Re: [RFT 03/13] sh: Constify ioreadX() iomem argument (as in generic implementation)

Re: [RFT 03/13] sh: Constify ioreadX() iomem argument (as in generic implementation)

[Arnd Bergmann]

On Tue, Jan 7, 2020 at 5:54 PM Krzysztof Kozlowski <krzk@kernel.org> wrote:
>
> The ioreadX() helpers have inconsistent interface.  On some architectures
> void *__iomem address argument is a pointer to const, on some not.
>
> Implementations of ioreadX() do not modify the memory under the address
> so they can be converted to a "const" version for const-safety and
> consistency among architectures.
>
> Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>

The patch looks good, but I think this has to be done together with the powerpc
version and the header that declares the function, for bisectibility.

       Arnd

Re: [RFT 06/13] arc: Constify ioreadX() iomem argument (as in generic implementation)

[Arnd Bergmann]

On Tue, Jan 7, 2020 at 5:54 PM Krzysztof Kozlowski <krzk@kernel.org> wrote:
>
> The ioreadX() helpers have inconsistent interface.  On some architectures
> void *__iomem address argument is a pointer to const, on some not.
>
> Implementations of ioreadX() do not modify the memory under the
> address so they can be converted to a "const" version for const-safety
> and consistency among architectures.
>
> Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>

The patch looks correct, but I would not bother here, as it has no
effect on the compiler or its output.

      Arnd

Re: [RFT 00/13] iomap: Constify ioreadX() iomem argument

[Geert Uytterhoeven]

Hi Krzysztof,

On Tue, Jan 7, 2020 at 5:53 PM Krzysztof Kozlowski <krzk@kernel.org> wrote:
> The ioread8/16/32() and others have inconsistent interface among the
> architectures: some taking address as const, some not.
>
> It seems there is nothing really stopping all of them to take
> pointer to const.

Shouldn't all of them take const volatile __iomem pointers?
It seems the "volatile" is missing from all but the implementations in
include/asm-generic/io.h.

> Patchset was really tested on all affected architectures.
> Build testing is in progress - I hope auto-builders will point any issues.
>
>
> Todo
> ====
> Convert also string versions (ioread16_rep() etc) if this aproach looks OK.
>
>
> Merging
> =======
> The first 5 patches - iomap, alpha, sh, parisc and powerpc - should probably go
> via one tree, or even squashed into one.

Yes, they should be squashed, cfr. Arnd's comment.
I also wouldn't bother doing the updates in patches 6-10.

The rest looks good to me.
Thanks a lot!

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@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: [RFT 02/13] alpha: Constify ioreadX() iomem argument (as in generic implementation)

[Geert Uytterhoeven]

Hi Krzysztof,

On Tue, Jan 7, 2020 at 5:53 PM Krzysztof Kozlowski <krzk@kernel.org> wrote:
> The ioreadX() helpers have inconsistent interface.  On some architectures
> void *__iomem address argument is a pointer to const, on some not.
>
> Implementations of ioreadX() do not modify the memory under the address
> so they can be converted to a "const" version for const-safety and
> consistency among architectures.
>
> Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>

> --- a/arch/alpha/include/asm/io.h
> +++ b/arch/alpha/include/asm/io.h
> @@ -151,9 +151,9 @@ static inline void generic_##NAME(TYPE b, QUAL void __iomem *addr)  \
>         alpha_mv.mv_##NAME(b, addr);                                    \
>  }
>
> -REMAP1(unsigned int, ioread8, /**/)
> -REMAP1(unsigned int, ioread16, /**/)
> -REMAP1(unsigned int, ioread32, /**/)
> +REMAP1(unsigned int, ioread8, const)
> +REMAP1(unsigned int, ioread16, const)
> +REMAP1(unsigned int, ioread32, const)

If these would become "const volatile", there would no longer be a need
for the last parameter of the REMAP1() macro.

>  REMAP1(u8, readb, const volatile)
>  REMAP1(u16, readw, const volatile)
>  REMAP1(u32, readl, const volatile)

Same for REMAP2() macro below, for iowrite*().

