linux-apus-devel

2.4 test 8

[Ken T. <ke...@we...>]

Compiled and booted 2.4 today - hooray !

Is affs still broken ?

If yes, what's wrong with it and how 'dangerous' is it considering my affs
test partition is on the same drive as ext2 partitions.

Ken.

Re: 2.4 test 8

[<fp...@zu...>]

On Thu, Sep 21, 2000 at 06:41:15PM +1100, Ken Tyler wrote:
> 
> Compiled and booted 2.4 today - hooray !
> 
> Is affs still broken ?
> 
> If yes, what's wrong with it and how 'dangerous' is it considering my affs
> test partition is on the same drive as ext2 partitions.

AFFS works for me (in -test7).
-- 
Frank Petzold, IBM Zurich Research Laboratory, Säumerstrasse 4,
CH-8803 Rüschlikon/Switzerland, Tel.  +41-1-724-84-42  Fax.   +41-1-724-89-56
 Business email: fp...@zu...          Private email: pe...@he...
   The opinions expressed here are mine and not necessarily those of IBM.

Re: 2.4 test 8

[Ken T. <ke...@we...>]

On Thu, 21 Sep 2000, Frank Petzold wrote:

> AFFS works for me (in -test7).

OK, it was reported as broken early on.

I'll give it try.

Ken.

affs in test8, was Re: 2.4 test 8

[Ken T. <ke...@we...>]

On Thu, 21 Sep 2000, Frank Petzold wrote:

> On Thu, Sep 21, 2000 at 06:41:15PM +1100, Ken Tyler wrote:

> > Is affs still broken ?
 
> AFFS works for me (in -test7).

Gave it a go with test8 - no good.

It looks like it has a problem similar to last time, copy a file to an
affs partition (where the file does not exist), md5sum matches OK.  Copy
again over the just made copy, the second copy fails md5sum checksum.

Last time the problem was to do with extension block caching, maybe
something similar this time - I know how it works now !

Ken.

Re: 2.4 test 8

[Roman Z. <zi...@fh...>]

Hi,

On Thu, 21 Sep 2000, Ken Tyler wrote:

> Is affs still broken ?
> 
> If yes, what's wrong with it and how 'dangerous' is it considering my affs
> test partition is on the same drive as ext2 partitions.

Sorry for not answering earlier, but I have a week off right now. :-)
I was working on this before I left, but it isn't finished yet, so the cvs
version is still in the same state.

bye, Roman

Re: 2.4 test 8

[Ken T. <ke...@we...>]

On Wed, 27 Sep 2000, Roman Zippel wrote:

> > Is affs still broken ?

> Sorry for not answering earlier, but I have a week off right now. :-)
> I was working on this before I left, but it isn't finished yet, so the cvs
> version is still in the same state.

Are you getting anywhere with it ?

The affs subdir has changed quite a bit since 2.2.  I had a look and did
some copying tests. The corruption was similar to what was happening in
2.2.

Ken.

Re: 2.4 test 8

[Roman Z. <zi...@fh...>]

Hi,

On Thu, 28 Sep 2000, Ken Tyler wrote:

> Are you getting anywhere with it ?

Slowly. :)
I want to clean up a few things and that will take some time...

> The affs subdir has changed quite a bit since 2.2.  I had a look and did
> some copying tests. The corruption was similar to what was happening in
> 2.2.

Was it fixed there? I can't remember hearing of (serious) problems with
2.2.

bye, Roman

Re: 2.4 test 8

[Michel <da...@re...>]

Roman Zippel wrote:

> > The affs subdir has changed quite a bit since 2.2.  I had a look and did
> > some copying tests. The corruption was similar to what was happening in
> > 2.2.
> 
> Was it fixed there?

Yep, Ken fixed it AFAIK.


> I can't remember hearing of (serious) problems with 2.2.

I have experienced the same problems with 2.2 that Ken is describing with 2.4.


Michel


-- 
Earthling Michel Dänzer (MrCooper)  \  CS student and free software enthusiast
Debian GNU/Linux (powerpc,i386) user \   member of XFree86 and the DRI project

Re: 2.4 test 8

[Roman Z. <zi...@fh...>]

Hi,

> Yep, Ken fixed it AFAIK.

I see two changes to fs/affs/file.c, but these are also in 2.4.

bye, Roman

Re: 2.4 test 8

[Ken T. <ke...@we...>]

On Tue, 3 Oct 2000, Roman Zippel wrote:

> Hi,
> 
> > Yep, Ken fixed it AFAIK.
> 
> I see two changes to fs/affs/file.c, but these are also in 2.4.

The symptoms in 2.4 don't look at all similar to the 2.2 problem. 
Started to look at it before I went on holiday but didn't get very far.  
Couldn't find any consistent pattern to the bad blocks, unlike 2.2 which
was very consistent (on big enough files) and was related to the magic
number 72.

I'll look some more when I catch up on a few things.

Ken.

Re: 2.4 test 8

[Roman Z. <zi...@fh...>]

Hi,

On Fri, 27 Oct 2000, Ken Tyler wrote:

> I'll look some more when I catch up on a few things.

Please wait one or two weeks, I'm dumping the current block lookup
mechanism anyway.

bye, Roman

Re: 2.4 test 8

[Ken T. <ke...@we...>]

On Fri, 27 Oct 2000, Roman Zippel wrote:

[I said re: affs errors]

> > I'll look some more when I catch up on a few things.

> Please wait one or two weeks, I'm dumping the current block lookup
> mechanism anyway.

I just compiled 2.4 test 9 from sourceforge, after a few copies of a 2 Meg
file to an affs partition I'm not getting any errors.

Have you/anyone done anything that might have fixed it ?

Ken. 

Re: 2.4 test 8

[Roman Z. <zi...@fh...>]

Hi,

On Wed, 8 Nov 2000, Ken Tyler wrote:

> I just compiled 2.4 test 9 from sourceforge, after a few copies of a 2 Meg
> file to an affs partition I'm not getting any errors.

No.
BTW I just imported test10, but there were some bigger mm changes, so it's
broken again... 

bye, Roman

Re: 2.4 test 8

[Ken T. <ke...@we...>]

On Thu, 9 Nov 2000, Roman Zippel wrote:

> On Wed, 8 Nov 2000, Ken Tyler wrote:
> 
> > I just compiled 2.4 test 9 from sourceforge, after a few copies of a 2 Meg
> > file to an affs partition I'm not getting any errors.
> 
> No.
> BTW I just imported test10, but there were some bigger mm changes, so it's
> broken again... 

Broken again ?  Is that just affs or the lot ?

Ken.

Re: 2.4 test 8

[Roman Z. <zi...@fh...>]

Hi,

On Sat, 11 Nov 2000, Ken Tyler wrote:

> Broken again ?  Is that just affs or the lot ?

Affs is still the same.
What got broken due to the test10 import should be fixed now.

bye, Roman