Subject: Re: How about to config Hayes ESP with glue device rather than magical option?
To: enami tsugutomo <enami@but-b.or.jp>
From: Michael L. VanLoon -- HeadCandy.com <michaelv@MindBender.serv.net>
List: current-users
Date: 09/08/1997 11:18:12
>> > Note that this patch doesn't include the patch to use the card over
>> > 115200 bps, since it is separate function.

>Matthias Drochner <drochner@zelux6.zel.kfa-juelich.de> writes:
>> Did you know that there is Linux driver source available
>> which uses enhanced features? (via www.hayes.com)

Unless someone has broken the com driver in current, it certainly knew
how to use the 1024-byte FIFO correctly.  This worked in 1.2.

>I didn't look the driver, but I got a document from anonymous ftp
>server of hayes.com.  I wrote above notice so that someone won't
>expect the feature which I implemented in my old PR.

>> I don't know the ESP cards, and I don't know if this would
>> give us more performance, however.

>I think ESP cards have four major enhanced feature; (1) 1024 byte
>fifo, (2) over 115200 bps rate, (3) memory mapped io and (4) DMA.
>Even in ns16550 compatile mode, (1) is already achieved with current
>com driver and (2) is probably available with my diff in old PR (I got
>one report that it doesn't work, but I can't debug it yet).  To
>achieve (3) and/or (4), the card should be operated in enhanced mode
>and probably new driver needs, but I don't know how much does it
>desired.  I'm currently using a ESP card on 486DX2/66 box to connect
>my ISP via 128kbps ISDN TA at 230400bps, and at least with this rate,
>it seems works well enough.

The 1.2 com driver knew how to do (1).  I haven't looked in the
current com driver to see if that code is still there.

You can do (2) just by setting the clock multiplier with the espca.exe
program that comes with the card, then running the card at a lower
multiple of the desired clock speed (at least I see no reason this
shouldn't work -- haven't actually attmpted it myself.

(3) and (4) would require a new driver, or at least a radically
different driver than the current com driver.  I don't think anyone
has actually attempted that.

-----------------------------------------------------------------------------
  Michael L. VanLoon                           michaelv@MindBender.serv.net
      Contract software development for Windows NT, Windows 95 and Unix.
             Windows NT and Unix server development in C++ and C.

        --<  Free your mind and your machine -- NetBSD free un*x  >--
    NetBSD working ports: 386+PC, Mac 68k, Amiga, Atari 68k, HP300, Sun3,
        Sun4/4c/4m, DEC MIPS, DEC Alpha, PC532, VAX, MVME68k, arm32...
    NetBSD ports in progress: PICA, others...
-----------------------------------------------------------------------------