path: root/wiki/Chat_log/20160608-mm-chatlog

Multimedia-chat-meeting-2016-06-08

<neg> hi all  [16:52]
<morimoto> hi  [16:53]
<uli___> hello
*** dammsan (~dammsan@s214090.ppp.asahi-net.or.jp) has joined channel
    #periperi  [16:55]
<morimoto> hi mag-mag
<dammsan> hi mori-mori
<kbingham> Morning'
<morimoto> Hi kbingham  [16:56]
<dammsan> hi!
<kbingham> I guess it's 'good afternoon' for morimoto + dammsan :)  [17:01]
<morimoto> Yes :)  [17:02]
<pinchartl> sorry, I'm late
<pinchartl> hello everybody
<morimoto> hi
<kbingham> Hola!  [17:03]
<pinchartl> I don't know if it's a good sign or not, but last night I was
	    dreaming we were having this meeting :-)
<kbingham> pinchartl: Well it was accurate ?
<kbingham> were any lottery numbers mentioned?  [17:04]
<pinchartl> it involved people crashing at my place and three different IRC
	    clients, so hopefully not
<morimoto> It is called as "nightmare" in Japan :)
<pinchartl> :-)  [17:05]
<pinchartl> so, topics for today
<pinchartl> Topic 1. Status check for the multimedia tasks  [17:06]
<pinchartl> Topic 2. Additional '50%' tasks (both May and June)
<pinchartl> Topic 3. Next meeting
<pinchartl> anything else ?
<dammsan> i have one  [17:07]
<dammsan> ideas for upcoming additional tasks for next quarter
<pinchartl> OK  [17:08]
<pinchartl> let's start with the status update  [17:09]
<pinchartl> in alphabetical order again, which I'll try not to mess up this
	    time
<pinchartl> kbingham: you're first
<pinchartl> FDP,v4.8,plan,kieran,Develop and upstream driver
<kbingham> I have the FDP1 powered up and successfully performing streaming
	   format conversion from all my tested inputs to all my tested
	   outputs.  [17:10]
<kbingham> It is not de-interlacing yet - as that is the next step
<kbingham> but as we now have  driver which is functional and performs 'a job'
	   I intend to post this as V1 to commence overall reviews, and while
	   that goes on, I will finish off the de-interlacing parts.
								        [17:11]
<kbingham> Any questions or more detail desired?  [17:12]
<pinchartl> when do you plan to post the driver ?
<kbingham> I hope either end of today or tomorrow morning  [17:13]
<kbingham> so .. imminently :)
<pinchartl> nice :-)
<pinchartl> and your target for de-interlacing ?
<kbingham> I'll be starting to push buffers through with both fields, and I
	   hope to have some basic functionality by the end of the week
								        [17:15]
<pinchartl> perfect
<pinchartl> so we're on track  [17:16]
<kbingham> but that will be only 2d de-int, only supporting interleaved
	   buffers containing a top and a bottom field to start with to keep
	   it simple.
<kbingham> Yes, I hope so :)
<pinchartl> next, me
<pinchartl> DU,?,plan,laurent,IPMMU integration on Gen3
<pinchartl> DU,?,plan,laurent,IPMMU support on Gen3 (through VSPD+FCP)
<pinchartl> no work done  [17:17]
<pinchartl> DU,v4.8,public,laurent,VSPD Z-order support upstream (Gen3)
<pinchartl> I've pinged Mauro to get this merged, without success so far
								        [17:18]
<pinchartl> I'll keep trying
<pinchartl> VSP,?,plan,laurent,Fixed alpha support (VI6_DPR_*_ROUTE.FXA)
								        [17:19]
<pinchartl> no progress
<pinchartl> VSP,?,plan,laurent,CLU WARN_ON fix
<pinchartl> VSP,?,plan,laurent,CLU 2D and 3D mode support
<pinchartl> VSP,?,plan,laurent,CLU/LUT test application
<pinchartl> VSP,?,plan,laurent,CLU/LUT upstream API
<pinchartl> I've implemented and tested this  [17:20]
<pinchartl> I will post the patches today
<morimoto> nice !
