summaryrefslogtreecommitdiff
path: root/wiki/Chat_log/20200611-io-chatlog
diff options
context:
space:
mode:
Diffstat (limited to 'wiki/Chat_log/20200611-io-chatlog')
-rw-r--r--wiki/Chat_log/20200611-io-chatlog94
1 files changed, 94 insertions, 0 deletions
diff --git a/wiki/Chat_log/20200611-io-chatlog b/wiki/Chat_log/20200611-io-chatlog
new file mode 100644
index 0000000..18a5def
--- /dev/null
+++ b/wiki/Chat_log/20200611-io-chatlog
@@ -0,0 +1,94 @@
+09:03 < wsa> welcome to the IO meeting
+09:03 < wsa> collected status updates
+09:03 < wsa> Status updates
+09:03 < wsa> ==============
+09:03 < wsa> A - what have I done since last time
+09:03 < wsa> ------------------------------------
+09:03 < wsa> Geert
+09:03 < wsa> : DT binding doc conversion to json-schema for em,uart, em,sti, mtu2, and tmu (updated), rtc-sh (updated), SDHI Runtime PM testing/investigation, fixed (worked around) RAVB regression due to Micrel PHY using internal delays now, RSPI/QSPI bit rate improvements
+09:03 < wsa> Niklas
+09:03 < wsa> : did clean ups for both thermal drivers
+09:03 < wsa> Shimoda-san
+09:03 < wsa> : Checked a BSP team's report about an SDHI DMA map/unmap unbalance and sent a fix, checked a RAVB driver patch from Bosch and submitted a patch, upported a thermal patch from BSP, reviewed SDHI driver's patches from Wolfram
+09:03 < wsa> Ulrich
+09:03 < wsa> : sent atomic transfers implementation for i2c-sh_mobile
+09:03 < wsa> Wolfram
+09:03 < wsa> : reviewed and tested RPM and clk_imbalance updates for SDHI, removed manual clock handling from renesas_sdhi_core, reworked and sent out series 'fix stalled SCC' and 'implement manual, calibration' for SDHI, sent out cleanup patches for I2C core, and I2C slave eeprom, review patches for IIC, I2C core, and r8a7742
+09:03 < wsa> B - what I want to do until next time
+09:03 < wsa> -------------------------------------
+09:03 < wsa> Geert
+09:03 < wsa> : wants to implement explicit internal delay setting from DT for RAVB, DT binding doc conversion to json-schema
+09:03 < wsa> Shimoda-san
+09:03 < wsa> : wants to continue investigating how to avoid mmc suspend/resume issue, ping Cogent about RAVB driver patch, convert usb-xhci, rcar-pci and sdhi dt docs to json-schema
+09:03 < wsa> Ulrich
+09:03 < wsa> : wants to implement atomic transfers implementation for i2c-rcar
+09:03 < wsa> Wolfram
+09:03 < wsa> : wants to resend series 'fix stalled SCC' and 'implement manual calibration' for SDHI, remove final bits of i2c_new_device API, new version of the series about blocked I2C devices in DT, review Ulrich's atomic_xfer patches for IIC
+09:03 < wsa> C - problems I currently have
+09:03 < wsa> -----------------------------
+09:03 < wsa> Shimoda-san
+09:03 < wsa> : wonders if any driver should not call cma_alloc() in system resume state?
+09:04 < wsa> uli__, geertu: the 'never-disable' series for WDT clocks is still under review?
+09:05 < geertu> wsa: My suggestion from last meeting was to mark the RWDT clock with CLK_IS_CRITICAL if it's already enabled.
+09:05 < wsa> uli__: and I am looking forward to test the atomic_xfer patch :)
+09:05 < wsa> geertu: ah, right, I remember
+09:06 < wsa> I think uli__ agreed?
+09:06 < uli__> i think i did.
+09:06 < uli__> didn't make it to my todo list for some reason...
+09:06 < wsa> uli__: can you work on that before atomic_xfer for i2c-rcar?
