FAQ Search Today's Posts Mark Forums Read
» Video Reviews

» Linux Archive

Linux-archive is a website aiming to archive linux email lists and to make them easily accessible for linux users/developers.


» Sponsor

» Partners

» Sponsor

Go Back   Linux Archive > Debian > Debian Development

 
 
LinkBack Thread Tools
 
Old 10-19-2011, 07:24 AM
Martin Wuertele
 
Default RFC: usb-modeswitch 1.2.0 release embedding jimtcl

* Roland Mas <lolando@debian.org> [2011-10-18 18:46]:

> > I'm not even sure network-manager is needed on the desktop install,
> > but eh.
>
> $ LANG=C aptitude why network-manager
> i gnome-core Depends network-manager-gnome (>= 0.8.999)
> i A network-manager-gnome Depends network-manager (>= 0.9)

That's the reason why it gets pulled in, however that's not a reason why
it is needed on desktop installations.

Yours Martin


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 20111019072448.GC10604@anguilla.debian.or.at">http ://lists.debian.org/20111019072448.GC10604@anguilla.debian.or.at
 
Old 10-19-2011, 09:22 AM
Adam Borowski
 
Default RFC: usb-modeswitch 1.2.0 release embedding jimtcl

On Tue, Oct 18, 2011 at 06:46:26PM +0200, Roland Mas wrote:
> Yves-Alexis Perez, 2011-10-18 18:35:04 +0200 :
> > I'm not even sure network-manager is needed on the desktop install,
> > but eh.
>
> $ LANG=C aptitude why network-manager
> i gnome-core Depends network-manager-gnome (>= 0.8.999)
> i A network-manager-gnome Depends network-manager (>= 0.9)

And since network-manager disallows even basic networking above "a single
interface with DHCP", this makes the gnome-core metapackage useless.

I'm not talking about router setups. Try a regular desktop with IPv6 that
runs virtual machines (VirtualBox, lxc) and has usb networking[1] enabled so
I can copy things to/from the phone ten times faster than over wifi. You
can't have any of this with n-m.

Also, http://bugs.debian.org/542095 -- this came up already before and after
a brief flamewar has been promptly fixed. Yet since a few days ago the hard
dependency on n-m is back.



[1]. Many thanks to new udev which makes the usb0 interface work out of the
box, including hotplugging.

--
1KB // Yo momma uses IPv4!
 
Old 10-19-2011, 12:17 PM
Didier Raboud
 
Default RFC: usb-modeswitch 1.2.0 release embedding jimtcl

Hi all again,

first of all, thanks for the healthy discussion.

Now, given the feedback, I plan to go with the brand new option 6 which is an
option 2, "done right" (eh, IMHO).

6) "Allow interpretation using separate libjim, from jimtcl"
This means packaging jimtcl and allow the usb-modeswitch-dispatcher to be
dynamically linked against libjim.
(That, plus "repacking to avoid the embedded jimtcl copy")
Pros: relatively easy, avoids the binary embedding of jimtcl.
Cons: replaces the need of the desktop install on a "tcl interpreter" to
"libjim"; it weighting (with default options that are more than what
is needed by the usb-modeswitch-dispatcher) 268k.

As far as I could see, it's possible to win 32k by providing a libjim-tiny
alternative library, but it doesn't seem worth it.

So, ITP on its way I guess.

Cheers,

OdyX


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 201110191417.12386.odyx@debian.org">http://lists.debian.org/201110191417.12386.odyx@debian.org
 
Old 10-20-2011, 05:27 AM
Philipp Matthias Hahn
 
Default RFC: usb-modeswitch 1.2.0 release embedding jimtcl

Hello,

On Wed, Oct 19, 2011 at 01:48:17PM +0100, Ben Hutchings wrote:
> Virtual machine networking works just fine for me with virt-manager and
> network-manager.

As far as I know not (fully) for the bridged setup [0]: NM recognized
the bridge, but since it doesn't have a "link status" [1], VPNs are
disabled.

