Re: Q: Extending the sysctl MIB for Linuxulator variables

[Daniel C. Sobral]

Matthew Dillon wrote:
> 
> :Well, it's also a module, so perhaps we should create the whole subtree
> :for modules (as was already discussed several times..)
> :
> :Andrzej Bialecki
> 
> Yes, this is very true.  But I think we are fooling ourselves if we
> believe linux emulation will not become 'standard' in the near future.
> Then we'll kick ourselves for giving the sysctl's convoluted names :-)

Also, the way we choose to tread leds to a highly modularized
kernel. Placing stuff in a "module" category would be somewhat
redundant, then.

--
Daniel C. Sobral(8-DCS)
[EMAIL PROTECTED]
[EMAIL PROTECTED]

"You intend to eat me, then?" he asked the dragon.
"Well, I must admit, more for the amusement than the taste."




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Dominic Mitchell]

On Mon, Aug 16, 1999 at 02:54:53PM -0700, Steve Kargl wrote:
> kern.modules seems to be slightly more general in that you can
> have kern.modules.xxx where xxx is anything under /modules that
> needs/wants to some tuning via sysctl.

This is daft.  Given that we are planning on makeing everything
dynamically configurable, there should be little to no differences as to
whether something is a module or not.  And besides, what about all the
modules whose static equivalents already have entries in the sysctl
space, like nfs?
-- 
Dom Mitchell -- Palmer & Harvey McLane -- Unix Systems Administrator

"Finally, when replying to messages only quote the parts of the message
 your will be discussing or that are relevant. Quoting whole messages
 and adding two lines at the top is not good etiquette." -- Elias Levy
-- 
**
This email and any files transmitted with it are confidential and 
intended solely for the use of the individual or entity to whom they   
are addressed. If you have received this email in error please notify 
the system manager.

This footnote also confirms that this email message has been swept by 
MIMEsweeper for the presence of computer viruses.
**


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Mark Murray]

> I kinda like the idea of a top-level compat category; it will keep the
> top level uncluttered when sysv and iBCS compatibility start requiring
> their own knobs, and if you put linux at the top level this will later
> be used as justification for putting all the other "compat" stuff up
> there too.  I think it's a slippery slope.

Agreed. Sysctl's have a decent hierarchical structure; let's use it.

Examples of broken/unused hierarchical structures:

o .COM namespace.
o Average DOS user's hard disk management (everything off root).

M
--
Mark Murray
Join the anti-SPAM movement: http://www.cauce.org


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Jeroen Ruigrok/Asmodai]

* Doug ([EMAIL PROTECTED]) [990817 03:40]:
>   In case anyone cares I'd like to put in a vote for compat.linux.
>From the design standpoint this balances the needs of prominence and clean
>top level name space nicely. 

Count me as another in favor of Mike's explanation.

Like Mike said, there were a few mistakes already at the top, and IMHO
linux ABI stuff doesn't justify a top-level assignment since in fact
it's a remapping of Linux calls to FreeBSD equivalent calls, not a true
implementation of Linux. So compat.* sounds way more sensible to shim-
like implementations.

Just my 0.02 euro's.

-- 
Jeroen Ruigrok van der Werven  asmodai(at)wxs.nl
The BSD Programmer's Documentation Project 
Network/Security SpecialistBSD: Technical excellence at its best
Take thy beak from out my heart and take thy form from off my door!


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Poul-Henning Kamp]

In message <[EMAIL PROTECTED]>, Mike Smith writes:
>>  In case anyone cares I'd like to put in a vote for compat.linux.
>> >From the design standpoint this balances the needs of prominence and clean
>> top level name space nicely. 
>
>And in case it's not clear from the exposition in my message to Poul, I 
>would find this most agreeable too.

Cool! Concensus, (I thought "compat" was on the black list with "emulation" ?)

Now everybody shut up and code!

--
Poul-Henning Kamp FreeBSD coreteam member
[EMAIL PROTECTED]   "Real hackers run -current on their laptop."
FreeBSD -- It will take a long time before progress goes too far!


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Poul-Henning Kamp]

In message <[EMAIL PROTECTED]>, Mike Smith writes:

>> I think that is too obscure considering the exposure this will get.
>
>What "exposure"?  It's a backend to a tuning interface for our ABI 
>compatibility...

