* [PATCH] sky2: Serialize access to PCI config space
@ 2009-08-06 11:10 Mike McCormack
2009-08-06 15:34 ` Stephen Hemminger
0 siblings, 1 reply; 5+ messages in thread
From: Mike McCormack @ 2009-08-06 11:10 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: netdev, Rene Mayrhofer, Richard Leitner
PCI config space may be shared across two device instances,
or accessed from an interrupt
, so serialize access so
read-update-write operations do not interfere with each other.
Only tested on a single port card, as my sky2 doesn't have dual ports.
Signed-off-by: Mike McCormack <mikem@ring3k.org>
---
drivers/net/sky2.c | 23 ++++++++++++++++-------
drivers/net/sky2.h | 1 +
2 files changed, 17 insertions(+), 7 deletions(-)
diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
index 1415a83..4d09632 100644
--- a/drivers/net/sky2.c
+++ b/drivers/net/sky2.c
@@ -618,6 +618,7 @@ static void sky2_phy_power_up(struct sky2_hw *hw, unsigned port)
{
u32 reg1;
+ spin_lock(&hw->pci_cfg_lock);
sky2_write8(hw, B2_TST_CTRL1, TST_CFG_WRITE_ON);
reg1 = sky2_pci_read32(hw, PCI_DEV_REG1);
reg1 &= ~phy_power[port];
@@ -628,6 +629,7 @@ static void sky2_phy_power_up(struct sky2_hw *hw, unsigned port)
sky2_pci_write32(hw, PCI_DEV_REG1, reg1);
sky2_write8(hw, B2_TST_CTRL1, TST_CFG_WRITE_OFF);
sky2_pci_read32(hw, PCI_DEV_REG1);
+ spin_unlock(&hw->pci_cfg_lock);
if (hw->chip_id == CHIP_ID_YUKON_FE)
gm_phy_write(hw, port, PHY_MARV_CTRL, PHY_CT_ANE);
@@ -681,11 +683,13 @@ static void sky2_phy_power_down(struct sky2_hw *hw, unsigned port)
gm_phy_write(hw, port, PHY_MARV_CTRL, PHY_CT_PDOWN);
}
+ spin_lock(&hw->pci_cfg_lock);
sky2_write8(hw, B2_TST_CTRL1, TST_CFG_WRITE_ON);
reg1 = sky2_pci_read32(hw, PCI_DEV_REG1);
reg1 |= phy_power[port]; /* set PHY to PowerDown/COMA Mode */
sky2_pci_write32(hw, PCI_DEV_REG1, reg1);
sky2_write8(hw, B2_TST_CTRL1, TST_CFG_WRITE_OFF);
+ spin_unlock(&hw->pci_cfg_lock);
}
/* Force a renegotiation */
@@ -2596,30 +2600,34 @@ static void sky2_hw_intr(struct sky2_hw *hw)
if (status & (Y2_IS_MST_ERR | Y2_IS_IRQ_STAT)) {
u16 pci_err;
+ spin_lock(&hw->pci_cfg_lock);
sky2_write8(hw, B2_TST_CTRL1, TST_CFG_WRITE_ON);
pci_err = sky2_pci_read16(hw, PCI_STATUS);
- if (net_ratelimit())
- dev_err(&pdev->dev, "PCI hardware error (0x%x)\n",
- pci_err);
-
sky2_pci_write16(hw, PCI_STATUS,
pci_err | PCI_STATUS_ERROR_BITS);
sky2_write8(hw, B2_TST_CTRL1, TST_CFG_WRITE_OFF);
+ spin_unlock(&hw->pci_cfg_lock);
+
+ if (net_ratelimit())
+ dev_err(&pdev->dev, "PCI hardware error (0x%x)\n",
+ pci_err);
}
if (status & Y2_IS_PCI_EXP) {
/* PCI-Express uncorrectable Error occurred */
u32 err;
+ spin_lock(&hw->pci_cfg_lock);
sky2_write8(hw, B2_TST_CTRL1, TST_CFG_WRITE_ON);
err = sky2_read32(hw, Y2_CFG_AER + PCI_ERR_UNCOR_STATUS);
sky2_write32(hw, Y2_CFG_AER + PCI_ERR_UNCOR_STATUS,
0xfffffffful);
- if (net_ratelimit())
- dev_err(&pdev->dev, "PCI Express error (0x%x)\n", err);
-
sky2_read32(hw, Y2_CFG_AER + PCI_ERR_UNCOR_STATUS);
sky2_write8(hw, B2_TST_CTRL1, TST_CFG_WRITE_OFF);
+ spin_unlock(&hw->pci_cfg_lock);
+
+ if (net_ratelimit())
+ dev_err(&pdev->dev, "PCI Express error (0x%x)\n", err);
}
if (status & Y2_HWE_L1_MASK)
@@ -4485,6 +4493,7 @@ static int __devinit sky2_probe(struct pci_dev *pdev,
}
hw->pdev = pdev;
+ spin_lock_init(&hw->pci_cfg_lock);
hw->regs = ioremap_nocache(pci_resource_start(pdev, 0), 0x4000);
if (!hw->regs) {
diff --git a/drivers/net/sky2.h b/drivers/net/sky2.h
index 4486b06..5a9b0cf 100644
--- a/drivers/net/sky2.h
+++ b/drivers/net/sky2.h
@@ -2087,6 +2087,7 @@ struct sky2_hw {
struct timer_list watchdog_timer;
struct work_struct restart_work;
wait_queue_head_t msi_wait;
+ spinlock_t pci_cfg_lock;
};
static inline int sky2_is_copper(const struct sky2_hw *hw)
--
1.5.6.5
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH] sky2: Serialize access to PCI config space
2009-08-06 11:10 [PATCH] sky2: Serialize access to PCI config space Mike McCormack
@ 2009-08-06 15:34 ` Stephen Hemminger
0 siblings, 0 replies; 5+ messages in thread
From: Stephen Hemminger @ 2009-08-06 15:34 UTC (permalink / raw)
To: Mike McCormack; +Cc: netdev, Rene Mayrhofer, Richard Leitner
On Thu, 06 Aug 2009 20:10:27 +0900
Mike McCormack <mikem@ring3k.org> wrote:
> PCI config space may be shared across two device instances,
> or accessed from an interrupt
> , so serialize access so
> read-update-write operations do not interfere with each other.
