<!-- Some styling for better description lists --><style type='text/css'>dt { font-weight: bold;float: left;display:inline;margin-right: 1em} dd { display:block; margin-left: 2em}</style>

   mchehab: -:73: WARNING: msleep &lt; 20ms can sleep for up to 20ms; see Documentation/timers/timers-howto.rst
   <br> @all, there's now a SPDX header for GFDL
   <br> (well, it is actually a commit not merged yet at SPDX spec, but it is a result of ~2 years of discussion there - so I'm starting to use it)
   <br> I submitted a patch fixing it, but I would prefer to have this one merged via the docs tree
   <br> So, old userspace-api docs should have: SPDX-License-Identifier: GFDL-1.1-or-later-no-invariants
   <br> while new ones should be dual-licensed:
   <br> SPDX-License-Identifier: GPL-2.0 OR GFDL-1.1-or-later-no-invariants
   <br> 
   <br> <u>hverkuil</u>: btw, a pull request from you has problems with that regards:
   <br> +++ b/Documentation/userspace-api/media/v4l/dev-encoder.rst
   <br> @@ -0,0 +1,728 @@
   <br> +.. SPDX-License-Identifier: GPL-2.0
   <br> as this is included inside the media uAPI, we need the file to be dual-licensed
   <br> we'll need Tomasz Figa ack on such change
   <br> as he's signing the original patch
   <br> btw, it took a lot of time for me to pull from your tree... it downloaded 1,2 GB
   hverkuil: <u>mchehab</u>: I asked Tomasz if this change it OK.
   <br> is OK
   mchehab: ok. please send then a followup patch with his SOB
   <br> I'll write an e-mail just to let this registered at the ML.
   <br> email sent
   hverkuil: well, it's now in duplicate :-)
   mchehab: :-)
   <br> sorry for not checking it before sending... very busy here... opened a window to handle pending PRs, as I don't want them to stay pending for a long time
   sailus: <u>mchehab</u>: Sleeping a little longer is not a problem in the two patches, really.
   mchehab: <u>sailus</u>: if this is not a problem, use msleep(20)...
   <br> or use the hires clock variants
   <br> usleep_range()
   <br> we had bugs in the past on media due to bad usages of msleep()
   <br> causing some code to stop working depending on how CONFIG_HZ was set
   <br> so better to never use msleep(low_delay) as this is just a call for problems ;-)
   <br> (it may not affect your code, but someone could copy it on some other code, and then the other code could be affected)
   <br> so, better to not give bad examples :-D
   <br> if anything between 10ms and 20ms is ok, just replace them by:
   <br> usleep_range(10000, 20000);
   <br> <u>hverkuil</u>: hmm... did you receive the Tomas SOB? I saw your e-mail to the ML, but not his reply
   hverkuil: <u>mchehab</u>: I didn't get a reply.
   <br> Was it mailed to you directly? That's fine by me as well.
   sailus: <u>mchehab</u>: To be fair, I agree this is not exemplary use of msleep. Could you still pull the set nonetheless? I'll ask the original submitter to address this in a follow-up patch.
   mchehab: <u>hverkuil</u>: no, he didn't. Oh, I misread your note:
   <br> (12:37:15) hverkuil: mchehab: I asked Tomasz if this change it OK.
   <br> somehow I understood that "you asked Tomasz and he agreed"
   <br> ok. I placed the PR on hold here... I'm handling the next ones
   <br> <u>sailus</u>: ok
   hverkuil: <u>mchehab</u>: Tomasz replied with his SoB. If you want, you can just take my PR and change the license before applying. I can also post a new patch, but I'm going into a meeting and so that will take longer.
   mchehab: no need. I have it already on my tree, and pulling from your PR is taking too long
   <br> not sure why some trees are taking too long to pull
   <br> probably due to that gcc version2 bug
   <br> ops
   <br> I meant to say git version 2 protocol bug
   <br> <u>hverkuil</u>: could you please reply there with your SOB too?
   <br> gah, Kate Stewart asked us to refrain use the new SPDX header for now... it sounds that they'll change it before merging at SPDX specs
   <br> I guess we'll need to keep using the old way for more time
   <br> <u>hverkuil</u>: sent the patch I'm intending to apply. Please answer with your SOB
   <br> I'm handling it the old way, until the discussions at SPDX upstream finishes
   hverkuil: Replied.
   mchehab: ok, all pending PRs applied
   hverkuil: Fantastic.
   sailus: <u>mchehab</u>: Obrigado!
   mchehab: I suspect that the SPDX will be solved soon, but Kate asked to change the name of the GFDL SPDX markup....
   <br> it will probably be:
   <br> GFDL-1.2-no-invariants-or-later
   <br> https://github.com/spdx/license-list-XML/pull/1048#issuecomment-648135498