* new reiser4 snapshot
@ 2003-10-17 15:42 Nikita Danilov
2003-10-17 18:51 ` Hans Reiser
2003-10-19 14:38 ` Nishant Sharma
0 siblings, 2 replies; 10+ messages in thread
From: Nikita Danilov @ 2003-10-17 15:42 UTC (permalink / raw)
To: Reiserfs mail-list
Cc: Reiserfs developers mail-list, Linux Kernel Mailing List
Hello,
Reiser4 snapshot against 2.6.0-test7 has been released:
http://namesys.com/snapshots/2003.10.17/
Description is in http://namesys.com/snapshots/2003.10.17/READ.ME
Nikita.
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: new reiser4 snapshot
2003-10-17 15:42 new reiser4 snapshot Nikita Danilov
@ 2003-10-17 18:51 ` Hans Reiser
2003-10-19 14:38 ` Nishant Sharma
1 sibling, 0 replies; 10+ messages in thread
From: Hans Reiser @ 2003-10-17 18:51 UTC (permalink / raw)
To: Nikita Danilov
Cc: Reiserfs mail-list, Reiserfs developers mail-list,
Linux Kernel Mailing List
This snapshot fixes two fatal bugs, and improves performance. Nikita
can still make it crash, sigh, and we have discovered some unnecessary
seeks when replicating Mike Benoit's settings for bonnie++ that are due
to block allocation bugs. Sigh. We are probably still at least 2 weeks
away from code that we can seriously ask people to use. The performance
is good enough to be well worth using despite the allocation bugs, but
we don't like to ask people to use code we know how to crash.
--
Hans
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: new reiser4 snapshot
2003-10-17 15:42 new reiser4 snapshot Nikita Danilov
2003-10-17 18:51 ` Hans Reiser
@ 2003-10-19 14:38 ` Nishant Sharma
2003-10-19 18:09 ` Hans Reiser
2003-10-20 9:36 ` Nikita Danilov
1 sibling, 2 replies; 10+ messages in thread
From: Nishant Sharma @ 2003-10-19 14:38 UTC (permalink / raw)
To: Nikita Danilov; +Cc: Reiserfs mail-list
Hi,
On Fri, 17 Oct 2003, Nikita Danilov wrote:
> Hello,
> Reiser4 snapshot against 2.6.0-test7 has been released:
> http://namesys.com/snapshots/2003.10.17/
> Description is in http://namesys.com/snapshots/2003.10.17/READ.ME
I downloaded the 2.6.0-test7 kernel and patched it for reiser4.
The kernel compiled successfully, with reiser4 enabled, except that some
symbols were not found. When I booted the new kernel, I found that reiser4
module doesnot load.
-bash-2.05b# modprobe reiser4
FATAL: Error inserting reiser4
(/lib/modules/2.6.0-test7/kernel/fs/reiser4/reiser4.ko): Unknown symbol in
module, or unknown parameter (see dmesg)
-bash-2.05b# dmesg |tail
nfs warning: mount version older than kernel
reiser4: Unknown symbol max_sane_readahead
reiser4: Unknown symbol __remove_from_page_cache
reiser4: Unknown symbol radix_tree_preload
reiser4: Unknown symbol generic_sync_sb_inodes
reiser4: Unknown symbol truncate_mapping_pages_range
reiser4: Unknown symbol balance_dirty_pages_ratelimited
reiser4: Unknown symbol inodes_stat
reiser4: Unknown symbol nr_free_pagecache_pages
reiser4: Unknown symbol destroy_inode
Following is the relevant section of my .config in kernel src-dir.
CONFIG_REISER4_FS=m
# CONFIG_REISER4_FS_SYSCALL is not set
# CONFIG_REISER4_LARGE_KEY is not set
# CONFIG_REISER4_CHECK is not set
CONFIG_REISER4_USE_EFLUSH=y
# CONFIG_REISER4_COPY_ON_CAPTURE is not set
# CONFIG_REISER4_BADBLOCKS is not set
What am I doing wrong? I believe that I followed the instructions
in http://namesys.com/snapshots/2003.10.17/READ.ME to the letter.
