* [PATCH] drm/sti: Fix return type of sti_{dvo, hda, hdmi}_connector_mode_valid()
@ 2022-11-02 15:56 ` Nathan Chancellor
0 siblings, 0 replies; 10+ messages in thread
From: Nathan Chancellor @ 2022-11-02 15:56 UTC (permalink / raw)
To: Alain Volmat
Cc: Kees Cook, Tom Rix, llvm, Nick Desaulniers, linux-kernel,
dri-devel, patches, Nathan Chancellor, Sami Tolvanen
With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
indirect call targets are validated against the expected function
pointer prototype to make sure the call target is valid to help mitigate
ROP attacks. If they are not identical, there is a failure at run time,
which manifests as either a kernel panic or thread getting killed. A
proposed warning in clang aims to catch these at compile time, which
reveals:
drivers/gpu/drm/sti/sti_hda.c:637:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
.mode_valid = sti_hda_connector_mode_valid,
^~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/sti/sti_dvo.c:376:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
.mode_valid = sti_dvo_connector_mode_valid,
^~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/sti/sti_hdmi.c:1035:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
.mode_valid = sti_hdmi_connector_mode_valid,
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
->mode_valid() in 'struct drm_connector_helper_funcs' expects a return
type of 'enum drm_mode_status', not 'int'. Adjust the return type of
sti_{dvo,hda,hdmi}_connector_mode_valid() to match the prototype's to
resolve the warning and CFI failure.
Link: https://github.com/ClangBuiltLinux/linux/issues/1750
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
---
drivers/gpu/drm/sti/sti_dvo.c | 5 +++--
drivers/gpu/drm/sti/sti_hda.c | 5 +++--
drivers/gpu/drm/sti/sti_hdmi.c | 5 +++--
3 files changed, 9 insertions(+), 6 deletions(-)
diff --git a/drivers/gpu/drm/sti/sti_dvo.c b/drivers/gpu/drm/sti/sti_dvo.c
index b6ee8a82e656..076d5f30a09c 100644
--- a/drivers/gpu/drm/sti/sti_dvo.c
+++ b/drivers/gpu/drm/sti/sti_dvo.c
@@ -346,8 +346,9 @@ static int sti_dvo_connector_get_modes(struct drm_connector *connector)
#define CLK_TOLERANCE_HZ 50
-static int sti_dvo_connector_mode_valid(struct drm_connector *connector,
- struct drm_display_mode *mode)
+static enum drm_mode_status
+sti_dvo_connector_mode_valid(struct drm_connector *connector,
+ struct drm_display_mode *mode)
{
int target = mode->clock * 1000;
int target_min = target - CLK_TOLERANCE_HZ;
diff --git a/drivers/gpu/drm/sti/sti_hda.c b/drivers/gpu/drm/sti/sti_hda.c
index 03cc401ed593..a53b5a15c2a9 100644
--- a/drivers/gpu/drm/sti/sti_hda.c
+++ b/drivers/gpu/drm/sti/sti_hda.c
@@ -601,8 +601,9 @@ static int sti_hda_connector_get_modes(struct drm_connector *connector)
#define CLK_TOLERANCE_HZ 50
-static int sti_hda_connector_mode_valid(struct drm_connector *connector,
- struct drm_display_mode *mode)
+static enum drm_mode_status
+sti_hda_connector_mode_valid(struct drm_connector *connector,
+ struct drm_display_mode *mode)
{
int target = mode->clock * 1000;
int target_min = target - CLK_TOLERANCE_HZ;
diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c
index cb82622877d2..09e0cadb6368 100644
--- a/drivers/gpu/drm/sti/sti_hdmi.c
+++ b/drivers/gpu/drm/sti/sti_hdmi.c
@@ -1004,8 +1004,9 @@ static int sti_hdmi_connector_get_modes(struct drm_connector *connector)
#define CLK_TOLERANCE_HZ 50
-static int sti_hdmi_connector_mode_valid(struct drm_connector *connector,
- struct drm_display_mode *mode)
+static enum drm_mode_status
+sti_hdmi_connector_mode_valid(struct drm_connector *connector,
+ struct drm_display_mode *mode)
{
int target = mode->clock * 1000;
int target_min = target - CLK_TOLERANCE_HZ;
base-commit: 9abf2313adc1ca1b6180c508c25f22f9395cc780
