There are two bugs:
1) If ida_simple_get() fails then this code calls put_device(carrier)
but we haven't yet called get_device(carrier) and probably that
leads to a use after free.
2) After device_initialize() then we need to use put_device() to
release the bus. This will free the internal resources tied to the
device and call mcb_free_bus() which will free the rest.
Fixes: 5d9e2ab9fea4 ("mcb: Implement bus->dev.release callback")
Fixes: 18d288198099 ("mcb: Correctly initialize the bus's device")
Signed-off-by: Dan Carpenter <[email protected]>
---
drivers/mcb/mcb-core.c | 12 ++++++------
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/drivers/mcb/mcb-core.c b/drivers/mcb/mcb-core.c
index 38fbb3b59873..38cc8340e817 100644
--- a/drivers/mcb/mcb-core.c
+++ b/drivers/mcb/mcb-core.c
@@ -277,8 +277,8 @@ struct mcb_bus *mcb_alloc_bus(struct device *carrier)
bus_nr = ida_simple_get(&mcb_ida, 0, 0, GFP_KERNEL);
if (bus_nr < 0) {
- rc = bus_nr;
- goto err_free;
+ kfree(bus);
+ return ERR_PTR(bus_nr);
}
bus->bus_nr = bus_nr;
@@ -293,12 +293,12 @@ struct mcb_bus *mcb_alloc_bus(struct device *carrier)
dev_set_name(&bus->dev, "mcb:%d", bus_nr);
rc = device_add(&bus->dev);
if (rc)
- goto err_free;
+ goto err_put;
return bus;
-err_free:
- put_device(carrier);
- kfree(bus);
+
+err_put:
+ put_device(&bus->dev);
return ERR_PTR(rc);
}
EXPORT_SYMBOL_NS_GPL(mcb_alloc_bus, MCB);
--
2.30.2
Am 01.06.21 um 11:18 schrieb Dan Carpenter:
> There are two bugs:
> 1) If ida_simple_get() fails then this code calls put_device(carrier)
> but we haven't yet called get_device(carrier) and probably that
> leads to a use after free.
> 2) After device_initialize() then we need to use put_device() to
> release the bus. This will free the internal resources tied to the
> device and call mcb_free_bus() which will free the rest.
>
> Fixes: 5d9e2ab9fea4 ("mcb: Implement bus->dev.release callback")
> Fixes: 18d288198099 ("mcb: Correctly initialize the bus's device")
> Signed-off-by: Dan Carpenter <[email protected]>
>
Thanks applied
On Tue, Jun 01, 2021 at 03:23:55PM +0200, Johannes Thumshirn wrote:
>
>
> Am 01.06.21 um 11:18 schrieb Dan Carpenter:
> > There are two bugs:
> > 1) If ida_simple_get() fails then this code calls put_device(carrier)
> > but we haven't yet called get_device(carrier) and probably that
> > leads to a use after free.
> > 2) After device_initialize() then we need to use put_device() to
> > release the bus. This will free the internal resources tied to the
> > device and call mcb_free_bus() which will free the rest.
> >
> > Fixes: 5d9e2ab9fea4 ("mcb: Implement bus->dev.release callback")
> > Fixes: 18d288198099 ("mcb: Correctly initialize the bus's device")
> > Signed-off-by: Dan Carpenter <[email protected]>
> >
>
> Thanks applied
I don't think this was actually applied and that's a good thing because
my patch wasn't correct. It needs to call device_put() on both
carrier and &bus->dev.
I'm going to send a v2 fix for this.
regards,
dan carpenter
On Fri, Aug 06, 2021 at 04:32:40PM +0300, Dan Carpenter wrote:
> On Tue, Jun 01, 2021 at 03:23:55PM +0200, Johannes Thumshirn wrote:
> >
> >
> > Am 01.06.21 um 11:18 schrieb Dan Carpenter:
> > > There are two bugs:
> > > 1) If ida_simple_get() fails then this code calls put_device(carrier)
> > > but we haven't yet called get_device(carrier) and probably that
> > > leads to a use after free.
> > > 2) After device_initialize() then we need to use put_device() to
> > > release the bus. This will free the internal resources tied to the
> > > device and call mcb_free_bus() which will free the rest.
> > >
> > > Fixes: 5d9e2ab9fea4 ("mcb: Implement bus->dev.release callback")
> > > Fixes: 18d288198099 ("mcb: Correctly initialize the bus's device")
> > > Signed-off-by: Dan Carpenter <[email protected]>
> > >
> >
> > Thanks applied
>
> I don't think this was actually applied and that's a good thing because
> my patch wasn't correct. It needs to call device_put() on both
> carrier and &bus->dev.
>
> I'm going to send a v2 fix for this.
Actually, no... My first patch was correct. Calling put_device(&bus->dev);
will call the release function which will put the carrier. My first
patch was correct.
:P
regards,
dan carpenter
On Tue, Jun 01, 2021 at 03:23:55PM +0200, Johannes Thumshirn wrote:
>
>
> Am 01.06.21 um 11:18 schrieb Dan Carpenter:
> > There are two bugs:
> > 1) If ida_simple_get() fails then this code calls put_device(carrier)
> > but we haven't yet called get_device(carrier) and probably that
> > leads to a use after free.
> > 2) After device_initialize() then we need to use put_device() to
> > release the bus. This will free the internal resources tied to the
> > device and call mcb_free_bus() which will free the rest.
> >
> > Fixes: 5d9e2ab9fea4 ("mcb: Implement bus->dev.release callback")
> > Fixes: 18d288198099 ("mcb: Correctly initialize the bus's device")
> > Signed-off-by: Dan Carpenter <[email protected]>
> >
>
> Thanks applied
This is still not in linux-next.
regards,
dan carpenter
Am 30.08.21 um 16:08 schrieb Dan Carpenter:
> On Tue, Jun 01, 2021 at 03:23:55PM +0200, Johannes Thumshirn wrote:
>>
>> Am 01.06.21 um 11:18 schrieb Dan Carpenter:
>>> There are two bugs:
>>> 1) If ida_simple_get() fails then this code calls put_device(carrier)
>>> but we haven't yet called get_device(carrier) and probably that
>>> leads to a use after free.
>>> 2) After device_initialize() then we need to use put_device() to
>>> release the bus. This will free the internal resources tied to the
>>> device and call mcb_free_bus() which will free the rest.
>>>
>>> Fixes: 5d9e2ab9fea4 ("mcb: Implement bus->dev.release callback")
>>> Fixes: 18d288198099 ("mcb: Correctly initialize the bus's device")
>>> Signed-off-by: Dan Carpenter <[email protected]>
>>>
>> Thanks applied
> This is still not in linux-next.
>
> regards,
> dan carpenter
Hi Dan,
Greg asked me to resend the patch with Cc stable and I totally forgot to
do that, my bad.
I'll do that ASAP once the merge window opens again or earlier if Greg
is OK with it.
Byte,
Johannes