Re: bootstrap oddities: cvs vs. arch? Long pathnames?

[Reiner Steib]

On Wed, Apr 21 2004, Reiner Steib wrote:

[ "make bootstrap" fails when built from an arch (not cvs) checkout if
  pathname is longer than a certain limit. ]

> I got same behavior (success in /var/tmp/ste/test; failure with "Lisp
> nesting exceeds max-lisp-eval-depth" in the directory with long
> pathname on two additional machines (with the same source tree).  One
> with SuSE 8.2 as above (Dual Pentium III machine), the other with SuSE
> 7.2 (Pentium 4).
>
> bootstrap also fails when in /var/tmp/ste/abcefghijk/test, but with a
> different error location:
[...]

Now checked out a fresh tree from cvs.  It bootstrapped successfully.
Then in the arch tree: failed with "Lisp nesting exceeds
max-lisp-eval-depth" as before.

Then I renamed '{arch}' to '.arch' (mv '{arch}' .arch).  After this,
"make bootstrap" ran successfully in with the sources from the arch
repository.  Therefore, I think the guess from your previous mail
probably is correct:

,----[ <news:address@hidden> ]
| I'd think that the byte-compiler shouldn't trigger the VC machinery
| at all, but maybe it does, and nobody noticed it before.
`----

Do your compile problems disappear too after renaming the {arch}
directory?

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo--- PGP key available via WWW   http://rsteib.home.pages.de/