[PATCH 0/3] ucm: miscellaneous fixes

[PATCH 0/3] ucm: miscellaneous fixes

[Tanu Kaskinen]

While working on PulseAudio's UCM code, I noticed strange use of const
in snd_use_case_get(). The third patch fixes that.

While working on the third patch, I also noticed strange code in
execute_sequence(). The first patch fixes one obvious bug, but there
was another issue which I didn't know how to fix, so I just added a
FIXME comment instead (that's just as an RFC, I hope the discussion
will reveal what the proper fix should be).

Tanu Kaskinen (3):
  ucm: fix variable mixup
  ucm: add a FIXME comment
  ucm: fix inappropriate use of const

 include/use-case.h |  2 +-
 src/ucm/main.c     | 34 +++++++++++++++++++++-------------
 2 files changed, 22 insertions(+), 14 deletions(-)

-- 
1.9.3

[PATCH 1/3] ucm: fix variable mixup

[Tanu Kaskinen]

I assume the intention was to use cdev1 for PlaybackCTL and cdev2 for
CaptureCTL, but cdev1 was being used for both and cdev2 was not used
for anything.
---
 src/ucm/main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/ucm/main.c b/src/ucm/main.c
index 182f174..3924aee 100644
--- a/src/ucm/main.c
+++ b/src/ucm/main.c
@@ -308,7 +308,7 @@ static int execute_sequence(snd_use_case_mgr_t *uc_mgr,
 					uc_error("cdev is not defined!");
 					return err;
 				}
-				err = get_value3(&cdev1, "CaptureCTL",
+				err = get_value3(&cdev2, "CaptureCTL",
 						 value_list1,
 						 value_list2,
 						 value_list3);
-- 
1.9.3

[PATCH 2/3] ucm: add a FIXME comment

[Tanu Kaskinen]

I'm pretty sure the current code will crash with some inputs, but I
don't know what the original author intended this code to do, so I
don't know how to fix it.
---
 src/ucm/main.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/src/ucm/main.c b/src/ucm/main.c
index 3924aee..62bc374 100644
--- a/src/ucm/main.c
+++ b/src/ucm/main.c
@@ -317,6 +317,14 @@ static int execute_sequence(snd_use_case_mgr_t *uc_mgr,
 					uc_error("cdev is not defined!");
 					return err;
 				}
+                                /* FIXME: If cdev1 is NULL and cdev2 is not,
+                                 * then set cdev to cdev1... makes no sense!
+                                 * Also, what if both cdev1 and cdev2 are NULL?
+                                 * What should happen? Later in this function we
+                                 * call open_ctl(), which assumes non-NULL cdev,
+                                 * so leaving cdev to NULL here is not an
+                                 * option (or at least cdev has to be checked
+                                 * before calling open_ctl()). */
 				if (cdev1 == NULL || cdev2 == NULL ||
                                     strcmp(cdev1, cdev2) == 0) {
 					cdev = (char *)cdev1;
-- 
1.9.3

[PATCH 3/3] ucm: fix inappropriate use of const

[Tanu Kaskinen]

The caller is expected to free the string returned by
snd_use_case_get(), but for some reason the string is marked as const,
so the caller is forced to do type casting to get rid of compiler
warnings. I don't see any reason for using const in snd_use_case_get(),
so let's remove that.

This will cause warnings in application code, if applications
currently pass const variables to the function, which they very likely
do. I don't know if that's acceptable. If not, then it's too late to
fix this bug...
---
 include/use-case.h |  2 +-
 src/ucm/main.c     | 24 ++++++++++++------------
 2 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/include/use-case.h b/include/use-case.h
index f30168f..28d786f 100644
--- a/include/use-case.h
+++ b/include/use-case.h
@@ -286,7 +286,7 @@ int snd_use_case_get_list(snd_use_case_mgr_t *uc_mgr,
  */
 int snd_use_case_get(snd_use_case_mgr_t *uc_mgr,
                      const char *identifier,
-                     const char **value);
+                     char **value);
 
 /**
  * \brief Get current - integer
diff --git a/src/ucm/main.c b/src/ucm/main.c
index 62bc374..8fbf94f 100644
--- a/src/ucm/main.c
+++ b/src/ucm/main.c
@@ -40,9 +40,9 @@
  * misc
  */
 