--
2.38.1
^ permalink raw reply related [flat|nested] 10+ messages in thread* Re: [PATCH] drm/sti: Fix return type of sti_{dvo,hda,hdmi}_connector_mode_valid()
2022-11-02 15:56 ` [PATCH] drm/sti: Fix return type of sti_{dvo, hda, hdmi}_connector_mode_valid() Nathan Chancellor
@ 2022-11-02 19:06 ` Kees Cook
-1 siblings, 0 replies; 10+ messages in thread
From: Kees Cook @ 2022-11-02 19:06 UTC (permalink / raw)
To: Nathan Chancellor
Cc: Tom Rix, llvm, Nick Desaulniers, linux-kernel, dri-devel,
patches, Sami Tolvanen, Alain Volmat
On Wed, Nov 02, 2022 at 08:56:23AM -0700, Nathan Chancellor wrote:
> With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
> indirect call targets are validated against the expected function
> pointer prototype to make sure the call target is valid to help mitigate
> ROP attacks. If they are not identical, there is a failure at run time,
> which manifests as either a kernel panic or thread getting killed. A
> proposed warning in clang aims to catch these at compile time, which
> reveals:
>
> drivers/gpu/drm/sti/sti_hda.c:637:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> .mode_valid = sti_hda_connector_mode_valid,
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
> drivers/gpu/drm/sti/sti_dvo.c:376:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> .mode_valid = sti_dvo_connector_mode_valid,
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
> drivers/gpu/drm/sti/sti_hdmi.c:1035:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> .mode_valid = sti_hdmi_connector_mode_valid,
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> ->mode_valid() in 'struct drm_connector_helper_funcs' expects a return
> type of 'enum drm_mode_status', not 'int'. Adjust the return type of
> sti_{dvo,hda,hdmi}_connector_mode_valid() to match the prototype's to
> resolve the warning and CFI failure.
>
> Link: https://github.com/ClangBuiltLinux/linux/issues/1750
> Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
--
Kees Cook
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH] drm/sti: Fix return type of sti_{dvo,hda,hdmi}_connector_mode_valid()
@ 2022-11-02 19:06 ` Kees Cook
0 siblings, 0 replies; 10+ messages in thread
From: Kees Cook @ 2022-11-02 19:06 UTC (permalink / raw)
To: Nathan Chancellor
Cc: Alain Volmat, dri-devel, Nick Desaulniers, Tom Rix,
Sami Tolvanen, llvm, linux-kernel, patches
On Wed, Nov 02, 2022 at 08:56:23AM -0700, Nathan Chancellor wrote:
> With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
> indirect call targets are validated against the expected function
> pointer prototype to make sure the call target is valid to help mitigate
> ROP attacks. If they are not identical, there is a failure at run time,
> which manifests as either a kernel panic or thread getting killed. A
> proposed warning in clang aims to catch these at compile time, which
> reveals:
>
> drivers/gpu/drm/sti/sti_hda.c:637:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> .mode_valid = sti_hda_connector_mode_valid,
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
> drivers/gpu/drm/sti/sti_dvo.c:376:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> .mode_valid = sti_dvo_connector_mode_valid,
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
> drivers/gpu/drm/sti/sti_hdmi.c:1035:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> .mode_valid = sti_hdmi_connector_mode_valid,
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> ->mode_valid() in 'struct drm_connector_helper_funcs' expects a return
> type of 'enum drm_mode_status', not 'int'. Adjust the return type of
> sti_{dvo,hda,hdmi}_connector_mode_valid() to match the prototype's to
> resolve the warning and CFI failure.
