Re: M4 assembler error on compile

On 8/17/23 08:57, Eric Blake wrote:
> m4 is single-threaded, so an asynsafe-spin operation is
> not going to be vital to performance

Gnulib has various options for omitting unnecessary locking code in 
single-threaded apps. If you have time, you might investigate why that 
is not happening for GNU m4. This should improve performance on most 
platforms, which would be worth doing even if Solaris 7 is no longer 
supported. And it should also fix this Solaris 7 porting glitch as a 
side effect.