Merge remote-tracking branch 'gitlab/wiki' into HEAD

1 files changed, 148 insertions, 0 deletions

diff --git a/wiki/Chat_log/20170116-io-chatlog b/wiki/Chat_log/20170116-io-chatlog
new file mode 100644
index 0000000..3cf4ad2
--- /dev/null
+++ b/wiki/Chat_log/20170116-io-chatlog
@@ -0,0 +1,148 @@
+--- Log opened Mon Jan 16 08:59:32 2017
+08:59 -!- wsa_ [~wsa@p54B33FFD.dip0.t-ipconnect.de] has joined #periperi-io
+08:59 -!- Irssi: #periperi-io: Total of 5 nicks [1 ops, 0 halfops, 0 voices, 4 normal]
+08:59 -!- Irssi: Join to #periperi-io was synced in 1 secs
+09:00 < wsa_> hey guys!
+09:00 <@morimoto> hi
+09:00 -!- shimoda [~shimoda@relprex3.renesas.com] has joined #periperi-io
+09:00 < neg> horms: I accidently sent my ABC awnsers from my private email address to periperi, I have now resent the anwsers but feel free to delete and not to add my private email to the periperi whitelist :-)
+09:00 -!- morimoto [~user@relprex3.renesas.com] has left #periperi-io ["ERC Version 5.3 (IRC client for Emacs)"]
+09:00 < neg> wsa_: hi
+09:00 < horms> neg: ok, will do
+09:01 < neg> horms: thanks
+09:01 -!- geertu [~geert@d54C189FD.access.telenet.be] has joined #periperi-io
+09:01 < wsa_> that reminds me, simon you can remove the wsa-dev address from periperi as well
+09:01 < geertu> Bummer, I was in #peripero-io
+09:01 < horms> wsa_: sure
+09:01 < wsa_> wsa@the-dreams.de seems to be back at full strength
+09:02 < wsa_> geertu: "we are the popular front"
+09:02 < wsa_> :D
+09:03 -!- Irssi: #periperi-io: Total of 6 nicks [0 ops, 0 halfops, 0 voices, 6 normal]
+09:04 < wsa_> ok, let's start
+09:04 < wsa_> thanks for all the updates
+09:04 < wsa_> despite holidays, there were some things going on :)
+09:05 < wsa_> i have a few questions
+09:05 < wsa_> first one about USB & IPMMU, I lost track a little of it
+09:05 < wsa_> shimoda: does a workaround exist meanwhile?
+09:06 < shimoda> wsa_: a workaround exist for some devices
+09:07 < wsa_> and the SD workaround is basically the same issue
+09:07 < wsa_> IPMMU issue
+09:07 < wsa_> ?
+09:08 < shimoda> wsa_: no. SDHI's restriction differs with IPMMU restriction...
+09:09 < wsa_> OK, then I need to reread about that
+09:10 < shimoda> wsa_: OK. But, I forgot that I wrote the SD restriction on periperi ML or not
+09:10 < shimoda> so, i will summarize them later
+09:10 < neg> is this be related to the IPMMU warnings seen on Gen2 while tuning SD or is it unrelated?
+09:11 < wsa_> shimoda: thank you
+09:11 < shimoda> neg: i'm not sure, but gen2 SYS-DMAC with IPMMU also has a problem
+09:12 < geertu> There's also ipmmu-vmsa e67b0000.mmu: DMA-API: device driver tries to sync DMA memory it has not allocated
+09:13 < neg> shimoda: ok thank you, I will read your upcoming summary on periperi and take it from there :-)
+09:14 < wsa_> sounds good
+09:14 < wsa_> i'm also looking forward to the desc :)
+09:15 < wsa_> next question I have is about thermal driver testing
+09:15 < wsa_> just so I know how to handle upstreaming
+09:16 < wsa_> how likely is it that we can get Renesas testing from Gomi-san like this week or next week?