We will be judged on how well we run linux more than many other sane
factors in the future.

>>From the perspective of an integrated namespace, we've already made the 
>wrong moves insofar as vm.* should be kern.vm.*, vfs.* should be 
>kern.vfs.*, etc.  Either the entire kernel namespace should have a 
>presumed leading kern. (and the existing kern.* nodes need to move) or 
>we should relocate stuff to reflect a more ubuquitous naming 
>arrangement.

You know, all your argument here supports is that the "kern" toplevel
tree is a mistake :-)

--
Poul-Henning Kamp FreeBSD coreteam member
[EMAIL PROTECTED]   "Real hackers run -current on their laptop."
FreeBSD -- It will take a long time before progress goes too far!


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[John-Mark Gurney]

Brian F. Feldman scribbled this message on Aug 16:
> > be used as justification for putting all the other "compat" stuff up
> > there too.  I think it's a slippery slope.
> 
> much as possible. Just like the ports, it makes things easier on
> everyone if we use lower-case (compat.linux, compat.ibcs2,
> compat.svr4.)

just to amend my previous comment, I would prefer compat.linux too...

-- 
  John-Mark Gurney  Voice: +1 541 684 8449
  Cu Networking   P.O. Box 5693, 97405

  "The soul contains in itself the event that shall presently befall it.
  The event is only the actualizing of its thought." -- Ralph Waldo Emerson


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[John-Mark Gurney]

Mike Smith scribbled this message on Aug 16:
> >From the perspective of an integrated namespace, we've already made the 
> wrong moves insofar as vm.* should be kern.vm.*, vfs.* should be 
> kern.vfs.*, etc.  Either the entire kernel namespace should have a 
> presumed leading kern. (and the existing kern.* nodes need to move) or 
> we should relocate stuff to reflect a more ubuquitous naming 
> arrangement.

hmmm... last I thought, all of sysctl was information about the kernel,
so why don't we stick vm, vfs, net, hw, machdep, user, and p1003_1b all
under kern??  I mean, net is all the parameters of the kernel networking,
the hw is the hardware the kernel is running on, machdep is more machine
dependant (though what is so machine dependance about msgbuf?) info about
the kernel, even user contains kernel paramters for the user process...

everything could be stuck under kern.* if you really wanted to, and then
it just moves the "top" level up to kern.* and doesn't help any... and
if we avoid expanding our top levels, we will soon end up w/ sysctl names
that are 80+ characters long because they are 10 levels deep...

if you can't tell, I'm in favor of the linux.*

-- 
  John-Mark Gurney  Voice: +1 541 684 8449
  Cu Networking   P.O. Box 5693, 97405

  "The soul contains in itself the event that shall presently befall it.
  The event is only the actualizing of its thought." -- Ralph Waldo Emerson


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Doug]

On Mon, 16 Aug 1999, Mike Smith wrote:

> > In case anyone cares I'd like to put in a vote for compat.linux.
> > >From the design standpoint this balances the needs of prominence and clean
> > top level name space nicely. 
> 
> And in case it's not clear from the exposition in my message to Poul, I 
> would find this most agreeable too.

It dawns on me that this might look like I'm trying to take credit
for the idea, which I'm not. I'm just agreeing with those who proposed it
already and stating my reasons for agreeing. 

Doug
-- 
On account of being a democracy and run by the people, we are the only
nation in the world that has to keep a government four years, no matter
what it does.
-- Will Rogers



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Mike Smith]

>   In case anyone cares I'd like to put in a vote for compat.linux.
> >From the design standpoint this balances the needs of prominence and clean
> top level name space nicely. 

And in case it's not clear from the exposition in my message to Poul, I 
would find this most agreeable too.

-- 
\\  The mind's the standard   \\  Mike Smith
\\  of the man.   \\  [EMAIL PROTECTED]
\\-- Joseph Merrick   \\  [EMAIL PROTECTED]




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Doug]

In case anyone cares I'd like to put in a vote for compat.linux.
>From the design standpoint this balances the needs of prominence and clean
top level name space nicely. 

