Re: Being more anal about iospace accesses..

[linux]

> Nice.
>
> But it still leaves me confused.  Before I had this code:
>
>	struct regs {
>		uint32_t status;
>		...
>	}
>
>	...
>
>	struct regs *regs = ioremap(...);
>	uint32_t status = regs->status;
>	...
> 
> So now I should do it like this:
>
>	#define REG_STATUS 0
>
>	...
>
>	void __iomem *regs = ioremap(...);
>	uint32_t status = readl(regs + REG_STATUS);
>	...
>
> But wait, that only works when long is 32bit wide.  Plus I could be
> stupid enough and "#define REG_STATUS 64" while the register space is
> just 64 bytes long.  It solves the confusion about address spaces,
> agreed, but overall I'm more confused now.  Hope it's just temporary.

No, you should do:

	struct regs {
	      uint32_t status;
	      ...
	}

	...

	struct regs __iomem *regs = ioremap(...);
	uint32_t status = ioread32(&regs->status);
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/