Re: A WISH: Reiser4 patch for 2.6.14

Re: A WISH: Reiser4 patch for 2.6.14

[Vladimir V. Saveliev]

Hello

Yusuf Iskenderoglu wrote:
> Okay, thanks for the information. I will continue waiting.
> 

Well, here is the first attempt to relelase reiser4 for 2.6.14. Please try to
use it.
ftp://ftp.namesys.com/pub/reiser4-for-2.6/2.6.14/reiser4-for-2.6.14-1.patch.gz


Consider it as an experimental thing. Note, it will probably not compile as a
module. Adding a line to mm/page-writeback.c the following line

EXPORT_SYMBOL(clear_page_dirty_for_io);

after clear_page_dirty_for_io()'s definition should fix that, though.



> On Mon, 2005-11-07 at 19:20 +0300, Vladimir V. Saveliev wrote:
>>Hello
>>
>>Yusuf Iskenderoglu wrote:
>>>Hello,
>>>
>>>I am not having a support issue, just a wish.
>>>It always takes many weeks until you put a patch for the reiser4
>>>filesystem for the most actual Linux version into your ftp server.
>>>In this case, 2.6.14 is already released for one week now, and there is
>>>still no patch available.
>>>
>>>I am not interested in patching the kernel with stuff from other/older
>>>patches.
>>>I just wish you be up-to-date
>>>
>>>Creating a patch for the actual kernel version simply cannot take you
>>>more than 1 Minute.
>>>
>>Sorry for delay with that. Code got big changes recently, we would like to test
>>it a bit.
> 
> 
> 

Vanilla 2.6.14.2 with reiser4-for-2.6.14-1.patch

[jp]

Hi,

I've compiled the Vanilla 2.6.14.2 with reiser4-for-2.6.14-1.patch, 
built static.

My first tests are very positive :

I Launched the following "disk operations" at the same time, while Gxine 
was playing a MPG :
- a 400MB ISO building
- complete kernel source recursive directory copy
- Openoffice startup
- Firefox startup

There was no hangup in the video, and all disk operations finished 
without problems.

I'm going to keep using this kernel, and report any problem.

Regards

JP



Vladimir V. Saveliev wrote:

>Hello
>
>Yusuf Iskenderoglu wrote:
>  
>
>>Okay, thanks for the information. I will continue waiting.
>>
>>    
>>
>
>Well, here is the first attempt to relelase reiser4 for 2.6.14. Please try to
>use it.
>ftp://ftp.namesys.com/pub/reiser4-for-2.6/2.6.14/reiser4-for-2.6.14-1.patch.gz
>
>
>Consider it as an experimental thing. Note, it will probably not compile as a
>module. Adding a line to mm/page-writeback.c the following line
>
>EXPORT_SYMBOL(clear_page_dirty_for_io);
>
>after clear_page_dirty_for_io()'s definition should fix that, though.
>
>
>
>  
>
>>On Mon, 2005-11-07 at 19:20 +0300, Vladimir V. Saveliev wrote:
>>    
>>
>>>Hello
>>>
>>>Yusuf Iskenderoglu wrote:
>>>      
>>>
>>>>Hello,
>>>>
>>>>I am not having a support issue, just a wish.
>>>>It always takes many weeks until you put a patch for the reiser4
>>>>filesystem for the most actual Linux version into your ftp server.
>>>>In this case, 2.6.14 is already released for one week now, and there is
>>>>still no patch available.
>>>>
>>>>I am not interested in patching the kernel with stuff from other/older
>>>>patches.
>>>>I just wish you be up-to-date
>>>>
>>>>Creating a patch for the actual kernel version simply cannot take you
>>>>more than 1 Minute.
>>>>
>>>>        
>>>>
>>>Sorry for delay with that. Code got big changes recently, we would like to test
>>>it a bit.
>>>      
>>>
>>    
>>
>
>
>
>  
>

Re: A WISH: Reiser4 patch for 2.6.14

[Jake Maciejewski]

[-- Attachment #1: Type: text/plain, Size: 1677 bytes --]

It looks good so far on amd64, but I haven't had the chance to stress
test it yet.

