linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH v2] sh: Convert nommu io{re,un}map() to static inline functions
@ 2022-06-20  7:01 Geert Uytterhoeven
  2022-06-20  7:06 ` Christoph Hellwig
  0 siblings, 1 reply; 2+ messages in thread
From: Geert Uytterhoeven @ 2022-06-20  7:01 UTC (permalink / raw)
  To: Yoshinori Sato, Rich Felker, Linus Torvalds, Andrew Morton
  Cc: Jonathan Cameron, linux-sh, linux-kernel, Geert Uytterhoeven,
	kernel test robot, Christoph Hellwig

Recently, nommu iounmap() was converted from a static inline function to
a macro again, basically reverting commit 4580ba4ad2e6b8dd ("sh: Convert
iounmap() macros to inline functions").  With -Werror, this leads to
build failures like:

    drivers/iio/adc/xilinx-ams.c: In function ‘ams_iounmap_ps’:
    drivers/iio/adc/xilinx-ams.c:1195:14: error: unused variable ‘ams’ [-Werror=unused-variable]
     1195 |  struct ams *ams = data;
	  |              ^~~

Fix this by replacing the macros for ioremap() and iounmap() by static
inline functions, based on <asm-generic/io.h>.

Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Fixes: 13f1fc870dd74713 ("sh: move the ioremap implementation out of line")
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Acked-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
v2:
  - Add Acked-by, Reviewed-by.

This is actually the third time this change was made, as Christoph
converted iounmap() to a macro before in commit 98c90e5ea34e98bd ("sh:
remove __iounmap"), reverting commit 733f0025f0fb43e3 ("sh: prevent
warnings when using iounmap").

Probably sh-nommu should include <asm-generic/io.h>, but that would
require a lot more changes.
---
 arch/sh/include/asm/io.h | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/arch/sh/include/asm/io.h b/arch/sh/include/asm/io.h
index cf9a3ec32406f856..fba90e670ed41d48 100644
--- a/arch/sh/include/asm/io.h
+++ b/arch/sh/include/asm/io.h
@@ -271,8 +271,12 @@ static inline void __iomem *ioremap_prot(phys_addr_t offset, unsigned long size,
 #endif /* CONFIG_HAVE_IOREMAP_PROT */
 
 #else /* CONFIG_MMU */
-#define iounmap(addr)		do { } while (0)
-#define ioremap(offset, size)	((void __iomem *)(unsigned long)(offset))
+static inline void __iomem *ioremap(phys_addr_t offset, size_t size)
+{
+	return (void __iomem *)(unsigned long)offset;
+}
+
+static inline void iounmap(volatile void __iomem *addr) { }
 #endif /* CONFIG_MMU */
 
 #define ioremap_uc	ioremap
-- 
2.25.1


^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [PATCH v2] sh: Convert nommu io{re,un}map() to static inline functions
  2022-06-20  7:01 [PATCH v2] sh: Convert nommu io{re,un}map() to static inline functions Geert Uytterhoeven
@ 2022-06-20  7:06 ` Christoph Hellwig
  0 siblings, 0 replies; 2+ messages in thread
From: Christoph Hellwig @ 2022-06-20  7:06 UTC (permalink / raw)
  To: Geert Uytterhoeven
  Cc: Yoshinori Sato, Rich Felker, Linus Torvalds, Andrew Morton,
	Jonathan Cameron, linux-sh, linux-kernel, kernel test robot,
	Christoph Hellwig

On Mon, Jun 20, 2022 at 09:01:43AM +0200, Geert Uytterhoeven wrote:
> Recently, nommu iounmap() was converted from a static inline function to
> a macro again, basically reverting commit 4580ba4ad2e6b8dd ("sh: Convert
> iounmap() macros to inline functions").  With -Werror, this leads to
> build failures like:

Stupid question:  Is there any reason sh-nommu can't just use the
generic nommu ioremap from asm-generic/io.h?

Then again how long it takes to to fix any sh bugs I'm still wondering
if we shouldn't just drop sh entirely given how much of a drag it is
on everyone.

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2022-06-20  7:06 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-06-20  7:01 [PATCH v2] sh: Convert nommu io{re,un}map() to static inline functions Geert Uytterhoeven
2022-06-20  7:06 ` Christoph Hellwig

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).