From: Ben Hutchings Date: Tue, 24 Sep 2013 22:21:57 +0000 (+0100) Subject: sfc: Tighten the check for RX merged completion events X-Git-Url: https://git.karo-electronics.de/?a=commitdiff_plain;h=92a04168390a2cb67cb60891e5870626d58c2abb;p=linux-beck.git sfc: Tighten the check for RX merged completion events The addition of RX event merging support means we don't reliably detect dropped RX events now. Currently we will only detect them if the previous event for the RX queue had the CONT bit set. Only accept RX completion events as merged if the GET_CAPABILITIES_OUT_RX_BATCHING bit is set in datapath_caps (which it won't be for the low-latency datapath) and the CONT bit is not set on the event. Signed-off-by: Ben Hutchings --- diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index 5d46d155b642..2f77359607d2 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -1766,6 +1766,8 @@ static int efx_ef10_handle_rx_event(struct efx_channel *channel, ((1 << ESF_DZ_RX_DSC_PTR_LBITS_WIDTH) - 1)); if (n_descs != rx_queue->scatter_n + 1) { + struct efx_ef10_nic_data *nic_data = efx->nic_data; + /* detect rx abort */ if (unlikely(n_descs == rx_queue->scatter_n)) { WARN_ON(rx_bytes != 0); @@ -1773,10 +1775,13 @@ static int efx_ef10_handle_rx_event(struct efx_channel *channel, return 0; } - if (unlikely(rx_queue->scatter_n != 0)) { - /* Scattered packet completions cannot be - * merged, so something has gone wrong. - */ + /* Check that RX completion merging is valid, i.e. + * the current firmware supports it and this is a + * non-scattered packet. + */ + if (!(nic_data->datapath_caps & + (1 << MC_CMD_GET_CAPABILITIES_OUT_RX_BATCHING_LBN)) || + rx_queue->scatter_n != 0 || rx_cont) { efx_ef10_handle_rx_bad_lbits( rx_queue, next_ptr_lbits, (rx_queue->removed_count +