On 15 January 2014 03:16, Roy Franz <roy.franz@xxxxxxxxxx> wrote: > On Tue, Jan 14, 2014 at 5:47 PM, Roy Franz <roy.franz@xxxxxxxxxx> wrote: >> On Tue, Jan 14, 2014 at 1:05 AM, Ard Biesheuvel >> <ard.biesheuvel@xxxxxxxxxx> wrote: >>> On 10 January 2014 17:30, Roy Franz <roy.franz@xxxxxxxxxx> wrote: >>>> This patch adds EFI stub support for the ARM Linux kernel. The EFI stub >>>> operates similarly to the x86 stub: it is a shim between the EFI firmware >>>> and the normal zImage entry point, and sets up the environment that the >>>> zImage is expecting. This includes loading the initrd (optionaly) and >>>> device tree from the system partition based on the kernel command line. >>>> The stub updates the device tree as necessary, adding entries for EFI >>>> runtime services. The PE/COFF "MZ" header at offset 0 results in the >>>> first instruction being an add that corrupts r5, which is not used by >>>> the zImage interface. >>>> >>>> Signed-off-by: Roy Franz <roy.franz@xxxxxxxxxx> >>>> Acked-by: Grant Likely <grant.likely@xxxxxxxxxx> >>>> --- >>> >>> [...] >>> >>>> diff --git a/arch/arm/boot/compressed/efi-header.S b/arch/arm/boot/compressed/efi-header.S >>>> new file mode 100644 >>>> index 0000000..dbb7101 >>>> --- /dev/null >>>> +++ b/arch/arm/boot/compressed/efi-header.S >>>> @@ -0,0 +1,117 @@ >>>> +@ Copyright (C) 2013 Linaro Ltd; <roy.franz@xxxxxxxxxx> >>>> +@ >>>> +@ This file contains the PE/COFF header that is part of the >>>> +@ EFI stub. >>>> +@ >>>> + >>>> + .org 0x3c >>>> + @ >>>> + @ The PE header can be anywhere in the file, but for >>>> + @ simplicity we keep it together with the MSDOS header >>>> + @ The offset to the PE/COFF header needs to be at offset >>>> + @ 0x3C in the MSDOS header. >>>> + @ The only 2 fields of the MSDOS header that are used are this >>>> + @ PE/COFF offset, and the "MZ" bytes at offset 0x0. >>>> + @ >>>> + .long pe_header @ Offset to the PE header. >>>> + >>>> + .align 3 Btw you also have a whitespace error here. >>>> +pe_header: >>>> + .ascii "PE" >>>> + .short 0 >>>> + >>>> +coff_header: >>>> + .short 0x01c2 @ ARM or Thumb >>> >>> Could you explain why you are using 0x1c2 (Thumb) here and not 0x1c0 (ARM) ? >>> >>> Cheers, >>> Ard. >> >> Nope. It should be 0x1c0. >> >> Roy > > OK, now I resolved the nagging feeling that I had already fixed > this... Right now, the EDK2 UEFI implementation > requires the machine type for ARM to be "0x1c2". I don't think that > this is correct, but correcting this in EDK2 slipped through > the cracks, but is now back on my todo list. > > I think that for now we should leave this as 0x1c2 so that the > unpatched EDK2 builds will boot it, and some time after EDK2 is > updated > this can be changed. I'll work on a patch for EDK2 and get the > discussion going on that list to resolve this in EDK2. > OK. I have updated sbsigntool (Linaro's version) so it supports either, but I agree that using the ARM constant is the correct way. -- Ard. -- To unsubscribe from this list: send the line "unsubscribe linux-efi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