--
Nishant
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: new reiser4 snapshot
2003-10-19 14:38 ` Nishant Sharma
@ 2003-10-19 18:09 ` Hans Reiser
2003-10-19 22:41 ` Carl-Daniel Hailfinger
2003-10-20 9:37 ` Nikita Danilov
2003-10-20 9:36 ` Nikita Danilov
1 sibling, 2 replies; 10+ messages in thread
From: Hans Reiser @ 2003-10-19 18:09 UTC (permalink / raw)
To: Nishant Sharma; +Cc: Nikita Danilov, Reiserfs mail-list
Nishant Sharma wrote:
>Hi,
>
>On Fri, 17 Oct 2003, Nikita Danilov wrote:
>
>
>>Hello,
>>Reiser4 snapshot against 2.6.0-test7 has been released:
>>http://namesys.com/snapshots/2003.10.17/
>>Description is in http://namesys.com/snapshots/2003.10.17/READ.ME
>>
>>
> I downloaded the 2.6.0-test7 kernel and patched it for reiser4.
>The kernel compiled successfully, with reiser4 enabled, except that some
>symbols were not found. When I booted the new kernel, I found that reiser4
>module doesnot load.
>
>-bash-2.05b# modprobe reiser4
>FATAL: Error inserting reiser4
>(/lib/modules/2.6.0-test7/kernel/fs/reiser4/reiser4.ko): Unknown symbol in
>module, or unknown parameter (see dmesg)
>-bash-2.05b# dmesg |tail
>nfs warning: mount version older than kernel
>reiser4: Unknown symbol max_sane_readahead
>reiser4: Unknown symbol __remove_from_page_cache
>reiser4: Unknown symbol radix_tree_preload
>reiser4: Unknown symbol generic_sync_sb_inodes
>reiser4: Unknown symbol truncate_mapping_pages_range
>reiser4: Unknown symbol balance_dirty_pages_ratelimited
>reiser4: Unknown symbol inodes_stat
>reiser4: Unknown symbol nr_free_pagecache_pages
>reiser4: Unknown symbol destroy_inode
>
> Following is the relevant section of my .config in kernel src-dir.
>
>CONFIG_REISER4_FS=m
># CONFIG_REISER4_FS_SYSCALL is not set
># CONFIG_REISER4_LARGE_KEY is not set
># CONFIG_REISER4_CHECK is not set
>CONFIG_REISER4_USE_EFLUSH=y
># CONFIG_REISER4_COPY_ON_CAPTURE is not set
># CONFIG_REISER4_BADBLOCKS is not set
>
> What am I doing wrong? I believe that I followed the instructions
>in http://namesys.com/snapshots/2003.10.17/READ.ME to the letter.
>
>
Nikita, please determine why reiser4 tells users they can configure it
as a module when they cannot.
--
Hans
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: new reiser4 snapshot
2003-10-19 18:09 ` Hans Reiser
@ 2003-10-19 22:41 ` Carl-Daniel Hailfinger
2003-10-20 9:44 ` Nikita Danilov
2003-10-20 9:37 ` Nikita Danilov
1 sibling, 1 reply; 10+ messages in thread
From: Carl-Daniel Hailfinger @ 2003-10-19 22:41 UTC (permalink / raw)
To: Hans Reiser; +Cc: Nishant Sharma, Nikita Danilov, Reiserfs mail-list
Hans Reiser wrote:
> Nishant Sharma wrote:
>
>> Hi,
>>
>> On Fri, 17 Oct 2003, Nikita Danilov wrote:
>>
>>
>>> Hello,
>>> Reiser4 snapshot against 2.6.0-test7 has been released:
>>> http://namesys.com/snapshots/2003.10.17/
>>> Description is in http://namesys.com/snapshots/2003.10.17/READ.ME
>>>
>> Following is the relevant section of my .config in kernel src-dir.