Background: I have one desktop PC which is my main work station at home,
which alse can run several VMs. Since I also own other mobile devices, I
like to connect from them to the VMs.
ut I also use OpenVPN to open tunnels to other sides on demand.

[0] A dedicated bridge for all VMs works fine, as then the ethernet
interface is fully handled by NM itself.

[1] Only the physical ethernet card has a link status, which is part of
the bridge, but NM doesn't handle this.

BYtE
Philipp
--
Philipp Matthias Hahn <pmhahn@debian.org>
GPG/PGP: 9A540E39 @ keyrings.debian.org


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 20111020052733.GA4709@pmhahn.de">http://lists.debian.org/20111020052733.GA4709@pmhahn.de
 
Old 10-20-2011, 07:34 AM
Timo Jyrinki
 
Default RFC: usb-modeswitch 1.2.0 release embedding jimtcl

2011/10/18 Yves-Alexis Perez <corsac@debian.org>:
> Not sure it'd help in this case but imho this is the case only on
> laptops. People installing desktop usually don't need a 3G modem (and if
> they do, they can manually install it). I'm not even sure
> network-manager is needed on the desktop install, but eh.

Actually, I already know a lot of people whose only desktop machine
network connection is a 3G dongle, since unlimited data with it is
about half the price of ADSL connection.

-Timo


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: CAJtFfxkSdHPd=Vo84HKbwmO3oBKMor3zvBo8aruAERKjrE+rs A@mail.gmail.com">http://lists.debian.org/CAJtFfxkSdHPd=Vo84HKbwmO3oBKMor3zvBo8aruAERKjrE+rs A@mail.gmail.com
 
Old 10-20-2011, 07:58 AM
Russell Coker
 
Default RFC: usb-modeswitch 1.2.0 release embedding jimtcl

On Thu, 20 Oct 2011, Timo Jyrinki <timo.jyrinki@gmail.com> wrote:
> Actually, I already know a lot of people whose only desktop machine
> network connection is a 3G dongle, since unlimited data with it is
> about half the price of ADSL connection.

My parents have been using USB 3G net access on their Desktop system for about
two years now. It costs them $150 per annum instead of $30 per month. If
they were to stop using a land-line phone service and use only mobile phones
then they could save another $20 per month.

In Australia ADSL and cable are only good value in urban areas if you need to
transfer more than about 2G of data per month or if you need reliable low
latency.

--
My Main Blog http://etbe.coker.com.au/
My Documents Blog http://doc.coker.com.au/


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 201110201858.45111.russell@coker.com.au">http://lists.debian.org/201110201858.45111.russell@coker.com.au
 
Old 10-20-2011, 12:53 PM
Yves-Alexis Perez
 
Default RFC: usb-modeswitch 1.2.0 release embedding jimtcl

On jeu., 2011-10-20 at 18:58 +1100, Russell Coker wrote:
> On Thu, 20 Oct 2011, Timo Jyrinki <timo.jyrinki@gmail.com> wrote:
> > Actually, I already know a lot of people whose only desktop machine
> > network connection is a 3G dongle, since unlimited data with it is
> > about half the price of ADSL connection.
>
> My parents have been using USB 3G net access on their Desktop system for about
> two years now. It costs them $150 per annum instead of $30 per month. If
> they were to stop using a land-line phone service and use only mobile phones
> then they could save another $20 per month.
>
> In Australia ADSL and cable are only good value in urban areas if you need to
> transfer more than about 2G of data per month or if you need reliable low
> latency.

I don't exactly want to debate on that. I never said nobody used that,
it's just that, imho, that's not really needed by default. Now if we
consider people using default install don't know how to install more
stuff, and use that as a basis for selecting what's in the default
install, I think we might have quite some issues.

Regards,
--
Yves-Alexis
 
Old 08-27-2012, 05:21 PM
Andrew Shadura
 
Default RFC: usb-modeswitch 1.2.0 release embedding jimtcl

Hello,

(As a Tcler I have to comment on this.)

