On 6/17/19 4:49 PM, Jan Kiszka wrote:
> On 17.06.19 15:13, Mario Mintel wrote:
>> As it improves readability of hypervisor messages for debugging purposes.
>>
>> Signed-off-by: Mario Mintel <[email protected]>
>> ---
>>   hypervisor/arch/x86/vtd.c | 9 +++++----
>>   1 file changed, 5 insertions(+), 4 deletions(-)
>>
>> diff --git a/hypervisor/arch/x86/vtd.c b/hypervisor/arch/x86/vtd.c
>> index 1cae0dcb..8e076b4b 100644
>> --- a/hypervisor/arch/x86/vtd.c
>> +++ b/hypervisor/arch/x86/vtd.c
>> @@ -621,8 +621,9 @@ static int vtd_reserve_int_remap_region(u16
>> device_id, unsigned int length)
>>           if (start < 0)
>>               start = n;
>>           if (n + 1 == start + length) {
>> -            printk("Reserving %u interrupt(s) for device %04x "
>> -                   "at index %d\n", length, device_id, start);
>> +            printk("Reserving %u interrupt(s) for device "
>> +                   "%02x:%02x.%x at index %d\n", length,
>> +                   PCI_BDF_PARAMS(device_id), start);
>>               for (n = start; n < start + length; n++) {
>>                   int_remap_table[n].field.assigned = 1;
>>                   int_remap_table[n].field.sid = device_id;
>> @@ -639,8 +640,8 @@ static void vtd_free_int_remap_region(u16
>> device_id, unsigned int length)
>>       int pos = vtd_find_int_remap_region(device_id);
>>         if (pos >= 0) {
>> -        printk("Freeing %u interrupt(s) for device %04x at index %d\n",
>> -               length, device_id, pos);
>> +        printk("Freeing %u interrupt(s) for device %02x:%02x.%x at "
>> +               "index %d\n", length, PCI_BDF_PARAMS(device_id), pos);
>>           while (length-- > 0)
>>               vtd_update_irte(pos++, free_irte);
>>       }
>>
> 
> Not all devices listed here are PCI devices (actually only the IOAPIC,
> we do not support the HPET yet). I guess the original idea was to avoid
> confusion when a device shows up that is not visible as PCI. But
> avoiding to parse the value for all the other devices remains a value of
> its own.

Ah. This explains why we see, e.g., the 'non-existing' device f0:1f.0
for the root APIC. All other apics are 'real' PCI devices that can also
be seen in lspci.

  Ralf

> 
> Applied to next, thanks.
> 
> Jan
> 

-- 
You received this message because you are subscribed to the Google Groups 
"Jailhouse" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jailhouse-dev/53528d36-2c0d-abcc-95ad-f59fe4e4238a%40oth-regensburg.de.
For more options, visit https://groups.google.com/d/optout.

Reply via email to