Recent posts

#1

A64 / Re: Bookworm upgrade availabil...

Last post by mossroy - May 11, 2024, 02:57:23 PM

As for whom it may be relevant, now, halve a year after installing bookworm from scratch on a lime2 board, is is still rock-solid. Even better, homeassistant-supervised is stable and the regular updates install without any issues. If someone is interesed in the installation process for homeassistant, just drop me a personal message

You're certainly right on your device (A20-OLinuXino-LIME2?), and I would love to be able to use the regular debian installer as you do: much more installation options (like full-disk encryption), much more frequent kernel security updates etc

But stock debian is not working on the devices I have.
I have several A20-OLinuXino-MICRO, where I was happily using the regular debian until buster. But, since Bullseye, there are kernel panics that make it unsusable, see my blog post in 2022: https://blog.mossroy.fr/2022/03/31/upgrade-vers-bullseye-des-serveurs-olinuxino/

I also have several A64-OLinuXino-2Ge8G-IND, on which I never could install the regular debian.

I re-tested with debian bookworm in November 2023:
On A20-OLinuXino-MICRO board, the installation worked fine but, after logging in, and using the board a few minutes, there was this kernel panic:

Code Select Expand

[  855.175881] 8<--- cut here ---
[  855.179204] Unable to handle kernel paging request at virtual address fffffffe
[  855.186510] [fffffffe] *pgd=80000040207003, *pmd=6ffce003, *pte=00000000
[  855.193343] Internal error: Oops: 80000207 [#1] SMP ARM
[  855.198608] Modules linked in: binfmt_misc ecb des_generic aes_arm_bs crypto_simd cryptd sun4i_backend axp20x_battery axp20x_adc axp20x_ac_power lima axp20x_pek at24 gpu_sched drm_shmem_helper sun4i_codec snd_soc_core ftdi_sio sun4i_ts snd_pcm_dmaengine nvmem_sunxi_sid snd_pcm usbserial snd_timer snd soundcore sunxi_cedrus(C) sunxi_wdt v4l2_mem2mem videobuf2_dma_contig videobuf2_memops videobuf2_v4l2 videobuf2_common sun4i_drm_hdmi cec sun4i_ss videodev rc_core libdes mc display_connector leds_gpio cpufreq_dt evdev loop fuse dm_mod configfs ip_tables x_tables autofs4 ext4 crc16 mbcache jbd2 crc32c_generic axp20x_usb_power industrialio axp20x_regulator pinctrl_axp209 sunxi phy_generic ahci_sunxi dwmac_sunxi stmmac_platform stmmac pcs_xpcs phylink of_mdio fixed_phy fwnode_mdio libphy musb_hdrc libahci_platform libahci i2c_mv64xxx libata sun4i_lradc_keys ohci_platform scsi_mod ohci_hcd udc_core ehci_platform ehci_hcd scsi_common phy_sun4i_usb usbcore spi_sun4i sunxi_mmc sun4i_drm
[  855.199059]  sun4i_frontend sun4i_tcon sun8i_tcon_top drm_dma_helper drm_kms_helper drm
[  855.293448] CPU: 0 PID: 2719 Comm: dpkg-deb Tainted: G         C         6.1.0-12-armmp-lpae #1  Debian 6.1.52-1
[  855.303621] Hardware name: Allwinner sun7i (A20) Family
[  855.308843] PC is at 0xfffffffe
[  855.312001] LR is at free_pipe_info+0x90/0xc4
[  855.316375] pc : [<fffffffe>]    lr : [<c06f260c>]    psr: a0080033
[  855.322637] sp : f0d5def8  ip : 00000010  fp : b6e66368
[  855.327859] r10: 000000f8  r9 : 00000000  r8 : c87f1c38
[  855.333080] r7 : 00000000  r6 : 0000000e  r5 : 00000150  r4 : c3930800
[  855.339601] r3 : ffffffff  r2 : e08db1e0  r1 : c23eed38  r0 : c3930800
[  855.346124] Flags: NzCv  IRQs on  FIQs on  Mode SVC_32  ISA Thumb  Segment user
[  855.353429] Control: 30c5387d  Table: 43ce01c0  DAC: fffffffd
[  855.359169] Register r0 information: slab kmalloc-cg-128 start c3930800 pointer offset 0 size 128
[  855.368065] Register r1 information: slab kmalloc-cg-512 start c23eec00 pointer offset 312 size 512
[  855.377122] Register r2 information: non-slab/vmalloc memory
[  855.382785] Register r3 information: non-paged memory
[  855.387837] Register r4 information: slab kmalloc-cg-128 start c3930800 pointer offset 0 size 128
[  855.396719] Register r5 information: non-paged memory
[  855.401770] Register r6 information: non-paged memory
[  855.406820] Register r7 information: NULL pointer
[  855.411523] Register r8 information: slab dentry start c87f1c38 pointer offset 0 size 36
[  855.419624] Register r9 information: NULL pointer
[  855.424327] Register r10 information: non-paged memory
[  855.429464] Register r11 information: non-paged memory
[  855.434600] Register r12 information: zero-size pointer
[  855.439823] Process dpkg-deb (pid: 2719, stack limit = 0x2b4da20b)
[  855.446008] Stack: (0xf0d5def8 to 0xf0d5e000)
[  855.450366] dee0:                                                       c3930800 c212b900
[  855.458539] df00: c358b440 c21d19d0 c87f1c38 c06f2760 c212b900 002c0002 c358b440 c06e8f00
[  855.466713] df20: c0450b80 c3aed500 c22bdcc0 000000f8 b6e66368 c93fa840 c26ea8f4 c26ea100
[  855.474886] df40: 00000824 f0d5df70 00000000 c046e1bc c26ea100 c39d5c00 c26ea920 c39d5c38
[  855.483058] df60: f0d5df70 c0450b94 f0d5dfb0 b6e90c04 c1612d98 c0451438 00000000 c3a7b300
[  855.491232] df80: c26ea100 42654441 c04002e4 c9fd1500 b6e66368 c04513f8 000000f8 c045145c
[  855.499403] dfa0: 00000001 c04000c0 00000001 b6e63888 00000000 b6f9e220 00000000 00000000
[  855.507575] dfc0: 00000001 b6e63888 00000000 000000f8 b6e65000 b6e66368 b6e66368 b6e66368
[  855.515749] dfe0: 000000f8 beddba44 b6de2129 b6d77616 60080030 00000000 00000000 00000000
[  855.523935]  free_pipe_info from pipe_release+0xbc/0xc4
[  855.529182]  pipe_release from __fput+0x8c/0x270
[  855.533821]  __fput from task_work_run+0x94/0xc4
[  855.538457]  task_work_run from do_exit+0x30c/0x99c
[  855.543351]  do_exit from do_group_exit+0x40/0x8c
[  855.548071]  do_group_exit from __wake_up_parent+0x0/0x20
[  855.553490] Code: bad PC value
[  855.556668] ---[ end trace 0000000000000000 ]---
[  855.561385] Fixing recursive fault but reboot is needed!
On A64-OLinuXino-2Ge8G-IND, the installation itself worked fine too, but the first boot never reaches the login phase. The console just remains black

