ungeskriptet has quit [Remote host closed the connection]
<jernej>
Lightsword: so you have everything figured out and there is no issue with DRAM driver? :)
<Lightsword>
jernej, seems ok on my system at least after I fixed dram params, but I think in next it may incorrectly think it detected a lot more ram than is even possible, so may make sense to clean things up to better handle detection failures
<jernej>
well, code in next actually fixed all of the reported cases of DRAM size misdetection, including cases found by my own experimenting on H616 and A523
<jernej>
yeah, with wrong parameters anything can happen :)
<jernej>
what I mean to say is that you can't fully error proof DRAM code
<Lightsword>
jernej, any easy way to prevent false detections like those for example?
<Lightsword>
I think one would normally assume if dram params are misconfigured they would generally only see less ram than exists, not more
<jernej>
yes, in this case if DRAM size is bigger than supported on the platform, it can report the error and hang instead of continue
radxanaoki has quit [Remote host closed the connection]
radxanaoki has joined #linux-sunxi
<jernej>
but you have to account for H6, which supports max. 3 GB of RAM, but usually this is done by installing 4 GB of RAM
Schimsalabim has quit [Ping timeout: 480 seconds]
Schimsalabim has joined #linux-sunxi
<jernej>
which is properly detected as 4 GB
<jernej>
so the easiest and the only error handling, which I would put in, is the one you mentioned. if for loop doesn't detect any row or column, hang the boot.
junari has joined #linux-sunxi
bauen1 has joined #linux-sunxi
Schimsalabim has quit [Read error: Connection reset by peer]
Schimsalabim has joined #linux-sunxi
hramrach has quit [Ping timeout: 480 seconds]
dfas has joined #linux-sunxi
sdfgsdfs has quit [Ping timeout: 480 seconds]
apritzel has quit [Ping timeout: 480 seconds]
Schimsalabim has quit [Ping timeout: 480 seconds]
Schimsalabim has joined #linux-sunxi
Schimsalabim has quit [Read error: Connection reset by peer]
Schimsalabim has joined #linux-sunxi
bauen1 has quit [Ping timeout: 480 seconds]
paulk-bis has quit []
paulk has joined #linux-sunxi
apritzel has joined #linux-sunxi
<apritzel>
Lightsword: btw: since there was a U-Boot release on Monday, next now became the new master. So if there is anything wrong with jernej's fixes, now would be the time to complain, so that we can fix it
<apritzel>
also for everyone else: if you have any U-Boot changes, now would be a good time to push for them, as the merge window is open. I will try to push as much of the A523 and A133 code in as possible
<jernej>
apritzel: I have half ready series for merging H6 and H616 DRAM size detection code. I can finish it over the weekend, but I won't be able to test it on HW.
<jernej>
should I post it anyway?
<apritzel>
jernej: that's much appreciated, I was hoping for you to merge that! I can surely do some testing
<apritzel>
yes please
<jernej>
I can squeeze Lightsword suggestion in too
<apritzel>
the next release is in July, so we have plenty of time to fix things
evgeny_boger has joined #linux-sunxi
<apritzel>
jernej: many thanks for the U-Boot reviews, much appreciated!
<apritzel>
I figured lately that USB was broken in U-Boot on most A10/A20 boards since about November, with no simple fix, so we need this code to get this going again
evgeny_boger1 has joined #linux-sunxi
evgeny_boger has quit [Ping timeout: 480 seconds]
evgeny_boger1 has quit [Remote host closed the connection]
bauen1 has joined #linux-sunxi
tiop has joined #linux-sunxi
munchaus1n has quit [Ping timeout: 480 seconds]
cnxsoft has quit [Remote host closed the connection]
munchausen has joined #linux-sunxi
hramrach has joined #linux-sunxi
radxanaoki has quit [Quit: radxanaoki]
ungeskriptet has joined #linux-sunxi
ungeskriptet has quit [Remote host closed the connection]
montjoie has quit [reticulum.oftc.net helix.oftc.net]
dsimic has quit [reticulum.oftc.net helix.oftc.net]
mripard has quit [reticulum.oftc.net helix.oftc.net]
kasesag has quit [reticulum.oftc.net helix.oftc.net]
szemzoa has quit [reticulum.oftc.net helix.oftc.net]
linkmauve has quit [reticulum.oftc.net helix.oftc.net]
psydroid has quit [reticulum.oftc.net helix.oftc.net]
hlauer[m] has quit [reticulum.oftc.net helix.oftc.net]
gamiee has quit [reticulum.oftc.net helix.oftc.net]
igraltist has quit [reticulum.oftc.net helix.oftc.net]
rtp has quit [reticulum.oftc.net helix.oftc.net]
rellla has quit [reticulum.oftc.net helix.oftc.net]
immibis has quit [reticulum.oftc.net helix.oftc.net]
ynezz has quit [reticulum.oftc.net helix.oftc.net]
