xirc2ps_cs driver timeouts/errors

xirc2ps_cs driver timeouts/errors

[dilinger]

My logs show the following:

xirc2ps_cs.c 1.31 1998/12/09 19:32:55 (dd9jn+kvh)
eth0: Xircom: port 0x300, irq 3, hwaddr 00:80:C7:1E:28:2A
eth0: MII link partner: 0021
eth0: MII selected
eth0: media 10BaseT, silicon revision 4
UDP: short packet: 137/58
NETDEV WATCHDOG: eth0: transmit timed out
eth0: transmit timed out
eth0: MII link partner: 0021
eth0: MII selected
eth0: media 10BaseT, silicon revision 4
NETDEV WATCHDOG: eth0: transmit timed out
eth0: transmit timed out
eth0: MII link partner: 0021
eth0: MII selected
eth0: media 10BaseT, silicon revision 4
UDP: short packet: 0/213
UDP: short packet: 0/58

Each time I get a transmit timeout, or UDP: short packet error,
networking on my laptop seems to go down.  Reinsertion of the
card temporarily fixes it, and if I leave it long enough it
also fixes itself.

Hardware info:

Socket 0:
  product info: "Xircom", "CreditCard 10/100", "CE3-10/100", "1.00"
  manfid: 0x0105, 0x010a
  function: 6 (network)

Socket 0:
  Vcc 5.0V  Vpp1 0.0V  Vpp2 0.0V
  interface type is "memory and I/O"
  irq 3 [exclusive] [level]
  function 0:
    config base 0x0800
      option 0x41 status 0x00
    io 0x0300-0x030f [16bit]

  Bus  0, device   3, function  0:
    CardBus bridge: Texas Instruments PCI1225 (rev 1).
      IRQ 11.
      Master Capable.  Latency=168.  Min Gnt=192.Max Lat=5.
      Non-prefetchable 32 bit memory at 0x10000000 [0x10000fff].
  Bus  0, device   3, function  1:
    CardBus bridge: Texas Instruments PCI1225 (#2) (rev 1).
      IRQ 11.
      Master Capable.  Latency=168.  Min Gnt=192.Max Lat=5.
      Non-prefetchable 32 bit memory at 0x10001000 [0x10001fff].

Kernel is stock 2.4.1.  If any more info is needed, let me know.
Please CC replies to me, as I'm not on the list.

-- 
"... being a Linux user is sort of like living in a house inhabited
by a large family of carpenters and architects. Every morning when
you wake up, the house is a little different. Maybe there is a new
turret, or some walls have moved. Or perhaps someone has temporarily
removed the floor under your bed." - Unix for Dummies, 2nd Edition
        -- found in the .sig of Rob Riggs, rriggs@tesser.com
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: xirc2ps_cs driver timeouts/errors

[David Hinds]

On Fri, Feb 02, 2001 at 11:54:31PM -0500, dilinger@mp3revolution.net wrote:
> 
> Each time I get a transmit timeout, or UDP: short packet error,
> networking on my laptop seems to go down.  Reinsertion of the
> card temporarily fixes it, and if I leave it long enough it
> also fixes itself.

Does the same happen with a 2.2 kernel and the 3.1.24 PCMCIA drivers?
There is a bug fix in the 3.1.24 xirc2ps_cs driver that hasn't been
merged into the kernel tree yet.

-- Dave
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: xirc2ps_cs driver timeouts/errors

[dilinger]

[-- Attachment #1: Type: text/plain, Size: 1574 bytes --]

I downloaded the pcmcia-cs 3.1.24 package and hand-merged the bugfix(es?)
into 2.4.1's xirc2ps_cs.c.  A(n attempt at a) patch to bring 2.4.1
up to pcmcia-cs's version is attached to this email.  So far, no
problems, but it'll be at least 48 hours before I can say whether
it happens or not..  This is also available at
http://incandescent.mp3revolution.net/kernel/xirc2ps_cs/.


Looking through logs, it appears as though the problem first
appeared in kernel 2.4.1-pre7, and has become frequent enough
to annoy me as of 2.4.1.  Did anything change dealing w/ this
driver/card between 2.4.0 and 2.4.1-pre7?



