Core-chat-meeting-2016-12-06 09:07 < geertu> Welcome to today's Core Group Meeting 09:07 < geertu> First, let me introduce a special guest: Jacopo! 09:07 < geertu> (or let him introduce himself ;) 09:07 < jmondi> ahah, hi Geert, thanks :) 09:08 < morimoto> Hi Jacopo ! 09:08 < jmondi> hello everyone, I met most of you in Berlin, for anyone else my name's Jacopo, I'm based in Italy, and I'm working on 2 trail tasks with Magnus for this months 09:08 < jmondi> s/trail/trial 09:09 < jmondi> hope to meet you all in person soon ;) 09:09 < geertu> Thanks, Jacopo 09:09 < geertu> Agenda: 09:09 < geertu> 1. Status updates 09:09 < geertu> 2. Proposal from BSP team about PFC I2C ch0, 3 and 5. 09:09 < geertu> 3. FOSDEM+core confirmation: who? 09:09 < geertu> Topic 1. Status updates 09:10 < geertu> A) What have I done since last time 09:10 < geertu> B) What I plan to do till next time 09:10 < geertu> C) Problems I have currently 09:10 < geertu> First (according to sort -r) is Simon 09:10 -!- horms [~horms@217.111.208.18] has quit [Ping timeout: 265 seconds] 09:11 < geertu> Nice try... 09:11 < neg> You scared him away :-) 09:11 < morimoto> :) 09:11 < geertu> Next is Shimoda-san 09:12 < shimoda> A) B) C) is nothing to me 09:12 < shimoda> --- END --- 09:12 -!- horms [~horms@217.111.208.18] has joined #periperi 09:12 < geertu> Thanks, Shimoda-san 09:13 < geertu> (2nd try) Next is Simon 09:13 < horms> sorry about that 09:13 < horms> network troubles as usual 09:13 < horms> Status: 09:13 < horms> A) What have I done since last time 09:13 < horms> * Analysed device tree binding status 09:13 < horms> - Presence and use of SoC-specific and fallback compatibility strings 09:13 < horms> - Posted some fixups 09:13 < horms> B) What I plan to do till next time 09:13 < horms> * Continue above 09:13 < horms> * See above 09:13 < horms> C) Problems I have currently 09:13 < horms> * Order of struct of_device_id entries. 09:13 < horms> - Fallback last seems nice but unnecessary given my reading of 09:13 < horms> __of_match_node() 09:13 < horms> - If it is required then a behaviour change is implied for 09:13 < horms> in-tree users of i2c-sh-mobile, intc-rcar 09:13 < horms> -- end -- 09:14 < geertu> You are right, __of_match_node() assigns a higher score to the first compatible value in the node. 09:14 < geertu> So it shouldn't be needed. 09:14 < geertu> Still, I think it's nice to list "fallback" values last. 09:15 < horms> ok, understood 09:15 < horms> i need to repost the pci series anyway 09:15 < horms> i'll reword the changelog 09:15 < horms> and also look to fixing up other caes 09:15 < geertu> Thanks Simon. 09:15 < geertu> Next is Niklas 09:16 < neg> A) Nothing 09:16 < neg> B) Start to work on 'GPIO-RCAR,?,noplan,?,Fix Runtime PM with GPIO interrupts (depends on irqchip PM) 09:16 < neg> C) None 09:16 < neg> EOT 09:17 < geertu> Thanks, Niklas 09:17 < geertu> Next is Morimoto-san 09:17 < morimoto> struct core_task *a = *b = *c = NULL; 09:17 < morimoto> return 0; 09:17 < geertu> Thanks. Morimoto-san 09:18 < geertu> Laurent seems to be away 09:18 < geertu> Next is myself 09:18 < geertu> A) Worked with Simon and Arnd to get our pull requests for arm-soc accepted. 09:18 < geertu> Sent pull requests for soc_device_match() and RZ/G clock definitions 09:18 < geertu> B) Coerce Simon into taking all MODEMR related patches 09:18 < geertu> 2017Q1 additional tasks 09:19 < geertu> C) What's wrong with the IPMMU? (more like a retorical quesiton) 09:19 < geertu> EOT 09:19 < horms> a) thanks! 09:19 < horms> b) happy to negotiate :) 09:19 < geertu> From the emails, looks like the HW team is investigating the IPMMU? 09:20 < geertu> Last (present) is Jacopo 09:21 < jmondi> A) PFC patches for Gyro-ADC: v3 in review 09:21 < jmondi> Enabled and tested MAX11100 ADC on genmai board. RFC patches sent out 09:22 < jmondi> B) Not sure is a core task, but I should write an IIO driver for that same ADC this week 09:22 < jmondi> C) none 09:22 < jmondi> that's all from me 09:22 < jmondi> -- end -- 09:22 < shimoda> geertu: yes, the HW team is investigating the IPMMU now. 09:23 < geertu> IIO drivers are indeed not core, but I/O. I don't know if Wolfram plans an I/O chat later this week. 09:23 < horms> jmondi: please consider patches to enable options in defconfigs if appropriate 09:23 < geertu> Thank Jacopo 09:23 < geertu> horms: The MAX11100 ADC is an add-on, not an on-board device 09:24 < jmondi> horms: for ADC you mean? 09:24 < jmondi> yes, it's temporary connected to one of Genmai's expansion header 09:24 < horms> It was just a general request. I'd need to look at the patches more closely to make a specific request regarding them. 09:25 < jmondi> need to sync with Magnus because we'll use another board, but it will remain connected to some expansion connector I guess 09:25 < horms> Ok 09:25 < horms> understood 09:25 < horms> my comment is probably not relevant in that case 09:25 < jmondi> horms: it may be for future! thanks anyway :) 09:26 < geertu> Topic 2. Proposal from BSP team about PFC I2C ch0, 3 and 5. 09:26 < geertu> Submitted by Shimoda-san 09:27 < shimoda> yes, should i explain? 09:27 < geertu> Yes please. 09:27 < neg> shimoda: yes please 09:28 < shimoda> I sent an email. BSP team would like to set the such pins via PFC driver 09:29 < shimoda> the pins spec differs than other pins. so bsp team think that the PFC driver should have other way 09:29 < geertu> IIUIC, multiple functions can be muxed to the i2c pins, but they cannot be used for GPIO? 09:30 < shimoda> geertu: let me check... 09:30 < geertu> It uses PIN_NUMBER('B', 11) 09:32 < geertu> But it also touches avb, hscif, msiof, pwm, sdhi 09:33 < geertu> Does it mean those functions are mutually exclusive with the i2c function, although they use different pins? 09:33 < geertu> And is this ES1.0 or ES2.0? 09:35 < shimoda> geertu: i don't know this is ro es1.0 or es2.0 09:37 < shimoda> about this mod_sel, as you said, they are muxed pins. 09:39 < shimoda> and these i2c pins cannot bd used for GPIO 09:40 < shimoda> this patch reuses POC/DRVCTRL register setting thing 09:41 < shimoda> BSP team would like to know this policy is good or not 09:41 < geertu> Ah, it's for e.g. "Phisically muxed with AVB_AVTP_MATCH_A" inside some floating comment box in (preliminary)CSD15-676_R-CarH3_pinfunction(Rev0p290)_customer.xlsx, which you cannot search for using LibreOffice 09:44 < geertu> According to that xlsx, it uses the same pin numbers as the "Phisically muxed" pins, but in a different row of the sheet. 09:45 < neg> It's pinmux for none-GPIO pins right? 09:45 < geertu> Can't we just use the corresponding RCAR_GP_PIN()? 09:46 < geertu> E.g. SCL0 is SoC pin AA33, SiP pin C29 09:47 < geertu> But SD1_CD is also AA33 / C29, and GP3_14 09:48 < geertu> so we could use RCAR_GP_PIN(3, 14) instead of PIN_NUMBER('C', 29)? 09:48 < geertu> What am I missing? 09:49 < geertu> I think this needs more clarificiation / investigation, offline. 09:50 < geertu> shimoda-san: Do you agree? 09:51 < shimoda> geertu: about red line in the xlsx file, it seems your comment is match, i will ask hw team or/and bsp team 09:51 < shimoda> and offline, it's nice to me :) 09:51 < shimoda> s/red line/red rectangles/ 09:52 < neg> yes, I think if there are a RCAR_GP_PIN() for that pin it should be used, or I'm also missing something 09:52 < geertu> It depends on what "Phisically muxed" really means 09:54 < geertu> (in gnumeric, it shows a black rectangle. I also cannot search for its contents, but I can move (not edit) it ;-) 09:54 < geertu> Let's continue 09:54 < geertu> Topic 3. FOSDEM+core confirmation: who? 09:55 < geertu> So it seems like Morimoto-san will miss the rabbits, and will go to trump(et)land instead ;-) 09:55 < geertu> Who's still planning to attend FOSDEM, and a core meeting on Friday before? 09:55 < geertu> +1 Geert 09:56 < neg> I will go to FOSDEM but have yet to book my tickets 09:57 < neg> (and attend the core meeting on Friday ofc) 09:57 < horms> If there is a Core meeting then I expect to be there 09:57 < horms> I may also come up with other reasons to be there, but that is the main one at this time 09:57 < jmondi> I'll be at fosdem. Hopefully I'll still have a reason to join these meetings :) 09:59 < morimoto> Please enjoy rabbits in FOSDEM :) 10:00 < geertu> I prepared rabbit two weeks ago 10:01 < geertu> I think we're finished, unless someone has still something to add? 10:01 < neg> are there any periperi plans for ELC timing since Morimoto-san will not join at FOSDEM? I have not made any plans for ELC but if I'm to go I think it's best to book soon :-) 10:02 < geertu> deadline for proposals is this Saturday. 10:02 < geertu> I don't plan to go to ELC this time. 10:03 < neg> ok sorry for proloning the meeting just wanted to check so I don't have to buy expensive tickets later on 10:03 < horms> I am considering going but so far have no plans 10:04 < geertu> neg: the tickets may become cheaper after the inauguration of the next president ;-) 10:06 < neg> geertu: haha yes and his inauguration is at end of Jan so maybe the riots in portland have settled down by ELC 10:08 < geertu> Thanks for joining, and have a nice continued day!