00:00
ErwinH has joined #linux-sunxi
00:04
ErwinH has quit [Ping timeout: 260 seconds]
00:08
ErwinH has joined #linux-sunxi
00:12
ErwinH has quit [Ping timeout: 258 seconds]
00:18
Andy-D has quit [Ping timeout: 252 seconds]
00:21
ErwinH has joined #linux-sunxi
00:21
jernej has quit [Ping timeout: 248 seconds]
00:22
bbrezillon has quit [Ping timeout: 265 seconds]
00:22
bbrezill1 has joined #linux-sunxi
00:25
ErwinH has quit [Ping timeout: 264 seconds]
00:27
mcan_ has quit [Ping timeout: 256 seconds]
00:29
ErwinH has joined #linux-sunxi
00:29
vagrantc has quit [Quit: leaving]
00:33
ErwinH has quit [Ping timeout: 248 seconds]
00:38
ErwinH has joined #linux-sunxi
00:39
TheLinuxBug has quit [Ping timeout: 240 seconds]
00:42
ErwinH has quit [Ping timeout: 250 seconds]
00:43
ccaione has quit [Ping timeout: 260 seconds]
00:43
hramrach has quit [Ping timeout: 260 seconds]
00:43
ccaione has joined #linux-sunxi
00:43
ccaione has joined #linux-sunxi
00:44
hramrach has joined #linux-sunxi
00:44
TheLinuxBug has joined #linux-sunxi
00:52
interrobangd has quit [Ping timeout: 246 seconds]
00:54
mcan has joined #linux-sunxi
00:54
ErwinH has joined #linux-sunxi
00:57
victhor has quit [Ping timeout: 258 seconds]
00:58
ErwinH has quit [Ping timeout: 260 seconds]
01:03
[Awaxx] has quit [Quit: "how do I feel when the pad comes with the pin! I need a sledge hammer to finishIt!"]
01:08
ErwinH has joined #linux-sunxi
01:11
interrobangd has joined #linux-sunxi
01:12
ErwinH has quit [Ping timeout: 268 seconds]
01:18
[Awaxx] has joined #linux-sunxi
01:18
ErwinH has joined #linux-sunxi
01:22
ErwinH has quit [Ping timeout: 250 seconds]
01:32
ErwinH has joined #linux-sunxi
01:36
ErwinH has quit [Ping timeout: 246 seconds]
01:38
ErwinH has joined #linux-sunxi
01:42
ErwinH has quit [Ping timeout: 258 seconds]
01:51
ErwinH has joined #linux-sunxi
01:56
ErwinH has quit [Ping timeout: 268 seconds]
01:59
ErwinH has joined #linux-sunxi
02:00
cnxsoft has joined #linux-sunxi
02:03
ErwinH has quit [Ping timeout: 245 seconds]
02:03
deskwizard has quit [Ping timeout: 264 seconds]
02:08
ErwinH has joined #linux-sunxi
02:12
ErwinH has quit [Ping timeout: 250 seconds]
02:18
ErwinH has joined #linux-sunxi
02:22
ErwinH has quit [Ping timeout: 256 seconds]
02:28
ErwinH has joined #linux-sunxi
02:28
egbert has quit [Disconnected by services]
02:28
egbert has joined #linux-sunxi
02:31
Net147 has quit [Ping timeout: 256 seconds]
02:32
ErwinH has quit [Ping timeout: 250 seconds]
02:32
Net147 has joined #linux-sunxi
02:37
libv_ has joined #linux-sunxi
02:38
ErwinH has joined #linux-sunxi
02:40
libv has quit [Ping timeout: 248 seconds]
02:42
ErwinH has quit [Ping timeout: 260 seconds]
02:53
ErwinH has joined #linux-sunxi
02:53
ninolein has quit [Ping timeout: 240 seconds]
02:53
kronicd has joined #linux-sunxi
02:54
ninolein has joined #linux-sunxi
02:57
ErwinH has quit [Ping timeout: 250 seconds]
03:01
popolon has quit [Quit: WeeChat 1.4]
03:09
ErwinH has joined #linux-sunxi
03:14
ErwinH has quit [Ping timeout: 264 seconds]
03:21
ErwinH has joined #linux-sunxi
03:25
ErwinH has quit [Ping timeout: 246 seconds]
03:26
KB3VGW has joined #linux-sunxi
03:26
KB3VGW has quit [Remote host closed the connection]
03:29
ErwinH has joined #linux-sunxi
03:33
ErwinH has quit [Ping timeout: 245 seconds]
03:38
ErwinH has joined #linux-sunxi
03:40
interrobangd has quit [Quit: Leaving]
03:42
ErwinH has quit [Ping timeout: 260 seconds]
03:49
ErwinH has joined #linux-sunxi
03:52
nOOb__ has quit [Ping timeout: 265 seconds]
03:53
ErwinH has quit [Ping timeout: 265 seconds]
04:00
ErwinH has joined #linux-sunxi
04:04
ErwinH has quit [Ping timeout: 265 seconds]
04:08
ErwinH has joined #linux-sunxi
04:12
ErwinH has quit [Ping timeout: 246 seconds]
04:14
chomwitt1 has joined #linux-sunxi
04:16
chomwitt has quit [Ping timeout: 256 seconds]
