From mint-bounce@lists.fishpool.fi Mon Jul 26 12:40:05 2010 X-Virus-Scanned: amavisd-new at demon.co.uk Subject: Re: [MiNT] "pid 80 (conholio) : unifs: couldn't match a drive with a directory" From: Alan Hourihane To: Peter Persson Cc: Vincent =?ISO-8859-1?Q?Rivi=E8re?= , mint@lists.fishpool.fi In-Reply-To: References: <4C4DB3C9.4050104@freesbee.fr> <1280161514.3622.8359.camel@jetpack.demon.co.uk> Content-Type: text/plain; charset="us-ascii" Date: Mon, 26 Jul 2010 17:38:14 +0100 Message-ID: <1280162294.3622.8382.camel@jetpack.demon.co.uk> Mime-Version: 1.0 X-Mailer: Evolution 2.28.3.1 Content-Transfer-Encoding: 7bit X-ecartis-version: Ecartis v1.0.0 Sender: mint-bounce@lists.fishpool.fi Errors-to: mint-bounce@lists.fishpool.fi X-original-sender: alanh@fairlite.co.uk Precedence: bulk List-help: List-unsubscribe: List-Id: X-List-ID: List-subscribe: List-owner: List-post: On Mon, 2010-07-26 at 18:31 +0200, Peter Persson wrote: > 26 jul 2010 kl. 18.25 skrev Alan Hourihane: > > > You should take that with some context. > > > > If you build new software against the newer MiNTlib then it can take > > advantage of new system calls which are not present in older kernels. > > Depending on how the software is written it may not fail gracefully if a > > system call fails. > > Phew, that sounds better. > > What's the policy on TOS compatibility and compatibility with older kernels? Mintlib used to have some degree of emulation for TOS built in, right? And that's still the case. > > Now, this is really with new unix style software, but for Atari software > > compiled against the mintlib it's unlikely to take advantage of these > > new system calls so should work just as well with old kernels as new > > kernels. > > Any particular system/library calls to worry about? New calls are fchdir, fdopendir and dirfd which don't have TOS requirements so aren't TOS related. Alan.