Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 20 Aug 1998 13:51:25 -0400 (EDT)
From:      "Matthew N. Dodd" <winter@jurai.net>
To:        Doug White <dwhite@resnet.uoregon.edu>
Cc:        Dag-Erling Coidan =?iso-8859-1?Q?Sm=F8rgrav?= <dag-erli@ifi.uio.no>, Bill Paul <wpaul@FreeBSD.ORG>, cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG
Subject:   Re: cvs commit: src/sys/pci if_xl.c
Message-ID:  <Pine.BSF.4.02.9808201349480.5813-100000@sasami.jurai.net>
In-Reply-To: <Pine.BSF.4.00.9808201015180.14968-100000@resnet.uoregon.edu>

next in thread | previous in thread | raw e-mail | index | archive | help

You've only addressed the specific case of the xl driver, and ignored the
other cases that have the same problem.

If the behavior is changed, and is consistent, using `grep -B 1' won't
kill you.

I vote for this change.

On Thu, 20 Aug 1998, Doug White wrote:

> 
> On 20 Aug 1998, Dag-Erling Coidan [iso-8859-1] Smørgrav wrote:
> 
> > Bill Paul <wpaul@FreeBSD.ORG> writes:
> > >   Log:
> > >   Fix small printf() bogon (forgot newline, and the message was longer that
> > >   80 cols).
> > 
> > Hmm, speaking of longer-than-80-col, I have a few PCI devices (Adaptec
> > 2940UW SCSI controller, Intel EtherExpress 100B LAN controller, 3Com
> > Etherlink XL LAN controller) that have names so long the probe message
> > is longer than 80 columns. Could the "int x irq x on pcix.xx.x" stuff
> > be written out on a separate line to avoid this?
> > 
> > Here's an example:
> > 
> > xl0: <3Com 3c905 Fast Etherlink XL 10/100BaseTX> rev 0x00 int a irq 11 on pci0.1
> > 7.0
> > 
> > should IMHO be
> > 
> > xl0: <3Com 3c905 Fast Etherlink XL 10/100BaseTX> rev 0x00
> > xl0: int a irq 11 on pci0.17.0
> 
> Eh, this is kinda yucky.  It's handy to be able to do 'dmesg | grep irq'
> and see what's where at a glance without having to backtrack and say 'oh,
> what's xl0? Hm, better run dmesg|grep xl0 to find out.'
> 
> The probe string could be shortened, though; we know it's 100mbit already
> (or will shortly), so just say <3com 3c905 Fast Etherlink XL> and get it
> over with:
> 
> xl0: <3Com 3c905 Fast Etherlink XL> rev 0x00 int a irq 11 on pci 0.17.0
> 
> Perhaps we should establish an informal limit on the size of the probe
> message, as well as some style-isms, like
> 
> .  Always preface messages from a driver with the driver name and unit
>    number.  bktr0 is a non-compliant example that I've been meaning to fix
>    for a while. (I deem violators of this item 'Linux-style probe
>    messages.' :-) )
> .  Don't tell us the speeds or capabilities of the device, use an
>    additional line for that.
> .  Omit 'Controller' and similar qualifiers.
> .  Ideally, say who made it, it's model number, and/or the mfr's official
>    designation for the device.
> 
> Besides, isn't that message put out by the kernel's standard probe code,
> so you'd have to do interesting gyrations to make it wrap nicely like
> that?
> 
> My US$.02.
> 
> Doug White                              | University of Oregon  
> Internet:  dwhite@resnet.uoregon.edu    | Residence Networking Assistant
> http://gladstone.uoregon.edu/~dwhite    | Computer Science Major
> 
> PS: Don't forget to hack the ISA drivers too.
> 
> 

-- 
| Matthew N. Dodd  |This space | '78 Datsun 280Z | FreeBSD/NetBSD/Sprite/VMS |
| winter@jurai.net |is for rent| '84 Volvo 245DL | ix86,sparc,m68k,pmax,vax  |
| http://www.jurai.net/~winter | Are you k-rad elite enough for my webpage?  |




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.02.9808201349480.5813-100000>