04:17
pg12 has quit [Ping timeout: 258 seconds]
04:18
ErwinH has joined #linux-sunxi
04:19
pg12 has joined #linux-sunxi
04:22
ErwinH has quit [Ping timeout: 258 seconds]
04:23
dave0x6d has quit [Quit: Connection closed for inactivity]
04:27
dave0x6d has joined #linux-sunxi
04:32
ErwinH has joined #linux-sunxi
04:32
terra854 has joined #linux-sunxi
04:36
ErwinH has quit [Ping timeout: 256 seconds]
04:38
ErwinH has joined #linux-sunxi
04:42
ErwinH has quit [Ping timeout: 252 seconds]
04:50
ErwinH has joined #linux-sunxi
04:54
ErwinH has quit [Ping timeout: 246 seconds]
05:00
ErwinH has joined #linux-sunxi
05:04
ErwinH has quit [Ping timeout: 256 seconds]
05:08
ErwinH has joined #linux-sunxi
05:12
ErwinH has quit [Ping timeout: 252 seconds]
05:15
JohnDoe_71Rus has joined #linux-sunxi
05:24
ErwinH has joined #linux-sunxi
05:29
ErwinH has quit [Ping timeout: 258 seconds]
05:38
ErwinH has joined #linux-sunxi
05:42
ErwinH has quit [Ping timeout: 264 seconds]
05:49
ErwinH has joined #linux-sunxi
05:53
ErwinH has quit [Ping timeout: 258 seconds]
05:55
reinforce has joined #linux-sunxi
05:57
TheSeven has quit [Ping timeout: 240 seconds]
05:57
TheSeven has joined #linux-sunxi
06:04
f0xx has joined #linux-sunxi
06:05
ErwinH has joined #linux-sunxi
06:07
ErwinH has quit [Read error: No route to host]
06:08
ErwinH has joined #linux-sunxi
06:12
ErwinH has quit [Ping timeout: 264 seconds]
06:22
ErwinH has joined #linux-sunxi
06:26
ErwinH has quit [Ping timeout: 252 seconds]
06:29
ErwinH has joined #linux-sunxi
06:30
f0xx has quit [Ping timeout: 248 seconds]
06:31
chomwitt1 has quit [Ping timeout: 256 seconds]
06:33
f0xx has joined #linux-sunxi
06:33
ErwinH has quit [Ping timeout: 258 seconds]
06:36
a1d3s has joined #linux-sunxi
06:37
<
a1d3s >
good morning
06:38
ErwinH has joined #linux-sunxi
06:42
ErwinH has quit [Ping timeout: 256 seconds]
06:46
<
KotCzarny >
likewise: btrfs + snapshots?
06:46
<
KotCzarny >
or any other fs with that feature
06:48
leviathanch has joined #linux-sunxi
06:50
ErwinH has joined #linux-sunxi
06:51
leviathanch has quit [Remote host closed the connection]
06:52
leviathanch has joined #linux-sunxi
06:54
ErwinH has quit [Ping timeout: 246 seconds]
06:59
indy has quit [Ping timeout: 264 seconds]
07:03
ErwinH has joined #linux-sunxi
07:07
ErwinH has quit [Ping timeout: 250 seconds]
07:14
IgorPec has joined #linux-sunxi
07:18
ErwinH has joined #linux-sunxi
07:23
ErwinH has quit [Ping timeout: 265 seconds]
07:30
ErwinH has joined #linux-sunxi
07:34
ErwinH has quit [Ping timeout: 250 seconds]
07:34
iamfrankenstein has joined #linux-sunxi
07:38
ErwinH has joined #linux-sunxi
07:42
ErwinH has quit [Ping timeout: 256 seconds]
07:47
cnxsoft1 has joined #linux-sunxi
07:49
cnxsoft has quit [Ping timeout: 258 seconds]
07:49
cnxsoft1 is now known as cnxsoft
07:50
ErwinH has joined #linux-sunxi
07:53
dave0x6d has quit [Quit: Connection closed for inactivity]
07:54
perr has joined #linux-sunxi
07:54
ErwinH has quit [Ping timeout: 258 seconds]
07:54
perr has quit [Remote host closed the connection]
07:54
perr has joined #linux-sunxi
07:54
perr has quit [Changing host]
07:54
perr has joined #linux-sunxi
07:54
f0xx has quit [Ping timeout: 250 seconds]
07:56
dave0x6d has joined #linux-sunxi
07:59
ErwinH has joined #linux-sunxi
08:03
ErwinH has quit [Ping timeout: 258 seconds]
08:06
netlynx has joined #linux-sunxi
08:08
ErwinH has joined #linux-sunxi
08:12
ErwinH has quit [Ping timeout: 250 seconds]
08:24
ErwinH has joined #linux-sunxi
08:28
ErwinH has quit [Ping timeout: 256 seconds]
08:31
indy has joined #linux-sunxi
08:35
<
igraltist >
on long term compiling my orangipi-pc got hot to 75°C
08:35
<
igraltist >
i think i have to put some cooling stuff
08:38
ErwinH has joined #linux-sunxi
08:38
<
scelestic >
igraltist: with a heatsink?
