Re: [PATCH] BaseTools/GenFw ARM: don't permit R_ARM_GOT_PREL relocations

[Leif Lindholm <leif.lindholm@linaro.org>]

On Tue, Dec 11, 2018 at 10:37:15AM +0100, Ard Biesheuvel wrote:
> We currently permit R_ARM_GOT_PREL relocations in the ELF32 conversion
> routines, under the assumption that relative relocations are fine as
> long as the section layout is the same between ELF and PE/COFF.
> 
> However, as is the case with any proxy generating relocation, it is
> up to the linker to emit an entry in the GOT table and populate it
> with the correct absolute address, which should also be fixed up at
> PE/COFF load time. Unfortunately, the relocations covering the GOT
> section are not emitted into the static relocation sections processed
> by GenFw, but only in the dynamic relocation section as a R_ARM_RELATIVE
> relocation, and so GenFw fails to emit the correct PE/COFF relocation
> data for GOT entries.
> 
> Since GOT indirection is pointless anyway for PE/COFF modules running
> in UEFI context, let's just drop the references to R_ARM_GOT_PREL from
> GenFw, resulting in a build time failure rather than a runtime failure
> if such relocations do occur.
> 
> Cc: Bob Feng <bob.c.feng@intel.com>
> Cc: Liming Gao <liming.gao@intel.com>
> Cc: Leif Lindholm <leif.lindholm@linaro.org>
> Contributed-under: TianoCore Contribution Agreement 1.1
> Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>

Reviewed-by: Leif Lindholm <leif.lindholm@linaro.org>

Ouch. This sounds like the best move for now. But how do we deal with
builds that actually break?

> ---
> 
> Without this patch, CLANG38 builds of ArmVirtQemuKernel-ARM (in LTO mode)
> succeed, but the resulting binaries are broken. This is due to the fact that
> the PIE linker running in LTO mode insists on emitting GOT based relocations,
> while we don't have the code to fix up the contents of the GOT. This change
> puts it in line with the AARCH64 build of the same platform/toolchains,
> which chokes on GOT based relocations as well. Since the use of the PIE
> linker is a peculiarity of ArmVirtQemuKernel/ArmVirtXen, and the fact that
> it is impossible to prevent the linker from emitting GOT based relocations,
> let's not go out of our way to fix it in the tools, but just drop CLANG38
> support from those platforms.
> 
>  BaseTools/Source/C/GenFw/Elf32Convert.c | 2 --
>  1 file changed, 2 deletions(-)
> 
> diff --git a/BaseTools/Source/C/GenFw/Elf32Convert.c b/BaseTools/Source/C/GenFw/Elf32Convert.c
> index 3d7de6d5c123..23e8065756e6 100644
> --- a/BaseTools/Source/C/GenFw/Elf32Convert.c
> +++ b/BaseTools/Source/C/GenFw/Elf32Convert.c
> @@ -837,7 +837,6 @@ WriteSections32 (
>            case R_ARM_LDC_PC_G0:
>            case R_ARM_LDC_PC_G1:
>            case R_ARM_LDC_PC_G2:
> -          case R_ARM_GOT_PREL:
>            case R_ARM_THM_JUMP11:
>            case R_ARM_THM_JUMP8:
>            case R_ARM_TLS_GD32:
> @@ -964,7 +963,6 @@ WriteRelocations32 (
>              case R_ARM_LDC_PC_G0:
>              case R_ARM_LDC_PC_G1:
>              case R_ARM_LDC_PC_G2:
> -            case R_ARM_GOT_PREL:
>              case R_ARM_THM_JUMP11:
>              case R_ARM_THM_JUMP8:
>              case R_ARM_TLS_GD32:
> -- 
> 2.19.2
>