make check?

Gintautas gmlists at
Thu Jul 12 13:55:23 PDT 2001


SWK> I vote no. I used to run `make check' when it was available, and

Heh I thought I was the only one against, so I kept silent ;))
I agree with Seth W. Klein completely and I don't have much to add to
his point. Maybe it should be noted in the beginning of the book that "the
user can perform 'make check' to check the installation programs which
support this command" but I don't think you should include this line
in every package. If you did, then you should include a lot of notes
about erroneous 'make check' messages just for the sake of
completeness. My position may be related to my little Linux
experience, but I think that if 'make' doesn't spit out an error, then
everything is OK. I, personally, usually try to simply execute one of the
programs which are in the freshly installed package and see if it
works and that is usually enough to verify that a package works
correctly. This might not be the case with errors related with too
much optimization...

Bottom line: if a package works incorrectly, you'll see it in the near
future if you use it. Then you can go back and start
tampering/testing... If you don't use it then there's no point in
checking anyway ;-/  Having in mind that most errors happen in the
'make' process (or './configure' ;)) I say this is not worth adding. I
also doubt the effectiveness of these checks in catching those
not-compile-time errors...

Flame me ;)

 1948 - A novel for dyslexics by George Orwell.

Unsubscribe: send email to lfs-discuss-request at
and put unsubscribe in the subject header of the message

More information about the lfs-dev mailing list