Re: here we go again with the weird kernel build errors...

From: Kris Kennaway <kris_at_obsecurity.org>
Date: Fri, 29 Jul 2005 15:40:42 -0400
On Fri, Jul 29, 2005 at 01:58:05PM -0400, Brandon S. Allbery KF8NH wrote:
> On Fri, 2005-07-29 at 13:32 -0400, Brandon S. Allbery KF8NH wrote:
> > > ===> aic7xxx (depend)
> > > ===> aic7xxx/aicasm (depend)
> > > ===> aic7xxx/ahc (depend)
> > > ( cd /usr/src/sys/modules/aic7xxx/ahc/../aicasm; make aicasm; )
> > > cc -O -pipe -g -march=pentium4 -nostdinc -I/usr/include -I. -I/usr/src/sys/modules/aic7xxx/aicasm/../../../dev/aic7xxx/aicasm -g  -c aicasm_gram.c
> > > /usr/src/sys/modules/aic7xxx/aicasm/../../../dev/aic7xxx/aicasm/aicasm_gram.y:1933: error: conflicting types for 'yyerror'
> > > aicasm_gram.c:2976: error: previous implicit declaration of 'yyerror' was here
> > > *** Error code 1
> > 
> > Erh?
> > 
> 
> Answering my own question:  the culprit is /usr/local/bin/yacc from
> devel/bison.  Perhaps the kernel build needs to protect against this
> when not done as part of building a full world?  or is it no longer
> supported to reconfigure and rebuild a kernel without a full buildworld?

AFAIK you have to buildworld first, yes.  This builds the bootstrap
tools needed to safely build the kernel.

Kris

Received on Fri Jul 29 2005 - 17:40:50 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:40 UTC