Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 7 Dec 2009 14:04:33 +0100
From:      Luigi Rizzo <rizzo@iet.unipi.it>
To:        Dag-Erling Sm??rgrav <des@des.no>
Cc:        svn-src-head@freebsd.org, Ed Schouten <ed@80386.nl>, svn-src-all@freebsd.org, src-committers@freebsd.org, Hajimu UMEMOTO <ume@freebsd.org>
Subject:   Re: the need for safe dynamic string libraries
Message-ID:  <20091207130433.GA71902@onelab2.iet.unipi.it>
In-Reply-To: <86ein7t5m5.fsf@ds4.des.no>
References:  <200912052009.nB5K9okL098577@svn.freebsd.org> <ygek4wzpdv3.wl%ume@mahoroba.org> <20091207055752.GD64905@hoeg.nl> <20091207085927.GC57764@onelab2.iet.unipi.it> <86iqcjt93c.fsf@ds4.des.no> <20091207105343.GA62012@onelab2.iet.unipi.it> <86ein7t5m5.fsf@ds4.des.no>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Dec 07, 2009 at 11:56:18AM +0100, Dag-Erling Sm??rgrav wrote:
> Luigi Rizzo <rizzo@iet.unipi.it> writes:
> > What do you mean exactly ?
> 
> When libsbuf was ported to userland, sbuf_printf() was pessimized due to
> the limitations of libc's printf().  This makes certain parts of

but, isn't this an implementation detail ? What prevents the same
API to have two different implementation, one for kernel, one for userland ?

cheers
luigi

> pseudofs extremely inefficient: for instance, if you read from a
> /proc/whatever that's backed by an sbuf, it has to generate the entire
> file even if you only asked for a small chunk at a large offset.
> 
> DES
> -- 
> Dag-Erling Sm??rgrav - des@des.no



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20091207130433.GA71902>