[m68k] asm regression with 3.3/3.4/3.5

[Zack Weinberg]

Andreas Schwab <schwab@suse.de> writes:

> Zack Weinberg <zack@codesourcery.com> writes:
>
>> Andreas Schwab <schwab@suse.de> writes:
>>
>>> Glibc depends heavily on being able to force a specific register to be
>>> able to do inline syscalls.
>>
>> On the x86 we have a complete set of single-register constraint
>> letters so this isn't a problem.
>
> Yes, but all the other sane architectures don't.

But do all the 'other sane architectures' also use wacky argument-
passing conventions for system calls?  I thought x86 was the only
Linux architecture where system calls didn't use the normal calling
convention (hence, the only architecture where glibc has a problem).

zw