-static int get_value1(const char **value, struct list_head *value_list,
+static int get_value1(char **value, struct list_head *value_list,
                       const char *identifier);
-static int get_value3(const char **value,
+static int get_value3(char **value,
 		      const char *identifier,
 		      struct list_head *value_list1,
 		      struct list_head *value_list2,
@@ -299,7 +299,7 @@ static int execute_sequence(snd_use_case_mgr_t *uc_mgr,
 		case SEQUENCE_ELEMENT_TYPE_CSET:
 		case SEQUENCE_ELEMENT_TYPE_CSET_BIN_FILE:
 			if (cdev == NULL) {
-				const char *cdev1 = NULL, *cdev2 = NULL;
+				char *cdev1 = NULL, *cdev2 = NULL;
 				err = get_value3(&cdev1, "PlaybackCTL",
 						 value_list1,
 						 value_list2,
@@ -313,7 +313,7 @@ static int execute_sequence(snd_use_case_mgr_t *uc_mgr,
 						 value_list2,
 						 value_list3);
 				if (err < 0 && err != ENOENT) {
-					free((char *)cdev1);
+					free(cdev1);
 					uc_error("cdev is not defined!");
 					return err;
 				}
@@ -327,11 +327,11 @@ static int execute_sequence(snd_use_case_mgr_t *uc_mgr,
                                  * before calling open_ctl()). */
 				if (cdev1 == NULL || cdev2 == NULL ||
                                     strcmp(cdev1, cdev2) == 0) {
-					cdev = (char *)cdev1;
-					free((char *)cdev2);
+					cdev = cdev1;
+					free(cdev2);
 				} else {
-					free((char *)cdev1);
-					free((char *)cdev2);
+					free(cdev1);
+					free(cdev2);
 				}
 			}
 			if (ctl == NULL) {
@@ -1233,7 +1233,7 @@ int snd_use_case_get_list(snd_use_case_mgr_t *uc_mgr,
 	return err;
 }
 
-static int get_value1(const char **value, struct list_head *value_list,
+static int get_value1(char **value, struct list_head *value_list,
                       const char *identifier)
 {
         struct ucm_value *val;
@@ -1254,7 +1254,7 @@ static int get_value1(const char **value, struct list_head *value_list,
         return -ENOENT;
 }
 
-static int get_value3(const char **value,
+static int get_value3(char **value,
 		      const char *identifier,
 		      struct list_head *value_list1,
 		      struct list_head *value_list2,
@@ -1284,7 +1284,7 @@ static int get_value3(const char **value,
  */
 static int get_value(snd_use_case_mgr_t *uc_mgr,
 			const char *identifier,
-			const char **value,
+			char **value,
 			const char *mod_dev_name,
 			const char *verb_name,
 			int exact)
@@ -1354,7 +1354,7 @@ static int get_value(snd_use_case_mgr_t *uc_mgr,
  */      
 int snd_use_case_get(snd_use_case_mgr_t *uc_mgr,
 		     const char *identifier,
-		     const char **value)
+		     char **value)
 {
 	const char *slash1, *slash2, *mod_dev_after;
 	const char *ident, *mod_dev, *verb;
-- 
1.9.3

Re: [PATCH 2/3] ucm: add a FIXME comment

[Takashi Iwai]

At Tue, 10 Feb 2015 22:42:33 +0200,
Tanu Kaskinen wrote:
> 
> I'm pretty sure the current code will crash with some inputs, but I
> don't know what the original author intended this code to do, so I
> don't know how to fix it.
> ---
>  src/ucm/main.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/src/ucm/main.c b/src/ucm/main.c
> index 3924aee..62bc374 100644
> --- a/src/ucm/main.c
> +++ b/src/ucm/main.c
> @@ -317,6 +317,14 @@ static int execute_sequence(snd_use_case_mgr_t *uc_mgr,
>  					uc_error("cdev is not defined!");
>  					return err;
>  				}
> +                                /* FIXME: If cdev1 is NULL and cdev2 is not,
> +                                 * then set cdev to cdev1... makes no sense!
> +                                 * Also, what if both cdev1 and cdev2 are NULL?
> +                                 * What should happen? Later in this function we
> +                                 * call open_ctl(), which assumes non-NULL cdev,
> +                                 * so leaving cdev to NULL here is not an
> +                                 * option (or at least cdev has to be checked
> +                                 * before calling open_ctl()). */
>  				if (cdev1 == NULL || cdev2 == NULL ||
>                                      strcmp(cdev1, cdev2) == 0) {
>  					cdev = (char *)cdev1;

Ouch, the code is really buggy there.  We must fix it instead of
leaving FIXME.

