linuxppc-dev.lists.ozlabs.org archive mirror
 help / color / mirror / Atom feed
From: Sascha Hauer <s.hauer@pengutronix.de>
To: Juergen Beisert <jbe@pengutronix.de>
Cc: linuxppc-embedded@ozlabs.org
Subject: Re: [RFC PATCH v0.1] net driver: mpc52xx fec
Date: Mon, 8 Oct 2007 11:01:35 +0200	[thread overview]
Message-ID: <20071008090135.GB11231@pengutronix.de> (raw)
In-Reply-To: <200709281112.18480.jbe@pengutronix.de>

On Fri, Sep 28, 2007 at 11:12:17AM +0200, Juergen Beisert wrote:
> 
> While network stress testing: 
> 
> [...]
> NETDEV WATCHDOG: eth0: transmit timed out
> net eth0: transmit timed out
> net eth0: queues didn't drain
> net eth0:   tx: index: 35, outdex: 36
> net eth0:   rx: index: 24, outdex: 25
> PHY: f0003000:00 - Link is Down
> PHY: f0003000:00 - Link is Up - 100/Full
> 
> The link is up again, but any connection is dead (no answers to ping etc.). 
> But the serial console is still working. I'm not sure if the RT-Preempt patch 
> *causes* this behavior or only *discover* it. Any idea?

While the previous patch I sent fixed the reset path for the fec
controller this patch makes sure the chip does not have to be resetted.
Problem was that we ran out of receive buffers when we nmapped our
board (nmap sends lots of small packages in a short period of time).
By increasing the number of rx buffers this problem does not appear
anymore.
This patch produces a significant memory overhead to the driver of about
340k, so we might want to have this configurable as a module parameter.
Let me know what the preferred way is, I can update the patch
accordingly.

Sascha


Increase the number of RX packets in the fec_mpc52xx driver. This is
necessary to no run out of rx packets when lots of small packets arrive
in short time (for example when nmapping the board)

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

---
 drivers/net/fec_mpc52xx/fec.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Index: linux-2.6/drivers/net/fec_mpc52xx/fec.h
===================================================================
--- linux-2.6.orig/drivers/net/fec_mpc52xx/fec.h
+++ linux-2.6/drivers/net/fec_mpc52xx/fec.h
@@ -19,7 +19,7 @@
 /* Tunable constant */
 /* FEC_RX_BUFFER_SIZE includes 4 bytes for CRC32 */
 #define FEC_RX_BUFFER_SIZE	1522	/* max receive packet size */
-#define FEC_RX_NUM_BD		64
+#define FEC_RX_NUM_BD		256
 #define FEC_TX_NUM_BD		64
 
 #define FEC_RESET_DELAY		50 	/* uS */

-- 
Pengutronix - Linux Solutions for Science and Industry
Entwicklungszentrum Nord     http://www.pengutronix.de

  parent reply	other threads:[~2007-10-08  9:01 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-08-10  9:51 [RFC PATCH v0.1] net driver: mpc52xx fec Domen Puncer
2007-08-10 13:02 ` Arnaldo Carvalho de Melo
2007-08-13  7:21   ` Domen Puncer
2007-08-18 10:06 ` Domen Puncer
2007-08-19 15:39   ` Matt Sealey
2007-08-20  8:31     ` Domen Puncer
2007-08-20 13:13       ` Domen Puncer
2007-08-20 19:02         ` Matt Sealey
2007-08-21  5:49           ` Domen Puncer
2007-09-02  7:41 ` [RFC PATCH v0.2] " Domen Puncer
2007-09-03 15:57   ` Grant Likely
2007-09-03 16:09     ` Jon Smirl
2007-09-03 16:41       ` Grant Likely
2007-09-15 12:14     ` Domen Puncer
2007-09-17  9:53       ` Sven Luther
2007-09-17 20:21         ` [PATCH] phy: export phy_mii_ioctl Domen Puncer
2007-09-17 22:08           ` Jon Smirl
2007-09-18 15:16             ` Domen Puncer
2007-09-18 19:17               ` Jon Smirl
2007-09-19 11:56                 ` Domen Puncer
2007-09-19 18:44                   ` Jon Smirl
2007-09-19 21:18                     ` Jon Smirl
2007-09-18 19:29               ` Jon Smirl
2007-09-19  8:54                 ` Pedro Luis D. L.
2007-09-19 10:37                   ` Juergen Beisert
2007-09-19 11:38                     ` Pedro Luis D. L.
2007-09-19 14:51                       ` Juergen Beisert
2007-09-19 15:11                         ` Pedro Luis D. L.
2007-09-19 13:56                   ` Jon Smirl
2007-09-19 14:31                     ` Pedro Luis D. L.
2007-09-19  8:54                 ` Pedro Luis D. L.
2007-09-20  6:36           ` Jeff Garzik
2007-10-02 12:49   ` [RFC PATCH v0.2] net driver: mpc52xx fec Sascha Hauer
2007-10-02 14:32     ` Domen Puncer
2007-10-02 15:46       ` Robert Schwebel
2007-09-27 17:07 ` [RFC PATCH v0.1] " Juergen Beisert
2007-09-27 18:12   ` Jon Smirl
2007-09-27 18:43     ` Scott Wood
2007-09-28  9:12       ` Juergen Beisert
2007-09-28 15:40         ` Scott Wood
2007-10-08  8:48         ` Sascha Hauer
2007-10-08  9:01         ` Sascha Hauer [this message]
2007-10-08 16:46           ` Jon Smirl
2007-09-28 15:07   ` Juergen Beisert
2007-09-28 15:38     ` Jon Smirl
2007-10-01  8:35       ` Juergen Beisert
2007-10-01 16:24         ` Juergen Beisert

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=20071008090135.GB11231@pengutronix.de \
    --to=s.hauer@pengutronix.de \
    --cc=jbe@pengutronix.de \
    --cc=linuxppc-embedded@ozlabs.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).