your change "x86: Also pass -P to $(CPP) when processing i386-opc.tbl"

your change "x86: Also pass -P to $(CPP) when processing i386-opc.tbl"

[Jan Beulich]

H.J.,

the description there talks about dealing with \ uses for line
continuation. -P, however, is about the suppression of line markers.
In recent work I've been wondering why i386-gen gives far off line
numbers in error diagnostics, making it sometimes quite difficult to
spot what is actually wrong. Furthermore i386-gen.c specifically has
code to consume the line markers.

Can you please explain to me what your change really was about? And
can you please propose an alternative way of having i386-gen give
useful diagnostics (unless you agree with reverting that change of
yours)?

Thanks, Jan

Re: your change "x86: Also pass -P to $(CPP) when processing i386-opc.tbl"

[H.J. Lu]

On Thu, Jul 28, 2022 at 5:33 AM Jan Beulich <jbeulich@suse.com> wrote:
>
> H.J.,
>
> the description there talks about dealing with \ uses for line
> continuation. -P, however, is about the suppression of line markers.
> In recent work I've been wondering why i386-gen gives far off line
> numbers in error diagnostics, making it sometimes quite difficult to
> spot what is actually wrong. Furthermore i386-gen.c specifically has
> code to consume the line markers.
>
> Can you please explain to me what your change really was about? And
> can you please propose an alternative way of having i386-gen give
> useful diagnostics (unless you agree with reverting that change of
> yours)?
>

Here it is:

https://sourceware.org/pipermail/binutils/2022-July/122060.html


-- 
H.J.