home *** CD-ROM | disk | FTP | other *** search
Wrap
Received: from ppsw3.cam.ac.uk (mauve.csi.cam.ac.uk [131.111.8.38]) by nacm.com (8.6.10/8.6.9) with ESMTP id HAA16739 for <executor@nacm.com>; Tue, 9 May 1995 07:35:18 -0700 Received: from mole.bio.cam.ac.uk by mauve.csi.cam.ac.uk with SMTP-CAM (XTPP8.1) as ppsw.cam.ac.uk; Tue, 9 May 1995 15:34:14 +0100 Received: from localhost by mole.bio.cam.ac.uk (931110.SGI/MDTG-V1.3@mole.bio.cam.ac.uk) id AA12312; Tue, 9 May 95 15:38:11 +0100 Message-Id: <9505091438.AA12312@mole.bio.cam.ac.uk> To: ctm@ardi.com (Clifford Thomas Matthews) Cc: executor@nacm.com Subject: Re: status report In-Reply-To: Your message of "Tue, 09 May 1995 01:09:00 MDT." <m0s8jQA-00000BC@beaut.ardi.com> Date: Tue, 09 May 1995 15:38:10 +0100 From: Tim Cutts (Zoology) <tjrc1@mole.bio.cam.ac.uk> Sender: owner-executor@nacm.com Precedence: bulk In your message (Tue, 9 May 95 01:09 MDT), you wrote: >1.99n is *at least* a week away. > >Mat is in the middle of his new blitter mods, so he needs to finish up >his work before we can begin testing what will become 1.99n. There >are also a variety of bug fixes that we need to make before we can >release 1.99n. > >The reason only some Linux users are finding 1.99m to be ridiculously >slow is related to some new cursor handling code we put in. Some Mac >applications frequently (once through the event loop, for instance) >set the cursor to the same value. Executor used to detect this >special case and ignore these extra calls. However, 1.99m doesn't >ignore them, it passes the calls right on to the low-level cursor >manipulating code, which results in *a lot* of traffic between >Executor and the X-server. If the traffic is done via shared memory, >it doesn't affect things too much, but if you can't use shared memory >(say you're using an X-terminal), then you lose big time. I'm not sure I agree with this. I find some apps almost completely unuseable even using MITSHM on the local console. Tim.