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@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: [RFT 00/13] iomap: Constify ioreadX() iomem argument

[Geert Uytterhoeven]

Hi Krzysztof,

On Wed, Jan 8, 2020 at 9:07 AM Geert Uytterhoeven <geert@linux-m68k.org> wrote:
> On Tue, Jan 7, 2020 at 5:53 PM Krzysztof Kozlowski <krzk@kernel.org> wrote:
> > The ioread8/16/32() and others have inconsistent interface among the
> > architectures: some taking address as const, some not.
> >
> > It seems there is nothing really stopping all of them to take
> > pointer to const.
>
> Shouldn't all of them take const volatile __iomem pointers?
> It seems the "volatile" is missing from all but the implementations in
> include/asm-generic/io.h.

As my "volatile" comment applies to iowrite*(), too, probably that should be
done in a separate patch.

Hence with patches 1-5 squashed, and for patches 11-13:
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@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: [RFT 00/13] iomap: Constify ioreadX() iomem argument

[Krzysztof Kozlowski]

On Wed, 8 Jan 2020 at 09:08, Geert Uytterhoeven <geert@linux-m68k.org> wrote:
>
> Hi Krzysztof,
>
> On Tue, Jan 7, 2020 at 5:53 PM Krzysztof Kozlowski <krzk@kernel.org> wrote:
> > The ioread8/16/32() and others have inconsistent interface among the
> > architectures: some taking address as const, some not.
> >
> > It seems there is nothing really stopping all of them to take
> > pointer to const.
>
> Shouldn't all of them take const volatile __iomem pointers?
> It seems the "volatile" is missing from all but the implementations in
> include/asm-generic/io.h.

It's kind of separate issue although I could squash it to limit
redundant changes.

> > Patchset was really tested on all affected architectures.

I just spot an error in my first message. I wanted to say:
"Patchset was NOT really tested on all affected architectures."

Obviously.


> > Build testing is in progress - I hope auto-builders will point any issues.
> >
> >
> > Todo
> > ====
> > Convert also string versions (ioread16_rep() etc) if this aproach looks OK.
> >
> >
> > Merging
> > =======
> > The first 5 patches - iomap, alpha, sh, parisc and powerpc - should probably go
> > via one tree, or even squashed into one.
>
> Yes, they should be squashed, cfr. Arnd's comment.
> I also wouldn't bother doing the updates in patches 6-10.

Indeed, thanks for comments.

Best regards,
Krzysztof

Re: [RFT 00/13] iomap: Constify ioreadX() iomem argument

[Krzysztof Kozlowski]

On Wed, 8 Jan 2020 at 09:13, Geert Uytterhoeven <geert@linux-m68k.org> wrote:
>
> Hi Krzysztof,
>
> On Wed, Jan 8, 2020 at 9:07 AM Geert Uytterhoeven <geert@linux-m68k.org> wrote:
> > On Tue, Jan 7, 2020 at 5:53 PM Krzysztof Kozlowski <krzk@kernel.org> wrote:
> > > The ioread8/16/32() and others have inconsistent interface among the
> > > architectures: some taking address as const, some not.
> > >
> > > It seems there is nothing really stopping all of them to take
> > > pointer to const.
> >
> > Shouldn't all of them take const volatile __iomem pointers?
> > It seems the "volatile" is missing from all but the implementations in
> > include/asm-generic/io.h.
>
> As my "volatile" comment applies to iowrite*(), too, probably that should be
> done in a separate patch.
>
> Hence with patches 1-5 squashed, and for patches 11-13:
> Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>

I'll add to this one also changes to ioreadX_rep() and add another
patch for volatile for reads and writes. I guess your review will be
appreciated once more because of ioreadX_rep()

Thanks,
Krzysztof

Re: [RFT 00/13] iomap: Constify ioreadX() iomem argument

[Christophe Leroy]

Le 08/01/2020 à 09:18, Krzysztof Kozlowski a écrit :
> On Wed, 8 Jan 2020 at 09:13, Geert Uytterhoeven <geert@linux-m68k.org> wrote:
>>
>> Hi Krzysztof,
>>
>> On Wed, Jan 8, 2020 at 9:07 AM Geert Uytterhoeven <geert@linux-m68k.org> wrote:
>>> On Tue, Jan 7, 2020 at 5:53 PM Krzysztof Kozlowski <krzk@kernel.org> wrote:
>>>> The ioread8/16/32() and others have inconsistent interface among the
>>>> architectures: some taking address as const, some not.
>>>>
>>>> It seems there is nothing really stopping all of them to take
>>>> pointer to const.
>>>
>>> Shouldn't all of them take const volatile __iomem pointers?
>>> It seems the "volatile" is missing from all but the implementations in
>>> include/asm-generic/io.h.
>>
>> As my "volatile" comment applies to iowrite*(), too, probably that should be
>> done in a separate patch.
>>
>> Hence with patches 1-5 squashed, and for patches 11-13:
>> Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
> 
> I'll add to this one also changes to ioreadX_rep() and add another
> patch for volatile for reads and writes. I guess your review will be
> appreciated once more because of ioreadX_rep()
> 

