Re: linux-next: Tree for Aug 23

[Peter Zijlstra <peterz@infradead.org>]

On Sun, Aug 24, 2014 at 09:57:41PM -0700, Guenter Roeck wrote:
> On Sat, Aug 23, 2014 at 08:04:22AM +1000, Stephen Rothwell wrote:
> > Hi all,
> > 
> > Changes since 20140822:
> > 
> > The mfd tree gained a build failure so I used the version from
> > next-20140822.
> > 
> > The usb-gadget tree gained a conflict against the usb-gadget-fixes tree.
> > 
> > The pwm tree gained a build failure for which I reverted a couple of
> > commits.
> > 
> > The staging tree still had its build failure for which I applied a
> > fix patch.
> > 
> > Non-merge commits (relative to Linus' tree): 1435
> >  1372 files changed, 34371 insertions(+), 25442 deletions(-)
> > 
> 
> Images build from this tree crash in mips, mips64, and sparc64 qemu tests
> with memory allocation errors. For mips, bisect points to commit ef31563e950c
> (locking,arch,mips: Fold atomic_ops).
> 
> # bad: [eeed658b5a92d957aab7bdb7c469a90016420c7d] Add linux-next specific files for 20140823
> # good: [7d1311b93e58ed55f3a31cc8f94c4b8fe988a2b9] Linux 3.17-rc1
> git bisect start 'HEAD' 'v3.17-rc1'
> # good: [a364ac83d553cd223cf1b898a738594f866242ac] Merge remote-tracking branch 'input/next'
> git bisect good a364ac83d553cd223cf1b898a738594f866242ac
> # bad: [6804e0868ce1f5f55a16270de65587e35b80231e] Merge remote-tracking branch 'usb-gadget/next'
> git bisect bad 6804e0868ce1f5f55a16270de65587e35b80231e
> # bad: [8590e100eca66f5763b4c8a5775e38e4a0c9939d] Merge branch 'x86/xsave'
> git bisect bad 8590e100eca66f5763b4c8a5775e38e4a0c9939d
> # good: [e71e79457b79a52827039d9d7f253321bfd342bd] perf symbols: Don't demangle parameters and such by default
> git bisect good e71e79457b79a52827039d9d7f253321bfd342bd
> # bad: [100b1cb29333c8ac37e68025b291e4b273c8ad43] Merge branch 'locking/core'
> git bisect bad 100b1cb29333c8ac37e68025b291e4b273c8ad43
> # bad: [c6470150dff9aff682063890c9b8eac71b695def] locking,arch,sh: Fold atomic_ops
> git bisect bad c6470150dff9aff682063890c9b8eac71b695def
> # good: [50f853e38b0b90a5703ab14b70e20eb5a8ccd5de] locking,arch,hexagon: Fold atomic_ops
> git bisect good 50f853e38b0b90a5703ab14b70e20eb5a8ccd5de
> # good: [d6dfe2509da935a15583cace7cd3837b1e8addef] locking,arch,metag: Fold atomic_ops
> git bisect good d6dfe2509da935a15583cace7cd3837b1e8addef
> # bad: [e69a0ef76627005e3e83d0e086e6bb1d247bb65b] locking,arch,mn10300: Fold atomic_ops
> git bisect bad e69a0ef76627005e3e83d0e086e6bb1d247bb65b
> # bad: [ef31563e950c60bb41b97c2b61c32de874f3c949] locking,arch,mips: Fold atomic_ops
> git bisect bad ef31563e950c60bb41b97c2b61c32de874f3c949
> # first bad commit: [ef31563e950c60bb41b97c2b61c32de874f3c949] locking,arch,mips: Fold atomic_ops
> 
> Reverting the first bad commit fixes the problem for mips and mips64.
> 
> For sparc64, given the above, I did not bother to bisect, but just reverted
> commit 4f3316c (locking,arch,sparc: Fold atomic_ops). This fixes the problem
> for sparc64.

Urgh, I take it this is still current? Would you perhaps have these qemu
images available so I can try and fix this?