diff mbox

[RFC] arm64: ioremap: add ioremap_cache macro

Message ID 1444616898-5057-1-git-send-email-yalin.wang2010@gmail.com (mailing list archive)
State New, archived
Headers show

Commit Message

yalin wang Oct. 12, 2015, 2:28 a.m. UTC
Add ioremap_cache macro, because some code will test if this macro
is defined or not, and will generate a generric version if not defined,
for example, memremap.c do like this.

Signed-off-by: yalin wang <yalin.wang2010@gmail.com>
---
 arch/arm64/include/asm/io.h | 1 +
 1 file changed, 1 insertion(+)

Comments

Catalin Marinas Oct. 13, 2015, 2:10 p.m. UTC | #1
On Mon, Oct 12, 2015 at 10:28:18AM +0800, yalin wang wrote:
> Add ioremap_cache macro, because some code will test if this macro
> is defined or not, and will generate a generric version if not defined,
> for example, memremap.c do like this.
> 
> Signed-off-by: yalin wang <yalin.wang2010@gmail.com>
> ---
>  arch/arm64/include/asm/io.h | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/arch/arm64/include/asm/io.h b/arch/arm64/include/asm/io.h
> index 44be1e0..d6b620c 100644
> --- a/arch/arm64/include/asm/io.h
> +++ b/arch/arm64/include/asm/io.h
> @@ -172,6 +172,7 @@ extern void __iomem *ioremap_cache(phys_addr_t phys_addr, size_t size);
>  #define ioremap_wc(addr, size)		__ioremap((addr), (size), __pgprot(PROT_NORMAL_NC))
>  #define ioremap_wt(addr, size)		__ioremap((addr), (size), __pgprot(PROT_DEVICE_nGnRE))
>  #define iounmap				__iounmap
> +#define ioremap_cache                  ioremap_cache

Applied. Thanks.
Will Deacon Oct. 13, 2015, 2:23 p.m. UTC | #2
On Tue, Oct 13, 2015 at 03:10:55PM +0100, Catalin Marinas wrote:
> On Mon, Oct 12, 2015 at 10:28:18AM +0800, yalin wang wrote:
> > Add ioremap_cache macro, because some code will test if this macro
> > is defined or not, and will generate a generric version if not defined,
> > for example, memremap.c do like this.
> > 
> > Signed-off-by: yalin wang <yalin.wang2010@gmail.com>
> > ---
> >  arch/arm64/include/asm/io.h | 1 +
> >  1 file changed, 1 insertion(+)
> > 
> > diff --git a/arch/arm64/include/asm/io.h b/arch/arm64/include/asm/io.h
> > index 44be1e0..d6b620c 100644
> > --- a/arch/arm64/include/asm/io.h
> > +++ b/arch/arm64/include/asm/io.h
> > @@ -172,6 +172,7 @@ extern void __iomem *ioremap_cache(phys_addr_t phys_addr, size_t size);
> >  #define ioremap_wc(addr, size)		__ioremap((addr), (size), __pgprot(PROT_NORMAL_NC))
> >  #define ioremap_wt(addr, size)		__ioremap((addr), (size), __pgprot(PROT_DEVICE_nGnRE))
> >  #define iounmap				__iounmap
> > +#define ioremap_cache                  ioremap_cache
> 
> Applied. Thanks.

