From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
To: davem@davemloft.net
Cc: Alexander Duyck <alexander.h.duyck@intel.com>,
netdev@vger.kernel.org, gospo@redhat.com, sassmann@redhat.com,
Greg Rose <gregory.v.rose@intel.com>,
Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Subject: [net-next 5/8] ixgbevf: Cleanup accounting for space needed at start of xmit_frame
Date: Tue, 17 Jul 2012 03:09:16 -0700 [thread overview]
Message-ID: <1342519759-25137-6-git-send-email-jeffrey.t.kirsher@intel.com> (raw)
In-Reply-To: <1342519759-25137-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: Alexander Duyck <alexander.h.duyck@intel.com>
This change cleans up the accounting needed at the start of xmit_frame so
that we can avoid doing too much work to determine how many descriptors we
will need.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Greg Rose <gregory.v.rose@intel.com>
Tested-by: Sibai Li <sibai.li@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c | 46 +++++++++++----------
1 file changed, 24 insertions(+), 22 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
index 75af192..855bb21 100644
--- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
+++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
@@ -175,10 +175,8 @@ static void ixgbevf_unmap_and_free_tx_resource(struct ixgbevf_adapter *adapter,
#define IXGBE_MAX_DATA_PER_TXD (1 << IXGBE_MAX_TXD_PWR)
/* Tx Descriptors needed, worst case */
-#define TXD_USE_COUNT(S) (((S) >> IXGBE_MAX_TXD_PWR) + \
- (((S) & (IXGBE_MAX_DATA_PER_TXD - 1)) ? 1 : 0))
-#define DESC_NEEDED (TXD_USE_COUNT(IXGBE_MAX_DATA_PER_TXD) /* skb->data */ + \
- MAX_SKB_FRAGS * TXD_USE_COUNT(PAGE_SIZE) + 1) /* for context */
+#define TXD_USE_COUNT(S) DIV_ROUND_UP((S), IXGBE_MAX_DATA_PER_TXD)
+#define DESC_NEEDED (MAX_SKB_FRAGS + 4)
static void ixgbevf_tx_timeout(struct net_device *netdev);
@@ -2932,33 +2930,37 @@ static int ixgbevf_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
unsigned int tx_flags = 0;
u8 hdr_len = 0;
int r_idx = 0, tso;
- int count = 0;
-
- unsigned int f;
+ u16 count = TXD_USE_COUNT(skb_headlen(skb));
+#if PAGE_SIZE > IXGBE_MAX_DATA_PER_TXD
+ unsigned short f;
+#endif
tx_ring = &adapter->tx_ring[r_idx];
+ /*
+ * need: 1 descriptor per page * PAGE_SIZE/IXGBE_MAX_DATA_PER_TXD,
+ * + 1 desc for skb_headlen/IXGBE_MAX_DATA_PER_TXD,
+ * + 2 desc gap to keep tail from touching head,
+ * + 1 desc for context descriptor,
+ * otherwise try next time
+ */
+#if PAGE_SIZE > IXGBE_MAX_DATA_PER_TXD
+ for (f = 0; f < skb_shinfo(skb)->nr_frags; f++)
+ count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size);
+#else
+ count += skb_shinfo(skb)->nr_frags;
+#endif
+ if (ixgbevf_maybe_stop_tx(netdev, tx_ring, count + 3)) {
+ adapter->tx_busy++;
+ return NETDEV_TX_BUSY;
+ }
+
if (vlan_tx_tag_present(skb)) {
tx_flags |= vlan_tx_tag_get(skb);
tx_flags <<= IXGBE_TX_FLAGS_VLAN_SHIFT;
tx_flags |= IXGBE_TX_FLAGS_VLAN;
}
- /* four things can cause us to need a context descriptor */
- if (skb_is_gso(skb) ||
- (skb->ip_summed == CHECKSUM_PARTIAL) ||
- (tx_flags & IXGBE_TX_FLAGS_VLAN))
- count++;
-
- count += TXD_USE_COUNT(skb_headlen(skb));
- for (f = 0; f < skb_shinfo(skb)->nr_frags; f++)
- count += TXD_USE_COUNT(skb_frag_size(&skb_shinfo(skb)->frags[f]));
-
- if (ixgbevf_maybe_stop_tx(netdev, tx_ring, count)) {
- adapter->tx_busy++;
- return NETDEV_TX_BUSY;
- }
-
first = tx_ring->next_to_use;
if (skb->protocol == htons(ETH_P_IP))
--
1.7.10.4
next prev parent reply other threads:[~2012-07-17 10:09 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-07-17 10:09 [net-next 0/8][pull request] Intel Wired LAN Driver Updates Jeff Kirsher
2012-07-17 10:09 ` [net-next 1/8] ixgbevf: Drop all dead or unnecessary code Jeff Kirsher
2012-07-17 10:09 ` [net-next 2/8] ixgbevf: Drop netdev_registered value since that is already stored in netdev Jeff Kirsher
2012-07-17 10:09 ` [net-next 3/8] ixgbevf: Make use of NETIF_F_RXCSUM instead of keeping our own flag Jeff Kirsher
2012-07-17 10:09 ` [net-next 4/8] ixgbevf: Drop use of eitr_low and eitr_high for hard coded values Jeff Kirsher
2012-07-17 10:09 ` Jeff Kirsher [this message]
2012-07-17 10:09 ` [net-next 6/8] ixgbevf: Update q_vector to contain ring pointers instead of bitmaps Jeff Kirsher
2012-07-17 10:09 ` [net-next 7/8] ixgbevf: Move Tx clean-up into NAPI context Jeff Kirsher
2012-07-17 10:09 ` [net-next 8/8] ixgbevf: Use igb style interrupt masks instead of ixgbe style Jeff Kirsher
2012-07-17 10:22 ` [net-next 0/8][pull request] Intel Wired LAN Driver Updates David Miller
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1342519759-25137-6-git-send-email-jeffrey.t.kirsher@intel.com \
--to=jeffrey.t.kirsher@intel.com \
--cc=alexander.h.duyck@intel.com \
--cc=davem@davemloft.net \
--cc=gospo@redhat.com \
--cc=gregory.v.rose@intel.com \
--cc=netdev@vger.kernel.org \
--cc=sassmann@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).