<pinchartl> the target is v4.8
<morimoto> "?" -> "v4.8" ?  [17:21]
<pinchartl> yes
<pinchartl> VSP,?,plan,laurent,UDS regression fix
<pinchartl> I've reproduced the problem but haven't been able to fix it yet
<morimoto> Do you have any plan for it ?
<pinchartl> yes, I plan to continue working on it next week  [17:22]
<morimoto> nice !
<pinchartl> I'm not sure when a fix will be available as I don't know the root
	    cause yet  [17:23]
<pinchartl> VSP,v4.8,plan,laurent,Fix suspend/resume crash
<pinchartl> no progress
<pinchartl> this will likely be delayed as it hasn't been included in the 50%
	    tasks for June
<pinchartl> VSP,v4.8,public,laurent,HGO operation mode selection  [17:24]
<pinchartl> VSP,v4.8,public,laurent,HGO support upstream on Gen3
<pinchartl> VSP,v4.8,public,laurent,HGO test application
<pinchartl> I've incorporated all review comments and will post a pull request
	    as soon as I can get Mauro to pull the other pending patches
								        [17:25]
<pinchartl> VSP,v4.8,public,laurent,V4L2 request API usable prototype  [17:26]
<pinchartl> I have a new test application
<pinchartl> the latest code and test application will be posted by the end of
	    the week
<morimoto> How about this ?  [17:27]
<morimoto> VSP,2016-06-30,request,laurent,update request API on renesas-driver
<morimoto> 
<morimoto> do you have this plan ?
<pinchartl> I'll ask Geert to merge my working branches  [17:28]
<morimoto> OK, I love you 
<pinchartl> :-)
<pinchartl> that's it for me  [17:29]
<pinchartl> dammsan: your turn
<pinchartl> VIN,?,plan,magnus,IPMMU integration on Gen2
<pinchartl> VIN,?,plan,magnus,IPMMU integration on Gen3
<pinchartl> VIN,?,plan,magnus,IPMMU support on Gen2
<pinchartl> VIN,?,plan,magnus,IPMMU support on Gen3
<pinchartl> no update I assume ?
<dammsan> yes, no progress
<morimoto> nice !
<dammsan> except i fixed Gen2 IPMMU support a little while ago
<dammsan> will poke around with VIN next week
<pinchartl> was it broken ?  [17:30]
<dammsan> i broke it in my multiarch v3 series
<dammsan> so i posted a v4
<pinchartl> ok :-)
<dammsan> v3 and v4 include     grouping of devices
<dammsan> based on their parent IPMMU instance
<pinchartl> I've seen the patches, I'll try to review them  [17:31]
<dammsan> however i'd like to add tasks for audio-dmac and IPMMU =)
<dammsan> thanks!
<dammsan> it would be nice to get the multi-arch code accepted at some point
<pinchartl> would you like to add the tasks now ?
<dammsan> i'll try to keep my focus on this
<dammsan> yes, that would be nice
<pinchartl> go ahead, I'm all ears  [17:32]
<dammsan> morimoto: how do you describe IPMMU integration for sound?
<dammsan> what is the device called? Audio-DMAC?
<morimoto> yes, but it using "SYS-DMAC".  [17:33]
<dammsan> i propose copy-paste the VIN IPMMU tasks and use Audio-DMAC instead
<morimoto> I mean driver
<pinchartl> right, it's not audio-dmac, but sys-dmac  [17:34]
<dammsan> ?
<dammsan> it is a separate DMAC for audio, right?
<morimoto> it is AUDIO-DMAC
<morimoto> Yes, defferent DMAC, but using same driver, I mean  [17:35]
<pinchartl> sorry, rcar-dmac
<pinchartl> and the driver already has IPMMU support, doesn't it ?
