From: Sekhar Nori <hidden> Date: 2013-09-19 10:22:24
The following changes since commit 272b98c6455f00884f0350f775c5342358ebb73f:
Linux 3.12-rc1 (2013-09-16 16:17:51 -0400)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/nsekhar/linux-davinci.git tags/davinci-fixes-for-v3.12-rc2
for you to fetch changes up to caf9269f883bf0d5b83384efb96e2aa572d1d013:
ARM: edma: dt: create unused channel list (2013-09-19 15:13:32 +0530)
----------------------------------------------------------------
DaVinci fixes for v3.12-rc2
---------------------------
This pull request includes:
1) A patch to fix build breakage with DEBUG_LL
2) A patch to quell a build-time warning.
3) A patch to complete the DT support for EDMA. Without
this patch, the DT support for EDMA cannot be used.
There are no in-tree DT users of EDMA currently but
nevertheless the patch fixes a previously incomplete
implementation.
----------------------------------------------------------------
Brian Norris (1):
ARM: davinci: dm365 evm: fix unused variable warning
Joel Fernandes (1):
ARM: edma: dt: create unused channel list
Sekhar Nori (1):
ARM: davinci: fix build breakage with DEBUG_LL
arch/arm/common/edma.c | 39 ++++++++++++++++++++++-----
arch/arm/mach-davinci/board-dm365-evm.c | 2 +-
arch/arm/mach-davinci/include/mach/serial.h | 4 +--
3 files changed, 35 insertions(+), 10 deletions(-)
From: Olof Johansson <hidden> Date: 2013-09-19 17:48:58
Hi Sekhar,
On Thu, Sep 19, 2013 at 3:22 AM, Sekhar Nori [off-list ref] wrote:
The following changes since commit 272b98c6455f00884f0350f775c5342358ebb73f:
Linux 3.12-rc1 (2013-09-16 16:17:51 -0400)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/nsekhar/linux-davinci.git tags/davinci-fixes-for-v3.12-rc2
for you to fetch changes up to caf9269f883bf0d5b83384efb96e2aa572d1d013:
ARM: edma: dt: create unused channel list (2013-09-19 15:13:32 +0530)
----------------------------------------------------------------
DaVinci fixes for v3.12-rc2
---------------------------
This pull request includes:
1) A patch to fix build breakage with DEBUG_LL
2) A patch to quell a build-time warning.
3) A patch to complete the DT support for EDMA. Without
this patch, the DT support for EDMA cannot be used.
There are no in-tree DT users of EDMA currently but
nevertheless the patch fixes a previously incomplete
implementation.
Finishing incomplete work is not a fix, especially if the users aren't
in-tree. The other patches are just fine to include though.
Want to rebuild the branch (and stage the EDMA work for 3.13), or
should I pick the patches into branches at our end? Your choice.
-Olof
From: Sekhar Nori <hidden> Date: 2013-09-20 04:54:46
On Thursday 19 September 2013 11:18 PM, Olof Johansson wrote:
Hi Sekhar,
On Thu, Sep 19, 2013 at 3:22 AM, Sekhar Nori [off-list ref] wrote:
quoted
The following changes since commit 272b98c6455f00884f0350f775c5342358ebb73f:
Linux 3.12-rc1 (2013-09-16 16:17:51 -0400)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/nsekhar/linux-davinci.git tags/davinci-fixes-for-v3.12-rc2
for you to fetch changes up to caf9269f883bf0d5b83384efb96e2aa572d1d013:
ARM: edma: dt: create unused channel list (2013-09-19 15:13:32 +0530)
----------------------------------------------------------------
DaVinci fixes for v3.12-rc2
---------------------------
This pull request includes:
1) A patch to fix build breakage with DEBUG_LL
2) A patch to quell a build-time warning.
3) A patch to complete the DT support for EDMA. Without
this patch, the DT support for EDMA cannot be used.
