wiki: Porting wiki: Porting Chat Log

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>

1 files changed, 135 insertions, 0 deletions

diff --git a/wiki/Chat_log/20161214-io-chatlog b/wiki/Chat_log/20161214-io-chatlog
new file mode 100644
index 0000000..7dec763
--- /dev/null
+++ b/wiki/Chat_log/20161214-io-chatlog
@@ -0,0 +1,135 @@
+--- Log opened Wed Dec 14 08:57:55 2016
+08:57 -!- wsa_ [~wsa@p54B33B2F.dip0.t-ipconnect.de] has joined #periperi-io
+08:57 -!- ServerMode/#periperi-io [+ns] by leguin.freenode.net
+08:57 -!- Irssi: #periperi-io: Total of 1 nicks [1 ops, 0 halfops, 0 voices, 0 normal]
+08:57 -!- Irssi: Join to #periperi-io was synced in 0 secs
+08:58 -!- geertu [~geert@d54C189FD.access.telenet.be] has joined #periperi-io
+08:58 < geertu> Mornin'
+08:58 -!- neg [~neg@unaffiliated/neg] has joined #periperi-io
+08:59 <@wsa_> good morning
+09:00 < neg> morning
+09:02 <@wsa_> i think we convinced Eduardo to soc_device_match
+09:02 < neg> yes :-)
+09:03 < geertu> Good
+09:03 < neg> I had a chat with Geert yesterday and I think I will switch to binary scaling instead of decimal scaling in next version to increase accuracy, I think it's better Renesas test that version then the deciaml one. Do you agree?
+09:03 <@wsa_> yes, i do
+09:03 <@wsa_> i wondered why it didn't work for you before
+09:04 <@wsa_> On C64, we always do binary scaling ;)
+09:04 < neg> it worked but the diff with the orignal algo which used deciaml scaling was greater so I stuck with the one which where closer to the orignal one
+09:05 <@wsa_> it's a pity neither morimoto nor shimoda are here; i wanted to discuss testing this driver inside Renesas
+09:05 < geertu> neg: it's strange the diff was greather. I can imagine it's greather for some specific values, but on average it should be smaller.
+09:05 <@wsa_> neg: given that the precision is higher, this was "non-working" in my book :)
+09:06 <@wsa_> it is also a pity that Ulrich is not here, I wonder about Geert's question, too
+09:06 < geertu> Perhaps a silly programming error?
+09:06 <@wsa_> "BTW, if DR already indicates timeout, what is the added value of TO?"
+09:07  * geertu still has to review Uli's SCIF series
+09:07 < geertu> The various "feature bits" on SCIF are a collection of wild ideas that came up through multiple brainstorm sessions?
+09:08 <@wsa_> heh
+09:09 <@wsa_> geertu: any news about the hotel in Brusseles for the pre-FOSDEM meetings?
+09:09 < geertu> No, haven't booked anything yet.
+09:09 < geertu> I'm trying to find out how many persons will attend ;-)
+09:10 <@wsa_> I'll arrive on Thursday and leave on Monday
+09:10 < geertu> For Core, we'll have Niklas, Jacopo, Simon, Laurent, and Me.
+09:10 < geertu> No recent response from Magnus or Uli
+09:11 < geertu> So that makes 5-7
+09:11 < geertu> What about I/O?
+09:11 <@wsa_> I recall that Magnus wanted to come?
+09:12 < geertu> His current (last I heard from Morimoto-san) plan is to arrive on Fri at 16h
+09:12 < geertu> ... which is in-time for the beer event, but nof the meeting, I guess ;-)
+09:13 <@wsa_> well, from the last chat meeting, I know of:
+09:13 <@wsa_> you, neg, simon, and me
+09:13 <@wsa_> probably uli
+09:13 <@wsa_> maybe
+09:14 <@wsa_> didn't know about Jacopo but he is welcome of course
+09:14 < geertu> So incl. J., that's also 5-7
+09:14 <@wsa_> pretty much the same group :)
+09:15 <@wsa_> i am likely interested in the core meeting as well
+09:16 <@wsa_> we might have cross over tasks anyhow
+09:16 <@wsa_> (TDSEL)
+09:16 < geertu> ok, 6-8
+09:16 <@wsa_> speaking of, what to do with the TDSEL patch for r8a7790?
+09:17 < geertu> That's the one with the hardcoded register values? I don't think we can use it as-is.
+09:19 < geertu> There doesn't seem to be a PIN_CONFIG_*  for capacitance in enum pin_config_param yet.
+09:19 <@wsa_> well
+09:19 <@wsa_> the docs don't really allow configuration
+09:19 < geertu> (about the hardcoded values) Or don't we need explicit configuration, and should we also use the same value?
+09:20 <@wsa_> yes
+09:20 < geertu> The docs are conflicting.
+09:20 <@wsa_> for H2, they say bits "should be 01"
+09:20 <@wsa_> which is not the reset value
+09:21 <@wsa_> it even says "must be 01"
+09:21 < geertu> Isn't there another doc that says "must be 00"?
+09:21 < geertu> While the HW engineers responded that it's the same hardware block everywhere.
+09:23 <@wsa_> Yes, M2-W says "must be 00"
+09:23 < geertu> If we don't need configuration, doing it in .init() is fine for me.
+09:24 <@wsa_> That's a good item for the FOSDEM meeting:
