Jesse Tie Ten Quee
highos at highos.com
Wed Nov 15 14:41:50 PST 2000
On Thu, Nov 16, 2000 at 12:08:01AM +0100, Matthias Benkmann wrote:
> The developers know that and make those failures non-fatal. On a Linux
> system a "make check" is expected to return status 0. I have done a make
> check for all pre-chroot LFS packages and not a single one return exit
> status !=0 except for glibc and glibc is the single one package for which
> "make check" MUST NOT fail. If it does, it means the library is broken and
> should not be used. I guess I'll have to write a bug report. Damn, I guess
> I'll have to try glibc2.2 first. I'm sure they don't accept bug reports
> for "old" versions.
i said "fail" not talking about a return status code.
> I consider looking for obscure bugs while in chroot a lot more complicated
> than knowing right away when something is wrong and what package it occurs
Well i would blame your host distrobution, as glibc right now gets
compiled outside of the static (and safe if i may add) environment.
As for obscure, if it's that bad, why haven't i seen a post about this
I've finished LFS quite a number of times under quite a number of
different environments, and i personally feel that a make test will just
complete things more then the user needs to know, and i've _never_ had a
a messed up install, there was once when i had some problems with
glibc, but i never did a make install, which a make test would not have
helped ;) (as it is, i continued and finished that install after i
realized what it was)
Try compiling glibc under your static environment, that's my advice.
This is just my opinion *shrugs*
Jesse Tie Ten Quee - highos at highos dot com
Unsubscribe: send email to lfs-discuss-request at linuxfromscratch.org
and put unsubscribe in the subject header of the message
More information about the lfs-dev