Schimsalabim has quit [Read error: Connection reset by peer]
apritzel has joined #linux-sunxi
colinsane has joined #linux-sunxi
aggi has quit [Remote host closed the connection]
aggi has joined #linux-sunxi
<MasterR3C0RD>
Finally got around to building U-Boot for the Cubie A5E; I see the boot failure directly after boot (and DRAM being autodetected as 8192M instead of 4096M once)
<jernej>
on avaota board, but no reason why it shouldn't work cubie board too
<MasterR3C0RD>
jernej: Sweet, I'll test that out later. Are there any parts of the DRAM code that might warrant a closer look in particular, off the top of your head?
<jernej>
mctl_phy_write_leveling() is untested, I'm not sure if I ever corrected mctl_phy_read_training() and mctl_phy_write_training() is H616 copy, which is completely wrong
<jernej>
for code which is actually used on avaota board gives 100% same register settings as BSP, so for that I'm sure it's correct
<jernej>
and it turns out that MR14 needs to become Kconfig option, because cubie board had it set slightly differently
<MasterR3C0RD>
Bleh, can't find a copy of JEDEC to see what that's for, but it's probably important
<MasterR3C0RD>
s/JEDEC/JESD209-4/
Schimsalabim has joined #linux-sunxi
<MasterR3C0RD>
According to random TI docs, MR14 corresponds to VREF_DQ setting/range. That'd definitely do the trick
<apritzel>
loki666: but that log is indeed good info: it gives the DRAM parameters, and also dumps the PMIC settings
wingrime1 has joined #linux-sunxi
gsz has joined #linux-sunxi
wingrime-ww has quit [Ping timeout: 480 seconds]
<loki666>
apritzel: no nothing, not even even return line
<apritzel>
loki666: you could check for a transistor footprint, like on that picture on the Eachlink wiki page. Maybe you find a trace connecting between RX and some pads, or you can try a continuity test
<apritzel>
mmh, hard to say, the other side of this empty resistor footprint seems to go to this connector and this cable?
<loki666>
no clues, nothing was visibly connected to the pads, probably from a inner layer
<apritzel>
and I'd assume that the traces go more or less directly to the SoC there. Do you see traces connected to those pads? Maybe on the underside?
<loki666>
didn't remove the pcb from the shell yet, I guess I can try, you think that for a reason or another (bad solder, missing res) RX is not working properly?
<loki666>
I guess we can confirma that with a u-boot no ?
<apritzel>
yeah, I guess. Did you convert the DRAM parameters into a new defconfig?
<loki666>
not, yet no
<apritzel>
also, since you have nice pictures, and I think it's a somewhat "sexy" device: any chance you can create a wiki page for that?
<apritzel>
that would be good place for the bootlog and those pictures, so other people could have a look
<loki666>
ofc, but I'm not sure I want to show my whack of a soldering job :p
<loki666>
I'd need access to it
<loki666>
I think I'll try to hack a init script to extract the BSP dtb
<apritzel>
you can leave that one picture out then ;-)
<loki666>
and maybe dump some info also (if that methods works)
Schimsalabim has quit [Read error: Connection reset by peer]
<loki666>
apritzel: continuity test, is not very conclusive... I don't have clear signal with RX to these res group, it reacts a bit, but nothing on the empty res
<loki666>
in anycase, with my soldering skills, I don't see myself doing that
<loki666>
I need a electronic magnifier
gsz has quit [Ping timeout: 480 seconds]
Schimsalabim has joined #linux-sunxi
Schimsalabim has quit [Read error: Connection reset by peer]