Issue in my API re-implementation in my kernel 3.11

Issue in my API re-implementation in my kernel 3.11

[Mathieu Belanger]

In my kernel 3.11 on my git, I have re-implemented an API because I  
did not know how to port the reiser4 to the new API (readdir ->  
iterate).

I did not get any problem with reiser4 or reiserfs -but- after  
installing the 3.12-rc1 kernel and rebooting, the kernel was missing  
and the grub.conf was total garbage. The /boot is EXT2 on my system.

readdir is for reading dir mostly, I don't know if it's the cause of  
the corruption or if it did not sync properly but still, if you want  
to play with a ext2 partition and you use the kernel 3.11 from my git,  
reboot to another kernel before, just as precaution.

The kernel 3.12-rc1 is not affected if it's because of that readdir  
API reimplementation because it's based on the Ivan port + shrinker  
new API patchs for reiser4.

Re: Issue in my API re-implementation in my kernel 3.11

[Christopher Gentile]

I've experienced corruption in the 3.12-rc1 on the git tree and just
did a fsck.reiser4 --build-fs. Also re-downloaded the tree and
recompiled as it looks like you've made changes I'll keep you posted.



On Sat, Sep 21, 2013 at 3:18 AM, Mathieu Belanger <m@thieu.so> wrote:
> In my kernel 3.11 on my git, I have re-implemented an API because I did not
> know how to port the reiser4 to the new API (readdir -> iterate).
>
> I did not get any problem with reiser4 or reiserfs -but- after installing
> the 3.12-rc1 kernel and rebooting, the kernel was missing and the grub.conf
> was total garbage. The /boot is EXT2 on my system.
>
> readdir is for reading dir mostly, I don't know if it's the cause of the
> corruption or if it did not sync properly but still, if you want to play
> with a ext2 partition and you use the kernel 3.11 from my git, reboot to
> another kernel before, just as precaution.
>
> The kernel 3.12-rc1 is not affected if it's because of that readdir API
> reimplementation because it's based on the Ivan port + shrinker new API
> patchs for reiser4.
> --
> To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Issue in my API re-implementation in my kernel 3.11

[Chris Gentile]

Mathieu et, all ,

Fixes seemed to have fixed the corruption issue thus far. Re-emerging
1000+ packages on Gentoo (tons of small reads/writes on SSD) with no
problems yet. Also, just downloaded the Linux-next patch from this
morning
(https://www.kernel.org/pub/linux/kernel/next/patch-v3.12-rc1-next-20130923.xz)
and applied on top of your Reiser4-for-3.12-rc1 branch recompiled
re-installed NVidia (thanks for fixing that btw!) and resuming compiling
as we speak. Going strong still and fast.

Not sure if it makes much difference but I use no compression just
mkfs.reiser4 -o formatting=smart on SSD drives.

Are there any special options I should be using on SSD arrays?

Addressing the compression angle, does formatting an array with one of
the compression options (lzo perhaps?) have significant overhead?
Storage space is not a concern however if there is very little overhead
I'd like to backup my data and do some benchmarking.  Please advise at
your convenience.


Thanks again,
Chris

On 09/21/13 19:53, Christopher Gentile wrote:
> I've experienced corruption in the 3.12-rc1 on the git tree and just
> did a fsck.reiser4 --build-fs. Also re-downloaded the tree and
> recompiled as it looks like you've made changes I'll keep you posted.
>
>
>
> On Sat, Sep 21, 2013 at 3:18 AM, Mathieu Belanger <m@thieu.so> wrote:
>> In my kernel 3.11 on my git, I have re-implemented an API because I did not
>> know how to port the reiser4 to the new API (readdir -> iterate).
>>
>> I did not get any problem with reiser4 or reiserfs -but- after installing
>> the 3.12-rc1 kernel and rebooting, the kernel was missing and the grub.conf
>> was total garbage. The /boot is EXT2 on my system.
>>
>> readdir is for reading dir mostly, I don't know if it's the cause of the
>> corruption or if it did not sync properly but still, if you want to play
>> with a ext2 partition and you use the kernel 3.11 from my git, reboot to
>> another kernel before, just as precaution.
>>
>> The kernel 3.12-rc1 is not affected if it's because of that readdir API
>> reimplementation because it's based on the Ivan port + shrinker new API
>> patchs for reiser4.
>> --
>> To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Issue in my API re-implementation in my kernel 3.11

[Edward Shishkin]

23.09.2013 13:01, Chris Gentile пишет:
> Mathieu et, all ,
>
> Fixes seemed to have fixed the corruption issue thus far. Re-emerging
> 1000+ packages on Gentoo (tons of small reads/writes on SSD) with no
> problems yet. Also, just downloaded the Linux-next patch from this
> morning
> (https://www.kernel.org/pub/linux/kernel/next/patch-v3.12-rc1-next-20130923.xz)
> and applied on top of your Reiser4-for-3.12-rc1 branch recompiled
> re-installed NVidia (thanks for fixing that btw!) and resuming compiling
> as we speak. Going strong still and fast.
>
> Not sure if it makes much difference but I use no compression just
> mkfs.reiser4 -o formatting=smart on SSD drives.


Which version of mkfs (mkfs.reiser4 --version)?
In 1.0.8 compression is enabled by default.


>
> Are there any special options I should be using on SSD arrays?
>
> Addressing the compression angle, does formatting an array with one of
> the compression options (lzo perhaps?) have significant overhead?


What overhead do you mean?


> Storage space is not a concern however if there is very little overhead
> I'd like to backup my data and do some benchmarking.  Please advise at
> your convenience.
>
>
> Thanks again,
> Chris
>
> On 09/21/13 19:53, Christopher Gentile wrote:
>> I've experienced corruption in the 3.12-rc1 on the git tree and just
>> did a fsck.reiser4 --build-fs. Also re-downloaded the tree and
>> recompiled as it looks like you've made changes I'll keep you posted.
>>
>>
>>
>> On Sat, Sep 21, 2013 at 3:18 AM, Mathieu Belanger <m@thieu.so> wrote:
>>> In my kernel 3.11 on my git, I have re-implemented an API because I did not
>>> know how to port the reiser4 to the new API (readdir -> iterate).
>>>
>>> I did not get any problem with reiser4 or reiserfs -but- after installing
>>> the 3.12-rc1 kernel and rebooting, the kernel was missing and the grub.conf
>>> was total garbage. The /boot is EXT2 on my system.
>>>
>>> readdir is for reading dir mostly, I don't know if it's the cause of the
>>> corruption or if it did not sync properly but still, if you want to play
>>> with a ext2 partition and you use the kernel 3.11 from my git, reboot to
>>> another kernel before, just as precaution.
>>>
>>> The kernel 3.12-rc1 is not affected if it's because of that readdir API
>>> reimplementation because it's based on the Ivan port + shrinker new API
>>> patchs for reiser4.
>>> --
>>> To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
>>> the body of a message to majordomo@vger.kernel.org
>>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> --
> To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

--
To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html