I see there are multiple bugs (in addition to your patch#1):

- the error check is wrong, it should be compared with -ENOENT
	if (err < 0 && err != ENOENT)

  This leaves cdev1 or cdev2 NULL as non-error.

- The intention of the code should be (as far as I understand):
  - if only one of cdev1 and cdev2 is defined, take it as cdev
  - if cdev1 and cdev2 are defined and have the same string, keep
    cdev1 and free cdev2
  - in the rest cases, free both cdev1 and cdev2 without changing cdev


Takashi

Re: [PATCH 3/3] ucm: fix inappropriate use of const

[Takashi Iwai]

At Tue, 10 Feb 2015 22:42:34 +0200,
Tanu Kaskinen wrote:
> 
> The caller is expected to free the string returned by
> snd_use_case_get(), but for some reason the string is marked as const,
> so the caller is forced to do type casting to get rid of compiler
> warnings. I don't see any reason for using const in snd_use_case_get(),
> so let's remove that.
> 
> This will cause warnings in application code, if applications
> currently pass const variables to the function, which they very likely
> do. I don't know if that's acceptable. If not, then it's too late to
> fix this bug...

I'm afraid it's too late to change this.


Takashi

> ---
>  include/use-case.h |  2 +-
>  src/ucm/main.c     | 24 ++++++++++++------------
>  2 files changed, 13 insertions(+), 13 deletions(-)
> 
> diff --git a/include/use-case.h b/include/use-case.h
> index f30168f..28d786f 100644
> --- a/include/use-case.h
> +++ b/include/use-case.h
> @@ -286,7 +286,7 @@ int snd_use_case_get_list(snd_use_case_mgr_t *uc_mgr,
>   */
>  int snd_use_case_get(snd_use_case_mgr_t *uc_mgr,
>                       const char *identifier,
> -                     const char **value);
> +                     char **value);
>  
>  /**
>   * \brief Get current - integer
> diff --git a/src/ucm/main.c b/src/ucm/main.c
> index 62bc374..8fbf94f 100644
> --- a/src/ucm/main.c
> +++ b/src/ucm/main.c
> @@ -40,9 +40,9 @@
>   * misc
>   */
>  
> -static int get_value1(const char **value, struct list_head *value_list,
> +static int get_value1(char **value, struct list_head *value_list,
>                        const char *identifier);
> -static int get_value3(const char **value,
> +static int get_value3(char **value,
>  		      const char *identifier,
>  		      struct list_head *value_list1,
>  		      struct list_head *value_list2,
> @@ -299,7 +299,7 @@ static int execute_sequence(snd_use_case_mgr_t *uc_mgr,
>  		case SEQUENCE_ELEMENT_TYPE_CSET:
>  		case SEQUENCE_ELEMENT_TYPE_CSET_BIN_FILE:
>  			if (cdev == NULL) {
> -				const char *cdev1 = NULL, *cdev2 = NULL;
> +				char *cdev1 = NULL, *cdev2 = NULL;
>  				err = get_value3(&cdev1, "PlaybackCTL",
>  						 value_list1,
>  						 value_list2,
> @@ -313,7 +313,7 @@ static int execute_sequence(snd_use_case_mgr_t *uc_mgr,
>  						 value_list2,
>  						 value_list3);
>  				if (err < 0 && err != ENOENT) {
> -					free((char *)cdev1);
> +					free(cdev1);
>  					uc_error("cdev is not defined!");
>  					return err;
>  				}
> @@ -327,11 +327,11 @@ static int execute_sequence(snd_use_case_mgr_t *uc_mgr,
>                                   * before calling open_ctl()). */
>  				if (cdev1 == NULL || cdev2 == NULL ||
>                                      strcmp(cdev1, cdev2) == 0) {
> -					cdev = (char *)cdev1;
> -					free((char *)cdev2);
> +					cdev = cdev1;
> +					free(cdev2);
>  				} else {
> -					free((char *)cdev1);
> -					free((char *)cdev2);
> +					free(cdev1);
> +					free(cdev2);
>  				}
>  			}
>  			if (ctl == NULL) {
> @@ -1233,7 +1233,7 @@ int snd_use_case_get_list(snd_use_case_mgr_t *uc_mgr,
>  	return err;
>  }
>  
> -static int get_value1(const char **value, struct list_head *value_list,
> +static int get_value1(char **value, struct list_head *value_list,
>                        const char *identifier)
>  {
>          struct ucm_value *val;
> @@ -1254,7 +1254,7 @@ static int get_value1(const char **value, struct list_head *value_list,
>          return -ENOENT;
>  }
>  
> -static int get_value3(const char **value,
> +static int get_value3(char **value,
>  		      const char *identifier,
>  		      struct list_head *value_list1,
>  		      struct list_head *value_list2,
> @@ -1284,7 +1284,7 @@ static int get_value3(const char **value,
>   */
>  static int get_value(snd_use_case_mgr_t *uc_mgr,
>  			const char *identifier,
> -			const char **value,
> +			char **value,
>  			const char *mod_dev_name,
>  			const char *verb_name,
>  			int exact)
> @@ -1354,7 +1354,7 @@ static int get_value(snd_use_case_mgr_t *uc_mgr,
>   */      
>  int snd_use_case_get(snd_use_case_mgr_t *uc_mgr,
>  		     const char *identifier,
> -		     const char **value)
> +		     char **value)
>  {
>  	const char *slash1, *slash2, *mod_dev_after;
>  	const char *ident, *mod_dev, *verb;
> -- 
> 1.9.3
> 

