Login | Register For Free | Help
Search for: (Advanced)

Mailing List Archive: Linux: Kernel

the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion

 

 

First page Previous page 1 2 3 4 5 6 7 8 9 Next page Last page  View All Linux kernel RSS feed   Index | Next | Previous | View Threaded


andrea at cpushare

Jul 27, 2006, 4:52 AM

Post #101 of 223 (10359 views)
Permalink
Re: the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On Thu, Jul 27, 2006 at 08:56:03AM +0200, Adrian Bunk wrote:
> On Wed, Jul 26, 2006 at 11:17:41PM +0200, andrea [at] cpushare wrote:
> > On Wed, Jul 26, 2006 at 10:50:22PM +0200, Adrian Bunk wrote:
> > > But depending on the nature of the error, the worst case might be the
> > > common case (as I've already explained in another email).
> > >
> > > If you can't ensure the quality of your data, please don't use this data
> > > to wrongly draw any conclusions from them [1].
> >
> > Please read the footer of the KLive pages:
> >
> > "The use of the information and of the software in this website is at
> > your own risk. KLive probably doesn't represent a reliable sample of
> > the real usage of the Linux Kernel."
>
> It was you who wrongly said:
> "With KLive I can attempt to estimate market share of _kernel_ code"
>
> Hadn't you read your own disclaimer?

There is no contradiction in the two statements. To attempt to
estimate something I don't need a reliable sample of the whole
population. Estimation is still a statistical thing. Also I said
attempt to estimate, it doesn't mean I will make it.

If you don't consider those results a positive for reiser4, it can
only mean you expected reiser4 to have a much higher share among the
KLive users. This is obvious.

> Every time someone will repeat the "1:5 ratio for reiser4:ext3 users",
> this will be an additional proof it's really worse than no data.

If they say "1:5 ratio for reiser4:ext3 KLive users" everything will
be correct and nobody can object because it's a fact.

I said myself that I'm no reiserfs user, and I don't plan to become
one any time soon (especially on my production systems), I'm only
reporting plain numbers as KLive measured the stuff. I'm surprised as
much as you are, but then I've to report facts, and not my own
opinions.

As far as I'm concerned the thing I like less of reiser4 is the plugin
thing, I'd be less concerned if that was a microkernel (fuse-like)
userland plugin system. Anyway with time perhaps things can change and
become userland based, and the stuff can be moved into vfs if that
code really belongs there as some kernel developer says. That doesn't
mean reiser4 can't be merged first and the stuff moved into vfs
later. xfs when was merged also pratically rewrote a vfs internally
that was meant to work with irix, if Christoph didn't complain about
xfs being merged, I don't see what's the problem of reiser4 being
merged even if it rewrites some part of vfs. xfs is also still having
various special features like the pinhole one that only belongs to the
vfs instead but nobody complains.

And if reiser4 is really so bad as they say, once people starts losing
data they will spread the word of not using it. As long as it's marked
experimental I don't see a big issue, the wireless driver for broadcom
chip will eat your filesystem too if the reverse engineered dma
operations writes into a buffer header instead of an skb.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


bunk at stusta

Jul 27, 2006, 4:56 AM

Post #102 of 223 (10342 views)
Permalink
Re: the ' 'official' point of view' expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On Thu, Jul 27, 2006 at 01:43:53PM +0200, Luigi Genoni wrote:

> I answered a mail about how klive data should not be took in account, and
> could even be dangerous...

... and you talking about people who might need reiser4 was therefore
a wrong answer.

I never said reiser4 shouldn't be merged, but this wasn't the topic of
this subthread.

cu
Adrian

--

"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


bunk at stusta

Jul 27, 2006, 5:18 AM

Post #103 of 223 (10348 views)
Permalink
Re: the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On Thu, Jul 27, 2006 at 01:52:29PM +0200, andrea [at] cpushare wrote:
> On Thu, Jul 27, 2006 at 08:56:03AM +0200, Adrian Bunk wrote:
>...
> > It was you who wrongly said:
> > "With KLive I can attempt to estimate market share of _kernel_ code"
> >
> > Hadn't you read your own disclaimer?
>
> There is no contradiction in the two statements. To attempt to
> estimate something I don't need a reliable sample of the whole
> population. Estimation is still a statistical thing. Also I said

Sure, you don't need any data to estimate anything.

