On 29-May-06, 03:57 (CDT), Ralf Wildenhues <[EMAIL PROTECTED]> wrote:
> FWIW, libtool scripts are a bit more complex. Unrelated though,
> Libtool records the shell and its features; if you change /bin/sh
> from bash to dash, the installed /usr/bin/libtool will have its
> $echo setting wrong, and
Sorry for the late message to this thread.
* Gabor Gombas wrote:
> On Sat, May 20, 2006 at 01:58:14AM +0200, Adam Borowski wrote:
> > Thus, it's bash's start-up which is the slow part, in the terms of
> > actual speed, bash is not that far behind.
>
> It would be interesting to compare something
Michal Politowski <[EMAIL PROTECTED]>:
> On Thu, 18 May 2006 22:38:08 +0200, Goswin von Brederlow wrote:
> [...]
> > 3. Make sh an alternative
>
> dash already optionally diverts it. Isn't it good enough?
Both of these are a really bad idea. If anything goes wrong at the
wrong moment, /bin/sh wo
Hello Gabor,
Gabor Gombas <[EMAIL PROTECTED]> wrote:
> On Sat, May 20, 2006 at 01:58:14AM +0200, Adam Borowski wrote:
>
>> Thus, it's bash's start-up which is the slow part, in the terms of
>> actual speed, bash is not that far behind.
>
> It would be interesting to compare something more complex
On Sat, May 20, 2006 at 01:58:14AM +0200, Adam Borowski wrote:
> The interesting fact is that, contrary to what I expected, running
> "configure" was improved by only about 1% on average.
That may come from the fact that "configure" uses only the most basic
shell constructs (it does not even use
On Tue, May 23, 2006 at 01:19:44PM +0200, Gabor Gombas wrote:
> difference. Some very basic tests using callgrind show that bash uses
> 20-30 times more CPU cache than dash. And when things are running
Er, fat fingers. The difference is just 2-3 times.
Gabor
--
---
On Mon, May 22, 2006 at 01:55:54PM +, Jörg Sommer wrote:
> But what counts more in the comparison dash vs. bash is the shell
> startup. And the shell is started for every script not name foo.sh.
Also, if init scripts will be really parallel (meaning lots of
concurrent scripts, not just 2-3),
gregor herrmann wrote:
> IIRC lintian does this already.
And devscripts contains a 'checkbashishms' tool
--
Felipe Sateler
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Hello Wouter,
Wouter Verhelst <[EMAIL PROTECTED]> wrote:
> On Fri, May 19, 2006 at 04:49:49PM +0200, Olaf van der Spek wrote:
>> On 5/19/06, Wouter Verhelst <[EMAIL PROTECTED]> wrote:
>> Why would they have to work with dash?
>
> If the difference in speed is indeed that insane, that's nice.
>
> I
On Fri, May 19, 2006 at 03:34:17PM +0200, Wouter Verhelst wrote:
> On Fri, May 19, 2006 at 05:45:46AM +0200, David Weinehall wrote:
> > Well, most of those scripts can be fixed quite easily, some require
> > a bit more work. I hereby promise to help fixing them to the extent
> > of my capability.
On Fri, May 19, 2006 at 04:49:49PM +0200, Olaf van der Spek wrote:
> On 5/19/06, Wouter Verhelst <[EMAIL PROTECTED]> wrote:
> >On Fri, May 19, 2006 at 05:45:46AM +0200, David Weinehall wrote:
> >> Well, most of those scripts can be fixed quite easily, some require
> >> a bit more work. I hereby pr
On May 19, Lars Wirzenius <[EMAIL PROTECTED]> wrote:
> > Some may need bash for a reason. Making them more complex just to have
> > them work with dash is not useful.
> If a script really needs bash, then starting it with #!/bin/bash is
> perfectly acceptable. I doubt there are very many such scri
On Thu, May 18, 2006 at 05:27:23PM -0300, Margarita Manterola wrote:
> During some tests I've performed, I've found that making the init
> scripts run with dash as default shell instead of bash makes the boot
> time a 10% faster (6 seconds in a 60 second boot).
This speed-up is not limited to boot
pe, 2006-05-19 kello 09:35 +0200, Marco d'Itri kirjoitti:
> On May 19, David Weinehall <[EMAIL PROTECTED]> wrote:
>
> > > That would mean having 2 shells since some scripts need bash. What a
> > > waste on small systems.
