Re: EFI-dualbooting OSX and Linux on iMac with T7400-CPU

Eeri Kask wrote: bibo,mao wrote: >> 2.6.18.5. > There exists one bug in Linux kernel only EFI bios relative at > http://marc.theaimsgroup.com/?l=linux-kernel&m=116157536316034&w=2 <http://marc.theaimsgroup.com/?l=linux-kernel&m=116157536316034&w=2>

Re: EFI-dualbooting OSX and Linux on iMac with T7400-CPU

Eeri Kask wrote: bibo,mao wrote: >> menuentry "MacOSX" { >> set root=(hd0,2) >> chainloader /System/Library/CoreServices/boot.efi >> } >> >> menuentry "GNU/Linux" { >> set root=(hd0,5) >> linux /boot/vmlinuz root

Re: EFI-dualbooting OSX and Linux on iMac with T7400-CPU

Eeri Kask wrote: Yoshinori K. Okuji wrote: >> (line 2-2) >> syntax error >> Incorrect command >> ... >> (line 12-12) >> Press any key to continue... > > Hmm.. I think you need to put the open braces in the same line as "menuentry" > commands. Thank you for the hint! :-) So now my grub

grub EFI signature should be little endian

hi, EFI bootloader signature is "EFIL" string, x86 machine is little-endian. This patch changes grub efi booloader sigature as little-endian. thanks bibo,mao --- grub2.org/include/grub/i386/linux.h 2006-12-06 17:51:19.0 +0800 +++ grub2/include/grub/i386/linux.h 2006-1

Re: multiboot2: variable data size

Johan Rydberg wrote: "bibo,mao" <[EMAIL PROTECTED]> writes: > If kernel image is bzImage, x64 efi bootloader need switch to 32 bit > protect mode(or real mode) from 64 bit long mode, and if kernel > image is gzipped/plain format, efi bootloader can directly jump

Re: multiboot2: variable data size

Yoshinori K. Okuji wrote: On Tuesday 28 November 2006 13:46, bibo,mao wrote: > yes, x84_64EFI starts with 64-bit long mode and page enabled(virtual > address equals physical address) if it is x86_64 efi bios, it is defined in > section 2.3.4 of UEFI Specification Version 2.0. >

Re: multiboot2: variable data size

ication Version 2.0. If kernel image is bzImage, x64 efi bootloader need switch to 32 bit protect mode(or real mode) from 64 bit long mode, and if kernel image is gzipped/plain format, efi bootloader can directly jump to 64-bit kernel entry address without mode switch. thanks bibo,mao

Re: [PATCH 3/3] grub EFI disk device enumberating

least, it is not tested. I tested on my EFI IA32 bios, it works for me, your method is better than me, I incorporated your function in my second patch. thanks bibo,mao In gnufi I have a device_path_iterate function that could be used for these kind of things. Maybe we should bring it in to GRUB2.

Re: [patch 2/3] grub efi initrd image memory allocation 4K alignment

This patch is sent for second time. thanks bibo,mao diff -Nrup grub2.org/loader/i386/efi/linux.c grub2/loader/i386/efi/linux.c --- grub2.org/loader/i386/efi/linux.c 2006-10-25 12:15:25.0 +0800 +++ grub2/loader/i386/efi/linux.c 2006-10-25 11:40:24.0 +0800 @@ -587,7 +587,6

Re: [PATCH 1/3] grub efi memory map patch

hi johan, Thanks for review my patch, I reply in lines. Johan Rydberg wrote: "bibo,mao" <[EMAIL PROTECTED]> writes: > This patch moves find_mmap_size from i386/efi/linux.c to > kern/efi/mm.c, and renamed as grub_efi_find_mmap_size, so that > other arch on

[PATCH 3/3] grub EFI disk device enumberating

find efi disk when grub.efi is executed on logical partition. Previously I posted this patch, now I repost again. Any suggestion is welcome. thanks bibo,mao diff -Nruap grub2.org/disk/efi/efidisk.c grub2/disk/efi/efidisk.c --- grub2.org/disk/efi/efidisk.c2006-10-24 13:23:42.0

[patch 2/3] grub efi initrd image memory allocation 4K alignment

Hi, On grub efi platform, initrd image memory allocation start address must be 4K alignment, else it will fail to alloc memory for initrd image. thanks bibo,mao diff -Nruap grub2.org/loader/i386/efi/linux.c grub2/loader/i386/efi/linux.c --- grub2.org/loader/i386/efi/linux.c 2006-10-24 13

[PATCH 1/3] grub efi memory map patch

EFI memory map table. any suggestion is welcome. thanks bibo,mao diff -Nruap grub2.org/include/grub/efi/api.h grub2/include/grub/efi/api.h --- grub2.org/include/grub/efi/api.h2006-10-24 13:05:48.0 +0800 +++ grub2/include/grub/efi/api.h2006-10-23 14:35:12.0 +0800

Re: Grub for ia64

It's already here! The binutils issue is mainly for ia64. For i386, grub-mkimage.c works. Tristan. On i386 objcopy in binutils can convert ELF format into PE32, but application need dynamically relocate itself so that PE32 image can be loaded to any place to execute. For grub-mkimage.c ap

Re: Grub for ia64

can fit for all architectures. why write different grub_efi_allocate_pages/grub_efi_free_boot_pages, we can modify these two functions to fit for all architecture EFI platform. thanks bibo,mao [EMAIL PROTECTED] wrote: Hi this is a port of grub2 to ia64. ia64 systems (itanium) are EFI based

grub2 64bit system compatible

Hi, The following is grub2 64-bit system compatible patch, in 64-bit system unsigned long is defined as 64-bit, but not 32-bit. This patch modify this point. Thanks bibo,mao --- grub2/fs/fat.c 2006-06-04 23:56:54.0 +0800 +++ grub2.new/fs/fat.c 2006-09-07 16:49:50.0 +0800

Re: EFI disk probing problem

gards blk5 as one block device. Actually it is only one extended partition. My patch discards all partition device(GRUB_EFI_MEDIA_DEVICE_PATH_TYPE). thanks bibo,mao blk1:Floppy - Alias (null) Acpi(PNP0A03,0)/Pci(1F|0)/Acpi(PNP0604,0) blkA:BlockDevice - Alias (null)

EFI disk probing problem

grub mailing list, but there is no response, I do not know why. thanks bibo,mao --- grub-1.94.org/disk/efi/efidisk.c2006-05-01 05:09:37.0 +0800 +++ grub-1.94/disk/efi/efidisk.c2006-07-26 13:00:09.0 +0800 @@ -87,6 +87,51 @@ find_last_device_path (const grub_efi_de ret

patch grub EFI setjmp in IA32

suggestion is welcome. Thanks bibo,mao --- grub-1.94.org/normal/i386/setjmp.S 2004-04-04 21:46:03.0 +0800 +++ grub-1.94/normal/i386/setjmp.S 2006-08-03 11:00:02.0 +0800 @@ -27,11 +27,13 @@ * int grub_setjmp (grub_jmp_buf env) */ FUNCTION(grub_setjmp) + popl