Re: toolchain/53034

["Roland Illig via gnats" <gnats-admin%NetBSD.org@localhost>]

The following reply was made to PR toolchain/53034; it has been noted by GNATS.

From: Roland Illig <roland.illig%gmx.de@localhost>
To: gnats-bugs%netbsd.org@localhost
Cc: 
Subject: Re: toolchain/53034
Date: Sat, 28 Jun 2025 12:07:30 +0200

 The fix from 2018 results in this output:
 > `all' is up to date.
 
 That seems wrong to me, as there is no rule to make foo.c.
 
 $ make -r -f 53034.mk -dm
 > Setting main node to "all"
 > Make_ExpandUse: examine all
 > ExamineLater: need to examine "lib.a"
 > Make_ExpandUse: examine lib.a
 > ExamineLater: need to examine "lib.a(foo.o)"
 > Make_ExpandUse: examine lib.a(foo.o)
 > ExamineLater: need to examine "foo.o"
 > Make_ExpandUse: examine foo.o
 > Examining foo.o....JOIN node...source not made...up-to-date.
 > Examining lib.a(foo.o)...nonexistent...up-to-date.
 > Examining lib.a...nonexistent...up-to-date.
 > Examining all...nonexistent...up-to-date.
 > `all' is up to date.
 > Examining .END...nonexistent...nonexistent and no sources...out-of-date.
 >  recheck(.END): update time from nonexistent to now
 
 I find the "nonexistent...up-to-date" lines confusing, as they don't
 reveal the reason why the target should be up-to-date despite being
 nonexistent.
 
 I also find it confusing that foo.c is not mentioned in the debug log at
 all. I guess that matches David's analysis around OP_MADE.