Date: Thu, 23 Sep 2010 15:00:23 +0100 From: Karl Pielorz <kpielorz_lst@tdx.co.uk> To: Ruslan Mahmatkhanov <cvs-src@yandex.ru> Cc: freebsd-ports@freebsd.org Subject: Re: 'mbuffer' broken for network support / pipes under 7.2-R / 8.1-R? Message-ID: <F766D219E8A6859B1F507A76@HexaDeca64.dmpriest.net.uk> In-Reply-To: <65031285246145@web49.yandex.ru> References: <B266E3BF267EA77C5E099324@HexaDeca64.dmpriest.net.uk> <4C9A44A3.5000502@yandex.ru> <498B06B4701B41C283B7A5BF@HexaDeca64.dmpriest.net.uk> <65031285246145@web49.yandex.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
--On 23 September 2010 16:49 +0400 Ruslan Mahmatkhanov <cvs-src@yandex.ru> wrote: > Hi Karl, > sorry about webmail formatting. Yeah, it certainly mangled the original mail when quoting it! <g> > Try to add this line into port's Makefile, and then reinstall mbuffer: > CFLAGS+=-m64 > > See if your 'network i/o' test case now passes. Makes no difference - it *doesn't fail* for network I/O (it completes fine) - it just dumps a single line to stderr at the start, which says: " mbuffer: warning: unable to set socket buffer size: No buffer space available " Adding '-m64' didn't get rid of that, or any of the compile time warnings... I can live with the warning (better than it not working at all! :) - unless it might be doing something nasty (like messing up network performance - though it seems to be getting close to wire speed for transfers). -Karl
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?F766D219E8A6859B1F507A76>