volatile should really only be used where deemed necessary:

https://www.kernel.org/doc/html/latest/process/volatile-considered-harmful.html

It is said: " ...  accessor functions might use volatile on 
architectures where direct I/O memory access does work. Essentially, 
each accessor call becomes a little critical section on its own and 
ensures that the access happens as expected by the programmer."

Christophe

_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

Re: [RFT 00/13] iomap: Constify ioreadX() iomem argument

[Geert Uytterhoeven]

Hi Christophe,

On Wed, Jan 8, 2020 at 9:35 AM Christophe Leroy <christophe.leroy@c-s.fr> wrote:
> Le 08/01/2020 à 09:18, Krzysztof Kozlowski a écrit :
> > On Wed, 8 Jan 2020 at 09:13, Geert Uytterhoeven <geert@linux-m68k.org> wrote:
> >> On Wed, Jan 8, 2020 at 9:07 AM Geert Uytterhoeven <geert@linux-m68k.org> wrote:
> >>> On Tue, Jan 7, 2020 at 5:53 PM Krzysztof Kozlowski <krzk@kernel.org> wrote:
> >>>> The ioread8/16/32() and others have inconsistent interface among the
> >>>> architectures: some taking address as const, some not.
> >>>>
> >>>> It seems there is nothing really stopping all of them to take
> >>>> pointer to const.
> >>>
> >>> Shouldn't all of them take const volatile __iomem pointers?
> >>> It seems the "volatile" is missing from all but the implementations in
> >>> include/asm-generic/io.h.
> >>
> >> As my "volatile" comment applies to iowrite*(), too, probably that should be
> >> done in a separate patch.
> >>
> >> Hence with patches 1-5 squashed, and for patches 11-13:
> >> Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
> >
> > I'll add to this one also changes to ioreadX_rep() and add another
> > patch for volatile for reads and writes. I guess your review will be
> > appreciated once more because of ioreadX_rep()
>
> volatile should really only be used where deemed necessary:
>
> https://www.kernel.org/doc/html/latest/process/volatile-considered-harmful.html
>
> It is said: " ...  accessor functions might use volatile on
> architectures where direct I/O memory access does work. Essentially,
> each accessor call becomes a little critical section on its own and
> ensures that the access happens as expected by the programmer."

That is exactly the use case here: all above are accessor functions.

Why would ioreadX() not need volatile, while readY() does?

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@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: [RFT 00/13] iomap: Constify ioreadX() iomem argument

[Arnd Bergmann]

On Wed, Jan 8, 2020 at 9:36 AM Christophe Leroy <christophe.leroy@c-s.fr> wrote:
> Le 08/01/2020 à 09:18, Krzysztof Kozlowski a écrit :
> > On Wed, 8 Jan 2020 at 09:13, Geert Uytterhoeven <geert@linux-m68k.org> wrote:
> > I'll add to this one also changes to ioreadX_rep() and add another
> > patch for volatile for reads and writes. I guess your review will be
> > appreciated once more because of ioreadX_rep()
> >
>
> volatile should really only be used where deemed necessary:
>
> https://www.kernel.org/doc/html/latest/process/volatile-considered-harmful.html
>
> It is said: " ...  accessor functions might use volatile on
> architectures where direct I/O memory access does work. Essentially,
> each accessor call becomes a little critical section on its own and
> ensures that the access happens as expected by the programmer."

The I/O accessors are one of the few places in which 'volatile' generally
makes sense, at least for the implementations that do a plain pointer
dereference (probably none of the ones in question here).

In case of readl/writel, this is what we do in asm-generic:

static inline u32 __raw_readl(const volatile void __iomem *addr)
{
        return *(const volatile u32 __force *)addr;
}

The __force-cast that removes the __iomem here also means that
the 'volatile' keyword could be dropped from the argument list,
as it has no real effect any more, but then there are a few drivers
that mark their iomem pointers as either 'volatile void __iomem*' or
(worse) 'volatile void *', so we keep it in the argument list to not
add warnings for those drivers.

