Linux 4.4-rc1

Linux 4.4-rc1

[Linus Torvalds]

So it's Sunday, two weeks has passed, and so 4.4-rc1 is out there and
the merge window is closed.

As usual, the full shortlog is much too big to post, so appended is
the usual shortlog of just my merges, which just shows who I did pulls
from, with a very short comment on each merge.

Just looking at the patch itself, things look fairly normal at a high
level, possibly a bit more driver-heavy than usual with about 75% of
the patch being drivers, and 10% being architecture updates. The
remaining 15% is documentation, filesystem, core networking (as
opposed to network drivers), tooling and some core infrastructure.

The driver changes are all over, although staging, networking and GPU
drivers stand out (and those three areas account for over half of the
driver changes - roughly 40% of the whole patch).

On the architecture side, ARM (when counting both 32-bit and 64-bit)
accounts for about half the changes, with x86, powerpc, mips, chris
and s390 accounting for the other half.

Go out and test.

                    Linus

---

Al Viro (2):
    vfs update
    vfs xattr cleanups

Alex Williamson (1):
    VFIO updates

Alexandre Belloni (1):
    RTC updates

Andrew Morton (4):
    patch-bomb
    second patch-bomb
    third patch-bomb
    final patch-bomb

Arnd Bergmann (1):
    asm-generic cleanups

Bjorn Helgaas (1):
    PCI updates

Bob Peterson (1):
    gfs2 updates

Borislav Petkov (1):
    EDAC updates

Brian Norris (1):
    MTD updates

Bruce Fields (1):
    nfsd updates

Catalin Marinas (2):
    arm64 updates
    arm64 fixes and clean-ups

Chris Mason (2):
    btrfs updates
    btrfs fixes and cleanups

Chris Zankel (1):
    xtensa updates

Dan Williams (2):
    libnvdimm updates
    libnvdimm fixes

Darren Hart (2):
    x86 platform driver update
    another x86 platform driver update

Dave Airlie (4):
    drm updates
    drm fixes
    drm fixes
    drm sti driver updates

Dave Chinner (1):
    xfs updates

David Miller (4):
    networking updates
    IDE fixlet
    sparc updates
    networking fixes

David Teigland (1):
    dlm update

David Vrabel (1):
    xen updates

David Woodhouse (1):
    intel iommu updates

Dmitry Torokhov (2):
    input updates
    more input updates

Doug Ledford (1):
    rdma updates

Eric Biederman (1):
    userns hardlink capability check fix

Geert Uytterhoeven (1):
    m68k update

Greg KH (5):
    USB updates
    tty/serial driver updates
    staging driver updates
    driver core updates
    char/misc driver updates

Greg Ungerer (1):
    m68knommu/coldfire fix

Guenter Roeck (1):
    hwmon updates

Hans-Christian Noren Egtvedt (1):
    avr32 update

Helge Deller (2):
    parisc updates
    parisc updates

Herbert Xu (2):
    crypto update
    crypto fix

Ingo Molnar (17):
    wchan kernel address hiding
    EFI changes
    RCU changes
    locking changes
    perf updates
    RAS changes
    scheduler changes
    x86 apic changes
    x86 asm changes
    x86 boot cleanup
    x86 cleanups
    x86 cpu changes
    x86 kgdb fixlet
    x86 fpu changes
    x86 sigcontext header cleanups
    x86 mm changes
    x86 platform changes

Jacek Anaszewski (1):
    LED updates

Jaegeuk Kim (1):
    f2fs updates

James Bottomley (2):
    SCSI updates
    final round of SCSI updates

James Hogan (1):
    metag arch updates

James Morris (2):
    security subsystem update
    security subsystem fixes

Jan Kara (1):
    ext2 fix

Jassi Brar (1):
    mailbox updates

Jean Delvare (1):
    tiny hwmon update

Jeff Layton (1):
    file locking updates

Jens Axboe (7):
    core block updates
    block driver updates
    lightnvm support
    block integrity updates
    block reservation support
    block IO poll support
    misc block fixes

Jesper Nilsson (1):
    CRIS changes

Jiri Kosina (3):
    livepatching fix
    HID updates
    trivial updates

Joerg Roedel (1):
    iommu updates

Jon Corbet (2):
    documentation update
    more documentation updates

Jon Mason (1):
    NTB bugfixes

Lee Jones (2):
    MFD updates
    backlight updates

Ley Foon Tan (1):
    nios2 updates

Linus Walleij (2):
    pin control updates
    GPIO updates

Mark Brown (3):
    regmap updates
    regulator updates
    spi updates

Martin Schwidefsky (1):
    s390 updates

Mauro Carvalho Chehab (1):
    media updates

Michael Ellerman (1):
    powerpc updates

Michal Marek (3):
    kbuild update
    kconfig updates
    misc kbuild updates

Mike Snitzer (1):
    device mapper updates

Neil Brown (2):
    md updates
    config fix for md

Nicholas Bellinger (1):
    SCSI target updates

Olof Johansson (7):
    ARM SoC non-urgent fixes
    ARM SoC cleanups
    ARM SoC platform updates
    ARM SoC driver updates
    ARM DT updates
    ARM SoC defconfig updates
    chrome platform updates

Paolo Bonzini (2):
    KVM updates
    second batch of kvm updates

Paul Moore (1):
    audit updates

Rafael Wysocki (2):
    power management and ACPI updates
    more power management and ACPI updates

Ralf Baechle (1):
    MIPS updates

Richard Weinberger (2):
    UML updates
    UBI/UBIFS updates

Rob Herring (2):
    DeviceTree updates
    DeviceTree fixes

Russell King (1):
    ARM updates

Rusty Russell (1):
    module updates

Sage Weil (1):
    Ceph updates

Sebastian Reichel (3):
    HSI updates
    power supply and reset updates
    power supply fix

Shuah Khan (1):
    kselftest updates

Stefan Richter (1):
    firewire fix

Stephen Boyd (1):
    clk updates

Steve French (1):
    SMB3 updates

Steven Rostedt (2):
    tracking updates
    trace cleanups

Takashi Iwai (2):
    sound updates
    sound fixes

Ted Ts'o (1):
    ext4 updates

Tejun Heo (3):
    workqueue update
    libata updates
    cgroup updates

Thierry Reding (1):
    pwm updates

Thomas Gleixner (7):
    timer updates
    irq updates
    irq and timer fixes
    x86 fixes
    liblockdep fixes
    scheduler fix
    perf updates

Tomi Valkeinen (1):
    fbdev updates

Tony Luck (1):
    pstore updates

Trond Myklebust (1):
    NFS client updates

Ulf Hansson (1):
    MMC updates

Vineet Gupta (2):
    ARC updates
    ARC fixes

Vinod Koul (1):
    dmaengine updates

Wim Van Sebroeck (1):
    watchdog update

Wolfram Sang (1):
    i2c updates

Yoshinori Sato (1):
    h8300 updates

Zhang Rui (1):
    thermal updates

Re: Linux 4.4-rc1

[Ken Moffat]

On Sun, Nov 15, 2015 at 05:24:37PM -0800, Linus Torvalds wrote:
> So it's Sunday, two weeks has passed, and so 4.4-rc1 is out there and
> the merge window is closed.
> 
[...]
> 
> Go out and test.
> 
>                     Linus
> 
After what I picked up during the 4.3 cycle, I tried using 'xzcat |
git apply -' but I got the following (after messages about spaces
before tabs) :

error: cannot apply binary patch to
'drivers/staging/ft1000/ft1000-pcmcia/ft1000.img' without full index
line
error: drivers/staging/ft1000/ft1000-pcmcia/ft1000.img: patch does
not apply
error: cannot apply binary patch to
'drivers/staging/ft1000/ft1000-usb/ft3000.img' without full index
line
error: drivers/staging/ft1000/ft1000-usb/ft3000.img: patch does not
apply

That is with git-2.6.3.

Interestingly, using 'xzcat | patch -p1' seems to apply fine, or at
least it doesn't seem to report any errors, and I have no particular
interest in ft1000 so I'll go with that.

It boots ok on this machine (AMD K10, with my current config), but
that doesn't say a lot, so I'll shut up.

ĸen
-- 
This email was written using 100% recycled letters.

Re: Linux 4.4-rc1

[Guenter Roeck]

On Sun, Nov 15, 2015 at 05:24:37PM -0800, Linus Torvalds wrote:
> So it's Sunday, two weeks has passed, and so 4.4-rc1 is out there and
> the merge window is closed.
> 
> As usual, the full shortlog is much too big to post, so appended is
> the usual shortlog of just my merges, which just shows who I did pulls
> from, with a very short comment on each merge.
> 
> Just looking at the patch itself, things look fairly normal at a high
> level, possibly a bit more driver-heavy than usual with about 75% of
> the patch being drivers, and 10% being architecture updates. The
> remaining 15% is documentation, filesystem, core networking (as
> opposed to network drivers), tooling and some core infrastructure.
> 
> The driver changes are all over, although staging, networking and GPU
> drivers stand out (and those three areas account for over half of the
> driver changes - roughly 40% of the whole patch).
> 
> On the architecture side, ARM (when counting both 32-bit and 64-bit)
> accounts for about half the changes, with x86, powerpc, mips, chris
> and s390 accounting for the other half.
> 
> Go out and test.
> 
Build results:
	total: 144 pass: 143 fail: 1
Failed builds:
	cris:allnoconfig

Qemu test results:
	total: 95 pass: 94 fail: 1
Failed tests:
	cris:crisv32_defconfig

cris:allnoconfig fails to build because of section mismatches, which is now
fatal with allnoconfig builds (since 47490ec141b9, "modpost: Add flag -E for
making section mismatches fatal"). Proposed patches to fix the problems are
available at [1] and [2]. Other allnoconfig builds (untested by me) probably
fail as well, though.

The cris:crisv32_defconfig qemu test crashes because of interference between
commit 7d8c70d8048c ("serial: mctrl-gpio: rename init function") and commit
7b9c5162c182 ("serial: etraxfs-uart: use mctrl_gpio helpers for handling
modem signals"). A proposed patch is available at [3].

Further details are available at http://server.roeck-us.net:8010/builders.

Guenter

---
[1] https://patchwork.kernel.org/patch/7592471/
[2] https://patchwork.kernel.org/patch/7592461/
[3] https://patchwork.kernel.org/patch/7533041/

Re: Linux 4.4-rc1

[Shuah Khan]

On Sun, Nov 15, 2015 at 6:24 PM, Linus Torvalds
<torvalds@linux-foundation.org> wrote:
> So it's Sunday, two weeks has passed, and so 4.4-rc1 is out there and
> the merge window is closed.
>
> As usual, the full shortlog is much too big to post, so appended is
> the usual shortlog of just my merges, which just shows who I did pulls
> from, with a very short comment on each merge.
>
> Just looking at the patch itself, things look fairly normal at a high
> level, possibly a bit more driver-heavy than usual with about 75% of
> the patch being drivers, and 10% being architecture updates. The
> remaining 15% is documentation, filesystem, core networking (as
> opposed to network drivers), tooling and some core infrastructure.
>
> The driver changes are all over, although staging, networking and GPU
> drivers stand out (and those three areas account for over half of the
> driver changes - roughly 40% of the whole patch).
>
> On the architecture side, ARM (when counting both 32-bit and 64-bit)
> accounts for about half the changes, with x86, powerpc, mips, chris
> and s390 accounting for the other half.
>
> Go out and test.
>

I ran into the same VPN bug that was reported in 4.3. This bug is very pesky.
VPN connection works just fine and subsequent web access fails. It took me
a couple of hours to trace it back to the kernel, after investigating
the obvious
suspects (routers, network connections etc.)

