2019-04-05 14:13:11

by Colin King

Subject: [PATCH][next] acpi/hmat: fix uninitialized pointer dereference on pointer 'target'

From: Colin Ian King <[email protected]>

The pointer 'target' is not initialized and is only assigned when the
ACPI_HMAT_MEMORY_PD_VALID bit in p->flags is set. There is a later null
check on target that leads to an uninitialized pointer read and
dereference when assigning target->processor_pxm when target contains a
non-null garbage value. Fix this by initializing targer to null.

Fixes: 665ac7e92757 ("acpi/hmat: Register processor domain to its memory")
Addresses-Coverity: ("Uninitialized pointer read")
Signed-off-by: Colin Ian King <[email protected]>
---
drivers/acpi/hmat/hmat.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/acpi/hmat/hmat.c b/drivers/acpi/hmat/hmat.c
index b7824a0309f7..b275016ff648 100644
--- a/drivers/acpi/hmat/hmat.c
+++ b/drivers/acpi/hmat/hmat.c
@@ -366,7 +366,7 @@ static int __init hmat_parse_proximity_domain(union acpi_subtable_headers *heade
const unsigned long end)
{
struct acpi_hmat_proximity_domain *p = (void *)header;
- struct memory_target *target;
+ struct memory_target *target = NULL;

if (p->header.length != sizeof(*p)) {
pr_notice("HMAT: Unexpected address range header length: %d\n",
--
2.20.1

2019-04-05 15:37:29

by Keith Busch

Subject: Re: [PATCH][next] acpi/hmat: fix uninitialized pointer dereference on pointer 'target'

On Fri, Apr 05, 2019 at 07:12:15AM -0700, Colin King wrote:
> From: Colin Ian King <[email protected]>
>
> The pointer 'target' is not initialized and is only assigned when the
> ACPI_HMAT_MEMORY_PD_VALID bit in p->flags is set. There is a later null
> check on target that leads to an uninitialized pointer read and
> dereference when assigning target->processor_pxm when target contains a
> non-null garbage value. Fix this by initializing targer to null.
>
> Fixes: 665ac7e92757 ("acpi/hmat: Register processor domain to its memory")
> Addresses-Coverity: ("Uninitialized pointer read")
> Signed-off-by: Colin Ian King <[email protected]>

I would have sworn this was fixed as it's in my tree already, but the
submitted patch sure enough doesn't have it.

I've double checked to see if there are any other discrepencies, and
there are no other differences after this.

Thanks for the fix!

Reviewed-by: Keith Busch <[email protected]>

> ---
> drivers/acpi/hmat/hmat.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/acpi/hmat/hmat.c b/drivers/acpi/hmat/hmat.c
> index b7824a0309f7..b275016ff648 100644
> --- a/drivers/acpi/hmat/hmat.c
> +++ b/drivers/acpi/hmat/hmat.c
> @@ -366,7 +366,7 @@ static int __init hmat_parse_proximity_domain(union acpi_subtable_headers *heade
> const unsigned long end)
> {
> struct acpi_hmat_proximity_domain *p = (void *)header;
> - struct memory_target *target;
> + struct memory_target *target = NULL;
>
> if (p->header.length != sizeof(*p)) {
> pr_notice("HMAT: Unexpected address range header length: %d\n",
> --

2019-04-05 16:07:37

by Mukesh Ojha

Subject: Re: [PATCH][next] acpi/hmat: fix uninitialized pointer dereference on pointer 'target'

On 4/5/2019 7:42 PM, Colin King wrote:
> From: Colin Ian King <[email protected]>
>
> The pointer 'target' is not initialized and is only assigned when the
> ACPI_HMAT_MEMORY_PD_VALID bit in p->flags is set. There is a later null
> check on target that leads to an uninitialized pointer read and
> dereference when assigning target->processor_pxm when target contains a
> non-null garbage value. Fix this by initializing targer to null.
>
> Fixes: 665ac7e92757 ("acpi/hmat: Register processor domain to its memory")
> Addresses-Coverity: ("Uninitialized pointer read")
> Signed-off-by: Colin Ian King <[email protected]>
Reviewed-by: Mukesh Ojha <[email protected]>