Doug



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Brian F. Feldman]

On Mon, 16 Aug 1999, Jordan K. Hubbard wrote:

> I kinda like the idea of a top-level compat category; it will keep the
> top level uncluttered when sysv and iBCS compatibility start requiring
> their own knobs, and if you put linux at the top level this will later
> be used as justification for putting all the other "compat" stuff up
> there too.  I think it's a slippery slope.
> 
> - Jordan
> 

This is the most convincing argument so far. It's got my vote. One
caveat: let's _please_ not use capitalization in the MIB tree as
much as possible. Just like the ports, it makes things easier on
everyone if we use lower-case (compat.linux, compat.ibcs2,
compat.svr4.)

 Brian Fundakowski Feldman  _ __ ___   ___ ___ ___  
 [EMAIL PROTECTED]   _ __ ___ | _ ) __|   \ 
 FreeBSD: The Power to Serve!_ __ | _ \._ \ |) |
   http://www.FreeBSD.org/  _ |___/___/___/ 



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Mike Smith]

> > Given that "ABI" is a bit obscure, kern.compat is the only sensible 
> > choice.
> 
> One one hand you're right (it is a compatibility stub) but OTOH it is also
> a kernel module... ;-)
> 
> Perhaps modules like this will want to have their stuff in BOTH places,
> i.e. in kernel.compat and in kernel.modules, depending what the given
> sysctl does.

Tuning parameters should be organised by function, not by 
implementation.

(Otherwise, think for a moment about a parameter that's used in more 
than one place...)

-- 
\\  The mind's the standard   \\  Mike Smith
\\  of the man.   \\  [EMAIL PROTECTED]
\\-- Joseph Merrick   \\  [EMAIL PROTECTED]




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Andrzej Bialecki]

On Mon, 16 Aug 1999, Mike Smith wrote:

> > > Yes, this is very true.  But I think we are fooling ourselves if we 
> > > believe linux emulation will not become 'standard' in the near future.
> > > Then we'll kick ourselves for giving the sysctl's convoluted names :-)
> > 
> > Yeah... Then, the next in line after "linux" are: ibcs2 and svr4 and
> > whatever comes next. Can you live with them as main sysctl categories?
> 
> Adding anything at the top level would be a terrible mistake.
> 
> Given that "ABI" is a bit obscure, kern.compat is the only sensible 
> choice.

One one hand you're right (it is a compatibility stub) but OTOH it is also
a kernel module... ;-)

Perhaps modules like this will want to have their stuff in BOTH places,
i.e. in kernel.compat and in kernel.modules, depending what the given
sysctl does.

Andrzej Bialecki

//  <[EMAIL PROTECTED]> WebGiro AB, Sweden (http://www.webgiro.com)
// ---
// -- FreeBSD: The Power to Serve. http://www.freebsd.org 
// --- Small & Embedded FreeBSD: http://www.freebsd.org/~picobsd/ 



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Mike Smith]

> >Given that "ABI" is a bit obscure, kern.compat is the only sensible 
> >choice.
> 
> I think that is too obscure considering the exposure this will get.

What "exposure"?  It's a backend to a tuning interface for our ABI 
compatibility...

> It doesn't really matter much what we feel about it, linux will be
> a native and 100% normal binary format for us, if we try to
> marginalize it we loose in perception.

I don't see how organising the sysctl namespace in a tidy fashion 
constitutes "marginalising" anything.

> We have things to make us posix compatible at the top level already,
> I don't see why the linux stuff should live under the top level too.

One wrong...

> And as father of sysctl, I think this discussion needs to come to
> a close rather than waste more bandwidth, so unless Mike can convince
> us why "Adding anything at the top level would be a terrible mistake"
> I think the conclusion is "linux.*"

For the same reason that fattening any top-level namespace is a bad 
idea.  I mean, why not just put all the Linux libraries in /lib where 
they expect to be?

>From the perspective of an integrated namespace, we've already made the 
wrong moves insofar as vm.* should be kern.vm.*, vfs.* should be 
kern.vfs.*, etc.  Either the entire kernel namespace should have a 
presumed leading kern. (and the existing kern.* nodes need to move) or 
we should relocate stuff to reflect a more ubuquitous naming 
arrangement.

