forcedeth: hardirq lockdep warning
PATCH 1/1 forcedeth: hardirq lockdep warning
BUG: warning at kernel/lockdep.c:1816/trace_hardirqs_on() (Not tainted) Call Trace: show_trace dump_stack trace_hardirqs_on :forcedeth:nv_nic_irq_other handle_IRQ_event __do_IRQ do_IRQ ret_from_intr
DWARF2 barf
default_idle
cpu_idle
rest_init
start_kernel
_sinittext
These 3 functions nv_nic_irq_tx(), nv_nic_irq_rx() and nv_nic_irq_other() are reachable from IRQ context and process context. Make use of the irq-save/restore spinlock variant.
(Compile tested only, since I do not have the hardware)
--- drivers/net/forcedeth.c | 31 +++++++++++++++++-------------- 1 file changed, 17 insertions(+), 14 deletions(-)
drivers/net/forcedeth.c +17 -14
@@ -2497,6 +2497,7 @@ static irqreturn_t nv_nic_irq_tx(int foou8 __iomem *base =get_hwbase (dev);u32 events;int i; +unsigned long flags;dprintk (KERN_DEBUG "%s: nv_nic_irq_tx\n" , dev->name);
@@ -2508,16 +2509,16 @@ static irqreturn_t nv_nic_irq_tx(int fooif (!(events & np->irqmask))break ; -spin_lock_irq (&np->lock); +spin_lock_irqsave (&np->lock, flags);nv_tx_done (dev); -spin_unlock_irq (&np->lock); +spin_unlock_irqrestore (&np->lock, flags);if (events & (NVREG_IRQ_TX_ERR )) {dprintk (KERN_DEBUG "%s: received irq with events 0x%x. Probably TX fail.\n" , dev->name, events); }if (i > max_interrupt_work) { -spin_lock_irq (&np->lock); +spin_lock_irqsave (&np->lock, flags);/* disable interrupts on the nic */ writel (NVREG_IRQ_TX_ALL , base + NvRegIrqMask);pci_push (base);
@@ -2527,7 +2528,7 @@ static irqreturn_t nv_nic_irq_tx(int foomod_timer (&np->nic_poll, jiffies +POLL_WAIT ); }printk (KERN_DEBUG "%s: too many iterations (%d) in nv_nic_irq_tx.\n" , dev->name, i); -spin_unlock_irq (&np->lock); +spin_unlock_irqrestore (&np->lock, flags);break ; }
@@ -2601,6 +2602,7 @@ static irqreturn_t nv_nic_irq_rx(int foou8 __iomem *base =get_hwbase (dev);u32 events;int i; +unsigned long flags;dprintk (KERN_DEBUG "%s: nv_nic_irq_rx\n" , dev->name);
@@ -2614,14 +2616,14 @@ static irqreturn_t nv_nic_irq_rx(int foonv_rx_process (dev, dev->weight);if (nv_alloc_rx (dev)) { -spin_lock_irq (&np->lock); +spin_lock_irqsave (&np->lock, flags);if (!np->in_shutdown)mod_timer (&np->oom_kick, jiffies +OOM_REFILL ); -spin_unlock_irq (&np->lock); +spin_unlock_irqrestore (&np->lock, flags); }if (i > max_interrupt_work) { -spin_lock_irq (&np->lock); +spin_lock_irqsave (&np->lock, flags);/* disable interrupts on the nic */ writel (NVREG_IRQ_RX_ALL , base + NvRegIrqMask);pci_push (base);
@@ -2631,7 +2633,7 @@ static irqreturn_t nv_nic_irq_rx(int foomod_timer (&np->nic_poll, jiffies +POLL_WAIT ); }printk (KERN_DEBUG "%s: too many iterations (%d) in nv_nic_irq_rx.\n" , dev->name, i); -spin_unlock_irq (&np->lock); +spin_unlock_irqrestore (&np->lock, flags);break ; } }
@@ -2648,6 +2650,7 @@ static irqreturn_t nv_nic_irq_other(intu8 __iomem *base =get_hwbase (dev);u32 events;int i; +unsigned long flags;dprintk (KERN_DEBUG "%s: nv_nic_irq_other\n" , dev->name);
@@ -2660,14 +2663,14 @@ static irqreturn_t nv_nic_irq_other(intbreak ;if (events &NVREG_IRQ_LINK ) { -spin_lock_irq (&np->lock); +spin_lock_irqsave (&np->lock, flags);nv_link_irq (dev); -spin_unlock_irq (&np->lock); +spin_unlock_irqrestore (&np->lock, flags); }if (np->need_linktimer &&time_after (jiffies, np->link_timeout)) { -spin_lock_irq (&np->lock); +spin_lock_irqsave (&np->lock, flags);nv_linkchange (dev); -spin_unlock_irq (&np->lock); +spin_unlock_irqrestore (&np->lock, flags); np->link_timeout = jiffies +LINK_TIMEOUT ; }if (events & (NVREG_IRQ_UNKNOWN )) {
@@ -2675,7 +2678,7 @@ static irqreturn_t nv_nic_irq_other(int dev->name, events); }if (i > max_interrupt_work) { -spin_lock_irq (&np->lock); +spin_lock_irqsave (&np->lock, flags);/* disable interrupts on the nic */ writel (NVREG_IRQ_OTHER , base + NvRegIrqMask);pci_push (base);
@@ -2685,7 +2688,7 @@ static irqreturn_t nv_nic_irq_other(intmod_timer (&np->nic_poll, jiffies +POLL_WAIT ); }printk (KERN_DEBUG "%s: too many iterations (%d) in nv_nic_irq_other.\n" , dev->name, i); -spin_unlock_irq (&np->lock); +spin_unlock_irqrestore (&np->lock, flags);break ; }
Quoted 1 line · On Tue, 19 Sep 2006 14:55:22 +0200 Peter Zijlstra <[email protected]>…
BUG: warning at kernel/lockdep.c:1816/trace_hardirqs_on() (Not tainted)
I wonder what line that was. DEBUG_LOCKS_WARN_ON(current->hardirq_context), I suppose.