#2

A20 / Re: LIME2 Rev.L lock-up freeze...

Last post by LubOlimex - May 10, 2024, 08:14:47 AM

But what does it say at start of boot over the serial? It is important that the revision is properly listed as revision L. Often people would delete the contents of the EEPROM (thus disabling automatic board recognition by the Olimage) and then also fail to manually configure the board via the u-boot-tools and then the board would load some generic revision A preset that might have settings incompatible with the actual board revision used. Like RAM timings, different LAN controller and so.

#3

A20 / Re: LIME2 Rev.L lock-up freeze...

Last post by mbosschaert - May 09, 2024, 10:29:18 PM

CPU:   Allwinner A20 (SUN7I)
ID:    A20-OLinuXino-LIME2 Rev.L

It identifies like this:

Code Select Expand

CPU: ARMv7 Processor [410fc074] revision 4 (ARMv7), cr=10c5387d
OF: fdt: Machine model: Olimex A20-OLinuXino-LIME2
I'm running an up to date olimex debian bullseye

#4

A20 / Re: LIME2 Rev.L lock-up freeze...

Last post by LubOlimex - May 09, 2024, 02:48:31 PM

Compare the BOMs (Bill-Of-Materials) to confirm what is the part. But I doubt change in MCP120T475I-TT to MCP121T475I-TT can be any sort of issue.

