Re: [patches] [PATCH] irqchip/riscv-plic: fix address alignment of the plic enable bits

[Palmer Dabbelt]

On Mon, 18 Dec 2017 23:44:38 PST (-0800), zongbox@xxxxxxxxx wrote:
> When the interrupt sourece id > 31, the register address is not 4 byte
> alignment. Arithmetic on void pointer has no size extension, it just
> add the result of 'hwirq/32' directly.
>
> Signed-off-by: Zong Li <zongbox@xxxxxxxxx>
> ---
>  drivers/irqchip/irq-riscv-plic.c | 15 +++++++++------
>  1 file changed, 9 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/irqchip/irq-riscv-plic.c b/drivers/irqchip/irq-riscv-plic.c
> index b5182b4..0a43763 100644
> --- a/drivers/irqchip/irq-riscv-plic.c
> +++ b/drivers/irqchip/irq-riscv-plic.c
> @@ -92,6 +92,8 @@
>   */
>  #define ENABLE_BASE		0x2000
>  #define ENABLE_PER_HART		0x80
> +#define ENABLE_SOURCES		32
> +#define ENABLE_SOURCES_OFFSET	4
>
>  /*
>   * Each hart context has a set of control registers associated with it.  Right
> @@ -128,9 +130,10 @@ struct plic_data {
>
>  /* Addressing helper functions. */
>  static inline
> -void __iomem *plic_enable_vector(struct plic_data *data, int contextid)
> +void __iomem *plic_enable_vector(struct plic_data *data, int contextid, int hwirq)
>  {
> -	return data->reg + ENABLE_BASE + contextid * ENABLE_PER_HART;
> +	return data->reg + ENABLE_BASE + contextid * ENABLE_PER_HART +
> +		(hwirq / ENABLE_SOURCES) * ENABLE_SOURCES_OFFSET;
>  }
>
>  static inline
> @@ -172,8 +175,8 @@ void plic_complete(struct plic_data *data, int contextid, u32 claim)
>  /* Explicit interrupt masking. */
>  static void plic_disable(struct plic_data *data, int contextid, int hwirq)
>  {
> -	void __iomem *reg = plic_enable_vector(data, contextid) + (hwirq / 32);
> -	u32 mask = ~(1 << (hwirq % 32));
> +	void __iomem *reg = plic_enable_vector(data, contextid, hwirq);
> +	u32 mask = ~(1 << (hwirq % ENABLE_SOURCES));
>
>  	spin_lock(&data->lock);
>  	writel(readl(reg) & mask, reg);
> @@ -182,8 +185,8 @@ static void plic_disable(struct plic_data *data, int contextid, int hwirq)
>
>  static void plic_enable(struct plic_data *data, int contextid, int hwirq)
>  {
> -	void __iomem *reg = plic_enable_vector(data, contextid) + (hwirq / 32);
> -	u32 bit = 1 << (hwirq % 32);
> +	void __iomem *reg = plic_enable_vector(data, contextid, hwirq);
> +	u32 bit = 1 << (hwirq % ENABLE_SOURCES);
>
>  	spin_lock(&data->lock);
>  	writel(readl(reg) | bit, reg);

Whoops, sorry about that.  I think it's cleaner to use a 'u32 *' instead of 
'void *' here, something more like this (which I haven't even compiled)

commit 2424a65ce079ecbe16fbedc704b88b6f04f8b97a
Author: Palmer Dabbelt <palmer@xxxxxxxxxx>
Date:   Wed Dec 20 19:31:41 2017 -0800

   void* -> u32* in PLIC

diff --git a/drivers/irqchip/irq-riscv-plic.c b/drivers/irqchip/irq-riscv-plic.c
index b5182b43b377..99e79713abc6 100644
--- a/drivers/irqchip/irq-riscv-plic.c
+++ b/drivers/irqchip/irq-riscv-plic.c
@@ -128,25 +128,25 @@ struct plic_data {

/* Addressing helper functions. */
static inline
-void __iomem *plic_enable_vector(struct plic_data *data, int contextid)
+u32 __iomem *plic_enable_vector(struct plic_data *data, int contextid)
{
	return data->reg + ENABLE_BASE + contextid * ENABLE_PER_HART;
}

static inline
-void __iomem *plic_priority(struct plic_data *data, int hwirq)
+u32 __iomem *plic_priority(struct plic_data *data, int hwirq)
{
	return data->reg + PRIORITY_BASE + hwirq * PRIORITY_PER_ID;
}

static inline
-void __iomem *plic_hart_threshold(struct plic_data *data, int contextid)
+u32 __iomem *plic_hart_threshold(struct plic_data *data, int contextid)
{
	return data->reg + CONTEXT_BASE + CONTEXT_PER_HART * contextid + CONTEXT_THRESHOLD;
}

static inline
-void __iomem *plic_hart_claim(struct plic_data *data, int contextid)
+u32 __iomem *plic_hart_claim(struct plic_data *data, int contextid)
{
	return data->reg + CONTEXT_BASE + CONTEXT_PER_HART * contextid + CONTEXT_CLAIM;
}
@@ -172,7 +172,7 @@ void plic_complete(struct plic_data *data, int contextid, u32 claim)
/* Explicit interrupt masking. */
static void plic_disable(struct plic_data *data, int contextid, int hwirq)
{
-	void __iomem *reg = plic_enable_vector(data, contextid) + (hwirq / 32);
+	u32 __iomem *reg = plic_enable_vector(data, contextid) + (hwirq / 32);
	u32 mask = ~(1 << (hwirq % 32));

	spin_lock(&data->lock);
@@ -182,7 +182,7 @@ static void plic_disable(struct plic_data *data, int contextid, int hwirq)

static void plic_enable(struct plic_data *data, int contextid, int hwirq)
{
-	void __iomem *reg = plic_enable_vector(data, contextid) + (hwirq / 32);
+	u32 __iomem *reg = plic_enable_vector(data, contextid) + (hwirq / 32);
	u32 bit = 1 << (hwirq % 32);

	spin_lock(&data->lock);

Note that the PLIC driver isn't upstream yet, and hasn't been reviewed because 
(as you've discovered) it's a bit of a mess.  Thanks for the patch, I'll be 
sure to fix it up before sending out the PLIC driver for review.