Show all posts by user

Re: Systemdpocalypse 258 - 3 months ago

I'm running Debian testing (13 in progress), armel support is still going there. As of this morning they're shipping Systemd 257.2-1. I expect they'll move to 258 before releasing 13 barring some major snafu or timing mismatch, so should have most things Debian ships already setup for this brave new world. Meanwhile, I can't see them moving 12 to 258, I think they'll just

by Kurlon - Debian

Re: Hardware Crypto Poking - OpenSSH? - 4 months ago

--- ../../openssh-orig/openssh-9.9p1/sandbox-seccomp-filter.c 2024-09-19 18:20:48.000000000 -0400 +++ sandbox-seccomp-filter.c 2025-01-03 18:20:19.803149104 -0500 @@ -402,6 +402,12 @@ SC_ALLOW_ARG(__NR_socketcall, 0, SYS_SHUTDOWN), SC_DENY(__NR_socketcall, EACCES), #endif + + /* Kurlon testing alfag */ +#ifdef __NR_socket + SC_ALLOW_ARG(__NR_socket, 0, AF_ALG), +#endif + #if d

by Kurlon - Debian

Hardware Crypto Poking - OpenSSH? - 4 months ago

I've started poking at getting Kirkwood hardware crypto playing ball again with OpenSSH. Digging around I found two major showstoppers with the current versions: - Privilege Separation - Digests On Priv Sep, the answer in the past was to disable it in OpenSSH, which seems wrong given how many safeties it provides. Testing with af_alg (It's in kernel, openssl supports it out of th

by Kurlon - Debian

Re: Debian preloading modules excessively? - 4 months ago

If I use 'dep' and don't clear out the excess modules before running update-initramfs, it sees them all and sets them to be loaded in the initramfs. Using 'list' and not providing any delays the excess load in until after rootfs is mounted. It's almost like it's just loading nearly every module built, as I see sisusb in that list even though it's absolutely

by Kurlon - Debian

Debian preloading modules excessively? - 4 months ago

I haven't tested with stable Debian, but on my 13 test setup on a GoFlex Net I'm seeing a massive influx of modules get loaded by... something at startup and I haven't been able to nail down why. [ 0.000000] [ T0] Booting Linux on physical CPU 0x0 [ 0.000000] [ T0] Linux version 6.12.5-kirkwood-tld-1 (root@gfn-test) (gcc (Debian 14.2.0-8) 14.2.0, GNU ld (GNU Binut

by Kurlon - Debian

Re: DisplayLink USB 2.0 FB driver removal - 4 months ago

My current setup I don't use an xorg conf, I let it auto-detect and it correctly identifies the card as 'modesetting'. Once the box is booted, you may need to kick the displaylink device to get it out of 'presetup' mode acting like a mass storage device with an antique Windows installer on it an into video mode. I do this via udev rule: /etc/udev/rules.d/60-displayl

by Kurlon - Displays

Re: DisplayLink USB 2.0 FB driver removal - 4 months ago

So, just went through this battle. I can say, the new DRM version of the driver works well, damage is supported, screen corruption seems minimal. Running Firefox remote to my GoFlex via direct tcp X11 (no SSH tunnel) I was able to do full screen video at 1280x720. Now, on the pain side for some reason the DRM driver seems REALLY dumb about picking outputs and screen modes vs the older FB version

by Kurlon - Displays

Re: Debian on AK Systems Thin client rev 1.3 - 4 months ago

I've awoken from my slumber, fired up my ye olde GoFlex Net and realize, I don't have time to shuffle through getting Gentoo current after a year+ of idling. Saw Bodhi has been keeping the lights on with an updated UBoot, Debian kernel... used those to get on Debian testing, recompiled the kernel to allow DisplayLink support and have gotten my rig back up and running as a full fledged &

by Kurlon - Debian

Re: Kirkwood U-boot - Getting all supported Kirkwoods on-board - 11 years ago

Hah, I *just* started poking at the linux-kirkwood-dt kernel package again today. If there is going to be an effort to get an updated FDT u-boot rolling for as many of the Kirkwood boards as possible, I'm game to help as much as I can.

by Kurlon - uBoot

Re: kernel 3.5 and device-tree support... - 12 years ago

Why would you load to 0x00600000 and then jump to 0x00600200 to kick off the chain loaded uboot? (uboot newb here, sorry if this is obvious/common knowledge)

by Kurlon - Debian

Re: Newer uBoot as workaround to 3.2 kernel problem? - 12 years ago

The Arch GoFlex instructions don't say to skip the uboot check anywhere that I've seen. Could you post a link to where you saw that bit? Davy, if you're respinning the uboot images, would you be up for doing some with Device Tree support enabled for testing?

by Kurlon - uBoot

Re: GoFlex Net: Boot from USB while SATA disks connected - 12 years ago

To use the the label based function you need an initrd. The kernel needs to know where to find root via a device, it cannot scan for labels, UUIDs, etc. By using an initrd you present a temporary rootfs the kernel can attach to, along with enough userland tools to do fancy volume searches based on UUID, labels, etc, and to then REMOUNT root and continue booting. Because Arch ARM doesn't

by Kurlon - uBoot

Re: Linux 3.3.2 kernel for Dockstar, GoFlex and many others - 12 years ago

FYI, I've got 3.5.2 working smoothly now, fixed my (stupid) reboot issue.