Re: [PATCH 1/3] ucm: fix variable mixup

[Takashi Iwai]

At Tue, 10 Feb 2015 22:42:32 +0200,
Tanu Kaskinen wrote:
> 
> I assume the intention was to use cdev1 for PlaybackCTL and cdev2 for
> CaptureCTL, but cdev1 was being used for both and cdev2 was not used
> for anything.

Applied, thanks.


Takashi

> ---
>  src/ucm/main.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/src/ucm/main.c b/src/ucm/main.c
> index 182f174..3924aee 100644
> --- a/src/ucm/main.c
> +++ b/src/ucm/main.c
> @@ -308,7 +308,7 @@ static int execute_sequence(snd_use_case_mgr_t *uc_mgr,
>  					uc_error("cdev is not defined!");
>  					return err;
>  				}
> -				err = get_value3(&cdev1, "CaptureCTL",
> +				err = get_value3(&cdev2, "CaptureCTL",
>  						 value_list1,
>  						 value_list2,
>  						 value_list3);
> -- 
> 1.9.3
> 

Re: [PATCH 2/3] ucm: add a FIXME comment

[Tanu Kaskinen]

On Wed, 2015-02-11 at 12:38 +0100, Takashi Iwai wrote:
> At Tue, 10 Feb 2015 22:42:33 +0200,
> Tanu Kaskinen wrote:
> > 
> > I'm pretty sure the current code will crash with some inputs, but I
> > don't know what the original author intended this code to do, so I
> > don't know how to fix it.
> > ---
> >  src/ucm/main.c | 8 ++++++++
> >  1 file changed, 8 insertions(+)
> > 
> > diff --git a/src/ucm/main.c b/src/ucm/main.c
> > index 3924aee..62bc374 100644
> > --- a/src/ucm/main.c
> > +++ b/src/ucm/main.c
> > @@ -317,6 +317,14 @@ static int execute_sequence(snd_use_case_mgr_t *uc_mgr,
> >  					uc_error("cdev is not defined!");
> >  					return err;
> >  				}
> > +                                /* FIXME: If cdev1 is NULL and cdev2 is not,
> > +                                 * then set cdev to cdev1... makes no sense!
> > +                                 * Also, what if both cdev1 and cdev2 are NULL?
> > +                                 * What should happen? Later in this function we
> > +                                 * call open_ctl(), which assumes non-NULL cdev,
> > +                                 * so leaving cdev to NULL here is not an
> > +                                 * option (or at least cdev has to be checked
> > +                                 * before calling open_ctl()). */
> >  				if (cdev1 == NULL || cdev2 == NULL ||
> >                                      strcmp(cdev1, cdev2) == 0) {
> >  					cdev = (char *)cdev1;
> 
> Ouch, the code is really buggy there.  We must fix it instead of
> leaving FIXME.
> 
> I see there are multiple bugs (in addition to your patch#1):
> 
> - the error check is wrong, it should be compared with -ENOENT
> 	if (err < 0 && err != ENOENT)
> 
>   This leaves cdev1 or cdev2 NULL as non-error.
> 
> - The intention of the code should be (as far as I understand):
>   - if only one of cdev1 and cdev2 is defined, take it as cdev
>   - if cdev1 and cdev2 are defined and have the same string, keep
>     cdev1 and free cdev2
>   - in the rest cases, free both cdev1 and cdev2 without changing cdev

Regarding that last case, just leaving cdev unchanged is not a good
idea. cdev will always be NULL if we don't set it here, and when we call
open_ctl(), it must be non-NULL.

