[llvm] r221639 - [yaml2obj] Support AArch64 relocations.
Daniel Stewart
stewartd at codeaurora.org
Tue Nov 11 08:38:12 PST 2014
It does seem like there are too many places the relocations are defined, and
I'd like to see it consolidated. This particular patch just followed the
pattern that was already there. I would think Simon and Shankar would be
able to give a better answer as to why this is defined as it is.
Daniel
-----Original Message-----
From: llvm-commits-bounces at cs.uiuc.edu
[mailto:llvm-commits-bounces at cs.uiuc.edu] On Behalf Of Hal Finkel
Sent: Tuesday, November 11, 2014 1:43 AM
To: Chad Rosier
Cc: llvm-commits at cs.uiuc.edu
Subject: Re: [llvm] r221639 - [yaml2obj] Support AArch64 relocations.
----- Original Message -----
> From: "Chad Rosier" <mcrosier at codeaurora.org>
> To: llvm-commits at cs.uiuc.edu
> Sent: Monday, November 10, 2014 5:02:04 PM
> Subject: [llvm] r221639 - [yaml2obj] Support AArch64 relocations.
>
> Author: mcrosier
> Date: Mon Nov 10 17:02:03 2014
> New Revision: 221639
>
> URL: http://llvm.org/viewvc/llvm-project?rev=221639&view=rev
> Log:
> [yaml2obj] Support AArch64 relocations.
>
> Patch by Daniel Stewart <stewartd at codeaurora.org>!
> Phabricator Revision: http://reviews.llvm.org/D6192
>
> Added:
> llvm/trunk/test/Object/AArch64/
> llvm/trunk/test/Object/AArch64/yaml2obj-elf-aarch64-rel.yaml
> Modified:
> llvm/trunk/lib/Object/ELFYAML.cpp
>
> Modified: llvm/trunk/lib/Object/ELFYAML.cpp
> URL:
>
http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Object/ELFYAML.cpp?rev=22
1639&r1=221638&r2=221639&view=diff
>
============================================================================
==
> --- llvm/trunk/lib/Object/ELFYAML.cpp (original)
> +++ llvm/trunk/lib/Object/ELFYAML.cpp Mon Nov 10 17:02:03 2014
> @@ -672,6 +672,92 @@ void ScalarEnumerationTraits<ELFYAML::EL
> ECase(R_386_IRELATIVE)
> ECase(R_386_NUM)
> break;
Is this function supposed to be synchronized with lib/Object/ELF.cpp (which
has one function: getELFRelocationTypeName). Keeping these in synch manually
seems error-prone (and somewhat silly). Can we refactor this so that we
don't have two essentially-identical lists of relocation names in the two
files?
Thanks again,
Hal
> + case ELF::EM_AARCH64:
> + ECase(R_AARCH64_NONE)
> + ECase(R_AARCH64_ABS64)
> + ECase(R_AARCH64_ABS32)
> + ECase(R_AARCH64_ABS16)
> + ECase(R_AARCH64_PREL64)
> + ECase(R_AARCH64_PREL32)
> + ECase(R_AARCH64_PREL16)
> + ECase(R_AARCH64_MOVW_UABS_G0)