08:42
chomwitt1 has joined #linux-sunxi
08:43
ErwinH has quit [Ping timeout: 268 seconds]
08:44
<
igraltist >
yes this i need
08:44
jernej has joined #linux-sunxi
08:48
f0xx has joined #linux-sunxi
08:50
Wizzup has quit [Remote host closed the connection]
08:50
Wizzup has joined #linux-sunxi
08:52
ErwinH has joined #linux-sunxi
08:56
ErwinH has quit [Ping timeout: 250 seconds]
08:59
ErwinH has joined #linux-sunxi
09:03
ErwinH has quit [Ping timeout: 258 seconds]
09:05
<
a1d3s >
but it didnt find the root partition right
09:08
<
igraltist >
lol with tor can't access
09:08
ErwinH has joined #linux-sunxi
09:09
<
igraltist >
i should build a map which provider blocking tor
09:10
<
beeble >
a1d3s: check the polarity of your CD gpio
09:11
<
a1d3s >
beeble for the first try i only used the m2-plus dts , i know them is wrong for ultra
09:11
<
a1d3s >
but some things work :)
09:12
ErwinH has quit [Ping timeout: 260 seconds]
09:13
<
beeble >
in that case the mmc controller doesn't match in more than the CD section :)
09:15
<
a1d3s >
i found in sinovoip bsp tree cd-gpios = <&pio PF 6 0 1 2 0>; and in mainline plus: cd-gpios = <&pio 5 6 GPIO_ACTIVE_HIGH>; /* PF6 */
09:15
<
a1d3s >
what to do with 0 1 2 0 ?
09:24
ErwinH has joined #linux-sunxi
09:27
f0xx has quit [Ping timeout: 250 seconds]
09:27
<
MoeIcenowy >
I really got a sample of the V3s-based Lichee Pi Zero
09:28
ErwinH has quit [Ping timeout: 260 seconds]
09:29
ErwinH has joined #linux-sunxi
09:32
perr has quit [Quit: Leaving]
09:33
ErwinH has quit [Ping timeout: 265 seconds]
09:38
ErwinH has joined #linux-sunxi
09:38
jstein_ has joined #linux-sunxi
09:40
jstein_ is now known as jstein
09:42
ErwinH has quit [Ping timeout: 264 seconds]
09:42
<
beeble >
a1d3s: mux, pu/pd, drive, output level
09:44
Mr__Anderson has joined #linux-sunxi
09:48
ErwinH has joined #linux-sunxi
09:49
f0xx has joined #linux-sunxi
09:52
ErwinH has quit [Ping timeout: 246 seconds]
09:55
<
MoeIcenowy >
V3s's SoC ID is 0x1681, only one more than H3
09:59
ErwinH has joined #linux-sunxi
10:01
petr has quit [Remote host closed the connection]
10:02
petr has joined #linux-sunxi
10:03
ErwinH has quit [Ping timeout: 245 seconds]
10:08
ErwinH has joined #linux-sunxi
10:12
ErwinH has quit [Ping timeout: 258 seconds]
10:12
kaspter has joined #linux-sunxi
10:14
f0xx has quit [Ping timeout: 268 seconds]
10:24
ErwinH has joined #linux-sunxi
10:27
BenG83 has joined #linux-sunxi
10:28
ErwinH has quit [Ping timeout: 245 seconds]
10:32
f0xx has joined #linux-sunxi
10:40
ErwinH has joined #linux-sunxi
10:45
ErwinH has quit [Ping timeout: 264 seconds]
10:49
ErwinH has joined #linux-sunxi
10:53
ErwinH has quit [Ping timeout: 260 seconds]
10:58
iamfrankenstein has quit [Ping timeout: 252 seconds]
11:02
ErwinH has joined #linux-sunxi
11:06
ErwinH has quit [Ping timeout: 245 seconds]
11:08
ErwinH has joined #linux-sunxi
11:12
ErwinH has quit [Ping timeout: 245 seconds]
11:20
ErwinH has joined #linux-sunxi
11:20
kaspter has quit [Ping timeout: 245 seconds]
11:24
ErwinH has quit [Ping timeout: 260 seconds]
11:27
<
beeble >
a1d3s: take a look at your mmc0 pin definition. could it be that it is set to xxx_NO_PULL? if if so try it with xxx_PULL_UP
11:28
<
a1d3s >
mmc0_pins_a: mmc0@0 allwinner,pull = <SUN4I_PINCTRL_NO_PULL>; mmc0_cd_pin: mmc0_cd_pin@0 { allwinner,pull = <SUN4I_PINCTRL_PULL_UP>;
11:29
ErwinH has joined #linux-sunxi
11:29
<
wens >
mmc0_pins_a should be PULL_UP
11:29
<
a1d3s >
wens ok will try
11:29
<
a1d3s >
and cd_pin NO_PULL?
11:30
<
a1d3s >
or both up?
11:30
<
beeble >
depends on the hardware. keep it PU since it seems to work now
11:30
<
a1d3s >
i'm not a programmer , all i do is try and error :)
11:33
ErwinH has quit [Ping timeout: 258 seconds]
11:33
paulk-aldrin has joined #linux-sunxi
11:34
<
a1d3s >
wens your u-boot works , but why it is only 1GB Dram?
11:37
fl_0 has quit [Ping timeout: 260 seconds]
11:38
<
a1d3s >
ok i thought i did something wrong :)
11:38
ErwinH has joined #linux-sunxi
11:38
<
MoeIcenowy >
wens: I got an Allwinner V3s board now
11:38
<
MoeIcenowy >
how can I port U-Boot to it?
11:39
<
jelle >
allwinner v3?
11:40
<
KotCzarny >
cheapo one core, dual cam soc
11:40
<
jelle >
all allwinners are cheap :D
11:40
<
wens >
MoeIcenowy: figure out dram :/
11:40
<
MoeIcenowy >
oh libdram...
11:41
<
MoeIcenowy >
the disaster
11:41
<
jelle >
oh action cam stuff
11:41
<
a1d3s >
the PULL_UP for mmc0 changed nothing .. sunxi-mmc 1c11000.mmc: smc 2 err, cmd 18, RD EBE !! <- this error is from eMMC (mmc2)
11:42
fl_0 has joined #linux-sunxi
11:42
ErwinH has quit [Ping timeout: 246 seconds]
11:42
<
beeble >
a1d3s: does your board have emmc? if so then enable PU there too
11:43
<
a1d3s >
beeble yes it have 8GB
11:46
<
wens >
a1d3s: add pull_up for it as well?
