diff options
Diffstat (limited to 'wiki/Chat_log/20180809-io-chatlog')
-rw-r--r-- | wiki/Chat_log/20180809-io-chatlog | 112 |
1 files changed, 112 insertions, 0 deletions
diff --git a/wiki/Chat_log/20180809-io-chatlog b/wiki/Chat_log/20180809-io-chatlog new file mode 100644 index 0000000..56623bc --- /dev/null +++ b/wiki/Chat_log/20180809-io-chatlog @@ -0,0 +1,112 @@ +09:05 < wsa> so, let's start the IO meeting +09:06 < wsa> first collected updates, then questions +09:06 < wsa> Status updates +09:06 < wsa> ============== +09:06 < wsa> A - what have I done since last time +09:06 < wsa> ------------------------------------ +09:06 < wsa> Geert +09:06 < wsa> : fixed QEMU SCIF emulation with Linux v4.11-rc1+ and SCIF regressions due to +09:06 < wsa> RZ/A2 support. +09:06 < wsa> Kaneko-san +09:06 < wsa> : works on D3 MSIOF upporting. +09:06 < wsa> Marek +09:06 < wsa> : has revisited PCIe L0s/L1 handling on Gen3 in Linux. +09:06 < wsa> Niklas +09:06 < wsa> : sent a new version of the SCC error patches and first version of ES handling +09:06 < wsa> for SDHI. Also worked into SDHI tuning and sorted out more patches for +09:06 < wsa> upporting. +09:06 < wsa> Shimoda-san +09:06 < wsa> : fixed ep0 for USB3 UDC SuperSpeed and upported PWM support for r8a77990. +09:06 < wsa> Ulrich +09:06 < wsa> : tested QEMU SCIF FIFO timeout fix. +09:06 < wsa> Wolfram +09:06 < wsa> : upported more SATA patches and a reworked version of I2C REP_START generation, +09:06 < wsa> reviewed Niklas' various SDHI series, tested SDR104 on Gen3, discussed SDHI +09:06 < wsa> feedback from the HW team, focalized an upstream discussion about I2C transfers +09:06 < wsa> with disabled irq, and worked on a needed DMA safe buffer API update. +09:06 < wsa> B - what I want to do until next time +09:06 < wsa> ------------------------------------- +09:06 < wsa> Kaneko-san +09:06 < wsa> : wants to continue D3 MSIOF upporting. +09:06 < wsa> Niklas +09:06 < wsa> : wants to keep on upporting SDHI patches and act on the responses we got from +09:06 < wsa> the HW team. +09:06 < wsa> Shimoda-san +09:06 < wsa> : wants to ping HW team to get USB unknown things and modify USB2 PHY for E3 +09:06 < wsa> accordingly, and continue to learn KVM (and virtualization) with USB host. +09:06 < wsa> Simon +09:06 < wsa> : wants to revisit RAVB upporting candidates. +09:06 < wsa> Ulrich +09:06 < wsa> : wants to implement checks for other SCIF error flags. +09:06 < wsa> Wolfram +09:06 < wsa> : wants to upstream SDR104 on Gen3, finish the DMA safe buffer API, recheck +09:06 < wsa> the watchdog task mentioned by Shimoda-san, check Yamada-san's patches +09:06 < wsa> for SDHI, and reactivate the work on I2C core PM improvements. +09:06 < wsa> C - problems I currently have +09:06 < wsa> ----------------------------- +09:06 < wsa> Geert +09:06 < wsa> : mentions the SCIF driver is very fragile. +09:06 < wsa> Shimoda-san +09:06 < wsa> : is blocked on some USB tasks because of waiting for answers from the HW team. +09:06 < wsa> Topics +09:06 < wsa> ====== +09:06 < wsa> Holidays +09:06 < wsa> -------- +09:06 < wsa> The following people have holidays the next week. +09:06 < wsa> * Geert +09:06 < wsa> * Niklas +09:06 < wsa> * Shimoda-san +09:06 < wsa> * Simon +09:06 < wsa> Next meeting +09:06 < wsa> ============ +09:06 < wsa> 2018-08-23, 09:00 CEST, 16:00 JST +09:07 < wsa> please double check and ask if you have questions +09:07 < wsa> my questions: +09:07 < wsa> Marex: what does "revisited PCIe L0s/L1 handling" mean? did you get the errata docs now? +09:09 < Marex> wsa: nope, I didnt find any +09:09 < Marex> wsa: I poked the ML thread since I was getting nothing there either +09:09 < wsa> maybe it is an idea to ask the HW team about it? +09:10 < Marex> wsa: yes +09:10 < wsa> I have limited hopes we get useful information out of that mail thread in time +09:11 < Marex> wsa: pretty much ... the discussion went somewhat off-topic, but Bjorn seemed to be quite helpful, so I'll check with him on IRC too +09:12 < wsa> good idea, too +09:12 < wsa> thanks! +09:12 < Marex> np +09:13 < wsa> that's mainly it, I think +09:14 < wsa> not super much happening at the moment, because people are busy with other things (LTSI) or enjoy the summer +09:14 < Marex> wsa: the summer that's roasting everyone, yeah +09:15 < Marex> wsa: I was barely able to do anything in the last month or so due to the heat +09:15 < Marex> wsa: it's getting better now +09:16 < wsa> so, questions from your side? +09:16 < wsa> JapaPERI maybe? +09:17 < shimoda> nothing from my side +09:17 < wsa> one word about my SDR104 results: transfer rates on M3N are awesom, H3 ES2.0 are terrible :/ +09:17 < wsa> "awesome" +09:17 < Marex> wsa: SDR104 with SD card in the full-sized slots ? +09:17 < wsa> yes +09:18 < Marex> wsa: same exact card I presume ? +09:18 < wsa> yes +09:18 < wsa> M3N is > 60MB/s, H3 is ~10MB/s +09:18 < shimoda> it's depends on swiotlb? +09:19 < shimoda> M3N don't use swiotlb because it has 1GiB +09:19 < wsa> oh, that might be a pointer. Totally forgot about that... +09:19 < geertu> Sho it should improve when booted with mem=1G? +09:19 < shimoda> oh, i think H3 seems terrible too.. +09:20 < geertu> ... or when enabling the IOMMU? ;-) +09:20 < shimoda> geertu: yeah, but should be 768M instead of 1G because all soc reserves 128M +09:21 < shimoda> or 256M +09:21 < wsa> I will play with all that +09:21 < geertu> shimoda: right, thx! +09:21 < horms> wsa: I also noticed "awesome" on M3-N, fwiiw +09:22 < wsa> horms: I think I never had >60MB/s with any other board so far... +09:22 * geertu is always confused by awesome and awful +09:22 < horms> yeah, something like that for me too +09:23 < Marex> wsa: you can also always check how the card behaves in U-Boot on that platform, it supports sdr104 +09:24 < wsa> Marex: true +09:24 < horms> geertu: its like chotto vs chou. Sometimes they are the opposite, sometimes they are the same. For extra fun the latter is a homoniem +09:24 < wsa> although it is more how the board behaves ;) +09:25 < wsa> so, i think we are done? +09:25 < wsa> summer influences also the meeting time... +09:26 < wsa> geertu: shall we continue with core right away or have a short break? +09:27 < wsa> anyway, thanks for this meeting everyone! +09:27 < wsa> enjoy your time! |