From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx1.redhat.com ([209.132.183.28]:25804 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752665Ab1AYPxK (ORCPT ); Tue, 25 Jan 2011 10:53:10 -0500 Date: Tue, 25 Jan 2011 10:53:04 -0500 From: Jeff Layton To: Trond Myklebust Cc: Bryan Schumaker , Benny Halevy , Chuck Lever , Thomas Haynes , linux-nfs@vger.kernel.org Subject: Re: Patched up cthon source? Message-ID: <20110125105304.0b300c57@tlielax.poochiereds.net> In-Reply-To: <1295969912.6867.2.camel@heimdal.trondhjem.org> References: <83EC814B-1043-46B2-8F5E-E1FB1F5B56BF@oracle.com> <4D3DFE40.1070605@panasas.com> <4D3ED37D.6020606@netapp.com> <20110125095649.446212fd@tlielax.poochiereds.net> <1295969912.6867.2.camel@heimdal.trondhjem.org> Content-Type: text/plain; charset=US-ASCII Sender: linux-nfs-owner@vger.kernel.org List-ID: MIME-Version: 1.0 On Tue, 25 Jan 2011 10:38:32 -0500 Trond Myklebust wrote: > On Tue, 2011-01-25 at 09:56 -0500, Jeff Layton wrote: > > On Tue, 25 Jan 2011 08:43:25 -0500 > > Bryan Schumaker wrote: > > > > > On 01/24/2011 05:33 PM, Benny Halevy wrote: > > > > this is a problem with the version of nroff on newer distributions. > > > > > > My version of groff (1.21) is splitting the error into two lines, so I am using the following patch. > > > > > > Bryan > > > > > > > Yep, this workaround stopped working around for me on my rawhide box > > broke yesterday for the same reason. > > > > I think this patch may be a better approach than trying to grep out > > random stuff, but I don't know whether non-GNU tbl programs will barf > > on it. If anyone has a solaris box handy, could you let me know if > > this breaks on it? > > > > ----------------------[snip]------------------ > > [PATCH] cthon04: add "nowarn" option to nroff.in tables > > > > This prevents the tbl preprocessor from stuffing warnings into the > > resulting file, which subsequently can make nroff spew warnings like > > this on stderr: > > > > warning: file `nroff.in', around line 47: > > table wider than line width > > > > Unfortunately, I'm not sure how non-gnu tbl programs will react to > > this option, so YMMV. > > > > Signed-off-by: Jeff Layton > > --- > > general/nroff.in | 4 ++-- > > 1 files changed, 2 insertions(+), 2 deletions(-) > > > > diff --git a/general/nroff.in b/general/nroff.in > > index 522ac80..a8acb3a 100644 > > --- a/general/nroff.in > > +++ b/general/nroff.in > > @@ -18,7 +18,7 @@ benchmarks. > > .LP > > I ran the benchmarks on the following configurations: > > .TS > > -center; > > +center nowarn; > > l l l l. > > CPU Unix Fortran Memory > > .sp 4p > > @@ -38,7 +38,7 @@ A bubble sort of integers. > > A prime number generator. > > .LP > > .TS > > -center; > > +center nowarn; > > c|c|c s|c s| > > c|c|c s|c s| > > c|c|c s|c s| > > Here's a radical idea: what say we change the nroff.in file so that it > no longer tries to fit a 130ish character wide table into an 80 > character wide page? > Agreed... I actually gave that a try first, but after wasting a half hour or so on it, I just went "nowarn". I'd be all for fixing it the right way if someone can suggest a patch... An even more radical idea that Peter S. suggested a while back would be to update this test to use something more modern than nroff (XML parsing maybe?)... -- Jeff Layton