On 2015-11-23 19:25, Peter Oh wrote: > > On 11/23/2015 10:02 AM, Sebastian Gottschall wrote: >> Am 23.11.2015 um 18:25 schrieb Peter Oh: >>> Hi, >>> >>> Have you measured the peak throughput? >>> The pre-allocated coherent memory concept was introduced as once of >>> peak throughput improvement. >>> IIRC, dma_map_single takes about 4 us on Cortex A7 and >>> dma_unmap_single also takes time to invalid cache. >>> Please share your tput number before and after, so I don't need to >>> worry about performance degrade. >> yes. and this concept fucks up the qualcom ipq806x platform (which has >> by default 2 QCA99XX cards). it does not work, since the preallocated >> concept allocates too much memory > Can you specify the number that you're seeing? > The tx buffer descriptor uses 56KB, so it would require 64KB x 2 slab > for 2 cards which is 128KB. > Are you saying your system does not work because of this 128KB usage? The reason might be the GFP_DMA flag, which is unnecessary.
- Felix _______________________________________________ ath10k mailing list ath10k@lists.infradead.org http://lists.infradead.org/mailman/listinfo/ath10k