Re: [PATCH 03/12] m68k/mac: Don't remap SWIM MMIO region

[Michael Schmitz]

Hi Geert,

I haven't seen that weird behaviour in quite some time - we discussed
changing the arch_initcall to some later priority at that time but I
never got around to trying that. No change to mappings in head.S
either as far as I could see.

Was there any rearrangement of MM init relative to arch init since?

Cheers,

  Michael


On Tue, Apr 10, 2018 at 12:54 AM, Geert Uytterhoeven
 wrote:
> Hi Finn,
>
> On Sun, Apr 1, 2018 at 3:41 AM, Finn Thain  wrote:
>> For reasons I don't understand, calling ioremap() then iounmap() on
>> the SWIM MMIO region causes a hang on 68030 (but not on 68040).
>
> Michael Schmitz also notices strange things with ioremap() on '030.
>
>> There's no need to call ioremap() for the SWIM address range, as it lies
>> within the usual IO device region at 0x5000 , which is already mapped.
>
> by head.S, right?
>
>> --- a/drivers/block/swim.c
>> +++ b/drivers/block/swim.c
>> @@ -911,7 +911,7 @@ static int swim_probe(struct platform_device *dev)
>> goto out;
>> }
>>
>> -   swim_base = ioremap(res->start, resource_size(res));
>> +   swim_base = (struct swim __iomem *)res->start;
>
> I guess you need a __force to please sparse?
>
> Gr{oetje,eeting}s,
>
> Geert
>
> --
> Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- 
> ge...@linux-m68k.org
>
> In personal conversations with technical people, I call myself a hacker. But
> when I'm talking to journalists I just say "programmer" or something like 
> that.
> -- Linus Torvalds
> --
> To unsubscribe from this list: send the line "unsubscribe linux-m68k" in
> the body of a message to majord...@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 03/12] m68k/mac: Don't remap SWIM MMIO region

[Michael Schmitz]

Hi Geert,

I haven't seen that weird behaviour in quite some time - we discussed
changing the arch_initcall to some later priority at that time but I
never got around to trying that. No change to mappings in head.S
either as far as I could see.

Was there any rearrangement of MM init relative to arch init since?

Cheers,

  Michael


On Tue, Apr 10, 2018 at 12:54 AM, Geert Uytterhoeven
 wrote:
> Hi Finn,
>
> On Sun, Apr 1, 2018 at 3:41 AM, Finn Thain  wrote:
>> For reasons I don't understand, calling ioremap() then iounmap() on
>> the SWIM MMIO region causes a hang on 68030 (but not on 68040).
>
> Michael Schmitz also notices strange things with ioremap() on '030.
>
>> There's no need to call ioremap() for the SWIM address range, as it lies
>> within the usual IO device region at 0x5000 , which is already mapped.
>
> by head.S, right?
>
>> --- a/drivers/block/swim.c
>> +++ b/drivers/block/swim.c
>> @@ -911,7 +911,7 @@ static int swim_probe(struct platform_device *dev)
>> goto out;
>> }
>>
>> -   swim_base = ioremap(res->start, resource_size(res));
>> +   swim_base = (struct swim __iomem *)res->start;
>
> I guess you need a __force to please sparse?
>
> Gr{oetje,eeting}s,
>
> Geert
>
> --
> Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- 
> ge...@linux-m68k.org
>
> In personal conversations with technical people, I call myself a hacker. But
> when I'm talking to journalists I just say "programmer" or something like 
> that.
> -- Linus Torvalds
> --
> To unsubscribe from this list: send the line "unsubscribe linux-m68k" in
> the body of a message to majord...@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 03/12] m68k/mac: Don't remap SWIM MMIO region

[Finn Thain]

On Mon, 9 Apr 2018, Geert Uytterhoeven wrote:

> Hi Finn,
> 
> On Sun, Apr 1, 2018 at 3:41 AM, Finn Thain  
> wrote:
> > For reasons I don't understand, calling ioremap() then iounmap() on 
> > the SWIM MMIO region causes a hang on 68030 (but not on 68040).
> 
> Michael Schmitz also notices strange things with ioremap() on '030.
> 
> > There's no need to call ioremap() for the SWIM address range, as it 
> > lies within the usual IO device region at 0x5000 , which is 
> > already mapped.
> 
> by head.S, right?
> 

Right. I'll mention that in the commit log when I re-submit this series.

> > --- a/drivers/block/swim.c
> > +++ b/drivers/block/swim.c
> > @@ -911,7 +911,7 @@ static int swim_probe(struct platform_device *dev)
> > goto out;
> > }
> >
> > -   swim_base = ioremap(res->start, resource_size(res));
> > +   swim_base = (struct swim __iomem *)res->start;
> 
> I guess you need a __force to please sparse?
> 

No, sparse did not find any new issues. It appears there's an old issue 
elsewhere in the driver though. I'll prepare a new patch for that.

