This option is passed by clang driver if the target triple is aarch64-unknown-linux.
Details
Details
Diff Detail
Diff Detail
- Repository
- rL LLVM
Event Timeline
Comment Actions
Do you know why the flag is so inconsistent? (Naturally I expected something like elf_aarch64.)
Comment Actions
$ aarch64-linux-gnu-ld -V GNU ld (GNU Binutils for Ubuntu) 2.25 Supported emulations: aarch64linux aarch64elf aarch64elf32 aarch64elf32b aarch64elfb armelf armelfb aarch64linuxb aarch64linux32 aarch64linux32b armelfb_linux_eabi armelf_linux_eabi
As far as I know, "aarch64linux" is the only variant sent by clang. And if the triple is "aarch64-pc-freebsd" it doesn't pass "-m" switch at all.
Comment Actions
LGTM with a nti.
test/ELF/emulation.s | ||
---|---|---|
167–173 ↗ | (On Diff #40674) | Remove these lines as the other test cases doesn't test these fields. |
test/ELF/emulation.s | ||
---|---|---|
167–173 ↗ | (On Diff #40674) | As far as I can see, both X86-64 and PPC64 have the same lines, X86 has similar lines and only MIPS and MIPSEL miss them. |
test/ELF/emulation.s | ||
---|---|---|
167–173 ↗ | (On Diff #40674) | Then maybe we should remove from them too? I think the purpose of this test is to test is Machine field, so checking values of miscellaneous Size fields seem excessive. |
test/ELF/emulation.s | ||
---|---|---|
167–173 ↗ | (On Diff #40674) | OK, but I don't think that cleaning the other cases is relevant to this patch. |
test/ELF/emulation.s | ||
---|---|---|
167–173 ↗ | (On Diff #40674) | Agreed. This should be done in another patch. |