summaryrefslogtreecommitdiff
path: root/wiki/Chat_log
diff options
context:
space:
mode:
Diffstat (limited to 'wiki/Chat_log')
-rw-r--r--wiki/Chat_log/20211007-io-chatlog103
1 files changed, 103 insertions, 0 deletions
diff --git a/wiki/Chat_log/20211007-io-chatlog b/wiki/Chat_log/20211007-io-chatlog
new file mode 100644
index 0000000..560627f
--- /dev/null
+++ b/wiki/Chat_log/20211007-io-chatlog
@@ -0,0 +1,103 @@
+09:04 < wsa> okay, then let's start the IO meeting
+09:04 < wsa> welcome everyone!
+09:04 * jmondi spills some strong coffee to marex
+09:04 < jmondi> ahah too much coffee for Marek now
+09:04 < jmondi> sorry wsa, go on :)
+09:04 < wsa> here are the status updates
+09:05 < wsa> Status updates
+09:05 < wsa> ==============
+09:05 < wsa> A - what have I done since last time
+09:05 < wsa> ------------------------------------
+09:05 < wsa> Geert
+09:05 < wsa> : fixed Ethernet after kexec/rebind and added missing Ethernet PHY resets on all boards, looked up when alt/gose ee100000.mmc broke
+09:05 < wsa> Niklas
+09:05 < wsa> : sent next version of reading calibration from fuses for thermal
+09:05 < wsa> Shimoda-san
+09:05 < wsa> : fixed freeing wrong tx descriptor for SH ethernet, fixed suspend/resume for CAN, sent patch to add c_sync parameter to the USB audio gadget, continued to improve R-Car S4 Ethernet driver
+09:05 < wsa> Ulrich
+09:05 < wsa> : sent series to integrate V3U registers into the CAN driver and enable CAN on V3U
+09:05 < wsa> Wolfram
+09:05 < wsa> : tested Ulrich's SCIF patch for SysRq, created test case and implemented and sent out proper bugfix for data corruption issue with RPC-IF, enabled RPC-IF on V3U, sent out new version of the GPIO logic analyzer, sent out RFC for making SDnH a seperate clock, sent series to also abort tuning with CMD12 for SD, found two alternatives to make SanDisk cards work better with SDHI,
+09:05 < wsa> reviewed RPC-IF enablement for G2L, discussed handling of undocumented bits for RPC-IF, upported minor patches for TPU, upported patch to avoid a theoretical race for I2C, helped investigating resume issue on Ebisu, sent a treewide cleanup series for getting .driver_data
+09:05 < wsa> B - what I want to do until next time
+09:05 < wsa> -------------------------------------
+09:05 < wsa> Shimoda-san
+09:05 < wsa> : wants to continue to improve R-Car S4 Ethernet driver, and to continue to make more eLinux articles
+09:05 < wsa> Ulrich
+09:05 < wsa> : wants to address issues in the V3U CAN series and send new version
+09:05 < wsa> Wolfram
+09:05 < wsa> : wants to handle bus-recovery consistently in the I2C subsystem, handle responses to patches for SDnH separation, send proper patch to avoid usage of undocumented bits for RPC-IF, more investigation for resume issue on Ebisu, find best solution to handle SanDisk cards with SDHI, pick up TPU patches again which need refactoring, send hopefully final version of the GPIO logic analyzer
+09:05 < geertu> moriperi: Seems like people complained about it on lkml in 2020, which migth have hit Ubuntu now
+09:05 < wsa> C - problems I currently have
+09:05 < wsa> -----------------------------
+09:05 < wsa> Niklas
+09:05 < wsa> : still wonders how to upstream the patches when there seems now HW for it
+09:05 < wsa> and brand-new:
+09:05 < wsa> Marek
+09:05 < wsa> : tries to wake up
+09:05 < wsa> :D
+09:06 < wsa> geertu: when did gose/alt break? I can't recall a mail about that, but it is interesting, of course
+09:07 < geertu> wsa: https://lore.kernel.org/all/CAMuHMdUuqQe6cM-Zt1GWq6xgydv1po8FOOW9qWe+5hK=ZRqE+w@mail.gmail.com/
+09:07 < wsa> shimoday: I think the CAN patches from uli are already good enough for testing. geertu had some comments but they don't affect functionality. IIRC Goda-san wanted to test the patches on V3U?
+09:07 < moriperi> geertu: thanks. do you know the url or something what I can check ?
+09:07 < geertu> - gose worked in v4.15, and is broken since at least v4.17,
+09:07 < geertu> - alt worked in v5.1-rc6, and is broken now.
