Re: [PATCH] PCI, x86: clear initial value for root info resources

[Bjorn Helgaas]

On Tue, Sep 18, 2012 at 5:49 PM, Yinghai Lu <yinghai@xxxxxxxxxx> wrote:
> On Tue, Sep 18, 2012 at 3:46 PM, Bjorn Helgaas <bhelgaas@xxxxxxxxxx> wrote:
>> On Fri, Sep 14, 2012 at 6:48 PM, Yinghai Lu <yinghai@xxxxxxxxxx> wrote:
>>> Found one system one root bus hot remove get panic.
>>> Panic happens when try to release hostbridge resource.
>>>
>>> It turns out that resource get reject during put into resource tree
>>> because of conflicts.
>>> Also that resource parent pointer have random value.
>>>
>>> That invalid value cause it pass through check __release_pci_root_info
>>> and panic in release_resource.
>>>
>>> Try to use kzalloc instead.
>>
>> Don't we need the same fix for ia64 in pci_acpi_scan_root()?  Here's
>> what it does:
>>
>>         if (windows) {
>>                 controller->window =
>>                         kmalloc_node(sizeof(*controller->window) * windows,
>>                                      GFP_KERNEL, controller->node);
>>
>
>
> yes, but they don't support pci_set_host_bridge_release yet. so they
> should not meet this problem yet.

Why should we wait to fix that bug until later?  I'm not interested in
debugging this in the future.  If you fix a bug, you should always try
to fix other occurrences of the same bug at the same time.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/