[PATCH 1/3] e1000: FIX: be ready for incoming irq at pci_request_irq

[PATCH 1/3] e1000: FIX: be ready for incoming irq at pci_request_irq

[Auke Kok]

From: Auke Kok <auke-jan.h.kok@intel.com>

DEBUG_SHIRQ code exposed that e1000 was not ready for incoming interrupts
after having called pci_request_irq. This obviously requires us to finish
our software setup which assigns the irq handler before we request the
irq.

Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
---

 drivers/net/e1000/e1000_main.c |   66 +++++++++++++++++++++++++++-------------
 1 files changed, 45 insertions(+), 21 deletions(-)

diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c
index 98215fd..fff3bc0 100644
--- a/drivers/net/e1000/e1000_main.c
+++ b/drivers/net/e1000/e1000_main.c
@@ -522,14 +522,15 @@ e1000_release_manageability(struct e1000_adapter *adapter)
 	}
 }
 
-int
-e1000_up(struct e1000_adapter *adapter)
+/**
+ * e1000_configure - configure the hardware for RX and TX
+ * @adapter = private board structure
+ **/
+static void e1000_configure(struct e1000_adapter *adapter)
 {
 	struct net_device *netdev = adapter->netdev;
 	int i;
 
-	/* hardware has been reset, we need to reload some things */
-
 	e1000_set_multi(netdev);
 
 	e1000_restore_vlan(adapter);
@@ -548,14 +549,20 @@ e1000_up(struct e1000_adapter *adapter)
 	}
 
 	adapter->tx_queue_len = netdev->tx_queue_len;
+}
+
+int e1000_up(struct e1000_adapter *adapter)
+{
+	/* hardware has been reset, we need to reload some things */
+	e1000_configure(adapter);
+
+	clear_bit(__E1000_DOWN, &adapter->flags);
 
 #ifdef CONFIG_E1000_NAPI
-	netif_poll_enable(netdev);
+	netif_poll_enable(adapter->netdev);
 #endif
 	e1000_irq_enable(adapter);
 
-	clear_bit(__E1000_DOWN, &adapter->flags);
-
 	/* fire a link change interrupt to start the watchdog */
 	E1000_WRITE_REG(&adapter->hw, ICS, E1000_ICS_LSC);
 	return 0;
@@ -640,15 +647,15 @@ e1000_down(struct e1000_adapter *adapter)
 	 * reschedule our watchdog timer */
 	set_bit(__E1000_DOWN, &adapter->flags);
 
+#ifdef CONFIG_E1000_NAPI
+	netif_poll_disable(netdev);
+#endif
 	e1000_irq_disable(adapter);
 
 	del_timer_sync(&adapter->tx_fifo_stall_timer);
 	del_timer_sync(&adapter->watchdog_timer);
 	del_timer_sync(&adapter->phy_info_timer);
 
-#ifdef CONFIG_E1000_NAPI
-	netif_poll_disable(netdev);
-#endif
 	netdev->tx_queue_len = adapter->tx_queue_len;
 	adapter->link_speed = 0;
 	adapter->link_duplex = 0;
@@ -1410,21 +1417,17 @@ e1000_open(struct net_device *netdev)
 		return -EBUSY;
 
 	/* allocate transmit descriptors */
-	if ((err = e1000_setup_all_tx_resources(adapter)))
+	err = e1000_setup_all_tx_resources(adapter);
+	if (err)
 		goto err_setup_tx;
 
 	/* allocate receive descriptors */
-	if ((err = e1000_setup_all_rx_resources(adapter)))
-		goto err_setup_rx;
-
-	err = e1000_request_irq(adapter);
+	err = e1000_setup_all_rx_resources(adapter);
 	if (err)
-		goto err_req_irq;
+		goto err_setup_rx;
 
 	e1000_power_up_phy(adapter);
 
-	if ((err = e1000_up(adapter)))
-		goto err_up;
 	adapter->mng_vlan_id = E1000_MNG_VLAN_NONE;
 	if ((adapter->hw.mng_cookie.status &
 			  E1000_MNG_DHCP_COOKIE_STATUS_VLAN_SUPPORT)) {
@@ -1437,12 +1440,33 @@ e1000_open(struct net_device *netdev)
 	    e1000_check_mng_mode(&adapter->hw))
 		e1000_get_hw_control(adapter);
 
