85xx Address space query

85xx Address space query

[kernel mailz]

[-- Attachment #1: Type: text/plain, Size: 768 bytes --]

Hi,

I am a newbie, trying to learn but have a few queries, nice if you could
respond
For linux on 85xx systems...

(a) Kernel code runs in PR=0 AS=0 and PID=0, which user space application
run in PR=1 AS=0 and PID 1-255.
Is this correct.

(b) I am writing a small program where the application code opens invokes a
ioctl call and passes a buffer pointer ( say 0x10000 in user space)
Now the driver code is using copy_from_user.
How this works internally ?

1. User code executes ioctl
2. interrupt goes to the kernel
3. ioctl handler in driver gets invoked
The buffer pointer still contains 0x10000.

How kernel code running in PR=0 accesses it and does the copy. I am not able
to see a address space switch in the asm code of copy_tofrom_user.

Please respond

-TRX

[-- Attachment #2: Type: text/html, Size: 855 bytes --]

Re: 85xx Address space query

[Kumar Gala]

On Jun 24, 2009, at 4:44 AM, kernel mailz wrote:

> Hi,
>
> I am a newbie, trying to learn but have a few queries, nice if you  
> could respond
> For linux on 85xx systems...
>
> (a) Kernel code runs in PR=0 AS=0 and PID=0, which user space  
> application run in PR=1 AS=0 and PID 1-255.
> Is this correct.

correct.

> (b) I am writing a small program where the application code opens  
> invokes a ioctl call and passes a buffer pointer ( say 0x10000 in  
> user space)
> Now the driver code is using copy_from_user.
> How this works internally ?
>
> 1. User code executes ioctl
> 2. interrupt goes to the kernel

On the interrupt the PR changes from 0 -> 1

> 3. ioctl handler in driver gets invoked
> The buffer pointer still contains 0x10000.
>
> How kernel code running in PR=0 accesses it and does the copy. I am  
> not able to see a address space switch in the asm code of  
> copy_tofrom_user.

There isn't a address space switch.  But address spaces exist at the  
same time.  The user app is given 0..0xc000_0000 and the kernel uses  
0xc000_0000..0xffff_ffff.

- k

Re: 85xx Address space query

[kernel mailz]

On Wed, Jun 24, 2009 at 6:45 PM, Kumar Gala<galak@kernel.crashing.org> wrot=
e:
>
> On Jun 24, 2009, at 4:44 AM, kernel mailz wrote:
>
>> Hi,
>>
>> I am a newbie, trying to learn but have a few queries, nice if you could
>> respond
>> For linux on 85xx systems...
>>
>> (a) Kernel code runs in PR=3D0 AS=3D0 and PID=3D0, which user space appl=
ication
>> run in PR=3D1 AS=3D0 and PID 1-255.
>> Is this correct.
>
> correct.
>
>> (b) I am writing a small program where the application code opens invoke=
s
>> a ioctl call and passes a buffer pointer ( say 0x10000 in user space)
>> Now the driver code is using copy_from_user.
>> How this works internally ?
>>
>> 1. User code executes ioctl
>> 2. interrupt goes to the kernel
>
> On the interrupt the PR changes from 0 -> 1
>
>> 3. ioctl handler in driver gets invoked
>> The buffer pointer still contains 0x10000.
>>
>> How kernel code running in PR=3D0 accesses it and does the copy. I am no=
t
>> able to see a address space switch in the asm code of copy_tofrom_user.
>
> There isn't a address space switch. =A0But address spaces exist at the sa=
me
> time. =A0The user app is given 0..0xc000_0000 and the kernel uses
> 0xc000_0000..0xffff_ffff.
>
Ah Ok, I get it Thanks Kumar

But If the app was running with PID=3D1, interrupt occurs, kernel code
gets executed in PID=3D1, how does the kernel handle this ? and goes
back to PID=3D0, since its translations are all in PID=3D0

> - k
>

Re: 85xx Address space query

[Scott Wood]

kernel mailz wrote:
> But If the app was running with PID=1, interrupt occurs, kernel code
> gets executed in PID=1, how does the kernel handle this ? and goes
> back to PID=0, since its translations are all in PID=0

PID 0 is special, it's mappings are present regardless of the value of 
the PID register.

-Scott

Re: 85xx Address space query

[kernel mailz]

So this means
when kernel gets interrupted by app which may be in PID=5 (say)
kernel translations for PID=0 remain valid ?
I am not able to follow Scott

-TZ

On Thu, Jun 25, 2009 at 4:22 AM, Scott Wood<scottwood@freescale.com> wrote:
> kernel mailz wrote:
>>
>> But If the app was running with PID=1, interrupt occurs, kernel code
>> gets executed in PID=1, how does the kernel handle this ? and goes
>> back to PID=0, since its translations are all in PID=0
>
> PID 0 is special, it's mappings are present regardless of the value of the
> PID register.
>
> -Scott
>

Re: 85xx Address space query

[Benjamin Herrenschmidt]

> > 1. User code executes ioctl
> > 2. interrupt goes to the kernel
> 
> On the interrupt the PR changes from 0 -> 1

The other way around actually :-)

> > 3. ioctl handler in driver gets invoked
> > The buffer pointer still contains 0x10000.
> >
> > How kernel code running in PR=0 accesses it and does the copy. I am  
> > not able to see a address space switch in the asm code of  
> > copy_tofrom_user.
> 
> There isn't a address space switch.  But address spaces exist at the  
> same time.  The user app is given 0..0xc000_0000 and the kernel uses  
> 0xc000_0000..0xffff_ffff.

