[opensuse-kernel] Dirty COW fix causes some apps to freeze; patch is available in upstream
Hello, I run across a problem with an application using wine on OpenSUSE 42.1. After some search, I found out that it is a known bug and that there is an fix available: https://bugs.winehq.org/show_bug.cgi?id=41637 Basically the (backported) Dirty COW fix misses some adaptions in a second file and that leads to problems in some very special cases. For details please see the LKML thread with the fix: https://lkml.org/lkml/2017/1/5/13 Is there any chance that this patch will be backported to the OpenSUSE 42.1 Kernel too? If I have seen it right, the OpenSUSE 42.2 Kernel is also affected. If necessary I'm able to apply the patch and build a kernel myself but I have no idea what I would have to do to submit it to the official OpenSUSE kernel. Any hints? Kind regards Christian Mauderer -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
(adds a couple CCs) On Fri, 2017-03-17 at 22:09 +0100, Christian Mauderer wrote:
Hello,
I run across a problem with an application using wine on OpenSUSE 42.1. After some search, I found out that it is a known bug and that there is an fix available:
https://bugs.winehq.org/show_bug.cgi?id=41637
Basically the (backported) Dirty COW fix misses some adaptions in a second file and that leads to problems in some very special cases. For details please see the LKML thread with the fix:
https://lkml.org/lkml/2017/1/5/13
Is there any chance that this patch will be backported to the OpenSUSE 42.1 Kernel too? If I have seen it right, the OpenSUSE 42.2 Kernel is also affected.
Hm, it's tagged for stable, but doesn't seem to be taking wing. -Mike -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
On 03/17/2017 11:23 PM, Mike Galbraith wrote:
(adds a couple CCs)
On Fri, 2017-03-17 at 22:09 +0100, Christian Mauderer wrote:
Hello,
I run across a problem with an application using wine on OpenSUSE 42.1. After some search, I found out that it is a known bug and that there is an fix available:
https://bugs.winehq.org/show_bug.cgi?id=41637
Basically the (backported) Dirty COW fix misses some adaptions in a second file and that leads to problems in some very special cases. For details please see the LKML thread with the fix:
https://lkml.org/lkml/2017/1/5/13
Is there any chance that this patch will be backported to the OpenSUSE 42.1 Kernel too? If I have seen it right, the OpenSUSE 42.2 Kernel is also affected.
Hm, it's tagged for stable, but doesn't seem to be taking wing.
When this situation happens, a ping to GregKH usually solves the problem. Larry -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
On Sat, 2017-03-18 at 11:26 -0500, Larry Finger wrote:
On 03/17/2017 11:23 PM, Mike Galbraith wrote:
(adds a couple CCs)
On Fri, 2017-03-17 at 22:09 +0100, Christian Mauderer wrote:
Hello,
I run across a problem with an application using wine on OpenSUSE 42.1. After some search, I found out that it is a known bug and that there is an fix available:
https://bugs.winehq.org/show_bug.cgi?id=41637
Basically the (backported) Dirty COW fix misses some adaptions in a second file and that leads to problems in some very special cases. For details please see the LKML thread with the fix:
https://lkml.org/lkml/2017/1/5/13
Is there any chance that this patch will be backported to the OpenSUSE 42.1 Kernel too? If I have seen it right, the OpenSUSE 42.2 Kernel is also affected.
Hm, it's tagged for stable, but doesn't seem to be taking wing.
When this situation happens, a ping to GregKH usually solves the problem.
Sure, for kernels he maintains, and where backporting isn't required. An mm expert would know exactly where all the fix needs to go, and whether any of those spots require more than simple reject fixup. -Mike -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
On Sat 18-03-17 05:23:12, Mike Galbraith wrote:
(adds a couple CCs)
On Fri, 2017-03-17 at 22:09 +0100, Christian Mauderer wrote:
Hello,
I run across a problem with an application using wine on OpenSUSE 42.1. After some search, I found out that it is a known bug and that there is an fix available:
https://bugs.winehq.org/show_bug.cgi?id=41637
Basically the (backported) Dirty COW fix misses some adaptions in a second file and that leads to problems in some very special cases. For details please see the LKML thread with the fix:
https://lkml.org/lkml/2017/1/5/13
Is there any chance that this patch will be backported to the OpenSUSE 42.1 Kernel too? If I have seen it right, the OpenSUSE 42.2 Kernel is also affected.