I'm not sure we want this. See:

  https://lkml.org/lkml/2015/10/9/699

Will
Arnd Bergmann Oct. 13, 2015, 2:24 p.m. UTC | #3
On Tuesday 13 October 2015 15:10:55 Catalin Marinas wrote:
> On Mon, Oct 12, 2015 at 10:28:18AM +0800, yalin wang wrote:
> > Add ioremap_cache macro, because some code will test if this macro
> > is defined or not, and will generate a generric version if not defined,
> > for example, memremap.c do like this.
> > 
> > Signed-off-by: yalin wang <yalin.wang2010@gmail.com>
> > ---
> >  arch/arm64/include/asm/io.h | 1 +
> >  1 file changed, 1 insertion(+)
> > 
> > diff --git a/arch/arm64/include/asm/io.h b/arch/arm64/include/asm/io.h
> > index 44be1e0..d6b620c 100644
> > --- a/arch/arm64/include/asm/io.h
> > +++ b/arch/arm64/include/asm/io.h
> > @@ -172,6 +172,7 @@ extern void __iomem *ioremap_cache(phys_addr_t phys_addr, size_t size);
> >  #define ioremap_wc(addr, size)               __ioremap((addr), (size), __pgprot(PROT_NORMAL_NC))
> >  #define ioremap_wt(addr, size)               __ioremap((addr), (size), __pgprot(PROT_DEVICE_nGnRE))
> >  #define iounmap                              __iounmap
> > +#define ioremap_cache                  ioremap_cache
> 
> Applied. Thanks.
> 

I think this will conflict with Dan Williams' series that removes
ioremap_cache from the kernel and replaces it with memremap.c

	Arnd
Catalin Marinas Oct. 13, 2015, 3:20 p.m. UTC | #4
On Tue, Oct 13, 2015 at 03:23:27PM +0100, Will Deacon wrote:
> On Tue, Oct 13, 2015 at 03:10:55PM +0100, Catalin Marinas wrote:
> > On Mon, Oct 12, 2015 at 10:28:18AM +0800, yalin wang wrote:
> > > Add ioremap_cache macro, because some code will test if this macro
> > > is defined or not, and will generate a generric version if not defined,
> > > for example, memremap.c do like this.
> > > 
> > > Signed-off-by: yalin wang <yalin.wang2010@gmail.com>
> > > ---
> > >  arch/arm64/include/asm/io.h | 1 +
> > >  1 file changed, 1 insertion(+)
> > > 
> > > diff --git a/arch/arm64/include/asm/io.h b/arch/arm64/include/asm/io.h
> > > index 44be1e0..d6b620c 100644
> > > --- a/arch/arm64/include/asm/io.h
> > > +++ b/arch/arm64/include/asm/io.h
> > > @@ -172,6 +172,7 @@ extern void __iomem *ioremap_cache(phys_addr_t phys_addr, size_t size);
> > >  #define ioremap_wc(addr, size)		__ioremap((addr), (size), __pgprot(PROT_NORMAL_NC))
> > >  #define ioremap_wt(addr, size)		__ioremap((addr), (size), __pgprot(PROT_DEVICE_nGnRE))
> > >  #define iounmap				__iounmap
> > > +#define ioremap_cache                  ioremap_cache
> > 
> > Applied. Thanks.
> 
> I'm not sure we want this. See:
> 
>   https://lkml.org/lkml/2015/10/9/699