> + ECase(R_AARCH64_MOVW_UABS_G0_NC)
> + ECase(R_AARCH64_MOVW_UABS_G1)
> + ECase(R_AARCH64_MOVW_UABS_G1_NC)
> + ECase(R_AARCH64_MOVW_UABS_G2)
> + ECase(R_AARCH64_MOVW_UABS_G2_NC)
> + ECase(R_AARCH64_MOVW_UABS_G3)
> + ECase(R_AARCH64_MOVW_SABS_G0)
> + ECase(R_AARCH64_MOVW_SABS_G1)
> + ECase(R_AARCH64_MOVW_SABS_G2)
> + ECase(R_AARCH64_LD_PREL_LO19)
> + ECase(R_AARCH64_ADR_PREL_LO21)
> + ECase(R_AARCH64_ADR_PREL_PG_HI21)
> + ECase(R_AARCH64_ADD_ABS_LO12_NC)
> + ECase(R_AARCH64_LDST8_ABS_LO12_NC)
> + ECase(R_AARCH64_TSTBR14)
> + ECase(R_AARCH64_CONDBR19)
> + ECase(R_AARCH64_JUMP26)
> + ECase(R_AARCH64_CALL26)
> + ECase(R_AARCH64_LDST16_ABS_LO12_NC)
> + ECase(R_AARCH64_LDST32_ABS_LO12_NC)
> + ECase(R_AARCH64_LDST64_ABS_LO12_NC)
> + ECase(R_AARCH64_LDST128_ABS_LO12_NC)
> + ECase(R_AARCH64_GOTREL64)
> + ECase(R_AARCH64_GOTREL32)
> + ECase(R_AARCH64_ADR_GOT_PAGE)
> + ECase(R_AARCH64_LD64_GOT_LO12_NC)
> + ECase(R_AARCH64_TLSLD_MOVW_DTPREL_G2)
> + ECase(R_AARCH64_TLSLD_MOVW_DTPREL_G1)
> + ECase(R_AARCH64_TLSLD_MOVW_DTPREL_G1_NC)
> + ECase(R_AARCH64_TLSLD_MOVW_DTPREL_G0)
> + ECase(R_AARCH64_TLSLD_MOVW_DTPREL_G0_NC)
> + ECase(R_AARCH64_TLSLD_ADD_DTPREL_HI12)
> + ECase(R_AARCH64_TLSLD_ADD_DTPREL_LO12)
> + ECase(R_AARCH64_TLSLD_ADD_DTPREL_LO12_NC)
> + ECase(R_AARCH64_TLSLD_LDST8_DTPREL_LO12)
> + ECase(R_AARCH64_TLSLD_LDST8_DTPREL_LO12_NC)
> + ECase(R_AARCH64_TLSLD_LDST16_DTPREL_LO12)
> + ECase(R_AARCH64_TLSLD_LDST16_DTPREL_LO12_NC)
> + ECase(R_AARCH64_TLSLD_LDST32_DTPREL_LO12)
> + ECase(R_AARCH64_TLSLD_LDST32_DTPREL_LO12_NC)
> + ECase(R_AARCH64_TLSLD_LDST64_DTPREL_LO12)
> + ECase(R_AARCH64_TLSLD_LDST64_DTPREL_LO12_NC)
> + ECase(R_AARCH64_TLSIE_MOVW_GOTTPREL_G1)
> + ECase(R_AARCH64_TLSIE_MOVW_GOTTPREL_G0_NC)
> + ECase(R_AARCH64_TLSIE_ADR_GOTTPREL_PAGE21)
> + ECase(R_AARCH64_TLSIE_LD64_GOTTPREL_LO12_NC)
> + ECase(R_AARCH64_TLSIE_LD_GOTTPREL_PREL19)
> + ECase(R_AARCH64_TLSLE_MOVW_TPREL_G2)
> + ECase(R_AARCH64_TLSLE_MOVW_TPREL_G1)
> + ECase(R_AARCH64_TLSLE_MOVW_TPREL_G1_NC)
> + ECase(R_AARCH64_TLSLE_MOVW_TPREL_G0)
> + ECase(R_AARCH64_TLSLE_MOVW_TPREL_G0_NC)
> + ECase(R_AARCH64_TLSLE_ADD_TPREL_HI12)
> + ECase(R_AARCH64_TLSLE_ADD_TPREL_LO12)
> + ECase(R_AARCH64_TLSLE_ADD_TPREL_LO12_NC)
> + ECase(R_AARCH64_TLSLE_LDST8_TPREL_LO12)
> + ECase(R_AARCH64_TLSLE_LDST8_TPREL_LO12_NC)
> + ECase(R_AARCH64_TLSLE_LDST16_TPREL_LO12)
> + ECase(R_AARCH64_TLSLE_LDST16_TPREL_LO12_NC)
> + ECase(R_AARCH64_TLSLE_LDST32_TPREL_LO12)
> + ECase(R_AARCH64_TLSLE_LDST32_TPREL_LO12_NC)