-- 
Tanu

Re: [PATCH 2/3] ucm: add a FIXME comment

[Takashi Iwai]

At Wed, 11 Feb 2015 13:44:40 +0200,
Tanu Kaskinen wrote:
> 
> On Wed, 2015-02-11 at 12:38 +0100, Takashi Iwai wrote:
> > At Tue, 10 Feb 2015 22:42:33 +0200,
> > Tanu Kaskinen wrote:
> > > 
> > > I'm pretty sure the current code will crash with some inputs, but I
> > > don't know what the original author intended this code to do, so I
> > > don't know how to fix it.
> > > ---
> > >  src/ucm/main.c | 8 ++++++++
> > >  1 file changed, 8 insertions(+)
> > > 
> > > diff --git a/src/ucm/main.c b/src/ucm/main.c
> > > index 3924aee..62bc374 100644
> > > --- a/src/ucm/main.c
> > > +++ b/src/ucm/main.c
> > > @@ -317,6 +317,14 @@ static int execute_sequence(snd_use_case_mgr_t *uc_mgr,
> > >  					uc_error("cdev is not defined!");
> > >  					return err;
> > >  				}
> > > +                                /* FIXME: If cdev1 is NULL and cdev2 is not,
> > > +                                 * then set cdev to cdev1... makes no sense!
> > > +                                 * Also, what if both cdev1 and cdev2 are NULL?
> > > +                                 * What should happen? Later in this function we
> > > +                                 * call open_ctl(), which assumes non-NULL cdev,
> > > +                                 * so leaving cdev to NULL here is not an
> > > +                                 * option (or at least cdev has to be checked
> > > +                                 * before calling open_ctl()). */
> > >  				if (cdev1 == NULL || cdev2 == NULL ||
> > >                                      strcmp(cdev1, cdev2) == 0) {
> > >  					cdev = (char *)cdev1;
> > 
> > Ouch, the code is really buggy there.  We must fix it instead of
> > leaving FIXME.
> > 
> > I see there are multiple bugs (in addition to your patch#1):
> > 
> > - the error check is wrong, it should be compared with -ENOENT
> > 	if (err < 0 && err != ENOENT)
> > 
> >   This leaves cdev1 or cdev2 NULL as non-error.
> > 
> > - The intention of the code should be (as far as I understand):
> >   - if only one of cdev1 and cdev2 is defined, take it as cdev
> >   - if cdev1 and cdev2 are defined and have the same string, keep
> >     cdev1 and free cdev2
> >   - in the rest cases, free both cdev1 and cdev2 without changing cdev
> 
> Regarding that last case, just leaving cdev unchanged is not a good
> idea. cdev will always be NULL if we don't set it here, and when we call
> open_ctl(), it must be non-NULL.

Right.  I *guess* if cdev1 != cdev2, it should bail out as an error.


Takashi

Re: [PATCH 2/3] ucm: add a FIXME comment

[Tanu Kaskinen]