by Kurlon - Debian

Device Tree enabled u-boot for kirkwood? - 12 years ago

I've been playing with 3.5 and now 3.6 kernels using Device Tree on my GoFlex. I've been trying to come up with a config that will allow DT kernels to boot using the updated u-boot without config/env changes if possible. So far, I can fire up kernels if I append the .dtb to the kernel and boot with no bootenv. In theory the kernel supports options to append legacy kernel options to t

by Kurlon - uBoot

Re: kernel 3.5 and device-tree support... - 12 years ago

For those not following the arm-kernel list, the problem wasn't with clocking, it was too small of a reserved memory pool at boot for DMA. Add 'coherent_pool=1M' to your boot args and see if that clears things up when you enable both sata ports.

by Kurlon - Debian

Re: kernel 3.5 and device-tree support... - 12 years ago

I've reported the bug upstream, and have been given a suggestion on where to look codewise for the culprit. Buttzy, if you read the arm-kernel archives you'll see patch submissions and how to format them. I can try to bang out a crude template in a bit if you'd like.

by Kurlon - Debian

Re: kernel 3.5 and device-tree support... - 12 years ago

Ok, I forgot to set CONFIG_SERIAL_OF_PLATFORM=y on my 3.6-rc1 test kernel. That with SATA set to 1 port has me booting into 3.6-rc1 cleanly. I'm going to retest with SATA set to two ports next. Edit: 2 ports, same hang I've been chasing...

by Kurlon - Debian

Re: kernel 3.5 and device-tree support... - 12 years ago

My GoFlex Net still won't boot next 8/13 with nr_ports set to 1, no visible change in console output from nr_ports set at 2. Can you pastebin your config so I can compare it against mine to see what's different?

by Kurlon - Debian

Re: kernel 3.5 and device-tree support... - 12 years ago

I haven't tried dropping nr_ports yet, lemmie test that real quick. In the mean time, you're saying you have a GoFlex Net with 256MB RAM?

by Kurlon - Debian

Re: kernel 3.5 and device-tree support... - 12 years ago

I'm also not seeing my serial port binding after the driver loads in that dmesg output, so something is wacky. Not sure where to dig just yet, but I've got some theories.

by Kurlon - Debian

Re: kernel 3.5 and device-tree support... - 12 years ago

Looks like there was a regression somewhere, I'm seeing the same hang on 3.6-rc1 and the current next cut, the last thing I see is: [ 15.283577] VFS: Disk quotas dquot_6.5.2 [ 15.287832] Dquot-cache hash table entries: 1024 (order 0, 4096 bytes) [ 15.296545] jffs2: version 2.2. (NAND) ?© 2001-2006 Red Hat, Inc. [ 15.303

by Kurlon - Debian

Re: kernel 3.5 and device-tree support... - 12 years ago

My bootargs get set as part of the normal boot script, so when I want to test a DT kernel I interrupt the boot before the auto cmds are run.

by Kurlon - Debian

Re: kernel 3.5 and device-tree support... - 12 years ago

The Arch stuff isn't using DT, it's a rehash of the 3.1 patchset updated for 3.5 and polished up. My DT stuff I append the DTB to the end of the zImage, make a uImage off that and boot without any bootargs or arcnumber set. I go by the bootargs in the DTB.

by Kurlon - Debian

Re: kernel 3.5 and device-tree support... - 12 years ago

I finally got 3.5.1 for kirkwoods out the door in Arch, I'm doing some test builds now of 3.6-rc1 and next-08132012 to see what I can sniff out on my GFN. (It's times like this I really wish I had an OpenRD or a good ARMv7 system to do my builds on...)

by Kurlon - Debian

Re: kernel 3.5 and device-tree support... - 12 years ago

Use the append option if your u-boot doesn't let you load the dtb directly. The amount of ram is specified in the DTS file, if you copied the GoFlex you'll only see 128MB.

by Kurlon - Debian

Re: kernel 3.5 and device-tree support... - 12 years ago

I'm just about finished with a 3.5.1 Arch Arm kernel, once that's done I'll bang on 3.6-rc1 on my GFN and see if I can figure out what you guys are running into. FYI, I do use a root delay of 10 seconds on my GFN, might be worth a try if you're bombing due to trying to mount the rootfs too soon.

by Kurlon - Debian

Re: kernel 3.5 and device-tree support... - 12 years ago

Well, it means you're getting well past kernel init and running multiuser (udev is firing off), which you were before as well. As to why it's going pearshaped for udev I'm not certain.

by Kurlon - Debian

Re: kernel 3.5 and device-tree support... - 12 years ago

You would only define i2c if your board has it. I want to say you had an i2c connection defined in your board file, in which case you can remove it from there and add it to the DTS, but I don't believe that's the hangup. SATA is complaining about a completely optional item, I've got to fire my GFN up when I get home to see if it makes a similar complaint.

by Kurlon - Debian

Re: kernel 3.5 and device-tree support... - 12 years ago

I'm not 100% convinced it's the SATA, AFAIK the address is fixed for them on Kirkwood. Do you have two ports?

by Kurlon - Debian

Re: kernel 3.5 and device-tree support... - 12 years ago

When you boot a DT kernel with a .dtb appended the arcNumber shouldn't matter. Can't hurt to try booting with the arcNumber unset though?

by Kurlon - Debian