Re: [PATCH v3 0/8] x86/boot: Rework PE header generation

[Ingo Molnar]

* Ard Biesheuvel <ardb@xxxxxxxxxx> wrote:

> From: Ard Biesheuvel <ardb@xxxxxxxxxx>
> 
> Now that the EFI stub boot flow no longer relies on memory that is
> executable and writable at the same time, we can reorganize the PE/COFF
> view of the kernel image and expose the decompressor binary's code and
> r/o data as a .text section and data/bss as a .data section, using 4k
> alignment and limited permissions.
> 
> Doing so is necessary for compatibility with hardening measures that are
> being rolled out on x86 PCs built to run Windows (i.e., the majority of
> them). The EFI boot environment that the Linux EFI stub executes in is
> especially sensitive to safety issues, given that a vulnerability in the
> loader of one OS can be abused to attack another.
> 
> In true x86 fashion, this is a lot more complicated than on other
> architectures, which have implemented this code/data split with 4k
> alignment from the beginning. The complicating factor here is that the
> boot image consists of two different parts, which are stitched together
> and fixed up using a special build tool.
> 
> After this series is applied, the only remaining task performed by the
> build tool is generating the CRC-32. Even though this checksum is
> usually wrong (given that distro kernels are signed for secure boot in a
> way that corrupts the CRC), this feature is retained as we cannot be
> sure that nobody is relying on this.
> 
> This supersedes the work proposed by Evgeniy last year, which did a
> major rewrite of the build tool in order to clean it up, before updating
> it to generate the new 4k aligned image layout. As this series proves,
> the build tool is mostly unnecessary, and we have too many of those
> already.
> 
> Changes since v2:
> - rebase onto tip/master
> - drop patches that have been picked up already
> - fix issue in the linker script that resulted in a bogus setup_size in
>   some cases when using ld.bfd
> - fix comment capitalization
> 
> Changes since v1:
> - drop patch that removed the CRC and the build tool
> - do not use fixed setup_size but derive it in the setup.ld linker
>   script
> - reorganize the PE header so the .compat section only covers its
>   payload and the padding that follows it
> - add hpa's ack to patch #4
> 
> Cc: Evgeniy Baskov <baskov@xxxxxxxxx>
> Cc: Borislav Petkov <bp@xxxxxxxxx>
> Cc: Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>
> Cc: Ingo Molnar <mingo@xxxxxxxxxx>
> Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> Cc: Peter Jones <pjones@xxxxxxxxxx>
> Cc: Matthew Garrett <mjg59@xxxxxxxxxxxxx>
> Cc: Gerd Hoffmann <kraxel@xxxxxxxxxx>
> Cc: Kees Cook <keescook@xxxxxxxxxxxx>
> Cc: "H. Peter Anvin" <hpa@xxxxxxxxx>
> 
> 
> Ard Biesheuvel (8):
>   x86/boot: Grab kernel_info offset from zoffset header directly
>   x86/boot: Set EFI handover offset directly in header asm
>   x86/boot: Define setup size in linker script
>   x86/boot: Derive file size from _edata symbol
>   x86/boot: Construct PE/COFF .text section from assembler
>   x86/boot: Drop PE/COFF .reloc section
>   x86/boot: Split off PE/COFF .data section
>   x86/boot: Increase section and file alignment to 4k/512
> 
>  arch/x86/boot/Makefile                 |   2 +-
>  arch/x86/boot/compressed/vmlinux.lds.S |   5 +-
>  arch/x86/boot/header.S                 | 146 +++++++------
>  arch/x86/boot/setup.ld                 |   7 +-
>  arch/x86/boot/tools/build.c            | 223 +-------------------
>  5 files changed, 97 insertions(+), 286 deletions(-)

Applied to tip:x86/boot, thanks!

	Ingo