Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 29 Apr 1995 10:41:33 -0400 (EDT)
From:      Peter Dufault <dufault@hda.com>
To:        jkh@time.cdrom.com (Jordan K. Hubbard)
Cc:        rgrimes@gndrsh.aac.dev.com, hackers@FreeBSD.org
Subject:   Re: What I'd *really like* for 2.0.5
Message-ID:  <199504291441.KAA13651@hda.com>
In-Reply-To: <13702.799154490@time.cdrom.com> from "Jordan K. Hubbard" at Apr 29, 95 04:21:30 am

next in thread | previous in thread | raw e-mail | index | archive | help
Jordan K. Hubbard writes:

Gentle Readers: Change the subject line when you begin a new thread!

I keep changing the subject, but I guess it screws up the threads
since other people are already replying to the previous thread.

All of those who are picky about what they read in -hackers are
missing this.

We now return to Jordan's proposal that should be resubjected as
"Jordan's message rampage proposal".

> 
> 1. Eliminate all but the "found" messages, allowing you to make kernels
>    with additional device support that aren't as chatty as the ones now.
>    When you boot a generic kernel on a given machine, it should show you
>    what it finds there and nothing more.  If you need more information
>    about what's being searched for, what's being found and what's not being
>    found, then I think that's a special circumstance and you should boot
>    with -v.

The verbose flag could become bit patterns.  One bit can be "log
failures".  And we could zap the kernel or the boot block
to keep the flag settings (zapping boot block is nice since you could
zap in your boot file name as well, including a "no pause before boot"
zap).

> 2. Carefully examine each and every probe message and attempt to regularize
>    it to some standard form, e.g.:
> 
> 	<devname><unit#>: <description>, port <port>, irq <irq>, drq <drq>, ioaddr <ioaddr>, iosize <iosize> [optional attributes] on <bus>.
> 	[<devname><unit#>: [optional information]]


Move all the print outs out of the drivers and into the probe/attach
code so that it is applied uniformly.  Drivers that just can't
resist saying SOMETHING with every boot can call a standard kernel
entry point to print out the start of the line.  We have the kdc
code with most of the info we need.  The notion is that the attach
only prints out the same information that you can later see with
something like "lsdev -lv".

>    The goal being to get principle probe messages on one line again and
>    to display the various parameters using the SAME naming conventions used
>    in userconfig to minimize confusion.  Some examples:
> 
> sc0: Syscons Console Driver, port 0x60-0x6f, irq 1 [VGA color] on ISA.
> sc0: [16 virtual consoles]

Or (switching to use the current devconf descriptions):

sc0: Graphics console [VGA color, 16 consoles]: port 0x60-0x6f, irq 1 on ISA

"Graphics console" is the description now in kern devconf.
A possible new "more detail" entry is "VGA color, 16 consoles".

Where Jordan has:
> ed0: General Ethernet Driver, port 0x280-0x29f, irq 15, ioaddr 0xd8000, iosize 16384 on ISA.
> ed0: [address 00:00:c0:58:99:72, type SMC8416C/SMC8416BT (16 bit)]

I'll have:

ed0: Ethernet adapter
ed0: [address00:00:c0:58:99:72, type SMC8416C/SMC8416BT (16 bit)]
ed0: port 0x280-0x29f, irq 15, ioaddr 0xd8000, iosize 16384 on ISA.

The devconf description is "Ethernet adapter"
The more_detail is "address00:00:c0:58:99:72, type SMC8416C/SMC8416BT (16 bit)"
The attach code knows how long these strings are and decides when to split
things to multiple lines.

You could then have a silent flag that would boot up like:

sc0: Graphics console
ed0: Ethernet adapter
...

I have a system going in at a client site that I wish could come up like
that.

Peter

-- 
Peter Dufault               Real Time Machine Control and Simulation
HD Associates, Inc.         Voice: 508 433 6936
dufault@hda.com             Fax:   508 433 5267



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199504291441.KAA13651>