From finrod@ppp012.uio.no  Sat Jan 31 12:31:01 1998
Received: from ppp012.uio.no (ppp012.uio.no [129.240.240.13])
          by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA00909
          for <FreeBSD-gnats-submit@freebsd.org>; Sat, 31 Jan 1998 12:30:59 -0800 (PST)
          (envelope-from finrod@ppp012.uio.no)
Received: (from root@localhost)
	by valinor.ewox.org (8.8.8/8.8.5) id UAA04381;
	Sat, 31 Jan 1998 20:29:12 GMT
Message-Id: <199801312029.UAA04381@valinor.ewox.org>
Date: Sat, 31 Jan 1998 20:29:12 GMT
From: dag-erli@ifi.uio.no
Reply-To: dag-erli@ifi.uio.no
To: FreeBSD-gnats-submit@freebsd.org
Subject: man page typos
X-Send-Pr-Version: 3.2

>Number:         5614
>Category:       docs
>Synopsis:       Typos in ifconfig(8) and ep(4) man pages
>Confidential:   no
>Severity:       non-critical
>Priority:       low
>Responsible:    steve
>State:          closed
>Quarter:        
>Keywords:       
>Date-Required:  
>Class:          doc-bug
>Submitter-Id:   current-users
>Arrival-Date:   Sat Jan 31 12:40:00 PST 1998
>Closed-Date:    Mon Feb 23 07:07:59 PST 1998
>Last-Modified:  Mon Feb 23 07:08:39 PST 1998
>Originator:     Dag-Erling Coidan Smrgrav
>Release:        FreeBSD 2.2.5-STABLE i386
>Organization:
>Environment:

FreeBSD 2.2.5-STABLE, last CVSup'ed on January 25th.

>Description:

One minor typo in ifconfig(8) man page (impliments instead of implements)

One slightly more serious typo in ep(4) man page: according to TFS the
Etherlink ID port is 0x100, and not 0x110 as the man page states. I
can however confirm from personal experience that Trouble Awaits He
Who Places Another Card At Port 0x100.

OBTW, there is nothing in the source code that will produce the third
diagnostic message listed in the ep(4) man page... Should I be
surprised? ;)

Whoever wrote that man page, if you're ever in Oslo, Norway, I'll buy
you a beer :)

>How-To-Repeat:

% man ifconfig
% man ep

>Fix:

Below are unified diffs for both pages:

--- ifconfig.8.orig     Sat Jan 31 21:12:27 1998
+++ ifconfig.8  Sat Jan 31 21:12:37 1998
@@ -267,7 +267,7 @@
 Under appletalk, set the interface to respond to a 
 .Em netrange.
 of the form startnet-endnet. Appletalk uses this scheme instead of 
-netmasks though FreeBSD impliments it internally as a set of netmasks.
+netmasks though FreeBSD implements it internally as a set of netmasks.
 .It Cm phase
 The argument following this specifies the version (phase) of the 
 Appletalk network attached to the interface. Values of 1 or 2 are permitted.

--- ep.4.orig   Sat Jan 31 21:18:10 1998
+++ ep.4        Sat Jan 31 21:19:35 1998
@@ -89,9 +89,11 @@
 The 3c509 card has no jumpers to set the address.
 3Com supplies software to set the address of the card in software.
 To find the card on the ISA bus, the kernel performs a complex
-scan operation at IO address 0x110.
+scan operation at IO address 0x100.
 Beware!
 Avoid placing other cards at that address!
+Be particularly wary of Creative Labs SB32/AWE32/AWE64 cards, which by
+default use the I/O range 0x100-0x107 for their 3D effects processor.
 .Pp
 .Sh DIAGNOSTICS
 ep0: reset (status: %x)
>Release-Note:
>Audit-Trail:

From: Steve Price <sprice@hiwaay.net>
To: freebsd-gnats-submit@freebsd.org, dag-erli@ifi.uio.no
Cc:  Subject: Re: docs/5614: Typos in ifconfig(8) and ep(4) man pages
Date: Sun, 01 Feb 1998 01:01:16 -0600

 >Fix
 >
 >      Below are unified diffs for both pages:
 >
 [patch for ifconfig.8 elided]
 
 I will apply this patch shortly.
 
