If user uses wrong ioctl command with _IOC_NONE and argument size greater than 0, it can cause NULL pointer access from memset of line 463. If _IOC_NONE, don't memset to 0 for kdata.
Signed-off-by: Zhaowei Yuan zhaowei.yuan@samsung.com --- drivers/gpu/drm/drm_drv.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) mode change 100644 => 100755 drivers/gpu/drm/drm_drv.c
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c old mode 100644 new mode 100755 index 2ab782c..1a92bcb --- a/drivers/gpu/drm/drm_drv.c +++ b/drivers/gpu/drm/drm_drv.c @@ -459,8 +459,9 @@ long drm_ioctl(struct file *filp, retcode = -EFAULT; goto err_i1; } - } else + } else if (cmd & IOC_OUT) { memset(kdata, 0, usize); + }
if (ioctl->flags & DRM_UNLOCKED) retcode = func(dev, kdata, file_priv); -- 1.7.9.5
Hi
On Wed, Jun 18, 2014 at 8:33 AM, Zhaowei Yuan zhaowei.yuan@samsung.com wrote:
If user uses wrong ioctl command with _IOC_NONE and argument size greater than 0, it can cause NULL pointer access from memset of line 463. If _IOC_NONE, don't memset to 0 for kdata.
Signed-off-by: Zhaowei Yuan zhaowei.yuan@samsung.com
Reviewed-by: David Herrmann dh.herrmann@gmail.com Cc: stable@vger.kernel.org
@Dave: Imo, this should go through stable. We initialize "kdata" to NULL, but "usize" can be up to 2^14, so it might write to real memory there. Most systems probably reserve the lower 64KB and the first write on the NULL-page should kill the process and have no other side-effects, however, who knows how memset() is implemented on different architectures. It might start writing at the end.. or might issue large writes.. what do I know..
Thanks David
drivers/gpu/drm/drm_drv.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) mode change 100644 => 100755 drivers/gpu/drm/drm_drv.c
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c old mode 100644 new mode 100755 index 2ab782c..1a92bcb --- a/drivers/gpu/drm/drm_drv.c +++ b/drivers/gpu/drm/drm_drv.c @@ -459,8 +459,9 @@ long drm_ioctl(struct file *filp, retcode = -EFAULT; goto err_i1; }
} else
} else if (cmd & IOC_OUT) { memset(kdata, 0, usize);} if (ioctl->flags & DRM_UNLOCKED) retcode = func(dev, kdata, file_priv);-- 1.7.9.5
dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
On 2014-06-18 08:33 +0200, Zhaowei Yuan wrote:
If user uses wrong ioctl command with _IOC_NONE and argument size greater than 0, it can cause NULL pointer access from memset of line 463. If _IOC_NONE, don't memset to 0 for kdata.
Signed-off-by: Zhaowei Yuan zhaowei.yuan@samsung.com
drivers/gpu/drm/drm_drv.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) mode change 100644 => 100755 drivers/gpu/drm/drm_drv.c
The mode change certainly was not intentional, was it? Just noticed it when I pulled 3.16-rc3 from Linus…
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c old mode 100644 new mode 100755 index 2ab782c..1a92bcb --- a/drivers/gpu/drm/drm_drv.c +++ b/drivers/gpu/drm/drm_drv.c @@ -459,8 +459,9 @@ long drm_ioctl(struct file *filp, retcode = -EFAULT; goto err_i1; }
} else
} else if (cmd & IOC_OUT) { memset(kdata, 0, usize);}if (ioctl->flags & DRM_UNLOCKED) retcode = func(dev, kdata, file_priv);
Cheers, Sven
dri-devel@lists.freedesktop.org
-
David Herrmann -
Sven Joachim -
Zhaowei Yuan