Patchwork gm45 intel gfx can generate non-MSI irq# in MSI mode (was Re: [PATCH] drm/i915: stop using GMBUS IRQs on Gen4 chips (was Re: [3.9-rc1] irq 16: nobody cared (was [3.9-rc1] very poor interrupt respo

login
register
mail settings
Submitter Jiri Kosina
Date March 20, 2013, 4:29 p.m.
Message ID <alpine.LNX.2.00.1303201728070.17295@pobox.suse.cz>
Download mbox | patch
Permalink /patch/229429/
State Not Applicable
Headers show

Comments

Jiri Kosina - March 20, 2013, 4:29 p.m.
On Tue, 19 Mar 2013, Alan Stern wrote:

> > > That might be misleading.  It's possible that the erroneous IRQs _are_
> > > being issued but you're simply not aware of them.  If the kernel thinks
> > > that no device is using IRQ 16 then it will leave that IRQ disabled.
> > 
> > I guess I should have phrased it more precisely, but that's exactly
> > what I expect is happening on my machine: I don't have anything on
> > irq16 (i.e. in non-msi mode the gfx interrupt isn't shared) and hence
> > the irq is completely disabled. Which obviously makes it impossible
> > for me to reproduce the issue. To test that theory, is there a quick
> > way to force-enable a given interrupt, short of just hacking up a 2nd
> > dummy irq handler in my driver?
> 
> I don't know of any way.  In fact, I have been thinking of writing a 
> test driver module, with a module parameter telling it which IRQ number 
> to register for.  It seems like the sort of thing that would be useful 
> to have, from time to time.

Ok, so how about this?
Daniel, is it enough to make the problem appear on your system (by 
building this into the kernel and booting with dummy-irq.irq=16)?

Thanks.





From: Jiri Kosina <jkosina@suse.cz>
Subject: [PATCH] dummy-irq: introduce a dummy IRQ handler driver

This module accepts a single 'irq' parameter, which it should register for.

Its sole purpose is to help with debugging of IRQ sharing problems, by
force-enabling IRQ that would otherwise be disabled.

Suggested-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 drivers/misc/Kconfig     |    8 ++++++
 drivers/misc/Makefile    |    1 +
 drivers/misc/dummy-irq.c |   59 ++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 68 insertions(+), 0 deletions(-)
 create mode 100644 drivers/misc/dummy-irq.c
Alan Stern - March 20, 2013, 5:23 p.m.
On Wed, 20 Mar 2013, Jiri Kosina wrote:

> > I don't know of any way.  In fact, I have been thinking of writing a 
> > test driver module, with a module parameter telling it which IRQ number 
> > to register for.  It seems like the sort of thing that would be useful 
> > to have, from time to time.
> 
> Ok, so how about this?
> Daniel, is it enough to make the problem appear on your system (by 
> building this into the kernel and booting with dummy-irq.irq=16)?
> 
> Thanks.
> 
> 
> 
> 
> 
> From: Jiri Kosina <jkosina@suse.cz>
> Subject: [PATCH] dummy-irq: introduce a dummy IRQ handler driver
> 
> This module accepts a single 'irq' parameter, which it should register for.
> 
> Its sole purpose is to help with debugging of IRQ sharing problems, by
> force-enabling IRQ that would otherwise be disabled.
> 
> Suggested-by: Alan Stern <stern@rowland.harvard.edu>
> Signed-off-by: Jiri Kosina <jkosina@suse.cz>

This is just what I was thinking of.  Three extremely minor 
suggestions...

> +static irqreturn_t dummy_interrupt(int irq, void *dev_id)
> +{
> +	static int count = 0;
> +
> +	if (count == 0) {
> +		printk("dummy-irq: interrupt occured on IRQ %d\n", irq);

You probably should put a severity level here.  KERN_INFO?

> +		count++;
> +	}
> +
> +	return IRQ_NONE;
> +}
> +
> +static int __init dummy_irq_init(void)
> +{
> +	if (request_irq(irq, &dummy_interrupt, IRQF_SHARED, "dummy_irq", &irq)) {
> +		printk(KERN_ERR "dummy-irq: cannot register IRQ %d\n", irq);
> +		return -EIO;
> +	}
> +	printk(KERN_INFO "dummy-irq: registered for IRQ %d\n", irq);
> +	return 0;
> +}
> +
> +static void __exit dummy_irq_exit(void)
> +{
> +	printk(KERN_INFO "dummy-irq unloaded\n");
> +	free_irq(irq, &irq);
> +	return;

A return statement isn't needed here.

