Re: [PATCH] New object format
Hi, Update: Add support for i386-ieee1275, i386-qemu and i386-coreboot, now the x86 platform is complete, you can build either one of it in Linux/OSX/Windows: Note: grub-mkfont doesn't compile in OSX, so you should add --disable-grub-mkfont, grub-emu could also cause problem in OSX/Windows, you can use --disable-grub-emu to disable it. Here are some information on how to compile them: i386-pc: (Linux/OSX/Windows) ./configure --with-platform=pc make . i386-efi: (Linux/OSX/Windows) ./configure --with-platform=efi make i386-ieee1275 (Linux/OSX/Windows) ./configure --with-platform=ieee1275 make i386-qemu (Linux/OSX/Windows) ./configure --with-platform=qemu make i386-coreboot (Linux/OSX/Windows) ./configure --with-platform=coreboot make x86_64-efi Linux: (needs gcc-multilib on 32-bit system) ./configure --with-platform=efi --target=x86_64 make OSX: ./configure --with-platform=efi --target=x86_64 make Windows: (needs mingw-w64) ./configure --with-platform=efi --target=x86_64-w64-mingw32 make -- Bean ___ Grub-devel mailing list Grub-devel@gnu.org http://lists.gnu.org/mailman/listinfo/grub-devel
Re: Fwd: [PATCH 1/2] Framebuffer split
Hello, both in the previous complete patch and Vladimir's git repository introduce a grub_video_rect_t type. Is there some intention to change the interface to use this structure? I was looking at the the various places that use points and rectangles. It's quite inconsistent and unclear now. There is an anonymous viewport structure inside render target which is the same as the rectangle type. Many coordinates and sizes are signed and many are unsigned leading to numerous casts and/or signed vs unsigned comparison warnings. The blit functions take a rectangle and a point specified as six separate integers which makes it somewhat unclear what is what. The other thing that bothers me is that the video drivers fill in the render target structure during video initialization. In my view the render target structure should not be public. The video driver should fill in the video mode structure and then call a function that fills in render target based on video mode and video data (framebuffer memory). This code is currently replicated needlessly in the vbe and sdl drivers. Bitmaps could be created in the same way. Thanks Michal ___ Grub-devel mailing list Grub-devel@gnu.org http://lists.gnu.org/mailman/listinfo/grub-devel
Re: booting iso image having syslinux
On Sun, 9 Aug 2009 16:55:57 +1000 Peter Cros wrote: > It could be easier to use the partedmagic live iso, pmagic-4.3.iso, > (kernel 2.6.30.1-pmagic), which does loop mount iso with 64 bit > grub.efi loopback on Apple mac (It has gparted). > Thanks but it seems pmagic is much more than a partitioning tool. In that case I can use puppy linux. Though my aim is to collect little live iso tools in usb stick. > > On Sat, Aug 8, 2009 at 9:50 PM, J. Bakshi wrote: > > > On Sat, 8 Aug 2009 12:34:12 +0200 > > "Vladimir 'phcoder' Serbinenko" wrote: > > > > > > I don't know if it is a coincident but your timing to share > > > > these info is directly connected with my problem. I am > > > > presently trying to collect some live iso in my usb stick and > > > > like to call those by grub2. gparted is one of them. I have in > > > > mind to use gparted iso but you have already mentioned that it > > > > can't be. Is there any other work around within grub to cope > > > > with these type of iso ? > > > It's not a question about grub but about initrd being or not being > > > able to loopmount. Ask gparted live guys to add this feature to > > > their scripts > > > > > > > > I'll do it. > > > > > > ___ > > Grub-devel mailing list > > Grub-devel@gnu.org > > http://lists.gnu.org/mailman/listinfo/grub-devel > > > > > ___ Grub-devel mailing list Grub-devel@gnu.org http://lists.gnu.org/mailman/listinfo/grub-devel