>>
>> CONFIG_REISER4_FS=m
>> # CONFIG_REISER4_FS_SYSCALL is not set
>> # CONFIG_REISER4_LARGE_KEY is not set
I thought CONFIG_REISER4_LARGE_KEY was force-enabled by now?
>> # CONFIG_REISER4_CHECK is not set
>> CONFIG_REISER4_USE_EFLUSH=y
>> # CONFIG_REISER4_COPY_ON_CAPTURE is not set
>> # CONFIG_REISER4_BADBLOCKS is not set
>>
>> What am I doing wrong? I believe that I followed the instructions
>> in http://namesys.com/snapshots/2003.10.17/READ.ME to the letter.
>>
>>
> Nikita, please determine why reiser4 tells users they can configure it
> as a module when they cannot.
Carl-Daniel
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: new reiser4 snapshot
2003-10-19 14:38 ` Nishant Sharma
2003-10-19 18:09 ` Hans Reiser
@ 2003-10-20 9:36 ` Nikita Danilov
1 sibling, 0 replies; 10+ messages in thread
From: Nikita Danilov @ 2003-10-20 9:36 UTC (permalink / raw)
To: Nishant Sharma; +Cc: Reiserfs mail-list
Nishant Sharma writes:
> Hi,
>
> On Fri, 17 Oct 2003, Nikita Danilov wrote:
> > Hello,
> > Reiser4 snapshot against 2.6.0-test7 has been released:
> > http://namesys.com/snapshots/2003.10.17/
> > Description is in http://namesys.com/snapshots/2003.10.17/READ.ME
> I downloaded the 2.6.0-test7 kernel and patched it for reiser4.
> The kernel compiled successfully, with reiser4 enabled, except that some
> symbols were not found. When I booted the new kernel, I found that reiser4
> module doesnot load.
>
> -bash-2.05b# modprobe reiser4
> FATAL: Error inserting reiser4
> (/lib/modules/2.6.0-test7/kernel/fs/reiser4/reiser4.ko): Unknown symbol in
> module, or unknown parameter (see dmesg)
> -bash-2.05b# dmesg |tail
> nfs warning: mount version older than kernel
> reiser4: Unknown symbol max_sane_readahead
> reiser4: Unknown symbol __remove_from_page_cache
> reiser4: Unknown symbol radix_tree_preload
> reiser4: Unknown symbol generic_sync_sb_inodes
> reiser4: Unknown symbol truncate_mapping_pages_range
> reiser4: Unknown symbol balance_dirty_pages_ratelimited
> reiser4: Unknown symbol inodes_stat
> reiser4: Unknown symbol nr_free_pagecache_pages
> reiser4: Unknown symbol destroy_inode
>
> Following is the relevant section of my .config in kernel src-dir.
>
> CONFIG_REISER4_FS=m
> # CONFIG_REISER4_FS_SYSCALL is not set
> # CONFIG_REISER4_LARGE_KEY is not set
> # CONFIG_REISER4_CHECK is not set
> CONFIG_REISER4_USE_EFLUSH=y
> # CONFIG_REISER4_COPY_ON_CAPTURE is not set
> # CONFIG_REISER4_BADBLOCKS is not set
>
> What am I doing wrong? I believe that I followed the instructions
> in http://namesys.com/snapshots/2003.10.17/READ.ME to the letter.
Reiser4 erroneously allows one to compile it as a module, while this is
not supported. Compile it as built-in. Thank you the report.
> --
> Nishant
>
Nikita.
