home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!gumby!wupost!emory!mathcs.emory.edu!km
- From: km@mathcs.emory.edu (Ken Mandelberg)
- Newsgroups: comp.sys.sun.misc
- Subject: Re: Why is initial screen I/O so slow?
- Message-ID: <9296@emory.mathcs.emory.edu>
- Date: 12 Aug 92 15:23:55 GMT
- References: <1992Aug12.141157.2277@lsl.co.uk>
- Sender: news@mathcs.emory.edu
- Reply-To: km@mathcs.emory.edu
- Organization: Emory University, Dept of Math and CS
- Lines: 22
-
- One of the Sun firmware developers gave a talk at a recent Usenix on the
- "openprom". Although, it wasn't the point of his talk, since the prom
- does have the code for the raw /dev/console terminal emulation he did
- address the slow scrolling. What he said was that the widely held
- belief that the slow scrolling was a consequence of the Forth interpter being
- slow was not correct. It turns out there was a bug in the terminal emulation
- code that they didn't notice until after the prom was frozen. I think it
- was something like they were reading or writing from some very expensive
- location continually in the code. He said that the scrolling speed was
- much improved in recent versions of the prom that fixed the bug.
-
- As everyone has noted, in most cases the slow scrolling of the full screen
- console is not a real concern. Tha prom driver is not used at all
- once a window system takes over the framebuffer and does its own terminal
- emulations.
-
- ---
- Ken Mandelberg | km@mathcs.emory.edu PREFERRED
- Emory University | {rutgers,gatech}!emory!km UUCP
- Dept of Math and CS | km@emory.bitnet NON-DOMAIN BITNET
- Atlanta, GA 30322 | Phone: Voice (404) 727-7963, FAX 727-5611
-
-