home  bbs  files  messages ]

      ZZLI4427             linux.debian.maint.boot             505 messages      

[ previous | next | reply ]

[ list messages | list forums ]

  Msg # 226 of 505 on ZZLI4427, Saturday 9-12-25, 3:47  
  From: =?UTF-8?Q?=E6=A2=81=E5=AE  
  To: KIRK REISER  
  Subj: Re: Deep Computing DC-ROMA ii won't boot  
 [continued from previous message] 
  
 > > Next, let's talk u-boot vs UEFI. U-boot requires a lot more than UEFI, as 
 has been mentioned previously. However, there is a lot more information 
 hidden 
 in that statement than you might think. For example, you can't just move the 
 dtb file to the 
 right location. The "dtb" extension stands for DeviceTree Binary, and is 
 directly tied to the kernel version being run. Simply moving the binary into 
 the right location will cause a version mis-match and cause the hardware to 
 not boot properly. Also, 
 with the version of U-boot the DC ROMA II uses, the entire boot stack is 
 stored on the storage medium (be it SD card or NVMe). This is opposed to 
 UEFI, 
 where you just have flash on the motherboard that is smart enough to reach 
 out 
 to the storage to find 
 your bootloader to get the process started. If you look in the installer ISO 
 for Debian, in /boot/dtbs I believe, it lists all the compatible chips, and 
 SpacemiT is not in there. Until it is, the Debian ISO will not work, and 
 given 
 the pace of the 
 SpacemiT crew, I'd hesitantly say expect that to be added in Forky. 
 > > 
 > > 
 > > To reiterate, simply using Debian Trixie on this device at this point in 
 time *will*. *not*. *work*. The standard Debian kernel does not support the 
 hardware yet, and the SpacemiT kernel you will likely have to rebuild from 
 scratch to get the modules 
 you need for accessibility purposes running, which in my experience is 
 *very* 
 hit or miss getting it to boot afterwards.€€ 
 > > 
 > > 
 > > I own this laptop, and as a person who is lucky enough to not need any 
 accessibility settings, it is frankly a nightmare to use in it's current 
 state. Simply running system updates is not an option, and I've had to 
 completely reinstall the operating 
 system on mine several times because I forgot. I've tried off-and-on since I 
 bought it at least a year ago, and it's currently gathering dust next to my 
 other K1/M1 system while I wait for the upstreaming effort to finish. Even 
 after the CPU gets 
 upstreamed, owners of this laptop will probably need to use DeepComputing's 
 custom ISO while Imagination Technologies (the GPU vendor) gets their act 
 together and finally merges their changes to mesa into upstream.€€ 
 > > 
 > > 
 > > I would highly recommend reading through the issues in the DC ROMA II 
 Github page (see here: https://github.com/DC-DeepComputing/DC-RO 
 A_Gen2_LAPTOP_K1_RV-L2A€€), just to get a sense for the state of the device 
 as 
 a whole. It's clunky, it's not ready, 
  and it's largely been forgotten by DeepComputing as far as I can tell while 
 they figure out their Framework Mainboard endeavour. The JH7110 SOC is kinda 
 the only good RISC-V chip to recommend right now for anything outside the 
 absolute most niche cases, 
 because it's been almost entirely upstreamed, and therefore is supported by 
 the Debian installer natively. RISC-V is a really cool technology, and I 
 love 
 it a lot, but the hardware ecosystem right now is about the same as the 
 Raspberry Pi 2 was when it 
 came out, and I mean that both from a software support standpoint and from a 
 hardware performance perspective. 
 > > 
 > > 
 > > To be perfectly frank, if I could talk to my past self, I would say to 
 not 
 buy this laptop and save myself the migraines. In a few years, it will be 
 better, but the hardware barely runs on the hardware manufacturer blessed 
 distro images. The fact 
 that anyone tried to cram this chip in a laptop is a testament to the 
 arrogance of man, because a laptop appeals to normal people, and this laptop 
 is at best a marketing stunt to drum up good PR for RISC-V on the whole. 
 > > 
 > > 
 > > 
 > > 
 > > 
         On Thu, 11 Sept 2025 at 18:01, Kirk Reiser  wrote: 
  
 > > > 
         Hello folks: Thank you for the responses. 
  