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: new reiser4 snapshot
2003-10-19 18:09 ` Hans Reiser
2003-10-19 22:41 ` Carl-Daniel Hailfinger
@ 2003-10-20 9:37 ` Nikita Danilov
1 sibling, 0 replies; 10+ messages in thread
From: Nikita Danilov @ 2003-10-20 9:37 UTC (permalink / raw)
To: Hans Reiser; +Cc: Reiserfs mail-list
Hans Reiser writes:
> Nishant Sharma wrote:
>
> >Hi,
> >
> >On Fri, 17 Oct 2003, Nikita Danilov wrote:
> >
> >
> >>Hello,
> >>Reiser4 snapshot against 2.6.0-test7 has been released:
> >>http://namesys.com/snapshots/2003.10.17/
> >>Description is in http://namesys.com/snapshots/2003.10.17/READ.ME
> >>
> >>
> > I downloaded the 2.6.0-test7 kernel and patched it for reiser4.
> >The kernel compiled successfully, with reiser4 enabled, except that some
> >symbols were not found. When I booted the new kernel, I found that reiser4
> >module doesnot load.
> >
> >-bash-2.05b# modprobe reiser4
> >FATAL: Error inserting reiser4
> >(/lib/modules/2.6.0-test7/kernel/fs/reiser4/reiser4.ko): Unknown symbol in
> >module, or unknown parameter (see dmesg)
> >-bash-2.05b# dmesg |tail
> >nfs warning: mount version older than kernel
> >reiser4: Unknown symbol max_sane_readahead
> >reiser4: Unknown symbol __remove_from_page_cache
> >reiser4: Unknown symbol radix_tree_preload
> >reiser4: Unknown symbol generic_sync_sb_inodes
> >reiser4: Unknown symbol truncate_mapping_pages_range
> >reiser4: Unknown symbol balance_dirty_pages_ratelimited
> >reiser4: Unknown symbol inodes_stat
> >reiser4: Unknown symbol nr_free_pagecache_pages
> >reiser4: Unknown symbol destroy_inode
> >
> > Following is the relevant section of my .config in kernel src-dir.
> >
> >CONFIG_REISER4_FS=m
> ># CONFIG_REISER4_FS_SYSCALL is not set
> ># CONFIG_REISER4_LARGE_KEY is not set
> ># CONFIG_REISER4_CHECK is not set
> >CONFIG_REISER4_USE_EFLUSH=y
> ># CONFIG_REISER4_COPY_ON_CAPTURE is not set
> ># CONFIG_REISER4_BADBLOCKS is not set
> >
> > What am I doing wrong? I believe that I followed the instructions
> >in http://namesys.com/snapshots/2003.10.17/READ.ME to the letter.
> >
> >
> Nikita, please determine why reiser4 tells users they can configure it
> as a module when they cannot.
Error. Will be fixed in the next snapshot.
>
> --
> Hans
>
Nikita.
>
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: new reiser4 snapshot
2003-10-19 22:41 ` Carl-Daniel Hailfinger
@ 2003-10-20 9:44 ` Nikita Danilov
2003-10-27 20:25 ` Mike Fedyk
0 siblings, 1 reply; 10+ messages in thread
From: Nikita Danilov @ 2003-10-20 9:44 UTC (permalink / raw)
To: Carl-Daniel Hailfinger; +Cc: Hans Reiser, Nishant Sharma, Reiserfs mail-list
Carl-Daniel Hailfinger writes:
> Hans Reiser wrote:
> > Nishant Sharma wrote:
> >
> >> Hi,
> >>
> >> On Fri, 17 Oct 2003, Nikita Danilov wrote:
> >>
> >>
> >>> Hello,
> >>> Reiser4 snapshot against 2.6.0-test7 has been released:
> >>> http://namesys.com/snapshots/2003.10.17/
> >>> Description is in http://namesys.com/snapshots/2003.10.17/READ.ME
> >>>
> >> Following is the relevant section of my .config in kernel src-dir.
> >>
> >> CONFIG_REISER4_FS=m
> >> # CONFIG_REISER4_FS_SYSCALL is not set
> >> # CONFIG_REISER4_LARGE_KEY is not set
>
> I thought CONFIG_REISER4_LARGE_KEY was force-enabled by now?