It may be time to change these drivers to not use volatile for __iomem
pointers, but that seems out of scope for what Krzysztof is trying
to do. Ideally we would be consistent here though, either using volatile
all the time or never.

        Arnd
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

Re: [RFT 00/13] iomap: Constify ioreadX() iomem argument

[Christophe Leroy]

Hi Geert,

Le 08/01/2020 à 09:43, Geert Uytterhoeven a écrit :
> Hi Christophe,
> 
> On Wed, Jan 8, 2020 at 9:35 AM Christophe Leroy <christophe.leroy@c-s.fr> wrote:
>> Le 08/01/2020 à 09:18, Krzysztof Kozlowski a écrit :
>>> On Wed, 8 Jan 2020 at 09:13, Geert Uytterhoeven <geert@linux-m68k.org> wrote:
>>>> On Wed, Jan 8, 2020 at 9:07 AM Geert Uytterhoeven <geert@linux-m68k.org> wrote:
>>>>> On Tue, Jan 7, 2020 at 5:53 PM Krzysztof Kozlowski <krzk@kernel.org> wrote:
>>>>>> The ioread8/16/32() and others have inconsistent interface among the
>>>>>> architectures: some taking address as const, some not.
>>>>>>
>>>>>> It seems there is nothing really stopping all of them to take
>>>>>> pointer to const.
>>>>>
>>>>> Shouldn't all of them take const volatile __iomem pointers?
>>>>> It seems the "volatile" is missing from all but the implementations in
>>>>> include/asm-generic/io.h.
>>>>
>>>> As my "volatile" comment applies to iowrite*(), too, probably that should be
>>>> done in a separate patch.
>>>>
>>>> Hence with patches 1-5 squashed, and for patches 11-13:
>>>> Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
>>>
>>> I'll add to this one also changes to ioreadX_rep() and add another
>>> patch for volatile for reads and writes. I guess your review will be
>>> appreciated once more because of ioreadX_rep()
>>
>> volatile should really only be used where deemed necessary:
>>
>> https://www.kernel.org/doc/html/latest/process/volatile-considered-harmful.html
>>
>> It is said: " ...  accessor functions might use volatile on
>> architectures where direct I/O memory access does work. Essentially,
>> each accessor call becomes a little critical section on its own and
>> ensures that the access happens as expected by the programmer."
> 
> That is exactly the use case here: all above are accessor functions.
> 
> Why would ioreadX() not need volatile, while readY() does?
> 

My point was: it might be necessary for some arches and not for others.

And as pointed by Arnd, the volatile is really only necessary for the 
dereference itself, should the arch use dereferencing.

So I guess the best would be to go in the other direction: remove 
volatile keyword wherever possible instead of adding it where it is not 
needed.

Christophe

Re: [RFT 02/13] alpha: Constify ioreadX() iomem argument (as in generic implementation)

[Krzysztof Kozlowski]

On Wed, Jan 08, 2020 at 09:10:06AM +0100, Geert Uytterhoeven wrote:
> Hi Krzysztof,
> 
> On Tue, Jan 7, 2020 at 5:53 PM Krzysztof Kozlowski <krzk@kernel.org> wrote:
> > The ioreadX() helpers have inconsistent interface.  On some architectures
> > void *__iomem address argument is a pointer to const, on some not.
> >
> > Implementations of ioreadX() do not modify the memory under the address
> > so they can be converted to a "const" version for const-safety and
> > consistency among architectures.
> >
> > Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
> 
> > --- a/arch/alpha/include/asm/io.h
> > +++ b/arch/alpha/include/asm/io.h
> > @@ -151,9 +151,9 @@ static inline void generic_##NAME(TYPE b, QUAL void __iomem *addr)  \
> >         alpha_mv.mv_##NAME(b, addr);                                    \
> >  }
> >
> > -REMAP1(unsigned int, ioread8, /**/)
> > -REMAP1(unsigned int, ioread16, /**/)
> > -REMAP1(unsigned int, ioread32, /**/)
> > +REMAP1(unsigned int, ioread8, const)
> > +REMAP1(unsigned int, ioread16, const)
> > +REMAP1(unsigned int, ioread32, const)
> 
> If these would become "const volatile", there would no longer be a need
> for the last parameter of the REMAP1() macro.
> 
> >  REMAP1(u8, readb, const volatile)
> >  REMAP1(u16, readw, const volatile)
> >  REMAP1(u32, readl, const volatile)
> 
> Same for REMAP2() macro below, for iowrite*().

