At Thu, 12 Dec 2013 02:32:13 -0500, Monty Montgomery wrote:
Monty, would you submit the emi driver patch to upstream?
It's been a few years since I've submitted a patch. Last I tried, upstream seriously frowned on any sort of 'I wrote a whole new driver' type patches. But I'm happy to try if you think it would go through.
I guess merging this code shouldn't be difficult. Of course, it'd be better to fix trivial coding style issues scripts/checkpatch.pl complains beforehand (and also avoid typedefs). Once when ready, I can give my reviewed-by tag, which will make the upstream merge easier.
We can keep the old EMI26 and EMI62 at first but make them conflicting with EMI driver via Kconfig. The external firmware files are maintained in linux-firmware tree nowadays. Once all things get merged, the old emi26/62 can be faded away.
Sure, that sounds perfectly sensible.
FWIW, a patch would be like below:
-- 8< -- --- a/drivers/usb/misc/Kconfig +++ b/drivers/usb/misc/Kconfig @@ -4,7 +4,8 @@ comment "USB Miscellaneous drivers"
config USB_EMI62 - tristate "EMI 6|2m USB Audio interface support" + tristate "EMI 6|2m USB Audio interface support (deprecated)" + depends on !USB_EMI ---help--- This driver loads firmware to Emagic EMI 6|2m low latency USB Audio and Midi interface. @@ -18,7 +19,8 @@ config USB_EMI62 module, say M here and read file:Documentation/kbuild/modules.txt.
config USB_EMI26 - tristate "EMI 2|6 USB Audio interface support" + tristate "EMI 2|6 USB Audio interface support (deprecated)" + depends on !USB_EMI ---help--- This driver loads firmware to Emagic EMI 2|6 low latency USB Audio interface. -- 8< --
Looking through the tarball and codes, my only concern is the license of the new firmware files...
Yes, that's what has always been problematic in the past. I see there's at least a patch now to restore it to the state it was in in early 2.6 when it worked but was just outdated.
That said, I don't want to go through this yet again where one problem gets fixed, everyone gets bogged down in arguing over the rest, and that's the state things stay in. That's where eMagic support has been since roughly 2008. If we're not going to fix it all, and fix it for real, I'd honestly rather not bother. No more partial solutions when the complete solution is already right here.
So before submitting a patch, I'd like to secure agreement over the firmware update from whomever needs to commit. Who would that be? If we can agree on that, I'm in for the rest.
The linux-firmware tree is maintained by Ben Hutchings email@example.com.
Of course, another option would be to distribute from other places. For example, ALSA still keeps some firmware files in alsa-firmware package because of their strange license condition.
I mean, it's not like we have explicit permission to distribute the firmware we're shipping now. :-P
Yeah, but the license issue is always picky (especially some people are), so I won't be surprised even if they won't land to linux-firmware tree but be distributed from somewhere else...