[openib-general] [PATCHv3][RFC] kDAPL: use cm timers instead of own

[James Lentini]

On Wed, 1 Jun 2005, Talpey, Thomas wrote:

> At 03:48 PM 6/1/2005, Tom Duffy wrote:
>> This patch takes into account some of the feedback by Sean, it sets the
>> retry to 0, so it will do the timeout right.  It also stops the
>> conversion of ms to ib at 63 since 64 overflows.  I have tested this
>> version using kdapltest in quit, performance, and transaction modes.
>>
>> James, do you think it good enough to apply?
>
> Do any of these tests intentionally trigger timer events? I don't think
> they do...

We don't have any tests that do that.

>
> BTW :-) surely you can think of an even more colorful name than this!
>
>> static inline u8 dapl_convert_ms_to_kookyib(unsigned long ms)
>
> Shouldn't this routine live in an IB-specific area, not core DAPL?

It is in IB-specific code.