[v3] habanalabs: fix potential spectre v1 gadgets

[PATCHv3] habanalabs: fix potential spectre v1 gadgets

Posted by Jordy Zomer 4 years, 4 months ago

It appears like nr could be a Spectre v1 gadget as it's supplied by a
user and used as an array index. Prevent the contents of kernel memory
being leaked  to userspace via speculative execution by using
array_index_nospec.

Signed-off-by: Jordy Zomer <jordy@pwning.systems>

---
Changes v1 -> v2: Added the correct offsets
Changes v2 -> v3: Fixed line-wrapping
---
 drivers/misc/habanalabs/common/habanalabs_ioctl.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/misc/habanalabs/common/habanalabs_ioctl.c b/drivers/misc/habanalabs/common/habanalabs_ioctl.c
index 3ba3a8ffda3e..c1cdf712a10d 100644
--- a/drivers/misc/habanalabs/common/habanalabs_ioctl.c
+++ b/drivers/misc/habanalabs/common/habanalabs_ioctl.c
@@ -14,6 +14,7 @@
 #include <linux/fs.h>
 #include <linux/uaccess.h>
 #include <linux/slab.h>
+#include <linux/nospec.h>
 
 static u32 hl_debug_struct_size[HL_DEBUG_OP_TIMESTAMP + 1] = {
 	[HL_DEBUG_OP_ETR] = sizeof(struct hl_debug_params_etr),
@@ -849,6 +850,7 @@ long hl_ioctl(struct file *filep, unsigned int cmd, unsigned long arg)
 	}
 
 	if ((nr >= HL_COMMAND_START) && (nr < HL_COMMAND_END)) {
+		nr = array_index_nospec(nr, HL_COMMAND_END);
 		ioctl = &hl_ioctls[nr];
 	} else {
 		dev_err(hdev->dev, "invalid ioctl: pid=%d, nr=0x%02x\n",
@@ -872,6 +874,7 @@ long hl_ioctl_control(struct file *filep, unsigned int cmd, unsigned long arg)
 	}
 
 	if (nr == _IOC_NR(HL_IOCTL_INFO)) {
+		nr = array_index_nospec(nr, _IOC_NR(HL_IOCTL_INFO)+1);
 		ioctl = &hl_ioctls_control[nr];
 	} else {
 		dev_err(hdev->dev_ctrl, "invalid ioctl: pid=%d, nr=0x%02x\n",
-- 
2.27.0

Re: [PATCHv3] habanalabs: fix potential spectre v1 gadgets

Posted by Jordy Zomer 4 years, 4 months ago

Sorry! Removed the message and hope the line-wrapping is correct now :)

> On 2 Feb 2022, at 20:11, Jordy Zomer <jordy@pwning.systems> wrote:
> 
> It appears like nr could be a Spectre v1 gadget as it's supplied by a
> user and used as an array index. Prevent the contents of kernel memory
> being leaked  to userspace via speculative execution by using
> array_index_nospec.
> 
> Signed-off-by: Jordy Zomer <jordy@pwning.systems>
> 
> ---
> Changes v1 -> v2: Added the correct offsets
> Changes v2 -> v3: Fixed line-wrapping
> ---
> drivers/misc/habanalabs/common/habanalabs_ioctl.c | 3 +++
> 1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/misc/habanalabs/common/habanalabs_ioctl.c b/drivers/misc/habanalabs/common/habanalabs_ioctl.c
> index 3ba3a8ffda3e..c1cdf712a10d 100644
> --- a/drivers/misc/habanalabs/common/habanalabs_ioctl.c
> +++ b/drivers/misc/habanalabs/common/habanalabs_ioctl.c
> @@ -14,6 +14,7 @@
> #include <linux/fs.h>
> #include <linux/uaccess.h>
> #include <linux/slab.h>
> +#include <linux/nospec.h>
> 
> static u32 hl_debug_struct_size[HL_DEBUG_OP_TIMESTAMP + 1] = {
> 	[HL_DEBUG_OP_ETR] = sizeof(struct hl_debug_params_etr),
> @@ -849,6 +850,7 @@ long hl_ioctl(struct file *filep, unsigned int cmd, unsigned long arg)
> 	}
> 
> 	if ((nr >= HL_COMMAND_START) && (nr < HL_COMMAND_END)) {
> +		nr = array_index_nospec(nr, HL_COMMAND_END);
> 		ioctl = &hl_ioctls[nr];
> 	} else {
> 		dev_err(hdev->dev, "invalid ioctl: pid=%d, nr=0x%02x\n",
> @@ -872,6 +874,7 @@ long hl_ioctl_control(struct file *filep, unsigned int cmd, unsigned long arg)
> 	}
> 
> 	if (nr == _IOC_NR(HL_IOCTL_INFO)) {
> +		nr = array_index_nospec(nr, _IOC_NR(HL_IOCTL_INFO)+1);
> 		ioctl = &hl_ioctls_control[nr];
> 	} else {
> 		dev_err(hdev->dev_ctrl, "invalid ioctl: pid=%d, nr=0x%02x\n",
> -- 
> 2.27.0
>