<morimoto> (many local talk happen in Renesas, sorry)  [17:36]
<dammsan> maybe so
<dammsan> but simply enabling it does not work
<dammsan> either issue with the IPMMU driver
<dammsan> or the DMA Engine slave side
<neg> I'm still trying to get the MMIO slave + IPMMU stuff merged  [17:37]
<dammsan> neg: thanks
<dammsan> AUDIO-DMAC,?,plan,magnus,IPMMU integration on Gen2
<dammsan> AUDIO-DMAC,?,plan,magnus,IPMMU integration on Gen3
<dammsan> AUDIO-DMAC,?,plan,magnus,IPMMU support on Gen2
<dammsan> AUDIO-DMAC,?,plan,magnus,IPMMU support on Gen3
<dammsan> it needs more effort anyway =)
<pinchartl> ok
<dammsan> that's it from my side  [17:38]
<pinchartl> morimoto: your turn  [17:39]
<pinchartl> RSND,2016-06-30,public,morimoto,DT bindings for HDMI sound
<pinchartl> RSND,2016-06-30,plan,morimoto,dw-hdmi-ahb-audio prototype on Gen3
<pinchartl> RSND,2016-06-30,plan,morimoto,HDMI SSI prototype on Gen3
<pinchartl> RSND,2016-06-30,plan,morimoto,HDMI sound Upstream support without
	    hotplug on Gen2
<pinchartl> RSND,2016-09-30,plan,morimoto,Hotplug support upstream on Gen3
<morimoto> for graph DT support on ALSA side, as 1st step, I posted sound card
	   cleanup patches to ML
<morimoto> I posted v2 now, but no responce at this point  [17:40]
<morimoto> It it was accepted, next step is I will post graph DT support for
	   ALSA SoC
<morimoto> it is already working in my local PC
<morimoto> but it is based on above cleanup patches, so, I'm now waiting.
								        [17:41]
<morimoto> And, after that, I can post HDMI sound support patches to ML
<morimoto> this is localy working, but have same background. So I'm waiting
<pinchartl> some background ?
<morimoto> About graph DT on ALSA, I have 1 concern
<morimoto> If video/sound both use graph DT, ALSA side want to know which port
	   is sound-endpoint  [17:42]
<morimoto> pinchartl: sorry /some/same/
<pinchartl> sorry, I misread :-)
<morimoto> I don't know how to solve this video/sound endpoint issue
<morimoto> 1 idea is adds new .type property.  [17:43]
<morimoto> like .type = sound
<morimoto> or, use new endpoint, like sound-endpoint
<morimoto> But I don't know. no response, no review  [17:44]
<morimoto> that's it from me
<pinchartl> shouldn't the type be a property of the port, not of the endpoint
	    ?
<morimoto> port/endpoint anything is OK.
<morimoto> just idea  [17:45]
<pinchartl> ok
<morimoto> I think CC:ed you 
<pinchartl> yes you did  [17:46]
<pinchartl> I haven't had time to review the patches I'm afraid
<pinchartl> I'll see what I can do
<morimoto> no problem
<morimoto> basically, current sound is using "simple" card. and my cleanup is
	   for it.
