Skip to contentSkip to conversation

ixgbe: Move ring features into an enum, allowing easier future maintenance

PATCH Applied Jeff Kirsher 2 participants 4 messages Feb 24, 2009 – Feb 25, 2009

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/2 ixgbe: Move ring features into an enum, allowing easier future maintenance

From: Shannon Nelson <[email protected]>

The ring_feature member of ixgbe_adapter is statically allocated based on the supported features of the device. When a new feature is added, we need to manually update the static allocation. This patch makes the feature list an enum, eliminating the need for multiple updates to the code when adding a new feature.

---

 drivers/net/ixgbe/ixgbe.h |   14 ++++++++++----
 1 files changed, 10 insertions(+), 4 deletions(-)
drivers/net/ixgbe/ixgbe.h +10 -4
index e98ace8..c0e56aa 100644
@@ -148,9 +148,15 @@ struct ixgbe_ring {
 	u16 rx_buf_len;
 };
 
-#define RING_F_DCB  0
-#define RING_F_VMDQ 1
-#define RING_F_RSS  2
+enum ixgbe_ring_f_enum {
+	RING_F_NONE = 0,
+	RING_F_DCB,
+	RING_F_VMDQ,
+	RING_F_RSS,
+
+	RING_F_ARRAY_SIZE      /* must be last in enum set */
+};
+
 #define IXGBE_MAX_DCB_INDICES   8
 #define IXGBE_MAX_RSS_INDICES  16
 #define IXGBE_MAX_VMDQ_INDICES 16
@@ -249,7 +255,7 @@ struct ixgbe_adapter {
 	u64 non_eop_descs;
 	int num_msix_vectors;
 	int max_msix_q_vectors;         /* true count of q_vectors for device */
-	struct ixgbe_ring_feature ring_feature[3];
+	struct ixgbe_ring_feature ring_feature[RING_F_ARRAY_SIZE];
 	struct msix_entry *msix_entries;
 
 	u64 rx_hdr_split;
#1 lore raw

PATCH 2/2 ixbge: fix bug when using large pages and jumbo frames

From: Jesse Brandeburg <[email protected]>

it was pointed out on the list that ixgbe was failing when using 64kB pages and large 16kB MTU.

since with a 64kB PAGE_SIZE MAX_SKB_FRAGS = 3, the way the driver was configuring page usage was assuming 2kB is half a page, and was only ever dmaing that much data to a half page.

(16kB - header size) / 2048 = 7 or 8 pages, which would far exceed 3

adjust the driver to account for these large pages, the hardware can support DMA to up to 16kB for each descriptor.

---

 drivers/net/ixgbe/ixgbe.h      |    1 +
 drivers/net/ixgbe/ixgbe_main.c |    9 ++++++++-
 2 files changed, 9 insertions(+), 1 deletions(-)
drivers/net/ixgbe/ixgbe.h +1 -0
index c0e56aa..2d877da 100644
@@ -71,6 +71,7 @@
 #define IXGBE_RXBUFFER_128   128    /* Used for packet split */
 #define IXGBE_RXBUFFER_256   256    /* Used for packet split */
 #define IXGBE_RXBUFFER_2048  2048
+#define IXGBE_MAX_RXBUFFER   16384  /* largest size for a single descriptor */
 
 #define IXGBE_RX_HDR_SIZE IXGBE_RXBUFFER_256
 
drivers/net/ixgbe/ixgbe_main.c +8 -1
index e0d736c..6564235 100644
@@ -1550,7 +1550,14 @@ static void ixgbe_configure_srrctl(struct ixgbe_adapter *adapter, int index)
 	srrctl &= ~IXGBE_SRRCTL_BSIZEPKT_MASK;
 
 	if (adapter->flags & IXGBE_FLAG_RX_PS_ENABLED) {
-		srrctl |= IXGBE_RXBUFFER_2048 >> IXGBE_SRRCTL_BSIZEPKT_SHIFT;
+		u16 bufsz = IXGBE_RXBUFFER_2048;
+		/* grow the amount we can receive on large page machines */
+		if (bufsz < (PAGE_SIZE / 2))
+			bufsz = (PAGE_SIZE / 2);
+		/* cap the bufsz at our largest descriptor size */
+		bufsz = min((u16)IXGBE_MAX_RXBUFFER, bufsz);
+
+		srrctl |= bufsz >> IXGBE_SRRCTL_BSIZEPKT_SHIFT;
 		srrctl |= IXGBE_SRRCTL_DESCTYPE_HDR_SPLIT_ALWAYS;
 		srrctl |= ((IXGBE_RX_HDR_SIZE <<
 		            IXGBE_SRRCTL_BSIZEHDRSIZE_SHIFT) &
maintainer Applied
#2 lore raw

Re: [net-next PATCH 2/2] ixbge: fix bug when using large pages and jumbo frames

From: Jeff Kirsher <[email protected]>
Date: Tue, 24 Feb 2009 15:40:56 -0800

Quoted 7 lines

it was pointed out on the list that ixgbe was failing when using 64kB pages and large 16kB MTU.

since with a 64kB PAGE_SIZE MAX_SKB_FRAGS = 3, the way the driver was configuring page usage was assuming 2kB is half a page, and was only ever dmaing that much data to a half page.

Quoted 7 lines

(16kB - header size) / 2048 = 7 or 8 pages, which would far exceed 3

adjust the driver to account for these large pages, the hardware can support DMA to up to 16kB for each descriptor.

Applied.

maintainer Applied
#3 lore raw
In reply to Jeff Kirsher

From: Jeff Kirsher <[email protected]>
Date: Tue, 24 Feb 2009 15:40:37 -0800

Quoted 10 lines

From: Shannon Nelson <[email protected]>

The ring_feature member of ixgbe_adapter is statically allocated based on the supported features of the device. When a new feature is added, we need to manually update the static allocation. This patch makes the feature list an enum, eliminating the need for multiple updates to the code when adding a new feature.

Applied.

I would appreciate it if, in the future, Shannon would signoff his patches :-)