>
> Link: https://github.com/ClangBuiltLinux/linux/issues/1750
> Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
--
Kees Cook
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH] drm/sti: Fix return type of sti_{dvo,hda,hdmi}_connector_mode_valid()
2022-11-02 15:56 ` [PATCH] drm/sti: Fix return type of sti_{dvo, hda, hdmi}_connector_mode_valid() Nathan Chancellor
@ 2022-11-14 22:03 ` Nathan Chancellor
-1 siblings, 0 replies; 10+ messages in thread
From: Nathan Chancellor @ 2022-11-14 22:03 UTC (permalink / raw)
To: Alain Volmat
Cc: dri-devel, Nick Desaulniers, Tom Rix, Kees Cook, Sami Tolvanen,
llvm, linux-kernel, patches
Hi all,
On Wed, Nov 02, 2022 at 08:56:23AM -0700, Nathan Chancellor wrote:
> With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
> indirect call targets are validated against the expected function
> pointer prototype to make sure the call target is valid to help mitigate
> ROP attacks. If they are not identical, there is a failure at run time,
> which manifests as either a kernel panic or thread getting killed. A
> proposed warning in clang aims to catch these at compile time, which
> reveals:
>
> drivers/gpu/drm/sti/sti_hda.c:637:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> .mode_valid = sti_hda_connector_mode_valid,
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
> drivers/gpu/drm/sti/sti_dvo.c:376:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> .mode_valid = sti_dvo_connector_mode_valid,
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
> drivers/gpu/drm/sti/sti_hdmi.c:1035:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> .mode_valid = sti_hdmi_connector_mode_valid,
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> ->mode_valid() in 'struct drm_connector_helper_funcs' expects a return
> type of 'enum drm_mode_status', not 'int'. Adjust the return type of
> sti_{dvo,hda,hdmi}_connector_mode_valid() to match the prototype's to
> resolve the warning and CFI failure.
>
> Link: https://github.com/ClangBuiltLinux/linux/issues/1750
> Signed-off-by: Nathan Chancellor <nathan@kernel.org>
> ---
> drivers/gpu/drm/sti/sti_dvo.c | 5 +++--
> drivers/gpu/drm/sti/sti_hda.c | 5 +++--
> drivers/gpu/drm/sti/sti_hdmi.c | 5 +++--
> 3 files changed, 9 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/gpu/drm/sti/sti_dvo.c b/drivers/gpu/drm/sti/sti_dvo.c
> index b6ee8a82e656..076d5f30a09c 100644
> --- a/drivers/gpu/drm/sti/sti_dvo.c
> +++ b/drivers/gpu/drm/sti/sti_dvo.c
> @@ -346,8 +346,9 @@ static int sti_dvo_connector_get_modes(struct drm_connector *connector)
>
> #define CLK_TOLERANCE_HZ 50
>
> -static int sti_dvo_connector_mode_valid(struct drm_connector *connector,
> - struct drm_display_mode *mode)
> +static enum drm_mode_status
> +sti_dvo_connector_mode_valid(struct drm_connector *connector,
> + struct drm_display_mode *mode)
> {
> int target = mode->clock * 1000;
> int target_min = target - CLK_TOLERANCE_HZ;
> diff --git a/drivers/gpu/drm/sti/sti_hda.c b/drivers/gpu/drm/sti/sti_hda.c
> index 03cc401ed593..a53b5a15c2a9 100644
> --- a/drivers/gpu/drm/sti/sti_hda.c
> +++ b/drivers/gpu/drm/sti/sti_hda.c