You may want to add, to make things clearer, that the HW treats PID 0
specially. Translations in the TLB that have PID (we call it TID in the
TLB entry) 0 match any value of the PID register.

So kernel pages all have TID = 0, which means they are visible to all
processes, but have permissions set such that only the supervisor (ie.
PR = 0) can actually access them. MSR:PR is automatically switched to 0
by the processor when taking an interrupt, allowing the kernel thus to
access both its own pages and the pages of the current process.

Cheers,
Ben.

Re: 85xx Address space query

[Benjamin Herrenschmidt]

On Wed, 2009-06-24 at 23:16 +0530, kernel mailz wrote:
> But If the app was running with PID=1, interrupt occurs, kernel code
> gets executed in PID=1, how does the kernel handle this ? and goes
> back to PID=0, since its translations are all in PID=0

PID 0 is a special case in HW. TLB entries with TID=0 will match any PID
value.

Cheers,
Ben.

Re: 85xx Address space query

[Kumar Gala]

That is correct.  The PID = 0 translations are always valid.

- k

On Jun 25, 2009, at 5:51 AM, kernel mailz wrote:

> So this means
> when kernel gets interrupted by app which may be in PID=5 (say)
> kernel translations for PID=0 remain valid ?
> I am not able to follow Scott
>
> -TZ
>
> On Thu, Jun 25, 2009 at 4:22 AM, Scott Wood<scottwood@freescale.com>  
> wrote:
>> kernel mailz wrote:
>>>
>>> But If the app was running with PID=1, interrupt occurs, kernel code
>>> gets executed in PID=1, how does the kernel handle this ? and goes
>>> back to PID=0, since its translations are all in PID=0
>>
>> PID 0 is special, it's mappings are present regardless of the value  
>> of the
>> PID register.
>>
>> -Scott
>>
> _______________________________________________
> Linuxppc-dev mailing list
> Linuxppc-dev@lists.ozlabs.org
> https://lists.ozlabs.org/listinfo/linuxppc-dev

Re: 85xx Address space query

[kernel mailz]

If an aap has to be placed in AS=1 and it issues an ioctl, kernel
needs to be modified ?
I guess the PID=0 trick will work when AS is same

right ?



On 6/25/09, Kumar Gala <galak@kernel.crashing.org> wrote:
> That is correct.  The PID = 0 translations are always valid.
>
> - k
>
> On Jun 25, 2009, at 5:51 AM, kernel mailz wrote:
>
>> So this means
>> when kernel gets interrupted by app which may be in PID=5 (say)
>> kernel translations for PID=0 remain valid ?
>> I am not able to follow Scott
>>
>> -TZ
>>
>> On Thu, Jun 25, 2009 at 4:22 AM, Scott Wood<scottwood@freescale.com>
>> wrote:
>>> kernel mailz wrote:
>>>>
>>>> But If the app was running with PID=1, interrupt occurs, kernel code
>>>> gets executed in PID=1, how does the kernel handle this ? and goes
>>>> back to PID=0, since its translations are all in PID=0
>>>
>>> PID 0 is special, it's mappings are present regardless of the value
>>> of the
>>> PID register.
>>>
>>> -Scott
>>>
>> _______________________________________________
>> Linuxppc-dev mailing list
>> Linuxppc-dev@lists.ozlabs.org
>> https://lists.ozlabs.org/listinfo/linuxppc-dev
>
>

Re: 85xx Address space query

[Kumar Gala]

On Jun 25, 2009, at 12:52 PM, kernel mailz wrote:

> If an aap has to be placed in AS=1 and it issues an ioctl, kernel
> needs to be modified ?

Correct, this would be a significant change to the kernel.

> I guess the PID=0 trick will work when AS is same
>
> right ?

correct.

- k

>
> On 6/25/09, Kumar Gala <galak@kernel.crashing.org> wrote:
>> That is correct.  The PID = 0 translations are always valid.
>>
>> - k
>>
>> On Jun 25, 2009, at 5:51 AM, kernel mailz wrote:
>>
>>> So this means
>>> when kernel gets interrupted by app which may be in PID=5 (say)
>>> kernel translations for PID=0 remain valid ?
>>> I am not able to follow Scott
>>>
>>> -TZ
>>>
>>> On Thu, Jun 25, 2009 at 4:22 AM, Scott Wood<scottwood@freescale.com>
>>> wrote:
>>>> kernel mailz wrote:
>>>>>
>>>>> But If the app was running with PID=1, interrupt occurs, kernel  
>>>>> code
>>>>> gets executed in PID=1, how does the kernel handle this ? and goes
>>>>> back to PID=0, since its translations are all in PID=0
>>>>
>>>> PID 0 is special, it's mappings are present regardless of the value
>>>> of the
>>>> PID register.
>>>>
>>>> -Scott
>>>>
>>> _______________________________________________
>>> Linuxppc-dev mailing list
>>> Linuxppc-dev@lists.ozlabs.org
>>> https://lists.ozlabs.org/listinfo/linuxppc-dev
>>
>>

Re: 85xx Address space query

[Scott Wood]

kernel mailz wrote:
> So this means
> when kernel gets interrupted by app which may be in PID=5 (say)
> kernel translations for PID=0 remain valid ?
> I am not able to follow Scott

Yes, exactly.  They're valid even in userspace, except to the extent that the 
kernel marks them supervisor-only.

-Scott