On Fri, Feb 02, 2001 at 09:20:40PM -0800, David Hinds wrote:
> 
> On Fri, Feb 02, 2001 at 11:54:31PM -0500, dilinger@mp3revolution.net wrote:
> > 
> > Each time I get a transmit timeout, or UDP: short packet error,
> > networking on my laptop seems to go down.  Reinsertion of the
> > card temporarily fixes it, and if I leave it long enough it
> > also fixes itself.
> 
> Does the same happen with a 2.2 kernel and the 3.1.24 PCMCIA drivers?
> There is a bug fix in the 3.1.24 xirc2ps_cs driver that hasn't been
> merged into the kernel tree yet.
> 
> -- Dave

-- 
"... being a Linux user is sort of like living in a house inhabited
by a large family of carpenters and architects. Every morning when
you wake up, the house is a little different. Maybe there is a new
turret, or some walls have moved. Or perhaps someone has temporarily
removed the floor under your bed." - Unix for Dummies, 2nd Edition
        -- found in the .sig of Rob Riggs, rriggs@tesser.com

[-- Attachment #2: xirc2ps_cs.c.diff --]
[-- Type: text/plain, Size: 1280 bytes --]

diff -urN linux.old/drivers/net/pcmcia/xirc2ps_cs.c linux/drivers/net/pcmcia/xirc2ps_cs.c
--- linux.old/drivers/net/pcmcia/xirc2ps_cs.c	Thu Oct 26 19:52:16 2000
+++ linux/drivers/net/pcmcia/xirc2ps_cs.c	Sat Feb  3 01:15:49 2001
@@ -382,6 +382,7 @@
 static void do_powerdown(struct net_device *dev);
 static int do_stop(struct net_device *dev);
 
+
 /*=============== Helper functions =========================*/
 static void
 flush_stale_links(void)
@@ -1348,7 +1349,6 @@
 	     * packets */
 	    lp->stats.rx_dropped++;
 	    DEBUG(2, "%s: RX drop, too much done\n", dev->name);
-	    PutWord(XIRCREG0_DO, 0x8000); /* issue cmd: skip_rx_packet */
 	} else if (rsr & PktRxOk) {
 	    struct sk_buff *skb;
 
@@ -1423,8 +1423,7 @@
 		if (!(rsr & PhyPkt))
 		    lp->stats.multicast++;
 	    }
-	    PutWord(XIRCREG0_DO, 0x8000); /* issue cmd: skip_rx_packet */
-	} else {
+	} else { /* bad packet */
 	    DEBUG(5, "rsr=%#02x\n", rsr);
 	}
 	if (rsr & PktTooLong) {
@@ -1439,6 +1438,9 @@
 	    lp->stats.rx_fifo_errors++; /* okay ? */
 	    DEBUG(3, "%s: Alignment error\n", dev->name);
 	}
+	
+	/* clear the received/dropped/error packet */
+	PutWord(XIRCREG0_DO, 0x8000); /* issue cmd: skip_rx_packet */
 
 	/* get the new ethernet status */
 	eth_status = GetByte(XIRCREG_ESR);

Re: xirc2ps_cs driver timeouts/errors

[dilinger]

A few minutes ago, I got the following:
Feb  3 13:09:10 pea kernel: UDP: short packet: 0/58

.. and the driver hung.  I had to reinsert the card to get
networking back.  This is after 10 hours of uptime, using
the patch i sent previously against 2.4.1's xirc2ps_cs.
So the answer is, yes, it happens w/ 3.1.24's driver.  Any
suggestions?

On Fri, Feb 02, 2001 at 09:20:40PM -0800, David Hinds wrote:
> 
> On Fri, Feb 02, 2001 at 11:54:31PM -0500, dilinger@mp3revolution.net wrote:
> > 
> > Each time I get a transmit timeout, or UDP: short packet error,
> > networking on my laptop seems to go down.  Reinsertion of the
> > card temporarily fixes it, and if I leave it long enough it
> > also fixes itself.
> 
> Does the same happen with a 2.2 kernel and the 3.1.24 PCMCIA drivers?
> There is a bug fix in the 3.1.24 xirc2ps_cs driver that hasn't been
> merged into the kernel tree yet.
> 
> -- Dave

-- 
"... being a Linux user is sort of like living in a house inhabited
by a large family of carpenters and architects. Every morning when
you wake up, the house is a little different. Maybe there is a new
turret, or some walls have moved. Or perhaps someone has temporarily
removed the floor under your bed." - Unix for Dummies, 2nd Edition
        -- found in the .sig of Rob Riggs, rriggs@tesser.com
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/