Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1260909 > unrolled thread

[v4.1.10-rt10][PATCH 0/2] PCI: dra7xx/dwc: fix "WARNING: CPU: 1 PID: 82 at kernel/irq/handle.c:150 handle_irq_event_percpu"

Back to article view | Back to linux.kernel

Contents

  [v4.1.10-rt10][PATCH 0/2] PCI: dra7xx/dwc: fix "WARNING: CPU: 1 PID: 82 at kernel/irq/handle.c:150 handle_irq_event_percpu" Grygorii Strashko <grygorii.strashko@ti.com> - 2015-11-02 20:40 +0100
    [v4.1.10-rt10][PATCH 1/2] genirq: introduce new generic_handle_irq_rt_wa() api Grygorii Strashko <grygorii.strashko@ti.com> - 2015-11-02 20:40 +0100
    [v4.1.10-rt10] [PATCH 2/2] PCI: dra7xx: fix "WARNING: CPU: 1 PID: 82 at kernel/irq/handle.c:150 handle_irq_event_percpu" Grygorii Strashko <grygorii.strashko@ti.com> - 2015-11-02 20:40 +0100
    Re: [v4.1.10-rt10][PATCH 0/2] PCI: dra7xx/dwc: fix "WARNING: CPU: 1  PID: 82 at kernel/irq/handle.c:150 handle_irq_event_percpu" Sebastian Andrzej Siewior <bigeasy@linutronix.de> - 2015-11-03 14:50 +0100
      Re: [v4.1.10-rt10][PATCH 0/2] PCI: dra7xx/dwc: fix "WARNING: CPU: 1  PID: 82 at kernel/irq/handle.c:150 handle_irq_event_percpu" Grygorii Strashko <grygorii.strashko@ti.com> - 2015-11-03 20:20 +0100
        Re: [v4.1.10-rt10][PATCH 0/2] PCI: dra7xx/dwc: fix "WARNING: CPU:  1 PID: 82 at kernel/irq/handle.c:150 handle_irq_event_percpu" Thomas Gleixner <tglx@linutronix.de> - 2015-11-03 21:10 +0100

#1260909 — [v4.1.10-rt10][PATCH 0/2] PCI: dra7xx/dwc: fix "WARNING: CPU: 1 PID: 82 at kernel/irq/handle.c:150 handle_irq_event_percpu"

Now in kernel below code pattern is used by many drivers:
static irqreturn_t driver_xx_hw_irq_handler(int irq, void *arg)
{
	<read IRQ status register>
  	<perform HW specific operations>

	for (<each set bit in IRQ status register>) {
		<get Linux IRQ number>
		generic_handle_irq(<Linux IRQ number>);
		|- handle_simple_irq()
		|-or- handle_level_irq()
		|-or- handle_edge_irq()
		   |-handle_irq_event()
		     |-handle_irq_event_percpu()
===
"WARNING: CPU: 1 PID: 82 at kernel/irq/handle.c:150 handle_irq_event_percpu+0x14c/0x174()
 irq 460 handler irq_default_primary_handler+0x0/0x14 enabled interrupts"
===
	}
}

On -RT above code will generate warnings, because driver_xx_hw_irq_handler()
will be forced threaded (by default) and, as result, generic_handle_irq()
will be called with IRQs enabled. To W/A this issue generic_handle_irq() can
be surrounded by raw_spin_lock_irqsave/irqrestore(wa_lock).

Instead of spreading this W/A directly in many drivers this series
introduces -RT specific version of generic_handle_irq() API -
generic_handle_irq_rt_wa(). This new generic_handle_irq_rt_wa() just calls
generic_handle_irq() surrounded by raw_spin_lock_irqsave/irqrestore().
If -RT is disabled It will fallback to generic_handle_irq().

And generic_handle_irq_rt_wa() is used then to W/A similar issues
in pcie-designware.c and pcie-dra7xx.c.

