kuba2k2 has quit [Read error: Connection reset by peer]
kuba2k2 has joined #linux-sunxi
kuba2k2 has quit [Read error: Connection reset by peer]
kuba2k2 has joined #linux-sunxi
kuba2k2 has quit [Read error: Connection reset by peer]
kuba2k2 has joined #linux-sunxi
szemzoa has quit []
kuba2k2 has quit [Read error: Connection reset by peer]
kuba2k2 has joined #linux-sunxi
warpme has joined #linux-sunxi
DarkNeutrino has joined #linux-sunxi
junari has joined #linux-sunxi
<junari>
apritzel: hi, did you do something with LPDDR4 driver on the opi zero3?
<junari>
I received my board and thought I'd do it while I still have time.
dsimic is now known as Guest3083
dsimic has joined #linux-sunxi
Guest3083 has quit [Ping timeout: 480 seconds]
<apritzel>
junari: I did test your driver, but it didn't work out of the box on the OPi Zero3
<apritzel>
I worked out the parameters, I think, but that didn't help
<apritzel>
I didn't have time yet to go through it more carefully
<apritzel>
but there are more people asking, and the DT is in mainline Linux now, so U-Boot should not lag behind...
<junari>
apritzel: understand, I will look
<apritzel>
I refactored one function, to remove those nasty "if (LPDDR3) ; else if (LPDDR4); else " cascades
<apritzel>
can post it tonight
<junari>
I tried to do this using other drivers. If there is a better option, then no problem
DarkNeutrino has quit [Ping timeout: 480 seconds]
Daanct12 has quit [Quit: WeeChat 4.0.5]
<junari>
looks like spi contains valid image and it run without microsd. What easy way to load from fel?
kuba2k2 has quit [Ping timeout: 480 seconds]
<junari>
a soldering iron solves any problem, almost
evgeny_boger1 has quit []
kuba2k2 has joined #linux-sunxi
bauen1 has quit [Ping timeout: 480 seconds]
evgeny_boger has joined #linux-sunxi
mps_ has left #linux-sunxi [#linux-sunxi]
junari_ has joined #linux-sunxi
kuba2k2 has quit [Ping timeout: 480 seconds]
evgeny_boger has quit [Quit: evgeny_boger]
evgeny_boger has joined #linux-sunxi
junari has quit [Ping timeout: 480 seconds]
<apritzel>
I put in an SD card with a magic binary on it
<apritzel>
junari_: fel-sdboot.sunxi from sunxi-tools
<apritzel>
dd to 8KB, like any other sunxi boot image
junari_ is now known as junari
<junari>
apritzel: thanks, now I will know
evgeny_boger has quit [Ping timeout: 480 seconds]
<apritzel>
other boards use a button that shorts the SPI chip select
<apritzel>
or some jumpers
<apritzel>
GrantM11235[m]: I don't think SPI NAND works easily on sunxi
<apritzel>
there are patches to enable it for the SPL (to load the rest of U-Boot), but I don't know what's the story for the "sf" command
<apritzel>
you would at least need to enable SPI NAND support, potentially a "driver" for the particular chip
<apritzel>
CONFIG_MTD_SPI_NAND, for a start
JohnDoe_71Rus has quit []
tnovotny_ has joined #linux-sunxi
tnovotny has quit [Ping timeout: 480 seconds]
bauen1 has joined #linux-sunxi
bauen1_ has joined #linux-sunxi
bauen1 has quit [Ping timeout: 480 seconds]
JohnDoe_71Rus has joined #linux-sunxi
tnovotny has joined #linux-sunxi
tnovotny_ has quit [Ping timeout: 480 seconds]
<jernej>
apritzel: speaking of DRAM, did you find proper way for memory barriee and delay in H6 driver?
<apritzel>
jernej: sorry, I was on holidays, and didn't touch a computer at all ;-)
<jernej>
no worries, I'm just curious
<jernej>
if I understand video you linked correctly, then barrier should be replaced with register read?
<apritzel>
that's only needed architecturally, I don't think we need it in this case
<apritzel>
the DSB should push it to the controller already
<jernej>
that talk actually reminded me on dummy read HDMI driver needs to do to descramble register addresses
<jernej>
(after correct code is written to config register)
kuba2k2 has joined #linux-sunxi
<apritzel>
jernej: did you try whether a DSB would suffice there?
