Messages in this thread | | | Date | Wed, 9 May 2012 19:52:04 +0200 | Subject | Re: [patch 0/5] Bug fix and win8 for hid-multitouch | From | Benjamin Tissoires <> |
| |
Hello Jiri,
On Wed, May 9, 2012 at 4:39 PM, Jiri Kosina <jkosina@suse.cz> wrote: > On Fri, 4 May 2012, benjamin.tissoires wrote: > >> The first patch fixes the bug I discovered lately. This bug was related to the out >> of bound bitfield test, so it concerns every known devices. Ideally, it should go >> to upstream-fixes, but this would requires additional adaptations. >> Jiri, How do you want to proceed? > > When did the bug get introduced, please? i.e. is it a regression?
Well, it's not exactly a regression: it's always been there... ;-)
More seriously, the bug was introduced in the commit "HID: multitouch: fix handling of buggy reports descriptors for Dell ST2220T" (in kernel 3.3) The thing is that this bug fix was not good and was working by miracle.
> > As we are getting close to 3.5 merge window being open, maybe the best > thing to do would be just push it there once it's open, and then > eventually backport it for stable@ if necessary.
That seems fair.
> >> The last 4 patches are the beginning of the support of Win8 devices. >> It's only the beginning as the specification is much more precise than >> the Win7, and we will need to do more tuning in the future. However, >> Win8 devices are retro-compatible with Win7, so they will work out of >> the box though. > > I understand that Henrik had some valid concerns, so I will wait for a > second respin and Henrik's ack, ok?
ok, Thanks, Benjamin
> > Thanks, > > -- > Jiri Kosina > SUSE Labs
| |