Cheers,
-Mukesh
> ---
> drivers/acpi/hmat/hmat.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/acpi/hmat/hmat.c b/drivers/acpi/hmat/hmat.c
> index b7824a0309f7..b275016ff648 100644
> --- a/drivers/acpi/hmat/hmat.c
> +++ b/drivers/acpi/hmat/hmat.c
> @@ -366,7 +366,7 @@ static int __init hmat_parse_proximity_domain(union acpi_subtable_headers *heade
> const unsigned long end)
> {
> struct acpi_hmat_proximity_domain *p = (void *)header;
> - struct memory_target *target;
> + struct memory_target *target = NULL;
>
> if (p->header.length != sizeof(*p)) {
> pr_notice("HMAT: Unexpected address range header length: %d\n",

2019-04-09 03:30:00

by Nathan Chancellor

Subject: Re: [PATCH][next] acpi/hmat: fix uninitialized pointer dereference on pointer 'target'

On Fri, Apr 05, 2019 at 03:12:15PM +0100, Colin King wrote:
> From: Colin Ian King <[email protected]>
>
> The pointer 'target' is not initialized and is only assigned when the
> ACPI_HMAT_MEMORY_PD_VALID bit in p->flags is set. There is a later null
> check on target that leads to an uninitialized pointer read and
> dereference when assigning target->processor_pxm when target contains a
> non-null garbage value. Fix this by initializing targer to null.
>
> Fixes: 665ac7e92757 ("acpi/hmat: Register processor domain to its memory")
> Addresses-Coverity: ("Uninitialized pointer read")
> Signed-off-by: Colin Ian King <[email protected]>

Reviewed-by: Nathan Chancellor <[email protected]>

> ---
> drivers/acpi/hmat/hmat.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/acpi/hmat/hmat.c b/drivers/acpi/hmat/hmat.c
> index b7824a0309f7..b275016ff648 100644
> --- a/drivers/acpi/hmat/hmat.c
> +++ b/drivers/acpi/hmat/hmat.c
> @@ -366,7 +366,7 @@ static int __init hmat_parse_proximity_domain(union acpi_subtable_headers *heade
> const unsigned long end)
> {
> struct acpi_hmat_proximity_domain *p = (void *)header;
> - struct memory_target *target;
> + struct memory_target *target = NULL;
>
> if (p->header.length != sizeof(*p)) {
> pr_notice("HMAT: Unexpected address range header length: %d\n",
> --
> 2.20.1
>

2019-04-09 07:52:45

by Rafael J. Wysocki

Subject: Re: [PATCH][next] acpi/hmat: fix uninitialized pointer dereference on pointer 'target'

On Fri, Apr 5, 2019 at 4:12 PM Colin King <[email protected]> wrote:
>
> From: Colin Ian King <[email protected]>
>
> The pointer 'target' is not initialized and is only assigned when the
> ACPI_HMAT_MEMORY_PD_VALID bit in p->flags is set. There is a later null
> check on target that leads to an uninitialized pointer read and
> dereference when assigning target->processor_pxm when target contains a
> non-null garbage value. Fix this by initializing targer to null.
>
> Fixes: 665ac7e92757 ("acpi/hmat: Register processor domain to its memory")
> Addresses-Coverity: ("Uninitialized pointer read")
> Signed-off-by: Colin Ian King <[email protected]>

Acked-by: Rafael J. Wysocki <[email protected]>

Greg, since the patch fixed by this one is in your tree, can you apply
this too, please?

> ---
> drivers/acpi/hmat/hmat.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/acpi/hmat/hmat.c b/drivers/acpi/hmat/hmat.c
> index b7824a0309f7..b275016ff648 100644
> --- a/drivers/acpi/hmat/hmat.c
> +++ b/drivers/acpi/hmat/hmat.c
> @@ -366,7 +366,7 @@ static int __init hmat_parse_proximity_domain(union acpi_subtable_headers *heade
> const unsigned long end)
> {
> struct acpi_hmat_proximity_domain *p = (void *)header;
> - struct memory_target *target;
> + struct memory_target *target = NULL;
>
> if (p->header.length != sizeof(*p)) {
> pr_notice("HMAT: Unexpected address range header length: %d\n",
> --
> 2.20.1
>