> > Well, most of those scripts can be fixed quite easily, some require
> > a
[Loïc Minier]
> Right now, if a script is named ".sh", it is sourced, perhaps it's
> enough to change /etc/init.d/rc to use dash, depend on dash there
> (sysv-rc), and progressively convert init scripts to use a symlink
> ending in .sh when they are POSIX?
Actually, if we want to run the scrip
On Fri, 19 May 2006, Goswin von Brederlow wrote:
> Henrique de Moraes Holschuh <[EMAIL PROTECTED]> writes:
> > On Fri, 19 May 2006, Goswin von Brederlow wrote:
> >> Alternatives are more suited for cases where one binary is provided by
> >> multiple packages. Currently we have bash, dash, sash, pos
On Friday 19 May 2006 05:12, Petter Reinholdtsen wrote:
> I was a bit surprised that parallel execution only shaved 2 seconds of
> the boot, but suspect it might be because of inefficient
> implementation in /etc/init.d/rc.
Not really a /etc/init.d/rc fault, is more likely a problem that there is
Gabor Gombas <[EMAIL PROTECTED]> writes:
> On Fri, May 19, 2006 at 03:17:20AM +0200, Goswin von Brederlow wrote:
>
>> Alternatives are more suited for cases where one binary is provided by
>> multiple packages. Currently we have bash, dash, sash, posh. Anything
>> else?
>
> Alternatives break on a
Adam Borowski <[EMAIL PROTECTED]> wrote:
> On Fri, May 19, 2006 at 10:18:46AM +0200, Michal Čihař wrote:
>> There are anyway users using dash as /bin/sh right now and broken
>> packages are bugged, so switching default should not reveal any new bug
>
> The policy says:
> # If a script requires non
On Fri, May 19, 2006 at 03:17:20AM +0200, Goswin von Brederlow wrote:
> Alternatives are more suited for cases where one binary is provided by
> multiple packages. Currently we have bash, dash, sash, posh. Anything
> else?
Alternatives break on a daily basis, I wouldn't trust them for something
a
On 5/19/06, Wouter Verhelst <[EMAIL PROTECTED]> wrote:
On Fri, May 19, 2006 at 05:45:46AM +0200, David Weinehall wrote:
> Well, most of those scripts can be fixed quite easily, some require
> a bit more work. I hereby promise to help fixing them to the extent
> of my capability.
Let's see. The
Wouter Verhelst <[EMAIL PROTECTED]> writes:
> On Fri, May 19, 2006 at 05:45:46AM +0200, David Weinehall wrote:
>> Well, most of those scripts can be fixed quite easily, some require
>> a bit more work. I hereby promise to help fixing them to the extent
>> of my capability.
>
> Let's see. The nbd-
On May 19, Adam Borowski <[EMAIL PROTECTED]> wrote:
> Besides, here's an idea:
> let's make it karma-mandatory for debian-devel readers to have /bin/sh point
> to foosh for sh!="ba". The more people people use alternate shells, the
> faster bugs are exposed.
Good idea, just don't try it with posh
On Fri, May 19, 2006 at 10:18:46AM +0200, Michal Čihař wrote:
> There are anyway users using dash as /bin/sh right now and broken
> packages are bugged, so switching default should not reveal any new bug
The policy says:
# If a script requires non-POSIX features from the shell interpreter, the
# a
Goswin von Brederlow wrote:
> "Margarita Manterola" <[EMAIL PROTECTED]> writes:
>
>> During some tests I've performed, I've found that making the init
>> scripts run with dash as default shell instead of bash makes the boot
>> time a 10% faster (6 seconds in a 60 second boot).
>>
>> To make this
On Fri, May 19, 2006 at 09:44:45AM +0200, Loïc Minier wrote:
> On Thu, May 18, 2006, Margarita Manterola wrote:
> > During some tests I've performed, I've found that making the init
> > scripts run with dash as default shell instead of bash makes the boot
> > time a 10% faster (6 seconds in a 60 se
On Fri, May 19, 2006 at 02:05:12AM +0200, Marco d'Itri wrote:
> On May 19, Wouter Verhelst <[EMAIL PROTECTED]> wrote:
>
> > Since bash does enable some features that are not specified in
> > POSIX, even when called as /bin/sh, I don't see what the problem
> > would be of installing "something else
On Fri, May 19, 2006 at 05:45:46AM +0200, David Weinehall wrote:
> Well, most of those scripts can be fixed quite easily, some require
> a bit more work. I hereby promise to help fixing them to the extent
> of my capability.