(btw, you're not the "father" of sysctl.  I might go for "perpetrator" 
 or "culprit" though.)

-- 
\\  The mind's the standard   \\  Mike Smith
\\  of the man.   \\  [EMAIL PROTECTED]
\\-- Joseph Merrick   \\  [EMAIL PROTECTED]




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Jordan K. Hubbard]

I kinda like the idea of a top-level compat category; it will keep the
top level uncluttered when sysv and iBCS compatibility start requiring
their own knobs, and if you put linux at the top level this will later
be used as justification for putting all the other "compat" stuff up
there too.  I think it's a slippery slope.

- Jordan


> In message <[EMAIL PROTECTED]>, Marcel Moolenaar writes:
> >Hi,
> >
> >There're a couple of variables in the Linuxulator that can be put under
> >sysctl. These include the kernel version and the OSS version, among
> >probably others.
> >
> >The question is simply were in the MIB to put them?
> >1) under "kern.linux"
> >2) under "kern.emu.linux"
> >3) under "linux"
> 
> I vote for 3.
> 
> --
> Poul-Henning Kamp FreeBSD coreteam member
> [EMAIL PROTECTED]   "Real hackers run -current on their laptop."
> FreeBSD -- It will take a long time before progress goes too far!
> 
> 
> To Unsubscribe: send mail to [EMAIL PROTECTED]
> with "unsubscribe freebsd-current" in the body of the message



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Steve Kargl]

Mike Smith wrote:
> > > Yes, this is very true.  But I think we are fooling ourselves if we 
> > > believe linux emulation will not become 'standard' in the near future.
> > > Then we'll kick ourselves for giving the sysctl's convoluted names :-)
> > 
> > Yeah... Then, the next in line after "linux" are: ibcs2 and svr4 and
> > whatever comes next. Can you live with them as main sysctl categories?
> 
> Adding anything at the top level would be a terrible mistake.
> 
> Given that "ABI" is a bit obscure, kern.compat is the only sensible 
> choice.
> 

kern.modules seems to be slightly more general in that you can
have kern.modules.xxx where xxx is anything under /modules that
needs/wants to some tuning via sysctl.

-- 
Steve


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Poul-Henning Kamp]

In message <[EMAIL PROTECTED]>, Mike Smith writes:
>> > Yes, this is very true.  But I think we are fooling ourselves if we 
>> > believe linux emulation will not become 'standard' in the near future.
>> > Then we'll kick ourselves for giving the sysctl's convoluted names :-)
>> 
>> Yeah... Then, the next in line after "linux" are: ibcs2 and svr4 and
>> whatever comes next. Can you live with them as main sysctl categories?

>Given that "ABI" is a bit obscure, kern.compat is the only sensible 
>choice.

I think that is too obscure considering the exposure this will get.
It doesn't really matter much what we feel about it, linux will be
a native and 100% normal binary format for us, if we try to
marginalize it we loose in perception.

We have things to make us posix compatible at the top level already,
I don't see why the linux stuff should live under the top level too.

And as father of sysctl, I think this discussion needs to come to
a close rather than waste more bandwidth, so unless Mike can convince
us why "Adding anything at the top level would be a terrible mistake"
I think the conclusion is "linux.*"

Last call Mike ?

--
Poul-Henning Kamp FreeBSD coreteam member
[EMAIL PROTECTED]   "Real hackers run -current on their laptop."
FreeBSD -- It will take a long time before progress goes too far!


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Mike Smith]

> > Yes, this is very true.  But I think we are fooling ourselves if we 
> > believe linux emulation will not become 'standard' in the near future.
> > Then we'll kick ourselves for giving the sysctl's convoluted names :-)
> 
> Yeah... Then, the next in line after "linux" are: ibcs2 and svr4 and
> whatever comes next. Can you live with them as main sysctl categories?

Adding anything at the top level would be a terrible mistake.

Given that "ABI" is a bit obscure, kern.compat is the only sensible 
choice.

