Hi Daniel and Lukasz,
On Tue, Feb 27, 2024 at 04:37:36PM +0100, Daniel Lezcano wrote:
> On 27/02/2024 12:09, Rafael J. Wysocki wrote:
> > On Tue, Feb 27, 2024 at 11:14 AM Daniel Lezcano
> > <daniel.lezc...@linaro.org> wrote:
> > >
> > > On 27/02/2024 01:54, Nathan Chancellor wrote:
> > > > When booting a CONFIG_FORTIFY_SOURCE=y kernel compiled with a toolchain
> > > > that supports __counted_by() (such as clang-18 and newer), there is a
> > > > panic on boot:
> > > >
> > > > [ 2.913770] memcpy: detected buffer overflow: 72 byte write of
> > > > buffer size 0
> > > > [ 2.920834] WARNING: CPU: 2 PID: 1 at lib/string_helpers.c:1027
> > > > __fortify_report+0x5c/0x74
> > > > ...
> > > > [ 3.039208] Call trace:
> > > > [ 3.041643] __fortify_report+0x5c/0x74
> > > > [ 3.045469] __fortify_panic+0x18/0x20
> > > > [ 3.049209] thermal_zone_device_register_with_trips+0x4c8/0x4f8
> > > >
> > > > This panic occurs because trips is counted by num_trips but num_trips is
> > > > assigned after the call to memcpy(), so the fortify checks think the
> > > > buffer size is zero because tz was allocated with kzalloc().
> > > >
> > > > Move the num_trips assignment before the memcpy() to resolve the panic
> > > > and ensure that the fortify checks work properly.
> > > >
> > > > Fixes: 9b0a62758665 ("thermal: core: Store zone trips table in struct
> > > > thermal_zone_device")
> > > > Signed-off-by: Nathan Chancellor <nat...@kernel.org>
> > > > ---
> > > > drivers/thermal/thermal_core.c | 2 +-
> > > > 1 file changed, 1 insertion(+), 1 deletion(-)
> > > >
> > > > diff --git a/drivers/thermal/thermal_core.c
> > > > b/drivers/thermal/thermal_core.c
> > > > index bb21f78b4bfa..1eabc8ebe27d 100644
> > > > --- a/drivers/thermal/thermal_core.c
> > > > +++ b/drivers/thermal/thermal_core.c
> > > > @@ -1354,8 +1354,8 @@ thermal_zone_device_register_with_trips(const
> > > > char *type,
> > > >
> > > > tz->device.class = thermal_class;
> > > > tz->devdata = devdata;
> > > > - memcpy(tz->trips, trips, num_trips * sizeof(*trips));
> > > > tz->num_trips = num_trips;
> > > > + memcpy(tz->trips, trips, num_trips * sizeof(*trips));
> > >
> > > IIUC, clang-18 is used and supports __counted_by().
> > >
> > > Is it possible sizeof(*trips) returns already the real trips array size
> > > and we are multiplying it again by num_trips ?
> > >
> > > While with an older compiler, __counted_by() does nothing and we have to
> > > multiply by num_trips ?
> > >
> > > IOW, the array size arithmetic is different depending if we have
> > > _counted_by supported or not ?
> >
> > IIUC it is just the instrumentation using the current value of
> > tz->num_trips (which is 0 before the initialization).
>
> Right, but I am wondering if
>
> memcpy(tz->trips, trips, num_trips * sizeof(*trips));
>
> is still correct with __counted_by because:
>
> (1) if the compiler supports it:
>
> sizeof(*trips) == 24 bytes * num_trips
No, this is incorrect. sizeof(*trips) == sizeof(struct thermal_trip),
which is never affected by __counted_by. As far as I am aware,
sizeof() in general is never affected by __counted_by because calling
sizeof() on a flexible array member (which are the only things currently
allowed to have __counted_by) is invalid because a FAM is by definition
an unknown size at compile time.
> then:
>
> memcpy(tz->trips, trips, num_trips * sizeof(*trips));
>
> memcpy(tz->trips, trips, num_trips * 24 * num_trips);
>
> ==> memory size = 24 * num_trips^2
>
> (2) if the compiler does not support it:
>
> sizeof(*trips) == 24 bytes
>
> then:
>
> memcpy(tz->trips, trips, num_trips * sizeof(*trips));
>
> memcpy(tz->trips, trips, num_trips * 24);
>
> ==> memory size = 24 * num_trips
>
> Or did I misunderstand __counted_by ?
I see there has been some confusion around this change from both
yourself and Lukasz, so I will try to make things as clear and concise
as I can.
The fortify routines in the kernel rely on __builtin_dynamic_object_size
to get the size of flexible array members (see the macro soup around
memcpy() in include/linux/fortify-string.h). Before __counted_by,
__builtin_dynamic_object_size() on a FAM would return -1. With
__counted_by, __builtin_dynamic_object_size() will be able to return the
correct size of the FAM by looking at the counted by member.
In this case, memcpy() was called with tz->num_trips (the counted by
member) == 0 (meaning that __bdos() will return 0) while trying to copy
'sizeof(*trips) * num_trips' bytes into it, triggering the fortify panic
routine. Moving the tz->num_trips assignment before the memcpy() call
ensures that __bdos() returns the correct amount for checking in the
fortify routines.
Hope that helps clear things up, I will make sure to write a clearer
commit message next time.
Cheers,
Nathan
