HEADS UP: merged llvm/clang 3.4

[Schaich Alonso]

On Sun, 23 Mar 2014 14:58:19 +0100
Dimitry Andric <dim at FreeBSD.org> wrote:

> On 23 Mar 2014, at 13:18, Schaich Alonso <alonsoschaich at fastmail.fm> wrote:
> > On Sun, 23 Mar 2014 12:40:07 +0200
> > Ivan Klymenko <fidaj at ukr.net> wrote:
> > 
> >> В Sun, 23 Mar 2014 02:29:38 -0700 (PDT)
> >> Jakub Lach <jakub_lach at mailplus.pl> пишет:
> >> 
> >>> Well, speaking of strangeness, I cannot build kernel on amd64.
> >>> 
> >>> cc -O2 -pipe -I. -I/usr/src/sys/dev/aic7xxx/aicasm -std=gnu99 
> >>> -Wsystem-headers -Werror -Wall -Wno-format-y2k -W
> >>> -Wno-unused-parameter -Wstrict-prototypes -Wmissing-prototypes
> >>> -Wpointer-arith -Wreturn-type -Wcast-qual -Wwrite-strings -Wswitch
> >>> -Wshadow -Wunused-parameter -Wcast-align -Wno-pointer-sign -c
> >>> aicasm_gram.c aicasm_gram.c:2:19: error: unused variable 'yysccsid'
> >>> [-Werror,-Wunused-const-variable]
> >>> static const char yysccsid[] = "@(#)yaccpar     1.9 (Berkeley)
> >>> 02/21/93"; ^
> >>> 1 error generated.
> >>> *** Error code 1
> >>> 
> >>> Stop.
> >>> make[2]: stopped in /usr/obj/usr/src/sys/GENERIC
> >>> *** Error code 1
> >>> 
> >>> Stop.
> >>> make[1]: stopped in /usr/src
> >>> *** Error code 1
> >>> 
> >> 
> >> I have the exact same problem...
> > 
> > As a workaround, you can avoid this by inserting
> > 
> >> NO_WERROR=
> >> WERROR=
> > 
> > into src.conf.
> 
> Aha, now I get it.  You seem to be building stable/9 from a stable/10 or
> newer host, which contains a newer version of yacc?  Apparently it
> inserts an unused banner into the produced parser, which clang then
> complains about.
> 
> The best solution is obviously to remove the banner, which is now being
> checked with the upstream yacc maintainer.  Since people will be stuck
> with the "wrong" version of yacc for some time, I guess the easiest way
> is to disable warnings for aicasm for now.
> 
> It is not as if anybody cares about warnings in aicasm; it has already
> been removed in head. :-)
> 
> -Dimitry
> 

Actually I was building 10-STABLE from a 2 week older snapshot of 10-STABLE.

The workaround is indeed from the FreeBSD-9 times' wiki article of building
FreeBSD with clang, however I remembered those yacc statements to already
have caused trouble in the past, so I look-up-ed the old articles and found
this.

Alonso