<jernej>
no, because I was just happy that it works and I didn't consider why it may be needed
maz_ has joined #linux-sunxi
ungeskriptet1 has joined #linux-sunxi
maz has quit [Remote host closed the connection]
ungeskriptet has quit [Quit: Ping timeout (120 seconds)]
kuba2k2 has quit [Ping timeout: 480 seconds]
bauen1_ has quit [Ping timeout: 480 seconds]
tnovotny has quit [Quit: Leaving]
junari has quit [Ping timeout: 480 seconds]
kuba2k2 has joined #linux-sunxi
apritzel has quit [Ping timeout: 480 seconds]
bauen1 has joined #linux-sunxi
bauen1_ has joined #linux-sunxi
tnovotny has joined #linux-sunxi
bauen1 has quit [Ping timeout: 480 seconds]
bauen1_ has quit [Ping timeout: 480 seconds]
vagrantc has joined #linux-sunxi
JohnDoe_71Rus has quit []
vagrantc has quit [Ping timeout: 480 seconds]
pg12_ has joined #linux-sunxi
pg12 has quit [Ping timeout: 480 seconds]
pg12_ is now known as pg12
kuba2k2 has quit [Ping timeout: 480 seconds]
warpme has quit []
kuba2k2 has joined #linux-sunxi
tnovotny has quit []
mkazantsev has joined #linux-sunxi
mkazantsev has quit []
kuba2k2 has quit []
apritzel has joined #linux-sunxi
<GrantM11235[m]>
Does u-boot support usb gadget on sunxi? I tried enabling it, but I got compiler errors like "undefined reference to `usb_gadget_register_driver'"
<apritzel>
what SoC is this? In general this works absolutely fine, many boards have this enabled by default and many people use it
<GrantM11235[m]>
F1C200s
<apritzel>
ah yeah, that's rather new, so there might be some bits missing
<apritzel>
let me have a look...
<GrantM11235[m]>
It looks like it was only added to linux in 6.3
vagrantc has joined #linux-sunxi
<apritzel>
GrantM11235[m]: what exactly? the DT for the LCtech Pi?
<apritzel>
U-Boot has quite some messy dependencies in Kconfig, to enable certain features
<apritzel>
my hunch is that suniv is missing something in this fabric
<Jookia>
apritzel: just an update on the t113 nand spl, i am actually in nightmare land with it still but i think the suniv spl code and t113 code is solid. is the suniv spl code abandoned or just in the patch queue?
<apritzel>
Jookia: I had a brief look over the SPL NAND-SPI patches, and didn't particularly like them. I was hoping to simplify them. Also I need to check how the code size behaves, for normal SPI NOR on other SoCs
<apritzel>
Jookia: to be clear: the SPL SPI code is not particularly pretty to begin with, because it's optimised like hell to be small
<apritzel>
and if we now make it uglier and blowing it up, that defeats the purpose
<Jookia>
i see
<Jookia>
any alternate design you could think of?
<apritzel>
Jookia: are you interested in SPL operation, U-Boot proper operation or both?
<Jookia>
SPL operation so i can boot in to second stage from NAND
<GrantM11235[m]>
Speaking of spi nand, I figured out the problem I was having earlier with the help of the u-boot irc channel. The sf command that I was using only works with spi-nor, but the mtd command works once I enabled all the necessary configs
<apritzel>
GrantM11235[m]: so just enabling CONFIG_USB_MUSB_GADGET=y on top of the lctech_pi_f1c200s_defconfig compiled for me ...
<apritzel>
now searching for that board ...
<GrantM11235[m]>
I haven't submitted anything for that board upstream yet
<apritzel>
ah, I see now, the compatible string is missing from sunxi.c
<GrantM11235[m]>
I'm trying to enable CONFIG_CMD_USB_MASS_STORAGE
<GrantM11235[m]>
Is it possible to automatically enable all the dependent configs for a certain config? I have been manually searching for each config with menuconfig, which is very annoying
<apritzel>
I once sent a patch to enable UMS by default for all of sunxi, let me dig that out ...
<GrantM11235[m]>
Okay, CONFIG_USB_MUSB_GADGET=y seems to have fixed the compiler errors
<apritzel>
I *think* that should be the only one needed, the rest comes automatically
<apritzel>
GrantM11235[m]: and if you connect "allwinner,suniv-f1c100s-musb" to sun4i_a10_cfg towards the end of drivers/usb/musb-new/sunxi.c, it should even probe ...
<apritzel>
but there might be more ...
<apritzel>
unfortunately U-Boot and Linux differ quite a bit in this driver, so a port is not straight-forward, but I think that should be enough
<apritzel>
we detect has_reset automatically, and no_configdata is already enabled by Kconfig
<GrantM11235[m]>
hmm, I still don't have the ums command even after I enabled CONFIG_CMD_USB_MASS_STORAGE
<apritzel>
which leaves .has_sram, but apparently this already works somehow for the A10, so we might be lucky
<apritzel>
GrantM11235[m]: but you did enable the other config options, so that it showed up in menuconfig?
<GrantM11235[m]>
yes
<apritzel>
did it compile cmd/usb_mass_storage.c?
<GrantM11235[m]>
Oops, my mistake. Somehow I messed up copying uboot to my sdcard, so it was running a previous version
<apritzel>
yeah, common problem. I learned to always check for the build date in the U-Boot banner
<apritzel>
the ums command works, but I get Controller uninitialized
<GrantM11235[m]>
me too
<apritzel>
so reset, configdata and SRAM look fine on the board
vagrantc has quit [Ping timeout: 480 seconds]
<GrantM11235[m]>
Does the a10 usb controller get initialized in the bootrom or something?
<apritzel>
that's independent. I booted via FEL, so OTG definitely works, but the U-Boot operation is separate from that
<GrantM11235[m]>
It looks like the error is because a static struct musb *gadget is null, but I'm not sure where it is supposed to be set
<apritzel>
the key is the "musbp = &gadget;" line towards the end, which means musb_init_controller() returns NULL