11:49
ErwinH has joined #linux-sunxi
11:51
<
a1d3s >
i pushed it to git
11:54
<
wens >
to be honest i haven't spent much time on the r40, other than adding u-boot support for it
11:56
ErwinH has quit [Ping timeout: 245 seconds]
11:57
<
MoeIcenowy >
wens: should I port U-Boot first, or should I port Linux first, to a new SoC?
11:58
ErwinH has joined #linux-sunxi
11:58
<
beeble >
a1d3s: i hope you fixed your conflicts in the dts. otherwise it will not build and you are still using the old one
11:59
<
wens >
MoeIcenowy: depends on if you have something to boot it or not :p
11:59
<
a1d3s >
beeble the pi starts but very slow and hangs around on systemd
11:59
<
wens >
though having u-boot makes it much more interesting for others
12:00
netlynx has quit [Quit: Ex-Chat]
12:00
<
beeble >
you have some more unresolved conflichts in the file
12:00
<
a1d3s >
beele yes i have fixed this
12:00
<
beeble >
ok, just wanted to be sure that you are not using old files
12:02
ErwinH has quit [Ping timeout: 252 seconds]
12:02
<
a1d3s >
this was the latest commit
12:03
victhor has joined #linux-sunxi
12:04
premoboss has joined #linux-sunxi
12:05
netlynx has joined #linux-sunxi
12:08
<
beeble >
looks fine otherwise. you could try to change the bus width to 4 as a workaround if it's a timing issue
12:12
cajg has joined #linux-sunxi
12:14
<
premoboss >
hi, i have NANOPI NEO, OS armbian 5.20, linux 3.4.112. in can connect via ethernet cross cable (Static ip). if i use minicom, i see on PC the uboot messages, kernel messages, and login prompt but i canot login (usb/rs232 csable works good with other boards). but mi BIG probblme is rhgat i cannot user usb0 from pc to NEO, so i enter via eth0 and do modprobe g_ether and usb0 go up (i placed auto usb0; iface usb0 inet static, address 192.168.10.1; net
12:14
<
premoboss >
mask 255.255.255.0 into /etc/network/interfaces of NEO). so on NEO usb0 is up but it dont make usb0 pf PC going up. cable is ok and PC confioguration is ok (it works woth other embedded linux board). what is frong? maybe some config of Armbian? i ask help.
12:18
foxx has joined #linux-sunxi
12:20
leviathanch has quit [Remote host closed the connection]
12:23
ErwinH has joined #linux-sunxi
12:27
ErwinH has quit [Ping timeout: 250 seconds]
12:29
premoboss has quit [Remote host closed the connection]
12:32
leio has quit [Read error: Connection reset by peer]
12:32
leio has joined #linux-sunxi
12:33
ErwinH has joined #linux-sunxi
12:33
ErwinH has quit [Client Quit]
12:34
<
MoeIcenowy >
wens, jemk: I remember you mentioned there's related documents about mctl?
12:35
kaspter has joined #linux-sunxi
12:37
<
wens >
MoeIcenowy: imx6 reference manual
12:38
<
MoeIcenowy >
wens: thx
12:38
<
MoeIcenowy >
I think for V3s there's another way to initialize DRAM...
12:39
<
MoeIcenowy >
directly restore a register dump to the mctl regs
12:41
<
NiteHawk >
MoeIcenowy: V3s pull request done - I'll now rebase the H3/SID fix on top of that and also merge it, unless you have any objections
12:41
<
MoeIcenowy >
nitehawk: thx ;-)
12:43
<
a1d3s >
my old 286er is faster :)
12:48
<
beeble >
it brings only one core up. but even with that it shouln't feel that slow
12:49
<
beeble >
and there could be a clocking issue
12:49
<
beeble >
do you get a shell?
12:52
foxx has quit [Ping timeout: 268 seconds]
12:53
<
a1d3s >
i'll try without systemd/udev if it runs faster an a login comes up
12:53
<
a1d3s >
the image is a minimal debootstraped jessie with kernel 4.9
12:54
<
a1d3s >
maybe i forgot some modules , didnt know :)
12:54
<
MoeIcenowy >
wens: In which situation will the controller returns "DRAM: 0 MiB" ?
12:56
<
beeble >
a1d3s: modules shouldn't matter for clockspeed. you can add init=/bin/sh to your kernel cmdline to just get a prompt
12:57
<
beeble >
and mybe look into sysfs cpufreq stats
13:02
<
KotCzarny >
MoeIcenowy: misdetection most likely
13:02
<
KotCzarny >
ie. wrong code
13:15
<
MoeIcenowy >
KotCzarny: I know that it's misdetection ;-)
13:20
<
wens >
yes you will likely need a new driver for both the ccu and pinctrl for r40
13:21
<
a1d3s >
ok this is a thing for tomorrow
13:21
<
a1d3s >
now i will go home
13:22
<
a1d3s >
thx beeble and wens
13:22
a1d3s has quit [Quit: Leaving]
13:29
victhor has quit [Ping timeout: 252 seconds]
13:37
<
MoeIcenowy >
could anyone help me to analysis V3s's DRAM controller dump?
13:37
interrobangd has joined #linux-sunxi
13:41
<
NiteHawk >
wens: did you by any chance find an opportunity to verify the SID quirk (observed on H3) for A80?
