This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Is -T meaningful as an assembler option?


I've been investigating a report that a command like

powerpc-eabi-gcc hello.c -T sim-hosted.ld

gives an assembler error from the -T option. It works OK if you omit the space after -T, and the syntax works with or without spaces on arm-none-eabi, mips-sde-elf, and m68k-elf toolchains I've tried.

It looks to me like the problem is that powerpc and a number of other back ends are including %{T} in ASM_SPEC. Is this supposed to be meaningful? GAS only documents -T as an assembler option for VAX, and says it's deprecated.

I've also noted that invoke.texi does not document -T in its normal use as a linker option. It's included in LINK_COMMAND_SPEC in gcc.c, so I assume this is the default interpretation for all targets. It's typically required to specify a linker script with -T when building for bare-metal targets, so this seems like a fairly critical hole in the manual.

I'm willing to prepare a patch that removes %{T} from all ASM_SPECs and documents -T as a linker option, but....

(a) is this the right thing to do?
(b) is this appropriate to submit for stage 3?

-Sandra


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]