[TG3]: 40-bit DMA workaround part 2

The 40-bit DMA workaround recently implemented for 5714, 5715, and
5780 needs to be expanded because there may be other tg3 devices
behind the EPB Express to PCIX bridge in the 5780 class device.

For example, some 4-port card or mother board designs have 5704 behind
the 5714.

All devices behind the EPB require the 40-bit DMA workaround.

Thanks to Chris Elmquist again for reporting the problem and testing
the patch.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

3 files changed, 45 insertions, 9 deletions

diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c
index b8f1524..caf4102 100644
--- a/drivers/net/tg3.c
+++ b/drivers/net/tg3.c
@@ -9552,12 +9552,36 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 		}
 	}
 
-	/* Find msi capability. */
+	/* The EPB bridge inside 5714, 5715, and 5780 cannot support
+	 * DMA addresses > 40-bit. This bridge may have other additional
+	 * 57xx devices behind it in some 4-port NIC designs for example.
+	 * Any tg3 device found behind the bridge will also need the 40-bit
+	 * DMA workaround.
+	 */
 	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5780 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5714) {
 		tp->tg3_flags2 |= TG3_FLG2_5780_CLASS;
+		tp->tg3_flags |= TG3_FLAG_40BIT_DMA_BUG;
 		tp->msi_cap = pci_find_capability(tp->pdev, PCI_CAP_ID_MSI);
 	}
+	else {
+		struct pci_dev *bridge = NULL;
+
+		do {
+			bridge = pci_get_device(PCI_VENDOR_ID_SERVERWORKS,
+						PCI_DEVICE_ID_SERVERWORKS_EPB,
+						bridge);
+			if (bridge && bridge->subordinate &&
+			    (bridge->subordinate->number <=
+			     tp->pdev->bus->number) &&
+			    (bridge->subordinate->subordinate >=
+			     tp->pdev->bus->number)) {
+				tp->tg3_flags |= TG3_FLAG_40BIT_DMA_BUG;
+				pci_dev_put(bridge);
+				break;
+			}
+		} while (bridge);
+	}
 
 	/* Initialize misc host control in PCI block. */
 	tp->misc_host_ctrl |= (misc_ctrl_reg &
@@ -10303,7 +10327,14 @@ static int __devinit tg3_test_dma(struct tg3 *tp)
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5704) {
 			u32 ccval = (tr32(TG3PCI_CLOCK_CTRL) & 0x1f);
 
-			if (ccval == 0x6 || ccval == 0x7)
+			/* If the 5704 is behind the EPB bridge, we can
+			 * do the less restrictive ONE_DMA workaround for
+			 * better performance.
+			 */
+			if ((tp->tg3_flags & TG3_FLAG_40BIT_DMA_BUG) &&
+			    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5704)
+				tp->dma_rwctrl |= 0x8000;
+			else if (ccval == 0x6 || ccval == 0x7)
 				tp->dma_rwctrl |= DMA_RWCTRL_ONE_DMA;
 
 			/* Set bit 23 to enable PCIX hw bug fix */
@@ -10759,19 +10790,20 @@ static int __devinit tg3_init_one(struct pci_dev *pdev,
 		goto err_out_iounmap;
 	}
 
-	/* 5714, 5715 and 5780 cannot support DMA addresses > 40-bit.
+	/* The EPB bridge inside 5714, 5715, and 5780 and any
+	 * device behind the EPB cannot support DMA addresses > 40-bit.
 	 * On 64-bit systems with IOMMU, use 40-bit dma_mask.
 	 * On 64-bit systems without IOMMU, use 64-bit dma_mask and
 	 * do DMA address check in tg3_start_xmit().
 	 */
-	if (tp->tg3_flags2 & TG3_FLG2_5780_CLASS) {
+	if (tp->tg3_flags2 & TG3_FLG2_IS_5788)
+		persist_dma_mask = dma_mask = DMA_32BIT_MASK;
+	else if (tp->tg3_flags & TG3_FLAG_40BIT_DMA_BUG) {
 		persist_dma_mask = dma_mask = DMA_40BIT_MASK;
 #ifdef CONFIG_HIGHMEM
 		dma_mask = DMA_64BIT_MASK;
 #endif
-	} else if (tp->tg3_flags2 & TG3_FLG2_IS_5788)
-		persist_dma_mask = dma_mask = DMA_32BIT_MASK;
-	else
+	} else
 		persist_dma_mask = dma_mask = DMA_64BIT_MASK;
 
 	/* Configure DMA attributes. */
@@ -10908,8 +10940,10 @@ static int __devinit tg3_init_one(struct pci_dev *pdev,
 	       (tp->tg3_flags & TG3_FLAG_SPLIT_MODE) != 0,
 	       (tp->tg3_flags2 & TG3_FLG2_NO_ETH_WIRE_SPEED) == 0,
 	       (tp->tg3_flags2 & TG3_FLG2_TSO_CAPABLE) != 0);
-	printk(KERN_INFO "%s: dma_rwctrl[%08x]\n",
-	       dev->name, tp->dma_rwctrl);
+	printk(KERN_INFO "%s: dma_rwctrl[%08x] dma_mask[%d-bit]\n",
+	       dev->name, tp->dma_rwctrl,
+	       (pdev->dma_mask == DMA_32BIT_MASK) ? 32 :
+	        (((u64) pdev->dma_mask == DMA_40BIT_MASK) ? 40 : 64));
 
 	return 0;
 
diff --git a/drivers/net/tg3.h b/drivers/net/tg3.h
index 7f4b7f6..7e3b613 100644
--- a/drivers/net/tg3.h
+++ b/drivers/net/tg3.h
@@ -2163,6 +2163,7 @@ struct tg3 {
 #define TG3_FLAG_10_100_ONLY		0x01000000
 #define TG3_FLAG_PAUSE_AUTONEG		0x02000000
 #define TG3_FLAG_IN_RESET_TASK		0x04000000
+#define TG3_FLAG_40BIT_DMA_BUG		0x08000000
 #define TG3_FLAG_BROKEN_CHECKSUMS	0x10000000
 #define TG3_FLAG_GOT_SERDES_FLOWCTL	0x20000000
 #define TG3_FLAG_SPLIT_MODE		0x40000000
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 1709b50..751eea5 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1365,6 +1365,7 @@
 #define PCI_DEVICE_ID_SERVERWORKS_HE	  0x0008
 #define PCI_DEVICE_ID_SERVERWORKS_LE	  0x0009
 #define PCI_DEVICE_ID_SERVERWORKS_GCNB_LE 0x0017
+#define PCI_DEVICE_ID_SERVERWORKS_EPB	  0x0103
 #define PCI_DEVICE_ID_SERVERWORKS_OSB4	  0x0200
 #define PCI_DEVICE_ID_SERVERWORKS_CSB5	  0x0201
 #define PCI_DEVICE_ID_SERVERWORKS_CSB6    0x0203