Date: Wed, 1 Apr 2015 14:57:29 +0300 From: Gleb Smirnoff <glebius@FreeBSD.org> To: Mateusz Guzik <mjguzik@gmail.com> Cc: svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org Subject: Re: svn commit: r280955 - in head/sys: modules/notrandom dev/notrandom Message-ID: <20150401115729.GA64665@FreeBSD.org> In-Reply-To: <20150401115204.GB16649@dft-labs.eu> References: <20150401113628.GA16649@dft-labs.eu> <20150401114313.GZ64665@FreeBSD.org> <20150401115204.GB16649@dft-labs.eu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Apr 01, 2015 at 01:52:04PM +0200, Mateusz Guzik wrote: M> /dev/zero returns 0s, while notrandom was implemented to return 7s. M> M> We could e.g. symlink /dev/notrandom to /dev/zero, but that would break M> compatiblity with Solaris apps depending on this functionality. M> M> It could also introduce hard-to-debug problems. After all, readers get a M> different value than they expect. M> M> Given that we have some parts of Solaris merged (zfs, dtrace) and work M> with open source forks on these technologies, I think we should diverge M> only when necessary. M> M> As a side note I'm surprised with the choice of 7. M> M> I would expect 3, no more, no less. 3 would be the number returned, and M> the number readers receive would be 3. You can add the seven_region variable to dev/null/null.c and implement the new device here. Now notrandom.c shares > 50% of code with null.c. -- Totus tuus, Glebius.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20150401115729.GA64665>