No. But:
1. one has to use -o key=key_short option for mkfs.reiser4 to create
file system with small keys.
2. lately we mostly test/care about large keys performance and
stability.
>
> >> # CONFIG_REISER4_CHECK is not set
> >> CONFIG_REISER4_USE_EFLUSH=y
> >> # CONFIG_REISER4_COPY_ON_CAPTURE is not set
> >> # CONFIG_REISER4_BADBLOCKS is not set
> >>
> >> What am I doing wrong? I believe that I followed the instructions
> >> in http://namesys.com/snapshots/2003.10.17/READ.ME to the letter.
> >>
> >>
> > Nikita, please determine why reiser4 tells users they can configure it
> > as a module when they cannot.
Nikita.
>
>
> Carl-Daniel
>
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: new reiser4 snapshot
2003-10-20 9:44 ` Nikita Danilov
@ 2003-10-27 20:25 ` Mike Fedyk
2003-10-27 21:28 ` Hans Reiser
0 siblings, 1 reply; 10+ messages in thread
From: Mike Fedyk @ 2003-10-27 20:25 UTC (permalink / raw)
To: Nikita Danilov
Cc: Carl-Daniel Hailfinger, Hans Reiser, Nishant Sharma,
Reiserfs mail-list
On Mon, Oct 20, 2003 at 01:44:22PM +0400, Nikita Danilov wrote:
> Carl-Daniel Hailfinger writes:
> > I thought CONFIG_REISER4_LARGE_KEY was force-enabled by now?
>
> No. But:
>
> 1. one has to use -o key=key_short option for mkfs.reiser4 to create
> file system with small keys.
>
> 2. lately we mostly test/care about large keys performance and
> stability.
So why would one want short keys in the first place then?
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: new reiser4 snapshot
2003-10-27 20:25 ` Mike Fedyk
@ 2003-10-27 21:28 ` Hans Reiser
0 siblings, 0 replies; 10+ messages in thread
From: Hans Reiser @ 2003-10-27 21:28 UTC (permalink / raw)
To: Mike Fedyk
Cc: Nikita Danilov, Carl-Daniel Hailfinger, Nishant Sharma,
Reiserfs mail-list
Mike Fedyk wrote:
>On Mon, Oct 20, 2003 at 01:44:22PM +0400, Nikita Danilov wrote:
>
>
>>Carl-Daniel Hailfinger writes:
>> > I thought CONFIG_REISER4_LARGE_KEY was force-enabled by now?
>>
>>No. But:
>>
>>1. one has to use -o key=key_short option for mkfs.reiser4 to create
>>file system with small keys.
>>
>>2. lately we mostly test/care about large keys performance and
>>stability.
>>
>>
>
>So why would one want short keys in the first place then?
>
>
>
>
If you are going to access your files in the order that you wrote them
rather than in readdir order, use short keys. You are unlikely to want
to use short keys. large keys were a controversial design decision that
worked. Someday we will write something to allow readdir order to be
creation time order (at the cost of directories not being able to scale
to large sizes effectively), and then short keys will be useful again
for a more significant fraction of users.
maybe config help needs to say that.... nikita, please fix....
--
Hans
^ permalink raw reply [flat|nested] 10+ messages in thread
end of thread, other threads:[~2003-10-27 21:28 UTC | newest]
Thread overview: 10+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-10-17 15:42 new reiser4 snapshot Nikita Danilov
2003-10-17 18:51 ` Hans Reiser
2003-10-19 14:38 ` Nishant Sharma
2003-10-19 18:09 ` Hans Reiser
2003-10-19 22:41 ` Carl-Daniel Hailfinger
2003-10-20 9:44 ` Nikita Danilov
2003-10-27 20:25 ` Mike Fedyk
2003-10-27 21:28 ` Hans Reiser
2003-10-20 9:37 ` Nikita Danilov
2003-10-20 9:36 ` Nikita Danilov
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.