13:46
mzki has quit [Ping timeout: 252 seconds]
13:48
kronicd has quit [Ping timeout: 252 seconds]
13:49
kronicd has joined #linux-sunxi
13:58
<
MoeIcenowy >
who can help me to check it?
13:59
<
MoeIcenowy >
jemk: thx!
13:59
<
MoeIcenowy >
what basic knowledge is needed to write DRAM controller drivers?
13:59
<
jemk >
looks a lot like h3, same zq quirk
14:00
<
MoeIcenowy >
but directly with H3 code path do not work
14:00
<
MoeIcenowy >
(the DRAM integrated is DDR2 according to BSP boot log
14:01
<
jemk >
oh, ddr2 isn't supported yet, it will be quite a bit different
14:03
<
MoeIcenowy >
what will be needed to add ddr2 support?
14:03
<
beeble >
for example ddr2 doesn't have zq calibration at all :)
14:04
<
jemk >
different timings, initialisation, some mode registers maybe
14:06
<
MoeIcenowy >
can I just use the register dump to initialize it?
14:06
<
beeble >
there are init sequences
14:06
cnxsoft has quit [Quit: cnxsoft]
14:06
<
jemk >
no, the initialisation sequences have to be run correctly
14:07
<
jemk >
... beeble is faster in typing ;)
14:07
<
beeble >
jemk: i cheat with shorter sentences :)
14:08
<
KotCzarny >
can it just work by register dump restore?
14:09
<
jemk >
the most important change would be MCTL_CR_DDR2 in mctl_set_cr i think
14:10
<
wens >
NiteHawk: no difference reading directly or through registers (i assume sid-registers forces this?)
14:10
<
wens >
NiteHawk: tested with sunxi-fel
14:11
<
jemk >
MoeIcenowy: two problems with this code, license and it isn't correct either
14:11
<
MoeIcenowy >
isn't correct either?!
14:13
<
jemk >
it doesn't have the zq calibration quirk and only sets fixed timings
14:13
<
beeble >
jemk: ddr2 doesn't have a zq calibration, thats a ddr3 feature
14:14
<
wens >
nice to have a hardware engineer here :)
14:14
<
jemk >
beeble: only the dram chips donthe controller can still do it, and the boot0 dump looks like it does
14:14
<
jemk >
*don't have it*
14:15
a1d3s_ has joined #linux-sunxi
14:19
<
MoeIcenowy >
by changing DDR3 to DDR2 in dram_sun8i_h3.c it still didn't work for V3s
14:19
<
MoeIcenowy >
or maybe we should also write fixed timings?
14:20
<
wens >
mainline u-boot doesn't support ddr2
14:20
<
beeble >
jemk: ah, i see. mctl does have zq calibration for ddr2. would probably work without too
14:21
<
MoeIcenowy >
wens: yes it's now the problem
14:21
<
beeble >
as long as you try different drive strenghts. but yeah, zq calibration for writes can help a lot
14:21
<
MoeIcenowy >
we got a chip which has only ddr2
14:22
<
jemk >
MoeIcenowy: there are more things of course, for example the dram mode registers, but they can be copied from the dump for now
14:23
<
jemk >
beeble: the zq cal isn't very good on sunxi anyway, so fixed values might be ok as well
14:24
<
jemk >
but you have to set the driver and odt strengths
14:24
<
NiteHawk >
wens: thx! yes, "sid-registers" would enforce the workaround method of reading via register - would only make a difference if the first "sunxi-fel sid" (memory access) would not yield the expected result
14:25
<
jemk >
MoeIcenowy: i usually add a register dump to u-boot and diff it with boot0 dump, then fix the differences one after another
14:25
<
MoeIcenowy >
could I create a dedicated driver for V3s which uses the BSP values to run the init process?
14:25
<
MoeIcenowy >
I think for V3s we have less situations
14:26
<
MoeIcenowy >
as all the DRAM wafers are the same
14:26
<
jemk >
hard-coded modifying the h3 code should be faster
14:27
<
MoeIcenowy >
which part is most important? DRAMCTL0 (started at 0x01c63000)?
14:28
<
jemk >
yes, and 0x01c62000
14:29
<
jemk >
did you set a low enough dram clock, i don't think it can do 672mhz
14:31
<
MoeIcenowy >
I set 360
14:36
reinforce has quit [Quit: Leaving.]
14:41
a1d3s__ has joined #linux-sunxi
14:45
a1d3s_ has quit [Ping timeout: 245 seconds]
14:45
a1d3s__ has quit [Client Quit]
14:52
<
MoeIcenowy >
mctl_channel_init function will return with error...
14:53
deskwizard has joined #linux-sunxi
14:59
<
jemk >
MoeIcenowy: try this
http://sprunge.us/fSVH i changed what i noticed from your boot0 dump, the remaining stuff could then be diffed with the dump you get from u-boot
15:05
libv_ is now known as libv
15:08
<
MoeIcenowy >
jemk: got 32MiB!
15:08
Ntemis has joined #linux-sunxi
15:10
<
MoeIcenowy >
oh it's because some my silly changes
15:10
<
MoeIcenowy >
just use your patch can give me 64 MiB
15:10
<
MoeIcenowy >
it seems that it didn't perfectly went back to main U-Boot...
15:12
massi has joined #linux-sunxi
15:26
<
MoeIcenowy >
oh forgot to add the first V3s SID on the wiki
15:26
<
MoeIcenowy >
NiteHawk: it seems that V3s FEL boot met some problem...