Hm, it's tagged for stable, but doesn't seem to be taking wing.
Just for the reference, it is 8310d48b125d ("mm/huge_memory.c: respect FOLL_FORCE/FOLL_COW for thp"). SLE12-SP2 doesn't have this fix either. Could you push this to 3.12 stable Jiri, or should I post it there? Please note that I am travelling until thursday so I might be slow at responding. -- Michal Hocko SUSE Labs -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
On 03/19/2017, 01:58 PM, Michal Hocko wrote:
On Sat 18-03-17 05:23:12, Mike Galbraith wrote:
(adds a couple CCs)
On Fri, 2017-03-17 at 22:09 +0100, Christian Mauderer wrote:
Hello,
I run across a problem with an application using wine on OpenSUSE 42.1. After some search, I found out that it is a known bug and that there is an fix available:
https://bugs.winehq.org/show_bug.cgi?id=41637
Basically the (backported) Dirty COW fix misses some adaptions in a second file and that leads to problems in some very special cases. For details please see the LKML thread with the fix:
https://lkml.org/lkml/2017/1/5/13
Is there any chance that this patch will be backported to the OpenSUSE 42.1 Kernel too? If I have seen it right, the OpenSUSE 42.2 Kernel is also affected.
Hm, it's tagged for stable, but doesn't seem to be taking wing.
Just for the reference, it is 8310d48b125d ("mm/huge_memory.c: respect FOLL_FORCE/FOLL_COW for thp"). SLE12-SP2 doesn't have this fix either. Could you push this to 3.12 stable Jiri, or should I post it there?
Please provide a backport and send to stable@vger, it does not apply at all. thanks, -- js suse labs -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
Am 20.03.2017 um 12:55 schrieb Jiri Slaby:
On 03/19/2017, 01:58 PM, Michal Hocko wrote:
On Sat 18-03-17 05:23:12, Mike Galbraith wrote:
(adds a couple CCs)
On Fri, 2017-03-17 at 22:09 +0100, Christian Mauderer wrote:
Hello,
I run across a problem with an application using wine on OpenSUSE 42.1. After some search, I found out that it is a known bug and that there is an fix available:
https://bugs.winehq.org/show_bug.cgi?id=41637
Basically the (backported) Dirty COW fix misses some adaptions in a second file and that leads to problems in some very special cases. For details please see the LKML thread with the fix:
https://lkml.org/lkml/2017/1/5/13
Is there any chance that this patch will be backported to the OpenSUSE 42.1 Kernel too? If I have seen it right, the OpenSUSE 42.2 Kernel is also affected.
Hm, it's tagged for stable, but doesn't seem to be taking wing.
Just for the reference, it is 8310d48b125d ("mm/huge_memory.c: respect FOLL_FORCE/FOLL_COW for thp"). SLE12-SP2 doesn't have this fix either. Could you push this to 3.12 stable Jiri, or should I post it there?
Please provide a backport and send to stable@vger, it does not apply at all.
thanks,
Hello, the Ubuntu people ported the patch back to their kernel. It applies cleanly onto the openSUSE-42.1 branch of git://kernel.opensuse.org/kernel.git You can find this version of the patch here: http://kernel.ubuntu.com/git/ubuntu/ubuntu-xenial.git/commit/?id=9c49adf93cd... I only tried whether it applies but not whether it would build. Would it be useful for you if I would build a test kernel with the patch and try it on one of my machines? Kind regards Christian Mauderer -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
On 03/20/2017, 01:41 PM, Christian Mauderer wrote:
Am 20.03.2017 um 12:55 schrieb Jiri Slaby:
On 03/19/2017, 01:58 PM, Michal Hocko wrote:
On Sat 18-03-17 05:23:12, Mike Galbraith wrote:
(adds a couple CCs)
On Fri, 2017-03-17 at 22:09 +0100, Christian Mauderer wrote:
Hello,
I run across a problem with an application using wine on OpenSUSE 42.1. After some search, I found out that it is a known bug and that there is an fix available:
https://bugs.winehq.org/show_bug.cgi?id=41637
Basically the (backported) Dirty COW fix misses some adaptions in a second file and that leads to problems in some very special cases. For details please see the LKML thread with the fix:
https://lkml.org/lkml/2017/1/5/13
Is there any chance that this patch will be backported to the OpenSUSE 42.1 Kernel too? If I have seen it right, the OpenSUSE 42.2 Kernel is also affected.
Hm, it's tagged for stable, but doesn't seem to be taking wing.
Just for the reference, it is 8310d48b125d ("mm/huge_memory.c: respect FOLL_FORCE/FOLL_COW for thp"). SLE12-SP2 doesn't have this fix either. Could you push this to 3.12 stable Jiri, or should I post it there?
Please provide a backport and send to stable@vger, it does not apply at all.
thanks,
Hello,
the Ubuntu people ported the patch back to their kernel. It applies cleanly onto the openSUSE-42.1 branch of git://kernel.opensuse.org/kernel.git
You can find this version of the patch here:
http://kernel.ubuntu.com/git/ubuntu/ubuntu-xenial.git/commit/?id=9c49adf93cd...
I only tried whether it applies but not whether it would build. Would it be useful for you if I would build a test kernel with the patch and try it on one of my machines?
Neither that applies. And if I force that, it does not build in 3.12: mm/huge_memory.c: In function ‘can_follow_write_pmd’: mm/huge_memory.c:1232:56: error: implicit declaration of function ‘pmd_dirty’ [-Werror=implicit-function-declaration] ((flags & FOLL_FORCE) && (flags & FOLL_COW) && pmd_dirty(pmd)); ^~~~~~~~~ thanks, -- js suse labs -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
Am 20.03.2017 um 13:44 schrieb Jiri Slaby:
On 03/20/2017, 01:41 PM, Christian Mauderer wrote:
Am 20.03.2017 um 12:55 schrieb Jiri Slaby:
On 03/19/2017, 01:58 PM, Michal Hocko wrote:
On Sat 18-03-17 05:23:12, Mike Galbraith wrote:
(adds a couple CCs)
On Fri, 2017-03-17 at 22:09 +0100, Christian Mauderer wrote:
Hello,
I run across a problem with an application using wine on OpenSUSE 42.1. After some search, I found out that it is a known bug and that there is an fix available:
https://bugs.winehq.org/show_bug.cgi?id=41637
Basically the (backported) Dirty COW fix misses some adaptions in a second file and that leads to problems in some very special cases. For details please see the LKML thread with the fix:
https://lkml.org/lkml/2017/1/5/13
Is there any chance that this patch will be backported to the OpenSUSE 42.1 Kernel too? If I have seen it right, the OpenSUSE 42.2 Kernel is also affected.
Hm, it's tagged for stable, but doesn't seem to be taking wing.
Just for the reference, it is 8310d48b125d ("mm/huge_memory.c: respect FOLL_FORCE/FOLL_COW for thp"). SLE12-SP2 doesn't have this fix either. Could you push this to 3.12 stable Jiri, or should I post it there?
Please provide a backport and send to stable@vger, it does not apply at all.
thanks,
Hello,
the Ubuntu people ported the patch back to their kernel. It applies cleanly onto the openSUSE-42.1 branch of git://kernel.opensuse.org/kernel.git
You can find this version of the patch here:
http://kernel.ubuntu.com/git/ubuntu/ubuntu-xenial.git/commit/?id=9c49adf93cd...
I only tried whether it applies but not whether it would build. Would it be useful for you if I would build a test kernel with the patch and try it on one of my machines?
