[PATCH 1/4] android: Fix build with GCC 10

[PATCH 1/4] android: Fix build with GCC 10

[Szymon Janc]

status and state are used to hold various enum types depending on test
and callback passed. Define them as int to avoid warnings about enum
assignment from invalid type.
---
 android/tester-main.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/android/tester-main.h b/android/tester-main.h
index 8a7384c57..6bdfdbcdb 100644
--- a/android/tester-main.h
+++ b/android/tester-main.h
@@ -653,8 +653,8 @@ struct map_inst_data {
  * matching with expected step data.
  */
 struct bt_callback_data {
-	bt_state_t state;
-	bt_status_t status;
+	int state;
+	int status;
 	int num_properties;
 	bt_property_t *properties;
 	bt_uuid_t *uuid;
-- 
2.26.0

[PATCH 2/4] sap: Fix compilation with GCC 10

[Szymon Janc]

---
 profiles/sap/sap.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/profiles/sap/sap.h b/profiles/sap/sap.h
index 16c333a9b..6e78d6c2d 100644
--- a/profiles/sap/sap.h
+++ b/profiles/sap/sap.h
@@ -76,7 +76,7 @@ struct sap_parameter {
 	uint8_t id;
 	uint8_t reserved;
 	uint16_t len;
-	uint8_t val[0];
+	uint8_t val[];
 	/*
 	 * Padding bytes 0-3 bytes
 	 */
@@ -86,7 +86,7 @@ struct sap_message {
 	uint8_t id;
 	uint8_t nparam;
 	uint16_t reserved;
-	struct sap_parameter param[0];
+	struct sap_parameter param[];
 } __attribute__((packed));
 
 #define SAP_BUF_SIZE		512
-- 
2.26.0

[PATCH 3/4] avctp: Fix compilation with GCC 10

[Szymon Janc]

This one is a false positive but since we never use more than
UINPUT_MAX_NAME_SIZE bytes of name we can silence GCC by reducing
size of source string.

  CC       profiles/audio/bluetoothd-avctp.o
In function ‘uinput_create’,
    inlined from ‘init_uinput’ at profiles/audio/avctp.c:1259:20:
profiles/audio/avctp.c:1188:3: error: ‘strncpy’ output may be truncated copying 79 bytes from a string of length 248 [-Werror=stringop-truncation]
 1188 |   strncpy(dev.name, name, UINPUT_MAX_NAME_SIZE);
      |   ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
---
 profiles/audio/avctp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/profiles/audio/avctp.c b/profiles/audio/avctp.c
index 37ffde9e7..058b44a8b 100644
--- a/profiles/audio/avctp.c
+++ b/profiles/audio/avctp.c
@@ -1246,7 +1246,7 @@ static int uinput_create(struct btd_device *device, const char *name,
 
 static void init_uinput(struct avctp *session)
 {
-	char name[248 + 1];
+	char name[UINPUT_MAX_NAME_SIZE];
 
 	device_get_name(session->device, name, sizeof(name));
 	if (g_str_equal(name, "Nokia CK-20W")) {
-- 
2.26.0

[PATCH 4/4] device: Fix compilation with GCC 10

[Szymon Janc]

Class is using only 3 bytes so make sure GCC is aware of that.

src/device.c:397:3: note: ‘sprintf’ output between 9 and 11 bytes into a destination of size 9
  397 |   sprintf(class, "0x%6.6x", device->class);
      |   ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
---
 src/device.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/device.c b/src/device.c
index 5f9ad227d..a8d95346a 100644
--- a/src/device.c
+++ b/src/device.c
@@ -394,7 +394,7 @@ static gboolean store_device_info_cb(gpointer user_data)
 		g_key_file_remove_key(key_file, "General", "Alias", NULL);
 
 	if (device->class) {
-		sprintf(class, "0x%6.6x", device->class);
+		sprintf(class, "0x%6.6x", device->class & 0xffffff);
 		g_key_file_set_string(key_file, "General", "Class", class);
 	} else {
 		g_key_file_remove_key(key_file, "General", "Class", NULL);
-- 
2.26.0

Re: [PATCH 3/4] avctp: Fix compilation with GCC 10

[Gix, Brian]

On Mon, 2020-04-20 at 14:07 +0200, Szymon Janc wrote:
> This one is a false positive but since we never use more than
> UINPUT_MAX_NAME_SIZE bytes of name we can silence GCC by reducing
> size of source string.
> 
>   CC       profiles/audio/bluetoothd-avctp.o
> In function ‘uinput_create’,
>     inlined from ‘init_uinput’ at profiles/audio/avctp.c:1259:20:
> profiles/audio/avctp.c:1188:3: error: ‘strncpy’ output may be truncated copying 79 bytes from a string of
> length 248 [-Werror=stringop-truncation]
>  1188 |   strncpy(dev.name, name, UINPUT_MAX_NAME_SIZE);
>       |   ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> cc1: all warnings being treated as errors
> ---
>  profiles/audio/avctp.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/profiles/audio/avctp.c b/profiles/audio/avctp.c
> index 37ffde9e7..058b44a8b 100644
> --- a/profiles/audio/avctp.c
> +++ b/profiles/audio/avctp.c
> @@ -1246,7 +1246,7 @@ static int uinput_create(struct btd_device *device, const char *name,
>  
>  static void init_uinput(struct avctp *session)
>  {
> -	char name[248 + 1];
> +	char name[UINPUT_MAX_NAME_SIZE];

Should this be nul terminated?  (UINPUT_MAX_NAME_SIZE + 1)  ?

