From: Thierry Reding <treding@nvidia.com>

The recent io_remap_pfn_range() rework applied the static and inline
specifiers to the implementation of io_remap_pfn_range_pfn() on MIPS
Alchemy, mirroring the same change on other platforms. However, this
function is defined in a source file and that definition causes a
conflict with its declaration. Fix this by dropping the specifiers.

Fixes: 8efb51ad1b05 ("mm: abstract io_remap_pfn_range() based on PFN")
Signed-off-by: Thierry Reding <treding@nvidia.com>
---
 arch/mips/alchemy/common/setup.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/arch/mips/alchemy/common/setup.c b/arch/mips/alchemy/common/setup.c
index c35b4f809d51..992134a8c23a 100644
--- a/arch/mips/alchemy/common/setup.c
+++ b/arch/mips/alchemy/common/setup.c
@@ -94,8 +94,7 @@ phys_addr_t fixup_bigphys_addr(phys_addr_t phys_addr, phys_addr_t size)
 	return phys_addr;
 }
 
-static inline unsigned long io_remap_pfn_range_pfn(unsigned long pfn,
-		unsigned long size)
+unsigned long io_remap_pfn_range_pfn(unsigned long pfn, unsigned long size)
 {
 	phys_addr_t phys_addr = fixup_bigphys_addr(pfn << PAGE_SHIFT, size);
 
-- 
2.51.0

On Wed, Oct 29, 2025 at 04:03:16PM +0100, Thierry Reding wrote:
> From: Thierry Reding <treding@nvidia.com>
> 
> The recent io_remap_pfn_range() rework applied the static and inline
> specifiers to the implementation of io_remap_pfn_range_pfn() on MIPS
> Alchemy, mirroring the same change on other platforms. However, this
> function is defined in a source file and that definition causes a
> conflict with its declaration. Fix this by dropping the specifiers.
> 
> Fixes: 8efb51ad1b05 ("mm: abstract io_remap_pfn_range() based on PFN")
> Signed-off-by: Thierry Reding <treding@nvidia.com>
> ---
>  arch/mips/alchemy/common/setup.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
> 
> diff --git a/arch/mips/alchemy/common/setup.c b/arch/mips/alchemy/common/setup.c
> index c35b4f809d51..992134a8c23a 100644
> --- a/arch/mips/alchemy/common/setup.c
> +++ b/arch/mips/alchemy/common/setup.c
> @@ -94,8 +94,7 @@ phys_addr_t fixup_bigphys_addr(phys_addr_t phys_addr, phys_addr_t size)
>  	return phys_addr;
>  }
>  
> -static inline unsigned long io_remap_pfn_range_pfn(unsigned long pfn,
> -		unsigned long size)
> +unsigned long io_remap_pfn_range_pfn(unsigned long pfn, unsigned long size)
>  {
>  	phys_addr_t phys_addr = fixup_bigphys_addr(pfn << PAGE_SHIFT, size);
>  
> -- 
> 2.51.0

applied to mips-fixes.

Thomas.

-- 
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea.                                                [ RFC1925, 2.3 ]

On 10/29/25 08:03, Thierry Reding wrote:
> From: Thierry Reding <treding@nvidia.com>
> 
> The recent io_remap_pfn_range() rework applied the static and inline
> specifiers to the implementation of io_remap_pfn_range_pfn() on MIPS
> Alchemy, mirroring the same change on other platforms. However, this
> function is defined in a source file and that definition causes a
> conflict with its declaration. Fix this by dropping the specifiers.
> 
> Fixes: 8efb51ad1b05 ("mm: abstract io_remap_pfn_range() based on PFN")
> Signed-off-by: Thierry Reding <treding@nvidia.com>

Tested-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>

Thanks!

Thomas, can you pick this up so we have working alchemy builds again? 
Thanks!
-- 
Florian

On Mon, Dec 08, 2025 at 10:16:29AM -0800, Florian Fainelli wrote:
> On 10/29/25 08:03, Thierry Reding wrote:
> > From: Thierry Reding <treding@nvidia.com>
> > 
> > The recent io_remap_pfn_range() rework applied the static and inline
> > specifiers to the implementation of io_remap_pfn_range_pfn() on MIPS
> > Alchemy, mirroring the same change on other platforms. However, this
> > function is defined in a source file and that definition causes a
> > conflict with its declaration. Fix this by dropping the specifiers.
> > 
> > Fixes: 8efb51ad1b05 ("mm: abstract io_remap_pfn_range() based on PFN")
> > Signed-off-by: Thierry Reding <treding@nvidia.com>
> 
> Tested-by: Florian Fainelli <f.fainelli@gmail.com>
> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
> 
> Thanks!
> 
> Thomas, can you pick this up so we have working alchemy builds again?

well, I've acked the patch for inclusion in the tree where the broken
commit was applied, but I could take it.

I'll correct the Fixes tag to

Fixes: c707a68f9468 ("mm: abstract io_remap_pfn_range() based on PFN")

when applying.

Thomas.

-- 
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea.                                                [ RFC1925, 2.3 ]

On Wed, Dec 10, 2025 at 05:38:52PM +0100, Thomas Bogendoerfer wrote:
> On Mon, Dec 08, 2025 at 10:16:29AM -0800, Florian Fainelli wrote:
> > On 10/29/25 08:03, Thierry Reding wrote:
> > > From: Thierry Reding <treding@nvidia.com>
> > > 
> > > The recent io_remap_pfn_range() rework applied the static and inline
> > > specifiers to the implementation of io_remap_pfn_range_pfn() on MIPS
> > > Alchemy, mirroring the same change on other platforms. However, this
> > > function is defined in a source file and that definition causes a
> > > conflict with its declaration. Fix this by dropping the specifiers.
> > > 
> > > Fixes: 8efb51ad1b05 ("mm: abstract io_remap_pfn_range() based on PFN")
> > > Signed-off-by: Thierry Reding <treding@nvidia.com>
> > 
> > Tested-by: Florian Fainelli <f.fainelli@gmail.com>
> > Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
> > 
> > Thanks!
> > 
> > Thomas, can you pick this up so we have working alchemy builds again?
> 
> well, I've acked the patch for inclusion in the tree where the broken
> commit was applied, but I could take it.

but this has to wait until -rc1, since the breaking commit is not
in my mips-next branch, yet (and mergign current upstream is not
an option there).

Thomas.

-- 
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea.                                                [ RFC1925, 2.3 ]

On Wed, Oct 29, 2025 at 04:03:16PM +0100, Thierry Reding wrote:
> From: Thierry Reding <treding@nvidia.com>
> 
> The recent io_remap_pfn_range() rework applied the static and inline
> specifiers to the implementation of io_remap_pfn_range_pfn() on MIPS
> Alchemy, mirroring the same change on other platforms. However, this
> function is defined in a source file and that definition causes a
> conflict with its declaration. Fix this by dropping the specifiers.
> 
> Fixes: 8efb51ad1b05 ("mm: abstract io_remap_pfn_range() based on PFN")

I guess this patch should be applied to the tree, where this commit
is in.

> Signed-off-by: Thierry Reding <treding@nvidia.com>

Acked-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>

-- 
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea.                                                [ RFC1925, 2.3 ]