Improper timings for the RAM memory are more likely. Are the boards and revisions properly listed at start of boot process? Like does it say:

CPU:   Allwinner A20 (SUN7I)
ID:    A20-OLinuXino-LIME2 Rev.L

#5

ESP32 / Re: ESP32-POE: esp_eth_driver_...

Last post by DirkB19 - May 09, 2024, 02:43:29 PM

@LubOlimex,

Thanks for the support, now it works.
I was getting a little desperate 
Because first I overlooked your tip to enable PSRAM ... this is definitely needed to make it work.
I would document this in Wiki or Documentation for other noobs like me who will encounter this.
Thanks,
DirkB19

#6

A20 / Re: LIME2 Rev.L lock-up freeze...

Last post by mbosschaert - May 08, 2024, 11:08:14 PM

I'm also still having these freezes with all of my Rev.L boards when a SATA HD (Or SSD) is connected, this is irrespective of the Olimex-debian version, the power supply, lipo connected, etc.

From this conversation I understand that Rev.K2 used to be stable with this respect.

Most important changes to consequtive versions of the board concern RAM, eMMC and the power supervisors. As I'm not using EMMC, and the suggestion that it is a power-handling problem make me consider to replace the power-supervisor component back to the one which is used in Rev.K2. Reading the revision document however confuses me a bit. On line 201 (Rev.K1) is stated that MCP121T475I-TT is used, however on line 215 (Rev.L) is stated that U14 is changed from MCP120T475I-TT to MCP121T475I-TT. The latter suggests that Rev.K1 has MCP120T475I-TT on U14. Or did I miss something here?

@LubOlimex, could you confirm this?

As around U14 there is sufficient space on the board I think exchanging this component should be possible.

Does this sound as test worthwile or is the idea not realistic or completely insane... ?

#7

A64 / Re: Bookworm upgrade availabil...

Last post by mbosschaert - May 08, 2024, 01:34:41 PM

As for whom it may be relevant, now, halve a year after installing bookworm from scratch on a lime2 board, is is still rock-solid. Even better, homeassistant-supervised is stable and the regular updates install without any issues. If someone is interesed in the installation process for homeassistant, just drop me a personal message

#8

ESP32 / Re: ESP32-POE: esp_eth_driver_...

Last post by LubOlimex - May 08, 2024, 10:20:08 AM

I use latest ESP32 package which appears to be version 3.0.0-rc1. Changed the defines of default example at start to:

Code Select Expand

// Important to be defined BEFORE including ETH.h for ETH.begin() to work.
// Example RMII LAN8720 (Olimex, etc.)
#ifndef ETH_PHY_TYPE
#define ETH_PHY_TYPE        ETH_PHY_LAN8720
#define ETH_PHY_ADDR        0
#define ETH_PHY_MDC        23
#define ETH_PHY_MDIO        18
#define ETH_PHY_POWER      12
#define ETH_CLK_MODE        ETH_CLOCK_GPIO0_OUT
#endif
Also reduced delay at end to 1000:

Code Select Expand

void loop()
{
  if (eth_connected) {
    testClient("google.com", 80);
  }
  delay(1000);
}
The board I selected is "Olimex ESP32-POE" and then changed PSRAM to enabled. Uploaded and it seems to work fine.

#9

AVR / Re: AVR-ISP MK2 Failed to get ...

Last post by LubOlimex - May 08, 2024, 08:12:50 AM

It should be some sort of driver issue. libusb-win32 is fine as I've tested previously.

It feels like this same problem: https://www.olimex.com/forum/index.php?topic=5975.0

#10

ESP32 / Re: ESP32-POE: esp_eth_driver_...

Last post by LubOlimex - May 08, 2024, 08:09:56 AM

I will test it again later today and if it works here I will give you instructions on what I use and what I did.

It is important to notice each ESP32-POE2 is tested empirically and the Ethernet is part of the test. Even if there is a problem it is not related to the hardware.

Edit: I found the issue - the board comes with ESP32-WROVER-E module. The information everywhere says it comes with WROOM but upon checking empirically the product it comes with ESP32-WROVER-E-N4R8. This causes two pin difference since WROVER module has PSRAM (and the PSRAM requires two pins extra). This is visible in the schematic top left corner of Ethernet box - now eMAC clock is GPIO0, not GPIO17.