| Message ID | 20180316154852.13206-2-logang@deltatee.com (mailing list archive) |
|---|---|
| State | Not Applicable |
| Delegated to: | Herbert Xu |
| Headers | show |
On Fri, Mar 16, 2018 at 09:48:44AM -0600, Logan Gunthorpe wrote: > Warnings of the form: > > lib/iomap.c:84:9: warning: cast to restricted __be16 > > Are found when running sparse checker with: > > make C=2 CF=-D__CHECK_ENDIAN__ > > This patch casts them with __force to suppress the warnings. Hi, it looks correct but if you should explain why the warning is not a valid one. Once done, you can then explain what problem you have fixed instead of saying you suppressed a warning. > Also, it looks like the wrong conversion function was used in I would suggest to split the patch in two since it does two different things which should have their own explanation. > diff --git a/lib/iomap.c b/lib/iomap.c > index 541d926da95e..a05d9fa21794 100644 > --- a/lib/iomap.c > +++ b/lib/iomap.c > @@ -106,8 +106,10 @@ EXPORT_SYMBOL(ioread32be); > #endif > > #ifndef mmio_write16be > -#define mmio_write16be(val,port) __raw_writew(be16_to_cpu(val),port) > -#define mmio_write32be(val,port) __raw_writel(be32_to_cpu(val),port) > +#define mmio_write16be(val, port) \ > + __raw_writew((u16 __force)cpu_to_be16(val), port) > +#define mmio_write32be(val, port) \ > + __raw_writel((u32 __force)cpu_to_be32(val), port) > #endif I think that in this sort of replacement, it's better to not split the lines (even if larger than 80, that's it). -- Luc Van Oostenryck
Hi Luc, Thanks for the feedback. I'll try to address your points in a v13 shortly. Logan On 18/03/18 10:13 AM, Luc Van Oostenryck wrote: > it looks correct but if you should explain why the warning > is not a valid one. Once done, you can then explain what problem > you have fixed instead of saying you suppressed a warning. > >> Also, it looks like the wrong conversion function was used in > > I would suggest to split the patch in two since it does two > different things which should have their own explanation. > >> diff --git a/lib/iomap.c b/lib/iomap.c >> index 541d926da95e..a05d9fa21794 100644 >> --- a/lib/iomap.c >> +++ b/lib/iomap.c >> @@ -106,8 +106,10 @@ EXPORT_SYMBOL(ioread32be); >> #endif >> >> #ifndef mmio_write16be >> -#define mmio_write16be(val,port) __raw_writew(be16_to_cpu(val),port) >> -#define mmio_write32be(val,port) __raw_writel(be32_to_cpu(val),port) >> +#define mmio_write16be(val, port) \ >> + __raw_writew((u16 __force)cpu_to_be16(val), port) >> +#define mmio_write32be(val, port) \ >> + __raw_writel((u32 __force)cpu_to_be32(val), port) >> #endif > > I think that in this sort of replacement, it's better to not split > the lines (even if larger than 80, that's it). > > -- Luc Van Oostenryck >
diff --git a/lib/iomap.c b/lib/iomap.c index 541d926da95e..a05d9fa21794 100644 --- a/lib/iomap.c +++ b/lib/iomap.c @@ -65,8 +65,8 @@ static void bad_io_access(unsigned long port, const char *access) #endif #ifndef mmio_read16be -#define mmio_read16be(addr) be16_to_cpu(__raw_readw(addr)) -#define mmio_read32be(addr) be32_to_cpu(__raw_readl(addr)) +#define mmio_read16be(addr) be16_to_cpu((__be16 __force)__raw_readw(addr)) +#define mmio_read32be(addr) be32_to_cpu((__be32 __force)__raw_readl(addr)) #endif unsigned int ioread8(void __iomem *addr) @@ -106,8 +106,10 @@ EXPORT_SYMBOL(ioread32be); #endif #ifndef mmio_write16be -#define mmio_write16be(val,port) __raw_writew(be16_to_cpu(val),port) -#define mmio_write32be(val,port) __raw_writel(be32_to_cpu(val),port) +#define mmio_write16be(val, port) \ + __raw_writew((u16 __force)cpu_to_be16(val), port) +#define mmio_write32be(val, port) \ + __raw_writel((u32 __force)cpu_to_be32(val), port) #endif void iowrite8(u8 val, void __iomem *addr)
Warnings of the form: lib/iomap.c:84:9: warning: cast to restricted __be16 Are found when running sparse checker with: make C=2 CF=-D__CHECK_ENDIAN__ This patch casts them with __force to suppress the warnings. Also, it looks like the wrong conversion function was used in the mmio_writeXXbe macros: 'val' is a cpu value to be written to a big endian register. Therefore cpu_to_beXX should have been used instead of beXX_to_cpu. Both are equivalent, functionally, but have different meanings. So there was no real bug just confusion for people reading the code and the sparse endian check. Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Philippe Ombredanne <pombredanne@nexb.com> Cc: Kate Stewart <kstewart@linuxfoundation.org> --- lib/iomap.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-)