On Tue, Aug 11, 2009 at 5:57 PM, Michael Krufky<mkrufky@xxxxxxxxxxxxxx> wrote: > On Tue, Aug 11, 2009 at 1:01 PM, Mauro Carvalho > Chehab<mchehab@xxxxxxxxxxxxx> wrote: >> Em Sat, 1 Aug 2009 14:03:24 -0400 >> Michael Krufky <mkrufky@xxxxxxxxxxxxxx> escreveu: >> >>> On Sat, Aug 1, 2009 at 12:40 AM, Mauro Carvalho >>> Chehab<mchehab@xxxxxxxxxxxxx> wrote: >>> > Em Fri, 31 Jul 2009 11:25:55 -0400 >>> > Michael Krufky <mkrufky@xxxxxxxxxxxxxx> escreveu: >>> > >>> >> Please review this patch yourself -- you will see I am simply removing >>> >> Hauppauge-specific handling that was incorrectly added by Uri, >>> >> resulting in the crippling of these devices under Linux. >>> > >>> > Fair enough. I'll apply the patch. Yet, we should address with Udi what else is >>> > needed in order to sync their tree with ours without breaking support for any >>> > existing device. >>> > >>> >> I cant stress enough how important it is that this changeset gets >>> >> merged upstream to Linus asap. The 2.6.31-rc kernel is broken without >>> >> this change. >>> > >>> > I'll add it on my next upstream changeset. >>> >>> >>> Thanks for applying the fix patch. >>> >>> Now, at least those devices will work again, but there is still a >>> regression since the previous kernels supported the LED and LNA >>> functionality that my next patch restores. >>> >>> Hopefully we will hear from Udi soon. >>> >>> Just to be clear, the patch that we'd like Udi's comments on is this one: >>> >>> http://kernellabs.com/hg/~mkrufky/sms1xxx/rev/tip >>> >>> As this code is in the smsdvb common code, it calls into sms-cards and >>> will return harmlessly to the caller on the non-hauppauge devices. >>> The functionality is only changed for those cards that have this >>> functionality defined in the sms-cards.c structures. >>> >>> Hopefully, Udi can agree to merge this into the 2.6.31 kernel, while >>> we can work on Siano's internal event interface for the next kernel. >>> Once that is working perfectly, we can remove the patch that I'm >>> proposing now, and convert the Hauppauge devices to the newer event >>> interface. >> >> Since: >> >> a) the pull request were sent on Jul, 28; >> b) we didn't have any answer from Siano up to today; >> c) a closer analysis showed that this patch won't affect non Hauppauge devices; >> >> I'm merging the fix today. >> >> Udi, the better is to work at the event interface in a way that it won't cause >> troubles with the existing Hauppauge devices. After having it done, we may just >> remove the legacy SMS code > > Mauro, > > Thank you for merging the fixes. Please be sure to send them both to > Linus for the 2.6.31 kernel, as these patches actually fix regressions > introduced only in this 2.6.31 kernel. > > Once the event interface works properly, I will be very happy to port > the existing device-specific GPIO handling functionality to it. This > will be a nice improvement in flexibility and in code cleanliness. > > Udi, Please let me know once this is ready -- I look forward to the testing. > > Regards, > > Mike > Mauro, Would you please send the final GPIO fix to Linus to fix the regression on the Hauppauge devices. 2.6.31 is in -rc7 now, and I would hate for this regression to not yet be solved before the kernel is released. I apologize for nagging -- I just prefer for fixes to be merged upstream *before* a kernel release. Thank you. Just as a reminder, the changeset that is still waiting for upstream merge is: * sms1xxx: restore GPIO functionality for all Hauppauge devices Thanks & regards, Mike Krufky -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
