On 1/25/23 22:25, Tom Tromey wrote:
>>>>>> "Tom" == Tom de Vries via Gdb-patches <gdb-patches@sourceware.org> writes:
> 
> Tom> Add new proc is_x86_64_m64_target and use it by running:
>   
> Tom> +proc is_x86_64_m64_target {} {
> 
> Probably should have an intro comment... it's kind of pedantic though
> since IMO the name is pretty clear.
> 

Done.

> Tom> +    return [istarget x86_64-*-* ] && [is_lp64_target]
> 
> Should wrap this in [expr {...}], since otherwise calling this will
> return a string like "1 && 0" or something.  Maybe we have a double-eval
> somewhere if this is actually working.

Done.

I've also rewritten the requires, as you suggested in the other reply, 
so we're now left with just these matches:
...
$ ( cd gdb/testsuite; find -type f -name "*.exp*" \
   | xargs grep 'x86_64.*lp64' )

./gdb.dwarf2/entry-value-typedef.exp:
   if { [istarget "x86_64-*-linux*"] && [is_lp64_target] } {

./gdb.base/jit-reader.exp:
   require {is_any_target "i?86-*-*" "x86_64-*-*"} is_lp64_target

./gdb.arch/amd64-i386-address.exp:
   require {is_any_target "x86_64-*-*" "i?86-*-*"} is_lp64_target

./lib/gdb.exp:
   return [expr [istarget x86_64-*-* ] && [is_lp64_target]]
...

Committed as attached.

[ FWIW, I did wonder for a bit about using the name is_amd64_m64_target 
instead, but didn't manage to convince myself one way or the other, so I 
just stuck with what I started with. ]

Thanks,
- Tom