Hi all, We're using a PXE / network install system for the SunFire x4100M2. We had a kernel that worked fine for the x4100 and x2100/x2100M2, but this new platform does something funny: after selecting the kernel at the PXElinux prompt, I don't see any kernel messages on the serial console, until the kernel has finished initialising. They are appearing on the VGA console, when I use console=tty0 console=ttyS0. With console=ttyS0 only, they appear nowhere. There is a BIOS setting called "Redirect" that has one option that looks like a smoking gun: "Redirection After BIOS POST". This has three settings: Disable: Turns off the redirection after POST Boot Loader: Redirection is active during POST and during Boot Loader. Always: Redirection is always active (Some OSs may not work if set to Always) It seems that "always" is behaving just like "Boot Loader" says it should - as soon as it would be saying "Loading kernel..." then I see no messages at all. Except perhaps a suprious character like "Á". The "Boot Loader" option then works like "disable" - as soon as PXElinux (or GRUB) loads, the console switches off and I see the "Á". What's got me confounded is that it suddenly wiki comes right after root gets mounted - be that an initrd image, or anything else. What is it that changes then, and how can I make that happen earlier? I found the console.txt file, and it seems that while boot is in progress the "system driver" is active vs the "modular driver". However, I'm not sure why there is a change of driver at that point; I didn't ask for it. I'm coming to the conclusion that the serial console redirection in the BIOS is broken somehow, anyone have any ideas on tracking this down further? -- Sam Vilain, Chief Yak Shaver, Catalyst IT (NZ) Ltd. phone: +64 4 499 2267 PGP ID: 0x66B25843 -
| Greg Kroah-Hartman | [PATCH 006/196] Chinese: add translation of oops-tracing.txt |
| Linus Torvalds | Linux 2.6.21-rc1 |
| david | Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3 |
| Vladislav Bolkhovitin | Re: Integration of SCST in the mainstream Linux kernel |
| Alexey Dobriyan | Re: [GIT]: Networking |
| Gerrit Renker | [PATCH 27/37] dccp: Integration of dynamic feature activation - part 2 (server side) |
| Jarek Poplawski | Re: [PATCH] pkt_sched: Destroy gen estimators under rtnl_lock(). |
| Evgeniy Polyakov | Re: [BUG] New Kernel Bugs |
git: | |
