Kernel 4.1 RC (Release Candidate) series

[Doug S]

Perhaps I should have started a new thread for the Wily Kernel. All the previous are for Vivid which is off topic.

Herein, we always have a thread going for the current RC (Release Candidate) series, independent of the version of Ubuntu running on the computer.


Can anyone confirm or deny: For any Kernel 4.1RC1 or 2 or 3: After a suspend, only CPU 0 comes back on-line.

For Kernel 4.0 things work fine for me upon resume from suspend.
This issue could well be processor dependent, mine is an i7-2600K.

[Doug S]

Can anyone confirm or deny: For any Kernel 4.1RC1 or 2 or 3: After a suspend, only CPU 0 comes back on-line.

... 16 kernel compiles later...

I bisected the kernel. The culprit seems to be:

3c18d447b3b36a8d3c90dc37dfbd363cdb685d0a is the first bad commit
commit 3c18d447b3b36a8d3c90dc37dfbd363cdb685d0a
Author: Juri Lelli <juri.lelli@arm.com>
Date:   Tue Mar 31 09:53:37 2015 +0100

    sched/core: Check for available DL bandwidth in cpuset_cpu_inactive()

    Hotplug operations are destructive w.r.t. cpusets. In case such an
    operation is performed on a CPU belonging to an exlusive cpuset, the
    DL bandwidth information associated with the corresponding root
    domain is gone even if the operation fails (in sched_cpu_inactive()).

    For this reason we need to move the check we currently have in
    sched_cpu_inactive() to cpuset_cpu_inactive() to prevent useless
    cpusets reconfiguration in the CPU_DOWN_FAILED path.

    Signed-off-by: Juri Lelli <juri.lelli@arm.com>
    Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
    Cc: Juri Lelli <juri.lelli@gmail.com>
    Link: http://lkml.kernel.org/r/1427792017-7356-2-git-send-email-juri.lelli@arm.com
    Signed-off-by: Ingo Molnar <mingo@kernel.org>

:040000 040000 10f8d81afdc8e625f8e6720883d3eb42c28d452b c08264528890941bad35d5d4cc134c03f259c534 M      kernel

EDIT: I made a mistake somewhere. I'll post the correct commit culprit once I figure it out.

EDIT 2: ... another bunch of kernel compiles later ... Proper culprit identified and edited above. The fix is 533445c6e533 "sched/core: Fix regression in cpuset_cpu_inactive() for suspend", which I am trying now.
EDIT 3: Patch works fine and fixes the issue.

[Cavsfan]

I don't know if you care about this or not but it is some news yesterday May 13th about kernel 4.1 being on the final release of Wily.

Being based on Vivid, Ubuntu 15.10 uses its Linux 3.19 kernel packages, but the Ubuntu
developers announced today, after an Ubuntu Kernel Team meeting, that they will switch
soon to Linux kernel 4.0 and that the final release of Ubuntu 15.10 (Wily Werewolf) will
most likely be powered by Linux kernel 4.1.

http://linux.softpedia.com/blog/Ubun...y-481002.shtml

I was wondering myself as to which one 15.10 would go with.

[dino99]

That one works smootly

linux (4.0.0-1.1) wily; urgency=low

[ Upstream Kernel Changes ]
* rebase to v4.0.1
* rebase to v4.0-rc5
- LP: #1428947
- LP: #1410704
- LP: #1412800
- LP: #1400215
- LP: #1411193
- LP: #1412800
- LP: #1408295
* rebase to v4.0-rc6
- LP: #1436745
* rebase to v4.0-rc7
* rebase to v4.0 final

-- Leann Ogasawara <email address hidden> Wed, 06 May 2015 06:29:00 -0700

[Doug S]

Can anyone confirm or deny: For any Kernel 4.1RC1 or 2 or 3: After a suspend, only CPU 0 comes back on-line.

The fix for this issue (533445c6e533 "sched/core: Fix regression in cpuset_cpu_inactive() for suspend") will be in kernel 4.1RC4.

[Doug S]

Kernel 4.1rc4 is available. I have been running it for 7 hours or so without issues.

[mrfelker]

I have not got Wily to boot with the latest Wily kernel from the lmainline (4.04). Has anybody else boot 4.04?

[Doug S]

Kernel 4.1RC5 is available.

Both the Ubuntu, and my own compiled, versions of kernels are not working for me at the moment. I am still recovering things after a hard disk failure two days ago on my test computer. My issues are NOT specific to this kernel series, but rather I have done something wrong while recovering things (and haven't looked into it yet).

[ronacc]

I have the 64bit rc5 ( ubuntu version ) running here , seems good so far .

[den_]

My system froze after a while with rc5, 14.04, Intel Sandy Bridge Graphics and oibaf repo for mesa and Intel drivers, while I was using VirtualBox and Win 7 guest. Besides that, other stuff seem to work well.