I remember Hans mentioning wanting to eliminate warnings, so I attached
a trivial patch for an ISO C90 mixed code/declarations warning.

On Fri, 2005-11-11 at 19:33 +0300, Vladimir V. Saveliev wrote:
> Hello
> 
> Yusuf Iskenderoglu wrote:
> > Okay, thanks for the information. I will continue waiting.
> > 
> 
> Well, here is the first attempt to relelase reiser4 for 2.6.14. Please try to
> use it.
> ftp://ftp.namesys.com/pub/reiser4-for-2.6/2.6.14/reiser4-for-2.6.14-1.patch.gz
> 
> 
> Consider it as an experimental thing. Note, it will probably not compile as a
> module. Adding a line to mm/page-writeback.c the following line
> 
> EXPORT_SYMBOL(clear_page_dirty_for_io);
> 
> after clear_page_dirty_for_io()'s definition should fix that, though.
> 
> 
> 
> > On Mon, 2005-11-07 at 19:20 +0300, Vladimir V. Saveliev wrote:
> >>Hello
> >>
> >>Yusuf Iskenderoglu wrote:
> >>>Hello,
> >>>
> >>>I am not having a support issue, just a wish.
> >>>It always takes many weeks until you put a patch for the reiser4
> >>>filesystem for the most actual Linux version into your ftp server.
> >>>In this case, 2.6.14 is already released for one week now, and there is
> >>>still no patch available.
> >>>
> >>>I am not interested in patching the kernel with stuff from other/older
> >>>patches.
> >>>I just wish you be up-to-date
> >>>
> >>>Creating a patch for the actual kernel version simply cannot take you
> >>>more than 1 Minute.
> >>>
> >>Sorry for delay with that. Code got big changes recently, we would like to test
> >>it a bit.
> > 
> > 
> > 
> 
-- 
Jake Maciejewski <maciejej@msoe.edu>

[-- Attachment #2: reiser4-cryptcompress-warning.patch --]
[-- Type: text/x-patch, Size: 587 bytes --]

--- cryptcompress.c.orig	2005-11-11 12:38:34.447766000 -0600
+++ cryptcompress.c	2005-11-11 12:39:22.066742000 -0600
@@ -505,6 +505,8 @@
 
 int open_cryptcompress(struct inode * inode, struct file * file)
 {
+	struct inode * parent;
+
 	assert("edward-1394", inode != NULL);
 	assert("edward-1395", file != NULL);
 	assert("edward-1396", file != NULL);
@@ -514,7 +516,6 @@
 	assert("edward-698",
 	       inode_file_plugin(inode) ==
 	       file_plugin_by_id(CRC_FILE_PLUGIN_ID));
-	struct inode * parent;
 	if (!need_cipher(inode))
 		/* the file is not to be ciphered */
 		return 0;

Re: A WISH: Reiser4 patch for 2.6.14

[Yusuf Iskenderoglu]

Hello,

I am now using this patch, it required me to export __pagevec_realloc @
mm/swap.c though, and not the one mentioned in the email.

I am using this partition with squid and will do some simultaneous tests
like building gcc on the same partition, and inform you, whether
interesting things happen.

Regards;
Yusuf.

On Fri, 2005-11-11 at 19:33 +0300, Vladimir V. Saveliev wrote:
> Hello
> 
> Yusuf Iskenderoglu wrote:
> > Okay, thanks for the information. I will continue waiting.
> > 
> 
> Well, here is the first attempt to relelase reiser4 for 2.6.14. Please try to
> use it.
> ftp://ftp.namesys.com/pub/reiser4-for-2.6/2.6.14/reiser4-for-2.6.14-1.patch.gz
> 
> 
> Consider it as an experimental thing. Note, it will probably not compile as a
> module. Adding a line to mm/page-writeback.c the following line
> 
> EXPORT_SYMBOL(clear_page_dirty_for_io);
> 
> after clear_page_dirty_for_io()'s definition should fix that, though.