+09:16 < shimoda> geertu: sorry I don't understand this "DMA-API" things
+09:17 < geertu> shimoda: It means the ipmmu driver does something on a wrong pointer
+09:17 < geertu> which may lead to corruption.
+09:18 < wsa_> and would it be OK for Renesas if we upstream the driver without internal testing to get into v4.11 and fix issues (if any) with incremental patches?
+09:18 < shimoda> wsa_: about thermel driver, Gomi-san and RVC are checking the v6 patch. and maybe i will get some feedack in this week .
+09:18 < wsa_> \\o//
+09:18 < wsa_> awesome!
+09:19 < wsa_> still, the answer to the above question would be nice to know (to have a "Plan B")
+09:21 < horms> wsa_: i would think the answer is yes unless there has been special requests made with regards to thermal from Renesas
+09:22 < wsa_> there are no special requests
+09:22 < shimoda> geertu: it's interesting. Is this easy to reproduce? 
+09:23 < geertu> shimoda: Yes, it happens during ipmmu init. Just enable CONFIG_DMA_API_DEBUG=y
+09:23 < wsa_> special is that we can do very limited testing on this one
+09:23 < horms> wsa_: then I think you should do whatever is best from an upstream pov
+09:24 < geertu> Sorry, during ravb_probe(), but I think it happens during init of the first user
+09:24 < horms> wsa_: right, obviously we don't want things catching fire :)
+09:24 < shimoda> geertu: thank you. i will try the config later
+09:24  * wsa_ sings "relight my fire..."
+09:25 < horms> haha: lets not have that be the theme for a patchset
+09:25 < wsa_> :)
+09:26 < wsa_> ok, so we'll try to get it into 4.11 anyhow, but hope that Gomi-san will be able to test this week
+09:26 < wsa_> which would be awesome
+09:28 < wsa_> is uli back from the holidays?
+09:29 < wsa_> well, he has been on the multimedia meeting
+09:29 < wsa_> would like to have had an update for his series -> will ask by mail
+09:29 < wsa_> any questions from your side?
+09:30 < horms> nothing in particular, other than that he should consider resending any outstanding patches
+09:31 < neg> wsa_: you got time to talk about IO task after meeting?
+09:32 < wsa_> yup
+09:32 < wsa_> it's about emergency shutdown right?
+09:33 < neg> yes
+09:33 < horms> you probably know, but there is code in the BSP for that
+09:33 < wsa_> well, there goes the last chance of really creating fire...
+09:34 < neg> horms: I had a quick look at it, but I only found it by luck so thanks for pointing it out :-)
+09:35 < horms> neg: any time
+09:35 < wsa_> ok, if there are no other questions, let's talk about further hacking thermal/emergency
+09:36 < wsa_> people not interested are free to leave, but everyone is welcome to stay, of course :)
+09:36 < horms> i had one question, sorry
+09:36 < wsa_> sure
+09:36 < horms> meeting next month in .be
+09:36 < horms> any firm plans?
+09:37 < shimoda> about emergency shutdown, this is also handled by Gomi-san and/or related member. so, i can ask him about this
+09:37 < wsa_> there will be an io meeting on friday
+09:37 < wsa_> when exactly depends on when we have the room
+09:37 < wsa_> geertu: ?
+09:38 < horms> ok, do we have a location: I only ask so I can arrange some accomodation for myself
+09:38 < geertu> horms: Yes, see "Meeting around FOSDEM"
+09:39 < geertu> on periperi-ML
+09:39 < horms> ok, thanks
+09:39 < geertu> wsa_: I don't have hours, it's booked for the full day
+09:39 < horms> I see it now
+09:39 < neg> shimoda: thanks for telling me Gomi-san is working with emergency shutdown
+09:40 < geertu> As I have to get their by train, I think I can be their at 9 or 9h30
+09:40 < geertu> Of course you can start earlier ;-)
+09:41 < geertu> s/be their/be there/
+09:41 < wsa_> i don't think i can make it earlier than 9 as well
+09:41 < horms> fwiw, i will likely arrive the day before
+09:42 < wsa_> 9-13 first meeting, lunch, 14-18 second meeting ?