+09:07 < uli__> sure
+09:07 < wsa> great, thanks
+09:07 < wsa> so, there is also the cma_alloc question from shimoda-san
+09:08 < wsa> shimoda: i am not much into the RAVB driver, it frees the CMA memory when the system suspends?
+09:09 < shimoda> shimoda: yes if user will not use wake-on-LAN
+09:09 < wsa> i'd think you don't need to free CMA memory until shutdown, but maybe I am naive here...
+09:09 < wsa> how much is it BTW?
+09:09 < geertu> I'm thinking the same
+09:10 < geertu> shimoda: BTW, who from Cogent are you expecting feedback from?
+09:10 < wsa> sergei
+09:10 < wsa> probably
+09:10 < Marex> wsa: no longer with cogent, no ?
+09:10 < geertu> Cogent is under restructuring, and (at last) Sergei has been laid off.
+09:10 < wsa> really?
+09:10 < wsa> wowo
+09:10 < Marex> sadly seems that way
+09:10 < geertu> FWIW, Sergei always worked on ravb/sh_eth in his spare time
+09:11 < shimoda> ah, this cma thing is for BSP and our customer (bosch). because critical issue happen on BSP only
+09:11 < geertu> s/at last/at least/
+09:11 < wsa> shimoda: so, this issue is not present upstream?
+09:11 < shimoda> about other patch (fix race condition on ravb driver), I expect Sergei reviews my patch.
+09:12 < wsa> phew, let's see if sergei still wants to donate time for the ethernet drivers
+09:12 < geertu> wsa: Probablt the issue was upstream, but fixed in the mean time. BSP is at v5.4 now, old BSp was v4.14.
+09:12 < shimoda> wsa: you're correct. because the DMA API calls other alloc function if cma_alloc() failed
+09:13 < shimoda> geertu: oops, I didn't know that.
+09:14 < wsa> well, as long as sergei is still listed as maintainer
+09:15 < wsa> shimoda: so, do you still need input on the cma issue?
+09:15 < wsa> are there other questions/issues?
+09:16 < geertu> wsa: Do you have plans to convert the i2c bindings to json-schema?
+09:17 < wsa> nope
+09:17 < wsa> too much other stuff to do
+09:17 < wsa> is it high priority?
+09:18 < shimoda> I don't have any question/issues for now. i wanted to feedback if some one knew
+09:18 < shimoda> wsa: ^
+09:18 * geertu just notices dt-schema/yaml-bindings/schemas/i2c/i2c-controller.yaml
+09:19 < wsa> i am not an expert with CMA (maybe the MM guys can add) but as I said I wonder why you won't keep the memory until shutdown
+09:19 * kbingham needs an i2c-mux schema too :-S
+09:19 < wsa> geertu: i think rob was working on tha
+09:19 < wsa> that
+09:20 < geertu> wsa: I was mainly waiting for the i2c maintainer to write the core i2c controller schema iunder Documentation/devicetree/bindings/i2c/, but apparently that exists in dt-schema already
+09:20 < shimoda> wsa: i don't know why. But, there is good point to fix the issue I think. Thank you for your feedback!
+09:21 < wsa> noob question: where is dt-schema/yaml-bindings/schemas/i2c/i2c-controller.yaml
+09:21 < wsa> ?
+09:21 < geertu> https://github.com/devicetree-org/dt-schema.git
+09:22 < geertu> For e.g. SPI, spi-controller.yaml is in Documentation/devicetree/bindings/spi/, not in dt-schema
+09:22 < wsa> github? well, ok...
+09:23 < wsa> so, this is the OS independent DT bindings repo?
+09:23 < wsa> and linux should pull it in?
+09:23 < kbingham> I don't think linux pulls it in - the dt-validator uses it ?
+09:23 < wsa> no, it's the tool
+09:23 < geertu> You need the dt-schema repo to run make dt_bindings_check and/or dtbs_check
+09:24 < wsa> anyhow, I recall Rob saying he had an issue to tackle for a proper i2c-controller.yaml
+09:25 < wsa> ok, i think this was it?
+09:26 < wsa> then, let's move to the core issues :)
+09:26 < wsa> geertu: enjoy!
+09:26 < wsa> thanks for this meeting and your work, everyone!