net/mlx5e: Fix napi poll with zero budget

napi->poll can be called with budget 0, e.g. in netpoll scenarios
where the caller only wants to poll TX rings
(poll_one_napi@net/core/netpoll.c).

The below commit changed RX polling from "while" loop to "do {} while",
which caused to ignore the initial budget and handle at least one RX
packet.

This fixes the following warning:
[ 2852.049194] mlx5e_napi_poll+0x0/0x260 [mlx5_core] exceeded budget in poll
[ 2852.049195] ------------[ cut here ]------------
[ 2852.049195] WARNING: CPU: 0 PID: 25691 at net/core/netpoll.c:171 netpoll_poll_dev+0x18a/0x1a0

Fixes: 4b7dfc9 ("net/mlx5e: Early-return on empty completion queues")
Signed-off-by: Saeed Mahameed <[email protected]>
Reviewed-by: Tariq Toukan <[email protected]>
Reported-by: Martin KaFai Lau <[email protected]>
Tested-by: Martin KaFai Lau <[email protected]>
Cc: [email protected]
Signed-off-by: Saeed Mahameed <[email protected]>

Author: Saeed Mahameed

drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c

@@ -49,7 +49,7 @@ int mlx5e_napi_poll(struct napi_struct *napi, int budget)
 	struct mlx5e_channel *c = container_of(napi, struct mlx5e_channel,
 					       napi);
 	bool busy = false;
-	int work_done;
+	int work_done = 0;
 	int i;
 
 	for (i = 0; i < c->num_tc; i++)
@@ -58,15 +58,17 @@ int mlx5e_napi_poll(struct napi_struct *napi, int budget)
 	if (c->xdp)
 		busy |= mlx5e_poll_xdpsq_cq(&c->rq.xdpsq.cq);
 
-	work_done = mlx5e_poll_rx_cq(&c->rq.cq, budget);
-	busy |= work_done == budget;
+	if (likely(budget)) { /* budget=0 means: don't poll rx rings */
+		work_done = mlx5e_poll_rx_cq(&c->rq.cq, budget);
+		busy |= work_done == budget;
+	}
 
 	busy |= c->rq.post_wqes(&c->rq);
 
 	if (busy) {
 		if (likely(mlx5e_channel_no_affinity_change(c)))
 			return budget;
-		if (work_done == budget)
+		if (budget && work_done == budget)
 			work_done--;
 	}