On Wed, 2015-02-11 at 14:01 +0100, Takashi Iwai wrote:
> At Wed, 11 Feb 2015 13:44:40 +0200,
> Tanu Kaskinen wrote:
> > 
> > On Wed, 2015-02-11 at 12:38 +0100, Takashi Iwai wrote:
> > > At Tue, 10 Feb 2015 22:42:33 +0200,
> > > Tanu Kaskinen wrote:
> > > > 
> > > > I'm pretty sure the current code will crash with some inputs, but I
> > > > don't know what the original author intended this code to do, so I
> > > > don't know how to fix it.
> > > > ---
> > > >  src/ucm/main.c | 8 ++++++++
> > > >  1 file changed, 8 insertions(+)
> > > > 
> > > > diff --git a/src/ucm/main.c b/src/ucm/main.c
> > > > index 3924aee..62bc374 100644
> > > > --- a/src/ucm/main.c
> > > > +++ b/src/ucm/main.c
> > > > @@ -317,6 +317,14 @@ static int execute_sequence(snd_use_case_mgr_t *uc_mgr,
> > > >  					uc_error("cdev is not defined!");
> > > >  					return err;
> > > >  				}
> > > > +                                /* FIXME: If cdev1 is NULL and cdev2 is not,
> > > > +                                 * then set cdev to cdev1... makes no sense!
> > > > +                                 * Also, what if both cdev1 and cdev2 are NULL?
> > > > +                                 * What should happen? Later in this function we
> > > > +                                 * call open_ctl(), which assumes non-NULL cdev,
> > > > +                                 * so leaving cdev to NULL here is not an
> > > > +                                 * option (or at least cdev has to be checked
> > > > +                                 * before calling open_ctl()). */
> > > >  				if (cdev1 == NULL || cdev2 == NULL ||
> > > >                                      strcmp(cdev1, cdev2) == 0) {
> > > >  					cdev = (char *)cdev1;
> > > 
> > > Ouch, the code is really buggy there.  We must fix it instead of
> > > leaving FIXME.
> > > 
> > > I see there are multiple bugs (in addition to your patch#1):
> > > 
> > > - the error check is wrong, it should be compared with -ENOENT
> > > 	if (err < 0 && err != ENOENT)
> > > 
> > >   This leaves cdev1 or cdev2 NULL as non-error.
> > > 
> > > - The intention of the code should be (as far as I understand):
> > >   - if only one of cdev1 and cdev2 is defined, take it as cdev
> > >   - if cdev1 and cdev2 are defined and have the same string, keep
> > >     cdev1 and free cdev2
> > >   - in the rest cases, free both cdev1 and cdev2 without changing cdev
> > 
> > Regarding that last case, just leaving cdev unchanged is not a good
> > idea. cdev will always be NULL if we don't set it here, and when we call
> > open_ctl(), it must be non-NULL.
> 
> Right.  I *guess* if cdev1 != cdev2, it should bail out as an error.

Ok, I'll make a patch that will implement the described behaviour.

-- 
Tanu

Re: [PATCH 2/3] ucm: add a FIXME comment

[Takashi Iwai]

At Wed, 11 Feb 2015 15:34:06 +0200,
Tanu Kaskinen wrote:
> 
> On Wed, 2015-02-11 at 14:01 +0100, Takashi Iwai wrote:
> > At Wed, 11 Feb 2015 13:44:40 +0200,
> > Tanu Kaskinen wrote:
> > > 
> > > On Wed, 2015-02-11 at 12:38 +0100, Takashi Iwai wrote:
> > > > At Tue, 10 Feb 2015 22:42:33 +0200,
> > > > Tanu Kaskinen wrote:
> > > > > 
> > > > > I'm pretty sure the current code will crash with some inputs, but I
> > > > > don't know what the original author intended this code to do, so I
> > > > > don't know how to fix it.
> > > > > ---
> > > > >  src/ucm/main.c | 8 ++++++++
> > > > >  1 file changed, 8 insertions(+)
> > > > > 
> > > > > diff --git a/src/ucm/main.c b/src/ucm/main.c
> > > > > index 3924aee..62bc374 100644
> > > > > --- a/src/ucm/main.c
> > > > > +++ b/src/ucm/main.c
> > > > > @@ -317,6 +317,14 @@ static int execute_sequence(snd_use_case_mgr_t *uc_mgr,
> > > > >  					uc_error("cdev is not defined!");
> > > > >  					return err;
> > > > >  				}
> > > > > +                                /* FIXME: If cdev1 is NULL and cdev2 is not,
> > > > > +                                 * then set cdev to cdev1... makes no sense!
> > > > > +                                 * Also, what if both cdev1 and cdev2 are NULL?
> > > > > +                                 * What should happen? Later in this function we
> > > > > +                                 * call open_ctl(), which assumes non-NULL cdev,
> > > > > +                                 * so leaving cdev to NULL here is not an
> > > > > +                                 * option (or at least cdev has to be checked
> > > > > +                                 * before calling open_ctl()). */
> > > > >  				if (cdev1 == NULL || cdev2 == NULL ||
> > > > >                                      strcmp(cdev1, cdev2) == 0) {
> > > > >  					cdev = (char *)cdev1;
> > > > 
> > > > Ouch, the code is really buggy there.  We must fix it instead of
> > > > leaving FIXME.
> > > > 
> > > > I see there are multiple bugs (in addition to your patch#1):
> > > > 
> > > > - the error check is wrong, it should be compared with -ENOENT
> > > > 	if (err < 0 && err != ENOENT)
> > > > 
> > > >   This leaves cdev1 or cdev2 NULL as non-error.
> > > > 
> > > > - The intention of the code should be (as far as I understand):
> > > >   - if only one of cdev1 and cdev2 is defined, take it as cdev
> > > >   - if cdev1 and cdev2 are defined and have the same string, keep
> > > >     cdev1 and free cdev2
> > > >   - in the rest cases, free both cdev1 and cdev2 without changing cdev
> > > 
> > > Regarding that last case, just leaving cdev unchanged is not a good
> > > idea. cdev will always be NULL if we don't set it here, and when we call
> > > open_ctl(), it must be non-NULL.
> > 
> > Right.  I *guess* if cdev1 != cdev2, it should bail out as an error.
> 
> Ok, I'll make a patch that will implement the described behaviour.