> @@ -601,8 +601,9 @@ static int sti_hda_connector_get_modes(struct drm_connector *connector)
>
> #define CLK_TOLERANCE_HZ 50
>
> -static int sti_hda_connector_mode_valid(struct drm_connector *connector,
> - struct drm_display_mode *mode)
> +static enum drm_mode_status
> +sti_hda_connector_mode_valid(struct drm_connector *connector,
> + struct drm_display_mode *mode)
> {
> int target = mode->clock * 1000;
> int target_min = target - CLK_TOLERANCE_HZ;
> diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c
> index cb82622877d2..09e0cadb6368 100644
> --- a/drivers/gpu/drm/sti/sti_hdmi.c
> +++ b/drivers/gpu/drm/sti/sti_hdmi.c
> @@ -1004,8 +1004,9 @@ static int sti_hdmi_connector_get_modes(struct drm_connector *connector)
>
> #define CLK_TOLERANCE_HZ 50
>
> -static int sti_hdmi_connector_mode_valid(struct drm_connector *connector,
> - struct drm_display_mode *mode)
> +static enum drm_mode_status
> +sti_hdmi_connector_mode_valid(struct drm_connector *connector,
> + struct drm_display_mode *mode)
> {
> int target = mode->clock * 1000;
> int target_min = target - CLK_TOLERANCE_HZ;
>
> base-commit: 9abf2313adc1ca1b6180c508c25f22f9395cc780
> --
> 2.38.1
>
>
Could someone please pick this up so that it makes 6.2? We would like
to try and get this warning turned on so that it can catch more
potential run time issues at compile time but that can only happen when
all the warnings are fixed.
Cheers,
Nathan
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH] drm/sti: Fix return type of sti_{dvo,hda,hdmi}_connector_mode_valid()
@ 2022-11-14 22:03 ` Nathan Chancellor
0 siblings, 0 replies; 10+ messages in thread
From: Nathan Chancellor @ 2022-11-14 22:03 UTC (permalink / raw)
To: Alain Volmat
Cc: Kees Cook, Tom Rix, llvm, Nick Desaulniers, linux-kernel,
dri-devel, patches, Sami Tolvanen
Hi all,
On Wed, Nov 02, 2022 at 08:56:23AM -0700, Nathan Chancellor wrote:
> With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
> indirect call targets are validated against the expected function
> pointer prototype to make sure the call target is valid to help mitigate
> ROP attacks. If they are not identical, there is a failure at run time,
> which manifests as either a kernel panic or thread getting killed. A
> proposed warning in clang aims to catch these at compile time, which
> reveals:
>
> drivers/gpu/drm/sti/sti_hda.c:637:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> .mode_valid = sti_hda_connector_mode_valid,
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
> drivers/gpu/drm/sti/sti_dvo.c:376:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> .mode_valid = sti_dvo_connector_mode_valid,
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
> drivers/gpu/drm/sti/sti_hdmi.c:1035:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> .mode_valid = sti_hdmi_connector_mode_valid,
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> ->mode_valid() in 'struct drm_connector_helper_funcs' expects a return
> type of 'enum drm_mode_status', not 'int'. Adjust the return type of
> sti_{dvo,hda,hdmi}_connector_mode_valid() to match the prototype's to
> resolve the warning and CFI failure.
