On Thu, Aug 16, 2012 at 4:49 PM, Stefan Fritsch <[email protected]> wrote: > Here is the next iteration: > > Changes from V5: > - Remove unused code for mutexes > - Remove explicit hardware memory barriers. If the host uses non-temporal > memory operations like movnti, it will hopefully issue the memory > barriers, too. The cheaper compiler memory barriers are still provided > by bus_dmamap_sync(). > - block: Fix error handling broken in v5 (the crash reported by johnw) > - net: set watermark for MCLGETI > - net: define constants for the queue numbers > - net: fix handling of tx queue full situations (this hopefully fixes > the scp stall observed by johnw) > > Thanks for all comments and testing. >
could you please tell me if you're using tx interrupt or not? if yes, why do you need to have a watchdog code implement a txeof path and not the actual reset? also you forget to undo the preps in vioif_populate_rx_mbufs by calling a virtio_enqueue_abort if i understand this spaghetti code correctly.
