Jump to content

Banana PRO: Testers wanted! SATA drive not working on some boards.

Gary Blake ヅ

By Gary Blake ヅ
in Allwinner sunxi

 Share

Recommended Posts

nihilista

nihilista

Posted
Posted

Tested both kernels, no sata disk present with both.

Got error while installing, but maybe thats not the problem:

Errors were encountered while processing:
 linux-dtb-next-next-sunxi_5.00_armhf.deb
main

main

Posted
Posted

I had another issue with SATA on Banana Pro, Armbian 5.0, kernel 4.4.1.

[    3.447472] ahci-sunxi 1c18000.sata: PHY power up failed.
[    3.447540] ahci-sunxi: probe of 1c18000.sata failed with error -5

Found the same problem here http://forum.armbian.com/index.php/topic/643-no-sata-on-fresh-compile/

 

But patch didn't helped me. I've hardcoded power up for SATA in u-boot for Banana Pro board, recompiled and now it's working.

I'm new here, so I've chosen bananapi in "Choose a board" dialog. Is it correct for Banana Pro?

zador.blood.stained

zador.blood.stained

Posted
Posted

@nihilista

Did you use 4.5 or 5.00 image? And can you please check your dmesg for any SATA/AHCI related messages like in @main's comment?

 

@main

So installing patched u-boot on top of Armbian 5.00 image without changing anything else gives you working SATA?

Edit: Patch didn't work probably because u-boot for all bananas use "bananapi_defconfig" and patch changes "bananapro_defconfig".

nihilista

nihilista

Posted
Posted

Ok, my easy way was using ext2explore on my windows pc to read sdcard, i looked in kern.log and found that:

ahci-sunxi 1c18000.sata: PHY power up failed.
ahci-sunxi: probe of 1c18000.sata failed with error -5

Installing on 4.5 image maybe later or tomorrow. But if i understand correct @main has got the solution? So its necessary to test with 4.5 images?

Gary Blake ヅ

Gary Blake ヅ

Posted
  • Author
Posted

I'm not sure. I don't see "power up failed" message here, and here installing only kernel and dtb packages without updating u-boot breaks SATA.

I just checked and my log has the power up issue so it sounds like main is onto the problem and has a fix.

zador.blood.stained

zador.blood.stained

Posted
Posted

OK, I pushed necessary patch to my repo, and if it doesn't break anything for other A20 Banana boards, Igor will probably push new u-boot and image soon.

 

@main, thanks for pointing out that that issue was related to this, it's hard to keep track of everything.

nihilista

nihilista

Posted
Posted

Thx for fast fixing that, after Igor pushed that we will only have to apt-get upgrade? (I'm asking because i spend half the day with getting old kernel back to sdcard while system on sata ssd was unbootable ;-) )

zador.blood.stained

zador.blood.stained

Posted
Posted

People with rootfs on SATA drives will have to update u-boot on their sdcard manually, otherwise if using SATA for data only, either "apt-get install --reinstall linux-u-boot-bananapi-next" if Igor doesn't change version or "apt-get upgrade" if version will be bumped.

nihilista

nihilista

Posted
Posted

People with rootfs on SATA drives will have to update u-boot on their sdcard manually, otherwise if using SATA for data only, either "apt-get install --reinstall linux-u-boot-bananapi-next" if Igor doesn't change version or "apt-get upgrade" if version will be bumped.

Allright, thx. Sorry for that (maybe) stupid question, but it must be installed to sdcard too (my rootfs is on sata drive, yes)? And if yes, how to do that as sdcard isnt used (mounted to /media/mmc with bind /boot to sata's /boot) ? As you see, i'm a bit noobish with linux ;-)

zador.blood.stained

zador.blood.stained

Posted
Posted

U-boot is loaded from a boot block on SD card, so you will need to unpack u-boot package and write it using "dd" manually. You can use card reader and any Linux system (including another Armbian).

For example, if your card will be detected as /dev/sdb, you will need to write it with

dd if=u-boot-sunxi-with-spl.bin of=/dev/sdb bs=1024 seek=8

Alternatively, you can just flash Armbian 5.00 on the card, update u-boot using apt and reconfigure boot.cmd (or copy backed up version from old card) to use rootfs on SATA drive.

tkaiser

tkaiser

Posted
Posted

OK, I pushed necessary patch to my repo, and if it doesn't break anything for other A20 Banana boards

 

Is it really safe to apply the patch for the Pro to both Banana Pi and Pro? Or is it time to start to differentiate between both boards?

nihilista

nihilista

Posted
Posted

U-boot is loaded from a boot block on SD card, so you will need to unpack u-boot package and write it using "dd" manually. You can use card reader and any Linux system (including another Armbian).

For example, if your card will be detected as /dev/sdb, you will need to write it with

dd if=u-boot-sunxi-with-spl.bin of=/dev/sdb bs=1024 seek=8

Alternatively, you can just flash Armbian 5.00 on the card, update u-boot using apt and reconfigure boot.cmd (or copy backed up version from old card) to use rootfs on SATA drive.

Thx again, that should work for me !

zador.blood.stained

zador.blood.stained

Posted
Posted

bananapi.fex and bananapipro.fex both have this entry:

ldo4_vol = 2800

sun7i-a20-bananapro.dts and sun7i-a20-bananapi.dts don't touch ldo4 settings at all; so safe enough for now, for testing.