-- 
\\  The mind's the standard   \\  Mike Smith
\\  of the man.   \\  [EMAIL PROTECTED]
\\-- Joseph Merrick   \\  [EMAIL PROTECTED]




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Marcel Moolenaar]

Matthew Dillon wrote:

> :Well, it's also a module, so perhaps we should create the whole subtree
> :for modules (as was already discussed several times..)
> 
> Yes, this is very true.  But I think we are fooling ourselves if we
> believe linux emulation will not become 'standard' in the near future.
> Then we'll kick ourselves for giving the sysctl's convoluted names :-)

I don't think we'll do ourselves a favor if we allow the Linuxulator to be
compiled into the kernel again. To much depends on it to be a module. I
write down "modules.linux" as an option (and a good one too, IMHO).

-- 
Marcel Moolenaar  mailto:[EMAIL PROTECTED]
SCC Internetworking & Databases http://www.scc.nl/
Amsterdam, The Netherlands tel: +31 20 4200655


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Matthew Dillon]

:
:Yeah... Then, the next in line after "linux" are: ibcs2 and svr4 and
:whatever comes next. Can you live with them as main sysctl categories?
:
:Andrzej Bialecki

I think Solaris has a chance, but I doubt any other traditional vendor
UNIXes do.  So it comes down to Solaris and Linux for the most part.

-Matt
Matthew Dillon 
<[EMAIL PROTECTED]>


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Andrzej Bialecki]

On Mon, 16 Aug 1999, Matthew Dillon wrote:

> 
> :
> :On Sun, 15 Aug 1999, Matthew Dillon wrote:
> :
> :> :>2) under "kern.emu.linux"
> :> :>3) under "linux"
> :> :
> :> :I vote for 3.
> :> :
> :> :--
> :> :Poul-Henning Kamp FreeBSD coreteam member
> :> :[EMAIL PROTECTED]   "Real hackers run -current on their laptop."
> :> :FreeBSD -- It will take a long time before progress goes too far!
> :> 
> :> Ditto.  Linux emulation is going to become increasingly important,
> :> burying deep would just make everyone's life more difficult.
> :
> :Well, it's also a module, so perhaps we should create the whole subtree
> :for modules (as was already discussed several times..)
> :
> :Andrzej Bialecki
> 
> Yes, this is very true.  But I think we are fooling ourselves if we 
> believe linux emulation will not become 'standard' in the near future.
> Then we'll kick ourselves for giving the sysctl's convoluted names :-)

Yeah... Then, the next in line after "linux" are: ibcs2 and svr4 and
whatever comes next. Can you live with them as main sysctl categories?

Andrzej Bialecki

//  <[EMAIL PROTECTED]> WebGiro AB, Sweden (http://www.webgiro.com)
// ---
// -- FreeBSD: The Power to Serve. http://www.freebsd.org 
// --- Small & Embedded FreeBSD: http://www.freebsd.org/~picobsd/ 



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Matthew Dillon]

:
:On Sun, 15 Aug 1999, Matthew Dillon wrote:
:
:> :>2) under "kern.emu.linux"
:> :>3) under "linux"
:> :
:> :I vote for 3.
:> :
:> :--
:> :Poul-Henning Kamp FreeBSD coreteam member
:> :[EMAIL PROTECTED]   "Real hackers run -current on their laptop."
:> :FreeBSD -- It will take a long time before progress goes too far!
:> 
:> Ditto.  Linux emulation is going to become increasingly important,
:> burying deep would just make everyone's life more difficult.
:
:Well, it's also a module, so perhaps we should create the whole subtree
:for modules (as was already discussed several times..)
:
:Andrzej Bialecki

Yes, this is very true.  But I think we are fooling ourselves if we 
believe linux emulation will not become 'standard' in the near future.
Then we'll kick ourselves for giving the sysctl's convoluted names :-)

-Matt
Matthew Dillon 
<[EMAIL PROTECTED]>


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Andrzej Bialecki]

On Sun, 15 Aug 1999, Matthew Dillon wrote:

> :>2) under "kern.emu.linux"
> :>3) under "linux"
> :
> :I vote for 3.
> :
> :--
> :Poul-Henning Kamp FreeBSD coreteam member
> :[EMAIL PROTECTED]   "Real hackers run -current on their laptop."
> :FreeBSD -- It will take a long time before progress goes too far!
> 
> Ditto.  Linux emulation is going to become increasingly important,
> burying deep would just make everyone's life more difficult.

