From: "Kok, Auke" <auke-jan.h.kok@intel.com>
To: "Garzik, Jeff" <jgarzik@pobox.com>
Cc: netdev@vger.kernel.org, "Brandeburg,
Jesse" <jesse.brandeburg@intel.com>,
"Kok, Auke" <auke-jan.h.kok@intel.com>,
"Kok, Auke" <auke@foo-projects.org>,
"Ronciak, John" <john.ronciak@intel.com>
Subject: [PATCH 21/23] ixgb: combine more rx descriptors to improve performance
Date: Tue, 19 Sep 2006 10:29:20 -0700 [thread overview]
Message-ID: <20060919172920.4605.24456.stgit@gitlost.site> (raw)
In-Reply-To: <20060919172623.4605.56860.stgit@gitlost.site>
We experimented with more descriptor buffer writebacks and found that
values larger than 8 give HW problems, but 8 is safe and gives us some
improved performance.
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
---
drivers/net/ixgb/ixgb.h | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/ixgb/ixgb.h b/drivers/net/ixgb/ixgb.h
index a1a7f3b..50ffe90 100644
--- a/drivers/net/ixgb/ixgb.h
+++ b/drivers/net/ixgb/ixgb.h
@@ -111,7 +111,7 @@ struct ixgb_adapter;
#define IXGB_RXBUFFER_16384 16384
/* How many Rx Buffers do we bundle into one write to the hardware ? */
-#define IXGB_RX_BUFFER_WRITE 4 /* Must be power of 2 */
+#define IXGB_RX_BUFFER_WRITE 8 /* Must be power of 2 */
/* only works for sizes that are powers of 2 */
#define IXGB_ROUNDUP(i, size) ((i) = (((i) + (size) - 1) & ~((size) - 1)))
---
Auke Kok <auke-jan.h.kok@intel.com>
next prev parent reply other threads:[~2006-09-19 17:20 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-09-19 17:26 [PATCH 00/23] e100, e1000, ixgb updates Kok, Auke
2006-09-19 17:28 ` [PATCH 01/23] e100, e1000, ixgb: update copyright header and remove LICENSE Kok, Auke
2006-09-19 17:28 ` [PATCH 02/23] e100, e1000, ixgb: Fix an impossible memory overwrite bug Kok, Auke
2006-09-19 17:28 ` [PATCH 03/23] e100: Add debugging code for cb cleaning and csum failures Kok, Auke
2006-09-19 19:10 ` Jeff Garzik
2006-09-19 22:25 ` Auke Kok
2006-09-19 21:33 ` Dave Jones
2006-09-19 21:40 ` Jeff Garzik
2006-09-19 21:46 ` Dave Jones
2006-09-19 22:21 ` Auke Kok
2006-09-19 17:28 ` [PATCH 04/23] e100: rework WoL and shutdown handling Kok, Auke
2006-09-19 19:10 ` Jeff Garzik
2006-09-19 20:25 ` Auke Kok
2006-09-19 20:30 ` Jeff Garzik
2006-09-19 21:38 ` Auke Kok
2006-09-19 17:28 ` [PATCH 05/23] e1000: rename flow control symbols Kok, Auke
2006-09-19 19:11 ` Jeff Garzik
2006-09-19 21:45 ` Auke Kok
2006-09-19 17:28 ` [PATCH 06/23] e1000: add enums for several link properties Kok, Auke
2006-09-19 17:28 ` [PATCH 07/23] e1000: remove unused code and make symbols static Kok, Auke
2006-09-19 19:28 ` Jeff Garzik
2006-09-19 17:28 ` [PATCH 08/23] e1000: add multicast stats counters Kok, Auke
2006-09-19 19:28 ` Jeff Garzik
2006-09-19 17:28 ` [PATCH 09/23] e1000: allow ethtool to pass arbitrary speed advertisment Kok, Auke
2006-09-19 17:28 ` [PATCH 10/23] e1000: Fix MANC detection for PCIE adapters Kok, Auke
2006-09-19 19:31 ` Jeff Garzik
2006-09-19 17:28 ` [PATCH 11/23] e1000: Jumbo frames fixes for 82573 Kok, Auke
2006-09-19 19:32 ` Jeff Garzik
2006-09-20 15:37 ` Auke Kok
2006-09-19 17:28 ` [PATCH 12/23] e1000: Maybe stop TX if not enough free descriptors Kok, Auke
2006-09-19 19:37 ` Jeff Garzik
2006-09-19 19:38 ` Jeff Garzik
2006-09-19 19:50 ` Stephen Hemminger
2006-09-19 20:45 ` Jeff Garzik
2006-09-19 20:59 ` Stephen Hemminger
2006-09-19 23:26 ` Auke Kok
2006-09-19 23:59 ` Stephen Hemminger
2006-09-20 15:30 ` Auke Kok
[not found] ` <4807377b0609211659mb0cb308hcbc78520148300a1@mail.gmail.com>
2006-09-24 11:30 ` Fwd: " Herbert Xu
2006-09-19 17:29 ` [PATCH 13/23] e1000: gather hardware bit tweaks Kok, Auke
2006-09-19 19:36 ` Jeff Garzik
2006-09-27 20:12 ` Auke Kok
2006-09-19 17:29 ` [PATCH 14/23] e1000: handle manageability for pci-e adapters at PHY powerdown Kok, Auke
2006-09-19 19:39 ` Jeff Garzik
2006-09-19 17:29 ` [PATCH 15/23] e1000: add PCI-E capability detection code Kok, Auke
2006-09-19 17:29 ` [PATCH 16/23] e1000: reduce RAR entries available for ICH8 Kok, Auke
2006-09-19 19:39 ` Jeff Garzik
2006-09-19 17:29 ` [PATCH 17/23] e1000: driver state fixes (race fix) Kok, Auke
2006-09-19 19:40 ` Jeff Garzik
2006-09-19 17:29 ` [PATCH 18/23] e1000: revert 'e1000: Remove 0x1000 as supported device' Kok, Auke
2006-09-19 19:40 ` Jeff Garzik
2006-09-19 20:34 ` Auke Kok
2006-09-19 20:43 ` Jeff Garzik
2006-09-19 17:29 ` [PATCH 19/23] e1000: rework polarity, NVM, eeprom code and fixes Kok, Auke
2006-09-19 17:29 ` [PATCH 20/23] e1000: don't strip vlan ID if 8021q claims it Kok, Auke
2006-09-19 19:41 ` Jeff Garzik
2006-09-19 17:29 ` Kok, Auke [this message]
2006-09-19 17:29 ` [PATCH 22/23] ixgb: convert to netdev_priv(netdev) Kok, Auke
2006-09-19 17:29 ` [PATCH 23/23] e100, e1000, ixgb: increment version numbers Kok, Auke
2006-09-19 19:41 ` Jeff Garzik
2006-09-19 19:42 ` [PATCH 00/23] e100, e1000, ixgb updates Jeff Garzik
2006-09-19 21:06 ` Auke Kok
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=20060919172920.4605.24456.stgit@gitlost.site \
--to=auke-jan.h.kok@intel.com \
--cc=auke@foo-projects.org \
--cc=jesse.brandeburg@intel.com \
--cc=jgarzik@pobox.com \
--cc=john.ronciak@intel.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).