There are no in-tree DT users of EDMA currently but
nevertheless the patch fixes a previously incomplete
implementation.
Finishing incomplete work is not a fix, especially if the users aren't
in-tree. The other patches are just fine to include though.
Yes, its not a regression. Doesn't break any in-tree support really. I
would not have sent it if it was late -rc cycle.
Want to rebuild the branch (and stage the EDMA work for 3.13), or
should I pick the patches into branches at our end? Your choice.
I just sent you an updated pull request with the offending patch removed.
Thanks,
Sekhar
From: Alexander Holler <hidden> Date: 2013-09-20 06:06:02
Am 19.09.2013 19:48, schrieb Olof Johansson:
Hi Sekhar,
On Thu, Sep 19, 2013 at 3:22 AM, Sekhar Nori [off-list ref] wrote:
quoted
The following changes since commit 272b98c6455f00884f0350f775c5342358ebb73f:
...
quoted
3) A patch to complete the DT support for EDMA. Without
this patch, the DT support for EDMA cannot be used.
There are no in-tree DT users of EDMA currently but
nevertheless the patch fixes a previously incomplete
implementation.
Finishing incomplete work is not a fix, especially if the users aren't
in-tree. The other patches are just fine to include though.
It fixes problems for all people which are not using the provided DT's
but using edma in their own DT. So I don't understand your arguing as DT
got introduced to not only support in-tree configs. It clearly fixes
problems.
Regards,
Alexander Holler
From: Olof Johansson <hidden> Date: 2013-09-20 17:40:19
Hi,
On Thu, Sep 19, 2013 at 11:06 PM, Alexander Holler [off-list ref] wrote:
Am 19.09.2013 19:48, schrieb Olof Johansson:
quoted
Hi Sekhar,
On Thu, Sep 19, 2013 at 3:22 AM, Sekhar Nori [off-list ref] wrote:
quoted
The following changes since commit 272b98c6455f00884f0350f775c5342358ebb73f:
...
quoted
quoted
3) A patch to complete the DT support for EDMA. Without
this patch, the DT support for EDMA cannot be used.
There are no in-tree DT users of EDMA currently but
nevertheless the patch fixes a previously incomplete
implementation.
Finishing incomplete work is not a fix, especially if the users aren't
in-tree. The other patches are just fine to include though.
It fixes problems for all people which are not using the provided DT's
but using edma in their own DT. So I don't understand your arguing as DT
got introduced to not only support in-tree configs. It clearly fixes
problems.
Right, that was the smaller part of the argument. The bigger is:
Completing a feature is not a fix. If the whole feature didn't make it
in for a release, then the rest goes into the next release.
-Olof
From: Alexander Holler <hidden> Date: 2013-09-20 17:45:59
Am 20.09.2013 19:40, schrieb Olof Johansson:
Hi,
On Thu, Sep 19, 2013 at 11:06 PM, Alexander Holler [off-list ref] wrote:
quoted
Am 19.09.2013 19:48, schrieb Olof Johansson:
quoted
Hi Sekhar,
On Thu, Sep 19, 2013 at 3:22 AM, Sekhar Nori [off-list ref] wrote:
quoted
The following changes since commit 272b98c6455f00884f0350f775c5342358ebb73f:
...
quoted
quoted
3) A patch to complete the DT support for EDMA. Without
this patch, the DT support for EDMA cannot be used.
There are no in-tree DT users of EDMA currently but
nevertheless the patch fixes a previously incomplete
implementation.
Finishing incomplete work is not a fix, especially if the users aren't
in-tree. The other patches are just fine to include though.
It fixes problems for all people which are not using the provided DT's
but using edma in their own DT. So I don't understand your arguing as DT
got introduced to not only support in-tree configs. It clearly fixes
problems.
Right, that was the smaller part of the argument. The bigger is:
Completing a feature is not a fix. If the whole feature didn't make it
in for a release, then the rest goes into the next release.
Bureaucrazy at it's best.
Regards,
Alexander Holler