But if your estimate is based on bad data it sounds better than if it
was based on no data although the value of the estimate is the same.

> attempt to estimate, it doesn't mean I will make it.

Giving low quality raw data and then saying "it wasn't me who did the
estimate" is silly.

> If you don't consider those results a positive for reiser4, it can
> only mean you expected reiser4 to have a much higher share among the
> KLive users. This is obvious.

That's completely wrong.
I consider those results neither positive nor negative for reiser4.

They could only be considered positive if someone expected less than
35 reiser4 users worldwide.

> > Every time someone will repeat the "1:5 ratio for reiser4:ext3 users",
> > this will be an additional proof it's really worse than no data.
>
> If they say "1:5 ratio for reiser4:ext3 KLive users" everything will
> be correct and nobody can object because it's a fact.

And 90% of all people will misread it.

And many people will spread it wrong.

That's exactly where such data is worse than no data.

You could claim you know your the suboptimal quality of the data and
that you had a disclaimer.

But did you read Hans' reaction on your email?

If you are intelligent (which I assume), you should have learned by this
how to not present your data.

> I said myself that I'm no reiserfs user, and I don't plan to become
> one any time soon (especially on my production systems), I'm only
> reporting plain numbers as KLive measured the stuff. I'm surprised as
> much as you are, but then I've to report facts, and not my own
> opinions.
>...

I have no plans to use reiser4 myself (I'm a happy ext2 user).
But I'm not against mergng reiser4, either.

I'll leave this technical discussion to the people who know more about
this.

cu
Adrian

--

"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


cat at zip

Jul 27, 2006, 5:21 AM

Post #104 of 223 (10350 views)
Permalink
Re: the ' 'official' point of view' expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On Thu, Jul 27, 2006 at 12:04:46PM +0200, Adrian Bunk wrote:
> I can't prove that the 1:5 ratio is wrong, but the point is that
> claiming a 1:5 ratio was true based on the klive data is not better than
> claiming it based on no data. But claiming it based on the klive data is
> worse since people like you are getting the wrong impression it was
> based on data that would have the quality for supporting such a
> statement.
>
> The data simply has not the quality for such a statement.
> Please read my two examples in [1] if you want to get an impression why
> such problems can occur.
>
> [1] http://lkml.org/lkml/2006/7/26/203

Also, one may wish to invest in the purchase of 'How to Lie with
Statistics'[1] by Darrel Huff (ISBN 0-393-31072-8). An easy yet
powerful little read.

[1] this is not meant to imply that anyone here is lieing, attempting to
lie, thinking of lieing, wishing they were lieing or having anything to
do with lieing, purposeful or otherwise.

--
"To the extent that we overreact, we proffer the terrorists the
greatest tribute."
- High Court Judge Michael Kirby
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


andrea at cpushare

Jul 27, 2006, 6:10 AM

Post #105 of 223 (10366 views)
Permalink
Re: the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On Thu, Jul 27, 2006 at 02:18:11PM +0200, Adrian Bunk wrote:
> They could only be considered positive if someone expected less than
> 35 reiser4 users worldwide.

The only thing we know for sure is that 35 out of 500 KLive user are
running reiser4, worldwide we have no clue.

> If you are intelligent (which I assume), you should have learned by this
> how to not present your data.

I think readers are intelligent enough too to interpret the KLive data
properly for themself, without you having to prevent their eyes to see
the raw KLive data.

When you tell me how I should not present my data, you're asking me to
censor part or all of the very output of the KLive project. I'd rather
wipe out KLive completely, than to censor it. The way I presented it
was absolutely not biased, if you can make more transparent and
unbiased sql queries than the ones I did, please post them and I'll be
glad to run them.

> [..] (I'm a happy ext2 user).

Oh my, I hope you're only choosing the fs for your own workstation.

Even though I don't pretend to fully understand someone who claims to
be happy with ext2, I've no idea why you hate so much the stuff
running at cpushare.com domain. But if it helps KLive is actually one
of the non commercial projects I'm hosting there, and the only reason
I keep it there, is to be sure not to find it filled by ads.