Good point, thanks!

Best regards,
Krzysztof

Re: [RFT 00/13] iomap: Constify ioreadX() iomem argument

[Krzysztof Kozlowski]

On Wed, Jan 08, 2020 at 09:44:36AM +0100, Arnd Bergmann wrote:
> On Wed, Jan 8, 2020 at 9:36 AM Christophe Leroy <christophe.leroy@c-s.fr> wrote:
> > Le 08/01/2020 à 09:18, Krzysztof Kozlowski a écrit :
> > > On Wed, 8 Jan 2020 at 09:13, Geert Uytterhoeven <geert@linux-m68k.org> wrote:
> > > I'll add to this one also changes to ioreadX_rep() and add another
> > > patch for volatile for reads and writes. I guess your review will be
> > > appreciated once more because of ioreadX_rep()
> > >
> >
> > volatile should really only be used where deemed necessary:
> >
> > https://www.kernel.org/doc/html/latest/process/volatile-considered-harmful.html
> >
> > It is said: " ...  accessor functions might use volatile on
> > architectures where direct I/O memory access does work. Essentially,
> > each accessor call becomes a little critical section on its own and
> > ensures that the access happens as expected by the programmer."
> 
> The I/O accessors are one of the few places in which 'volatile' generally
> makes sense, at least for the implementations that do a plain pointer
> dereference (probably none of the ones in question here).
> 
> In case of readl/writel, this is what we do in asm-generic:
> 
> static inline u32 __raw_readl(const volatile void __iomem *addr)
> {
>         return *(const volatile u32 __force *)addr;
> }

SuperH is another example:
1. ioread8_rep(void __iomem *addr, void *dst, unsigned long count)
   calls mmio_insb()

2. static inline void mmio_insb(void __iomem *addr, u8 *dst, int count)
   calls __raw_readb()

3. #define __raw_readb(a)          (__chk_io_ptr(a), *(volatile u8  __force *)(a))

Even if interface was not marked as volatile, in fact its implementation
was casting to volatile.

> The __force-cast that removes the __iomem here also means that
> the 'volatile' keyword could be dropped from the argument list,
> as it has no real effect any more, but then there are a few drivers
> that mark their iomem pointers as either 'volatile void __iomem*' or
> (worse) 'volatile void *', so we keep it in the argument list to not
> add warnings for those drivers.
> 
> It may be time to change these drivers to not use volatile for __iomem
> pointers, but that seems out of scope for what Krzysztof is trying
> to do. Ideally we would be consistent here though, either using volatile
> all the time or never.

Indeed. I guess there are no objections around const so let me send v2
for const only.

Best regards,
Krzysztof

_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

Re: [RFT 00/13] iomap: Constify ioreadX() iomem argument

[Arnd Bergmann]

On Wed, Jan 8, 2020 at 10:15 AM Krzysztof Kozlowski <krzk@kernel.org> wrote:

> > The __force-cast that removes the __iomem here also means that
> > the 'volatile' keyword could be dropped from the argument list,
> > as it has no real effect any more, but then there are a few drivers
> > that mark their iomem pointers as either 'volatile void __iomem*' or
> > (worse) 'volatile void *', so we keep it in the argument list to not
> > add warnings for those drivers.
> >
> > It may be time to change these drivers to not use volatile for __iomem
> > pointers, but that seems out of scope for what Krzysztof is trying
> > to do. Ideally we would be consistent here though, either using volatile
> > all the time or never.
>
> Indeed. I guess there are no objections around const so let me send v2
> for const only.

Ok, sounds good. Maybe mention in the changelog then that the
'volatile' in the interface is intentionally left out, and that only users
of readl/writel still have it to deal with existing drivers.

    Arnd

RE: [RFT 00/13] iomap: Constify ioreadX() iomem argument

[David Laight]

From: Christophe Leroy
> Sent: 08 January 2020 08:49
...
> And as pointed by Arnd, the volatile is really only necessary for the
> dereference itself, should the arch use dereferencing.

I've had trouble with some versions of gcc and reading of 'volatile unsigned char *'.
It tended to follow the memory read with an extra mask with 0xff.
(I suspect that internally the value landed into a temporary 'int' variable.)

I got better code using memory barriers.
So putting an asm barrier for the exact location of the memory read
either side of the read should have the desired effect without adding
extra instructions.
(You might think 'volatile' would mean that - but it doesn't.)

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)