>
> Only tested on a single port card, as my sky2 doesn't have dual ports.
>
> Signed-off-by: Mike McCormack <mikem@ring3k.org>
No, more locks is not better design. Better to just use existing
RTNL mutex consistently.
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH] sky2: Serialize access to PCI config space
@ 2009-08-09 11:37 Mike McCormack
2009-08-10 15:23 ` Stephen Hemminger
2009-08-10 15:27 ` Stephen Hemminger
0 siblings, 2 replies; 5+ messages in thread
From: Mike McCormack @ 2009-08-09 11:37 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: netdev
Disable NAPI when powering up or down the phy, as sky2_err_intr
may also access the PCI config space.
Signed-off-by: Mike McCormack <mikem@ring3k.org>
---
drivers/net/sky2.c | 7 ++++++-
1 files changed, 6 insertions(+), 1 deletions(-)
diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
index 8e05d29..824a319 100644
--- a/drivers/net/sky2.c
+++ b/drivers/net/sky2.c
@@ -1448,7 +1448,9 @@ static int sky2_up(struct net_device *dev)
if (!sky2->rx_ring)
goto err_out;
+ napi_disable(&hw->napi);
sky2_mac_init(hw, port);
+ napi_enable(&hw->napi);
/* Register is number of 4K blocks on internal RAM buffer. */
ramsize = sky2_read8(hw, B2_E_0) * 4;
@@ -1880,13 +1882,16 @@ static int sky2_down(struct net_device *dev)
sky2_write32(hw, B0_IMSK, imask);
sky2_read32(hw, B0_IMSK);
+ /* after this, there should be no more interrupts for this port */
synchronize_irq(hw->pdev->irq);
- napi_synchronize(&hw->napi);
+ napi_disable(&hw->napi);
spin_lock_bh(&sky2->phy_lock);
sky2_phy_power_down(hw, port);
spin_unlock_bh(&sky2->phy_lock);
+ napi_enable(&hw->napi);
+
/* turn off LED's */
sky2_write16(hw, B0_Y2LED, LED_STAT_OFF);
--
1.5.6.5
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH] sky2: Serialize access to PCI config space
2009-08-09 11:37 Mike McCormack
@ 2009-08-10 15:23 ` Stephen Hemminger
2009-08-10 15:27 ` Stephen Hemminger
1 sibling, 0 replies; 5+ messages in thread
From: Stephen Hemminger @ 2009-08-10 15:23 UTC (permalink / raw)
To: Mike McCormack; +Cc: netdev
On Sun, 09 Aug 2009 20:37:19 +0900
Mike McCormack <mikem@ring3k.org> wrote:
>
> Disable NAPI when powering up or down the phy, as sky2_err_intr
> may also access the PCI config space.
Maybe, but any sky2_err interrupt is a serious condition that
means hardware is being used incorrectly or dead.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] sky2: Serialize access to PCI config space
2009-08-09 11:37 Mike McCormack
2009-08-10 15:23 ` Stephen Hemminger
@ 2009-08-10 15:27 ` Stephen Hemminger
1 sibling, 0 replies; 5+ messages in thread
From: Stephen Hemminger @ 2009-08-10 15:27 UTC (permalink / raw)
To: Mike McCormack; +Cc: netdev
On Sun, 09 Aug 2009 20:37:19 +0900
Mike McCormack <mikem@ring3k.org> wrote:
>
> Disable NAPI when powering up or down the phy, as sky2_err_intr
> may also access the PCI config space.
>
Also access to PCI config is done through shared memory, so each
access is atomic. The only issue is the stupid config write bit;
this bit really doesn't protect from anything useful.
Probably better to just always leave it TST_CFG_WRITE_ON.
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2009-08-10 15:27 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-08-06 11:10 [PATCH] sky2: Serialize access to PCI config space Mike McCormack
2009-08-06 15:34 ` Stephen Hemminger
-- strict thread matches above, loose matches on Subject: below --
2009-08-09 11:37 Mike McCormack
2009-08-10 15:23 ` Stephen Hemminger
2009-08-10 15:27 ` Stephen Hemminger
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).