 > > > 
 > > > 
 > > > 
         €€The laptop did infact come with ubuntu-24.04 installed, except it 
 was 
  
 > > > 
         €€a greatly reduced package, they say because it wasn't expected 
 that a 
  
 > > > 
         €€lot of people would login. It was kind of bogus in my mind. 
  
 > > > 
 > > > 
 > > > 
         €€I upgraded the OS to their most recent version of 24.04.3 except 
 it 
  
 > > > 
         €€craps out when trying to upgrade to their idea of the latest 
 kernel 
  
 > > > 
         €€6.14.0. The distro as shipped has kernel 6.6.36 except with many 
  
 > > > 
         €€kernel modules missing. The upgrade of the kernel dies while 
 trying 
 to 
  
 > > > 
         €€copy the dtb files from a nonexistant spacemit directory in 
 /boot. I 
  
 > > > 
         €€have tried many things to attempt to spoof the new kernel from 
 copying 
  
 > > > 
         €€over the installed dtbs to a new 6.14.0 directory with no joy. 
  
 > > > 
 > > > 
 > > > 
         €€Oh, and yes, it is using u-boot. 
  
 > > > 
 > > > 
 > > > 
         €€Unfortunately Some of the kernel modules which are missing are the 
  
 > > > 
         €€speakup modules which I must have to provide speech in text 
 console 
  
 > > > 
         €€mode. That has also made it much more difficult because it means I 
  
 > > > 
         €€have to have a sighted person here to help me 
 install/break/reinstall 
  
 > > > 
         €€various packages. 
  
 > > > 
 > > > 
 > > > 
         €€Upon examining the two debian images I tried to boot I realized 
 that 
  
 > > > 
         €€the netinst is grub based so wouldn't boot. The dvd however does 
  
 > > > 
         €€appear to be u-boot based so I am still at a loss for the reason 
 it 
  
 > > > 
         €€won't boot. 
  
 > > > 
 > > > 
 > > > 
         €€I don't really understand the booting mechanism of u-boot, even 
 after 
  
 > > > 
         €€reading what I've been able to find. 
  
 > > > 
 > > > 
 > > > 
         €€Anyway, the spacemit is an 8-core SOC called K1X I believe. The 
  
 > > > 
         €€processor is a M1-8571. Here is the cpu line from dmesg on the 
 system: 
  
 > > > 
 > > > 
 > > > 
         €€spacemit-socinfo soc:socinfo@0: Spacemit: CPU[M1-8571] REV[C] 
 DRO[127] Detected 
  
 > > > 
 > > > 
 > > > 
         €€So thank you and if you have any other recommendations I would 
  
 > > > 
         €€appreciate them. 
  
 > > > 
 > > > 
 > > > 
         €€€€€€ Kirk 
  
 > > > 
 > > > 
 > > > €€ 
 > > > 
         €€On Thu, 11 Sep 2025, Greg Sterling wrote: 
  
 > > > 
 > > > 
 > > > 
         €€> Hello! 
  
 > > > 
         €€> 
  
 > > > 
         €€> My boss thought I might be able to get you contact with the 
 devboards community for help.€€ If you€€€re interested, you might be able to 
 get some help by sending an email to devboard-community@riscv.org.€€ There 
 are 
 a lot of helpful people 
 out there who may have experienced similar problems and may be able to offer 
 some help or suggestions. 
  
 > > > 
         €€> 
  
 > > > 
         €€> Thanks and have a great day! 
  
 > > > 
         €€> 
  
 > > > 
         €€> Greg Sterling (RISC-V International) 
  
 > > > 
         €€> Phone: +1 603 321 7320 
  
 > > > 
         €€> E-mail: greg@riscv.org 
 > > > 
         €€> 
  
 > > > 
         €€>> On Sep 10, 2025, at 8:04€€€PM, Jeff Scheel  
 wrote: 
  
 > > > 
         €€>> 
  
 > > > 
         €€>> You might want to try and connect with Kirk with the DevBoard 
 community to see€€ if someone can help here... 
  
 > > > 
         €€>> 
  
 > > > 
         €€>> -- 
  
 > > > 
         €€>> Jeff Scheel (he/him/his) 
  
 > > > 
  
 [continued in next message] 
  
 --- SoupGate-Win32 v1.05 
  * Origin: you cannot sedate... all the things you hate (1:229/2) 

[ list messages | list forums | previous | next | reply ]

search for:

328,123 visits
(c) 1994,  bbs@darkrealms.ca