Revision r193272 hoses kernel builds

[Steve Kargl]

On Mon, Jun 29, 2009 at 05:11:39PM -0400, John Baldwin wrote:
> Steve Kargl wrote:
> >On Mon, Jun 29, 2009 at 12:42:22PM -0700, Steve Kargl wrote:
> >>On Mon, Jun 29, 2009 at 11:44:51AM -0700, Steve Kargl wrote:
> >>>Am I missing something obvious?  r193272 was committed on 20090601
> >>>(over 28 days ago), and I haven't seen anyone else report this
> >>>build failure.
> >>>
> >>r193272 simply can't be correct.
> >>
> >>1.168 ! jhb   125: static int   nfs_clnt_tcp_soupcall(struct socket *so, 
> >>void *arg, int waitflag);
> >>
> >>1.116   ps    950: static void
> >>              951: nfs_clnt_tcp_soupcall(struct socket *so, void *arg, 
> >>              int waitflag)
> >
> >Making what appears to be the obvious fixes to
> >nfsclient/nfs_sockets.c leads to a failure in
> >/usr/src/sys/nfsserver/nfs_srvsock.c 
> >
> >cc -c -O2 -frename-registers -pipe -fno-strict-aliasing -march=opteron 
> >-std=c99 -g -Wall -Wredundant-decls -Wnested-externs -Wstrict-prototypes  
> >-Wmissing-prototypes -Wpointer-arith -Winline -Wcast-qual  -Wundef 
> >-Wno-pointer-sign -fformat-extensions -nostdinc  -I. -I/usr/src/sys 
> >-I/usr/src/sys/contrib/altq -D_KERNEL -DHAVE_KERNEL_OPTION_HEADERS 
> >-include opt_global.h -fno-common -finline-limit=8000 --param 
> >inline-unit-growth=100 --param large-function-growth=1000  
> >-fno-omit-frame-pointer -mcmodel=kernel -mno-red-zone  -mfpmath=387 
> >-mno-sse -mno-sse2 -mno-sse3 -mno-mmx -mno-3dnow  -msoft-float 
> >-fno-asynchronous-unwind-tables -ffreestanding -fstack-protector -Werror  
> >/usr/src/sys/nfsserver/nfs_srvsock.c
> >cc1: warnings being treated as errors
> >/usr/src/sys/nfsserver/nfs_srvsock.c: In function 'nfsrv_rcv':
> >/usr/src/sys/nfsserver/nfs_srvsock.c:530: warning: control reaches end of 
> >non-void function
> >*** Error code 1
> 
> I don't understand why this compiled fine.  I tested the NFS client (my 
> home dir was over NFS) with these changes as well.  I don't know how you 
> are getting a compile failure either.  Both the nfsclient and nfsserver 
> modules compile fine for me when built as modules.  Hmm, are you using
> NFS_LEGACYRPC?  That might explain it.  The patch at 
> http://www.FreeBSD.org/~jhb/patches/nfs_upcall.patch should fix the 
> build, but NFS_LEGACYRPC is also slated to be removed for 8.0 I believe.
> 

Yes, I had NFS_LEGACYRPC in the kernel config file.  I was updating
a cluster from a build that occurred within a few days of the new
RPC code, so I added the option as suggested by dfr(?).  Your patch
is the same as the patch I included in my reply to Rick.  It appears
that I must have been the only person using NFS_LEGACYRPC.

-- 
Steve