15:27
<
MoeIcenowy >
after executing SPL, it can go back to FEL, and sunxi-fel {sid,ver} works
15:27
<
MoeIcenowy >
however, it didn't allow to execute SPL again...
15:27
<
MoeIcenowy >
(For my former SoCs, this time SPL can be executed again
15:28
<
MoeIcenowy >
and fail with usb_bulk_send error
15:31
deskwizard has quit [Remote host closed the connection]
15:31
leviathanch has joined #linux-sunxi
15:32
IgorPec has joined #linux-sunxi
15:33
deskwizard has joined #linux-sunxi
15:34
<
jemk >
MoeIcenowy: can you paste the register dump from u-boot?
15:34
andoma_ has quit [Ping timeout: 268 seconds]
15:35
kaspter has quit [Ping timeout: 245 seconds]
15:35
<
NiteHawk >
MoeIcenowy: are you using the BSP SPL? I think I have observed similar behaviour on other SoCs with BSP vs. mainline U-Boot, this phenomenon is not entirely unknown
15:36
<
MoeIcenowy >
NiteHawk: I'm using my homebrew SPL
15:36
<
MoeIcenowy >
jemk: still cannot run the main U-Boot
15:37
<
jemk >
the spl should dump the regs too with my patch
15:38
andoma has joined #linux-sunxi
15:38
<
MoeIcenowy >
but I removed the message now
15:38
interrobangd has quit [Ping timeout: 260 seconds]
15:39
victhor has joined #linux-sunxi
15:41
<
MoeIcenowy >
or maybe just I clocked the CPU core too high
15:41
reinforce has joined #linux-sunxi
15:42
paulk-aldrin has quit [Remote host closed the connection]
15:45
Mr__Anderson has quit [Ping timeout: 265 seconds]
15:47
<
jemk >
thx, looks like there is a bit more to do
15:50
interrobangd has joined #linux-sunxi
15:50
Mr__Anderson has joined #linux-sunxi
15:51
<
MoeIcenowy >
yes...
15:51
<
jernej >
MoeIcenowy: Does A64 U-Boot work if it is loaded through FEL?
15:51
<
MoeIcenowy >
the DRAM content is in a mess
15:52
<
MoeIcenowy >
jernej: the U-Boot itself works...
15:52
<
MoeIcenowy >
but I still cannot get graphics work
15:52
<
jernej >
that's I will try to fix tonight :)
15:52
<
jernej >
so, which file? u-boot-dtb.bin?
16:00
<
jernej >
oh, sorry, nonstandard cable needed for FEL, I guess it will have to wait until I receive proper cable
16:01
Andy-D has joined #linux-sunxi
16:02
popolon has joined #linux-sunxi
16:05
mzki has joined #linux-sunxi
16:09
<
MoeIcenowy >
jemk: I found what I writes to DRAM is not equal to what I read
16:09
<
MoeIcenowy >
is it because the DRAM controller is not fully correctly populated?
16:10
<
jemk >
there are some strange things with the bus-width
16:11
<
jemk >
it only has 16 bit, but in mctl_cr it is set to 32 bit
16:12
<
MoeIcenowy >
oh... 0x4a000000 seems to be beyond 64MiB DRAM...
16:12
<
jemk >
how much ram does it really have btw?
16:12
<
MoeIcenowy >
64MiB .
16:14
<
MoeIcenowy >
the usual loading address of our mainline u-boot is 0x4a000000...
16:18
<
MoeIcenowy >
jemk: why do you comment out mctl_set_master_priority?
16:19
<
jemk >
because boot0 doesn't set it, i just try to do what boot0 does
16:20
<
jemk >
we can later think about what this means, first it just has to work
16:21
<
jemk >
found another mistake, we need bank detection, your ddr2 only has 4 banks, you should change to MCTL_CR_FOUR_BANKS in mctl_set_cr before trying
16:22
<
MoeIcenowy >
at that time I got 32 MiB ...
16:23
<
jemk >
you can't trust this output anymore, it doesn't take number of banks into account
16:24
<
MoeIcenowy >
and the displayed size become 32MiB
16:26
<
jemk >
but now the phy stuff looks correct, so after fixing size detect it should work
16:26
<
MoeIcenowy >
how to fix size detect?
16:28
f0xx has quit [Ping timeout: 260 seconds]
16:29
<
jemk >
i hope adding a bank detection is enough
16:29
<
MoeIcenowy >
but we hardcoded it to 4 banks...
16:29
<
jemk >
and the detection still assumes there are 8
16:30
<
MoeIcenowy >
I cannot understand this...
16:33
<
MoeIcenowy >
is there other things that could be forgot to be doubled?
16:34
|Jeroen| has joined #linux-sunxi
16:35
<
KotCzarny >
i think wens had similar trouble with a80
16:36
<
Ntemis >
hey guys does H3 support uEnv.txt ?
16:36
<
KotCzarny >
h3 doesnt
16:36
<
KotCzarny >
uboot does
16:36
<
MoeIcenowy >
KotCzarny: he's R40
16:37
<
MoeIcenowy >
which met the trouble
16:37
<
KotCzarny >
MoeIcenowy: maybe, dont remember exactly
16:37
<
Ntemis >
KotCzarny: so not possible?
16:39
<
KotCzarny >
ntemis: so your question wasnt exactly on point
16:39
<
KotCzarny >
its up to uboot, not soc
16:39
<
KotCzarny >
if you use legacy uboot, no, if you use mainline uboot, yes
16:39
<
Ntemis >
ah great thanks
16:41
<
KotCzarny >
also, how is your 800mhz? did you confirm its snake oil or real thing?