+09:08 < geertu> moriperi: https://lore.kernel.org/linux-nfs/635679406.70384074.1603272832846.JavaMail.zimbra@dneg.com/
+09:08 < geertu> might be a starting point. good luck!
+09:08 < wsa> geertu: ah, this mail
+09:08 < wsa> geertu: okay, i thought there was a follow-up
+09:09 < wsa> I have access to Gose now, too (thanks, Magnus!), so when time permits I can have a look as well
+09:09 < geertu> wsa: Yep, after that I tried "[PATCH 0/3] mmc: also abort tuning with CMD12 for SD"
+09:09 < geertu> didn't help
+09:10 < shimoday> wsa: I think goda-san will not test the can patch on V3U. is it ok for upstream?
+09:10 < wsa> geertu: I would have been seriously surprised, if it did :D
+09:10 < wsa> so, we got RPC in a much better state since the last meeting
+09:11 < wsa> I think it should be actually usable now ;)
+09:11 < wsa> neg: about your issue with HW with fuses actually set, did you ask Renesas Europe already?
+09:12 < neg> no I have only asked around in Renesas Japan
+09:12 < wsa> patersonc[m]: you here?
+09:12 < neg> So my question is if we think it's OK to send the next version of the patch on the public ML in hope someone with the fuses set can test it
+09:14 < wsa> that would be a customer, or?
+09:15 < wsa> new idea: maybe it makes sense to ask Dirk Behme (Bosch) directly if they have the fuses set?
+09:15 < wsa> I could do that...
+09:15 < neg> Good idea, I will try that
+09:16 < wsa> you can do that, too, of course :)
+09:16 < geertu> or
+09:16 < geertu> Eugeniu Rosca at ADIT?
+09:16 < neg> Or if you know him please go a head and do so, likely the success rate is higher then me cold calling him
+09:17 < wsa> neg: yes, I know him a little. I will do it
+09:17 < neg> Maybe the best way then is to post the patch to the public ML as a RFT and CC Dirk and Eugeniu asking if they have production hardware where the fuses are set?
+09:17 < wsa> geertu: yes, another good candidate
+09:19 < neg> OK let's start with wsa talking to Dirk and we can take it from there, thanks for the help
+09:19 * wsa gets attacked by a puppy
+09:19 < wsa> any questions from your side?
+09:20 < geertu> wsa: pitbull puppies grow into full-fledged pitbulls?
+09:20 < neg> wsa: I will summarise it in a mail to you
+09:20 < wsa> neg: thanks
+09:21 < wsa> geertu: this one attacks with cuteness
+09:22 < patersonc[m]> wsa: MOrnin
+09:23 < wsa> Hi patersonc[m]: do you know if any of the Gen3-alike SoCs has the fuses set for the Thermal IP cores?
+09:24 < wsa> our development boards don't have it, so we can't test the patches reading the parameters from fuses
+09:25 < neg> The test for this is to check if BIT(15) and BIT(14) in the Thermal THSCP is set
+09:25 < patersonc[m]> Oh. Why would they disable that?
+09:26 < patersonc[m]> wsa: I'm just on the school run atm. Will check when I get home
+09:26 < wsa> because they are not final when we get the prototypes AFAIU
+09:26 < wsa> patersonc[m]: thanks!
+09:27 < wsa> patersonc[m]: can you contact Niklas with your findings? that would be great
+09:28 < shimoday> wsa: neg: about the thermal fuse, should I ask BSP team?
+09:28 < wsa> shimoday: yes, please
+09:28 < neg> shimoday: please do
+09:28 < wsa> ah, now I see that shimoday replied to the CAN question
+09:29 < wsa> hmm, Uli's patch is too big to upstream it untested, I am afraid :(
+09:29 < shimoday> wsa: neg: i got it.
+09:29 < neg> shimoday: On internal ML please check '[RFT 0/2] thermal: rcar_gen3_thermal: Read calibration from fuses'
+09:30 < shimoday> neg: thank you for the indicate, i'll ask BSP team.
+09:31 < wsa> seems we have to check again how to test CAN on V3U ourselves
+09:31 < wsa> I guess we really want this patch as it paves the way for S4
+09:31 < wsa> which has 16 CAN channels
+09:32 < shimoday> wsa: i agree (the patch is too big)
+09:32 < wsa> will need to check the schematics
+09:34 < wsa> okay, I guess we can move to core then?
+09:34 < geertu> wsa: I think so
+09:34 < wsa> geertu: here is the mic!
+09:34 < wsa> thanks everyone!