After reading an article describing the High-priority list for the FSF I checked out our listing.
And found the following.
""" The way to solve the problem is to run a free BIOS. And our community has developed free BIOSes--for instance, LinuxBIOS, which is a modified version of the kernel, Linux. However, """
Seems to me this is not really true. LinuxBIOS is _vastly_ different than the Linux Kernel. We ripped some PCI code, printk code and a few other bits but I can't see us being a "modified version of the kernel..."
We can load a kernel and our ultimate goal is to use a kernel as a super-bootloader but not to be a kernel.
Anyone else think it needs changing?
-- Richard A. Smith
On Wed, May 03, 2006 at 12:07:47AM -0500, Richard Smith wrote:
Anyone else think it needs changing?
Yes. Who do we ask? Can Ward make the change?
I suggest "LinuxBIOS, that can boot several free operating systems on various hardware today, and hopes to utilize the Linux kernel as main bootloader in the future."
//Peter
Not future, with LinuxBIOS+kernel+kexec..we already can use kernel as bootloader...
current we only put some effects to provide tiny kernel and small ramdisk...
fastboot seems to provide the ramdisk...
YH
On 5/2/06, Peter Stuge stuge-linuxbios@cdy.org wrote:
On Wed, May 03, 2006 at 12:07:47AM -0500, Richard Smith wrote:
Anyone else think it needs changing?
Yes. Who do we ask? Can Ward make the change?
I suggest "LinuxBIOS, that can boot several free operating systems on various hardware today, and hopes to utilize the Linux kernel as main bootloader in the future."
//Peter
-- linuxbios mailing list linuxbios@linuxbios.org http://www.openbios.org/mailman/listinfo/linuxbios
On Wed, May 03, 2006 at 12:18:17AM -0700, yhlu wrote:
Not future, with LinuxBIOS+kernel+kexec..we already can use kernel as bootloader...
I agree, it can be done today, but..
current we only put some effects to provide tiny kernel and small ramdisk...
..because of these issues it isn't the norm.
Has anyone even gotten a LinuxBIOS with Linux payload into 512KB?
fastboot seems to provide the ramdisk...
Do you know if it's an initramfs? I haven't gotten around to experimenting with that yet, but it looks nice.
//Peter
Peter Stuge wrote:
Has anyone even gotten a LinuxBIOS with Linux payload into 512KB?
Oh yea. In 2000, it was very common to jam a 2.2 kernel into a 512kb flash part ... very easy. Etherboot etc. came much later, and were ONLY put in because vendors moved to 256KB flash parts, which killed us.
in fact, back then, there was an uncompressor in linuxbios itself, much like what stefan just added to V2. It was later removed -- long story. I was glad to see it return.
ron
On Wed, May 03, 2006 at 08:29:14AM +0200, Peter Stuge wrote:
On Wed, May 03, 2006 at 12:07:47AM -0500, Richard Smith wrote:
Anyone else think it needs changing?
Yes. Who do we ask? Can Ward make the change?
Yes; as soon as there is consensus on a better wording?
Ward.
Richard Smith wrote:
Anyone else think it needs changing?
The original goal of linuxbios was that linux would be the bios.
sadly, linux got big and flash got small.
That said, on OLPC, we're going to try again. In fact I'm gearing up to try putting some gcc code into linux and loading linux direct from romcc-based code. I think it might work.
Who knows what's true :-)
ron
On Wed, May 03, 2006 at 09:45:28AM -0600, Ronald G Minnich wrote:
sadly, linux got big and flash got small.
Linux can still be compiled down to under a meg, especially if many of the post-boot things are compiled as modules.
Arc Riley wrote:
On Wed, May 03, 2006 at 09:45:28AM -0600, Ronald G Minnich wrote:
sadly, linux got big and flash got small.
Linux can still be compiled down to under a meg, especially if many of the post-boot things are compiled as modules.
yeah, but it needs to be under (512-32)K. Plus, you need to stuff an initrd in there for basic utilities.
That's been harder.
ron
On 5/3/06, Ronald G Minnich rminnich@lanl.gov wrote:
That said, on OLPC, we're going to try again. In fact I'm gearing up to try putting some gcc code into linux and loading linux direct from romcc-based code. I think it might work.
So that may be true in the (near) future but its not true now and may not ever be true for a lot of motherboards. Perhaps then just a more detailed description that covers both where it is and the ultimate goal?
-- Richard A. Smith
Richard Smith wrote:
On 5/3/06, Ronald G Minnich rminnich@lanl.gov wrote:
That said, on OLPC, we're going to try again. In fact I'm gearing up to try putting some gcc code into linux and loading linux direct from romcc-based code. I think it might work.
So that may be true in the (near) future but its not true now and may not ever be true for a lot of motherboards.
maybe. But that can be said in general about linuxbios :-)
Perhaps then just a more detailed description that covers both where it is and the ultimate goal?
it's a phantasm in my brain at present.
ron
-
Arc Riley -
Peter Stuge -
Richard Smith -
Ronald G Minnich -
Ward Vandewege -
yhlu