Since "data" is u32, &data is a "u32 *" type, which means pointer math
will move in u32-sized steps. This was meant to be a byte offset, so
cast &data to "char *" to aim the copy into the correct location.
Seen with -Warray-bounds (and found by Coverity):
In file included from ./include/linux/string.h:269,
from ./arch/powerpc/include/asm/paca.h:15,
from ./arch/powerpc/include/asm/current.h:13,
from ./include/linux/mutex.h:14,
from ./include/linux/notifier.h:14,
from ./include/linux/clk.h:14,
from drivers/spi/spi-dw-bt1.c:12:
In function 'memcpy',
inlined from 'dw_spi_bt1_dirmap_copy_from_map' at drivers/spi/spi-dw-bt1.c:87:3:
./include/linux/fortify-string.h:20:29: warning: '__builtin_memcpy' offset 4 is out of the bounds [0, 4] of object 'data' with type 'u32' {aka 'unsigned int'} [-Warray-bounds]
20 | #define __underlying_memcpy __builtin_memcpy
| ^
./include/linux/fortify-string.h:191:9: note: in expansion of macro '__underlying_memcpy'
191 | return __underlying_memcpy(p, q, size);
| ^~~~~~~~~~~~~~~~~~~
drivers/spi/spi-dw-bt1.c: In function 'dw_spi_bt1_dirmap_copy_from_map':
drivers/spi/spi-dw-bt1.c:77:6: note: 'data' declared here
77 | u32 data;
| ^~~~
Addresses-Coverity: CID 1497771 Out-of-bounds access
Fixes: abf00907538e ("spi: dw: Add Baikal-T1 SPI Controller glue driver")
Signed-off-by: Kees Cook <[email protected]>
---
drivers/spi/spi-dw-bt1.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/spi/spi-dw-bt1.c b/drivers/spi/spi-dw-bt1.c
index 4aa8596fb1f2..5be6b7b80c21 100644
--- a/drivers/spi/spi-dw-bt1.c
+++ b/drivers/spi/spi-dw-bt1.c
@@ -84,7 +84,7 @@ static void dw_spi_bt1_dirmap_copy_from_map(void *to, void __iomem *from, size_t
if (shift) {
chunk = min_t(size_t, 4 - shift, len);
data = readl_relaxed(from - shift);
- memcpy(to, &data + shift, chunk);
+ memcpy(to, (char *)&data + shift, chunk);
from += chunk;
to += chunk;
len -= chunk;
--
2.25.1
On Thu, Feb 11, 2021 at 12:37:14PM -0800, Kees Cook wrote:
> Since "data" is u32, &data is a "u32 *" type, which means pointer math
> will move in u32-sized steps. This was meant to be a byte offset, so
> cast &data to "char *" to aim the copy into the correct location.
>
> Seen with -Warray-bounds (and found by Coverity):
>
> In file included from ./include/linux/string.h:269,
> from ./arch/powerpc/include/asm/paca.h:15,
> from ./arch/powerpc/include/asm/current.h:13,
> from ./include/linux/mutex.h:14,
> from ./include/linux/notifier.h:14,
> from ./include/linux/clk.h:14,
> from drivers/spi/spi-dw-bt1.c:12:
> In function 'memcpy',
> inlined from 'dw_spi_bt1_dirmap_copy_from_map' at drivers/spi/spi-dw-bt1.c:87:3:
> ./include/linux/fortify-string.h:20:29: warning: '__builtin_memcpy' offset 4 is out of the bounds [0, 4] of object 'data' with type 'u32' {aka 'unsigned int'} [-Warray-bounds]
> 20 | #define __underlying_memcpy __builtin_memcpy
> | ^
> ./include/linux/fortify-string.h:191:9: note: in expansion of macro '__underlying_memcpy'
> 191 | return __underlying_memcpy(p, q, size);
> | ^~~~~~~~~~~~~~~~~~~
> drivers/spi/spi-dw-bt1.c: In function 'dw_spi_bt1_dirmap_copy_from_map':
> drivers/spi/spi-dw-bt1.c:77:6: note: 'data' declared here
> 77 | u32 data;
> | ^~~~
Can't believe I missed that. Thanks!
Acked-by: Serge Semin <[email protected]>
>
> Addresses-Coverity: CID 1497771 Out-of-bounds access
> Fixes: abf00907538e ("spi: dw: Add Baikal-T1 SPI Controller glue driver")
> Signed-off-by: Kees Cook <[email protected]>
> ---
> drivers/spi/spi-dw-bt1.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/spi/spi-dw-bt1.c b/drivers/spi/spi-dw-bt1.c
> index 4aa8596fb1f2..5be6b7b80c21 100644
> --- a/drivers/spi/spi-dw-bt1.c
> +++ b/drivers/spi/spi-dw-bt1.c
> @@ -84,7 +84,7 @@ static void dw_spi_bt1_dirmap_copy_from_map(void *to, void __iomem *from, size_t
> if (shift) {
> chunk = min_t(size_t, 4 - shift, len);
> data = readl_relaxed(from - shift);
> - memcpy(to, &data + shift, chunk);
> + memcpy(to, (char *)&data + shift, chunk);
> from += chunk;
> to += chunk;
> len -= chunk;
> --
> 2.25.1
>
On Thu, 11 Feb 2021 12:37:14 -0800, Kees Cook wrote:
> Since "data" is u32, &data is a "u32 *" type, which means pointer math
> will move in u32-sized steps. This was meant to be a byte offset, so
> cast &data to "char *" to aim the copy into the correct location.
>
> Seen with -Warray-bounds (and found by Coverity):
>
> In file included from ./include/linux/string.h:269,
> from ./arch/powerpc/include/asm/paca.h:15,
> from ./arch/powerpc/include/asm/current.h:13,
> from ./include/linux/mutex.h:14,
> from ./include/linux/notifier.h:14,
> from ./include/linux/clk.h:14,
> from drivers/spi/spi-dw-bt1.c:12:
> In function 'memcpy',
> inlined from 'dw_spi_bt1_dirmap_copy_from_map' at drivers/spi/spi-dw-bt1.c:87:3:
> ./include/linux/fortify-string.h:20:29: warning: '__builtin_memcpy' offset 4 is out of the bounds [0, 4] of object 'data' with type 'u32' {aka 'unsigned int'} [-Warray-bounds]
> 20 | #define __underlying_memcpy __builtin_memcpy
> | ^
> ./include/linux/fortify-string.h:191:9: note: in expansion of macro '__underlying_memcpy'
> 191 | return __underlying_memcpy(p, q, size);
> | ^~~~~~~~~~~~~~~~~~~
> drivers/spi/spi-dw-bt1.c: In function 'dw_spi_bt1_dirmap_copy_from_map':
> drivers/spi/spi-dw-bt1.c:77:6: note: 'data' declared here
> 77 | u32 data;
> | ^~~~
Applied to
https://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git for-next
Thanks!
[1/1] spi: dw: Avoid stack content exposure
commit: 386f771aad15dd535f2368b4adc9958c0160edd4
All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.
You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.
If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.
Please add any relevant lists and maintainers to the CCs when replying
to this mail.
Thanks,
Mark