 [patch for ep.4 elided]
 
 IO address 0x110 is indeed correct.  See a recent discussion
 in -hackers (IIRC), PR kern/4459, and the comments Joerg Wunsch
 added in revision 1.22 of if_epreg.h by way of Justin Gibbs manual
 on the 3c509.
 
 Steve

From: dag-erli@ifi.uio.no (Dag-Erling Coidan Smrgrav)
To: Steve Price <sprice@hiwaay.net>
Cc: freebsd-gnats-submit@freebsd.org
Subject: Re: docs/5614: Typos in ifconfig(8) and ep(4) man pages
Date: 01 Feb 1998 20:06:20 +0100

 Steve Price <sprice@hiwaay.net> writes:
 > IO address 0x110 is indeed correct.  See a recent discussion
 > in -hackers (IIRC), PR kern/4459, and the comments Joerg Wunsch
 > added in revision 1.22 of if_epreg.h by way of Justin Gibbs manual
 > on the 3c509.
 
 Ah, I see. elink.h defines ELINK_ID_PORT to 0x100, hence the confusion.
 
 But how do you explain the problems I experienced (namely, "ep0 not
 found at 0x340" and "3c5x9 board(s) found at 0x220, 0x220, 0x220...")
 which disappeared when I mapped my SB32 to 0x108 instead of 0x100?
 
 -- 
  * Finrod (INTJ) * Unix weenie * dag-erli@ifi.uio.no * cellular +47-92835919 *
   RFC1123: "Be liberal in what you accept, and conservative in what you send"

From: Steve Price <sprice@hiwaay.net>
To: Dag-Erling Coidan Smrgrav <dag-erli@ifi.uio.no>, joerg@FreeBSD.ORG,
        gibbs@FreeBSD.ORG, freebsd-gnats-submit@FreeBSD.ORG
Cc:  Subject: Re: docs/5614: Typos in ifconfig(8) and ep(4) man pages
Date: Sun, 01 Feb 1998 13:52:39 -0600

 Dag-Erling Coidan Smrgrav wrote:
 > 
 > Steve Price <sprice@hiwaay.net> writes:
 > > IO address 0x110 is indeed correct.  See a recent discussion
 > > in -hackers (IIRC), PR kern/4459, and the comments Joerg Wunsch
 > > added in revision 1.22 of if_epreg.h by way of Justin Gibbs manual
 > > on the 3c509.
 > 
 > Ah, I see. elink.h defines ELINK_ID_PORT to 0x100, hence the confusion.
 
 Hmm...
 
 steve[/usr/src/sys]$ find . | xargs grep EP_ID_PORT
 ./i386/boot/netboot/3c509.c:    int data, j, io_base, id_port =
 EP_ID_PORT;
 ./i386/boot/netboot/3c509.h:#define EP_ID_PORT      0x100
 ./i386/isa/if_epreg.h:#define EP_ID_PORT      0x110
 ./pc98/boot/netboot/3c509.c:    int data, j, io_base, id_port =
 EP_ID_PORT;
 ./pc98/boot/netboot/3c509.h:#define EP_ID_PORT      0x100
 ./pc98/boot/netboot/if_epreg.h:#define EP_ID_PORT      0x100
 steve[/usr/src/sys]$ find . | xargs grep ELINK_ID_PORT
 ./i386/isa/elink.c:             outb(ELINK_ID_PORT, ELINK_RESET);
 ./i386/isa/elink.c:             outb(ELINK_ID_PORT, c);
 ./i386/isa/elink.h:#define      ELINK_ID_PORT   0x71d0
 ./i386/isa/elink.h:#define      ELINK_ID_PORT   0x100
 ./i386/isa/if_ep.c:    int data, i, j, id_port = ELINK_ID_PORT;
                                                  ^^^^^^^^^^^^^
 This is why you experienced the problem.  This should be
 EP_ID_PORT or ELINK_ID_PORT should be changed 0x110.  Joerg?
 Justin?
 