Grygorii Strashko (2):
  genirq: introduce new generic_handle_irq_rt_wa() api
  PCI: dra7xx: fix "WARNING: CPU: 1 PID: 82 at  kernel/irq/handle.c:150
    handle_irq_event_percpu"

 drivers/pci/host/pci-dra7xx.c      |  3 ++-
 drivers/pci/host/pcie-designware.c |  2 +-
 include/linux/irqdesc.h            |  5 +++++
 kernel/irq/irqdesc.c               | 19 +++++++++++++++++++
 4 files changed, 27 insertions(+), 2 deletions(-)

-- 
2.6.2

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1260910 — [v4.1.10-rt10][PATCH 1/2] genirq: introduce new generic_handle_irq_rt_wa() api

Now in kernel below code pattern is used by many drivers:
static irqreturn_t driver_xx_hw_irq_handler(int irq, void *arg)
{
	<read IRQ status register>
  	<perform HW specific operations>

	for (<each set bit in IRQ status register>) {
		<get Linux IRQ number>
		generic_handle_irq(<Linux IRQ number>);
		|- handle_simple_irq()
		|-or- handle_level_irq()
		|-or- handle_edge_irq()
		   |-handle_irq_event()
		     |-handle_irq_event_percpu()
===
"WARNING: CPU: 1 PID: 82 at kernel/irq/handle.c:150 handle_irq_event_percpu+0x14c/0x174()
 irq 460 handler irq_default_primary_handler+0x0/0x14 enabled interrupts"
===
	}
}

On -RT above code will generate warnings, because driver_xx_hw_irq_handler()
will be forced threaded (by default) and, as result, generic_handle_irq()
will be called with IRQs enabled. To W/A this issue generic_handle_irq() can
be surrounded by raw_spin_lock_irqsave/irqrestore(wa_lock).

Instead of spreading this W/A directly in many drivers this patch
introduces -RT specific version of generic_handle_irq() API -
generic_handle_irq_rt_wa(). This new generic_handle_irq_rt_wa() just calls
generic_handle_irq() surrounded by raw_spin_lock_irqsave/irqrestore().
If -RT is disabled It will fallback to generic_handle_irq().

Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
---
 include/linux/irqdesc.h |  5 +++++
 kernel/irq/irqdesc.c    | 19 +++++++++++++++++++
 2 files changed, 24 insertions(+)

diff --git a/include/linux/irqdesc.h b/include/linux/irqdesc.h
index 9d97cd5..beea9a5 100644
--- a/include/linux/irqdesc.h
+++ b/include/linux/irqdesc.h
@@ -131,6 +131,11 @@ static inline void generic_handle_irq_desc(unsigned int irq, struct irq_desc *de
 }
 
 int generic_handle_irq(unsigned int irq);
+#ifdef CONFIG_PREEMPT_RT_FULL
+int generic_handle_irq_rt_wa(unsigned int irq);
+#else
+#define generic_handle_irq_rt_wa generic_handle_irq
+#endif
 
 #ifdef CONFIG_HANDLE_DOMAIN_IRQ
 /*
diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c
index 99793b9..b77e01b 100644
--- a/kernel/irq/irqdesc.c
+++ b/kernel/irq/irqdesc.c
@@ -353,6 +353,25 @@ int generic_handle_irq(unsigned int irq)
 }
 EXPORT_SYMBOL_GPL(generic_handle_irq);
 
+#ifdef CONFIG_PREEMPT_RT_FULL
+int generic_handle_irq_rt_wa(unsigned int irq)
+{
+	struct irq_desc *desc = irq_to_desc(irq);
+	DEFINE_RAW_SPINLOCK(wa_lock);
+	unsigned long wa_lock_flags;
+
+	if (!desc)
+		return -EINVAL;
+
+	raw_spin_lock_irqsave(&wa_lock, wa_lock_flags);
+	generic_handle_irq_desc(irq, desc);
+	raw_spin_unlock_irqrestore(&wa_lock, wa_lock_flags);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(generic_handle_irq_rt_wa);
+#endif
+
 #ifdef CONFIG_HANDLE_DOMAIN_IRQ
 /**
  * __handle_domain_irq - Invoke the handler for a HW irq belonging to a domain
-- 
2.6.2

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1260912 — [v4.1.10-rt10] [PATCH 2/2] PCI: dra7xx: fix "WARNING: CPU: 1 PID: 82 at kernel/irq/handle.c:150 handle_irq_event_percpu"

