13:36 <mkorpershoek> narmstrong: what are your thoughts about text-based environment files: https://u-boot.readthedocs.io/en/latest/usage/environment.html#text-based-environment ?

13:36 <mkorpershoek> do you really hate it, or would you be open for include/configs/meson64_android.h to get converted (if I'm doing it) ?

13:37 <mkorpershoek> I think that text based will make things easier for future board additions/or reuse for other SoCs as well

13:40 <narmstrong> mkorpershoek: I'm not against it, but we would have a lot of duplicate right ?

13:40 <narmstrong> I think we should keep the .h for boards using the generic one, and add .env for boards that overload

13:43 <mkorpershoek> the idea is to get rid of `include/configs/meson64_android.h` completely and have include/environment/amlogic/android.env instead

13:45 <mkorpershoek> because writing long strings in C (such as `bootcmd_system`) feels painful and error-prone

13:46 <mkorpershoek> eventually, `include/configs/meson.h` would also be replaced by `include/environment/amlogic/default.env`

13:48 <mkorpershoek> For example, for the partition tables, it would look like this https://www.irccloud.com/pastebin/5A8SWGVX/

13:51 <mkorpershoek> I'm thinking about this because I came accross how TI handles this in `include/environment/ti/`

13:57 <narmstrong> if we can keep a common env, then sure let's do it!

14:00 <mkorpershoek> awesome, thanks for your quick feedback :)

14:03 <f_> I can really tell how Amlogic's DDR init is magic code

14:07 <f_> Anyway, I'll start documenting BL2

14:07 <f_> I started with the BL1 watchdog and how BL2 disables it.

14:09 <f_> https://git.vitali64.duckdns.org/misc/reversing-gxbb-bl2.git/about/docs/watchdog.md

14:11 <f_> I'm not very good at documentation though, but I'm trying my best.

16:37 <f_> It does seem like pctl init is very similar

17:23 <f_> I documented how BL2 loads BL21 stuff https://git.vitali64.duckdns.org/misc/reversing-gxbb-bl2.git/about/docs/bl21.md

17:58 <minute> xdarklight: i've integrated the rtw88 patch now with the debian kernel i'm testing (on a311d), but having some strange new problem with the driver now when loading the fw

17:59 <minute> [ 480.154680] rtw_8822cs mmc2:0001:1: firmware: direct-loading firmware rtw88/rtw8822c_wow_fw.bin

17:59 <minute> [ 480.157322] rtw_8822cs mmc2:0001:1: mac power on failed

18:16 <minute> xdarklight: do you know perchance which patches i need on top of linux-6.3.0 to make the main rtw88 patchset work with rtw_8822cs?

18:16 <minute> rtw88 sdio patchset i mean

19:03 <xdarklight> minute: RE rtw88 commits to apply on top of Linux 6.3 (in the order in which they should be applied): 96c79da2e4d1a25e73916c656fe4ccf7fc8176ae 8599ea40582d49afe437badde76e31bd7429c607 64e9d564653566812dfade36d20f1794407e6ca1 9e688784b8a13515ee186d0ee1cfab12b6d05241 ad0a677bce20c7fa2e8af3fd7f23c0686018202b b7ed9fa2cb76ca7a3c3cd4a6d35748fe1fbda9f6 15c8e267dfa62f207ee1db666c822324e3362b84 6a92566088b1a37c1cf2c4b6b5fb733dc5bdc6a6

19:03 <xdarklight> 65371a3f14e73979958aea0db1e3bb456a296149 b722e5b130bcaa8224c93c6625685d8276742cbd a5d25f9ff91831f3a87aca96480a8b080546ccb8 02461d9368c59510ef51cc8a1db1f0f31cfbf9ad 095e62dd74276dd801c04d9a2a9afcd922811218 6fdacb78f7999f5c14d9dae10d47de50959297d9 b2a777d68434375dc05a6fda5fec34a474bbf21f cb0ddaaa5db09d7d216fcbf0e68779be223a1128

19:04 <xdarklight> minute: I think the very last one has been backported to 6.3.x (where x > 0)

19:05 <minute> xdarklight: wow, thank you

19:05 <xdarklight> minute: also I assume that the "mac power on" failed error is with a previously working rtw88 driver that just had the patch I mentioned added on top

19:05 <minute> xdarklight: hmm, i didn't actually test it ~without~ your patch

19:06 <xdarklight> minute: the latest rtw88 series on the mailing list doesn't include all patches that are needed to make it work because some patches got applied early

19:07 <minute> xdarklight: what i did was i applied this patch series on top of 6.3.0 https://lore.kernel.org/netdev/20230405200729.632435-1-martin.blumenstingl@googlemail.com/T/

19:07 <minute> xdarklight: oh!

19:07 <minute> xdarklight: so possibly i could apply an older series?

19:08 <xdarklight> minute: you could apply an older series (which I don't recommend, who knows if there are bugs that have been fixed since) or just pick all commits I listed above from mainline

19:08 <minute> xdarklight: ok thanks, i'll try to do that then

20:36 <minute> xdarklight: hmm, 6fdacb78f7999f5c14d9dae10d47de50959297d9 is referring to SDIO_DEVICE_ID_REALTEK_RTW8822CS and SDIO_VENDOR_ID_REALTEK which are undefined, so i am missing some other patch yet

20:37 <minute> something like https://lore.kernel.org/lkml/20221227233020.284266-6-martin.blumenstingl@googlemail.com/

20:38 <minute> 7d6d2dd326a8a8d32091e9748f3428dd3be68367

21:20 <minute> xdarklight: i can report that with the above stack, rt8822cs works fine now on my platform!

21:21 <minute> xdarklight: incl the patch you asked me to test