bnxt_en: Fix IRQ coalescing regression.

Recent IRQ coalescing clean up has removed a guard-rail for the max DMA
buffer coalescing value.  This is a 6-bit value and must not be 0.  We
already have a check for 0 but 64 is equivalent to 0 and will cause
non-stop interrupts.  Fix it by adding the proper check.

Fixes: f8503969d27b ("bnxt_en: Refactor and simplify coalescing code.")
Reported-by: Andy Gospodarek <gospo@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index c3dfaa5..4e3d569 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -4548,9 +4548,13 @@ static void bnxt_hwrm_set_coal_params(struct bnxt_coal *hw_coal,
 
 	val = clamp_t(u16, hw_coal->coal_bufs, 1, max);
 	req->num_cmpl_aggr_int = cpu_to_le16(val);
+
+	/* This is a 6-bit value and must not be 0, or we'll get non stop IRQ */
+	val = min_t(u16, val, 63);
 	req->num_cmpl_dma_aggr = cpu_to_le16(val);
 
-	val = clamp_t(u16, hw_coal->coal_bufs_irq, 1, max);
+	/* This is a 6-bit value and must not be 0, or we'll get non stop IRQ */
+	val = clamp_t(u16, hw_coal->coal_bufs_irq, 1, 63);
 	req->num_cmpl_dma_aggr_during_int = cpu_to_le16(val);
 
 	tmr = BNXT_USEC_TO_COAL_TIMER(hw_coal->coal_ticks);