I am hoping the fix gets into 4.4-rc2 as well as 4.3.1

http://patchwork.ozlabs.org/patch/544307/

I tested the fix on 4.3 and 4.4-rc1 on my VPN setup

thanks,
-- Shuah

Re: Linux 4.4-rc1

[Linus Torvalds]

On Wed, Nov 18, 2015 at 6:46 AM, Shuah Khan <shuahkhan@gmail.com> wrote:
>
> I am hoping the fix gets into 4.4-rc2 as well as 4.3.1
>
> http://patchwork.ozlabs.org/patch/544307/
>
> I tested the fix on 4.3 and 4.4-rc1 on my VPN setup

That should be commit 5f8dc33e8ee7 ("net: fix feature changes on
devices without ndo_set_features") and it got merged through the
networking tree yesterday.

Of course, it's always a good idea to double-check, so go out and test
that the current -git tree works for you..

Thanks,
            Linus

Re: Linux 4.4-rc1

[David Miller]

From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Wed, 18 Nov 2015 08:17:17 -0800

> On Wed, Nov 18, 2015 at 6:46 AM, Shuah Khan <shuahkhan@gmail.com> wrote:
>>
>> I am hoping the fix gets into 4.4-rc2 as well as 4.3.1
>>
>> http://patchwork.ozlabs.org/patch/544307/
>>
>> I tested the fix on 4.3 and 4.4-rc1 on my VPN setup
> 
> That should be commit 5f8dc33e8ee7 ("net: fix feature changes on
> devices without ndo_set_features") and it got merged through the
> networking tree yesterday.
> 
> Of course, it's always a good idea to double-check, so go out and test
> that the current -git tree works for you..

Also, all of the commits in question up the chain of "Fixes: " tags
only apply to 4.4-rc1, so 4.3 should not need the fix.

linux-next: stats for v4.4-rc1

[Stephen Rothwell]

Hi all,

As usual, the executive friendly graph is at
http://neuling.org/linux-next-size.html :-)

(No merge commits counted, next-20151103 was the first linux-next after
the merge window opened.)

Commits in v4.4-rc1 (relative to v4.3):            11528
Commits in next-20151103:                          10823
Commits with the same SHA1:                         9957
Commits with the same patch_id:                      404 (1)
Commits with the same subject line:                   51 (1)

(1) not counting those in the lines above.

So commits in -rc1 that were in next-20151103:     10412 90%

Some breakdown of the list of extra commits (relative to next-20151103)
in -rc1:

Top ten first word of commit summary:

    136 drm
    116 btrfs
     40 mips
     36 cxlflash
     34 kvm
     28 hpsa
     27 perf
     27 net
     24 megaraid_sas
     17 vfio

Top ten authors:

     32 mrochs@linux.vnet.ibm.com
     30 a.hajda@samsung.com
     24 sumit.saxena@avagotech.com
     24 quwenruo@cn.fujitsu.com
     17 sreekanth.reddy@avagotech.com
     17 dsterba@suse.com
     16 jcmvbkbc@gmail.com
     16 chuck.lever@oracle.com
     15 edumazet@google.com
     15 don.brace@pmcs.com

Top ten commiters:

    107 martin.petersen@oracle.com
     94 davem@davemloft.net
     69 jbottomley@odin.com
     54 clm@fb.com
     48 dsterba@suse.com
     45 bskeggs@redhat.com
     42 torvalds@linux-foundation.org
     41 ralf@linux-mips.org
     32 pbonzini@redhat.com
     30 rostedt@goodmis.org

There are also 411 commits in next-20151103 that didn't make it into
v4.4-rc1.