On Tue, 18 Oct 2011 13:36:43 +0200
Didier Raboud <odyx@debian.org> wrote:

> 1) "Forget about jimtcl, rely on existing tcl interpreters"

> This is mostly "repacking to avoid the embedded jimtcl copy", "no
> packaging of it, go on as is done currently; by relying on
> existing tcl interpreters.
> Pros: easy, straightforward,avoids the binary embedding of jimtcl.
> Cons: does not solve the "desktop install needs tcl interpreter".

Jimsh is already available, and can be used separately. Also, libjim
allows linking dynamically. And also, jim and tcl are a bit different,
so it's not always jim-based script is able to run in plain tclsh
without additional shims.

> 2) "Allow interpretation using separate jimtcl"

> This means packaging jimtcl and allow usb-modeswitch to depend on
> it (That, plus "repacking to avoid the embedded jimtcl copy")
> Pros: relatively easy, avoids the binary embedding of jimtcl.
> Cons: replaces the need of the desktop install on a "tcl
> interpreter" to "jimtcl". Although it's probably smaller.

Already packaged, see above.

> 3) "Embed jimtcl using the internal copy"

> This means taking the upstream tarball as is.
> Pros: small standalone -dispatcher binary.
> Cons: code duplication, potential security issues with
> out-of-date jimtcl versions, …

I see no problems with this, if there's just one or two packages
linking against libjim statically.

> 4) "Embed jimtcl using a standalone package"

> This means packaging jimtcl and do some build-time trickery to
> include the jimtcl static library (if possible, only the needed
> parts) into usb- -modeswitch-dispatcher.
> Pros: small standalone -dispatcher binary, no code duplication.
> Cons: binNMU needed at each jimtcl upgrade, static linkeage.

Same as above.

> 5) "Rewrite the usb-modeswitch-dispatcher in C"

> This work has already been done by Mathieu Trudel-Lapierre for the
> Ubuntu ackage.
> For now, the upstream developer hasn't included this rewrite into
> the upstream package (for his own set of reasons). My current
> strategy is to avoid as much as possible to diverge from upstream,
> hence why it's not in Debian's usb-modeswitch for now.
> Pros: No tcl interpreter needed.
> Cons: as it's not an upstream effort, it can become out-of-sync
> in terms of functionality and bugfixes (and indeed currently is as of
> 1.2.0~beta).

Stupid and useless. Usb-modeswitch was originally written in C, and
later rewritten in Tcl partially, as it was very hard to maintain it.
What's wrong with having a minimalist tcl interpreter? It's no bigger
than bash, and actually much smaller, and it's faster and doesn't rely
on coreutils.

> What's your opinion ?

Just link it against libjim, statically or dynamically.

--
WBR, Andrew
 
Old 08-27-2012, 05:24 PM
Andrew Shadura
 
Default RFC: usb-modeswitch 1.2.0 release embedding jimtcl

Hello,

On Tue, 18 Oct 2011 10:59:36 -0400
Mathieu Trudel-Lapierre <mathieu-tl@ubuntu.com> wrote:

> It also doesn't solve a second case we're trying to cover: the fact
> that usb-modeswitch would be the only package in the boot path on
> *Ubuntu* that would rely on Tcl. That's another reason why a compiled
> language was chosen.

Please get ready: there will be one more.

> > 2) "Allow interpretation using separate jimtcl"

> Sounds like a good idea to ship jimtcl separately anyway. That said,
> the comments above apply again.

There's jim already packaged, as a library and as an interpreter!

> > * *For now, the upstream developer hasn't included this rewrite
> > into the upstream package (for his own set of reasons). My current
> > strategy is to avoid as much as possible to diverge from upstream,
> > hence why it's not in Debian's usb-modeswitch for now.

> Yup, it's already out-of-sync, though I'll try to get this fixed in
> the next two weeks. I've also sent another email to upstream about
> including the "rewrite". The end goal would be to have a tarball that
> provides both options: a tcl version and a C version of the
> -dispatcher code. The version to use could be chosen at build time.