Thanks Will and Arnd, I missed this. Patch reverted.
yalin wang Oct. 14, 2015, 2:41 a.m. UTC | #5
> On Oct 13, 2015, at 23:20, Catalin Marinas <catalin.marinas@arm.com> wrote:
> 
> On Tue, Oct 13, 2015 at 03:23:27PM +0100, Will Deacon wrote:
>> On Tue, Oct 13, 2015 at 03:10:55PM +0100, Catalin Marinas wrote:
>>> On Mon, Oct 12, 2015 at 10:28:18AM +0800, yalin wang wrote:
>>>> Add ioremap_cache macro, because some code will test if this macro
>>>> is defined or not, and will generate a generric version if not defined,
>>>> for example, memremap.c do like this.
>>>> 
>>>> Signed-off-by: yalin wang <yalin.wang2010@gmail.com>
>>>> ---
>>>> arch/arm64/include/asm/io.h | 1 +
>>>> 1 file changed, 1 insertion(+)
>>>> 
>>>> diff --git a/arch/arm64/include/asm/io.h b/arch/arm64/include/asm/io.h
>>>> index 44be1e0..d6b620c 100644
>>>> --- a/arch/arm64/include/asm/io.h
>>>> +++ b/arch/arm64/include/asm/io.h
>>>> @@ -172,6 +172,7 @@ extern void __iomem *ioremap_cache(phys_addr_t phys_addr, size_t size);
>>>> #define ioremap_wc(addr, size)		__ioremap((addr), (size), __pgprot(PROT_NORMAL_NC))
>>>> #define ioremap_wt(addr, size)		__ioremap((addr), (size), __pgprot(PROT_DEVICE_nGnRE))
>>>> #define iounmap				__iounmap
>>>> +#define ioremap_cache                  ioremap_cache
>>> 
>>> Applied. Thanks.
>> 
>> I'm not sure we want this. See:
>> 
>>  https://lkml.org/lkml/2015/10/9/699
> 
> Thanks Will and Arnd, I missed this. Patch reverted.
> 
i don’t understand why conflict with Dan Williams’ patch.
Dan Williams ’s patch also define ioremap_cache  for arch ia64  & arch sh & arch xtensa ,
i see this :
# git show   92281dee825f
am i miss something?

Thanks
Arnd Bergmann Oct. 14, 2015, 8:23 a.m. UTC | #6
On Wednesday 14 October 2015 10:41:26 yalin wang wrote:

> > On Oct 13, 2015, at 23:20, Catalin Marinas <catalin.marinas@arm.com> wrote:
> >> I'm not sure we want this. See:
> >> 
> >>  https://lkml.org/lkml/2015/10/9/699
> > 
> > Thanks Will and Arnd, I missed this. Patch reverted.
> > 
> i don’t understand why conflict with Dan Williams’ patch.
> Dan Williams ’s patch also define ioremap_cache  for arch ia64  & arch sh & arch xtensa ,
> i see this :
> # git show   92281dee825f
> am i miss something?

I meant the new series, see https://lkml.org/lkml/2015/10/9/716 for the
patch that removes it again.

	Arnd
yalin wang Oct. 14, 2015, 12:01 p.m. UTC | #7
> On Oct 14, 2015, at 16:23, Arnd Bergmann <arnd@arndb.de> wrote:
> 
> On Wednesday 14 October 2015 10:41:26 yalin wang wrote:
> 
>>> On Oct 13, 2015, at 23:20, Catalin Marinas <catalin.marinas@arm.com> wrote:
>>>> I'm not sure we want this. See:
>>>> 
>>>> https://lkml.org/lkml/2015/10/9/699
>>> 
>>> Thanks Will and Arnd, I missed this. Patch reverted.
>>> 
>> i don’t understand why conflict with Dan Williams’ patch.
>> Dan Williams ’s patch also define ioremap_cache  for arch ia64  & arch sh & arch xtensa ,
>> i see this :
>> # git show   92281dee825f
>> am i miss something?
> 
> I meant the new series, see https://lkml.org/lkml/2015/10/9/716 for the
> patch that removes it again.
> 
> 	Arnd
Got it ,
Thanks :)
diff mbox

Patch

diff --git a/arch/arm64/include/asm/io.h b/arch/arm64/include/asm/io.h
index 44be1e0..d6b620c 100644
--- a/arch/arm64/include/asm/io.h
+++ b/arch/arm64/include/asm/io.h
@@ -172,6 +172,7 @@  extern void __iomem *ioremap_cache(phys_addr_t phys_addr, size_t size);
 #define ioremap_wc(addr, size)		__ioremap((addr), (size), __pgprot(PROT_NORMAL_NC))
 #define ioremap_wt(addr, size)		__ioremap((addr), (size), __pgprot(PROT_DEVICE_nGnRE))
 #define iounmap				__iounmap
+#define ioremap_cache                  ioremap_cache
 
 /*
  * io{read,write}{16,32}be() macros