+	/* before we allocate an interrupt, we must be ready to handle it.
+	 * Setting DEBUG_SHIRQ in the kernel makes it fire an interrupt
+	 * as soon as we call pci_request_irq, so we have to setup our
+	 * clean_rx handler before we do so.  */
+	e1000_configure(adapter);
+
+	err = e1000_request_irq(adapter);
+	if (err)
+		goto err_req_irq;
+
+	/* From here on the code is the same as e1000_up() */
+	clear_bit(__E1000_DOWN, &adapter->flags);
+
+#ifdef CONFIG_E1000_NAPI
+	netif_poll_enable(netdev);
+#endif
+
+	e1000_irq_enable(adapter);
+
+	/* fire a link status change interrupt to start the watchdog */
+	E1000_WRITE_REG(&adapter->hw, ICS, E1000_ICS_LSC);
+
 	return E1000_SUCCESS;
 
-err_up:
-	e1000_power_down_phy(adapter);
-	e1000_free_irq(adapter);
 err_req_irq:
+	e1000_release_hw_control(adapter);
+	e1000_power_down_phy(adapter);
 	e1000_free_all_rx_resources(adapter);
 err_setup_rx:
 	e1000_free_all_tx_resources(adapter);

[PATCH 2/3] e1000: FIX: firmware handover bits

[Auke Kok]

From: Bruce Allan <bruce.w.allan@intel.com>

Upon code inspection it was spotted that the firmware handover bit get/set
mismatched, which may have resulted in management issues on PCI-E
adapters. Setting them correctly may fix some management issues such
as arp routing etc.

Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
---

 drivers/net/e1000/e1000_main.c |   33 ++++++++++++---------------------
 1 files changed, 12 insertions(+), 21 deletions(-)

diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c
index fff3bc0..3492f0b 100644
--- a/drivers/net/e1000/e1000_main.c
+++ b/drivers/net/e1000/e1000_main.c
@@ -409,25 +409,21 @@ e1000_release_hw_control(struct e1000_adapter *adapter)
 {
 	uint32_t ctrl_ext;
 	uint32_t swsm;
-	uint32_t extcnf;
 
 	/* Let firmware taken over control of h/w */
 	switch (adapter->hw.mac_type) {
-	case e1000_82571:
-	case e1000_82572:
-	case e1000_80003es2lan:
-		ctrl_ext = E1000_READ_REG(&adapter->hw, CTRL_EXT);
-		E1000_WRITE_REG(&adapter->hw, CTRL_EXT,
-				ctrl_ext & ~E1000_CTRL_EXT_DRV_LOAD);
-		break;
 	case e1000_82573:
 		swsm = E1000_READ_REG(&adapter->hw, SWSM);
 		E1000_WRITE_REG(&adapter->hw, SWSM,
 				swsm & ~E1000_SWSM_DRV_LOAD);
+		break;
+	case e1000_82571:
+	case e1000_82572:
+	case e1000_80003es2lan:
 	case e1000_ich8lan:
-		extcnf = E1000_READ_REG(&adapter->hw, CTRL_EXT);
+		ctrl_ext = E1000_READ_REG(&adapter->hw, CTRL_EXT);
 		E1000_WRITE_REG(&adapter->hw, CTRL_EXT,
-				extcnf & ~E1000_CTRL_EXT_DRV_LOAD);
+				ctrl_ext & ~E1000_CTRL_EXT_DRV_LOAD);
 		break;
 	default:
 		break;
@@ -450,26 +446,21 @@ e1000_get_hw_control(struct e1000_adapter *adapter)
 {
 	uint32_t ctrl_ext;
 	uint32_t swsm;
-	uint32_t extcnf;
 
 	/* Let firmware know the driver has taken over */
 	switch (adapter->hw.mac_type) {
-	case e1000_82571:
-	case e1000_82572:
-	case e1000_80003es2lan:
-		ctrl_ext = E1000_READ_REG(&adapter->hw, CTRL_EXT);
-		E1000_WRITE_REG(&adapter->hw, CTRL_EXT,
-				ctrl_ext | E1000_CTRL_EXT_DRV_LOAD);
-		break;
 	case e1000_82573:
 		swsm = E1000_READ_REG(&adapter->hw, SWSM);
 		E1000_WRITE_REG(&adapter->hw, SWSM,
 				swsm | E1000_SWSM_DRV_LOAD);
 		break;
+	case e1000_82571:
+	case e1000_82572:
+	case e1000_80003es2lan:
 	case e1000_ich8lan:
-		extcnf = E1000_READ_REG(&adapter->hw, EXTCNF_CTRL);
-		E1000_WRITE_REG(&adapter->hw, EXTCNF_CTRL,
-				extcnf | E1000_EXTCNF_CTRL_SWFLAG);
+		ctrl_ext = E1000_READ_REG(&adapter->hw, CTRL_EXT);
+		E1000_WRITE_REG(&adapter->hw, CTRL_EXT,
+				ctrl_ext | E1000_CTRL_EXT_DRV_LOAD);
 		break;
 	default:
 		break;

[PATCH 3/3] e1000: FIX: Stop raw interrupts disabled nag from RT

[Auke Kok]

From: Mark Huth <mhuth@mvista.com>

Current e1000_xmit_frame spews raw interrupt disabled nag messages when
used with RT kernel patches.  This patch uses spin_trylock_irqsave,
which allows RT patches to properly manage the irq semantics.

