Re: [PATCH 01/11] irqchip: Allow irq_reg_{readl,writel} to use __raw_{readl_writel}

[Thomas Gleixner]

On Wed, 29 Oct 2014, Kevin Cernekee wrote:
> Thomas:
> > How does that work with multi arch kernels?
> 
> I am assuming this question refers to e.g. CONFIG_ARCH_MULTIPLATFORM
> 
> If GENERIC_IRQ_CHIP is being used, the current implementation of
> generic-chip.c will have to pick one global definition of
> irq_reg_{readl,writel} for all supported SoCs.
> 
> One possibility is that individual irqchip drivers requiring special
> accessors can pass in their own function pointers, similar to this:
> 
> struct sdhci_ops {
> #ifdef CONFIG_MMC_SDHCI_IO_ACCESSORS
> u32 (*read_l)(struct sdhci_host *host, int reg);
> u16 (*read_w)(struct sdhci_host *host, int reg);
> u8 (*read_b)(struct sdhci_host *host, int reg);
> void (*write_l)(struct sdhci_host *host, u32 val, int reg);
> void (*write_w)(struct sdhci_host *host, u16 val, int reg);
> void (*write_b)(struct sdhci_host *host, u8 val, int reg);
> #endif
> 
> and fall back to readl/writel if none are supplied.  It would be nice
> if this provided common definitions for the __raw_ and maybe the BE
> variants too.
> 
> Or, we could add IRQ_GC_NATIVE_IO and/or IRQ_GC_BE_IO to enum irq_gc_flags.

I definitely prefer to have these options in the generic chip
implementation so we avoid that driver writers duplicate code in
creative and wrong ways.

Thanks,

	tglx
--
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/