Re: [PATCH V3 RFC 00/16] EFI stub for ARM

From: Roy Franz
Date: Tue Aug 13 2013 - 13:58:24 EST

On Fri, Aug 9, 2013 at 4:26 PM, Roy Franz <roy.franz@xxxxxxxxxx> wrote:
> This patch series adds EFI stub support for the ARM architecture.
> Some code that was previously only used by x86/x86_64 is now shared
> and has been made more general. The stub for ARM is implemented in
> a similar manner to x86 in that it is a shim layer between EFI and
> the normal zImage/bzImage boot process, and that an image with the
> stub configured is bootable as both a zImage and EFI application.
> This patch now (new for v3) series depends on the
> "arm: Add [U]EFI runtime services support" patches by leif.lindholm@xxxxxxxxxxx
> The Kconfig option now depends on the "CONFIG_EFI" option that his series
> adds, and this option will ensure a little endian configuration. Also, the
> EFI support is used to handle the EFI memory map the stub passes to the kernel.
> There are some minor changes to be coordinated with the EFI runtime services
> patch series, so I have put this back to RFC status. These changes should be
> minor and relate to final device tree bindings.
> Changes since v2:
> * EFI bugfix "correct call to free_pages" that patch series
> depends on now in mainline
> * remove "-fno-stack-protector" from decompressor Makefile. The current code doesn't
> trigger the stack protection, so the decompressor now compiles with it still
> on. Note that there has never been any stack protection in the decompressor
> since the stack usage doesn't trigger the heuristic in GCC, so right now
> the "-fno-stack-protector" is a noop.
> * Changed EFI command line handling to not have a fixed limit.
> * Change FDT memory allocation to retry with a larger allocation if
> first educated guess is inadequate.
> * Correctly set 'SizeOfCode' in PE/COFF header.
> * Reviewed ".setup" section that is in x86 PE/COFF header. This is used for x86
> to account for code that is not in the .text section. We don't need this
> for ARM, as all of our code is in the .text section, or in the PE/COFF header
> itself.
> * Moved EFI_STUB_ERROR #define to header file to share between stub C and ASM.
> * Variety of cleanups and fixes in head.S.
> * Changed update_fdt_and_exit_boot() to just update the device tree, and
> renamed appropriately. Memory allocations moved out of this function as
> well, which enables the retries if the initial FDT size is too small.
> Note that in order to do the retried allocations, the original FDT and
> command line memory regions are left allocated. This is OK since the kernel
> has the memory map and will free these allocations along with the initrd
> and new fdt allocations.
> * Added prefix to all prints, reduced number of prints, and reviewed all
> messages.
> * Change mixed usage of dtb/fdt to all be fdt or "device tree" in efi-stub.c
> * remove unnecessary zimage_size variable from relocate_kernel()
> * correct return types on EFI functions - should be efi_status_t, not int.
> Changes since V1:
> * Updated head.S based on feedback from Dave Martin. ARM/THUMB
> switches now much cleaner.
> * Broke up changes to x86 and common code into more patches.
> 10 more patches in this series.
> Roy Franz (16):
> Move common EFI stub code from x86 arch code to common location
> Add system pointer argument to shared EFI stub related functions
> so they no longer use global system table pointer as they did
> when part of eboot.c. This code is now shared, so using a
> global variable as part of the interface is not that nice.
> Also, by avoiding any global variables in the ARM EFI stub,
> this allows the code to be position independent without
> requiring GOT fixups.
> Rename memory allocation/free functions
> Add minimum address parameter to efi_low_alloc()
> rename __get_map() to efi_get_memory_map(), add parameter to
> optionally return mmap key. The mmap key is required to exit
> EFI boot services, and allows efi_get_memory_map() to be used
> for getting final memory map.
> Enforce minimum alignment of 1 page on allocations. The
> efi_high_alloc() and efi_low_alloc() functions use the
> EFI_ALLOCATE_ADDRESS option to the EFI function
> allocate_pages(), which requires a minimum of page alignment,
> and rejects all other requests.
> Allow efi_free() to be called with size of 0, and do nothing in that
> case.
> Generalize handle_ramdisks() and rename to handle_cmdline_files().
> Renames in handle_cmdline_files() to complete generalization.
> Move EFI_READ_CHUNK_SIZE define to shared location.
> Add proper definitions for some EFI function pointers.
> Fix types in EFI calls to match EFI function definitions.
> resolve warnings found on ARM compile
> Add strstr to compressed string.c for ARM.
> Add EFI stub for ARM
> Add config EFI_STUB for ARM to Kconfig
> arch/arm/Kconfig | 11 +
> arch/arm/boot/compressed/Makefile | 15 +-
> arch/arm/boot/compressed/efi-header.S | 111 +++++++
> arch/arm/boot/compressed/efi-stub.c | 448 ++++++++++++++++++++++++++++
> arch/arm/boot/compressed/efi-stub.h | 5 +
> arch/arm/boot/compressed/head.S | 90 +++++-
> arch/arm/boot/compressed/string.c | 21 ++
> arch/x86/boot/compressed/eboot.c | 490 ++-----------------------------
> arch/x86/boot/compressed/eboot.h | 9 -
> drivers/firmware/efi/efi-stub-helper.c | 505 ++++++++++++++++++++++++++++++++
> include/linux/efi.h | 51 +++-
> 11 files changed, 1263 insertions(+), 493 deletions(-)
> create mode 100644 arch/arm/boot/compressed/efi-header.S
> create mode 100644 arch/arm/boot/compressed/efi-stub.c
> create mode 100644 arch/arm/boot/compressed/efi-stub.h
> create mode 100644 drivers/firmware/efi/efi-stub-helper.c
> --

Hi Matt,

Do you have any more feedback on the X86 and common code (patches
1-13) that needs to be addressed? Mark Salter has a working ARM64 EFI
stub implemented based on these patches, so the common code has now
been tested with another architecture, and he has acked these patches.
If the current patches are OK, can this be queued for 3.12? (and into
linux-next, if appropriate)
The ARM portion may take a little longer based on the EFI runtime
services patch dependencies, but getting the common code merged would
allow the ARM64 EFI stub work to go forward independently.

I can resend patches 1-13 as a new series of x86/common only changes
if you would like.

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at