Signed-off-by: Mark Huth <mhuth@mvista.com>
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
---

 drivers/net/e1000/e1000_main.c |    5 +----
 1 files changed, 1 insertions(+), 4 deletions(-)

diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c
index 3492f0b..7bbefca 100644
--- a/drivers/net/e1000/e1000_main.c
+++ b/drivers/net/e1000/e1000_main.c
@@ -3378,12 +3378,9 @@ e1000_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
 	    (adapter->hw.mac_type == e1000_82573))
 		e1000_transfer_dhcp_info(adapter, skb);
 
-	local_irq_save(flags);
-	if (!spin_trylock(&tx_ring->tx_lock)) {
+	if (!spin_trylock_irqsave(&tx_ring->tx_lock, flags))
 		/* Collision - tell upper layer to requeue */
-		local_irq_restore(flags);
 		return NETDEV_TX_LOCKED;
-	}
 
 	/* need: count + 2 desc gap to keep tail from touching
 	 * head, otherwise try next time */

Re: [PATCH 1/3] e1000: FIX: be ready for incoming irq at pci_request_irq

[Jeff Garzik]

Auke Kok wrote:
> From: Auke Kok <auke-jan.h.kok@intel.com>
> 
> DEBUG_SHIRQ code exposed that e1000 was not ready for incoming interrupts
> after having called pci_request_irq. This obviously requires us to finish
> our software setup which assigns the irq handler before we request the
> irq.
> 
> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
> ---
> 
>  drivers/net/e1000/e1000_main.c |   66 +++++++++++++++++++++++++++-------------
>  1 files changed, 45 insertions(+), 21 deletions(-)

All these do indeed look like fixes to me.  But they look like low 
priority fixes that would need some public testing behind them, and it's 
pretty late in the 2.6.21-rc game.

I'll merge them into an e1000-fixes branch for now (and propagates 
through #ALL to akpm's -mm).  If replies to this email indicate we 
really should push these upstream for 2.6.21-rc, it will be easy enough 
to do so via #e1000-fixes.

	Jeff

Re: [PATCH 1/3] e1000: FIX: be ready for incoming irq at pci_request_irq

[Kok, Auke]

Jeff Garzik wrote:
> Auke Kok wrote:
>> From: Auke Kok <auke-jan.h.kok@intel.com>
>>
>> DEBUG_SHIRQ code exposed that e1000 was not ready for incoming interrupts
>> after having called pci_request_irq. This obviously requires us to finish
>> our software setup which assigns the irq handler before we request the
>> irq.
>>
>> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
>> ---
>>
>>  drivers/net/e1000/e1000_main.c |   66 +++++++++++++++++++++++++++-------------
>>  1 files changed, 45 insertions(+), 21 deletions(-)
> 
> All these do indeed look like fixes to me.  But they look like low 
> priority fixes that would need some public testing behind them, and it's 
> pretty late in the 2.6.21-rc game.
> 
> I'll merge them into an e1000-fixes branch for now (and propagates 
> through #ALL to akpm's -mm).  If replies to this email indicate we 
> really should push these upstream for 2.6.21-rc, it will be easy enough 
> to do so via #e1000-fixes.

Personally, I think this is really really needed. I'm surprised that you already 
didn't push this considering Andrew pulled this into -mm immediately.

We've also been crunching this patch in our labs for a whole week now doing all 
sorts of load/unload up/down torture and it's a real improvement. Especially 
ESB2 systems were hit with this bug irregardless of the DEBUG_SHIRQ code active 
or not, so it's a real bug with real fix.

but hey, it doesn't fix a new problem :)

Auke

Re: [PATCH 1/3] e1000: FIX: be ready for incoming irq at pci_request_irq

[Jeff Garzik]

Kok, Auke wrote:
> Personally, I think this is really really needed. I'm surprised that you 
> already didn't push this considering Andrew pulled this into -mm 
> immediately.


Since it affects a fragile area of e1000 for all [e1000] users, I much 
prefer to err on the side of caution.  I have a history of pulling 
wide-impact fixes like this, late in -rc cycle, and having yet more 
breakage appear.

But if Andrew or Linus disagree, feel free to override me.  This is just 
area where I am intentionally over-cautious.

	Jeff