Now let's try to get some work done instead of only sending emails ;)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


vonbrand at inf

Jul 27, 2006, 6:26 AM

Post #106 of 223 (10372 views)
Permalink
Re: the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

Hans Reiser <reiser [at] namesys> wrote:
> Adrian Bunk wrote:
> >[1] the conclusion itself might or might not be true
> > e.g. there _could_ be an 1:5 ratio between reiser4 and ext3 users
> > but your data is not in any way able to support or reject this
> > statement

> It does however suggest that my surprise at how people at the last
> Linux Conference I went to all seemed to know that there exists a
> Reiser4

I'd be rather surprised to find someone at a Linux conference who hasn't at
least heard of the recurrent flamewars on the topic here...

> may be due to it being more widely used than I would have
> guessed. Maybe there is some coolness factor to having the faster FS
> that you can't get from any Distro that is enough to overcome the hassle
> of compiling reiser4progs and a kernel before inserting the DVD.

Not seen any data backing up the "faster", let alone so much faster that
the hassle (and the risk) would make it worth trying... No, Gentoo folks
don't count, around here they are fond of claiming that their self-compiled
systems are at least twice as fast as a binary distribution with the exact
same software.

> I
> would not have guessed we had 1/5th of ext3's usage even among lkml
> readers.....

I'd guess something of the order of 1/100, for testing purposes and idle
curiosity.

> I guess the market contains more people who like
> technology than I was guessing. Maybe there is a positive word of mouth
> effect going on too.

LKML (and Linux conferences) are exactly the places where you /only/ find
this kind of people...

> It would be nice if SuSE and others at least made it an unsupported
> option at install time. I shall have to find the time to go asking them
> all.....

At least Fedora is trying hard to just follow upstream packages (in this
case, Linus' kernels) with the absolute minimum of local patches. It makes
good sense, as it reduces the up-front work, and (more important) minimizes
the pain when the later official version is somehow incompatible with the
previews.

If you want ReiserFS 4 to get more exposure, do the legwork to get it into
the official kernel. Distributions should be reluctant to pick it up as
long as its status as an official Linux filesystem is in question.
--
Dr. Horst H. von Brand User #22616 counter.li.org
Departamento de Informatica Fono: +56 32 654431
Universidad Tecnica Federico Santa Maria +56 32 654239
Casilla 110-V, Valparaiso, Chile Fax: +56 32 797513
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


vonbrand at inf

Jul 27, 2006, 6:30 AM

Post #107 of 223 (10344 views)
Permalink
Re: the ' 'official' point of view' expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

Luigi Genoni <genoni [at] sns> wrote:

[...]

> Anyway you have a datum.
> Some people need reiser4, period.

Nope. Some people run kernels that include reiser4. That is all you can
infer, and that I knew beforehand. They are at least 35, and that I'd have
guessed in any case.
--
Dr. Horst H. von Brand User #22616 counter.li.org
Departamento de Informatica Fono: +56 32 654431
Universidad Tecnica Federico Santa Maria +56 32 654239
Casilla 110-V, Valparaiso, Chile Fax: +56 32 797513
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


gmu2006 at gmail

Jul 27, 2006, 6:42 AM

Post #108 of 223 (10349 views)
Permalink
Re: the ' 'official' point of view' expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On 7/27/06, Horst H. von Brand <vonbrand [at] inf> wrote:
> Luigi Genoni <genoni [at] sns> wrote:
>
> [...]
>
> > Anyway you have a datum.
> > Some people need reiser4, period.
>
> Nope. Some people run kernels that include reiser4. That is all you can
> infer, and that I knew beforehand. They are at least 35, and that I'd have
> guessed in any case.

