Date: Sun, 1 Jul 2001 18:19:00 +0200 (CEST)
From: Adrian Bunk <bunk@fs.tum.de>
To: util-linux@math.uio.no
Subject: [patch] setterm -{dump,append} don't work on current virtual console

Hi Andries,

Colin Watson <cjw44@cam.ac.uk> reported the following Problem:

<--  snip  -->

When 'setterm -dump' or 'setterm -append' is called without further
arguments, it tries to dump the current virtual console, using
/dev/vcsa0. Unfortunately, according to the Linux Device List, the
current virtual console text/attribute contents are on /dev/vcsa.

<--  snip  -->

He sent the following patch:

--- util-linux-2.11g.orig/misc-utils/setterm.c
+++ util-linux-2.11g/misc-utils/setterm.c
@@ -1115,7 +1115,10 @@
     int fd, i, j;
     char *inbuf, *outbuf, *p, *q;

-    sprintf(infile, "/dev/vcsa%d", vcnum);
+    if (vcnum > 0)
+       sprintf(infile, "/dev/vcsa%d", vcnum);
+    else
+       sprintf(infile, "/dev/vcsa");
     fd = open(infile, 0);
     if (fd < 0 || read(fd, header, 4) != 4)
       goto try_ioctl;

cu
Adrian

-- 
A "No" uttered from deepest conviction is better and greater than a
"Yes" merely uttered to please, or what is worse, to avoid trouble.
                -- Mahatma Ghandi




Date: Sun, 1 Jul 2001 20:31:21 +0200 (MET DST)
From: Andries.Brouwer@cwi.nl
To: bunk@fs.tum.de, util-linux@math.uio.no
Cc: cjw44@cam.ac.uk, hpa@transmeta.com
Subject: Re: [util-linux] [patch] setterm -{dump,
    append} don't work on current virtual console

    From: Adrian Bunk <bunk@fs.tum.de>

    Colin Watson <cjw44@cam.ac.uk> reported the following Problem:

    When 'setterm -dump' or 'setterm -append' is called without further
    arguments, it tries to dump the current virtual console, using
    /dev/vcsa0. Unfortunately, according to the Linux Device List, the
    current virtual console text/attribute contents are on /dev/vcsa.

I don't think I'll make this change.

There are several machines here with /dev/vcsa0 and without /dev/vcsa.
So making this change would certainly give slight problems.

One could test both, but I prefer to leave the code as it is.
I consider the Linux Device List broken in this respect -
the vcs naming is exactly analogous to the tty naming, with tty0
on the same major as tty1, and there was no need to change anything.

Andries

[Let me cc hpa]
[A distribution should make both nodes, just to be on the safe side.]