Re: [patch V3 09/33] genirq/msi: Add range checking to msi_insert_desc()

[Thomas Gleixner]

On Wed, Dec 14 2022 at 10:42, Niklas Schnelle wrote:
> On Tue, 2022-12-13 at 11:04 -0800, Guenter Roeck wrote:
>> This patch results in various s390 qemu test failures.
>> There is a warning backtrace
>> 
>>    12.674858] WARNING: CPU: 0 PID: 1 at kernel/irq/msi.c:167 msi_ctrl_valid+0x2a/0xb0
>> 
>> followed by
>> 
>> [   12.684333] virtio_net: probe of virtio0 failed with error -34
>> 
>> and Ethernet interfaces don't instantiate.
> As far as I'm aware so far he tracked this down to code calling
> msi_domain_get_hwsize() which in turn calls msi_get_device_domain()
> which then returns NULL leading to msi_domain_get_hwsize() returning 0.
> I think this is related to the fact that we currently don't use IRQ
> domains.

Correct and for some stupid reason I thought 0 is a good return value
here :)



diff --git a/kernel/irq/msi.c b/kernel/irq/msi.c
index bd4d4dd626b4..8fb10f216dc0 100644
--- a/kernel/irq/msi.c
+++ b/kernel/irq/msi.c
@@ -609,8 +609,8 @@ static unsigned int msi_domain_get_hwsize(struct device *dev, unsigned int domid
 		info = domain->host_data;
 		return info->hwsize;
 	}
-	/* No domain, no size... */
-	return 0;
+	/* No domain, default to MSI_MAX_INDEX */
+	return MSI_MAX_INDEX;
 }
 
 static inline void irq_chip_write_msi_msg(struct irq_data *data,