>
> Link: https://github.com/ClangBuiltLinux/linux/issues/1750
> Signed-off-by: Nathan Chancellor <nathan@kernel.org>
> ---
> drivers/gpu/drm/sti/sti_dvo.c | 5 +++--
> drivers/gpu/drm/sti/sti_hda.c | 5 +++--
> drivers/gpu/drm/sti/sti_hdmi.c | 5 +++--
> 3 files changed, 9 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/gpu/drm/sti/sti_dvo.c b/drivers/gpu/drm/sti/sti_dvo.c
> index b6ee8a82e656..076d5f30a09c 100644
> --- a/drivers/gpu/drm/sti/sti_dvo.c
> +++ b/drivers/gpu/drm/sti/sti_dvo.c
> @@ -346,8 +346,9 @@ static int sti_dvo_connector_get_modes(struct drm_connector *connector)
>
> #define CLK_TOLERANCE_HZ 50
>
> -static int sti_dvo_connector_mode_valid(struct drm_connector *connector,
> - struct drm_display_mode *mode)
> +static enum drm_mode_status
> +sti_dvo_connector_mode_valid(struct drm_connector *connector,
> + struct drm_display_mode *mode)
> {
> int target = mode->clock * 1000;
> int target_min = target - CLK_TOLERANCE_HZ;
> diff --git a/drivers/gpu/drm/sti/sti_hda.c b/drivers/gpu/drm/sti/sti_hda.c
> index 03cc401ed593..a53b5a15c2a9 100644
> --- a/drivers/gpu/drm/sti/sti_hda.c
> +++ b/drivers/gpu/drm/sti/sti_hda.c
> @@ -601,8 +601,9 @@ static int sti_hda_connector_get_modes(struct drm_connector *connector)
>
> #define CLK_TOLERANCE_HZ 50
>
> -static int sti_hda_connector_mode_valid(struct drm_connector *connector,
> - struct drm_display_mode *mode)
> +static enum drm_mode_status
> +sti_hda_connector_mode_valid(struct drm_connector *connector,
> + struct drm_display_mode *mode)
> {
> int target = mode->clock * 1000;
> int target_min = target - CLK_TOLERANCE_HZ;
> diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c
> index cb82622877d2..09e0cadb6368 100644
> --- a/drivers/gpu/drm/sti/sti_hdmi.c
> +++ b/drivers/gpu/drm/sti/sti_hdmi.c
> @@ -1004,8 +1004,9 @@ static int sti_hdmi_connector_get_modes(struct drm_connector *connector)
>
> #define CLK_TOLERANCE_HZ 50
>
> -static int sti_hdmi_connector_mode_valid(struct drm_connector *connector,
> - struct drm_display_mode *mode)
> +static enum drm_mode_status
> +sti_hdmi_connector_mode_valid(struct drm_connector *connector,
> + struct drm_display_mode *mode)
> {
> int target = mode->clock * 1000;
> int target_min = target - CLK_TOLERANCE_HZ;
>
> base-commit: 9abf2313adc1ca1b6180c508c25f22f9395cc780
> --
> 2.38.1
>
>
Could someone please pick this up so that it makes 6.2? We would like
to try and get this warning turned on so that it can catch more
potential run time issues at compile time but that can only happen when
all the warnings are fixed.
Cheers,
Nathan
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH] drm/sti: Fix return type of sti_{dvo,hda,hdmi}_connector_mode_valid()
2022-11-14 22:03 ` Nathan Chancellor
@ 2022-11-17 23:19 ` Kees Cook
-1 siblings, 0 replies; 10+ messages in thread
From: Kees Cook @ 2022-11-17 23:19 UTC (permalink / raw)
To: Nathan Chancellor
Cc: Alain Volmat, dri-devel, Nick Desaulniers, Tom Rix,
Sami Tolvanen, llvm, linux-kernel, patches
On Mon, Nov 14, 2022 at 03:03:47PM -0700, Nathan Chancellor wrote:
> Hi all,
>
> On Wed, Nov 02, 2022 at 08:56:23AM -0700, Nathan Chancellor wrote:
> > With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
> > indirect call targets are validated against the expected function
> > pointer prototype to make sure the call target is valid to help mitigate
> > ROP attacks. If they are not identical, there is a failure at run time,
> > which manifests as either a kernel panic or thread getting killed. A
> > proposed warning in clang aims to catch these at compile time, which
> > reveals:
> >
> > drivers/gpu/drm/sti/sti_hda.c:637:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> > .mode_valid = sti_hda_connector_mode_valid,
> > ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
> > drivers/gpu/drm/sti/sti_dvo.c:376:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> > .mode_valid = sti_dvo_connector_mode_valid,
> > ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
> > drivers/gpu/drm/sti/sti_hdmi.c:1035:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> > .mode_valid = sti_hdmi_connector_mode_valid,
> > ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> >
> > ->mode_valid() in 'struct drm_connector_helper_funcs' expects a return
> > type of 'enum drm_mode_status', not 'int'. Adjust the return type of
> > sti_{dvo,hda,hdmi}_connector_mode_valid() to match the prototype's to
> > resolve the warning and CFI failure.
