Re: [U-Boot] [BUG] booting on the Tinker Board (tinker-rk3288_defconfig) is broken

David Wu Fri, 01 Mar 2019 04:36:26 -0800

Hi Heiko,

在 2019/2/28 下午9:29, Heiko Stübner 写道:

Hi Philipp,


Am Donnerstag, 28. Februar 2019, 13:46:44 CET schrieb Heiko Stübner:

Am Donnerstag, 28. Februar 2019, 13:36:52 CET schrieb Philipp Tomsich:

On 28.02.2019, at 11:50, Heiko Stübner <he...@sntech.de> wrote:

Hi David,

Am Montag, 18. Februar 2019, 02:05:12 CET schrieb David Wu:

Hi Heinrich and Michael,

Another thing i see is that I missed a patch, for the 3288 gpio0, its
iomux is special, there is no high 16-bit write-enabled bit. For Tinker
board, it uses I2C0, the current driver will overwrite the I2C0 iomux,
while request the GPIO0A4. It requires a patch:

http://patchwork.ozlabs.org/patch/1040541/
<http://patchwork.ozlabs.org/patch/1040541/>>


I've also just stumbled onto the issue of recent uboots not booting
on rk3288. While your patch seems to have reached patchwork, somehow
the u-boot list seems to have lost it - as it's neither in public
archives nor in my inbox it seems.


Please note that I had requested changes to that patch during review.
I’ll ping David again on when he expect to have those changes implemented
and ready.


yeah, I saw that in patchwork, but couldn't comment due to the patch
seemingly not having reach the actual u-boot list strangely.

That patch doesn't seem to help me right now anyway, as I somehow seem
be stuck on mem_malloc_init() hanging when called from initr_malloc()
(board_r.c) and am not yet sure what is going on there
(on u-boot-master from last friday).


This problem seems to be memory-related somehow?
My 2GB rk3288-evb comes up fine to the uboot prompt, while
my 1GB rk3288-phycore hangs in said mem_malloc_init().


Is this issue introduced by the pinctrl patch?


rk3288-evb (2GB):
[...]
Relocation Offset is: 7ed88000
Relocating to 7ed88000, new gd at 7cd7fed8, sp at 7cd76230
[...]
initr_malloc: before mem_malloc_init
using memory 0x7cd80000-0x7ed88000 for malloc()
initr_malloc: after mem_malloc_init
[...]


rk3288-phycore (1GB):
[...]
Relocation Offset is: 3ff7b000
Relocating to 3ff7b000, new gd at 3df72ee0, sp at 3df68f60
[...]
initr_malloc: before mem_malloc_init
using memory 0x3df73000-0x3ff7b000 for malloc()
[HANG]


Do you have some tips what I need to poke for this issue?


Heiko

Applying the patch does not make my board (phycore-rk3288 in this case)
get farther though - I'll investigate more.


But it looks like we should be having the same issue in the kernel, just
never triggered it, as the gpio0 are of more esotheric uses normally.

Anyway, I'm wondering if defining IOMUX_WRITABLE_32BIT alone
wouldn't be enough and use that for iomux, pull and drv, similar to
what we do for the pull/drv register source already.

That way we could refrain from introducing DRV_TYPE_WRITABLE_32BIT
and PULL_TYPE_WRITABLE_32BIT .


Heiko

在 2019/2/17 下午8:41, Heinrich Schuchardt 写道:

On 2/17/19 1:18 PM, Michael Nazzareno Trimarchi wrote:

Hi

[U-Boot] [PATCH 3/5] rockchip: rk3288-vyasa: increase heap space
after
relocation

Can you check it if you have the same problem?


Applying all the changes causes SPL not to start.

CONFIG_SYS_MALLOC_F_LEN=0x4000
does not solve the problem.

Best regards

Heinrich

Michael


On Sun., 17 Feb. 2019, 1:11 pm Heinrich Schuchardt
<xypron.g...@gmx.de

<mailto:xypron.g...@gmx.de> wrote:
     On 2/17/19 9:19 AM, David Wu wrote:

Hi Henrich,

在 2019/2/16 下午5:53, Heinrich Schuchardt 写道:

On 2/13/19 11:56 AM, Philipp Tomsich wrote:

     <snip>


Hello David, hello Philipp,

what are your ideas to reduce the SPL size to under 0x7800 again?
Or will you move all rk3288 boards to TPL like
TARGET_VYASA_RK3288?


CONFIG_SPL_I2C_SUPPORT is necessary for Tink spl? Remove this can
make
spl boot. As the Tinker does not use the i2c to read the MAC
address
from eeprom at the SPL stage, which is at uboot.

     Hello David,

the SPL image size now just fits:

     Image Type:   Rockchip RK32 (SD/MMC) boot image
     Data Size:    30720 bytes

SPL is successful. But MMC is failing in main U-Boot:```

     U-Boot SPL 2019.04-rc1-00239-gb89074f650 (Feb 17 2019 - 12:41:39
     +0100)
     Returning to boot ROM...

U-Boot 2019.04-rc1-00239-gb89074f650 (Feb 17 2019 - 12:41:39

     +0100)

Model: Tinker-RK3288

     DRAM:  2 GiB
     MMC:   dwmmc@ff0c0000: 1
     Loading Environment from MMC...
     ```

No further output here.With some debug output enabled:```

     U-Boot SPL 2019.04-rc1-00239-gb89074f650-dirty (Feb 17 2019 -
     13:05:10
     +0100)
     Returning to boot ROM...
     mmc_bind: alias ret=0, devnum=1
     env_init: Environment MMC init done (ret=-2)

U-Boot 2019.04-rc1-00239-gb89074f650-dirty (Feb 17 2019 -

     13:05:10
     +0100)

Model: Tinker-RK3288

     DRAM:  2 GiB
     mmc_bind: alias ret=0, devnum=1
     MMC:   dwmmc@ff0c0000: 1
     Loading Environment from MMC...
     ```

Any suggestion how to proceed?CC: Jaehoon.Best regardsHeinrich


_______________________________________________
U-Boot mailing list
U-Boot@lists.denx.de <mailto:U-Boot@lists.denx.de>
https://lists.denx.de/listinfo/u-boot
<https://lists.denx.de/listinfo/u-boot>>


_______________________________________________
U-Boot mailing list
U-Boot@lists.denx.de <mailto:U-Boot@lists.denx.de>
https://lists.denx.de/listinfo/u-boot
<https://lists.denx.de/listinfo/u-boot>



_______________________________________________
U-Boot mailing list
U-Boot@lists.denx.de
https://lists.denx.de/listinfo/u-boot

Re: [U-Boot] [BUG] booting on the Tinker Board (tinker-rk3288_defconfig) is broken

Reply via email to