35.5 as I'm testing it here on my workstation and it seems to be
faster when you test some things involving many copies of large
multi-level sourcetree directories each 3 to 6GiB big in size.
2.6.18-rc2-mm1 with Reiser4 looks ok so far and I had no sync() OOPS
like the last time with one -mm revision.

speed tells us nothing about reliability of course, but compared to
ext3 with dir_index,sparse_super Reiser4 seems to handle "du -sh" and
"rm -r" much faster and without eating all of the CPU cycles as it
finishes quicker although Reiser4 was meant to be CPU-heavy compared
to ext*/reiserfs3.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


andrea at cpushare

Jul 27, 2006, 6:50 AM

Post #109 of 223 (10340 views)
Permalink
Re: the ' 'official' point of view' expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On Thu, Jul 27, 2006 at 09:30:36AM -0400, Horst H. von Brand wrote:
> Nope. Some people run kernels that include reiser4. That is all you can
> infer, and that I knew beforehand. They are at least 35, and that I'd have

Well, if they were sure they could never get any benefit by reiser4
they wouldn't be testing it in the first place...

But certainly the fact they're testing it, doesn't mean they're
actually going to use it forever. You can monitor the 35 users live
with this link and to see if they increase or decrease over time ;)

http://klive.cpushare.com/?order_by=kernel_group&where_machine=all&branch=all&scheduler=all&smp=all&live=live&ip=all

They're ordered by cumulative uptime and not by the number of
users. For example there are only 10 jfs users but those 10 jfs users
have a much larger average uptime (59 days) so they score much higher
in the cumulative uptime too. Why the average reiser4 uptime is much
lower is unknown, it could be because they shutdown the system by
night, or because they upgrade kernel so frequently, or because the
system crashes.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


bunk at stusta

Jul 27, 2006, 6:58 AM

Post #110 of 223 (10359 views)
Permalink
Re: the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On Thu, Jul 27, 2006 at 03:10:32PM +0200, andrea [at] cpushare wrote:
> On Thu, Jul 27, 2006 at 02:18:11PM +0200, Adrian Bunk wrote:
> > They could only be considered positive if someone expected less than
> > 35 reiser4 users worldwide.
>
> The only thing we know for sure is that 35 out of 500 KLive user are
> running reiser4, worldwide we have no clue.
>
> > If you are intelligent (which I assume), you should have learned by this
> > how to not present your data.
>
> I think readers are intelligent enough too to interpret the KLive data
> properly for themself, without you having to prevent their eyes to see
> the raw KLive data.
>
> When you tell me how I should not present my data, you're asking me to
> censor part or all of the very output of the KLive project. I'd rather
> wipe out KLive completely, than to censor it. The way I presented it
> was absolutely not biased, if you can make more transparent and
> unbiased sql queries than the ones I did, please post them and I'll be
> glad to run them.

What about the following statement:

"Gentoo is 47 times as popular as SuSE among KLive users (a service
offered by a SuSE employee gathering data from many users worldwide)."

Is any part of this information wrong?

Is it therefore OK to put this information to /. ?

Surely not, but you might get the point.

> > [..] (I'm a happy ext2 user).
>
> Oh my, I hope you're only choosing the fs for your own workstation.
>
> Even though I don't pretend to fully understand someone who claims to
> be happy with ext2,

It's stable, and if the machine crashes the fsck is really great (Linus'
tree contains at least a dozen patches that had their temporary home in
my lost+found). (I don't care that the fsck takes 20 minutes on a 250 GB
partition.)

Other people have other preferences and do therefore prefer other
filesystems.

> I've no idea why you hate so much the stuff
> running at cpushare.com domain. But if it helps KLive is actually one
> of the non commercial projects I'm hosting there, and the only reason
> I keep it there, is to be sure not to find it filled by ads.

This isn't against cpushare.com, it's against publishing things people
will easily misinterpret.

The Linux Counter [1] is another non commercial project gathering
similar data (including live data about running kernels and uptimes)
from far more users, and I've already given an example in this thread
how someone could have misinterpreted the data offered there. And if
someone would argue with data from there in a similar easy to
misinterpret way on this list, I'd react similarly.

And another example would be that you could say "one third of all Linux
Counter machines are still running kernels < 2.6". This would be based
on current information submitted automatically from nearly 5000
machines. But people would read this is "one third of all machines are
still running kernels < 2.6". But the data doesn't have the quality for
this generalization. And this is exactly the problem.

