[GIT PULL] arch/tile update for 3.1

[GIT PULL] arch/tile update for 3.1

[Chris Metcalf]

Linus,

Please pull the following trivial change for 3.1 from:

  git://github.com/cmetcalf-tilera/linux-tile.git stable

This change allows 3.1-rc9 to build for the Tile architecture.  A bombing
of <asm/atomic.h> to <linux/atomic.h> erroneously modified a couple
of places where we were relying on some assembly constants in the tile
<asm/atomic.h>.  Thanks!

Chris Metcalf (1):
      tile: revert change from <asm/atomic.h> to <linux/atomic.h> in asm files

 arch/tile/kernel/intvec_32.S  |    2 +-
 arch/tile/lib/atomic_asm_32.S |    2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

-- 
Chris Metcalf, Tilera Corp.
http://www.tilera.com

Re: [GIT PULL] arch/tile update for 3.1

[Linus Torvalds]

On Wed, Oct 12, 2011 at 9:49 AM, Chris Metcalf <cmetcalf@tilera.com> wrote:
>
> Please pull the following trivial change for 3.1 from:
>
>  git://github.com/cmetcalf-tilera/linux-tile.git stable
>
> This change allows 3.1-rc9 to build for the Tile architecture.  A bombing
> of <asm/atomic.h> to <linux/atomic.h> erroneously modified a couple
> of places where we were relying on some assembly constants in the tile
> <asm/atomic.h>.  Thanks!

This is not *at*all* what I get when I pull that branch.

I get a merge and two other commits too.

Not pulled.

         Linus

Re: [GIT PULL] arch/tile update for 3.1

[Chris Metcalf]

On 10/13/2011 2:31 AM, Linus Torvalds wrote:
> On Wed, Oct 12, 2011 at 9:49 AM, Chris Metcalf <cmetcalf@tilera.com> wrote:
>> Please pull the following trivial change for 3.1 from:
>>
>>  git://github.com/cmetcalf-tilera/linux-tile.git stable
>>
>> This change allows 3.1-rc9 to build for the Tile architecture.  A bombing
>> of <asm/atomic.h> to <linux/atomic.h> erroneously modified a couple
>> of places where we were relying on some assembly constants in the tile
>> <asm/atomic.h>.  Thanks!
> This is not *at*all* what I get when I pull that branch.
>
> I get a merge and two other commits too.
>
> Not pulled.

Mea culpa.  I've fixed the tree, so if you could pull again, it should now
have what the original message described.

No doubt my screwup is just a variant on a story you've heard more than a
few times, but for the record, here's some detail.  I saw the extra two
commits, which made no sense to me, so I examined one of the files that the
request-pull message described as different and saw no difference, so I
assumed it was just some minor confusion (maybe due to github vs
kernel.org).  Turns out that commit was a duplicate that had been in my
tree since June that I'd forgotten about, and the merge represented git
merging in the same semantic commit from a different tree.  So, I've now
rolled back my stable tree to June, re-applied just the patch I wanted you
to pull, and retested the tree -- and confirmed that git request-pull
actually now produces the output in my earlier email.

I'll make sure this doesn't happen again.
-- 
Chris Metcalf, Tilera Corp.
http://www.tilera.com