> + ECase(R_AARCH64_TLSLE_LDST64_TPREL_LO12)
> + ECase(R_AARCH64_TLSLE_LDST64_TPREL_LO12_NC)
> + ECase(R_AARCH64_TLSDESC_ADR_PAGE)
> + ECase(R_AARCH64_TLSDESC_LD64_LO12_NC)
> + ECase(R_AARCH64_TLSDESC_ADD_LO12_NC)
> + ECase(R_AARCH64_TLSDESC_CALL)
> + ECase(R_AARCH64_COPY)
> + ECase(R_AARCH64_GLOB_DAT)
> + ECase(R_AARCH64_JUMP_SLOT)
> + ECase(R_AARCH64_RELATIVE)
> + ECase(R_AARCH64_TLS_DTPREL64)
> + ECase(R_AARCH64_TLS_DTPMOD64)
> + ECase(R_AARCH64_TLS_TPREL64)
> + ECase(R_AARCH64_TLSDESC)
> + ECase(R_AARCH64_IRELATIVE)
> + break;
> default:
> llvm_unreachable("Unsupported architecture");
> }
>
> Added: llvm/trunk/test/Object/AArch64/yaml2obj-elf-aarch64-rel.yaml
> URL:
>
http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Object/AArch64/yaml2obj-
elf-aarch64-rel.yaml?rev=221639&view=auto
>
============================================================================
==
> --- llvm/trunk/test/Object/AArch64/yaml2obj-elf-aarch64-rel.yaml
> (added)
> +++ llvm/trunk/test/Object/AArch64/yaml2obj-elf-aarch64-rel.yaml Mon
> Nov 10 17:02:03 2014
> @@ -0,0 +1,47 @@
> +# RUN: yaml2obj -format=elf %s > %t
> +# RUN: obj2yaml %t | FileCheck %s
> +
> +# CHECK: - Name: .rela.text
> +# CHECK-NEXT: Type: SHT_RELA
> +# CHECK-NEXT: Link: .symtab
> +# CHECK-NEXT: AddressAlign: 0x0000000000000008
> +# CHECK-NEXT: Info: .text
> +# CHECK-NEXT: Relocations:
> +# CHECK-NEXT: - Offset: 0x0000000000000000
> +# CHECK-NEXT: Symbol: main
> +# CHECK-NEXT: Type: R_AARCH64_ABS64
> +# CHECK-NEXT: Addend: 0
> +
> +FileHeader:
> + Class: ELFCLASS64
> + Data: ELFDATA2LSB
> + Type: ET_REL
> + Machine: EM_AARCH64
> +Sections:
> + - Type: SHT_PROGBITS
> + Name: .text
> + Flags: [ SHF_ALLOC, SHF_EXECINSTR ]
> + AddressAlign: 0x04
> + Content: 0000000000000000
> + - Type: SHT_RELA
> + Name: .rela.text
> + Link: .symtab
> + Info: .text
> + AddressAlign: 0x08
> + Relocations:
> + - Offset: 0
> + Symbol: main
> + Type: R_AARCH64_ABS64
> + Addend: 0
> +
> +Symbols:
> + Local:
> + - Name: .text
> + Type: STT_SECTION
> + Section: .text
> +
> + Global:
> + - Name: main
> + Type: STT_FUNC
> + Section: .text
> + Size: 0x08
>
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>
--
Hal Finkel
Assistant Computational Scientist
Leadership Computing Facility
Argonne National Laboratory
_______________________________________________
llvm-commits mailing list
llvm-commits at cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
More information about the llvm-commits
mailing list