Skip to contentSkip to conversation

ALSA: bebob: Could I still put some patches to 4.5-rc1?

DISCUSSION Takashi Iwai 2 participants 3 messages Jan 18, 2016 1 message not in the archive

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

#1 lore raw
Quoted 11 lines · Takashi Sakamoto

Hi,

Currently I got a help to add support of a new model into ALSA BeBoB driver.
http://sourceforge.net/p/ffado/mailman/ffado-devel/thread/CAKKsO5CYHVRvgo6AV%3D%3DZRJ2it53cF7UGHnnDyCLZY_jzJYHpSA%40mail.gmail.com/#msg34751986

I've already created the patch candidate and users test them.
https://github.com/takaswie/snd-firewire-improve/tree/soundcraft

If the test finished in a few days, could I put them to 4.5-rc1 or not in time?

In general, the development phase is already over when a merge window is opened. That means, only fix patches should be merged after that. As a rule of thumb, I'd take patches to stable kernels at any time, but the rest depends on urgency (or it has been already submitted / reviewed but I forgot to merge, etc).

thanks,

Takashi

#2 lore raw

Hi,

Quoted 11 lines · Takashi Iwai
Quoted 9 lines · Takashi Sakamoto

Hi,

Currently I got a help to add support of a new model into ALSA BeBoB driver.
http://sourceforge.net/p/ffado/mailman/ffado-devel/thread/CAKKsO5CYHVRvgo6AV%3D%3DZRJ2it53cF7UGHnnDyCLZY_jzJYHpSA%40mail.gmail.com/#msg34751986

I've already created the patch candidate and users test them.
https://github.com/takaswie/snd-firewire-improve/tree/soundcraft

Quoted 8 lines
Quoted 2 lines

If the test finished in a few days, could I put them to 4.5-rc1 or not in time?

In general, the development phase is already over when a merge window is opened. That means, only fix patches should be merged after that. As a rule of thumb, I'd take patches to stable kernels at any time, but the rest depends on urgency (or it has been already submitted / reviewed but I forgot to merge, etc).

Thanks for the clear instruction about your maintenance policy. I mostly agree with the policy, so stopped posting my patches during merge window opening (at least in recent releases).

While, the patches allow users to handle their unit. I think it valuable to add the patches in this merge window and increase supported devices in ALSA, as long as the changes is quite small (i.e. just adding device entry).

Today, a tester reports that snd-bebob cannot drive his unit just with the patches. As a result of additional investigation, I conclude that there may be quirks on the sequence of packets from the unit. This means that the changes may become larger, unfortunately.

I give up to request adding the patches into Linux 4.5, and postpone it to next merge window.

Thanks

Takashi Sakamoto

#3 lore raw
Quoted 24 lines · Takashi Sakamoto

Hi,

Quoted 11 lines · Takashi Iwai
Quoted 9 lines · Takashi Sakamoto

Hi,

Currently I got a help to add support of a new model into ALSA BeBoB driver.
http://sourceforge.net/p/ffado/mailman/ffado-devel/thread/CAKKsO5CYHVRvgo6AV%3D%3DZRJ2it53cF7UGHnnDyCLZY_jzJYHpSA%40mail.gmail.com/#msg34751986

I've already created the patch candidate and users test them.
https://github.com/takaswie/snd-firewire-improve/tree/soundcraft

Quoted 8 lines
Quoted 2 lines

If the test finished in a few days, could I put them to 4.5-rc1 or not in time?

In general, the development phase is already over when a merge window is opened. That means, only fix patches should be merged after that. As a rule of thumb, I'd take patches to stable kernels at any time, but the rest depends on urgency (or it has been already submitted / reviewed but I forgot to merge, etc).

Quoted 8 lines

Thanks for the clear instruction about your maintenance policy. I mostly agree with the policy, so stopped posting my patches during merge window opening (at least in recent releases).

While, the patches allow users to handle their unit. I think it valuable to add the patches in this merge window and increase supported devices in ALSA, as long as the changes is quite small (i.e. just adding device entry).

This is the point as I mentioned as "depends". A simple addition of device IDs is usually OK, even with a few more fixes. It implies that such a patch wouldn't give any intrusive changes in the core or the common driver part.

Quoted 7 lines

Today, a tester reports that snd-bebob cannot drive his unit just with the patches. As a result of additional investigation, I conclude that there may be quirks on the sequence of packets from the unit. This means that the changes may become larger, unfortunately.

I give up to request adding the patches into Linux 4.5, and postpone it to next merge window.

OK.

thanks,

Takashi