Use generic_handle_irq_rt_wa() to W/A below backtrace, which is
triggered because dra7xx_pcie_msi_irq_handler() forced to be
threaded on -RT:

------------[ cut here ]------------
WARNING: CPU: 1 PID: 82 at kernel/irq/handle.c:150 handle_irq_event_percpu+0x14c/0x174()
irq 460 handler irq_default_primary_handler+0x0/0x14 enabled interrupts
Modules linked in: ahci_platform(+) ti_vip(+) libahci_platform xhci_pci(+) c_can_platform(+)
libahci xhci_hcd ti_vpe c_can libata v4l2_mem2mem can_dev gpio_keys leds_gpio
snd_soc_simple_card usbcore spi_ti_qspi videobuf2_core extcon_usb_gpio omap_wdt
ti_vpdma videobuf2_dma_contig ti_soc_thermal videobuf2_memops dwc3_omap extcon
rtc_omap ov1063x v4l2_common videodev media snd_soc_tlv320aic3x omap_rng rng_core omap_remoteproc
CPU: 1 PID: 82 Comm: irq/26-dra7-pci Not tainted 4.1.10-rt10-02638-g6486d7e-dirty #1
Hardware name: Generic DRA74X (Flattened Device Tree)
Backtrace:
[<c0013060>] (dump_backtrace) from [<c0013280>] (show_stack+0x18/0x1c)
 r7:c07acca0 r6:00000000 r5:c09034e4 r4:00000000
[<c0013268>] (show_stack) from [<c061ef14>] (dump_stack+0x90/0xac)
[<c061ee84>] (dump_stack) from [<c00436e4>] (warn_slowpath_common+0x7c/0xb8)
 r7:c07acca0 r6:00000096 r5:00000009 r4:ee4d3e38
[<c0043668>] (warn_slowpath_common) from [<c0043758>] (warn_slowpath_fmt+0x38/0x40)
 r8:ee3fcc00 r7:000001cc r6:00000000 r5:00000002 r4:c07acc78
[<c0043724>] (warn_slowpath_fmt) from [<c0085fd0>] (handle_irq_event_percpu+0x14c/0x174)
 r3:000001cc r2:c07acc78
 r4:ecfcdd80
[<c0085e84>] (handle_irq_event_percpu) from [<c008607c>] (handle_irq_event+0x84/0xb8)
 r10:00000001 r9:ee4b59c0 r8:ee1d0900 r7:00000001 r6:00000000 r5:ecfcdd80
 r4:ee3fcc00
[<c0085ff8>] (handle_irq_event) from [<c0089240>] (handle_simple_irq+0x90/0x118)
 r5:ee4a9020 r4:ee3fcc00
[<c00891b0>] (handle_simple_irq) from [<c0085514>] (generic_handle_irq+0x30/0x44)
 r5:ee4a9020 r4:000001cc
[<c00854e4>] (generic_handle_irq) from [<c034a7d4>] (dra7xx_pcie_msi_irq_handler+0x7c/0x8c)
 r5:ee4a9020 r4:00000001
[<c034a758>] (dra7xx_pcie_msi_irq_handler) from [<c0086f08>] (irq_forced_thread_fn+0x28/0x5c)
 r5:ee1d0900 r4:ee4b59c0
[<c0086ee0>] (irq_forced_thread_fn) from [<c00871dc>] (irq_thread+0x128/0x204)
 r7:00000001 r6:00000000 r5:ee4d2000 r4:ee4b59e4
[<c00870b4>] (irq_thread) from [<c005e2fc>] (kthread+0xd4/0xec)
 r10:00000000 r9:00000000 r8:00000000 r7:c00870b4 r6:ee4b59c0 r5:ee4b5a00
 r4:00000000
[<c005e228>] (kthread) from [<c000faa8>] (ret_from_fork+0x14/0x2c)
 r7:00000000 r6:00000000 r5:c005e228 r4:ee4b5a00
---[ end trace 0000000000000002 ]---

Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
---
 drivers/pci/host/pci-dra7xx.c      | 3 ++-
 drivers/pci/host/pcie-designware.c | 2 +-
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/pci/host/pci-dra7xx.c b/drivers/pci/host/pci-dra7xx.c
index 6589e7e..8c241cb 100644
--- a/drivers/pci/host/pci-dra7xx.c
+++ b/drivers/pci/host/pci-dra7xx.c
@@ -220,7 +220,8 @@ static irqreturn_t dra7xx_pcie_msi_irq_handler(int irq, void *arg)
 	case INTB:
 	case INTC:
 	case INTD:
-		generic_handle_irq(irq_find_mapping(pp->irq_domain, ffs(reg)));
+		generic_handle_irq_rt_wa(irq_find_mapping(pp->irq_domain,
+					 ffs(reg)));
 		break;
 	}
 
diff --git a/drivers/pci/host/pcie-designware.c b/drivers/pci/host/pcie-designware.c
index 2e9f84f..6ba0ee3 100644
--- a/drivers/pci/host/pcie-designware.c
+++ b/drivers/pci/host/pcie-designware.c
@@ -177,7 +177,7 @@ irqreturn_t dw_handle_msi_irq(struct pcie_port *pp)
 				dw_pcie_wr_own_conf(pp,
 						PCIE_MSI_INTR0_STATUS + i * 12,
 						4, 1 << pos);
-				generic_handle_irq(irq);
+				generic_handle_irq_rt_wa(irq);
 				pos++;
 			}
 		}
-- 
2.6.2

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1261504 — Re: [v4.1.10-rt10][PATCH 0/2] PCI: dra7xx/dwc: fix "WARNING: CPU: 1 PID: 82 at kernel/irq/handle.c:150 handle_irq_event_percpu"

On 11/02/2015 08:30 PM, Grygorii Strashko wrote:
> On -RT above code will generate warnings, because driver_xx_hw_irq_handler()
> will be forced threaded (by default) and, as result, generic_handle_irq()
> will be called with IRQs enabled. To W/A this issue generic_handle_irq() can
> be surrounded by raw_spin_lock_irqsave/irqrestore(wa_lock).

and what happens without -RT if you boot threadedirqs?

Sebastian
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1261805 — Re: [v4.1.10-rt10][PATCH 0/2] PCI: dra7xx/dwc: fix "WARNING: CPU: 1 PID: 82 at kernel/irq/handle.c:150 handle_irq_event_percpu"

On 11/03/2015 03:46 PM, Sebastian Andrzej Siewior wrote:
> On 11/02/2015 08:30 PM, Grygorii Strashko wrote:
>> On -RT above code will generate warnings, because driver_xx_hw_irq_handler()
>> will be forced threaded (by default) and, as result, generic_handle_irq()
>> will be called with IRQs enabled. To W/A this issue generic_handle_irq() can
>> be surrounded by raw_spin_lock_irqsave/irqrestore(wa_lock).
> 
> and what happens without -RT if you boot threadedirqs?
> 

Yep. It will still warn :( Can be reworked to cover this case also, i think
(if overall concept will be accepted) 


Thanks.
-- 
regards,
-grygorii
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1261841 — Re: [v4.1.10-rt10][PATCH 0/2] PCI: dra7xx/dwc: fix "WARNING: CPU: 1 PID: 82 at kernel/irq/handle.c:150 handle_irq_event_percpu"

On Tue, 3 Nov 2015, Grygorii Strashko wrote:
> On 11/03/2015 03:46 PM, Sebastian Andrzej Siewior wrote:
> > On 11/02/2015 08:30 PM, Grygorii Strashko wrote:
> >> On -RT above code will generate warnings, because driver_xx_hw_irq_handler()
> >> will be forced threaded (by default) and, as result, generic_handle_irq()
> >> will be called with IRQs enabled. To W/A this issue generic_handle_irq() can
> >> be surrounded by raw_spin_lock_irqsave/irqrestore(wa_lock).
> > 
> > and what happens without -RT if you boot threadedirqs?
> > 
> 
> Yep. It will still warn :( Can be reworked to cover this case also, i think
> (if overall concept will be accepted) 

wa_lock - I read that "Warning Annihilation Lock" - does not really
qualify as a concept.

Thanks,

	tglx
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web