**** BEGIN LOGGING AT Fri Jun 19 02:59:57 2020
Jun 19 04:05:15 <jkridner[m]> probably want to jump up to the 5.4 kernel as ready.
Jun 19 04:06:51 <jkridner[m]> don't need a new image to switch kernels
Jun 19 04:07:04 <jkridner[m]> still lots of issues, but 5.4 should be partially reasonable and I think the plan is to have it ready by August.
Jun 19 04:27:59 <lorforlinux[m]1> Okay sure, will do that.
Jun 19 11:10:12 <vedant16[m]> jkridner: https://www.evelta.com/beaglebone-green-wireless-102010048/ surprisingly this India store has BB green wireless stocked in India
Jun 19 11:10:48 <vedant16[m]> This store has almost all beagleboard's board, stocked locally in India https://www.evelta.com/categories/boards-kits-and-programmers/single-board-computers/
Jun 19 12:28:03 <nwan[m]> jkridner: BBB Overlay is working now: https://github.com/NiklasWan/bb.org-overlays/commit/5a4280777760580fcd18399f47167f50062964bd I will work now on porting everything to 5.4-rt, to see how it goes
Jun 19 14:09:57 <nwan[m]> pdp7: First step was successfull. I was able to compile and run 5.4-rt on my BBAI without problems
Jun 19 14:21:42 <lorforlinux[m]1> nwan: you can also update to 5.4 by using `cd /opt/scripts/tools/` then `./update_kernel.sh --lts-5_4`
Jun 19 14:23:25 <nwan[m]> lorforlinux: Thanks but I need a custom kernel, because I am porting a driver to 5.4-rt
Jun 19 14:27:25 <lorforlinux[m]1> ohh okay, not problemo :)
Jun 19 14:27:34 <lorforlinux[m]1>  * ohh okay, no problemo :)
Jun 19 14:33:17 <lorforlinux[m]1> nwan: RNDIS and wifi working in your 5.4 kernel?
Jun 19 14:34:18 <nwan[m]> Wifi causes problem on 5.4-rt this is why I blacklisted the driver, to avoid problems
Jun 19 14:53:39 <lorforlinux[m]1> okay it's not working here either :(
Jun 19 14:54:37 <nwan[m]> I think they also blacklisted it.
Jun 19 14:57:01 <lorforlinux[m]1> yeah i think so.
Jun 19 14:58:25 <nwan[m]> Do you have the file /etc/modprobe.d/brcmfmac-blacklist.conf ?
Jun 19 15:01:24 <lorforlinux[m]1> I have two files for BT and wifi `btwilink-blacklist.conf & btwilink-blacklist.conf`
Jun 19 15:01:40 <lorforlinux[m]1>  * I have two files for BT and wifi `btwilink-blacklist.conf & rtl8723bu-blacklist.conf`
Jun 19 15:02:54 <nwan[m]> Okay pdp7 told that "blacklist brcmfmac" blacklists the wifi driver. Maybe it is in one of those files for you
Jun 19 15:03:28 <lorforlinux[m]1> Does RNDIS work for you on 5.4?
Jun 19 15:03:46 <lorforlinux[m]1> I don't see any connection through USB!
Jun 19 15:04:32 <nwan[m]> Have not tried it yet. I allways connect via seriell port or via ssh on eth0
Jun 19 15:05:31 <lorforlinux[m]1> ohh nice will try using ssh via ethernet :)
Jun 19 15:06:29 <lorforlinux[m]1> Serial port window is of fixed size and It creates problems when commands is big 😬
Jun 19 15:07:38 <nwan[m]> yes I know. But it is the only way to get proper U-Boot log information. So I am using it all the time
Jun 19 15:08:13 <lorforlinux[m]1> yeah me too, just to get the boot log info.
Jun 19 18:11:11 <deepankarmaithan> cwicks:  jkridner  I got a mail from DHL today asking for customs duty and when i called the customer care they said they dont have any information about sender paying the duties
Jun 19 19:11:41 <pdp7> nwan[m]: good to hear
Jun 19 19:38:17 <nwan[m]> jkridner: pdp7 Is it recommended to use linux/types.h over stdint.h in kernel space driver code?
Jun 19 19:44:15 <pdp7> do you mean ./include/uapi/linux/types.h ?
Jun 19 19:44:24 <pdp7> what symbols are you looking for?
Jun 19 19:47:27 <nwan[m]> I am currently digging my head around the existing avb driver And in this file( https://github.com/induarun9086/beagleboard-linux/blob/4.4/sound/drivers/avb_types.h) from line 220-227 several types are defined but I think this is wrong because you cannot be sure that an int is 32 bit on every architecture
Jun 19 19:49:25 <nwan[m]> I am looking for u32 and stuff like that. In standard user space world it would be like uint32_t. But my question is, is it rather preferred to use u32 from include/linux/types.h in kernel space?
Jun 19 19:52:43 <pdp7> do you care about accessing from user space?
Jun 19 19:52:53 <pdp7> https://www.kernel.org/doc/html/v4.10/process/coding-style.html#typedefs
Jun 19 19:53:12 <pdp7> oops https://www.kernel.org/doc/html/latest/process/coding-style.html#typedefs
Jun 19 19:54:51 <nwan[m]> Thank you. So use u32 if no access from user space is needed and _u32 when access from user space is required
Jun 19 19:55:24 <pdp7> __u32 is used in structs for userspace ABI
Jun 19 19:55:24 <pdp7> https://elixir.bootlin.com/linux/latest/source/include/uapi/linux/gpio.h
Jun 19 19:55:50 <pdp7> but internal kernel structs are just u32
Jun 19 19:55:51 <pdp7> https://elixir.bootlin.com/linux/latest/source/drivers/gpio/gpio-omap.c#L46
Jun 19 19:56:03 <pdp7> > Thank you. So use u32 if no access from user space is needed and _u32 when access from user space is required
Jun 19 19:56:03 <pdp7> yeah
Jun 19 19:56:45 <nwan[m]> Until now I am not sure if I need access from user space, because I need to dig deeper into avb.c file which is around 3000 lines. So this will take some time 😁
Jun 19 19:56:53 <nwan[m]> Thank you pdp7 for your help
Jun 19 20:00:56 <pdp7> i think it should be easy to change if you need to
Jun 19 20:01:41 <pdp7> and i believe the only reason it is __u32 is becasue userspace may not have c99 .... but i think you can assume that won't happen
Jun 19 20:01:52 <pdp7> i mean, who has a pre C99 compiler
Jun 19 20:02:48 <nwan[m]> Yes I allready did. Just deleted the typedefs and included the header. That was enough, because naming was identical
Jun 19 20:03:14 <nwan[m]> <pdp7 "i mean, who has a pre C99 compil"> no one using a BeagleBone I think :D
Jun 19 20:03:22 <pdp7> right
**** ENDING LOGGING AT Sat Jun 20 03:00:22 2020