<pinchartl> ok  [17:47]
<morimoto> graph support will be on "simple" card + graph feature
<pinchartl> neg: you're next  [17:48]
<pinchartl> ADV7482,v4.7,plan,niklas,Prototype on Gen3
<pinchartl> ADV7482,v4.9,plan,niklas,Gen3 support upstream
<pinchartl> ADV7482,v4.9,plan,niklas,Interlace support upstream
<pinchartl> VIN,v4.7,plan,niklas,CSI2 prototype (Gen3)
<pinchartl> VIN,v4.8,plan,niklas,Gen3 support
<pinchartl> VIN,v4.8,plan,niklas,Scaler support (on Gen3)
<pinchartl> VIN,v4.8,public,niklas,New VIN driver without soc-camera (tested
	    on Gen2)
<pinchartl> VIN,v4.9,plan,niklas,CSI2 interlace support upstream (Gen3)
<pinchartl> VIN,v4.9,plan,niklas,CSI2 support upstream (Gen3)
<pinchartl> VIN,v4.9,plan,niklas,Gen3 support upstream (without CSI-2)
<neg> VIN Gen3 is public, but no review comments so far which is not so good
      since it was a rather large series  [17:49]
<neg> CSI-2 are almost done and no work started on ADV7482 other then make it
      compile outside BSP  [17:50]
<pinchartl> regarding CSI-2, does that include interlace support ?  [17:51]
<neg> will hold of a bit on scaler and interlace support until I get some Ack
      that the driver itself looks ok with regard to the new group concept
      introduced for Gen3
<pinchartl> ok
<neg> yes CSI-2 includes interlace support as far as I can test it. It is only
      needed to figure out the bus speed AFIK  [17:52]
<pinchartl> when you say almost done, is that prototype code, or code ready to
	    be submitted upstream ?  [17:53]
<neg> I really would like to get some reviews on the VIN driver before I say
      CSI-2 is upstream ready since they are quiet interconnected
<neg> but the code itself for CSI-2 I'd say is ready for upstream  [17:54]
<pinchartl> ok
<pinchartl> can you ping me on VIN review at the beginning of next week ?
<neg> will do
<pinchartl> thanks  [17:55]
<dammsan> about the "New VIN driver" task, is v4.8 still likely?
<pinchartl> I believe so, it nearly made it to v4.7  [17:56]
<neg> yes I think so, Hans sent the pull request to mauro for v4.7 but we
      missed it. Have not seen anything sayng it wont be accepted for v4.8
      (that is for the Gen2 driver)
<dammsan> good, thanks!!
<pinchartl> how about "VIN,v4.7,plan,niklas,CSI2 prototype (Gen3)" ?  [17:57]
<neg> I hope to post the prototype early next week and if the reviews are OK I
      think it is ready for upstream  [17:58]
<dammsan> we are waiting for paperwork, i'm the reason for slow handling
<dammsan> will finalise tonight
<dammsan> just need to be able to leave the renesas office
*** horms (~horms@reginn.isobedori.kobe.vergenet.net) has quit: Quit: Leaving
<neg> but as I state above CSI-2 might need more work if the VIN Gen3 driver
      group concept gets rejected  [17:59]
<pinchartl> ok
<pinchartl> uli___: you're next  [18:00]
<pinchartl> DU,v4.7,plan,ulrich,Atomic API test program
<pinchartl> DU,v4.7,public,ulrich,HDMI output on Gen3 prototype
<pinchartl> DU,v4.7,prototype,ulrich,Test setup with HDMI output to HDMI input
	    loopback (without EDID)
<pinchartl> DU,v4.7,public,ulrich,EDID generation support for the HDMI
	    loopback test setup
<pinchartl> DU,v4.9,plan,ulrich,HDMI output on Gen3 upstream
<pinchartl> VIN,v4.8,public,ulrich,Add DV timings support to rcar-vin
<uli___> api test is public
<uli___> as is the new hdmi out prototype
<uli___> no progress on the rest  [18:01]
<pinchartl> I don't think I've been CC'ed, where have you posted them ?
<uli___> the soc mailing list
<uli___> yes, no cc's, sorry about that  [18:02]
<neg> I included all your VIN work for the Gen3 driver except the EDID parts
<pinchartl> no worries
<uli___> neg: saw that, thanks
<morimoto> do you have plan to update HDMI out for ES1.1 ?
<morimoto> I sent it on periperi ML  [18:03]
<uli___> yes, but not this month
<uli___> we postponed that
<uli___> the loopback test goes first
<morimoto> OK
<pinchartl> are we still on track with
<pinchartl> DU,v4.7,prototype,ulrich,Test setup with HDMI output to HDMI input
	    loopback (without EDID)
<pinchartl> DU,v4.7,public,ulrich,EDID generation support for the HDMI
	    loopback test setup  [18:04]
<pinchartl> will they be complete by the end of this month ?
