Hi Linus,
Please pull from:
git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media tags/media/v4.2-2
For some fixes:
- a regression fix at the videobuf2 core driver;
- Fix error handling at mantis probing code;
- Revert the IR encode patches, as the API is not mature enough.
So, better to postpone the changes to a latter Kernel;
- Fix Kconfig breakages on some randconfig scenarios.
Thanks!
Mauro
-
The following changes since commit faebbd8f134f0c054f372982c8ddd1bbcc41b440:
[media] lmedm04: fix the range for relative measurements (2015-06-24 08:38:30 -0300)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media tags/media/v4.2-2
for you to fetch changes up to ab3452a4da373944c3435ca33ec4604920a516f8:
[media] mantis: Fix error handling in mantis_dma_init() (2015-08-11 10:28:10 -0300)
----------------------------------------------------------------
media fixes for v4.2-rc8
----------------------------------------------------------------
David Härdeman (7):
[media] rc: nuvoton-cir: Add support for writing wakeup samples via sysfs filter callback"
[media] rc: rc-loopback: Add loopback of filter scancodes"
[media] rc: rc-core: Add support for encode_wakeup drivers"
[media] rc: ir-rc6-decoder: Add encode capability"
[media] rc: ir-rc5-decoder: Add encode capability"
[media] rc: rc-ir-raw: Add Manchester encoder (phase encoder) helper"
[media] rc: rc-ir-raw: Add scancode encoder callback"
Fabio Estevam (1):
[media] mantis: Fix error handling in _dmantisma_init()
Laurent Pinchart (1):
[media] vb2: Fix compilation breakage when !CONFIG_BUG
Randy Dunlap (2):
[media] media/dvb: fix ts2020.c Kconfig and build
[media] media/pci/cobalt: fix Kconfig and build when SND is not enabled
Sakari Ailus (1):
[media] vb2: Only requeue buffers immediately once streaming is started
drivers/media/dvb-frontends/Kconfig | 2 +-
drivers/media/pci/cobalt/Kconfig | 1 +
drivers/media/pci/cobalt/cobalt-irq.c | 2 +-
drivers/media/pci/mantis/mantis_dma.c | 5 +-
drivers/media/rc/ir-rc5-decoder.c | 116 -
drivers/media/rc/ir-rc6-decoder.c | 122 -
drivers/media/rc/nuvoton-cir.c | 127 -
drivers/media/rc/nuvoton-cir.h | 1 -
drivers/media/rc/rc-core-priv.h | 36 -
drivers/media/rc/rc-ir-raw.c | 139 --
drivers/media/rc/rc-loopback.c | 36 -
drivers/media/rc/rc-main.c | 7 +-
drivers/media/v4l2-core/videobuf2-core.c | 40 +-
include/media/rc-core.h | 7 -
include/media/videobuf2-core.h | 2 +
15 files changed, 34 insertions(+), 609 deletions(-)
On Tue, Aug 18, 2015 at 1:18 PM, Mauro Carvalho Chehab
<[email protected]> wrote:
>
> - Revert the IR encode patches, as the API is not mature enough.
> So, better to postpone the changes to a latter Kernel;
What the hell have you done with the commit messages?
The first line is completely corrupted for those reverts, and as a
result your own shortlog looks like crap and is completely misleading.
A revert is described as
Revert ".. old patch name goes here .."
but your reverts are broken, and are described as
.. old patch name goes here .."
presumably due to some horribly broken automation crap of yours that
adds the "[media]" prefix or something.
How did you not notice this when you sent the shortlog? Or even
earlier? This is some serious sh*t, since it basically means that your
log messages are very misleading, since the one-liner actually implies
exactly the reverse of what the commit does.
I unpulled this, because I think misleading commit messages is a
serious problem, and basically *half* (and patch-wise, the bulk) of
the commits in this queue are completely broken.
Linus