Why do you need this? What's wrong with having an ultrasmall
interpreter in default Ubuntu, which provides more features than bash,
which is much faster and much smaller?

Also, you're redoing upstream's work in an absolutely opposite
direction: they've moved away from C, and you're bringing it back!

> I'm obviously all for this option, but I agree it would be much better
> if it was included in the tarball.

No. Just keep it as is.

--
WBR, Andrew
 
Old 08-27-2012, 07:46 PM
Steve Langasek
 
Default RFC: usb-modeswitch 1.2.0 release embedding jimtcl

On Mon, Aug 27, 2012 at 07:21:22PM +0200, Andrew Shadura wrote:
> > 3) "Embed jimtcl using the internal copy"

> > This means taking the upstream tarball as is.
> > Pros: small standalone -dispatcher binary.
> > Cons: code duplication, potential security issues with
> > out-of-date jimtcl versions, …

> I see no problems with this, if there's just one or two packages
> linking against libjim statically.

Please familiarize yourself with Debian policy on static linking. It's not
ok to use static linking for such things, because it *is* a burden for the
security team, even if it's only done in one or two packages.

> > 5) "Rewrite the usb-modeswitch-dispatcher in C"
>
> > This work has already been done by Mathieu Trudel-Lapierre for the
> > Ubuntu ackage.
> > For now, the upstream developer hasn't included this rewrite into
> > the upstream package (for his own set of reasons). My current
> > strategy is to avoid as much as possible to diverge from upstream,
> > hence why it's not in Debian's usb-modeswitch for now.
> > Pros: No tcl interpreter needed.
> > Cons: as it's not an upstream effort, it can become out-of-sync
> > in terms of functionality and bugfixes (and indeed currently is as of
> > 1.2.0~beta).

> Stupid and useless. Usb-modeswitch was originally written in C, and
> later rewritten in Tcl partially, as it was very hard to maintain it.
> What's wrong with having a minimalist tcl interpreter? It's no bigger
> than bash, and actually much smaller, and it's faster and doesn't rely
> on coreutils.

The C rewrite, which is included in Ubuntu despite not being accepted into
Debian, addressed both image size issues and boot-time performance issues
with the tcl implementation. I don't know how jimtcl compares, but tclsh
had a measurable impact on boot speed due to its runtime having to be loaded
from disk for a single operation. (There were also some buggy udev rules
involved, but that's a different story.)

This is just one example of the costs of gratuitously different
implementation languages. I'm disappointed that, as the ifupdown
maintainer, you don't recognize that *being different* is a major drag on
the overall maintainability of this package and instead are advocating for
further variety.

On Mon, Aug 27, 2012 at 07:24:06PM +0200, Andrew Shadura wrote:
> On Tue, 18 Oct 2011 10:59:36 -0400
> Mathieu Trudel-Lapierre <mathieu-tl@ubuntu.com> wrote:

> > It also doesn't solve a second case we're trying to cover: the fact
> > that usb-modeswitch would be the only package in the boot path on
> > *Ubuntu* that would rely on Tcl. That's another reason why a compiled
> > language was chosen.

> Please get ready: there will be one more.

I don't think that's at all acceptable. Please reconsider this plan.

> Why do you need this? What's wrong with having an ultrasmall
> interpreter in default Ubuntu, which provides more features than bash,
> which is much faster and much smaller?

> Also, you're redoing upstream's work in an absolutely opposite
> direction: they've moved away from C, and you're bringing it back!

Being upstream does not mean they're always right. Even if they have sound
technical reasons for their decision, it may still not be the right decision
for the downstream. That's why distro patches exist.

--
Steve Langasek Give me a lever long enough and a Free OS
Debian Developer to set it on, and I can move the world.
Ubuntu Developer http://www.debian.org/
slangasek@ubuntu.com vorlon@debian.org
 

Thread Tools




All times are GMT. The time now is 08:29 PM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.
Copyright 2007 - 2008, www.linux-archive.org