Neither that applies. And if I force that, it does not build in 3.12: mm/huge_memory.c: In function ‘can_follow_write_pmd’: mm/huge_memory.c:1232:56: error: implicit declaration of function ‘pmd_dirty’ [-Werror=implicit-function-declaration] ((flags & FOLL_FORCE) && (flags & FOLL_COW) && pmd_dirty(pmd)); ^~~~~~~~~
thanks,
Hello Jiri Slaby, might I ask on which branch and which repository you are working? Like I said, I tried the openSUSE-42.1 branch where the patch applied. But it sounds like you are using another one which is based on an older kernel version. Kind regards Christian Mauderer -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
On Mon, 20 Mar 2017 13:48:27 +0100, Christian Mauderer wrote:
Am 20.03.2017 um 13:44 schrieb Jiri Slaby:
On 03/20/2017, 01:41 PM, Christian Mauderer wrote:
Am 20.03.2017 um 12:55 schrieb Jiri Slaby:
On 03/19/2017, 01:58 PM, Michal Hocko wrote:
On Sat 18-03-17 05:23:12, Mike Galbraith wrote:
(adds a couple CCs)
On Fri, 2017-03-17 at 22:09 +0100, Christian Mauderer wrote: > Hello, > > I run across a problem with an application using wine on OpenSUSE 42.1. > After some search, I found out that it is a known bug and that there is > an fix available: > > https://bugs.winehq.org/show_bug.cgi?id=41637 > > Basically the (backported) Dirty COW fix misses some adaptions in a > second file and that leads to problems in some very special cases. For > details please see the LKML thread with the fix: > > https://lkml.org/lkml/2017/1/5/13 > > Is there any chance that this patch will be backported to the OpenSUSE > 42.1 Kernel too? If I have seen it right, the OpenSUSE 42.2 Kernel is > also affected.
Hm, it's tagged for stable, but doesn't seem to be taking wing.
Just for the reference, it is 8310d48b125d ("mm/huge_memory.c: respect FOLL_FORCE/FOLL_COW for thp"). SLE12-SP2 doesn't have this fix either. Could you push this to 3.12 stable Jiri, or should I post it there?
Please provide a backport and send to stable@vger, it does not apply at all.
thanks,
Hello,
the Ubuntu people ported the patch back to their kernel. It applies cleanly onto the openSUSE-42.1 branch of git://kernel.opensuse.org/kernel.git
You can find this version of the patch here:
http://kernel.ubuntu.com/git/ubuntu/ubuntu-xenial.git/commit/?id=9c49adf93cd...
I only tried whether it applies but not whether it would build. Would it be useful for you if I would build a test kernel with the patch and try it on one of my machines?
Neither that applies. And if I force that, it does not build in 3.12: mm/huge_memory.c: In function ‘can_follow_write_pmd’: mm/huge_memory.c:1232:56: error: implicit declaration of function ‘pmd_dirty’ [-Werror=implicit-function-declaration] ((flags & FOLL_FORCE) && (flags & FOLL_COW) && pmd_dirty(pmd)); ^~~~~~~~~
thanks,
Hello Jiri Slaby,
might I ask on which branch and which repository you are working? Like I said, I tried the openSUSE-42.1 branch where the patch applied. But it sounds like you are using another one which is based on an older kernel version.
Well, how about opening a (open)SUSE bugzilla entry as a start? Then we can track patches more easily than over ML. I can merge openSUSE-42.1 fix quickly if provided there, for example. For other branches, we need more reviews. thanks, Takashi -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
Am 20.03.2017 um 14:15 schrieb Takashi Iwai:
On Mon, 20 Mar 2017 13:48:27 +0100, Christian Mauderer wrote:
Am 20.03.2017 um 13:44 schrieb Jiri Slaby:
On 03/20/2017, 01:41 PM, Christian Mauderer wrote:
Am 20.03.2017 um 12:55 schrieb Jiri Slaby:
On 03/19/2017, 01:58 PM, Michal Hocko wrote:
On Sat 18-03-17 05:23:12, Mike Galbraith wrote: > (adds a couple CCs) > > On Fri, 2017-03-17 at 22:09 +0100, Christian Mauderer wrote: >> Hello, >> >> I run across a problem with an application using wine on OpenSUSE 42.1. >> After some search, I found out that it is a known bug and that there is >> an fix available: >> >> https://bugs.winehq.org/show_bug.cgi?id=41637 >> >> Basically the (backported) Dirty COW fix misses some adaptions in a >> second file and that leads to problems in some very special cases. For >> details please see the LKML thread with the fix: >> >> https://lkml.org/lkml/2017/1/5/13 >> >> Is there any chance that this patch will be backported to the OpenSUSE >> 42.1 Kernel too? If I have seen it right, the OpenSUSE 42.2 Kernel is >> also affected. > > Hm, it's tagged for stable, but doesn't seem to be taking wing.