> >
> > Link: https://github.com/ClangBuiltLinux/linux/issues/1750
> > Signed-off-by: Nathan Chancellor <nathan@kernel.org>
> > ---
> > drivers/gpu/drm/sti/sti_dvo.c | 5 +++--
> > drivers/gpu/drm/sti/sti_hda.c | 5 +++--
> > drivers/gpu/drm/sti/sti_hdmi.c | 5 +++--
> > 3 files changed, 9 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/sti/sti_dvo.c b/drivers/gpu/drm/sti/sti_dvo.c
> > index b6ee8a82e656..076d5f30a09c 100644
> > --- a/drivers/gpu/drm/sti/sti_dvo.c
> > +++ b/drivers/gpu/drm/sti/sti_dvo.c
> > @@ -346,8 +346,9 @@ static int sti_dvo_connector_get_modes(struct drm_connector *connector)
> >
> > #define CLK_TOLERANCE_HZ 50
> >
> > -static int sti_dvo_connector_mode_valid(struct drm_connector *connector,
> > - struct drm_display_mode *mode)
> > +static enum drm_mode_status
> > +sti_dvo_connector_mode_valid(struct drm_connector *connector,
> > + struct drm_display_mode *mode)
> > {
> > int target = mode->clock * 1000;
> > int target_min = target - CLK_TOLERANCE_HZ;
> > diff --git a/drivers/gpu/drm/sti/sti_hda.c b/drivers/gpu/drm/sti/sti_hda.c
> > index 03cc401ed593..a53b5a15c2a9 100644
> > --- a/drivers/gpu/drm/sti/sti_hda.c
> > +++ b/drivers/gpu/drm/sti/sti_hda.c
> > @@ -601,8 +601,9 @@ static int sti_hda_connector_get_modes(struct drm_connector *connector)
> >
> > #define CLK_TOLERANCE_HZ 50
> >
> > -static int sti_hda_connector_mode_valid(struct drm_connector *connector,
> > - struct drm_display_mode *mode)
> > +static enum drm_mode_status
> > +sti_hda_connector_mode_valid(struct drm_connector *connector,
> > + struct drm_display_mode *mode)
> > {
> > int target = mode->clock * 1000;
> > int target_min = target - CLK_TOLERANCE_HZ;
> > diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c
> > index cb82622877d2..09e0cadb6368 100644
> > --- a/drivers/gpu/drm/sti/sti_hdmi.c
> > +++ b/drivers/gpu/drm/sti/sti_hdmi.c
> > @@ -1004,8 +1004,9 @@ static int sti_hdmi_connector_get_modes(struct drm_connector *connector)
> >
> > #define CLK_TOLERANCE_HZ 50
> >
> > -static int sti_hdmi_connector_mode_valid(struct drm_connector *connector,
> > - struct drm_display_mode *mode)
> > +static enum drm_mode_status
> > +sti_hdmi_connector_mode_valid(struct drm_connector *connector,
> > + struct drm_display_mode *mode)
> > {
> > int target = mode->clock * 1000;
> > int target_min = target - CLK_TOLERANCE_HZ;
> >
> > base-commit: 9abf2313adc1ca1b6180c508c25f22f9395cc780
> > --
> > 2.38.1
> >
> >
>
> Could someone please pick this up so that it makes 6.2? We would like
> to try and get this warning turned on so that it can catch more
> potential run time issues at compile time but that can only happen when
> all the warnings are fixed.
Same here; I'll pick this up tomorrow if no one else snags it.