>  
>  	device_get_name(session->device, name, sizeof(name));
>  	if (g_str_equal(name, "Nokia CK-20W")) {

Re: [PATCH 1/4] android: Fix build with GCC 10

[Gix, Brian]

Hi Luiz,

Modulo my question about nul termination on patch 3/4,  This patchset resolves all of the GCC 10 build
error/warnings and, I think, should be applied.

On Mon, 2020-04-20 at 14:07 +0200, Szymon Janc wrote:
> status and state are used to hold various enum types depending on test
> and callback passed. Define them as int to avoid warnings about enum
> assignment from invalid type.
> ---
>  android/tester-main.h | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/android/tester-main.h b/android/tester-main.h
> index 8a7384c57..6bdfdbcdb 100644
> --- a/android/tester-main.h
> +++ b/android/tester-main.h
> @@ -653,8 +653,8 @@ struct map_inst_data {
>   * matching with expected step data.
>   */
>  struct bt_callback_data {
> -	bt_state_t state;
> -	bt_status_t status;
> +	int state;
> +	int status;
>  	int num_properties;
>  	bt_property_t *properties;
>  	bt_uuid_t *uuid;

Re: [PATCH 3/4] avctp: Fix compilation with GCC 10

[Luiz Augusto von Dentz]

Hi Brian,

On Mon, Apr 20, 2020 at 9:44 AM Gix, Brian <brian.gix@intel.com> wrote:
>
> On Mon, 2020-04-20 at 14:07 +0200, Szymon Janc wrote:
> > This one is a false positive but since we never use more than
> > UINPUT_MAX_NAME_SIZE bytes of name we can silence GCC by reducing
> > size of source string.
> >
> >   CC       profiles/audio/bluetoothd-avctp.o
> > In function ‘uinput_create’,
> >     inlined from ‘init_uinput’ at profiles/audio/avctp.c:1259:20:
> > profiles/audio/avctp.c:1188:3: error: ‘strncpy’ output may be truncated copying 79 bytes from a string of
> > length 248 [-Werror=stringop-truncation]
> >  1188 |   strncpy(dev.name, name, UINPUT_MAX_NAME_SIZE);
> >       |   ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> > cc1: all warnings being treated as errors
> > ---
> >  profiles/audio/avctp.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/profiles/audio/avctp.c b/profiles/audio/avctp.c
> > index 37ffde9e7..058b44a8b 100644
> > --- a/profiles/audio/avctp.c
> > +++ b/profiles/audio/avctp.c
> > @@ -1246,7 +1246,7 @@ static int uinput_create(struct btd_device *device, const char *name,
> >
> >  static void init_uinput(struct avctp *session)
> >  {
> > -     char name[248 + 1];
> > +     char name[UINPUT_MAX_NAME_SIZE];
>
> Should this be nul terminated?  (UINPUT_MAX_NAME_SIZE + 1)  ?

I guess not since that is send over to the kernel which accepts up to
UINPUT_MAX_NAME_SIZE so if the name is exactly 248 then the kernel
should be the one adding the NULL termination, if it doesn't then we
need to truncate by doing name[247] = '\0'.

> >
> >       device_get_name(session->device, name, sizeof(name));
> >       if (g_str_equal(name, "Nokia CK-20W")) {



-- 
Luiz Augusto von Dentz

Re: [PATCH 1/4] android: Fix build with GCC 10

[Luiz Augusto von Dentz]

Hi Szymon,

On Mon, Apr 20, 2020 at 9:42 AM Gix, Brian <brian.gix@intel.com> wrote:
>
> Hi Luiz,
>
> Modulo my question about nul termination on patch 3/4,  This patchset resolves all of the GCC 10 build
> error/warnings and, I think, should be applied.
>
> On Mon, 2020-04-20 at 14:07 +0200, Szymon Janc wrote:
> > status and state are used to hold various enum types depending on test
> > and callback passed. Define them as int to avoid warnings about enum
> > assignment from invalid type.
> > ---
> >  android/tester-main.h | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/android/tester-main.h b/android/tester-main.h
> > index 8a7384c57..6bdfdbcdb 100644
> > --- a/android/tester-main.h
> > +++ b/android/tester-main.h
> > @@ -653,8 +653,8 @@ struct map_inst_data {
> >   * matching with expected step data.
> >   */
> >  struct bt_callback_data {
> > -     bt_state_t state;
> > -     bt_status_t status;
> > +     int state;
> > +     int status;
> >       int num_properties;
> >       bt_property_t *properties;
> >       bt_uuid_t *uuid;

Applied, thanks.

-- 
Luiz Augusto von Dentz