| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |
| Who | What | When |
|---|---|---|
| *** | mchehab has quit IRC (Ping timeout: 250 seconds) | [00:51] |
| ........................................................... (idle for 4h50mn) | ||
| koike has quit IRC (Ping timeout: 244 seconds) | [05:41] | |
| ..................................... (idle for 3h3mn) | ||
| Nikhil_D has quit IRC (Remote host closed the connection) | [08:44] | |
| hverkuil | mchehab: this is the linux-next repo, right: git://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git
and the master branch is the right branch to check out? I ask because the input patches of my CEC series apply cleanly there, but you had conflicts. So now I am confused. | [08:55] |
| ........... (idle for 52mn) | ||
| *** | agd5f has quit IRC (Ping timeout: 244 seconds) | [09:48] |
| ......................................... (idle for 3h23mn) | ||
| mchehab | hverkuil: I didn't actually tried to apply the patch. Just looked at the file via cgit
Still, I think that it is better/safer to send such patch via input tree | [13:11] |
| .... (idle for 19mn) | ||
| hverkuil | mchehab: do you want both patch 1 and 2 to go via input tree? How to handle this so patches 3 and up can still be merged in our tree? | [13:31] |
| mchehab | yes, patches 1 and 2
the RC map file can also be send to his tree, with my ack well, patch 1 actually seems ok to go via my tree as I doubt there are much updates for new types | [13:31] |
| hverkuil | I would prefer to have just patch 2 and 3 go via Dmitry. I'll have to double-check, but I think those are independent of the remainder of the series.
(compiling now without patches 2 & 3) hmm, dependency on RC_TYPE_CEC. | [13:33] |
| mchehab | patch 1 can go on both trees, I guess | [13:35] |
| hverkuil | OK, if I split patch 3 in two: one patch for rc-cec and one for the remainder, then I think it should work OK. | [13:35] |
| OK, just to check: if I split patch 3 into two: one patch for rc-cec.c and one for rc-main/rc-map.h, then patch 2 can go via Dmitry's tree and the rc-cec patch would be on hold in our tree until the input-event-codes.h changes are merged for 4.8. Everything else can be merged in our master. You just don't have the rc-cec keymap module.
(rc-cec.c depends on the rc-map.h changes AND on the input-event-codes.h changes) | [13:45] | |
| mchehab | ok
hmm... maybe it is better to just merge patch 1 on both this s/this/trees/ this way, I don't need to put it on a separate topic branch | [13:47] |
| hverkuil | I propose that I split up patch 3 as described above and that I post a v18 (I just fixed a s5p-cec warning as well that I missed) with a CC to linux-input.
Once done, you can Ack the input patch(es) that you want Dmitry to handle. You can then merge the cec patches except the one for the rc-cec keymap. Alternatively, I can also #if 0 the new keys in rc-cec.c so it can be merged in our tree, and for 4.8 a follow-up patch is added that removes the #if 0's. | [13:54] |
| mchehab | #if 0 works for me
well, it also works to merge the rc-cec late you mean the rc-cec keymap patch late, right? | [13:59] |
| hverkuil | right | [14:00] |
| mchehab | ok, let's do that | [14:00] |
| hverkuil | OK. | [14:00] |
| mchehab | can the rc-cec be merged via Dmitry's tree?
if so, we can send him patches 1 and 2 + rc-cec keymap and merge the rest via our tree | [14:01] |
| hverkuil | well, rc-cec depends on both the input-event-codes.h change AND the rc-map.h change, and the cec framework depends on rc-map.h.
It makes no sense to me to have Dmitry handle rc-cec.c. I think it is easiest to just keep rc-cec.c aside until both the input and media changes are merged in 4.8. | [14:02] |
| mchehab | I don't see any problem asking dmitry to merge patch 3 as a hole
it won't cause any conflicts on our side of course, I'll need to sign it or ack | [14:06] |
| hverkuil | you're talking about patch 3 as I posted it in v17? | [14:07] |
| mchehab | the remaining patches on this series depend on patch 1-3 to build?
yes | [14:07] |
| hverkuil | As it stands now in v17, the remaining patches depend on 1-3. | [14:07] |
| mchehab | what's de dependency?
+#define RC_MAP_CEC "rc-cec" ? | [14:08] |
| hverkuil | But only rc-cec.c depends on input-event-codes.h. | [14:08] |
| mchehab | I mean: what patches 4-16 depends on the stuff at 1-3? | [14:09] |
| hverkuil | the cec framework depends on patch 1 (BUS_CEC) and RC_BIT_CEC, RC_MAP_CEC and RC_TYPE_CEC defines from rc-map.h.
(patch 9) | [14:09] |
| mchehab | ok. let's then put patch 1 and a patch adding those 3 defines on a stable topic branch...
asking Dmitry to pull from it and merge patch 1 and 3 (without the above) on his tree | [14:11] |
| hverkuil | I think this is overly complex.
If I split patch 3 in two, one updating rc-main.c and rc-map.h and the other for rc-cec.c, then all that needs to be kept aside is the rc-cec patch. No need to keep a big topic branch. My opinion. Besides, rc-cec.c itself depends on RC_TYPE_CEC as well. and RC_MAP_CEC. | [14:11] |
| mchehab | ok
let's do that | [14:15] |
| hverkuil | You were planning to merge patch 1 through both trees, right? Since patch 9 depends on that.
(both trees == linux-input and linux-media) | [14:17] |
| mchehab | works for me
just let me and Dmitry to know what patches should go to each tree IMHO, the best is if you send 2 patch series: one for me; another for Dmitry I"ll ack on his series | [14:21] |
| hverkuil | OK, sounds good.
I'll do a third patch series as well with just the rc-cec patch, to keep it separate from the other two. Is there anything else I should do before I spam the lists with a v18? I hate having to do a v19 :-) | [14:24] |
| pinchartl | hverkuil: v18 ? looks like the patch set reached adulthood :-) | [14:26] |
| hverkuil | it does feel like that :-) | [14:27] |
| mchehab | I guess that's it | [14:31] |
| hverkuil | OK. I'm preparing the patch series. | [14:32] |
| pinchartl | pinchartl goes back to implementing rotation in the vsp driver
I got it half working http://www.ideasonboard.org/vsp-rotation.png but there seems to be still something missing | [14:35] |
| .... (idle for 16mn) | ||
| hverkuil | mchehab: posted the patch series for linux-input. | [14:51] |
| Hmm, the original RFC for CEC was posted March 2010. So this patch series is the equivalent of a nice 6 year old whisky :-)
posted v18 of the patch series. mchehab: do you want a pull request as well for v18? I'll just post one. | [15:04] | |
| ..................................... (idle for 3h2mn) | ||
| *** | _abbenormal has quit IRC (Read error: Connection reset by peer) | [18:09] |
| .................................... (idle for 2h57mn) | ||
| Bladelouse_ has quit IRC (Ping timeout: 240 seconds) | [21:06] | |
| ........................... (idle for 2h10mn) | ||
| WeaselSoup has quit IRC (Quit: WOOPS) | [23:16] | |
| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |
