linux-kernel - Re: [PATCH v3 02/14] PCI: prevent duplicate slot names

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20081002044825.GA18063@ldl.fc.hp.com>
Date:	Wed, 1 Oct 2008 22:48:26 -0600
From:	Alex Chiang <achiang@...com>
To:	Kenji Kaneshige <kaneshige.kenji@...fujitsu.com>
Cc:	linux-pci@...r.kernel.org, linux-kernel@...r.kernel.org,
	jbarnes@...tuousgeek.org, kristen.c.accardi@...el.com,
	matthew@....cx
Subject: Re: [PATCH v3 02/14] PCI: prevent duplicate slot names

Hi Kenji-san,

* Kenji Kaneshige <kaneshige.kenji@...fujitsu.com>:
> Alex Chiang wrote:
>> * Kenji Kaneshige <kaneshige.kenji@...fujitsu.com>:
>>> Kenji Kaneshige wrote:
>>>> Hi Alex-san,
>>>>
>>>> Here is one comment, though I have not finished reviewing/testing
>>>> your patches yet (sorry for the delay).
>>>>
>>>> Alex Chiang wrote:
>>>>
>>>> (snip.)
>>>>
>>>>> diff --git a/drivers/pci/hotplug/pci_hotplug_core.c   
>>>>> b/drivers/pci/hotplug/pci_hotplug_core.c
>>>>> index 3e37d63..46802dc 100644
>>>>> --- a/drivers/pci/hotplug/pci_hotplug_core.c
>>>>> +++ b/drivers/pci/hotplug/pci_hotplug_core.c
>>>>> @@ -570,39 +570,32 @@ int pci_hp_register(struct hotplug_slot 
>>>>> *slot,  struct pci_bus *bus, int slot_nr,
>>>>>          return -EINVAL;
>>>>>      }
>>>>>  -    /* Check if we have already registered a slot with the same 
>>>>> name. */
>>>>> -    if (get_slot_from_name(name))
>>>>> -        return -EEXIST;
>>>>> -
>>>>>      /*
>>>>> -     * No problems if we call this interface from both ACPI_PCI_SLOT
>>>>> -     * driver and call it here again. If we've already created the
>>>>> -     * pci_slot, the interface will simply bump the refcount.
>>>>> +     * Look for existing slot. If we find it, and it was created by a
>>>>> +     * slot detection driver (ie, doesn't have a ->hotplug()) then we
>>>>> +     * allow the hotplug driver calling us to rename the slot if 
>>>>>  desired.
>>>>> +     *
>>>>> +     * Otherwise, create the slot and carry on with life.
>>>>>       */
>>>>> -    pci_slot = pci_create_slot(bus, slot_nr, name);
>>>>> -    if (IS_ERR(pci_slot))
>>>>> -        return PTR_ERR(pci_slot);
>>>>> -
>>>>> -    if (pci_slot->hotplug) {
>>>>> -        dbg("%s: already claimed\n", __func__);
>>>>> -        pci_destroy_slot(pci_slot);
>>>>> -        return -EBUSY;
>>>>> +    pci_slot = pci_get_pci_slot(bus, slot_nr);
>>>> The pci_get_pci_slot() function refers pci_bus->slots list, so it
>>>> should be called with pci_bus_sem semaphore held as pci_create_slot()
>>>> does, or pci_bus_sem semaphore should be held by pci_get_pci_slot()
>>>> itself.
>>
>> Yes, I've changed pci_get_pci_slot() to acquire the pci_bus_sem
>> semaphore.
>>
>> Thank you for pointing this out.
>>
>> It will be fixed in v4 of this patch series, which I will send
>> out after I receive the rest of your review comments.
>>
>
> I noticed that changing pci_get_pci_slot() to acquire the pci_bus_sem
> might be not enough. If slot was created between pci_get_pci_slot() and
> pci_create_slot() by another thread in the following code, something
> wrong would happen I think.
>
>        pci_slot = pci_get_pci_slot(bus, slot_nr);
>        if (pci_slot) {
>                if (pci_slot->hotplug) {
>                        result = -EBUSY;
>                        goto err;
>                }
>
>                if (strcmp(kobject_name(&pci_slot->kobj), name))
>                        if ((result = pci_rename_slot(pci_slot, name)))
>                                goto err;
>        } else {
>                pci_slot = pci_create_slot(bus, slot_nr, name);
>                if ((result = IS_ERR(pci_slot)))
>                        goto out;
>        }

I'm sorry, I don't think I see the problem you are pointing out.

If pci_get_pci_slot() finds a pci_slot, we do not modify
pci_bus->slots any further, so even if a new slot is created, it
shouldn't affect the pci_slot that we already found.

I must be missing something, but I don't know what. Would you
mind explaining what you had in mind with your comment?

> I've finished reviewing and testing your patches. The rest of your
> patch looks good to me. Of course, we must not forget the comment
> from Taku Izumi.

Yes, I've modified my patch series to take into account the bugs
that Taku-san found.

Thank you both for your reviews.

/ac

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/