On Tue, Mar 20, 2012 at 5:16 AM, Johannes Berg
<johannes@xxxxxxxxxxxxxxxx> wrote:
> In other words:
>
>> #if (LINUX_VERSION_CODE > KERNEL_VERSION(2,6,27))
>
> I'd change this to 2.6.33 so
I think I tried this approach as well.
>> #include_next <linux/tracepoint.h>
>
> I don't have to think about this:
>
>> +/* Backports f42c85e7 */
>> +#if (LINUX_VERSION_CODE < KERNEL_VERSION(2,6,30))
>> +#undef TP_STRUCT__entry
>> +#define TP_STRUCT__entry(args...) args
>> +#endif
>> +
>> +#if (LINUX_VERSION_CODE < KERNEL_VERSION(2,6,33))
>> +/* Backports 091ad365, a rename */
>> +#define DECLARE_EVENT_CLASS TRACE_EVENT_TEMPLATE
>> +
>> +#if (LINUX_VERSION_CODE < KERNEL_VERSION(2,6,30))
>> +#define TP_PROTO TPPROTO
>> +#define TP_ARGS TPARGS
>> +#define TP_FMT TPFMT
>> +#define TP_RAW_FMT TPRAWFMT
>> +#endif
>> +
>> +#endif
>
>
> This seems worthwhile and looks good:
>
>> +#else /* just disable tracing */
>> +
>> +/* Disable all tracing */
>> +#undef TRACE_EVENT
>> +#define TRACE_EVENT(name, proto, ...) \
>> +static inline void trace_ ## name(proto) {}
>> +#undef DECLARE_EVENT_CLASS
>> +#define DECLARE_EVENT_CLASS(...)
>> +#undef DEFINE_EVENT
>> +#define DEFINE_EVENT(evt_class, name, proto, ...) \
>> +static inline void trace_ ## name(proto) {}
>> +
>> +#define TP_PROTO(args...) args
>> +#define TP_ARGS(args...) args
>> +#define TP_CONDITION(args...) args
>
> That doesn't seem to be needed since these are used inside the other
> macros only?
>
>> +struct tracepoint_iter {
>> +};
>
> Not sure why that would be needed?
Sure, yeah I think I tried disabling as you suggested and ran into
compilation issues with mac80211 on 2.6.31.
Luis
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
