Some 2.1.72 odd things.

James Mastros (
Fri, 12 Dec 1997 17:42:10 -0500 (EST)

NB: None of these have shown symptoms. Indeed, do to root's (i.e. my own)
stupidity (rm -r src in /usr, instead of /usr/src/bind-{whatever}), I
haven't been able to compile linux-2.1.72.

1) In Documentation/devices.t{xt|ex}, there are several odd things going on:
a) majors 45-47 (block) were reserved for "parallel port IDE disk",
"...ATAPI CD-ROM", and "...ATAPI disk". Why don't we handle
these like other IDE/ATAPI devices: have /dev/hdpar[a-f][1-90]?
b) \minor{152} has the description "Radio card (type?)": from
glancing at the driver, this is a generic interface (i.e. any
radio-card). Also, shouldn't this have a major, as more then
one radio-card is sensical?
c) All of major 27 has major changes reversed (originaly changed by
the merge of the new ftape circa 2.1.69 (.68?)).

2) In alpha's entry.S, many sys_ calls were changed to osf_ calls. Dosn't
this break source-level compatiblity for the old names? (But I don't
know the reason for those calls: perhaps we only need to worry about
binary compatiblity.

3) Wouldn't adding a printk when throbing the reset line be nice (in
process.c)? Also, shouldn't we printk when we enter the reboot-code (I
certainly think that rebooting the system is important enough to be
logged.) Also, can we gaurantee that udelay won't be fobared? (Likely;
it's to stupid a function to mess us (esph. since even if the timing is
off by a factor of ten here, it dosn't much matter.))

4) I think the "%%" -> "%" change in "100% soundblaster compatible" will
cause xconfig to barf; it's better to make [menu]config display an extra
"%". (in tcl/tk, % is a metachar)

5) Why this uncharesticly small patchlevel? I guess we really are getting
close to 2.2. Yeha!

-=- James Mastros

When the annals of distributed computing are written, and the name 'Bovine'
appears in there, I can say "Hey, I was a part of that, I checked .0012% of
the keyspace".
	-=- Brian Wilson <>
Go to before I come make you!