Well, it's also a module, so perhaps we should create the whole subtree
for modules (as was already discussed several times..)

Andrzej Bialecki

//  <[EMAIL PROTECTED]> WebGiro AB, Sweden (http://www.webgiro.com)
// ---
// -- FreeBSD: The Power to Serve. http://www.freebsd.org 
// --- Small & Embedded FreeBSD: http://www.freebsd.org/~picobsd/ 



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Andrew Gallatin]

Brian F. Feldman writes:
 > On Sun, 15 Aug 1999, Warner Losh wrote:
 > 
 > > In message <[EMAIL PROTECTED]> "Brian F. 
 >Feldman" writes:
 > > : I suppose, but wouldn't the proper place be under machdep? I agree that
 > > : a linux top-level MIB would be easiest to remember.
 > > 
 > > Linux isn't machdep.  It is MI since we could have Linux/Alpha or
 > > Linux/MIPS emulators...
 > 
 > Well then, we need to move it a directory level and take out machine
 > dependencies, and make it machine independent.
 > 

Linux is actually a little bit machdep:  we can't move it all up a
directory level because most of the syscall numbers, as well as many
flags (to mmap, ioctl, etc) are different between linux/i386 &
linux/alpha.  They bootstrapped themselves off of osf/1 & never
reverted back to their own flags/syscall numbers, they just grew a new
set instead. So I'd suggest moving everything but linux.h & the
syscall related files up a level.  There will need to be some ifdefs
in the code, but not many.

I've been working a little on getting the Linuxulator running on
FreeBSD/alpha.  Because linux is such a bloody cludge (the syscall &
flags differences), its a bit more difficult that I initially thought
it would be.  The also call some functions (osf1_setsysinfo, for
example) which are really osf/1 calls.  Luckily, I have my osf/1
compat code which I'm using to field these.

Right now it works well enough to install linux_base from ports (good
job in keeing it MI, Marcel!) & run things like ls, uname, etc.
However, its not useful for anything real just yet.  My goal is to get
em86 (the x86 emulator that lets linux/alpha run linux/i386 binaries)
running, as well as the Compaq compilers.

Drew

PS: Is anybody interested in reviewing my osf/1 compat code so that it
can be comitted?  It works quite well (SimOS, Netscape, Mathematica,
Matlab, S-Plus, emacs, etc all run), but I'd really like somebody to
look it over before I commit it. I've left it at 
http://www.freebsd.org/~gallatin/osf1.tar.gz

--
Andrew Gallatin, Sr Systems Programmer  http://www.cs.duke.edu/~gallatin
Duke University Email: [EMAIL PROTECTED]
Department of Computer Science  Phone: (919) 660-6590










To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Marcel Moolenaar]

"Brian F. Feldman" wrote:

> > : I suppose, but wouldn't the proper place be under machdep? I agree that
> > : a linux top-level MIB would be easiest to remember.
> >
> > Linux isn't machdep.  It is MI since we could have Linux/Alpha or
> > Linux/MIPS emulators...
> 
> Well then, we need to move it a directory level and take out machine
> dependencies, and make it machine independent.

A split of the Linux sources has been discussed already and isn't the issue
here. Please stay on topic or change the subject.

-- 
Marcel Moolenaar  mailto:[EMAIL PROTECTED]
SCC Internetworking & Databases http://www.scc.nl/
Amsterdam, The Netherlands tel: +31 20 4200655


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Brian F. Feldman]

On Sun, 15 Aug 1999, Warner Losh wrote:

> In message <[EMAIL PROTECTED]> "Brian F. 
>Feldman" writes:
> : I suppose, but wouldn't the proper place be under machdep? I agree that
> : a linux top-level MIB would be easiest to remember.
> 
> Linux isn't machdep.  It is MI since we could have Linux/Alpha or
> Linux/MIPS emulators...

Well then, we need to move it a directory level and take out machine
dependencies, and make it machine independent.

> 
> Warner
> 

 Brian Fundakowski Feldman  _ __ ___   ___ ___ ___  
 [EMAIL PROTECTED]   _ __ ___ | _ ) __|   \ 
 FreeBSD: The Power to Serve!_ __ | _ \._ \ |) |
   http://www.FreeBSD.org/  _ |___/___/___/ 



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Warner Losh]

In message <[EMAIL PROTECTED]> Mike Smith writes:
: We're staying away from the term "emulation" because it's being 
: associated with things like the abominable 'lxrun' and virtual-machine 
: emulators like VMware.