> Now let's try to get some work done instead of only sending emails ;)

Agreed. ;-)

cu
Adrian

[1] http://counter.li.org/

--

"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


genoni at sns

Jul 27, 2006, 7:31 AM

Post #111 of 223 (10373 views)
Permalink
Re: the ' 'official' point of view' expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

Since reiser4 is not something enabled by default into a default kernel
distribution, I assume they enabled it knowing what they where doing because
they wanted to use it.



On Thu, July 27, 2006 15:30, Horst H. von Brand wrote:
> Luigi Genoni <genoni [at] sns> wrote:
>
>
> [...]
>
>
>> Anyway you have a datum.
>> Some people need reiser4, period.
>>
>
> Nope. Some people run kernels that include reiser4. That is all you can
> infer, and that I knew beforehand. They are at least 35, and that I'd have
> guessed in any case. --
> Dr. Horst H. von Brand User #22616 counter.li.org
> Departamento de Informatica Fono: +56 32 654431
> Universidad Tecnica Federico Santa Maria +56 32 654239
> Casilla 110-V, Valparaiso, Chile Fax: +56 32 797513
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo [at] vger More majordomo info at
> http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
>
>

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


andrea at cpushare

Jul 27, 2006, 7:45 AM

Post #112 of 223 (10333 views)
Permalink
Re: the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On Thu, Jul 27, 2006 at 03:58:23PM +0200, Adrian Bunk wrote:
> What about the following statement:
>
> "Gentoo is 47 times as popular as SuSE among KLive users (a service
> offered by a SuSE employee gathering data from many users worldwide)."
>
> Is any part of this information wrong?
>
> Is it therefore OK to put this information to /. ?
>
> Surely not, but you might get the point.

I already told you once that such a claim is totally bogus because
KLive _doesn't_ record anything about the _userland_, it only records
kernel stuff.

It's not Dlive (distributon Live), it's KLive and K stands for
_kernel_ not distributions.

Not sure how many times I'll have to remind you about this.


Also note currently the current cumulative uptime ratio is 1:18 not
1:47 and even the ratio of the number of users is 1:42. The only claim
based on actual facts you can try to publish on the press is this:

"The Gentoo kernels are 18 times more popular than the SUSE kernels
among KLive users (a service offered in his spare time by a SUSE
contractor that tries to gather data from many users worldwide)."

You can also add:

"Another data point is that among the KLive users the Fedora kernels
are 24 times less popular then Gentoo ones."

Free to advertise the above if you find it interesting.

> This isn't against cpushare.com, it's against publishing things people
> will easily misinterpret.

I think people deserve to think for themself without you acting as
censorship filter for them.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


bunk at stusta

Jul 27, 2006, 8:05 AM

Post #113 of 223 (10371 views)
Permalink
Re: the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On Thu, Jul 27, 2006 at 04:45:29PM +0200, andrea [at] cpushare wrote:
> On Thu, Jul 27, 2006 at 03:58:23PM +0200, Adrian Bunk wrote:
> > What about the following statement:
> >
> > "Gentoo is 47 times as popular as SuSE among KLive users (a service
> > offered by a SuSE employee gathering data from many users worldwide)."
> >
> > Is any part of this information wrong?
> >
> > Is it therefore OK to put this information to /. ?
> >
> > Surely not, but you might get the point.
>
> I already told you once that such a claim is totally bogus because
> KLive _doesn't_ record anything about the _userland_, it only records
> kernel stuff.
>
> It's not Dlive (distributon Live), it's KLive and K stands for
> _kernel_ not distributions.
>
> Not sure how many times I'll have to remind you about this.
>
>
> Also note currently the current cumulative uptime ratio is 1:18 not
> 1:47 and even the ratio of the number of users is 1:42. The only claim
> based on actual facts you can try to publish on the press is this:

The 1:47 was what I calculated yesterday, but it seems to be 1:42 now.

But I'd measure popularity in current usage, not in cumulative uptime.

> "The Gentoo kernels are 18 times more popular than the SUSE kernels
> among KLive users (a service offered in his spare time by a SUSE
> contractor that tries to gather data from many users worldwide)."
>...

