On Fri, May 15, 2020 at 4:25 PM Andy Shevchenko
<andy.shevche...@gmail.com> wrote:
>
> On Fri, May 15, 2020 at 4:48 PM kbuild test robot <l...@intel.com> wrote:
>
> > sparse warnings: (new ones prefixed by >>)
> >
> > >> drivers/mtd/nand/raw/intel-nand-controller.c:441:24: sparse: sparse:
> > >> incorrect type in argument 1 (different base types) @@ expected
> > >> unsigned int val @@ got restricted __be32 [assignedunsigned int val @@
> > >> drivers/mtd/nand/raw/intel-nand-controller.c:441:24: sparse: expected
> > >> unsigned int val
> > >> drivers/mtd/nand/raw/intel-nand-controller.c:441:24: sparse: got
> > >> restricted __be32 [assigned] [usertype] reg
> > drivers/mtd/nand/raw/intel-nand-controller.c:444:24: sparse: sparse:
> > incorrect type in argument 1 (different base types) @@ expected unsigned
> > int val @@ got restricted __be32 [assignedunsigned int val @@
> > drivers/mtd/nand/raw/intel-nand-controller.c:444:24: sparse: expected
> > unsigned int val
> > drivers/mtd/nand/raw/intel-nand-controller.c:444:24: sparse: got
> > restricted __be32 [assigned] [usertype] reg
> >
> > 440 reg = cpu_to_be32(*pdata++);
> > > 441 writel(reg, ebu_host->hsnand + HSNAND_CMSG_0);
> > 442
> > 443 reg = cpu_to_be32(*pdata);
> > 444 writel(reg, ebu_host->hsnand + HSNAND_CMSG_1);
>
> On BE:
> cpu_to_be32 -> no-op
> writel() -> converts reg to LE
>
> On LE:
> cpu_to_be32 -> converts to BE
> writel() -> no-op (in terms of conversion)
>
> Seems to me that the proper API (if above is intended) should be swab32().
> But something tells me that above is broken (or my understanding is wrong).
iowrite_be32() is the correct way to store word into a big-endian mmio register,
if that is the intention here.
Arnd
