As seen from the following log (building stage3)
/
-c -g -gnatpg -gnata -g
\
-I- -I. -Iada
/ -o ada/s-memory.o
mkdir -p ada/bldtools/nmake_s
cp -p /
/
/ ada/nmake.adb
ada/bldtools/nmake_s
(cd ada/bldtools/nmake_s; /tmp/gnatmake -q xnmake ; ./xnmake -s //nmake.ads
)
/
-c -g -gnatpg -gnata -I-
-I. -Iada
/ -o ada/treepr.o
this is because we build gnattools only after building stage3.

Comment #1 from bonzini at gnu dot org 2007-01-23 14:23
Is this really a bug, or just an enhancement?
As far as I know, it's been like this forever (even when you used to "make
gnat_lib_and_tools" after "make bootstrap").

Comment #2 from rguenth at gcc dot gnu dot org 2007-01-23 14:26
It's an enhancement. Currently we do not use the finally built gnatmake, so
we'd
increase coverage if we'd do. (I just wanted to record this in a different
bug)

Comment #3 from charlet at gcc dot gnu dot org 2007-01-23 15:44
xnmake is an internal support tool used to automatically generate a file,
and gnattools should indeed only be built after a full bootstrap of the
compiler.

As for more coverage of gnatmake, this usage of gnatmake does not test anything
really, and make check-ada tests gnatmake much more extensively anyway.

If anything, the dependencies in the makefile should be enhanced to avoid
calling gnatmake/xnmake unnecessarily after the first time.

Arno

Comment #4 from bonzini at gnu dot org 2007-01-23 16:02
Is the file placed in srcdir? If not, it's *right* to generate the file on
every stage.

Comment #5 from charlet at gcc dot gnu dot org 2007-01-23 16:10
<<
Is the file placed in srcdir? If not, it's *right* to generate the file on
every stage.
>>

Fair enough. Then there's nothing to fix or improve here.

Arno