Slightly modified version is now in my .signature (I omitted the
"spare time" part since the information is still true and sounds better
without it - it would be cencorship if you'd try to force me to not
omit it).

cu
Adrian

--

The Gentoo kernels are 42 times more popular than the SUSE kernels
among KLive users (a service by SuSE contractor Andrea Arcangeli that
gathers data about kernels from many users worldwide).

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


kangur at polcom

Jul 27, 2006, 8:39 AM

Post #114 of 223 (10352 views)
Permalink
Re: the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On Wed, 26 Jul 2006, Pavel Machek wrote:

> Hi!
>
>>> of the story for me. There's nothing wrong about focusing on newer code,
>>> but the old code needs to be cared for, too, to fix remaining issues
>>> such as the "can only have N files with the same hash value".
>>>
>> Requires a disk format change, in a filesystem without plugins, to fix it.
>
> Well, too bad, if reiser3 is so broken it needs on-disk-format-change,
> then I guess doing that change is the right thing to do...

Sorry for my stupid question, but could you tell me why starting to make
incompatible changes to reiserfs3 now (when reiserfs3 "technology" is
rather old) and making reiserfs3 unstable (again), possibly for several
months or even years is better than fixing big issues with reiser4 (if
there are any really big left) merging it and trying to stabilize it?

For end user both ways will result in mkfs so...


Thanks,

Grzegorz Kulewski

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


galibert at pobox

Jul 27, 2006, 8:52 AM

Post #115 of 223 (10349 views)
Permalink
Re: the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On Wed, Jul 26, 2006 at 01:08:06PM +0000, Pavel Machek wrote:
> Hi!
>
> > > A much more important effect is that non-maintainers aren't familiar
> > > with coding and patch submission guidelines. For example, in
> > > suspend2, Nigel first tried with patches that were too monolithic,
> > > and then his next series was too broken down such that it was too
> > > hard to review (and "git bisect" wouldn't work).
> >
> > All his submissions since 2004 or so? It's a little easy to limit
> > oneself to the last two ones.
>
> Nigel did not do any submissions in 2004 or so. Check your fact, that
> stuff was marked 'RFC' and yes I did comment on it.

2004-09-16 submission:

http://lkml.org/lkml/2004/9/16/76
http://lkml.org/lkml/2004/9/16/77
http://lkml.org/lkml/2004/9/16/78
http://lkml.org/lkml/2004/9/16/81
http://lkml.org/lkml/2004/9/16/82
http://lkml.org/lkml/2004/9/16/83
http://lkml.org/lkml/2004/9/16/86
http://lkml.org/lkml/2004/9/16/87
http://lkml.org/lkml/2004/9/16/89
http://lkml.org/lkml/2004/9/16/90


2004-11-24 submission:

http://lkml.org/lkml/2004/11/24/93
http://lkml.org/lkml/2004/11/24/94
http://lkml.org/lkml/2004/11/24/95
http://lkml.org/lkml/2004/11/24/96
http://lkml.org/lkml/2004/11/24/97
http://lkml.org/lkml/2004/11/24/98
http://lkml.org/lkml/2004/11/24/100
http://lkml.org/lkml/2004/11/24/101
[58 mails or so total]


> He did 1 (one) submission that looked like SubmittingPatches at the
> first sight, and that was very recent.
>
> Stop spreading lies.

I am awaiting your apologies.

OG.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


andrea at cpushare

Jul 27, 2006, 9:11 AM

Post #116 of 223 (10338 views)
Permalink
Re: the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On Thu, Jul 27, 2006 at 05:05:08PM +0200, Adrian Bunk wrote:
> Slightly modified version is now in my .signature (I omitted the
> "spare time" part since the information is still true and sounds better
> without it - it would be cencorship if you'd try to force me to not
> omit it).

Well, a signature like that, perfectly match the IQ of who writes it,
please add a link to the klive homepage too if you don't mind.