Quoted 10 lines
Call Trace: show_trace dump_stack trace_hardirqs_on :forcedeth:nv_nic_irq_other handle_IRQ_event __do_IRQ do_IRQ ret_from_intrDWARF2 barf
It's good, isn't it?
Quoted 8 lines
default_idle
cpu_idle
rest_init
start_kernel
_sinittextThese 3 functions
nv_nic_irq_tx(),nv_nic_irq_rx()andnv_nic_irq_other()are reachable from IRQ context and process context.
That's "fix deadlock", not "fix lockdep warning". However it's often the case that it's not really deadlockable because (often) the card's IRQ line has been disabled. That appears to be the case in nv_do_nic_poll()'s call to nv_nic_irq_tx, for example.
Quoted 2 lines
Make use of the
irq-save/restore spinlock variant.
So this perhaps is another lockdep workaround..
Quoted 7 lines · Andrew Morton
Quoted 1 line · On Tue, 19 Sep 2006 14:55:22 +0200 Peter Zijlstra <[email protected]>…
BUG: warning at kernel/lockdep.c:1816/trace_hardirqs_on() (Not tainted)I wonder what line that was.
DEBUG_LOCKS_WARN_ON(current->hardirq_context), I suppose.
That's what it matches up to in the Fedora kernel. (We have a bunch of lockdep fixes scooped up from lkml over the last month or so, which may offset us).
Dave
Quoted 7 lines · Andrew Morton
Quoted 1 line · On Tue, 19 Sep 2006 14:55:22 +0200 Peter Zijlstra <[email protected]>…
BUG: warning at kernel/lockdep.c:1816/trace_hardirqs_on() (Not tainted)I wonder what line that was.
DEBUG_LOCKS_WARN_ON(current->hardirq_context), I suppose.
Correct indeed.
Quoted 11 lines
Quoted 10 lines
Call Trace: show_trace dump_stack trace_hardirqs_on :forcedeth:nv_nic_irq_other handle_IRQ_event __do_IRQ do_IRQ ret_from_intrDWARF2 barf
Quoted 1 line
It's good, isn't it?
Yeah, I hope we'll get it sorted out quickly....
Quoted 6 lines
Quoted 6 lines
default_idle
cpu_idle
rest_init
start_kernel
_sinittext
Quoted 7 lines
Quoted 2 lines
These 3 functions
nv_nic_irq_tx(),nv_nic_irq_rx()andnv_nic_irq_other()are reachable from IRQ context and process context.That's "fix deadlock", not "fix lockdep warning". However it's often the case that it's not really deadlockable because (often) the card's IRQ line has been disabled. That appears to be the case in nv_do_nic_poll()'s call to
nv_nic_irq_tx, for example.
Yeah, I saw some of that. But I'm not well versed in the various netdev irq receive paths nor in the driver. I couldn't find the actual maintainer in the MAINTAINERS file nor in the source, thanks for including him in the CC. Lets hope he can say if this is an actual fix or just a workaround.
Quoted 6 lines
That's "fix deadlock", not "fix lockdep warning". However it's often the case that it's not really deadlockable because (often) the card's IRQ line has been disabled. That appears to be the case in nv_do_nic_poll()'s call to
nv_nic_irq_tx, for example.
except when it's a shared irq line, then the game rules changed ;)
Jeff Garzik
Appliedapplied to #upstream-fixes