On Thu, 2008-03-13 at 12:04 +0200, Or Gerlitz wrote: > Roland, > > Just curious why you think extra pointer indirection for each send is > worse than TLB misses taken by using vmalloc(), are the latter being > excepted very un often? Also assuming keeping the dma addresses at the > TX descriptor is a practice taken also by Ethernet drivers who do have > big send queues, maybe we should look there to see how this problem was > solved? > > Or. > I think that a TLB miss will happen only once for each process while a cache miss on the pointer might happen more often. The question is whether it is significant or not. I think the biggest drawback of vmalloc is the scarce space in 32 bit systems - other than that it is preferred over kmalloc for each entry.
_______________________________________________ general mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