+09:25 <@wsa_> if somebody could bring a koelsch, then we could check if the same SDR104 problem with TDSEL=00 exists
+09:25 <@wsa_> or any other Gen2 board
+09:26 <@wsa_> okay, that was all the topics I had
+09:26 <@wsa_> anything left from your side?
+09:27 < neg> I have question is this BUG http://pastebin.com/wycYBEyF which can be triggerd on Koelsch by 'reboot' using shmobile_defconfig + CONFIG_DEBUG_ATOMIC_SLEEP known?
+09:27 < geertu> Is this card SDR-104? http://www.conrad.be/ce/nl/product/1388580/SanDisk-Ultra-SDHC-kaart-32-GB-Class-10-UHS-I
+09:27 < geertu> "systemd-shutdow Not tainted"? Shouldn't systemd taint the system? ;-)
+09:28 < geertu> I recently enabled CONFIG_DEBUG_ATOMIC_SLEEP in my .config, but haven't seen that message on my Koelsch. Usually not using shmobile_defconfig, though.
+09:28 < neg> I need to switch to init for my NFS, can't boot properly without CONFIG_CGROUPS and CONFIG_SECCOMP :-)
+09:29 < neg> the BUG is triggerd by running reboot and hits all the time for me
+09:30 <@wsa_> geertu: it should be. it has uhs-1 and i've never seen cards which only support SDR50 but not SDR104
+09:30 <@wsa_> geertu: but shouldn't you play safe and go for micro-sd?
+09:30 <@wsa_> I don't know about other Gen2 boards, but Lager has micro-sd only
+09:30 < geertu> wsa_: OK. I used my Koelsch a few weeks ago to copy music to that card, to be used in my car.
+09:31 <@wsa_> ah, you already have it
+09:31 < geertu> Yes.
+09:31 <@wsa_> well, the kernel log will tell you which mode was used
+09:32 < geertu> I also have a similar smaller uSD card http://www.conrad.be/ce/nl/product/1381015/SanDisk-Ultra-16-GB-microSDHC-kaart-Class-10-UHS-I-incl-SD-adapter
+09:32 < geertu> Also used koelsch to init that one ;-)
+09:32 <@wsa_> and if the problem is the same, then you will get timeouts after a second or so after removing the card
+09:32 < geertu> In SDHI0, of course (SDHI1 is slower)
+09:33 <@wsa_> cool
+09:33 <@wsa_> because it is only the san-disk card which causes the problem for me
+09:33 < geertu> Lemme look in kern.log
+09:33 <@wsa_> the samsung card works fine
+09:33 <@wsa_> (mine is 64gb, though)
+09:36 < geertu> mmc0: new ultra high speed SDR104 SDHC card at address 0007
+09:37 < geertu> mmcblk0: mmc0:0007 SL32G 29.0 GiB 
+09:37 < geertu> and in SDHI1:
+09:37 < geertu> mmc1: new ultra high speed SDR50 SDHC card at address 0007
+09:37 < geertu> mmcblk1: mmc1:0007 SL32G 29.0 GiB 
+09:37 < geertu> The other one:
+09:37 < geertu> mmc0: new ultra high speed SDR104 SDHC card at address aaaa
+09:37 < geertu> mmcblk0: mmc0:aaaa SL16G 14.8 GiB 
+09:38 < geertu> According to the log, I removed and reinserted the second one, and that worked
+09:38 <@wsa_> nice to know
+09:38 < geertu> If there's anything else I can test
+09:39 <@wsa_> don't think so
+09:39 <@wsa_> re-inserting is the test to do
+09:39 < geertu> You need my TDSEL values?
+09:40 <@wsa_> ah, yes
+09:40 <@wsa_> i assumed they were 0, but there is firmware involved
+09:43 < geertu> => md.l 0xe6060084 2
+09:43 < geertu> e6060084: 00000000 00000000    ........
+09:46 <@wsa_> ok
+09:46 <@wsa_> thanks!
+09:46 <@wsa_> i'll bring my "magic card" to Brussels nonetheless :)
+09:46 < geertu> If you need anything else to be tested, the cards are in my car resp. NanoPi
+09:48 <@wsa_> neg: this issue is known. i2c transactions want irqs but they are disabled already that late
+09:49 <@wsa_> neg: this is why we have a task for 'irqless i2c communication' in the io todo
+09:49 < neg> wsa_: ahh I see, good then I know :-)
+09:49 -!- horms [~horms@217.111.208.18] has joined #periperi-io
+09:50 <@wsa_> seems i should always mention in the invitation letter that we meet in #periperi-io
+09:50 <@wsa_> sorry simon
+09:51 < horms> No, the problem is entirely mine. I forgot to add the meeting to my schedule.
+09:51 < horms> Appologies for missing the meeting
+09:51 < neg> wsa_: could you in the report put a small remark that we would like Morimot-san or Khiem help to test the thermal driver?
+09:52 <@wsa_> yes
+09:52 <@wsa_> not only a small one :)
+09:53 < neg> thanks :-)
+09:54 <@wsa_> no prob, simon. i'll upload the log later today
+09:54 <@wsa_> so, i have another appointment soon
+09:54 <@wsa_> and i think we are done?
+09:55 <@wsa_> i'll ask the remaining questions in the summary mail
+09:55 <@wsa_> to uli and morimoto
+09:55 <@wsa_> until then, have a nice week
+09:56 < neg> I have nothing more, thanks all
+09:58 <@wsa_> cya!
+09:59 < geertu> thx, bye
+--- Log closed Wed Dec 14 10:03:36 2016