<uli___> should be ok
<pinchartl> ok, thanks
<pinchartl> that's it then
<pinchartl> next topic
<pinchartl> Topic 2. Additional '50%' tasks (May & June)
<pinchartl> let's start with May  [18:05]
<kbingham> sorry laurent
<pinchartl> yes ?
<kbingham> Could I ask one more Q on status :) - What is the status of the FCP
	   patches ?
<pinchartl> sure
<pinchartl> do you mean the ones you've posted ?  [18:06]
<kbingham> pinchartl: No - the ones you posted :)
<kbingham> Are they already en route to ML?
<pinchartl> I've sent a pull request for v4.7, it got rejected on the grounds
	    of being sent too late, and I'm trying to get it in v4.8
								        [18:07]
<kbingham> Ok. That clears that up for me :)
<pinchartl> so, additional tasks for May  [18:08]
<pinchartl> has everybody sent their report ?
<neg> yes, report is sent and test procedure on elinux.org
<pinchartl> great  [18:09]
<uli___> report and invoice sent :) also, instructions on elinux.org
<pinchartl> where on elinux.org by the way ?
<neg> http://elinux.org/R-Car/Tests:rcar-vin  [18:10]
<uli___> http://elinux.org/User:Uli/Tests:KMS  [18:11]
<dammsan> uli___: about resending the HDMI prototype, can you please try to do
	  it before the end of this month?
<dammsan> i'm asking because we want to check the result  [18:12]
<uli___> i'll try to squeeze it in then
<dammsan> and timing wise if it happens next month it is too late to decide
	  next quarter additional task
<dammsan> thanks, please do!!
<pinchartl> uli___: could you move http://elinux.org/User:Uli/Tests:KMS to
	    http://elinux.org/R-Car/Tests:KMS ?  [18:13]
<pinchartl> and add a link to http://elinux.org/R-Car/Devices ?
<dammsan> pinchartl: please keep the URLs half-stable, I've forwarded these to
	  people inside renesas for checking
<morimoto> dammsan: uli___: I believe it is for "HDMI-OUT for ES1.1" ?
<pinchartl> dammsan: we can add a redirection
<dammsan> pinchartl: sounds good, thanks!  [18:14]
<pinchartl> neg: could you add a link to the VIN tests to
	    http://elinux.org/R-Car/Devices ?
<uli___> will do
<neg> pinchartl: sure
<pinchartl> thank you
<pinchartl> you can add sections for VIN and DU in
	    http://elinux.org/User:Uli/Tests:KMS  [18:15]
<pinchartl> oops, I mean in http://elinux.org/R-Car/Devices
<pinchartl> no need for a long device description for now
<dammsan> uli__: can you please include the ES1.1 support in your HDMI out
	  prototype update?
<pinchartl> just create Tests section that link to the test pages
<neg> for future reference should we use 'R-Car/Tests:foo' ? There are also
      for example 'Tests:SCIF-FIFO'
<uli___> dammsan: yes, that's the one thing  [18:16]
<dammsan> great, thank you!
<pinchartl> neg: that sounds good to me
<pinchartl> now, about additional tasks for June  [18:17]
<pinchartl> are we on track ?
<pinchartl> any issue ?
<uli___> i'm ok so far  [18:18]
<neg> I'm a bit scared of doing both HDMI and i2c secondary device in one go
      for ADV7482 but have not yet started so maybe the i2c part is easy :-)
      Will let you know if I run into trouble  [18:19]
<pinchartl> ok. please let me know as early as possible
<neg> will do, thanks  [18:20]
<pinchartl> the I2C part should be easy
<pinchartl> Topic 3. Additional tasks for next quarter  [18:21]
<neg> good, then I feel I'm on track
<pinchartl> dammsan: do you want to drive this ?