Re: [PATCHv3] habanalabs: fix potential spectre v1 gadgets

Posted by Andrew Cooper 4 years, 4 months ago

On 02/02/2022 19:11, Jordy Zomer wrote:
> diff --git a/drivers/misc/habanalabs/common/habanalabs_ioctl.c b/drivers/misc/habanalabs/common/habanalabs_ioctl.c
> index 3ba3a8ffda3e..c1cdf712a10d 100644
> --- a/drivers/misc/habanalabs/common/habanalabs_ioctl.c
> +++ b/drivers/misc/habanalabs/common/habanalabs_ioctl.c
> @@ -849,6 +850,7 @@ long hl_ioctl(struct file *filep, unsigned int cmd, unsigned long arg)
>  	}
>  
>  	if ((nr >= HL_COMMAND_START) && (nr < HL_COMMAND_END)) {
> +		nr = array_index_nospec(nr, HL_COMMAND_END);
>  		ioctl = &hl_ioctls[nr];
>  	} else {
>  		dev_err(hdev->dev, "invalid ioctl: pid=%d, nr=0x%02x\n",
> @@ -872,6 +874,7 @@ long hl_ioctl_control(struct file *filep, unsigned int cmd, unsigned long arg)
>  	}
>  
>  	if (nr == _IOC_NR(HL_IOCTL_INFO)) {
> +		nr = array_index_nospec(nr, _IOC_NR(HL_IOCTL_INFO)+1);
>  		ioctl = &hl_ioctls_control[nr];
>  	} else {
>  		dev_err(hdev->dev_ctrl, "invalid ioctl: pid=%d, nr=0x%02x\n",

Sorry for jumping in the middle here, but array_index_nospec() is
finicky to use, and both examples here don't quite behave as intended.

(In reverse order), first, any competent compiler will turn:

    if (nr == _IOC_NR(HL_IOCTL_INFO)) {
        ioctl = &hl_ioctls_control[nr];

into

        ioctl = &hl_ioctls_control[_IOC_NR(HL_IOCTL_INFO)];

which can be evaluated at compile time without reference to nr.  By
using array_index_nospec(), you inhibit this optimisation, forcing the
load to be dependent on nr, and with the extra logic to try and make nr
be safe.

Second, array_index_nospec(nr, HL_COMMAND_END); only protects the upper
bound.  In this case, the valid lower bound is not zero.

i.e. while you've prevented an OoB read of the hl_ioctls[] array as a
whole, there is still a Spectre-v1 gadget where you can speculatively
continue using &hl_ioctls[0] which is out of bounds from the logic's
point of view.

One fix is to use:

    nr = HL_COMMAND_START +
        array_index_nospec(nr, HL_COMMAND_END - HL_COMMAND_START);

while another is to confirm that the subsequent logic is safe to
confused-deputy problems.  This analysis probably needs doing anyway as
there are multiple ioctl implementations that can be selected.

~Andrew