--
Kees Cook
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH] drm/sti: Fix return type of sti_{dvo,hda,hdmi}_connector_mode_valid()
@ 2022-11-17 23:19 ` Kees Cook
0 siblings, 0 replies; 10+ messages in thread
From: Kees Cook @ 2022-11-17 23:19 UTC (permalink / raw)
To: Nathan Chancellor
Cc: Tom Rix, llvm, Nick Desaulniers, linux-kernel, dri-devel,
patches, Sami Tolvanen, Alain Volmat
On Mon, Nov 14, 2022 at 03:03:47PM -0700, Nathan Chancellor wrote:
> Hi all,
>
> On Wed, Nov 02, 2022 at 08:56:23AM -0700, Nathan Chancellor wrote:
> > With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
> > indirect call targets are validated against the expected function
> > pointer prototype to make sure the call target is valid to help mitigate
> > ROP attacks. If they are not identical, there is a failure at run time,
> > which manifests as either a kernel panic or thread getting killed. A
> > proposed warning in clang aims to catch these at compile time, which
> > reveals:
> >
> > drivers/gpu/drm/sti/sti_hda.c:637:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> > .mode_valid = sti_hda_connector_mode_valid,
> > ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
> > drivers/gpu/drm/sti/sti_dvo.c:376:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> > .mode_valid = sti_dvo_connector_mode_valid,
> > ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
> > drivers/gpu/drm/sti/sti_hdmi.c:1035:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_connector *, struct drm_display_mode *)' with an expression of type 'int (struct drm_connector *, struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> > .mode_valid = sti_hdmi_connector_mode_valid,
> > ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> >
> > ->mode_valid() in 'struct drm_connector_helper_funcs' expects a return
> > type of 'enum drm_mode_status', not 'int'. Adjust the return type of
> > sti_{dvo,hda,hdmi}_connector_mode_valid() to match the prototype's to
> > resolve the warning and CFI failure.
> >
> > Link: https://github.com/ClangBuiltLinux/linux/issues/1750
> > Signed-off-by: Nathan Chancellor <nathan@kernel.org>
> > ---
> > drivers/gpu/drm/sti/sti_dvo.c | 5 +++--
> > drivers/gpu/drm/sti/sti_hda.c | 5 +++--
> > drivers/gpu/drm/sti/sti_hdmi.c | 5 +++--
> > 3 files changed, 9 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/sti/sti_dvo.c b/drivers/gpu/drm/sti/sti_dvo.c
> > index b6ee8a82e656..076d5f30a09c 100644
> > --- a/drivers/gpu/drm/sti/sti_dvo.c
> > +++ b/drivers/gpu/drm/sti/sti_dvo.c
> > @@ -346,8 +346,9 @@ static int sti_dvo_connector_get_modes(struct drm_connector *connector)
> >
> > #define CLK_TOLERANCE_HZ 50
> >
> > -static int sti_dvo_connector_mode_valid(struct drm_connector *connector,
> > - struct drm_display_mode *mode)
> > +static enum drm_mode_status
> > +sti_dvo_connector_mode_valid(struct drm_connector *connector,
> > + struct drm_display_mode *mode)
> > {
> > int target = mode->clock * 1000;
> > int target_min = target - CLK_TOLERANCE_HZ;
> > diff --git a/drivers/gpu/drm/sti/sti_hda.c b/drivers/gpu/drm/sti/sti_hda.c
> > index 03cc401ed593..a53b5a15c2a9 100644
> > --- a/drivers/gpu/drm/sti/sti_hda.c
> > +++ b/drivers/gpu/drm/sti/sti_hda.c
> > @@ -601,8 +601,9 @@ static int sti_hda_connector_get_modes(struct drm_connector *connector)
> >
> > #define CLK_TOLERANCE_HZ 50
> >
> > -static int sti_hda_connector_mode_valid(struct drm_connector *connector,
> > - struct drm_display_mode *mode)
> > +static enum drm_mode_status
