On 13/11/2015 16:26, Thomas Gleixner wrote:
> On Fri, 13 Nov 2015, Marc Gonzalez wrote:
>> On 13/11/2015 15:16, Daniel Lezcano wrote:
>>> For example:
>>>
>>> struct clockcommon {
>>> u32 mult;
>>> u32 shift;
>>> int rating;
>>> void __iomem *base;
>>> char *name;
>>> int irq;
>>> };
>>>
>>> struct clocksource {
>>> struct clockcommon common; /* MUST be the first field */
>>> cycle_t (*read)(struct clocksource *cs);
>>> cycle_t mask;
>>> ...
>>> };
>>
>> According to my notes, commit 369db4c952 grouped hot-path data
>> into a single cache line (hence ____cacheline_aligned).
>>
>> (AFAIR, ARMv7 ARCH_MULTIPLATFORM assumes CACHE_LINE=64)
>>
>> Not sure how to make the two concepts (common base struct and
>> grouping hot data) play nicely, without wasting a lot of space
>> on padding.
>
> It wont play well. We are not going to change the layout of struct
> clocksource because it will hurt the sane use cases for no reason.
You are referring to Daniel's proposed reorg, right?
What about my proposal? I inserted reg after maxadj, but it may be
inserted anywhere in the struct.
diff --git a/include/linux/clocksource.h b/include/linux/clocksource.h
index 278dd279a7a8..03807ca0d54e 100644
--- a/include/linux/clocksource.h
+++ b/include/linux/clocksource.h
@@ -74,6 +74,9 @@ struct clocksource {
u32 shift;
u64 max_idle_ns;
u32 maxadj;
+#ifdef CONFIG_CLKSRC_MMIO
+ void __iomem *reg;
+#endif
#ifdef CONFIG_ARCH_CLOCKSOURCE_DATA
struct arch_clocksource_data archdata;
#endif
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