 ./i386/isa/if_ie.c:     outb(ELINK_ID_PORT, 0x00);
 ./i386/isa/if_ie.c:     outb(ELINK_ID_PORT, 0xff);
 ./i386/isa/if_ie.c:     outb(ELINK_ID_PORT, 0x00);
 ./i386/isa/if_ie.c:     outb(ELINK_ID_PORT, 0x00);
 steve[/usr/src/sys]$ 
 
 > But how do you explain the problems I experienced (namely, "ep0 not
 > found at 0x340" and "3c5x9 board(s) found at 0x220, 0x220, 0x220...")
 > which disappeared when I mapped my SB32 to 0x108 instead of 0x100?
 
 See above explanation.
 
 Steve
 
 > --
 >  * Finrod (INTJ) * Unix weenie * dag-erli@ifi.uio.no * cellular +47-92835919 *
 >   RFC1123: "Be liberal in what you accept, and conservative in what you send"

From: J Wunsch <j@uriah.heep.sax.de>
To: Steve Price <sprice@hiwaay.net>
Cc: =?iso-8859-1?Q?Dag-Erling_Coidan_Sm=F8rgrav?= <dag-erli@ifi.uio.no>,
        gibbs@FreeBSD.ORG, freebsd-gnats-submit@FreeBSD.ORG
Subject: Re: docs/5614: Typos in ifconfig(8) and ep(4) man pages
Date: Tue, 3 Feb 1998 19:34:05 +0100

 As Steve Price wrote:
 
 > > Ah, I see. elink.h defines ELINK_ID_PORT to 0x100, hence the confusion.
 > 
 > Hmm...
 > 
 > steve[/usr/src/sys]$ find . | xargs grep EP_ID_PORT
 > ./i386/boot/netboot/3c509.c:    int data, j, io_base, id_port =
 > EP_ID_PORT;
 > ./i386/boot/netboot/3c509.h:#define EP_ID_PORT      0x100
 > ./i386/isa/if_epreg.h:#define EP_ID_PORT      0x110
 > ./pc98/boot/netboot/3c509.c:    int data, j, io_base, id_port =
 > EP_ID_PORT;
 > ./pc98/boot/netboot/3c509.h:#define EP_ID_PORT      0x100
 > ./pc98/boot/netboot/if_epreg.h:#define EP_ID_PORT      0x100
 > steve[/usr/src/sys]$ find . | xargs grep ELINK_ID_PORT
 > ./i386/isa/elink.c:             outb(ELINK_ID_PORT, ELINK_RESET);
 > ./i386/isa/elink.c:             outb(ELINK_ID_PORT, c);
 > ./i386/isa/elink.h:#define      ELINK_ID_PORT   0x71d0
 > ./i386/isa/elink.h:#define      ELINK_ID_PORT   0x100
 > ./i386/isa/if_ep.c:    int data, i, j, id_port = ELINK_ID_PORT;
 >                                                  ^^^^^^^^^^^^^
 > This is why you experienced the problem.  This should be
 > EP_ID_PORT or ELINK_ID_PORT should be changed 0x110.  Joerg?
 
 Hmm, i simply forgot about the netboot stuff.  It would probably be
 better if netboot included the /usr/src/sys/ files anyway.  (I think
 biosboot does it this way for the serial line stuff.)
 
 -- 
 cheers, J"org
 
 joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE
 Never trust an operating system you don't have sources for. ;-)
State-Changed-From-To: open->closed 
State-Changed-By: steve 
State-Changed-When: Mon Feb 23 07:07:59 PST 1998 
State-Changed-Why:  
Now fixed in both -current and -stable. 
>Unformatted:
