From: Didier Pallard <[email protected]> Sent: Tuesday, March 17, 2020 11:19 To: Slava Ovsiienko <[email protected]> Cc: [email protected]; [email protected]; Matan Azrad <[email protected]> Subject: Re: [dpdk-dev] [PATCH] net/mlx5: fix Rx descriptor status returned value >>>well, please do if you don't mind, OK, I will.
>>>You will validate it quicker, and I'm currently working on a different topic. >>>Btw, do you think it's possible to have an implementation of >>>[r,t]x_status_descriptor functions for vectorized implementation? >>>thanks Yes, we are planning this update. With best regards, Slava didier On Tue, Mar 17, 2020 at 9:33 AM Slava Ovsiienko <[email protected]<mailto:[email protected]>> wrote: >> From: Didier Pallard >> <[email protected]<mailto:[email protected]>> >> Sent: Monday, March 16, 2020 19:24 >> To: Slava Ovsiienko >> <[email protected]<mailto:[email protected]>> >> Cc: [email protected]<mailto:[email protected]>; >> [email protected]<mailto:[email protected]>; Matan Azrad >> <[email protected]<mailto:[email protected]>> >> Subject: Re: [dpdk-dev] [PATCH] net/mlx5: fix Rx descriptor status returned >> value >> Well, you're right, another way to fix the problem could be to set up the >> queue size assuming the provided number >> is a number of packets in queue rather than a number of mbufs in queue. Yes, it is intended in queue setup routine. But, for mlx5 we have a bug for regular mlx5_rx_burst if scattering is enabled, the Rx queue is created with size wqe_n elements, should be wqe_n << sges_n instead, to be able to receive the requested number of packets (wqe_n). I think we must fix. Would you like to update your patch, or should I provide mine? >> But not sure it's better, it's also important for the application/user to >> know the number of mbufs that could fit in a rx/tx queue, >> whatever the number of packets that it covers, since it is very important to >> size the memory pools correctly to avoid any >> mbuf shortage during system life. >> Thanks >> Didier To estimate - the number of "DPDK descriptors" should be multiplied by the maximal length of scattered packet chain. With best regards, Slava > -----Original Message----- > From: dev <mailto:[email protected]<mailto:[email protected]>> On > Behalf Of Didier Pallard > Sent: Friday, March 13, 2020 11:57 > To: mailto:[email protected]<mailto:[email protected]> > Cc: mailto:[email protected]<mailto:[email protected]> > Subject: [dpdk-dev] [PATCH] net/mlx5: fix Rx descriptor status returned value > > Two bugs in rx_queue_count function: > - One entry may contain several segments, so 'used' must be multiplied > by number of segments per entry to properly reflect the queue usage. > - rx_queue_count returns the number of entries used in queue, so it ranges > from 0 to max number of entries in queue, not this number minus > one. > > Fixes: 8788fec1f269 ("net/mlx5: implement descriptor status API") > Cc: mailto:[email protected]<mailto:[email protected]> > Signed-off-by: Didier Pallard > <mailto:[email protected]<mailto:[email protected]>> > --- > drivers/net/mlx5/mlx5_rxtx.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/net/mlx5/mlx5_rxtx.c b/drivers/net/mlx5/mlx5_rxtx.c index > 5ac63da8039d..17f80c25443e 100644 > --- a/drivers/net/mlx5/mlx5_rxtx.c > +++ b/drivers/net/mlx5/mlx5_rxtx.c > @@ -500,7 +500,7 @@ rx_queue_count(struct mlx5_rxq_data *rxq) > used += n; > cqe = &(*rxq->cqes)[cq_ci & cqe_cnt]; > } > - used = RTE_MIN(used, (1U << rxq->elts_n) - 1); > + used = RTE_MIN(used * (1 << rxq->sges_n), 1U << rxq->elts_n); > return used; > } > > -- > 2.24.1