+09:42 < geertu> Please reply to "Meeting around FOSDEM" to confirm your (un)attendence by Wednesday morning, thanks
+09:42 < neg> I think 9:30 is a fine time, I need to navigate the local public transport system which if last year was any indication of my skill will take some time :-)
+09:42 < geertu> BTW, Magnus just told me Ito-san may want to join the meetings
+09:43 < wsa_> i'll be likely in brussels on thursday as well, but just "likely" yet...
+09:43 < geertu> neg: I believe there's a direct subway from your hotel
+09:43 < wsa_> neg: ok
+09:43 < wsa_> 9:30-13:30 first meeting, lunch, 15-19 second meeting ?
+09:44 < geertu> 13:30 may be a bit late for lunch
+09:44 < neg> geertu: ahh thanks, will look it up
+09:45 < wsa_> 9:30-13:00 first meeting, lunch, 14:30-18 second meeting, then open_end hacking ?
+09:45 < geertu> What about 9:30-12:30 (I/O), 13:30-16:30 (Core), 16:30-18:00 (misc)
+09:45 < wsa_> also fine
+09:46 < wsa_> although, 1:30 for lunch might be more relaxing
+09:46 < geertu> Note that we have to think about Dinner, and FOSDEM Beer Event starts at 16:00 :-)
+09:47 < wsa_> even if it is lurking in the room for 30 minutes or having a short walk or something
+09:48 < wsa_> can't tell that yet, i have not made all FOSDEM appointments
+09:49 < wsa_> might be I'm busy on Friday evening meeting some ppl
+09:49 < wsa_> hopefully not
+09:49 < wsa_> :)
+09:50 < wsa_> ok, back to thermal?
+09:52 < wsa_> neg: fire away
+09:52 < neg> So my question is more or less if you think "Thermal,?,noplan,?,add emergency shutdown support" could be a base or additonal task for me duing Q1 or if you rather see me working on something else. At the moment I have no IO tasks planed for Q1
+09:53 < wsa_> i was rather thinking about "Thermal,?,noplan,?,add interrupt support" as your next task :)
+09:54 < wsa_> which might be a prerequisite for shutdown anyhow
+09:54 < wsa_> haven't checked
+09:54 < neg> OK that also sounds like a suitable task for me
+09:55 < wsa_> i have some drafts on my hdd somewhere, but it needs to be heavily refactored
+09:55 < wsa_> if not redesigned
+09:56 < wsa_> but at least we got the bindings in shape already
+09:56 < wsa_> so we have freedom to redesign
+09:57 < neg> OK that is fine, there is no rush we can hammer it out with email maybe before FOSDEM. My primary cencern where that there where some IO work for me to do :-)
+09:57 < wsa_> interrupts were the reason why I went from three DT nodes to one and to introduce the #thermal-sensor-cells to be 1
+09:58 < wsa_> you won't be bored with us
+09:58 < wsa_> if all fails, there is SDHI ;)
+09:59 < wsa_> but i think thermal might be your friend in the next time
+10:00 < neg> nice :-) Then maybe you can send me the draft you made and think about if you wish me to do this as part of base or if we should try to get some of the work as an additinal contract and we take it from there?
+10:00 < wsa_> exactly
+10:01 < neg> thank you
+10:03 < wsa_> i think we are done then
+10:03 < wsa_> thank you all!
+10:03 < wsa_> have a great week
+10:03 < neg> I have nothing more, thanks all
+10:04 < shimoda> thank you!
+10:05 < geertu> thx
+10:10 -!- shimoda [~shimoda@relprex3.renesas.com] has quit Quit: WeeChat 0.4.2
+--- Log closed Mon Jan 16 10:16:36 2017