mbox

[Pull,Request,Natty] sparse-keymap backports to aid in hotkey debugging

Message ID 20110406035554.GA20983@thinkpad-t410
State New
Headers show

Pull-request

git://kernel.ubuntu.com/sforshee/ubuntu-natty.git sparse-backports

Message

Seth Forshee April 6, 2011, 3:55 a.m. UTC
With these patches sparse-keymap will emit MSC_SCAN events for all keys
as well as KEY_UNKNOWN events for keys not in the keymap, allowing for
easy diagnosis of malfunctioning hotkeys using userspace utilities.


The following changes since commit 68270e62f769f441734287e5d629b49c795e20be:

  UBUNTU: Ubuntu-2.6.38-8.41 (2011-04-05 09:39:06 -0700)

are available in the git repository at:
  git://kernel.ubuntu.com/sforshee/ubuntu-natty.git sparse-backports

Seth Forshee (2):
      Input: sparse-keymap - report scancodes with key events
      Input: sparse-keymap - report KEY_UNKNOWN for unknown scan codes

 drivers/input/sparse-keymap.c |   14 ++++++++++++++
 1 files changed, 14 insertions(+), 0 deletions(-)

Comments

Tim Gardner April 6, 2011, 1:05 p.m. UTC | #1
On 04/05/2011 09:55 PM, Seth Forshee wrote:
> git://kernel.ubuntu.com/sforshee/ubuntu-natty.git sparse-backports
>

applied - don't forget to propose for stable (which seems appropriate in 
this case).
Seth Forshee April 6, 2011, 1:39 p.m. UTC | #2
On Wed, Apr 06, 2011 at 07:05:03AM -0600, Tim Gardner wrote:
> On 04/05/2011 09:55 PM, Seth Forshee wrote:
> >git://kernel.ubuntu.com/sforshee/ubuntu-natty.git sparse-backports
> >
> 
> applied - don't forget to propose for stable (which seems
> appropriate in this case).

Huh, I wasn't thinking of these as stable material, since they're just
helpful for troubleshooting and don't really fix anything. Under what
circumstances are non-fix patches accepted in the stable trees? Because
I don't really see it described in stable_kernel_rules.txt.

Thanks,
Seth
Tim Gardner April 6, 2011, 1:54 p.m. UTC | #3
On 04/06/2011 07:39 AM, Seth Forshee wrote:
> On Wed, Apr 06, 2011 at 07:05:03AM -0600, Tim Gardner wrote:
>> On 04/05/2011 09:55 PM, Seth Forshee wrote:
>>> git://kernel.ubuntu.com/sforshee/ubuntu-natty.git sparse-backports
>>>
>>
>> applied - don't forget to propose for stable (which seems
>> appropriate in this case).
>
> Huh, I wasn't thinking of these as stable material, since they're just
> helpful for troubleshooting and don't really fix anything. Under what
> circumstances are non-fix patches accepted in the stable trees? Because
> I don't really see it described in stable_kernel_rules.txt.
>
> Thanks,
> Seth
>

IMHO a stable patch should only address a regression, or be a 
prerequisite for a patch that addresses a regression.

What function do your 2 sparse-keymap patches serve?

rtg
Seth Forshee April 6, 2011, 2:11 p.m. UTC | #4
On Wed, Apr 06, 2011 at 07:54:31AM -0600, Tim Gardner wrote:
> On 04/06/2011 07:39 AM, Seth Forshee wrote:
> >On Wed, Apr 06, 2011 at 07:05:03AM -0600, Tim Gardner wrote:
> >>On 04/05/2011 09:55 PM, Seth Forshee wrote:
> >>>git://kernel.ubuntu.com/sforshee/ubuntu-natty.git sparse-backports
> >>>
> >>
> >>applied - don't forget to propose for stable (which seems
> >>appropriate in this case).
> >
> >Huh, I wasn't thinking of these as stable material, since they're just
> >helpful for troubleshooting and don't really fix anything. Under what
> >circumstances are non-fix patches accepted in the stable trees? Because
> >I don't really see it described in stable_kernel_rules.txt.
> >
> >Thanks,
> >Seth
> >
> 
> IMHO a stable patch should only address a regression, or be a
> prerequisite for a patch that addresses a regression.

That was my thought. This doesn't address a regression.

> What function do your 2 sparse-keymap patches serve?

When some user has hotkey issues and the ACPI/WMI driver for their
machine uses sparse-keymap, these patches make it possible to diagnose
the issues using userpsace utilities like keymap or input-events.
Without these patches sparse-keymap will either emit only the key code
(with no scan code) or nothing at all, and depending on the driver it
might be necessary to provide a special kernel build with some printk's
to find out what scan codes correspond to the malfunctioning hotkeys.

In short, the patches make it easier for us to debug hotkey issues
reported by users. So they don't seem to be stable material.

Seth