16:41
deskwizard has quit [Ping timeout: 264 seconds]
16:43
<
MoeIcenowy >
jemk: another problem
16:43
<
MoeIcenowy >
I hacked the loading address of u-boot to 0x41000000 and entered U-Boot main binary
16:43
<
MoeIcenowy >
but it stucks at "DRAM: 0 Bytes"
16:47
<
jemk >
one step after another, first we need a matching register dump, then we are pretty sure dram is initialised correctly
16:51
vagrantc has joined #linux-sunxi
16:54
scelestic has quit [Read error: Connection reset by peer]
16:54
<
MoeIcenowy >
I think the DRAM is in fact 32bit...
16:55
<
MoeIcenowy >
yes, it's why it is reported as 32 MB
16:56
<
MoeIcenowy >
oh your 2nd patch fixed this
16:56
<
beeble >
MoeIcenowy: a partnumber could clarify that
16:56
<
MoeIcenowy >
beeble: it's a in-package die
16:56
<
MoeIcenowy >
no partnumber
16:56
<
MoeIcenowy >
or I should say the DRAM's partnumber is just "Allwinner V3s"
16:57
<
beeble >
ah, multi die package. i see.
16:59
<
jemk >
there are only two data byte lanes in the phy, so only 16 bit
16:59
<
MoeIcenowy >
but according to BSP cr register, it's 32-bit
17:00
<
MoeIcenowy >
and with 32-bit settings at least U-Boot is not corrupted
17:01
<
jemk >
maybe this bit means full and half buswidth instead of 32 and 16, then it would fit
17:01
<
jemk >
i don't think we ever found a name anywhere
17:01
<
MoeIcenowy >
yes...
17:02
<
beeble >
ah, so it's a V3s
17:02
<
MoeIcenowy >
how can I detect whether DRAM is stable?
17:02
massi has quit [Remote host closed the connection]
17:02
<
beeble >
uboot has a memtester command. good for a first try
17:02
<
MoeIcenowy >
but I haven't entered U-Boot...
17:03
<
beeble >
ok, so still in spl i see
17:03
<
beeble >
then write to dram adresses and read back
17:03
<
beeble >
if it's the same you have a good chance that you are doing something right
17:04
<
beeble >
everything is moving on this new allwinner website. my head is spinning
17:09
<
jemk >
if i didn't miss anything and the detection is correct this should now produce a 1:1 boot0 identical dump (except some volatile status registers)
17:10
<
jemk >
ah, its on top of the last patch now...
17:11
<
MoeIcenowy >
should the bank detection be in H3?
17:11
<
jemk >
if it works we should enable it everywhere where ddr2 could happen
17:11
<
MoeIcenowy >
or we should only keep it for V3s?
17:12
<
MoeIcenowy >
I think finally all the code should be #ifdef CONFIG_MACH_SUN8I_H3_DDR2, right?
17:12
Pepe has joined #linux-sunxi
17:12
<
jemk >
all the stuff i did now needs massive cleanup if it works, too many things just hardcoded now
17:15
<
MoeIcenowy >
but many things are replacing one hardcode with another hardcode ;-)
17:15
Harrier has quit [Remote host closed the connection]
17:16
<
jemk >
can you paste a (hopefully last for now) dump with this patch, if everything fits cleanup can start
17:18
Harrier has joined #linux-sunxi
17:19
<
MoeIcenowy >
jemk: ^
17:19
_fortis has quit [K-Lined]
17:21
<
jemk >
perfect, looks like boot0, so dram should work from the hardware point of view
17:23
Mr__Anderson has quit [Remote host closed the connection]
17:23
<
MoeIcenowy >
you may want to rename the ifdefs, right? ;-)
17:23
<
jemk >
since thats more ddr2 specific than v3 specific, yes
17:25
<
MoeIcenowy >
but the U-Boot payload still argues with "0 Bytes DRAM"...
17:25
<
MoeIcenowy >
(Even if itself lies in DRAM
17:26
leviathanch has quit [Remote host closed the connection]
17:28
paulk-collins has joined #linux-sunxi
17:30
<
jemk >
i have never looked into the u-boot part, no idea how their dram size detect works, it just worked for me
17:36
LargePrime has joined #linux-sunxi
17:39
<
MoeIcenowy >
jernej: it's only some left testing code. You can just uncomment it, and it won't work :-(
17:40
<
jernej >
as I said, I will wait until A to A USB cable comes
17:40
<
KotCzarny >
make one yourself from 2 old mice?
17:40
<
jernej >
but please try this: uncomment above mentioned line and replace "clrbits_le32(SUN8I_DE_SEL_REG, 1);" with "setbits_le32(SUN8I_DE_SEL_REG, 1);"
17:41
<
KotCzarny >
or two old printer cables
17:44
<
jernej >
well, there is a box with cut cables at my work, but I'm on vacation till 9th...
