From: Nitin Gupta on
Hi All,

On 05/24/2010 07:48 PM, Nitin Gupta wrote:
> Creates RAM based block devices: /dev/zramX (X = 0, 1, ...).
> Pages written to these disks are compressed and stored in memory
> itself. These disks allow very fast I/O and compression provides
> good amounts of memory savings.
>
> This is enhancement over existing ramzswap driver which creates
> virtual block devices (/dev/ramzswapX) which could be used only
> as swap disks.
>
> Now, with the ability to handle any kind of I/O request, zram
> devices have lot more use cases:
> - /tmp storage
> - various caches under /var
> - swap disks
> - maybe even more! :)
>
> Performance numbers can be found at:
> http://code.google.com/p/compcache/wiki/zramperf
>

Any reviews/comments, please?

I also performed iozone tests to stress zram:
http://code.google.com/p/compcache/wiki/zramIOzone

Note that the effect of pagecache could not be completely
avoided, even when pagecache was dropped every 1 second.
So, this iozone test should not be considered for speed
measurements but as a stress test only. For speed tests,
you can refer 'dd' test results included in changelog (patch 1).

No errors were observed during either iozone or dd tests.

Thanks,
Nitin
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Nitin Gupta on
On 05/28/2010 09:56 PM, Greg KH wrote:
> On Fri, May 28, 2010 at 09:38:53PM +0530, Nitin Gupta wrote:
>> Hi All,
>>
>> On 05/24/2010 07:48 PM, Nitin Gupta wrote:
>>> Creates RAM based block devices: /dev/zramX (X = 0, 1, ...).
>>> Pages written to these disks are compressed and stored in memory
>>> itself. These disks allow very fast I/O and compression provides
>>> good amounts of memory savings.
>>>
>>> This is enhancement over existing ramzswap driver which creates
>>> virtual block devices (/dev/ramzswapX) which could be used only
>>> as swap disks.
>>>
>>> Now, with the ability to handle any kind of I/O request, zram
>>> devices have lot more use cases:
>>> - /tmp storage
>>> - various caches under /var
>>> - swap disks
>>> - maybe even more! :)
>>>
>>> Performance numbers can be found at:
>>> http://code.google.com/p/compcache/wiki/zramperf
>>>
>>
>> Any reviews/comments, please?
>
> We are all busy with the .35-rc1 merge work at the moment, sorry.
>

Thanks for your reply.

Ok, in the meantime, I will further test it and keep building upon it.
When rc1 workload is over, I think all this can be merged in linux-next?

Thanks,
Nitin

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Nitin Gupta on
Hi Greg,

On 05/28/2010 09:56 PM, Greg KH wrote:
> On Fri, May 28, 2010 at 09:38:53PM +0530, Nitin Gupta wrote:
>>
>> On 05/24/2010 07:48 PM, Nitin Gupta wrote:
>>> Creates RAM based block devices: /dev/zramX (X = 0, 1, ...).
>>> Pages written to these disks are compressed and stored in memory
>>> itself. These disks allow very fast I/O and compression provides
>>> good amounts of memory savings.
>>>
>>> This is enhancement over existing ramzswap driver which creates
>>> virtual block devices (/dev/ramzswapX) which could be used only
>>> as swap disks.
>>>
>>> Now, with the ability to handle any kind of I/O request, zram
>>> devices have lot more use cases:
>>> - /tmp storage
>>> - various caches under /var
>>> - swap disks
>>> - maybe even more! :)
>>>
>>> Performance numbers can be found at:
>>> http://code.google.com/p/compcache/wiki/zramperf
>>>
>>
>> Any reviews/comments, please?
>
> We are all busy with the .35-rc1 merge work at the moment, sorry.
>

Can you please apply these patches to linux-next? or maybe for .35-rc3/rc4 too?

Thanks,
Nitin
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Nitin Gupta on
On 06/17/2010 12:13 AM, Greg KH wrote:
> On Wed, Jun 16, 2010 at 07:44:20AM +0530, Nitin Gupta wrote:
>> Hi Greg,
>>
>> On 05/28/2010 09:56 PM, Greg KH wrote:
>>> On Fri, May 28, 2010 at 09:38:53PM +0530, Nitin Gupta wrote:
>>>>
>>>> On 05/24/2010 07:48 PM, Nitin Gupta wrote:
>>>>> Creates RAM based block devices: /dev/zramX (X = 0, 1, ...).
>>>>> Pages written to these disks are compressed and stored in memory
>>>>> itself. These disks allow very fast I/O and compression provides
>>>>> good amounts of memory savings.
>>>>>
>>>>> This is enhancement over existing ramzswap driver which creates
>>>>> virtual block devices (/dev/ramzswapX) which could be used only
>>>>> as swap disks.
>>>>>
>>>>> Now, with the ability to handle any kind of I/O request, zram
>>>>> devices have lot more use cases:
>>>>> - /tmp storage
>>>>> - various caches under /var
>>>>> - swap disks
>>>>> - maybe even more! :)
>>>>>
>>>>> Performance numbers can be found at:
>>>>> http://code.google.com/p/compcache/wiki/zramperf
>>>>>
>>>>
>>>> Any reviews/comments, please?
>>>
>>> We are all busy with the .35-rc1 merge work at the moment, sorry.
>>>
>>
>> Can you please apply these patches to linux-next?
>
> Yes, I'm working on it, I should get to it by the end of this week.
>
>> or maybe for .35-rc3/rc4 too?
>
> No, this is not a bugfix, why would it be acceptable for the .35 kernel
> release?
>

Since its a staging driver, I thought any kind of changes can be pulled in
mainline too. Maybe I still don't understand how -staging works :)

Anyways, thanks for taking it in linux-next.

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