net: mvpp2: release reference to txq_cpu[] entry after unmapping
authorThomas Petazzoni <thomas.petazzoni@free-electrons.com>
Tue, 21 Feb 2017 10:28:05 +0000 (11:28 +0100)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Sat, 21 Oct 2017 15:09:05 +0000 (17:09 +0200)
[ Upstream commit 36fb7435b6ac4d288a2d4deea8934f9456ab46b6 ]

The mvpp2_txq_bufs_free() function is called upon TX completion to DMA
unmap TX buffers, and free the corresponding SKBs. It gets the
references to the SKB to free and the DMA buffer to unmap from a per-CPU
txq_pcpu data structure.

However, the code currently increments the pointer to the next entry
before doing the DMA unmap and freeing the SKB. It does not cause any
visible problem because for a given SKB the TX completion is guaranteed
to take place on the CPU where the TX was started. However, it is much
more logical to increment the pointer to the next entry once the current
entry has been completely unmapped/released.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Acked-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <alexander.levin@verizon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/net/ethernet/marvell/mvpp2.c

index 0e67145bc418426fbeb9f3cfc2503e6be8597477..4f34e1b7970547ec5cd4e270c1be1d66ef70073a 100644 (file)
@@ -4415,13 +4415,12 @@ static void mvpp2_txq_bufs_free(struct mvpp2_port *port,
                struct mvpp2_txq_pcpu_buf *tx_buf =
                        txq_pcpu->buffs + txq_pcpu->txq_get_index;
 
-               mvpp2_txq_inc_get(txq_pcpu);
-
                dma_unmap_single(port->dev->dev.parent, tx_buf->phys,
                                 tx_buf->size, DMA_TO_DEVICE);
-               if (!tx_buf->skb)
-                       continue;
-               dev_kfree_skb_any(tx_buf->skb);
+               if (tx_buf->skb)
+                       dev_kfree_skb_any(tx_buf->skb);
+
+               mvpp2_txq_inc_get(txq_pcpu);
        }
 }