[libvirt] [PATCH] conf: validate that PCI controller index is < 256

[Laine Stump]

On 03/27/2017 03:41 AM, Martin Kletzander wrote:
> On Sun, Mar 26, 2017 at 08:49:28PM -0400, Laine Stump wrote:
>> This is the maximum for many reasons, for starters because index ==
>> bus number, and a controller's bus number is 8 bits.
>>
>> This incidentally resolves: https://bugzilla.redhat.com/1329090
>> ---
>> src/conf/domain_conf.c | 19 +++++++++++++++++++
>> 1 file changed, 19 insertions(+)
>>
>> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c
>> index 6bbc6a2..2139ab0 100644
>> --- a/src/conf/domain_conf.c
>> +++ b/src/conf/domain_conf.c
>> @@ -4846,6 +4846,24 @@ virDomainNetDefValidate(const virDomainNetDef
>> *net)
>>
>>
>> static int
>> +virDomainControllerDefValidate(const virDomainControllerDef *controller)
>> +{
>> +    if (controller->type == VIR_DOMAIN_CONTROLLER_TYPE_PCI) {
>> +        if (controller->idx > 255) {
>> +            virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
>> +                           _("PCI controller index %d too high, or
>> too many "
> 
> The first line might be enough, maybe adding the max value there.

Okay, I adjusted it.

> 
>> +                             "PCI controllers. A maximum of 256 PCI "
>> +                             "controllers is allowed, and the maximum
>> value "
>> +                             "of the index attribute is 255"),
>> +                           controller->idx);
>> +            return -1;
>> +        }
>> +    }
>> +    return 0;
>> +}
>> +
>> +
>> +static int
>> virDomainDeviceDefValidateInternal(const virDomainDeviceDef *dev,
>>                                    const virDomainDef *def)
>> {
>> @@ -4867,6 +4885,7 @@ virDomainDeviceDefValidateInternal(const
>> virDomainDeviceDef *dev,
>>     case VIR_DOMAIN_DEVICE_HOSTDEV:
>>     case VIR_DOMAIN_DEVICE_WATCHDOG:
>>     case VIR_DOMAIN_DEVICE_CONTROLLER:
>> +        return virDomainControllerDefValidate(dev->data.controller);
> 
> Do you really mean to fall-through from hostdev, watchdog and so on? ;)

No, it was just late in the day, I was on a roll and didn't pay
attention. I moved this up above the other NOP cases. I'm just attaching
the patch here. Is that good enough for an ACK, or do you want a full
re-send?


> 
>>     case VIR_DOMAIN_DEVICE_GRAPHICS:
>>     case VIR_DOMAIN_DEVICE_HUB:
>>     case VIR_DOMAIN_DEVICE_SMARTCARD:
>> -- 
>> 2.9.3
>>
>> -- 
>> libvir-list mailing list
>> libvir-list at redhat.com
>> https://www.redhat.com/mailman/listinfo/libvir-list

-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-conf-validate-that-PCI-controller-index-is-256.patch
Type: text/x-patch
Size: 2065 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/libvir-list/attachments/20170327/e58caf44/attachment-0001.bin>