> +}
> +
> +module_init(dummy_irq_init);
> +module_exit(dummy_irq_exit);
> +
> +MODULE_LICENSE("GPL");
> +MODULE_AUTHOR("Jiri Kosina");
> +module_param_named(irq, irq, uint, 0444);

module_param is good enough when the parameter's name is the same as 
the variable's name.

> +MODULE_PARM_DESC(irq, "The IRQ to register for");

Alan Stern

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

Patch

diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig
index e83fdfe..db24b79 100644
--- a/drivers/misc/Kconfig
+++ b/drivers/misc/Kconfig
@@ -93,6 +93,14 @@  config ATMEL_TCB_CLKSRC_BLOCK
 	  TC can be used for other purposes, such as PWM generation and
 	  interval timing.
 
+config DUMMY_IRQ
+	tristate "Dummy IRQ handler"
+	default n
+	---help---
+	  This module accepts a single 'irq' parameter, which it should register for.
+	  Its sole purpose is to help with debugging of IRQ sharing problems, by
+	  force-enabling IRQ that would otherwise be disabled.
+
 config IBM_ASM
 	tristate "Device driver for IBM RSA service processor"
 	depends on X86 && PCI && INPUT
diff --git a/drivers/misc/Makefile b/drivers/misc/Makefile
index 35a1463..28ff261 100644
--- a/drivers/misc/Makefile
+++ b/drivers/misc/Makefile
@@ -13,6 +13,7 @@  obj-$(CONFIG_ATMEL_TCLIB)	+= atmel_tclib.o
 obj-$(CONFIG_BMP085)		+= bmp085.o
 obj-$(CONFIG_BMP085_I2C)	+= bmp085-i2c.o
 obj-$(CONFIG_BMP085_SPI)	+= bmp085-spi.o
+obj-$(CONFIG_DUMMY_IRQ)		+= dummy-irq.o
 obj-$(CONFIG_ICS932S401)	+= ics932s401.o
 obj-$(CONFIG_LKDTM)		+= lkdtm.o
 obj-$(CONFIG_TIFM_CORE)       	+= tifm_core.o
diff --git a/drivers/misc/dummy-irq.c b/drivers/misc/dummy-irq.c
new file mode 100644
index 0000000..4fc13e0
--- /dev/null
+++ b/drivers/misc/dummy-irq.c
@@ -0,0 +1,59 @@ 
+/*
+ * Dummy IRQ handler driver.
+ *
+ * This module only registers itself as a handler that is specified to it
+ * by the 'irq' parameter.
+ *
+ * The sole purpose of this module is to help with debugging of systems on
+ * which spurious IRQs cause the IRQ to be disabled.
+ *
+ * Copyright (C) 2013 Jiri Kosina
+ */
+
+/*
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/irq.h>
+#include <linux/interrupt.h>
+
+static int irq;
+
+static irqreturn_t dummy_interrupt(int irq, void *dev_id)
+{
+	static int count = 0;
+
+	if (count == 0) {
+		printk("dummy-irq: interrupt occured on IRQ %d\n", irq);
+		count++;
+	}
+
+	return IRQ_NONE;
+}
+
+static int __init dummy_irq_init(void)
+{
+	if (request_irq(irq, &dummy_interrupt, IRQF_SHARED, "dummy_irq", &irq)) {
+		printk(KERN_ERR "dummy-irq: cannot register IRQ %d\n", irq);
+		return -EIO;
+	}
+	printk(KERN_INFO "dummy-irq: registered for IRQ %d\n", irq);
+	return 0;
+}
+
+static void __exit dummy_irq_exit(void)
+{
+	printk(KERN_INFO "dummy-irq unloaded\n");
+	free_irq(irq, &irq);
+	return;
+}
+
+module_init(dummy_irq_init);
+module_exit(dummy_irq_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Jiri Kosina");
+module_param_named(irq, irq, uint, 0444);
+MODULE_PARM_DESC(irq, "The IRQ to register for");