Pull request for 'jg-20061103-00' tag

Pull request for 'jg-20061103-00' tag

[Francois Romieu]

Please pull from tag 'jg-20061103-00' in repository

git://electric-eye.fr.zoreil.com/home/romieu/linux-2.6.git jg-20061103-00

to get the changes below.

Distance from 'upstream-fixes'
-----------------------------

17fddc34b36fc26aa8b5f130fe32b446d9d88fa2

Diffstat
--------

 drivers/net/r8169.c |   22 ++++++++++++++++++++--
 1 files changed, 20 insertions(+), 2 deletions(-)

Shortlog
--------

Francois Romieu:
      r8169: perform a PHY reset before any other operation at boot time

Patch
-----

diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c
index 27f90b2..50b753d 100644
--- a/drivers/net/r8169.c
+++ b/drivers/net/r8169.c
@@ -571,8 +571,8 @@ static void rtl8169_xmii_reset_enable(vo
 {
 	unsigned int val;
 
-	val = (mdio_read(ioaddr, MII_BMCR) | BMCR_RESET) & 0xffff;
-	mdio_write(ioaddr, MII_BMCR, val);
+	mdio_write(ioaddr, MII_BMCR, BMCR_RESET);
+	val = mdio_read(ioaddr, MII_BMCR);
 }
 
 static void rtl8169_check_link_status(struct net_device *dev,
@@ -1406,6 +1406,22 @@ static void rtl8169_release_board(struct
 	free_netdev(dev);
 }
 
+static void rtl8169_phy_reset(struct net_device *dev,
+			      struct rtl8169_private *tp)
+{
+	void __iomem *ioaddr = tp->mmio_addr;
+	int i;
+
+	tp->phy_reset_enable(ioaddr);
+	for (i = 0; i < 100; i++) {
+		if (!tp->phy_reset_pending(ioaddr))
+			return;
+		msleep(1);
+	}
+	if (netif_msg_link(tp))
+		printk(KERN_ERR "%s: PHY reset failed.\n", dev->name);
+}
+
 static void rtl8169_init_phy(struct net_device *dev, struct rtl8169_private *tp)
 {
 	void __iomem *ioaddr = tp->mmio_addr;
@@ -1434,6 +1450,8 @@ static void rtl8169_init_phy(struct net_
 
 	rtl8169_link_option(board_idx, &autoneg, &speed, &duplex);
 
+	rtl8169_phy_reset(dev, tp);
+
 	rtl8169_set_speed(dev, autoneg, speed, duplex);
 
 	if ((RTL_R8(PHYstatus) & TBI_Enable) && netif_msg_link(tp))
-- 
Ueimor

Re: Pull request for 'jg-20061103-00' tag

[Jeff Garzik]

Francois Romieu wrote:
> Please pull from tag 'jg-20061103-00' in repository
> 
> git://electric-eye.fr.zoreil.com/home/romieu/linux-2.6.git jg-20061103-00
> 
> to get the changes below.
> 
> Distance from 'upstream-fixes'
> -----------------------------
> 
> 17fddc34b36fc26aa8b5f130fe32b446d9d88fa2
> 
> Diffstat
> --------
> 
>  drivers/net/r8169.c |   22 ++++++++++++++++++++--
>  1 files changed, 20 insertions(+), 2 deletions(-)
> 
> Shortlog
> --------
> 
> Francois Romieu:
>       r8169: perform a PHY reset before any other operation at boot time

This warrants much more testing than pushing into 2.6.19-rc4 would give 
us, so I'm pulling it into #upstream.

In the past, with 10/100 hubs or ancient Cisco switches, we really 
didn't want to reset the phy and restart autonegotiation, because that 
might be problematic.

In any case, this is a behavior change that may solve problems... but 
also needs testing to insure that it doesn't also cause problems.

	Jeff

Re: Pull request for 'jg-20061103-00' tag

[Francois Romieu]

Jeff Garzik <jeff@garzik.org> :
[...]
> This warrants much more testing than pushing into 2.6.19-rc4 would give 
> us, so I'm pulling it into #upstream.
> 
> In the past, with 10/100 hubs or ancient Cisco switches, we really 
> didn't want to reset the phy and restart autonegotiation, because that 
> might be problematic.
> 
> In any case, this is a behavior change that may solve problems... but 
> also needs testing to insure that it doesn't also cause problems.

I don't have a single report of a 0x8136 user whose network adapter would
work out of the box without the patch. The sole users of a 0x8136 which I
have heard of need it. See the "r8169 mac address change" thread on l-k
last week. Those users should be able to workaround the issue on 2.6.19
with mii-tool (a bit sucky but...).

Regarding testing, see below.

I have done a quick grep through the l-k mailboxes of these last weeks to
see how much reports included the r8169 module (even in non-r8169 related
bugs of course):
200643: 1
200642: 0
200641: 1 (tainted, smp, preempt)
200640: 0
200639: 0
200638: 0
200637: 0
200636: 1 (Darren Salt, proud 0x8136 user)
200635: 0
200634: 0
200633: 1 (*doh* plain r8169 issue under oom)

Mid august here, i.e. 2 months and a half.

I'll ping known r8169 users, especially those who had link issues in
the past but I can not force them at gunpoint.

If 2.6.19 is more than 2 weeks away, we will probably trade two weeks
of close observation against two months of supplementary delay for no
real stability gain.

Well, MII likes to eat babies anyway. :o/

-- 
Ueimor