Re: x86_64 imporatant information

2005-08-20 Thread Yoshinori K. Okuji
On Sunday 21 August 2005 01:25, Ruslan Nikolaev wrote: > I think it's just because we need to use special memory model > "-mcmodel=kernel" that's designed especially for kernel. Does we have > problems when LMA != VMA and using this memory model? I can't find such > information... You have already

x86_64 imporatant information

2005-08-20 Thread Ruslan Nikolaev
In my previous post I forgot to say about the most imporatant reason to use negative addresses.   Take a look on the following information:   - /usr/src/linux/vmlinux: file format elf64-x86-64 Sections:Idx Name  Size  VMA   LMA

Re: [PATCH] x86_64 again (related with my previous topics)

2005-08-20 Thread Ruslan Nikolaev
Hi!   Hmmm... I'm a bit confused. May be I don't undestand something but... Yes it about virtual address. How operating system will have deal with kernel symbols? For example I have a system call "my_system_call". "my_system_call" have a physical address 0x0024 and virtual address 0x802

Re: sparc64 port

2005-08-20 Thread Vincent Pelletier
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Marco Gerards wrote: > Please start sentences with an uppercase letter, same for the other > entries. Oops. > Some files are not GPL'ed (cache.c, for example). The copyright years > are not always correct, like that of ieee1275.h (you wrote it, so i

Re: sparc64 port

2005-08-20 Thread Marco Gerards
Vincent Pelletier <[EMAIL PROTECTED]> writes: Hi Vincent, This is really nice! Thanks a lot. You asked me on IRC if I can test if it still builds on the PPC. I will test that (and let you know on IRC), after that you can commit the patch if you fix the things I mention in my email and Okuji me

Re: sparc64 port

2005-08-20 Thread Yoshinori K. Okuji
Go ahead. I don't see anything wrong any more. Okuji ___ Grub-devel mailing list Grub-devel@gnu.org http://lists.gnu.org/mailman/listinfo/grub-devel

Re: sparc64 port

2005-08-20 Thread Vincent Pelletier
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Yoshinori K. Okuji wrote: > On Saturday 20 August 2005 13:40, Vincent Pelletier wrote: >> * boot/sparc64/ieee1275/cmain.c: New file. > This is wrong. Corrected. >> * kern/sparc64/cache.c: New file. > The contents of this file is strange. W

Re: sparc64 port

2005-08-20 Thread Yoshinori K. Okuji
On Saturday 20 August 2005 13:40, Vincent Pelletier wrote: > * boot/sparc64/ieee1275/cmain.c: New file. This is wrong. I guess you just followed the powerpc port, but it is also wrong. Instead, this file must be put in the kernel. The directory boot is used to organize files for the bootst

sparc64 port

2005-08-20 Thread Vincent Pelletier
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 I fixed the sparc64 port. Current status is : rescue mode works, it can only be netbooted, no module support, no utils (grub-emu, grub-install,...) I separate the Changelog in 2 parts, one for the changes I made that could interfere with the other por

search command

2005-08-20 Thread Yoshinori K. Okuji
I've implement a new command "search" right now. This is similar to "find" in GRUB Legacy, but has more features. I renamed it to "search", because it is significantly different from "find" in GNU. Some supposed usage: # Find a partition whose label is "Boot Partition" and set the result in roo