Not all irq chips are IO-APIC chip.
In our system, there are many demux GPIO interrupts except for the
io-apic chip interrupts, and these GPIO interrupts are belonged
to other irq chips, the chip data is not type of struct irq_cfg
either.
But in function __setup_vector_irq(), it listed all allocated irqs,
and presume all irq chip is ioapic_chip and the chip data is type
of struct irq_cfg, it possibly causes the vector_irq is corrupted
randomly.
For example, one irq 258 is not io-apic chip irq, in __setup_vector_irq(),
the chip data is forced to be used as struct irq_cfg, then the value
cfg->domain and cfg->vector are wrong to be used to write vector_irq:
vector = cfg->vector;
per_cpu(vector_irq, cpu)[vector] = irq;
This patch use the .flags to identify if the irq chip is io-apic.
Signed-off-by: liu chuansheng <[email protected]>
---
arch/x86/kernel/apic/io_apic.c | 24 ++++++++++++++++++++++++
1 files changed, 24 insertions(+), 0 deletions(-)
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 1817fa9..f9cac47 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -68,6 +68,18 @@
#define for_each_irq_pin(entry, head) \
for (entry = head; entry; entry = entry->next)
+/* need more thoughts ... */
+#define CHIP_FLAG_IOAPIC 0x1000
+static inline bool is_ioapic_irq(int irq)
+{
+ struct irq_chip *chip;
+ chip = irq_get_chip(irq);
+ if ((chip) && (chip->flags == CHIP_FLAG_IOAPIC))
+ return true;
+
+ return false;
+}
+
#ifdef CONFIG_IRQ_REMAP
static void irq_remap_modify_chip_defaults(struct irq_chip *chip);
static inline bool irq_remapped(struct irq_cfg *cfg)
@@ -1238,6 +1250,9 @@ void __setup_vector_irq(int cpu)
raw_spin_lock(&vector_lock);
/* Mark the inuse vectors */
for_each_active_irq(irq) {
+ if (!is_ioapic_irq(irq))
+ continue;
+
cfg = irq_get_chip_data(irq);
if (!cfg)
continue;
@@ -1259,6 +1274,9 @@ void __setup_vector_irq(int cpu)
if (irq < 0)
continue;
+ if (!is_ioapic_irq(irq))
+ continue;
+
cfg = irq_cfg(irq);
if (!cpumask_test_cpu(cpu, cfg->domain))
per_cpu(vector_irq, cpu)[vector] = -1;
@@ -2596,6 +2614,7 @@ static struct irq_chip ioapic_chip __read_mostly = {
.irq_eoi = ack_apic_level,
.irq_set_affinity = ioapic_set_affinity,
.irq_retrigger = ioapic_retrigger_irq,
+ .flags = CHIP_FLAG_IOAPIC,
};
static inline void init_IO_APIC_traps(void)
@@ -2661,6 +2680,7 @@ static struct irq_chip lapic_chip __read_mostly = {
.irq_mask = mask_lapic_irq,
.irq_unmask = unmask_lapic_irq,
.irq_ack = ack_lapic_irq,
+ .flags = CHIP_FLAG_IOAPIC,
};
static void lapic_register_intr(int irq)
@@ -3146,6 +3166,7 @@ static struct irq_chip msi_chip = {
.irq_ack = ack_apic_edge,
.irq_set_affinity = msi_set_affinity,
.irq_retrigger = ioapic_retrigger_irq,
+ .flags = CHIP_FLAG_IOAPIC,
};
static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq)
@@ -3260,6 +3281,7 @@ static struct irq_chip dmar_msi_type = {
.irq_ack = ack_apic_edge,
.irq_set_affinity = dmar_msi_set_affinity,
.irq_retrigger = ioapic_retrigger_irq,
+ .flags = CHIP_FLAG_IOAPIC,
};
int arch_setup_dmar_msi(unsigned int irq)
@@ -3308,6 +3330,7 @@ static struct irq_chip hpet_msi_type = {
.irq_ack = ack_apic_edge,
.irq_set_affinity = hpet_msi_set_affinity,
.irq_retrigger = ioapic_retrigger_irq,
+ .flags = CHIP_FLAG_IOAPIC,
};
int arch_setup_hpet_msi(unsigned int irq, unsigned int id)
@@ -3375,6 +3398,7 @@ static struct irq_chip ht_irq_chip = {
.irq_ack = ack_apic_edge,
.irq_set_affinity = ht_set_affinity,
.irq_retrigger = ioapic_retrigger_irq,
+ .flags = CHIP_FLAG_IOAPIC,
};
int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev)
--
1.7.0.4
On Tue, 2012-10-30 at 00:15 +0800, Chuansheng Liu wrote:
> Not all irq chips are IO-APIC chip.
