intel.com>> wrote:
On 2014/03/11, 5:29 AM, "E.S. Rosenberg"
<esr+hpdd-discuss@mail.hebrew.edu<mailto:esr%2Bhpdd-discuss@mail.hebrew.edu><mailto:esr+hpdd-discuss@mail.hebrew.edu<mailto:esr%2Bhpdd-discuss@mail.hebrew.edu>>>
wrote:
On Mon, Mar 10, 2014 at 9:12 PM, Dilger, Andreas
<andreas.dilger@intel.com<mailto:andreas.dilger@intel.com><mailto:andreas.dilger@intel.com<mailto:andreas.dilger@intel.com>>>
wrote:
On 2014/03/10, 6:01 AM, "E.S. Rosenberg"
<esr@cs.huji.ac.il<mailto:esr@cs.huji.ac.il><mailto:esr@cs.huji.ac.il<mailto:esr@cs.huji.ac.il>><mailto:esr@cs.huji.ac.il<mailto:esr@cs.huji.ac.il><mailto:esr@cs.huji.ac.il<mailto:esr@cs.huji.ac.il>>>>
wrote:
On Tue, Mar 4, 2014 at 4:04 PM, Simmons, James A.
>If yes, you most likely stumbled on
https://jira.hpdd.intel.com/browse/LU-4209
We are running the in-kernel client, kernel 3.13.x on Debian machines
The lfs utility in use is from the lustre 2.4.2 branch
The patch was resubmitted, and Al Viro had relatively little to say about it (i.e. he
didn't completely hate it, which is good). I've since sent an updated patch to
Greg again. Any chance it'll be in 3.14?
The patch is currently in the staging tree, I'm not sure which release it will be in.
Do I need an updated version of lustre once I have a patched kernel or is this transparent
to lustre 2.4.2?
The Lustre userspace patch needed to work with the newest kernel (when it arrives, looks
like it will be 3.15) has landed for Lustre 2.5.3 and 2.6.0. It will also be in 2.4.4
when that is released. Once you have a newer lfs from one of those branches, it will work
on both older (e.g. 2.6.32) and newer kernels (3.15+), excluding the 3.11-3.14 range
unless the fix is also landed into the -stable kernels.
Did the kernel patch make it in this merge-window?
Cheers, Andreas
--
Andreas Dilger
Lustre Software Architect
Intel High Performance Data Division
3064
days inactive
3114
days old
4 comments
2 participants
participants (2)
-
Dilger, Andreas -
E.S. Rosenberg