17:52
scelestic has joined #linux-sunxi
17:58
JohnDoe_71Rus has joined #linux-sunxi
18:21
mrnuke has joined #linux-sunxi
18:24
mzki has quit [Ping timeout: 265 seconds]
18:26
joost_dtn has joined #linux-sunxi
18:27
<
MoeIcenowy >
got the reason for U-Boot's "0Bytes DRAM"
18:27
<
MoeIcenowy >
U-Boot reserved 64MB for malloc
18:27
<
MoeIcenowy >
but the board have only 64MB
18:27
joost_dtn has quit [Client Quit]
18:28
<
KotCzarny >
such a kidder
18:29
<
MoeIcenowy >
not U-Boot reserved 64MB
18:29
<
MoeIcenowy >
it's sunxi-common.h configs to reserve 64MB
18:32
<
NiteHawk >
malloc(the_world) :D
18:35
foxx has joined #linux-sunxi
18:36
<
MoeIcenowy >
ok it seems like some romantic sentence
18:36
<
MoeIcenowy >
give(you, malloc(the_whole_world)) ;-)
18:37
<
MoeIcenowy >
although I think we the mainline developer have never thought the situation that we have only 64MiB as the whole world ;-)
18:37
<
NiteHawk >
hey, 640K ought to be enough for everyone
18:38
<
beeble >
fyi: a large ubifs can require a lot of malloc size during recovery
18:38
<
beeble >
large >= 2GB
18:39
<
MoeIcenowy >
something to mention: the official system firmware media of V3s is... SPI NOR flash
18:42
<
NiteHawk >
interesting. so the lichee has an spi flash onboard?
18:48
<
KotCzarny >
how much?
19:03
<
MoeIcenowy >
I don't know how much either...
19:04
JohnDoe_71Rus has joined #linux-sunxi
19:04
<
MoeIcenowy >
my board is a very early sample, that needs a wire to solve the power problem ;-)
19:04
<
MoeIcenowy >
NiteHawk: yes, and the SPI flash is intended to be one of the main storages
19:11
<
KotCzarny >
MoeIcenowy: does it come with any os?
19:13
TheLinuxBug has quit [Ping timeout: 248 seconds]
19:16
Nacho has quit [Ping timeout: 246 seconds]
19:17
TheLinuxB has joined #linux-sunxi
19:17
gzamboni has quit [Quit: Ex-Chat]
19:17
gzamboni has joined #linux-sunxi
19:18
gzamboni has quit [Remote host closed the connection]
19:26
mzki has joined #linux-sunxi
19:31
terra854 has quit [Quit: Connection closed for inactivity]
19:37
nikre has joined #linux-sunxi
19:38
<
jonkerj_ >
I've ordered a opi one, to enhance my development rig. Cannot wait to show the result off :-)
19:38
jonkerj_ is now known as jonkerj
19:38
<
jonkerj >
it will replace the usb-uarts, and the usb-part in the usb-relay
19:39
<
jonkerj >
also, a 5-port ethernet switch which can be powered from my 5v brick
19:39
<
jonkerj >
so it will connect to mains and to some ethernet port, and offerrs remote UART and powercycle for my h3 boards
19:39
parazyd has joined #linux-sunxi
19:39
<
KotCzarny >
make a howto on your user page
19:39
<
KotCzarny >
with pics
19:40
<
jonkerj >
of course
19:40
<
jonkerj >
that will be half of the fun
19:46
Ntemis has quit [Remote host closed the connection]
19:57
Nacho has joined #linux-sunxi
20:02
netlynx has quit [Quit: Ex-Chat]
20:02
LargePrime has quit [Ping timeout: 256 seconds]
20:16
LargePrime has joined #linux-sunxi
20:29
interrobangd has quit [Ping timeout: 246 seconds]
20:32
paulk-collins has quit [Remote host closed the connection]
20:32
scream has joined #linux-sunxi
20:36
indy has quit [Ping timeout: 246 seconds]
20:44
indy has joined #linux-sunxi
20:50
interrobangd has joined #linux-sunxi
20:59
aballier has quit [Ping timeout: 258 seconds]
21:00
aballier has joined #linux-sunxi
21:08
iamfrankenstein has joined #linux-sunxi
21:08
deskwizard has joined #linux-sunxi
21:11
Putti has quit [Quit: Leaving]
21:14
TheLinuxB has quit [Ping timeout: 246 seconds]
21:18
foxx has quit [Ping timeout: 260 seconds]
21:19
reinforce has quit [Quit: Leaving.]
21:24
<
nikre >
is the gpio order different in opi one and pc?
21:31
TheLinuxBug has joined #linux-sunxi
21:31
Mr__Anderson has joined #linux-sunxi
21:32
Putti has joined #linux-sunxi
21:48
mzki has quit [Ping timeout: 246 seconds]
21:50
mzki has joined #linux-sunxi
21:51
Putti has quit [Quit: Leaving]
21:51
jstein__ has joined #linux-sunxi
21:51
jstein is now known as Guest18859
21:53
jstein__ is now known as jstein
21:55
miasma has quit [Ping timeout: 258 seconds]
21:55
Guest18859 has quit [Ping timeout: 246 seconds]
21:58
interrobangd has quit [Ping timeout: 260 seconds]
22:02
miasma has joined #linux-sunxi
22:04
iamfrankenstein1 has joined #linux-sunxi
22:05
iamfrankenstein has quit [Ping timeout: 248 seconds]
22:05
iamfrankenstein1 is now known as iamfrankenstein
22:16
|Jeroen| has quit [Quit: dada]
22:31
Putti has joined #linux-sunxi
22:41
nikre has quit [Quit: Leaving]
22:42
jstein has quit [Remote host closed the connection]
22:57
scream has quit [Remote host closed the connection]
23:07
Putti has quit [Quit: Leaving]
23:19
Putti has joined #linux-sunxi
23:22
Mr__Anderson has quit [Remote host closed the connection]
23:29
Putti has quit [Quit: Leaving]
23:32
deskwizard has quit [Ping timeout: 258 seconds]
23:38
cptG_ has joined #linux-sunxi
23:42
cptG has quit [Ping timeout: 268 seconds]
23:51
interrobangd has joined #linux-sunxi
23:54
bugzc has joined #linux-sunxi