Just for the reference, it is 8310d48b125d ("mm/huge_memory.c: respect FOLL_FORCE/FOLL_COW for thp"). SLE12-SP2 doesn't have this fix either. Could you push this to 3.12 stable Jiri, or should I post it there?
Please provide a backport and send to stable@vger, it does not apply at all.
thanks,
Hello,
the Ubuntu people ported the patch back to their kernel. It applies cleanly onto the openSUSE-42.1 branch of git://kernel.opensuse.org/kernel.git
You can find this version of the patch here:
http://kernel.ubuntu.com/git/ubuntu/ubuntu-xenial.git/commit/?id=9c49adf93cd...
I only tried whether it applies but not whether it would build. Would it be useful for you if I would build a test kernel with the patch and try it on one of my machines?
Neither that applies. And if I force that, it does not build in 3.12: mm/huge_memory.c: In function ‘can_follow_write_pmd’: mm/huge_memory.c:1232:56: error: implicit declaration of function ‘pmd_dirty’ [-Werror=implicit-function-declaration] ((flags & FOLL_FORCE) && (flags & FOLL_COW) && pmd_dirty(pmd)); ^~~~~~~~~
thanks,
Hello Jiri Slaby,
might I ask on which branch and which repository you are working? Like I said, I tried the openSUSE-42.1 branch where the patch applied. But it sounds like you are using another one which is based on an older kernel version.
Well, how about opening a (open)SUSE bugzilla entry as a start? Then we can track patches more easily than over ML.
I can merge openSUSE-42.1 fix quickly if provided there, for example. For other branches, we need more reviews.
thanks,
Takashi
Hello Takashi, I created the bug here: https://bugzilla.opensuse.org/show_bug.cgi?id=1030118 I've tried to add the relevant informations from this thread. Kind regards, Christian Mauderer -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
On 03/20/2017, 01:48 PM, Christian Mauderer wrote:
might I ask on which branch and which repository you are working
As I wrote, this is kernel 3.12 -- no branch. I maintain stable-3.12 which SLE12-SP1 is based upon. thanks, -- js suse labs -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
On Sat, 18 Mar 2017 05:23:12 +0100, Mike Galbraith wrote:
(adds a couple CCs)
On Fri, 2017-03-17 at 22:09 +0100, Christian Mauderer wrote:
Hello,
I run across a problem with an application using wine on OpenSUSE 42.1. After some search, I found out that it is a known bug and that there is an fix available:
https://bugs.winehq.org/show_bug.cgi?id=41637
Basically the (backported) Dirty COW fix misses some adaptions in a second file and that leads to problems in some very special cases. For details please see the LKML thread with the fix:
https://lkml.org/lkml/2017/1/5/13
Is there any chance that this patch will be backported to the OpenSUSE 42.1 Kernel too? If I have seen it right, the OpenSUSE 42.2 Kernel is also affected.
Hm, it's tagged for stable, but doesn't seem to be taking wing.
That's because cows can't fly. -- Jean Delvare SUSE L3 Support -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
Am 21.03.2017 um 18:29 schrieb Jean Delvare:
That's because cows can't fly.
But "With sufficient thrust, pigs fly just fine." (RFC1925 §3) SCNR -- Stefan Seyfried "For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." -- Richard Feynman -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
On mar., 2017-03-21 at 21:36 +0100, Stefan Seyfried wrote:
Am 21.03.2017 um 18:29 schrieb Jean Delvare:
That's because cows can't fly.
But "With sufficient thrust, pigs fly just fine."
(RFC1925 §3) SCNR
I know! This is how I manage to differentiate pigs and cows. -- Jean Delvare SUSE L3 Support -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
participants (8)
-
Christian Mauderer -
Jean Delvare -
Jiri Slaby -
Larry Finger -
Michal Hocko -
Mike Galbraith -
Stefan Seyfried -
Takashi Iwai