> > +sti_hda_connector_mode_valid(struct drm_connector *connector,
> > + struct drm_display_mode *mode)
> > {
> > int target = mode->clock * 1000;
> > int target_min = target - CLK_TOLERANCE_HZ;
> > diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c
> > index cb82622877d2..09e0cadb6368 100644
> > --- a/drivers/gpu/drm/sti/sti_hdmi.c
> > +++ b/drivers/gpu/drm/sti/sti_hdmi.c
> > @@ -1004,8 +1004,9 @@ static int sti_hdmi_connector_get_modes(struct drm_connector *connector)
> >
> > #define CLK_TOLERANCE_HZ 50
> >
> > -static int sti_hdmi_connector_mode_valid(struct drm_connector *connector,
> > - struct drm_display_mode *mode)
> > +static enum drm_mode_status
> > +sti_hdmi_connector_mode_valid(struct drm_connector *connector,
> > + struct drm_display_mode *mode)
> > {
> > int target = mode->clock * 1000;
> > int target_min = target - CLK_TOLERANCE_HZ;
> >
> > base-commit: 9abf2313adc1ca1b6180c508c25f22f9395cc780
> > --
> > 2.38.1
> >
> >
>
> Could someone please pick this up so that it makes 6.2? We would like
> to try and get this warning turned on so that it can catch more
> potential run time issues at compile time but that can only happen when
> all the warnings are fixed.
Same here; I'll pick this up tomorrow if no one else snags it.
--
Kees Cook
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH] drm/sti: Fix return type of sti_{dvo,hda,hdmi}_connector_mode_valid()
2022-11-02 15:56 ` [PATCH] drm/sti: Fix return type of sti_{dvo, hda, hdmi}_connector_mode_valid() Nathan Chancellor
@ 2022-11-18 17:40 ` Kees Cook
-1 siblings, 0 replies; 10+ messages in thread
From: Kees Cook @ 2022-11-18 17:40 UTC (permalink / raw)
To: nathan, alain.volmat
Cc: Kees Cook, llvm, trix, samitolvanen, patches, linux-kernel,
dri-devel, ndesaulniers
On Wed, 2 Nov 2022 08:56:23 -0700, Nathan Chancellor wrote:
> With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
> indirect call targets are validated against the expected function
> pointer prototype to make sure the call target is valid to help mitigate
> ROP attacks. If they are not identical, there is a failure at run time,
> which manifests as either a kernel panic or thread getting killed. A
> proposed warning in clang aims to catch these at compile time, which
> reveals:
>
> [...]
Applied to for-next/hardening, thanks!
[1/1] drm/sti: Fix return type of sti_{dvo,hda,hdmi}_connector_mode_valid()
https://git.kernel.org/kees/c/0ad811cc08a9
--
Kees Cook
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH] drm/sti: Fix return type of sti_{dvo, hda, hdmi}_connector_mode_valid()
@ 2022-11-18 17:40 ` Kees Cook
0 siblings, 0 replies; 10+ messages in thread
From: Kees Cook @ 2022-11-18 17:40 UTC (permalink / raw)
To: nathan, alain.volmat
Cc: Kees Cook, trix, llvm, ndesaulniers, linux-kernel, dri-devel,
patches, samitolvanen
On Wed, 2 Nov 2022 08:56:23 -0700, Nathan Chancellor wrote:
> With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
> indirect call targets are validated against the expected function
> pointer prototype to make sure the call target is valid to help mitigate
> ROP attacks. If they are not identical, there is a failure at run time,
> which manifests as either a kernel panic or thread getting killed. A
> proposed warning in clang aims to catch these at compile time, which
> reveals:
>
> [...]
Applied to for-next/hardening, thanks!
[1/1] drm/sti: Fix return type of sti_{dvo,hda,hdmi}_connector_mode_valid()
https://git.kernel.org/kees/c/0ad811cc08a9
--
Kees Cook
^ permalink raw reply [flat|nested] 10+ messages in thread