Pull request for 'r8169-for-jeff-20070919' branch

Pull request for 'r8169-for-jeff-20070919' branch

[Francois Romieu]

Please pull from branch 'r8169-for-jeff-20070919' in repository

git://electric-eye.fr.zoreil.com/home/romieu/linux-2.6.git r8169-for-jeff-20070919

to get the fixes below.

Distance from 'linux-2.6.git' (a88a8eff1e6e32d3288986a9d36c6a449c032d3a)
------------------------------------------------------------------------

d78ae2dcc2acebb9a1048278f47f762c069db75c
65d916d95314566f426cc40ff0f17b754a773b0b

Diffstat
--------

 drivers/net/r8169.c |   14 +++++++++++++-
 1 files changed, 13 insertions(+), 1 deletions(-)

Shortlog
--------

Edward Hsu (1):
      r8169: correct phy parameters for the 8110SC

Francois Romieu (1):
      r8169: workaround against ignored TxPoll writes (8168)

Patch
-----

diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c
index b85ab4a..c921ec3 100644
--- a/drivers/net/r8169.c
+++ b/drivers/net/r8169.c
@@ -1228,7 +1228,10 @@ static void rtl8169_hw_phy_config(struct net_device *dev)
 		return;
 	}
 
-	/* phy config for RTL8169s mac_version C chip */
+	if ((tp->mac_version != RTL_GIGA_MAC_VER_02) &&
+	    (tp->mac_version != RTL_GIGA_MAC_VER_03))
+		return;
+
 	mdio_write(ioaddr, 31, 0x0001);			//w 31 2 0 1
 	mdio_write(ioaddr, 21, 0x1000);			//w 21 15 0 1000
 	mdio_write(ioaddr, 24, 0x65c7);			//w 24 15 0 65c7
@@ -2567,6 +2570,15 @@ static void rtl8169_tx_interrupt(struct net_device *dev,
 		    (TX_BUFFS_AVAIL(tp) >= MAX_SKB_FRAGS)) {
 			netif_wake_queue(dev);
 		}
+		/*
+		 * 8168 hack: TxPoll requests are lost when the Tx packets are
+		 * too close. Let's kick an extra TxPoll request when a burst
+		 * of start_xmit activity is detected (if it is not detected,
+		 * it is slow enough). -- FR
+		 */
+		smp_rmb();
+		if (tp->cur_tx != dirty_tx)
+			RTL_W8(TxPoll, NPQ);
 	}
 }
 
-- 
Ueimor

Re: Pull request for 'r8169-for-jeff-20070919' branch

[Chuck Ebbert]

On 09/19/2007 03:56 PM, Francois Romieu wrote:
> Please pull from branch 'r8169-for-jeff-20070919' in repository
> 

People are still reporting hangs with this card in 2.6.22.6, are there
any fixes appropriate for that?

Re: Pull request for 'r8169-for-jeff-20070919' branch

[Francois Romieu]

Chuck Ebbert <cebbert@redhat.com> :
[...]
> People are still reporting hangs with this card in 2.6.22.6, are there
> any fixes appropriate for that?

$ grep 2.6.22.6 ~/Mail/linux/bugs/pending
$ grep 2.6.22.6 ~/Mail/romieu
[nothing besides Sylvain's success with 2.6.22.6 + the r8169 test driver]
$ less +/2.6.22.6 ~/Mail/linux/netdev/200709 
[nada]
$ less +/816[89] ~/Mail/linux/kernel/200737
[...]

/me scratches head...

The answer is probably "not yet".

Can you refresh my memory with some #BZ or an url ?

-- 
Ueimor

Re: Pull request for 'r8169-for-jeff-20070919' branch

[Chuck Ebbert]

On 09/20/2007 04:44 PM, Francois Romieu wrote:
> Chuck Ebbert <cebbert@redhat.com> :
> [...]
>> People are still reporting hangs with this card in 2.6.22.6, are there
>> any fixes appropriate for that?
> 
> $ grep 2.6.22.6 ~/Mail/linux/bugs/pending
> $ grep 2.6.22.6 ~/Mail/romieu
> [nothing besides Sylvain's success with 2.6.22.6 + the r8169 test driver]
> $ less +/2.6.22.6 ~/Mail/linux/netdev/200709 
> [nada]
> $ less +/816[89] ~/Mail/linux/kernel/200737
> [...]
> 
> /me scratches head...
> 
> The answer is probably "not yet".
> 
> Can you refresh my memory with some #BZ or an url ?
> 

Looking again, it may not be r8169's fault, but it is involved. It
used to hang and could be restarted by pulling the cable and reinserting
it, but now it just locks solid:

https://bugzilla.redhat.com/show_bug.cgi?id=252955

Will ask the reporter to test with a different adapter.

Re: Pull request for 'r8169-for-jeff-20070919' branch

[Sylvain Le Gall]

Hello,

On Thu, Sep 20, 2007 at 05:26:40PM -0400, Chuck Ebbert wrote:
> On 09/20/2007 04:44 PM, Francois Romieu wrote:
> > Chuck Ebbert <cebbert@redhat.com> :
> > [...]
> >> People are still reporting hangs with this card in 2.6.22.6, are there
> >> any fixes appropriate for that?
> > 
> > $ grep 2.6.22.6 ~/Mail/linux/bugs/pending
> > $ grep 2.6.22.6 ~/Mail/romieu
> > [nothing besides Sylvain's success with 2.6.22.6 + the r8169 test driver]
> > $ less +/2.6.22.6 ~/Mail/linux/netdev/200709 
> > [nada]
> > $ less +/816[89] ~/Mail/linux/kernel/200737
> > [...]
> > 
> > /me scratches head...
> > 
> 
> Looking again, it may not be r8169's fault, but it is involved. It
> used to hang and could be restarted by pulling the cable and reinserting
> it, but now it just locks solid:
> 
> https://bugzilla.redhat.com/show_bug.cgi?id=252955
> 
> Will ask the reporter to test with a different adapter.
> 

I permit myself to reply, since i was CCed.

I have this kind of error at the beginning. It was quite strange, but
caused by IRQ problems: 
* eth0, cdrom (ide1 or something like that), audio controler, usb on the
  same IRQ
* from time to time i get very strange behavior:
** lost CDROM
** cannot move mouse
** strange latency on the network (ping to google go from 65ms to 500ms)
* when i :
** use USB
** use audio

In fact, that was a hidden IRQ conflict... I solve it by upgrading BIOS.

Kind regards
Sylvain Le Gall

ps: maybe it has nothing to do with your problem, if so please forgive
me for the noise

Re: Pull request for 'r8169-for-jeff-20070919' branch

[Francois Romieu]

Chuck Ebbert <cebbert@redhat.com> :
[...]
> Looking again, it may not be r8169's fault, but it is involved. It

Sort of:
[...]
* Hung when moving data off a physical disk in an LVM. Twice during night.
* Hung when adding a disk to a software Raid. Once.

> used to hang and could be restarted by pulling the cable and reinserting
> it,

This is already worked around in 2.6.22-stable.

> but now it just locks solid:
> 
> https://bugzilla.redhat.com/show_bug.cgi?id=252955

A part of it look like the usual samba + 816x failure. The most recent
changeset in the patch submitted for inclusion should be applied.

-- 
Ueimor