On Mon, 21 Jul 2003, Christoph Hellwig wrote: > On Mon, Jul 21, 2003 at 01:00:54PM +0200, Javier Achirica wrote: > > > > Daniel, > > > > Thank you for your patch. Some comments about it: > > > > - I'd rather fix whatever is broken in the current code than going back to > > spinlocks, as they increase latency and reduce concurrency. In any case, > > please check your code. > > In general we prefer spinlocks in linux for drivers unless there's a > very good reason against it. If you have latency or concurrency problems > it seems you have problems with your algorithms or the length of your > critical sections. I don't think it is a problem with the algorithms. It is, of course, with the length of the critical sections, due to the fact that the Aironet firmware needs serialized commands and some of them take a long time to execute. To ensure serialization, a spinlock was used, but in that case it could be hold for hundreds of milliseconds in some cases (waiting for the command to finish). If you have any suggestion about how to better deal with that issue, I'd be happy to hear it. Javier Achirica - : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
