On Wed, 2014-02-05 at 20:44 -0800, Joe Perches wrote:
> Perhaps use a more common kernel style
>
> struct ieee80211_reason_descriptions {
> u16 code;
> const char * desc;
> }
>
> and enumerate the reason codes with #defines and use a
> macro to populate the descriptions
>
> #define IEEE80211_REASON_RESERVED 0
> #define IEEE80211_REASON_UNSPECIFIED 1
>
> etc.
>
> #define POPULATE_IEEE_REASON(code) \
> {.code = IEEE80211_REASON_##code, .desc = #code}
>
> static const struct ieee80211_reason_descriptions reasons[] = {
> POPULATE_IEEE_REASON(RESERVED),
> POPULATE_IEEE_REASON(UNSPECIFIED),
> [etc...]
> };
>
> So this function becomes something like:
>
> const char *ieee80211_get_reason_code_string(u16 reason_code)
> {
> int i;
>
> for (i = 0; i < ARRAY_SIZE(reasons); i++) {
> if (reasons[i].code == reason_code)
> return reasons[i].desc;
> }
Isn't it more efficient to just let the compiler generate it with a big
switch() statement? AFAICT gcc will typically generate code like
Calvin's original hand-rolled code and/or big lookup tables anyway, if
faced with something like
switch (reason) {
case 17: return "asdf";
... // all the others
default: return "<unknown>";
};
In any case, I agree with you and Jouni about leaving the number -
that's certainly needed - but I'm willing to merge a clean patch like
this too.
johannes
--
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