 

Edit: It may be necessary to separate u-boot configs for pi and pro (they have different default settings), but upgrading existing installations to use different u-boot package names will be tricky.

 

Edit 2: I checked BPi-M1 schematics, and LDO4 there is not used. And for now I don't want to register on lemaker's forum just to check other schematics.

main

main

Posted
Posted
@main

So installing patched u-boot on top of Armbian 5.00 image without changing anything else gives you working SATA?

Edit: Patch didn't work probably because u-boot for all bananas use "bananapi_defconfig" and patch changes "bananapro_defconfig".

While searching for solution, I've changed ext4load mmc 0 0x49000000 /boot/dtb/sun7i-a20-bananapro.dtb in /boot/boot.cmd

But seems, You provided working solution for now.

Igor

Igor

Posted
Posted

linux-u-boot-next-bananapi_5.01_armhf.deb is in repository - compiled with patch, so apt-get upgrade should work now (on SD media)

1
nihilista

nihilista

Posted
Posted

Did an apt-get upgrade, flashed new u-boot manually to mmc with dd method, rebooted and now its armbian 5.0 with working sata ssd, all works :thumbup::)

Gordon

Gordon

Posted
Posted

Did an apt-get upgrade, and also using dd to push uboot manually to mmc.

But my sata hard disk is still not coming up.

 

Here the dmesg I have.

root@bananapi:~# dmesg|grep ata
[    0.000000] CPU: PIPT / VIPT nonaliasing data cache, VIPT aliasing instruction cache
[    0.000000] Memory policy: Data cache writealloc
[    0.000000] Memory: 1010820K/1046952K available (6822K kernel code, 256K rwdata, 1760K rodata, 328K init, 280K bss, 19748K reserved, 16384K cma-reserved, 244136K highmem)
      .data : 0xc08bc000 - 0xc08fc1a0   ( 257 kB)
[    3.205286] libata version 3.00 loaded.
[    3.477962] ahci-sunxi 1c18000.sata: controller can't do PMP, turning off CAP_PMP
[    3.478019] ahci-sunxi 1c18000.sata: SSS flag set, parallel bus scan disabled
[    3.478050] ahci-sunxi 1c18000.sata: AHCI 0001.0100 32 slots 1 ports 3 Gbps 0x1 impl platform mode
[    3.478067] ahci-sunxi 1c18000.sata: flags: ncq sntf stag pm led clo only pio slum part ccc 
[    3.479650] ata1: SATA max UDMA/133 mmio [mem 0x01c18000-0x01c18fff] port 0x100 irq 33
[    3.827920] ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
[    8.827910] ata1.00: qc timeout (cmd 0xec)
[    8.827933] ata1.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[    9.177887] ata1: SATA link down (SStatus 0 SControl 300)

And this sata hard disk can show up as soon as I switch to old version 4.5 or other distro with 4.3.3 kernal version.

Igor

Igor

Posted
Posted

Now we have one success and one failure with the same fix on the same hardware. WTF ? :)

nihilista

nihilista

Posted
Posted

Did an apt-get upgrade, and also using dd to push uboot manually to mmc.

But my sata hard disk is still not coming up.

 

Here the dmesg I have.

root@bananapi:~# dmesg|grep ata
[    0.000000] CPU: PIPT / VIPT nonaliasing data cache, VIPT aliasing instruction cache
[    0.000000] Memory policy: Data cache writealloc
[    0.000000] Memory: 1010820K/1046952K available (6822K kernel code, 256K rwdata, 1760K rodata, 328K init, 280K bss, 19748K reserved, 16384K cma-reserved, 244136K highmem)
      .data : 0xc08bc000 - 0xc08fc1a0   ( 257 kB)
[    3.205286] libata version 3.00 loaded.
[    3.477962] ahci-sunxi 1c18000.sata: controller can't do PMP, turning off CAP_PMP
[    3.478019] ahci-sunxi 1c18000.sata: SSS flag set, parallel bus scan disabled
[    3.478050] ahci-sunxi 1c18000.sata: AHCI 0001.0100 32 slots 1 ports 3 Gbps 0x1 impl platform mode
[    3.478067] ahci-sunxi 1c18000.sata: flags: ncq sntf stag pm led clo only pio slum part ccc 
[    3.479650] ata1: SATA max UDMA/133 mmio [mem 0x01c18000-0x01c18fff] port 0x100 irq 33
[    3.827920] ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
[    8.827910] ata1.00: qc timeout (cmd 0xec)
[    8.827933] ata1.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[    9.177887] ata1: SATA link down (SStatus 0 SControl 300)

And this sata hard disk can show up as soon as I switch to old version 4.5 or other distro with 4.3.3 kernal version.

 

The only thing i did in addition was changing "ext4load mmc 0 0x49000000 /boot/dtb/sun7i-a20-bananapro.dtb in /boot/boot.cmd", but i think that should not make the difference.

Guest
This topic is now closed to further replies.
 Share
Go to topic listing
×
×
  • Create New...

Important Information

Terms of Use - Privacy Policy - Guidelines

  I accept