<dammsan> thanks
<dammsan> i'd like to specify tasks in two batches
<dammsan> one i'd like decide in the middle of this month
<dammsan> the next around meeting time in July
<dammsan> the deadlines for the two batches are supposed to be 8/M and 9/M
								        [18:22]
<dammsan> we may be able to do remote access for M3-W for the 8/M tasks
<dammsan> but real physical hardware access will not be possible for 8/M
	  target  [18:23]
<dammsan> so instead of focusing on the entire quarter i'd like to begin with
	  focusing on the 8/M target that we should decide this month
<pinchartl> will it be possible for 9/M ?
<dammsan> maybe
<dammsan> kind of unlikely to be honest
<dammsan> we will know more during the meeting in japan  [18:24]
<dammsan> as for what kind of tasks to deal with
<dammsan> i'd like to get a bunch of proposals
<pinchartl> ok
<dammsan> both from you guys
<dammsan> and also ask internal renesas if they have something for us  [18:25]
<dammsan> we will meet f2f with multimedia people next week
<dammsan> hopefully that will result in an updated list
<pinchartl> ok
<pinchartl> should we start now, or would you like to get them by e-mail ?
								        [18:26]
<dammsan> i'l like you to start the process in your head
<dammsan> but after that email would be good =)
<dammsan> also agreeing on a schedule for fixing tasks would be good  [18:27]
<dammsan> when do you intend to hold next chat meeting?
<pinchartl> in two weeks
<pinchartl> 22nd of June
<dammsan> thanks  [18:28]
<pinchartl> is that fine for everybody by the way ?
<pinchartl> same time as today
<dammsan> so can we have title proposals done by the 17th?
<pinchartl> (morimoto: thanks for your e-mail with the status update, I've
	    incorporated it)
<dammsan> for some overview
<neg> pinchartl: time and date works for me  [18:29]
<morimoto> pinchartl: thanks
<dammsan> pinchartl: scrap the 17th
<dammsan> pinchartl: i think we should provide you a list of requests from
	  renesas on the 17th
<dammsan> can we have the titles done by the 22nd then?  [18:30]
<pinchartl> that's fine with me
<dammsan> thanks
<dammsan> so the time around 20th and later is needed to fix some tasks
								        [18:31]
<neg> Hum are there any information about schematics for M3-W? To look for
      tasks it would help me to be able to look at how VIN is setup
<pinchartl> neg: uli___: kbingham: could you start thinking about tasks
	    proposals for Q3 in parallel ? I'd like to hear your ideas on what
	    is needed for multimedia support in the areas you're responsible
	    for (and possibly other areas as well)  [18:32]
<pinchartl> dammsan: as we've discussed previously, video codecs, video test
	    framework (extension of the VSP test framework to cover other
	    devices and more features) and suspend/resume support for the VSP
	    driver are possible candidates for Q3  [18:33]
<kbingham> pinchartl: Ack. I'll have an explore around to see what's what :)
								        [18:34]
<pinchartl> kbingham: thanks
<morimoto> neg: basically H3 Salvator <-> M3 Salvator are same board. I can
	   say "Basically" :)
<morimoto> pinchartl: now I got VSP1 bug-report from BSP team. I will report
	   it to you tomorrow  [18:36]
<pinchartl> morimoto: thank you. how many dozens of pages are there ? :-)
								        [18:37]
<neg> morimoto: ok thanks, so its the same utilisation of CSI-2 and no digital
      video source for VIN?
<morimoto> neg: if it was so in H3, basically yes  [18:38]
<morimoto> pinchartl: let me check
<pinchartl> we've covered all the topics for today. any last comment/question
	    from anyone ?
<dammsan> nope, thanks for your help!
<neg> I'm happy, thanks all  [18:39]
<morimoto> pinchartl: it seems that BSP team already found issue
<morimoto> only 100 dozens  [18:40]
<pinchartl> :-)
<pinchartl> I'll try not to have nightmares about them tonight
<pinchartl> thank you for attending everybody
<pinchartl> have a nice evening or day
<morimoto> Thanks, bye
<pinchartl> (report sent)
<morimoto> have a nice nightmares :)  [18:41]
<dammsan> thanks guys
<kbingham> Cheers!
<uli___> cu  [18:42]