Also, there is a perception that "emulation" is slower than native,
which isn't the case for the Linux ABI in FreeBSD.

Warner



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Warner Losh]

In message <[EMAIL PROTECTED]> "Brian F. 
Feldman" writes:
: I suppose, but wouldn't the proper place be under machdep? I agree that
: a linux top-level MIB would be easiest to remember.

Linux isn't machdep.  It is MI since we could have Linux/Alpha or
Linux/MIPS emulators...

Warner


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Mike Smith]

> Hi,
> 
> There're a couple of variables in the Linuxulator that can be put under
> sysctl. These include the kernel version and the OSS version, among
> probably others.
> 
> The question is simply were in the MIB to put them?
> 1) under "kern.linux"
> 2) under "kern.emu.linux"
> 3) under "linux"
> 4) non of the above, because...

kern.abi.linux or kern.compat.linux.

We're staying away from the term "emulation" because it's being 
associated with things like the abominable 'lxrun' and virtual-machine 
emulators like VMware.

-- 
\\  The mind's the standard   \\  Mike Smith
\\  of the man.   \\  [EMAIL PROTECTED]
\\-- Joseph Merrick   \\  [EMAIL PROTECTED]




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Poul-Henning Kamp]

In message <[EMAIL PROTECTED]>, "Brian F. 
Feldman" writes:
>On Sun, 15 Aug 1999, Poul-Henning Kamp wrote:
>
>> >The question is simply were in the MIB to put them?
>>...
>> >3) under "linux"
>> 
>> I vote for 3.
>
>I suppose, but wouldn't the proper place be under machdep? I agree that
>a linux top-level MIB would be easiest to remember.

We may have to emulate linux on other platforms as well...

--
Poul-Henning Kamp FreeBSD coreteam member
[EMAIL PROTECTED]   "Real hackers run -current on their laptop."
FreeBSD -- It will take a long time before progress goes too far!


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Brian F. Feldman]

On Sun, 15 Aug 1999, Poul-Henning Kamp wrote:

> >The question is simply were in the MIB to put them?
>...
> >3) under "linux"
> 
> I vote for 3.

I suppose, but wouldn't the proper place be under machdep? I agree that
a linux top-level MIB would be easiest to remember.

> 
> --
> Poul-Henning Kamp FreeBSD coreteam member
> [EMAIL PROTECTED]   "Real hackers run -current on their laptop."
> FreeBSD -- It will take a long time before progress goes too far!

 Brian Fundakowski Feldman  _ __ ___   ___ ___ ___  
 [EMAIL PROTECTED]   _ __ ___ | _ ) __|   \ 
 FreeBSD: The Power to Serve!_ __ | _ \._ \ |) |
   http://www.FreeBSD.org/  _ |___/___/___/ 



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Matthew Dillon]

:>2) under "kern.emu.linux"
:>3) under "linux"
:
:I vote for 3.
:
:--
:Poul-Henning Kamp FreeBSD coreteam member
:[EMAIL PROTECTED]   "Real hackers run -current on their laptop."
:FreeBSD -- It will take a long time before progress goes too far!

Ditto.  Linux emulation is going to become increasingly important,
burying deep would just make everyone's life more difficult.

-Matt
Matthew Dillon 
<[EMAIL PROTECTED]>


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Q: Extending the sysctl MIB for Linuxulator variables

[Poul-Henning Kamp]

In message <[EMAIL PROTECTED]>, Marcel Moolenaar writes:
>Hi,
>
>There're a couple of variables in the Linuxulator that can be put under
>sysctl. These include the kernel version and the OSS version, among
>probably others.
>
>The question is simply were in the MIB to put them?
>1) under "kern.linux"
>2) under "kern.emu.linux"
>3) under "linux"

I vote for 3.

--
Poul-Henning Kamp FreeBSD coreteam member
[EMAIL PROTECTED]   "Real hackers run -current on their laptop."
FreeBSD -- It will take a long time before progress goes too far!


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message