-- 

> Gr{oetje,eeting}s,
> 
> Geert
> 
> 

-- 

Re: [PATCH 03/12] m68k/mac: Don't remap SWIM MMIO region

[Finn Thain]

On Mon, 9 Apr 2018, Geert Uytterhoeven wrote:

> Hi Finn,
> 
> On Sun, Apr 1, 2018 at 3:41 AM, Finn Thain  
> wrote:
> > For reasons I don't understand, calling ioremap() then iounmap() on 
> > the SWIM MMIO region causes a hang on 68030 (but not on 68040).
> 
> Michael Schmitz also notices strange things with ioremap() on '030.
> 
> > There's no need to call ioremap() for the SWIM address range, as it 
> > lies within the usual IO device region at 0x5000 , which is 
> > already mapped.
> 
> by head.S, right?
> 

Right. I'll mention that in the commit log when I re-submit this series.

> > --- a/drivers/block/swim.c
> > +++ b/drivers/block/swim.c
> > @@ -911,7 +911,7 @@ static int swim_probe(struct platform_device *dev)
> > goto out;
> > }
> >
> > -   swim_base = ioremap(res->start, resource_size(res));
> > +   swim_base = (struct swim __iomem *)res->start;
> 
> I guess you need a __force to please sparse?
> 

No, sparse did not find any new issues. It appears there's an old issue 
elsewhere in the driver though. I'll prepare a new patch for that.

-- 

> Gr{oetje,eeting}s,
> 
> Geert
> 
> 

-- 

Re: [PATCH 03/12] m68k/mac: Don't remap SWIM MMIO region

[Luc Van Oostenryck]

On Mon, Apr 09, 2018 at 02:54:30PM +0200, Geert Uytterhoeven wrote:
> >
> > -   swim_base = ioremap(res->start, resource_size(res));
> > +   swim_base = (struct swim __iomem *)res->start;
> 
> I guess you need a __force to please sparse?

Hi,

Only pointer-to-pointer conversions may need __force.
The logic being if you use an integer type to hold an address
this address is typeless anyway and has no address-space, so
extra checking is not done in this case and the cast in itself
is enough.

Cheers,
-- Luc

Re: [PATCH 03/12] m68k/mac: Don't remap SWIM MMIO region

[Luc Van Oostenryck]

On Mon, Apr 09, 2018 at 02:54:30PM +0200, Geert Uytterhoeven wrote:
> >
> > -   swim_base = ioremap(res->start, resource_size(res));
> > +   swim_base = (struct swim __iomem *)res->start;
> 
> I guess you need a __force to please sparse?

Hi,

Only pointer-to-pointer conversions may need __force.
The logic being if you use an integer type to hold an address
this address is typeless anyway and has no address-space, so
extra checking is not done in this case and the cast in itself
is enough.

Cheers,
-- Luc

Re: [PATCH 03/12] m68k/mac: Don't remap SWIM MMIO region

[Geert Uytterhoeven]

Hi Finn,

On Sun, Apr 1, 2018 at 3:41 AM, Finn Thain  wrote:
> For reasons I don't understand, calling ioremap() then iounmap() on
> the SWIM MMIO region causes a hang on 68030 (but not on 68040).

Michael Schmitz also notices strange things with ioremap() on '030.

> There's no need to call ioremap() for the SWIM address range, as it lies
> within the usual IO device region at 0x5000 , which is already mapped.

by head.S, right?

> --- a/drivers/block/swim.c
> +++ b/drivers/block/swim.c
> @@ -911,7 +911,7 @@ static int swim_probe(struct platform_device *dev)
> goto out;
> }
>
> -   swim_base = ioremap(res->start, resource_size(res));
> +   swim_base = (struct swim __iomem *)res->start;

I guess you need a __force to please sparse?

Gr{oetje,eeting}s,

Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

Re: [PATCH 03/12] m68k/mac: Don't remap SWIM MMIO region

[Geert Uytterhoeven]

Hi Finn,

On Sun, Apr 1, 2018 at 3:41 AM, Finn Thain  wrote:
> For reasons I don't understand, calling ioremap() then iounmap() on
> the SWIM MMIO region causes a hang on 68030 (but not on 68040).

Michael Schmitz also notices strange things with ioremap() on '030.

> There's no need to call ioremap() for the SWIM address range, as it lies
> within the usual IO device region at 0x5000 , which is already mapped.

by head.S, right?

> --- a/drivers/block/swim.c
> +++ b/drivers/block/swim.c
> @@ -911,7 +911,7 @@ static int swim_probe(struct platform_device *dev)
> goto out;
> }
>
> -   swim_base = ioremap(res->start, resource_size(res));
> +   swim_base = (struct swim __iomem *)res->start;

I guess you need a __force to please sparse?

Gr{oetje,eeting}s,

Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds