From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
To: davem@davemloft.net
Cc: netdev@vger.kernel.org, gospo@redhat.com,
Jesse Brandeburg <jesse.brandeburg@intel.com>,
Breno Leitao <leitao@linux.vnet.ibm.com>,
Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Subject: [net-next PATCH 2/2] ixbge: fix bug when using large pages and jumbo frames
Date: Tue, 24 Feb 2009 15:40:56 -0800 [thread overview]
Message-ID: <20090224234056.29312.78842.stgit@lost.foo-projects.org> (raw)
In-Reply-To: <20090224234037.29312.54427.stgit@lost.foo-projects.org>
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
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.
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
CC: Breno Leitao <leitao@linux.vnet.ibm.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ixgbe/ixgbe.h | 1 +
drivers/net/ixgbe/ixgbe_main.c | 9 ++++++++-
2 files changed, 9 insertions(+), 1 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe.h b/drivers/net/ixgbe/ixgbe.h
index c0e56aa..2d877da 100644
--- a/drivers/net/ixgbe/ixgbe.h
+++ b/drivers/net/ixgbe/ixgbe.h
@@ -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
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index e0d736c..6564235 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -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) &
next prev parent reply other threads:[~2009-02-24 23:41 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-02-24 23:40 [net-next PATCH 1/2] ixgbe: Move ring features into an enum, allowing easier future maintenance Jeff Kirsher
2009-02-24 23:40 ` Jeff Kirsher [this message]
2009-02-25 0:37 ` [net-next PATCH 2/2] ixbge: fix bug when using large pages and jumbo frames David Miller
2009-02-25 0:37 ` [net-next PATCH 1/2] ixgbe: Move ring features into an enum, allowing easier future maintenance 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=20090224234056.29312.78842.stgit@lost.foo-projects.org \
--to=jeffrey.t.kirsher@intel.com \
--cc=davem@davemloft.net \
--cc=gospo@redhat.com \
--cc=jesse.brandeburg@intel.com \
--cc=leitao@linux.vnet.ibm.com \
--cc=netdev@vger.kernel.org \
/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).