Tooniis[m] has quit [reticulum.oftc.net helix.oftc.net]
DuClare has quit [reticulum.oftc.net helix.oftc.net]
Newbyte has quit [reticulum.oftc.net helix.oftc.net]
barni2000[m] has quit [reticulum.oftc.net helix.oftc.net]
montjoie_ has joined #linux-sunxi
gamiee has joined #linux-sunxi
rellla has joined #linux-sunxi
psydroid has joined #linux-sunxi
Newbyte has joined #linux-sunxi
hlauer[m] has joined #linux-sunxi
Tooniis[m] has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
Daanct12 has quit [Quit: WeeChat 4.6.1]
cnxsoft has quit [Ping timeout: 480 seconds]
barni2000[m] has joined #linux-sunxi
Schimsalabim has quit [Ping timeout: 480 seconds]
Schimsalabim has joined #linux-sunxi
apritzel has quit [Ping timeout: 480 seconds]
dok has quit [Remote host closed the connection]
dok has joined #linux-sunxi
junari has quit [Remote host closed the connection]
warpme has joined #linux-sunxi
Schimsalabim has quit [Read error: Connection reset by peer]
Schimsalabim has joined #linux-sunxi
paulk-bis has joined #linux-sunxi
<gamiee>
apritzel : the DTS you made for Banana Pi M4 Berry seems to be done according to PCB V00. V10 (which I have) have different RAM (the same one as on Orange Pi Zero 3, but 2GiB version instead of 4GiB), and WiFi is now connected through USB hub. BTW do you have downloaded the V00 schematics? The Google Drive link is dead.
<apritzel>
Lightsword: you cannot really use a config for an older BSP kernel on a recent mainline kernel
<apritzel>
besides: configuring a tailored kernel from scratch is non-trivial in these days, this is not the 90s anymore
<apritzel>
I'd recommend you start with defconfig, then disable the stuff you don't need, for instance other platforms, PCIe, and stuff. And keep notes of what you disabled, as you probably need to repeat this exercise
<Lightsword>
apritzel, ok, yeah, was hoping to avoid having to do most of that, I thought that kernel config I linked was based on mainline however
<apritzel>
the Allwinner BSP kernel is a nightmare. Sure, it was forked from mainline at some point, but they butchered it heavily afterwards
<apritzel>
plus v6.1 is now more than two years old
<Lightsword>
apritzel, btw should it be possible to enable networking in uboot on h616 boards like mine? was messing with various device tree configs but couldn't get anything to actually work
<apritzel>
no, not easily. We don't support the integrated PHY in the H616 in mainline, neither in Linux nor U-Boot. We don't even have any DT bindings agreed on.
<apritzel>
There are patches floating around, and some downstream distros have picked them up, but that's not mainline
<Lightsword>
how do I know if my h616 is even using the integrated phy?
<apritzel>
yes, the OrangePis use external Gigabit PHYs. The best way to check is to look at the board, to see if there is a chip close to the Ethernet jack
<Lightsword>
apritzel, I see a chip labeled PSF-16211
<apritzel>
if the board uses 100MBit/s, it's a high chance it's using the internal PHY
<apritzel>
this chip is just the magnetics/transformers, not a PHY
<Lightsword>
all the mainline stuff appears to be using 05020000 as the register address, while this is using 05030000, which is a bit confusing as well
<Lightsword>
apritzel, those patches floating around on the uboot/kernel mailing lists somewhere?
<apritzel>
502 is the MAC which exposes the RGMII pins, to use an external (Gigabit) PHY. 503 is a 2nd MAC, which is connected straight to the internal PHY only
<Lightsword>
apritzel, ah, so mine would have to then be using the internal PHY
<apritzel>
I don't think those patches have been posted, as they are not ready yet. They live in various github repos, I believe, don't know of a good example, and it's unreviewed and prone to change at best anyway
<Lightsword>
apritzel, did the patches work at all?
<apritzel>
yes, they did, but there is a big difference between "works for me" and "ready for mainline"
<Lightsword>
apritzel, remember who was working on them or where I might find those patches to test out?
<apritzel>
we will come back to them at some point in the next months, I think (since I want them too), but it's not high priority
<apritzel>
search for AC200 or AC300, that's the name of the chip co-packaged on the recent Allwinner SoCs providing the internal PHY functionality
<apritzel>
I don't know, it's one of those various branches, with unreviewed and WIP patches. The first one looks like to be against some mainline kernel, but the U-Boot one looks very hackish
tiop has quit []
junari has joined #linux-sunxi
ftg has quit [Read error: Connection reset by peer]