From: Anand Gadiyar <gadiyar@ti.com>
References: <201102051417.58953.edt@aei.ca> <201102072115.20858.edt@aei.ca>
 <AANLkTinpKwe2vWH3=Ur+N76a6KyfuFp0V-EznXXS7m+p@mail.gmail.com> <201102082331.27644.edt@aei.ca>
MIME-Version: 1.0
In-Reply-To: <201102082331.27644.edt@aei.ca>
Thread-Index: AcvIEj5H14XTge0NTGO0ln+xubdcWgADbaYw
Date: Wed, 9 Feb 2011 11:47:40 +0530
Message-ID: <c037b856cfa01b5c7014873850cd07a9@mail.gmail.com>
Subject: RE: [BUG] usb problems in .38-rc3+
To: Ed Tomlinson <edt@aei.ca>
Cc: Oliver Neukum <oneukum@suse.de>, linux-kernel@vger.kernel.org,
        linux-bluetooth@vger.kernel.org, linux-usb@vger.kernel.org
Content-Type: text/plain; charset=ISO-8859-1
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 3006
Lines: 72

Ed Tomlinson wrote:
> On Tuesday 08 February 2011 01:46:16 Gadiyar, Anand wrote:
> > On Tue, Feb 8, 2011 at 7:45 AM, Ed Tomlinson <edt@aei.ca> wrote:
> > > I tried bisecting without much luck.  It started with about 4000
commits to check.  It was still bad
> > > when it reached the first 1000 commits post .37.  Then all boots
started crashing.  I think its possible
> > > to restrict a bisect to a directory - if so, what dir should I try?
> > >
> >
> > Maybe drivers/ or drivers/usb would be enough for a first attempt?
> > I usually start there.
> >
> > You could just do:
> >
> > git bisect start -- [path]
> >
> > to restrict bisection to commits that touch [path].
>
> This is as far as I can get:
>
> # bad: [100b33c8bd8a3235fd0b7948338d6cbb3db3c63d] Linux 2.6.38-rc4
> # good: [3c0eee3fe6a3a1c745379547c7e7c904aa64f6d5] Linux 2.6.37
> # good: [387c31c7e5c9805b0aef8833d1731a5fe7bdea14] Linux 2.6.37-rc8
> # good: [90a8a73c06cc32b609a880d48449d7083327e11a] Linux 2.6.37-rc7
> # good: [c8ddb2713c624f432fa5fe3c7ecffcdda46ea0d4] Linux 2.6.37-rc1
> git bisect start 'v2.6.38-rc4' 'v2.6.37' 'v2.6.37-rc8' 'v2.6.37-rc7'
'v2.6.37-rc1' '--' 'drivers/usb' 'drivers/bluetooth'
> # good: [5cdc5bd8b2b1190cb54548c03194b154b4892e2a] Merge branch
'musb-hw' of git://gitorious.org/usb/usb into musb
> git bisect good 5cdc5bd8b2b1190cb54548c03194b154b4892e2a
> # bad: [f2c565e223af39ed38be5c84b1a37b591b22db83] xHCI: replace
dev_dbg() with xhci_dbg()
> git bisect bad f2c565e223af39ed38be5c84b1a37b591b22db83
> # good: [2af10844eb6ed104f9505bf3a7ba3ceb02264f31] USB: Merge 2.6.37-rc5
into usb-next
> git bisect good 2af10844eb6ed104f9505bf3a7ba3ceb02264f31
> # bad: [3e5b08cbbf78bedd316904ab0cf3b27119433ee5] Merge branch
'usb-next' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb-2.6
> git bisect bad 3e5b08cbbf78bedd316904ab0cf3b27119433ee5
> # good: [1051b9f0f9eab8091fe3bf98320741adf36b4cfa] Merge branch
'for-rmk' of git://git.pengutronix.de/git/imx/linux-2.6 into devel-stable
> git bisect good 1051b9f0f9eab8091fe3bf98320741adf36b4cfa
>
> After this the kernel does not build.
>
> Any clues?

Is it easier to just fix the build break locally and proceed?

I'm afraid I don't know how to skip a step during bisection.
I try narrowing things down manually.

Given that 2.6.37 is fine, can you check if v2.6.38-rc1 works?
If it doesn't work, then this break was caused by something that
went in during the merge window (which I believe is most likely).

You may be able to get somewhere by using snapshots of
linux-next as of particular days. They are available at [1]
as a single patch against a tag from Linus' tree.

Warning: This could be tedious, maybe not much more than a
git-bisect. Maybe easier to try and work it out from the
code.


- Anand
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/