Thanks!


Takashi

Re: [PATCH 2/3] ucm: add a FIXME comment

[Liam Girdwood]

On Wed, 2015-02-11 at 15:34 +0200, Tanu Kaskinen wrote:
> On Wed, 2015-02-11 at 14:01 +0100, Takashi Iwai wrote:
> > At Wed, 11 Feb 2015 13:44:40 +0200,
> > Tanu Kaskinen wrote:
> > > 
> > > On Wed, 2015-02-11 at 12:38 +0100, Takashi Iwai wrote:
> > > > At Tue, 10 Feb 2015 22:42:33 +0200,
> > > > Tanu Kaskinen wrote:
> > > > > 
> > > > > I'm pretty sure the current code will crash with some inputs, but I
> > > > > don't know what the original author intended this code to do, so I
> > > > > don't know how to fix it.
> > > > > ---
> > > > >  src/ucm/main.c | 8 ++++++++
> > > > >  1 file changed, 8 insertions(+)
> > > > > 
> > > > > diff --git a/src/ucm/main.c b/src/ucm/main.c
> > > > > index 3924aee..62bc374 100644
> > > > > --- a/src/ucm/main.c
> > > > > +++ b/src/ucm/main.c
> > > > > @@ -317,6 +317,14 @@ static int execute_sequence(snd_use_case_mgr_t *uc_mgr,
> > > > >  					uc_error("cdev is not defined!");
> > > > >  					return err;
> > > > >  				}
> > > > > +                                /* FIXME: If cdev1 is NULL and cdev2 is not,
> > > > > +                                 * then set cdev to cdev1... makes no sense!
> > > > > +                                 * Also, what if both cdev1 and cdev2 are NULL?
> > > > > +                                 * What should happen? Later in this function we
> > > > > +                                 * call open_ctl(), which assumes non-NULL cdev,
> > > > > +                                 * so leaving cdev to NULL here is not an
> > > > > +                                 * option (or at least cdev has to be checked
> > > > > +                                 * before calling open_ctl()). */
> > > > >  				if (cdev1 == NULL || cdev2 == NULL ||
> > > > >                                      strcmp(cdev1, cdev2) == 0) {
> > > > >  					cdev = (char *)cdev1;
> > > > 
> > > > Ouch, the code is really buggy there.  We must fix it instead of
> > > > leaving FIXME.
> > > > 
> > > > I see there are multiple bugs (in addition to your patch#1):
> > > > 
> > > > - the error check is wrong, it should be compared with -ENOENT
> > > > 	if (err < 0 && err != ENOENT)
> > > > 
> > > >   This leaves cdev1 or cdev2 NULL as non-error.
> > > > 
> > > > - The intention of the code should be (as far as I understand):
> > > >   - if only one of cdev1 and cdev2 is defined, take it as cdev
> > > >   - if cdev1 and cdev2 are defined and have the same string, keep
> > > >     cdev1 and free cdev2
> > > >   - in the rest cases, free both cdev1 and cdev2 without changing cdev
> > > 
> > > Regarding that last case, just leaving cdev unchanged is not a good
> > > idea. cdev will always be NULL if we don't set it here, and when we call
> > > open_ctl(), it must be non-NULL.
> > 
> > Right.  I *guess* if cdev1 != cdev2, it should bail out as an error.
> 
> Ok, I'll make a patch that will implement the described behaviour.
> 

It may also be worth adding a comment describing the cdev1/2 usage or
maybe even rename them to make it obvious ?

Thanks

Liam