[PATCH] qe_ic: fix a buffer overflow error and add check elsewhere

127 is the theoretical up boundary of QEIC number,
in fact there only be 44 qe_ic_info now.
add check to overflow for qe_ic_info

Signed-off-by: Zhao Qiang <qiang.zhao@nxp.com>
---
 drivers/soc/fsl/qe/qe_ic.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/soc/fsl/qe/qe_ic.c b/drivers/soc/fsl/qe/qe_ic.c
index 5419527..90c00b7 100644
--- a/drivers/soc/fsl/qe/qe_ic.c
+++ b/drivers/soc/fsl/qe/qe_ic.c

@@ -261,6 +261,11 @@ static int qe_ic_host_map(struct irq_domain *h, unsigned int virq,
 	struct qe_ic *qe_ic = h->host_data;
 	struct irq_chip *chip;
 
+	if (hw >= ARRAY_SIZE(qe_ic_info)) {
+		pr_err("%s: Invalid hw irq number for QEIC\n", __func__);
+		return -EINVAL;
+	}
+
 	if (qe_ic_info[hw].mask == 0) {
 		printk(KERN_ERR "Can't map reserved IRQ\n");
 		return -EINVAL;

@@ -409,7 +414,8 @@ int qe_ic_set_priority(unsigned int virq, unsigned int priority)
 
 	if (priority > 8 || priority == 0)
 		return -EINVAL;
-	if (src > 127)
+	if (WARN_ONCE(src >= ARRAY_SIZE(qe_ic_info),
+		      "%s: Invalid hw irq number for QEIC\n", __func__))
 		return -EINVAL;
 	if (qe_ic_info[src].pri_reg == 0)
 		return -EINVAL;

@@ -438,6 +444,9 @@ int qe_ic_set_high_priority(unsigned int virq, unsigned int priority, int high)
 
 	if (priority > 2 || priority == 0)
 		return -EINVAL;
+	if (WARN_ONCE(src >= ARRAY_SIZE(qe_ic_info),
+		      "%s: Invalid hw irq number for QEIC\n", __func__))
+		return -EINVAL;
 
 	switch (qe_ic_info[src].pri_reg) {
 	case QEIC_CIPZCC:

-- 
2.1.0.27.g96db324

On Thu, Jan 21, 2016 at 9:06 AM, Zhao Qiang [off-list ref] wrote:

127 is the theoretical up boundary of QEIC number,
in fact there only be 44 qe_ic_info now.
add check to overflow for qe_ic_info

Signed-off-by: Zhao Qiang <qiang.zhao@nxp.com>

Acked-by: Li Yang <redacted>

Regards,
Leo

On Thu, 21 Jan 2016, Zhao Qiang wrote:

127 is the theoretical up boundary of QEIC number,
in fact there only be 44 qe_ic_info now.
add check to overflow for qe_ic_info

How do you trigger that overflow? The above does not explain WHY we need these
checks.

Sigh. Another dump ground for SOC stuff? irq chip drivers belong into
drivers/irqchip.

Thanks,

	tglx

On Tue, 2016-01-26 at 18:31 +0100, Thomas Gleixner wrote:

On Thu, 21 Jan 2016, Zhao Qiang wrote:

quoted

127 is the theoretical up boundary of QEIC number,
in fact there only be 44 qe_ic_info now.
add check to overflow for qe_ic_info

How do you trigger that overflow? The above does not explain WHY we need
these
checks.

The check in qe_ic_host_map can be triggered by bad data in a device tree.

The set_priority functions do not appear to be used at all.

quoted

diff --git a/drivers/soc/fsl/qe/qe_ic.c b/drivers/soc/fsl/qe/qe_ic.c
index 5419527..90c00b7 100644
--- a/drivers/soc/fsl/qe/qe_ic.c
+++ b/drivers/soc/fsl/qe/qe_ic.c

Sigh. Another dump ground for SOC stuff?

Another?  Where are the others, besides arch?

 irq chip drivers belong into drivers/irqchip.

Yes.  This stuff was recently moved out of arch/powerpc to work toward being
able to use it on ARM.  I'm expecting followup patches to move things like
this that belong elsewhere.

-Scott