>
> In our system, there are many demux GPIO interrupts except for the
> io-apic chip interrupts, and these GPIO interrupts are belonged
> to other irq chips, the chip data is not type of struct irq_cfg
> either.
>
> But in function __setup_vector_irq(), it listed all allocated irqs,
> and presume all irq chip is ioapic_chip and the chip data is type
> of struct irq_cfg, it possibly causes the vector_irq is corrupted
> randomly.
>
> For example, one irq 258 is not io-apic chip irq, in __setup_vector_irq(),
> the chip data is forced to be used as struct irq_cfg, then the value
> cfg->domain and cfg->vector are wrong to be used to write vector_irq:
> vector = cfg->vector;
> per_cpu(vector_irq, cpu)[vector] = irq;
>
> This patch use the .flags to identify if the irq chip is io-apic.
I have a feeling that your gpio driver is abusing the 'chip_data' in the
struct irq_data. Shouldn't the driver be using 'handler_data' instead?
>From include/linux/irq.h:
* @handler_data: per-IRQ data for the irq_chip methods
* @chip_data: platform-specific per-chip private data for the chip
* methods, to allow shared chip implementations
Also, how are these routed to the processors and the mechanism of the
vector assignment for these irq's? I presume irq_cfg is needed for the
setup and the interrupt migration from one cpu to another.
What am I missing?
thanks,
suresh
> -----Original Message-----
> From: Siddha, Suresh B
> Sent: Tuesday, October 30, 2012 4:24 AM
> To: Liu, Chuansheng
> Cc: [email protected]; [email protected]; [email protected];
> [email protected]; [email protected]; [email protected]
> Subject: Re: [PATCH] x86/ioapic: Fix the vector_irq[] is corrupted randomly
>
> On Tue, 2012-10-30 at 00:15 +0800, Chuansheng Liu wrote:
> > Not all irq chips are IO-APIC chip.
> >
> > In our system, there are many demux GPIO interrupts except for the
> > io-apic chip interrupts, and these GPIO interrupts are belonged
> > to other irq chips, the chip data is not type of struct irq_cfg
> > either.
> >
> > But in function __setup_vector_irq(), it listed all allocated irqs,
> > and presume all irq chip is ioapic_chip and the chip data is type
> > of struct irq_cfg, it possibly causes the vector_irq is corrupted
> > randomly.
> >
> > For example, one irq 258 is not io-apic chip irq, in __setup_vector_irq(),
> > the chip data is forced to be used as struct irq_cfg, then the value
> > cfg->domain and cfg->vector are wrong to be used to write vector_irq:
> > vector = cfg->vector;
> > per_cpu(vector_irq, cpu)[vector] = irq;
> >
> > This patch use the .flags to identify if the irq chip is io-apic.
>
> I have a feeling that your gpio driver is abusing the 'chip_data' in the
> struct irq_data. Shouldn't the driver be using 'handler_data' instead?
Not abusing.
There are many driver codes which has their own chip and chip_data.
For example,
langwell_gpio.c, the chip_data type is struct lnw_gpio;
gpio_omap.c, the chip_data type is struct gpio_bank;
In these cases, if we abused the gpio chip_data type with struct irq_cfg,
we will get very wrong cfg->vector, the value maybe 1 or 1000, anyway,
it is a random value.
>
> From include/linux/irq.h:
> * @handler_data: per-IRQ data for the irq_chip methods
> * @chip_data: platform-specific per-chip private data for the chip
> * methods, to allow shared chip implementations
>
> Also, how are these routed to the processors and the mechanism of the
> vector assignment for these irq's? I presume irq_cfg is needed for the
> setup and the interrupt migration from one cpu to another.
Normally gpio chip just has only one base irq is related with io-apic interrupt,
other allocated *VIRTUAL* irqs are based on this base irq, so no vector assign to them.
But in code __setup_vector_irq(), it list all allocated irqs which include the *VIRTUAL* irqs,
It causes this case that chip_data type is abused.
>
> What am I missing?
>
> thanks,
> suresh
????{.n?+???????+%?????ݶ??w??{.n?+????{??G?????{ay? ʇڙ?,j??f???h?????????z_??(?階?ݢj"???m??????G????????????&???~???iO???z??v?^?m???? ????????I?