On Sat, Nov 20, 2021 at 10:48:41AM +0100, H. Nikolaus Schaller wrote: > Hi Grond, > > > Am 19.11.2021 um 23:36 schrieb Grond <[email protected]>: > > > > > > Hi. > > > > Having recently encountered a slew of software incompatibility issues on > > my Pandora related to it's ancient (3.2) kernel, I thought I give the > > letux kernel a try as I'd heard it could at least boot on the Pandora. > > Yes, just tested yesterday... > But with some Letux OS Debian. > > What I do not know if it boots well with the original OS. >
Hmm. That's odd. I was able to get kernels to boot (both the ones that came from the makesd script and my own custom-built ones), but the only way I was able to tell that anything was running was the heartbeat LED flashing. The LCD screen was never initialized. I've ordered a Pandora serial adapter to try and debug the problem, but in the meantime, what kernel did you get to boot? I built and tried letux-5.4.160 (7c7255c3463641b8f699472f9114b0435dbfe707) and whichever 5.4.160 kernel is currently installed via makesd (probably the same git revision). > > > > Unfortunately, it seems that there is something wrong with the git > > server HTTP(S) server instance running on git.goldelico.com. I've tried > > both cloning the repo from the URL given on the project page > > (https://git.goldelico.com/letux-kernel.git) and adding it as a remote > > to my existing kernel source trees. In both cases it gets stuck with the > > message "Fetching objects: 2" printed to the terminal, and never > > completes, even after an hour or two (though it does continue to consume > > bandwidth the whole time). > > > > > I have managed to download the letux kernel source from both the github > > mirror and over git protocol via git://git.goldelico.com/gta04-kernel.git > > without issues (although I have also encountered similar problems with > > downloading from the later source in the past). But I thought I'd send a > > problem report to this mailing list, as I suspect that the maintainer(s) > > of git.goldelico.com are subscribed and might not have encountered this > > issue before. > > Well, this is more or less expected. The git repo has grown to almost 3GB > and the server has a 10 MBit/s connection - and serves other things as well. > > So it takes at least 40 minutes to fetch all the raw files through http > before git starts to do something. An upgrade to 40 MBit/s is planned to > come soon. > > I do not know the details but the git protocol seems to be more clever in > knowing what to transfer first but is less robust for network interruptions. > > This is why we mirror to github. This allows our server to just upload > the changes. > > Anyways thank you very much for your report (since we do not regularily > look on the whole setup as a user does). > Ah. That makes sense. -- Attached is my PGP public key. Primary key fingerprint: B7C7 AD66 D9AF 4348 0238 168E 2C53 D8FA 55D8 9FD9 If you have a PGP key (and a minute to spare) please send it in reply to this email. If you have no idea what PGP is, feel free to ignore all this gobbledegook.
binXKifqF9S6T.bin
Description: PGP Key 0x2C53D8FA55D89FD9.
signature.asc
Description: PGP signature
_______________________________________________ Community mailing list [email protected] http://lists.goldelico.com/mailman/listinfo.cgi/community http://www.tinkerphones.org