This IMHO would be even more hilarious:

"I'm happy with ext2" by Adrain Bunk, the guy who pretends to provide
useful feedback in the reiser4 discussion.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


alan at lxorguk

Jul 27, 2006, 9:43 AM

Post #117 of 223 (10349 views)
Permalink
Re: the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

Ar Mer, 2006-07-26 am 13:08 +0000, ysgrifennodd Pavel Machek:
> He did 1 (one) submission that looked like SubmittingPatches at the
> first sight, and that was very recent.
>
> Stop spreading lies.

Pavel, I think you might want to check your facts and then apologize.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


matthias.andree at gmx

Jul 27, 2006, 10:28 AM

Post #118 of 223 (10363 views)
Permalink
Re: the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On Thu, 27 Jul 2006, Grzegorz Kulewski wrote:

> Sorry for my stupid question, but could you tell me why starting to make
> incompatible changes to reiserfs3 now (when reiserfs3 "technology" is
> rather old) and making reiserfs3 unstable (again), possibly for several
> months or even years is better than fixing big issues with reiser4 (if
> there are any really big left) merging it and trying to stabilize it?
>
> For end user both ways will result in mkfs so...

ext2fs and ext3fs, without "plugins", added dir_index as a compatible
upgrade, with an e2fsck option (that implies optional) to build indices
for directories without them.

ext3fs is a compatible upgrade from ext2fs, it's as simple as unmount,
tune2fs -j, mount.

reiserfs 3.6 could deal with 3.5 file systems, and "mount -o conv" with
a 3.6 driver would convert a 3.5 file system to 3.6 level
(ISTR it had to do with large file support and perhaps NFS
exportability, but don't quote me on that).

I wonder what makes the hash overflow issue so complicated (other than
differing business plans, that is) that upgrading in place isn't
possible. Changes introduce instability, but namesys were proud of their
regression testing - so how sustainable is their internal test suite?

Instead, we're told reiser4 would fix this (quite likely) and we should
wait until it's ready (OK, we shouldn't be using experimental stuff for
production but rather for /tmp, but the file system will take many
months to mature after integration) and it will be "mkfs" time - so
reiser4 better be mature before we go that way if there's no way back
short of "amrecover", "restore" or "tar -x".

Smashing out most of the Cc:s in order not to bore people.

--
Matthias Andree
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


jeff at garzik

Jul 27, 2006, 10:56 AM

Post #119 of 223 (10349 views)
Permalink
Re: the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

Pavel Machek wrote:
> Hi!
>
>>> of the story for me. There's nothing wrong about focusing on newer code,
>>> but the old code needs to be cared for, too, to fix remaining issues
>>> such as the "can only have N files with the same hash value".
>>>
>> Requires a disk format change, in a filesystem without plugins, to fix it.
>
> Well, too bad, if reiser3 is so broken it needs on-disk-format-change,
> then I guess doing that change is the right thing to do...

Actually, there is reiser4 brokenness lurking in Hans' statement, too:

A filesystem WITH plugins must still handle the standard Linux
compatibility stuff that other filesystems handle.

Plugins --do not-- mean that you can just change the filesystem format
willy-nilly, with zero impact.

Jeff



-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


jim at why

Jul 27, 2006, 11:37 AM

Post #120 of 223 (10354 views)
Permalink
Re: the ' 'official' point of view' expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On 07/27/06 04:31:07PM +0200, Luigi Genoni wrote:
> Since reiser4 is not something enabled by default into a default kernel
> distribution, I assume they enabled it knowing what they where doing because
> they wanted to use it.
>
>

Or because they did 'allmodconfig' or 'allyesconfig'. Whenever I build
a kernel I enabled everything possible as a module in case I ever need
it. For instance, a few weeks ago I had the reiserfs module loaded because
I was testing something, if I had klive running it would have said that I
use reiserfs when in fact I don't.

Jim.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


vonbrand at inf

Jul 27, 2006, 11:43 AM

Post #121 of 223 (10362 views)
Permalink
Re: the ' 'official' point of view' expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

Luigi Genoni <genoni [at] sns> wrote:

[...]

> Since reiser4 is not something enabled by default into a default kernel
> distribution, I assume they enabled it knowing what they where doing because
> they wanted to use it.

