Re: [kvm-commits] Revert "kvm: qemu: initialize ata ports as enabled"

Re: [kvm-commits] Revert "kvm: qemu: initialize ata ports as enabled"

[Gregory Haskins]

On Wed, 2007-07-18 at 12:20 +0300, Avi Kivity wrote:
> repository: /home/avi/kvm
> branch: master
> commit a4d8dd22718d59fbdbaddaaed0b0e0c4664f397d
> Author: Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w@public.gmane.org>
> Date:   Wed Jul 18 12:20:41 2007 +0300
> 
>     Revert "kvm: qemu: initialize ata ports as enabled"
>     
>     This reverts commit 55a3212bc2f5ecddcd4c5cdf2bfb37ad71e45ff2.  Turns out
>     it was due to a bad bios and everything works as expectd.
>     
>     Signed-off-by: Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w@public.gmane.org>

I'm not objecting to this change, but I am a little confused.  How does
reverting this fix anything?  Are the ports supposed to be on or off?
If turning them on causes things to slow down, what difference does it
make who turns them on (bios or qemu)?  Sorry if I missed the
conversation where this was already discussed.

-Greg


-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/

Re: [kvm-commits] Revert "kvm: qemu: initialize ata ports as enabled"

[Avi Kivity]

Gregory Haskins wrote:
> On Wed, 2007-07-18 at 12:20 +0300, Avi Kivity wrote:
>   
>> repository: /home/avi/kvm
>> branch: master
>> commit a4d8dd22718d59fbdbaddaaed0b0e0c4664f397d
>> Author: Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w@public.gmane.org>
>> Date:   Wed Jul 18 12:20:41 2007 +0300
>>
>>     Revert "kvm: qemu: initialize ata ports as enabled"
>>     
>>     This reverts commit 55a3212bc2f5ecddcd4c5cdf2bfb37ad71e45ff2.  Turns out
>>     it was due to a bad bios and everything works as expectd.
>>     
>>     Signed-off-by: Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w@public.gmane.org>
>>     
>
> I'm not objecting to this change, but I am a little confused.  How does
> reverting this fix anything?  Are the ports supposed to be on or off?
> If turning them on causes things to slow down, what difference does it
> make who turns them on (bios or qemu)?  Sorry if I missed the
> conversation where this was already discussed.
>
>   

Good question.  I initially guessed that when the bios turns them on, it
does other things as well, but that's not the case.



-- 
Do not meddle in the internals of kernels, for they are subtle and quick to panic.


-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/

Re: [kvm-commits] Revert "kvm: qemu: initialize ata ports as enabled"

[Gregory Haskins]

On Wed, 2007-07-18 at 15:42 +0300, Avi Kivity wrote:
>   
> 
> Good question.  I initially guessed that when the bios turns them on, it
> does other things as well, but that's not the case.


Ah, I was wondering that myself.

I noticed from Dave's diffed dmesg output that you do see the PIIX
driver reporting the ports disabled in one case, and enabled in the
other.  I also see that the drives still show up later in the dmesg
output regardless of the PIIX report, and that the hdparam tool shows
them with different DMA settings.

What I am wondering is if there is a way to see which driver actually
published that information to ide-core?  (Is this available in something
like sysfs?).  We have been assuming that the only difference is dma
mode enabled/disabled, but I am still wondering if a different driver
loaded all together.

-Greg


-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/