From: Rob Herring <[email protected]>
If skb allocation for the rx ring fails repeatedly, we can reach a point
were the ring is empty. In this condition, the driver is out of sync with
the h/w. While this has always been possible, the removal of the skb
recycling seems to have made triggering this problem easier.
Signed-off-by: Rob Herring <[email protected]>
---
drivers/net/ethernet/calxeda/xgmac.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/net/ethernet/calxeda/xgmac.c b/drivers/net/ethernet/calxeda/xgmac.c
index a170065..bb1e80b 100644
--- a/drivers/net/ethernet/calxeda/xgmac.c
+++ b/drivers/net/ethernet/calxeda/xgmac.c
@@ -1141,6 +1141,9 @@ static int xgmac_rx(struct xgmac_priv *priv, int limit)
struct sk_buff *skb;
int frame_len;
+ if (!dma_ring_cnt(priv->rx_head, priv->rx_tail, DMA_RX_RING_SZ))
+ break;
+
entry = priv->rx_tail;
p = priv->dma_rx + entry;
if (desc_get_owner(p))
--
1.7.10.4
From: Rob Herring <[email protected]>
WOL is broken because the magic packet status bit is getting set rather
than the enable bit. The PMT interrupt is not getting serviced because
the PMT interrupt is also enabled on the global interrupt, but not
cleared by the global interrupt and the global interrupt is higher
priority. This fixes both of these issues to get WOL working.
There's still a problem with receive after resume, but at least now we
can wake-up.
Signed-off-by: Rob Herring <[email protected]>
---
drivers/net/ethernet/calxeda/xgmac.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/calxeda/xgmac.c b/drivers/net/ethernet/calxeda/xgmac.c
index bb1e80b..b0ebc9f 100644
--- a/drivers/net/ethernet/calxeda/xgmac.c
+++ b/drivers/net/ethernet/calxeda/xgmac.c
@@ -163,6 +163,7 @@
#define XGMAC_FLOW_CTRL_FCB_BPA 0x00000001 /* Flow Control Busy ... */
/* XGMAC_INT_STAT reg */
+#define XGMAC_INT_STAT_PMTIM 0x00800000 /* PMT Interrupt Mask */
#define XGMAC_INT_STAT_PMT 0x0080 /* PMT Interrupt Status */
#define XGMAC_INT_STAT_LPI 0x0040 /* LPI Interrupt Status */
@@ -960,6 +961,9 @@ static int xgmac_hw_init(struct net_device *dev)
writel(DMA_INTR_DEFAULT_MASK, ioaddr + XGMAC_DMA_STATUS);
writel(DMA_INTR_DEFAULT_MASK, ioaddr + XGMAC_DMA_INTR_ENA);
+ /* Mask power mgt interrupt */
+ writel(XGMAC_INT_STAT_PMTIM, ioaddr + XGMAC_INT_STAT);
+
/* XGMAC requires AXI bus init. This is a 'magic number' for now */
writel(0x0077000E, ioaddr + XGMAC_DMA_AXI_BUS);
@@ -1828,7 +1832,7 @@ static void xgmac_pmt(void __iomem *ioaddr, unsigned long mode)
unsigned int pmt = 0;
if (mode & WAKE_MAGIC)
- pmt |= XGMAC_PMT_POWERDOWN | XGMAC_PMT_MAGIC_PKT;
+ pmt |= XGMAC_PMT_POWERDOWN | XGMAC_PMT_MAGIC_PKT_EN;
if (mode & WAKE_UCAST)
pmt |= XGMAC_PMT_POWERDOWN | XGMAC_PMT_GLBL_UNICAST;
--
1.7.10.4
From: Rob Herring <[email protected]>
Date: Thu, 28 Mar 2013 16:32:44 -0500
> From: Rob Herring <[email protected]>
>
> If skb allocation for the rx ring fails repeatedly, we can reach a point
> were the ring is empty. In this condition, the driver is out of sync with
> the h/w. While this has always been possible, the removal of the skb
> recycling seems to have made triggering this problem easier.
>
> Signed-off-by: Rob Herring <[email protected]>
Applied.
From: Rob Herring <[email protected]>
Date: Thu, 28 Mar 2013 16:32:45 -0500
> From: Rob Herring <[email protected]>
>
> WOL is broken because the magic packet status bit is getting set rather
> than the enable bit. The PMT interrupt is not getting serviced because
> the PMT interrupt is also enabled on the global interrupt, but not
> cleared by the global interrupt and the global interrupt is higher
> priority. This fixes both of these issues to get WOL working.
>
> There's still a problem with receive after resume, but at least now we
> can wake-up.
>
> Signed-off-by: Rob Herring <[email protected]>
Applied.