m4-1.4 again solved

Erika Pacholleck Pacholleck.E at knuut.de
Sun Jun 18 13:30:59 PDT 2000


Gerard Beekmans <-- Son, 18 Jun 2000 :
> 
> What if you just ignore the make check for now. M4 might rely on other
> programs to perform a test which aren't installed yet. Doesn't it
> compile at all (ignore that sigstack warning. Everybody gets it, it
> should not result in a compilation failure). 

The only ones left were m4 (still static) and psmisc (which I have to install
by hand cause syvinit is already there and overwrites parts of psmisc).
A warning is what we call "ein Pups gegen den Wind", so small you do
not even recognize it, so this did not bother me. But if it is connected to
all checks failing I don´t see any reason to replace my static (and get rid
of the nss-files).

> SNIP
> ./configure --prefix=/usr 1> configure-log 2>&1
> mmake 1> make.log 2>&1
>  SNIP

Done that, exactely how you wrote (including typo) - yep, happens if you
think copying is less mistakes than typing.
Added a third one for the check. Afterward - of course curious - fired up vim
on the check-log and YIPPIE, completely all checks passed. Guess I´ve never
been that fast in typing make install ;-)

Sure I kept the old one backed up, so I tried to find diffs. Results:
config-log = identical, make-log = identical, check-log = totally different
config.log = identical, config.cache = identical, config.h = identical
config.status = DIFF (verified with /root/script-dyn/Konfiguriere.m4.old)
--------------------------------------
./configure  --host=i586-pc-linux-gnu --prefix=/usr
./configure  --prefix=/usr
--------------------------------------
Transferred all to Windy partition and ran second checker on it.
Yep, that´s definetly the only diff, a real nice one. 

As I get with nearly every ./configure the message I should specify the
hostname all my scripts begin with ./configure --host=i586-pc-linux-gnu \
So I prewrote all my scripts in DLD (ger kbd) with just ./configure --help
to check for the options the package accepts. And host is a quite normal
one which should not lead into such troubles, specially not if --help
explicitly states this option. Still dont´t understand how a host option can
influence that - but solved for now.

Thanks very much for the hint. This way you´re at least off the checking.
Maybe you can try it the other way round if you don´t have anything funny
at hand (on my machine this is reproducable any time)
Erika
--
Mail archive: http://www.pcrdallas.com/mail-archives/lfs-discuss
IRC access: server: irc.linuxfromscratch.org port: 6667 channel: #LFS
Unsubscribe: email lfs-discuss-request at linuxfromscratch.org and put
"unsubscribe" (without the quotation marks) in the body of the message
(no subject is required)



More information about the lfs-dev mailing list