You know, I compile ReiserFS, JFS and XFS (and also CRAMFS and ROMFS) into
my self-compiled kernels too. Haven't used any of them in ages (if ever),
but one more go at compile-testing won't hurt...
--
Dr. Horst H. von Brand User #22616 counter.li.org
Departamento de Informatica Fono: +56 32 654431
Universidad Tecnica Federico Santa Maria +56 32 654239
Casilla 110-V, Valparaiso, Chile Fax: +56 32 797513

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


ninja at slaphack

Jul 27, 2006, 12:06 PM

Post #122 of 223 (10351 views)
Permalink
Re: the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

Jeff Garzik wrote:
> Pavel Machek wrote:
>> Hi!
>>
>>>> of the story for me. There's nothing wrong about focusing on newer
>>>> code,
>>>> but the old code needs to be cared for, too, to fix remaining issues
>>>> such as the "can only have N files with the same hash value".
>>> Requires a disk format change, in a filesystem without plugins, to
>>> fix it.

> A filesystem WITH plugins must still handle the standard Linux
> compatibility stuff that other filesystems handle.
>
> Plugins --do not-- mean that you can just change the filesystem format
> willy-nilly, with zero impact.

They --do-- mean that you can change much of the filesystem behavior
without requiring massive on-disk changes or massive interface changes.

After all, this is how many FUSE plugins work -- standard FS interface,
usually uses another standard FS as storage, but does crazy things like
compression, encryption, and other transformations in between.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


andrea at cpushare

Jul 27, 2006, 12:34 PM

Post #123 of 223 (10359 views)
Permalink
Re: the ' 'official' point of view' expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On Thu, Jul 27, 2006 at 02:37:33PM -0400, Jim Crilly wrote:
> Or because they did 'allmodconfig' or 'allyesconfig'. Whenever I build
> a kernel I enabled everything possible as a module in case I ever need
> it. For instance, a few weeks ago I had the reiserfs module loaded because
> I was testing something, if I had klive running it would have said that I
> use reiserfs when in fact I don't.

reiserfs would showup in the module list in such case, but _not_ in
the fs list. KLive records both the modules loaded _and_ the mounted
fs.

This is the code that records the FS:

if CONFIG_FS:
fs = {}
for _fs in [ re.search(r'^[^\s]+\s[^\s]+\s([^\s]+)', x).group(1)
for x in file('/proc/mounts').readlines() ]:
if _fs not in ['rootfs', 'proc', 'devpts']:
fs[_fs] = None
fs = ' '.join(fs.keys())
KERNEL += str_append(fs)

/proc/mounts only lists the _mounted_ fs, not the fs loaded into the
kernel statically or with insmod.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


andrea at cpushare

Jul 27, 2006, 1:11 PM

Post #124 of 223 (10370 views)
Permalink
Re: the ' 'official' point of view' expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

On Thu, Jul 27, 2006 at 02:43:48PM -0400, Horst H. von Brand wrote:
> You know, I compile ReiserFS, JFS and XFS (and also CRAMFS and ROMFS) into
> my self-compiled kernels too. Haven't used any of them in ages (if ever),

If you never mounted them shortly after boot time, they would have
never showed up in KLive fs list if that's what you mean.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


reiser at namesys

Jul 27, 2006, 6:47 PM

Post #125 of 223 (10369 views)
Permalink
Re: the " 'official' point of view" expressed by kernelnewbies.org regarding reiser4 inclusion [In reply to]

andrea [at] cpushare wrote:

>
>
>As far as I'm concerned the thing I like less of reiser4 is the plugin
>thing, I'd be less concerned if that was a microkernel (fuse-like)
>userland plugin system.
>

Performance. If your plugin is performance valuing, it needs to be in
kernel. Also, FUSE does not have per-file plugins (correct me if I err
here). It would be nice to see it pay attention to reiser4 pluginids,
and thus become per-file.

Thanks for your other words, which I will not comment on because I agree
with them.;-)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo [at] vger
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

First page Previous page 1 2 3 4 5 6 7 8 9 Next page Last page  View All Linux kernel RSS feed   Index | Next | Previous | View Threaded
 
 


Interested in having your list archived? Contact Gossamer Threads
 
  Web Applications & Managed Hosting Powered by Gossamer Threads Inc.