Top ten first word of commit summary:

     52 mm
     23 orangefs
     23 arm
     18 page-flags
     17 arm-soc
     15 soc
     15 ocfs2
     12 thp
     12 nfsd
     11 media

Top ten authors:

     63 kirill.shutemov@linux.intel.com
     30 akpm@linux-foundation.org
     20 hubcap@omnibond.com
     19 minchan@kernel.org
     19 jlayton@poochiereds.net
     13 olof@lixom.net
     11 sboyd@codeaurora.org
      8 ryan.ding@oracle.com
      8 pankaj.dubey@samsung.com
      7 mathieu.poirier@linaro.org

Some of Andrew's patches are fixes for other patches in his tree (and
have been merged into those).

Top ten commiters:

    199 sfr@canb.auug.org.au
     24 hubcap@omnibond.com
     20 bfields@redhat.com
     20 agross@codeaurora.org
     13 olof@lixom.net
     12 arnd@arndb.de
     11 wsa@the-dreams.de
     11 mchehab@osg.samsung.com
      9 kgene@kernel.org
      8 peter.chen@freescale.com

Those commits by me are from the quilt series (mainly Andrew's mmotm
tree).

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

Re: Linux 4.4-rc1

[Shuah Khan]

On Wed, Nov 18, 2015 at 9:17 AM, Linus Torvalds
<torvalds@linux-foundation.org> wrote:
> On Wed, Nov 18, 2015 at 6:46 AM, Shuah Khan <shuahkhan@gmail.com> wrote:
>>
>> I am hoping the fix gets into 4.4-rc2 as well as 4.3.1
>>
>> http://patchwork.ozlabs.org/patch/544307/
>>
>> I tested the fix on 4.3 and 4.4-rc1 on my VPN setup
>
> That should be commit 5f8dc33e8ee7 ("net: fix feature changes on
> devices without ndo_set_features") and it got merged through the
> networking tree yesterday.
>
> Of course, it's always a good idea to double-check, so go out and test
> that the current -git tree works for you..
>

I am running into the following depmod error:

depmod: ERROR: Found 2 modules in dependency cycles!
depmod: ERROR: Cycle detected: lnet -> libcfs -> lnet
Makefile:1139: recipe for target '_modinst_post' failed
make: *** [_modinst_post] Error 1

I am testing on your tree - the top commit is:

commit 34258a32d9a9fc9e38fb549efe1692301cc31f85
Merge: 0d77a12 f52c74f
Author: Linus Torvalds <torvalds@linux-foundation.org>
Date:   Wed Nov 18 08:59:29 2015 -0800

    Merge branch 'for-linus' of
git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux

thanks,
-- Shuah

Re: Linux 4.4-rc1

[Linus Torvalds]

On Thu, Nov 19, 2015 at 6:10 AM, Shuah Khan <shuahkhan@gmail.com> wrote:
>
> I am running into the following depmod error:
>
> depmod: ERROR: Found 2 modules in dependency cycles!
> depmod: ERROR: Cycle detected: lnet -> libcfs -> lnet

There should be a patch for this in the staging tree, but Greg hasn't
sent the pull request yet.

In the meantime, do you actually _use_ lustre? Because if not, you can
just compile it out for now.

                      Linus

Re: Linux 4.4-rc1

[Shuah Khan]

On Thu, Nov 19, 2015 at 12:25 PM, Linus Torvalds
<torvalds@linux-foundation.org> wrote:
> On Thu, Nov 19, 2015 at 6:10 AM, Shuah Khan <shuahkhan@gmail.com> wrote:
>>
>> I am running into the following depmod error:
>>
>> depmod: ERROR: Found 2 modules in dependency cycles!
>> depmod: ERROR: Cycle detected: lnet -> libcfs -> lnet
>
> There should be a patch for this in the staging tree, but Greg hasn't
> sent the pull request yet.
>
> In the meantime, do you actually _use_ lustre? Because if not, you can
> just compile it out for now.
>

I must have enabled a while back to experiment and forgot all about it.
Disabled Lustre. Verified VPN and everything looks good.

-- Shuah