On Wed, Jul 12, 2017 at 11:06:23PM +0300, Artturi Alm wrote: > On Wed, Jul 12, 2017 at 06:12:34PM +0200, Mark Kettenis wrote: > > > Date: Mon, 10 Jul 2017 23:18:59 +0300 > > > From: Artturi Alm <artturi....@gmail.com> > > > > > > Hi, > > > > > > this does clutter my diffs, and the XXX comment is correct, > > > > It probably isn't. None of the other architectures have those macros > > in their <machine/asm.h>. > > > > Ok, didn't consider that, you're right and the diff withdrawn, but anyway, > what i was imprecisely after was that i'd prefer > having something like RODATA(name) found from sparc64/include/asm.h, > instead of the useless _C_LABEL() to use in armv7 md .S assembly, and > just because of the weird name i didn't find intuitive enough, i didn't > even suggest the arm64 EENTRY().. > Don't get me wrong, i'm more than happy to drop all the labeling > macros out of my diffs, and choose my self what is global and what is not, > while it's against the minimalism i'm now aiming at to even get the diffs > read&commented when freetime&interest does meet:)
_C_LABEL() is a holdover from the pre-ELF days. At one time (in NetBSD) this code still compiled using an a.out toolchain. In that era, it was defined to be _C_LABEL(x) _/**/x (this was also pre-ansi) (or _ ## x post ansi). On ELF _C_LABEL evaluates to just the symbol. It cannot have any association to either text or data because the proper use of it can refer to either text or data eg: bl _C_LABEL(c_func). Basically it is a relic from the past, it might make sense to just remove all references to _C_LABEL() at this point and just the symbol itself. Dale Rahn dr...@dalerahn.com