Let's see. The nbd-client and nbd-server initscripts use bash arrays. Do
Henrique de Moraes Holschuh <[EMAIL PROTECTED]> writes:
> On Fri, 19 May 2006, Goswin von Brederlow wrote:
>> Alternatives are more suited for cases where one binary is provided by
>> multiple packages. Currently we have bash, dash, sash, posh. Anything
>> else?
>
> Are you prepared to put your li
Hi
On Fri, 19 May 2006 10:12:57 +0200
Petter Reinholdtsen <[EMAIL PROTECTED]> wrote:
> I believe this is a fairly safe thing to do, but believe it should be
> done shortly after etch releases, and not just a few months before we
> freeze for etch. There are other things we can manage before etch
[Margarita Manterola]
> During some tests I've performed, I've found that making the init
> scripts run with dash as default shell instead of bash makes the boot
> time a 10% faster (6 seconds in a 60 second boot).
I saw (parts of) your talk at debconf on this topic, and was happy to
see the resu
Jeroen van Wolffelaar wrote:
> Of course, all the packages in question must then depend on dash, but
> that's not really a problem IMHO. The only problem I do see is that dash
> currently asks a question upon installation, which I think it should
> simply stop doing so, administrators who want this
Olaf van der Spek wrote:
> Can't you only change the symlink if/when dash is installed?
Yes, dash manages /bin/sh in a sensible manner.
dpkg-reconfigure dash to switch the symlink to it.
--
see shy jo
signature.asc
Description: Digital signature
Margarita Manterola wrote:
> Option (1) implies making dash base and Essential: yes (currently dash
> is optional), and that would imply that until no Essential package
> depends on bash, we would have two shells in Essential.
Since initramfs-tools, yaird, and initrd-tools all depend on dash
anywa
On May 19, David Weinehall <[EMAIL PROTECTED]> wrote:
> > That would mean having 2 shells since some scripts need bash. What a
> > waste on small systems.
> Well, most of those scripts can be fixed quite easily, some require
> a bit more work. I hereby promise to help fixing them to the extent
>
On Thu, May 18, 2006, Margarita Manterola wrote:
> During some tests I've performed, I've found that making the init
> scripts run with dash as default shell instead of bash makes the boot
> time a 10% faster (6 seconds in a 60 second boot).
[...]
> 2. Change #!/bin/sh for #!/bin/dash in the script
On Thu, May 18, 2006 at 10:38:08PM +0200, Goswin von Brederlow wrote:
> "Margarita Manterola" <[EMAIL PROTECTED]> writes:
>
> > During some tests I've performed, I've found that making the init
> > scripts run with dash as default shell instead of bash makes the boot
> > time a 10% faster (6 secon
On Fri, 19 May 2006, Goswin von Brederlow wrote:
> Alternatives are more suited for cases where one binary is provided by
> multiple packages. Currently we have bash, dash, sash, posh. Anything
> else?
Are you prepared to put your life on the line that the alternatives system
will never, ever leav
On Thu, 18 May 2006, Florent Bayle wrote:
> Why no managing /bin/sh link with update-alternatives ?
Because it is essential.
--
"One disk to rule them all, One disk to find them. One disk to bring
them all and in the darkness grind them. In the Land of Redmond
where the shadows lie." -- Th
On Thu, May 18, 2006 at 05:27:00PM -0300, Margarita Manterola wrote:
> 1. Make /bin/sh point to /bin/dash
The biggest problem with this seems to be scripts that expect /bin/sh to
point to /bin/bash. Arguably those scripts are broken, but there exists
another choice that avoids that unnecessary
Michal Politowski <[EMAIL PROTECTED]> writes:
> On Thu, 18 May 2006 22:38:08 +0200, Goswin von Brederlow wrote:
> [...]
>> 3. Make sh an alternative
>
> dash already optionally diverts it. Isn't it good enough?
When I upload my fash (fast shell) package that would want to divert
sh too and then c
On May 19, Wouter Verhelst <[EMAIL PROTECTED]> wrote:
> Since bash does enable some features that are not specified in POSIX,
> even when called as /bin/sh, I don't see what the problem would be of
> installing "something else" as our default /bin/sh (ignoring the fact
That it would break all thes
On Thu, May 18, 2006 at 11:29:47PM +0200, Jeroen van Wolffelaar wrote:
> On Thu, May 18, 2006 at 05:27:23PM -0300, Margarita Manterola wrote:
> > This option also might imply some extra bugs, but it's believed that
> > not so many, since there are already quite a number of people with
> > /bin/sh -
On Thu, May 18, 2006 at 05:27:23PM -0300, Margarita Manterola wrote:
> Also, I've heard other options being posed, such as writing all init
> scripts (including /etc/init.d/rc) in python. But I do want to
> concentrate on what's possible to do for etch or etch+1.
If you're going to use a real scr
On May 18, Gustavo Franco <[EMAIL PROTECTED]> wrote:
> Good, but i think interested people here are all our users, so what
> we're going to do with the 10% performance hit for use bash? I think
Nothing. A 10% optimization of init scripts which we already know
are highly suboptimal is not worth pur
On Thu, 18 May 2006 22:38:08 +0200, Goswin von Brederlow wrote:
[...]
> 3. Make sh an alternative
dash already optionally diverts it. Isn't it good enough?
--
Michał Politowski
Talking has been known to lead to communication if practiced carelessly.
On 5/18/06, Marco d'Itri <[EMAIL PROTECTED]> wrote:
On May 18, Margarita Manterola <[EMAIL PROTECTED]> wrote:
> To make this speed up available to everyone, we have 2 main choices:
>
> 1. Make /bin/sh point to /bin/dash
> 2. Change #!/bin/sh for #!/bin/dash in the scripts
We have an even simple
On Thu, May 18, 2006 at 05:27:23PM -0300, Margarita Manterola wrote:
> During some tests I've performed, I've found that making the init
> scripts run with dash as default shell instead of bash makes the boot
> time a 10% faster (6 seconds in a 60 second boot).
>
> To make this speed up available
On May 18, Margarita Manterola <[EMAIL PROTECTED]> wrote:
> To make this speed up available to everyone, we have 2 main choices:
>
> 1. Make /bin/sh point to /bin/dash
> 2. Change #!/bin/sh for #!/bin/dash in the scripts
We have an even simpler choice which already works and does not require
cha
On Thu, May 18, 2006 at 06:10:06PM -0300, Daniel Ruoso wrote:
> > This option also might imply some extra bugs, but it's believed that
> > not so many, since there are already quite a number of people with
> > /bin/sh -> /bin/dash, and they do file bugs when "bashisms" appear.
> A tool searching f
Em Qui, 2006-05-18 às 17:27 -0300, Margarita Manterola escreveu:
> During some tests I've performed, I've found that making the init
> scripts run with dash as default shell instead of bash makes the boot
> time a 10% faster (6 seconds in a 60 second boot).
Nice...
> To make this speed up availab
"Margarita Manterola" <[EMAIL PROTECTED]> writes:
> During some tests I've performed, I've found that making the init
> scripts run with dash as default shell instead of bash makes the boot
> time a 10% faster (6 seconds in a 60 second boot).
>
> To make this speed up available to everyone, we hav
Le Jeudi 18 Mai 2006 22:31, Olaf van der Spek a écrit :
> On 5/18/06, Margarita Manterola <[EMAIL PROTECTED]> wrote:
> > During some tests I've performed, I've found that making the init
> > scripts run with dash as default shell instead of bash makes the boot
> > time a 10% faster (6 seconds in a
On 5/18/06, Margarita Manterola <[EMAIL PROTECTED]> wrote:
During some tests I've performed, I've found that making the init
scripts run with dash as default shell instead of bash makes the boot
time a 10% faster (6 seconds in a 60 second boot).
To make this speed up available to everyone, we ha
During some tests I've performed, I've found that making the init
scripts run with dash as default shell instead of bash makes the boot
time a 10% faster (6 seconds in a 60 second boot).
To make this speed up available to everyone, we have 2 main choices:
1. Make /bin/sh point to /bin/dash
2. Ch
55 matches
Mail list logo