Skip to contentSkip to conversation

kconfig/menu.c: fix uninitialized variable warning

PATCH In review Madhavan Srinivasan 3 participants 8 messages Sep 19, 2013 – Sep 20, 2013

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 kconfig/menu.c: fix uninitialized variable warning

In file included from scripts/kconfig/zconf.tab.c:2537:0:
scripts/kconfig/menu.c: In function ‘get_symbol_str’:
scripts/kconfig/menu.c:586:18: warning: ‘jump’ may be used uninitialized in this function [-Wmaybe-uninitialized]

     jump->offset = r->len - 1;
                  ^
scripts/kconfig/menu.c:547:19: note: ‘jump’ was declared here
  struct jump_key *jump;
                   ^
---
 scripts/kconfig/menu.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
scripts/kconfig/menu.c +1 -1
index c1d5320..23b1827 100644
@@ -544,7 +544,7 @@ static void get_prompt_str(struct gstr *r, struct property *prop,
 {
 	int i, j;
 	struct menu *submenu[8], *menu, *location = NULL;
-	struct jump_key *jump;
+	struct jump_key *jump = NULL;
 
 	str_printf(r, _("Prompt: %s\n"), _(prop->text));
 	menu = prop->menu->parent;

--
1.7.10.4

#1 lore raw
Quoted 4 lines · Madhavan Srinivasan

In file included from scripts/kconfig/zconf.tab.c:2537:0:
scripts/kconfig/menu.c: In function ‘get_symbol_str’:
scripts/kconfig/menu.c:586:18: warning: ‘jump’ may be used uninitialized in this function [-Wmaybe-uninitialized]

                   ^^^^^

from gcc(1):
"These warnings are made optional because GCC is not smart enough to see all the reasons why the code might be correct despite appearing to have an error."

I do not see this warning with gcc 4.7.2 or 4.8.1. Which version are you using?

The code does:

static void get_prompt_str(struct gstr *r, struct property *prop, [...]

	if (head && location) {
		jump = xmalloc(sizeof(struct jump_key));
[... different basic block ...]
			if (head && location && menu == location)
				jump->offset = r->len - 1;

If the second "if" is true, then the first one was true as well, and "jump" is initilized.

Quoted 26 lines
     jump->offset = r->len - 1;
                  ^
scripts/kconfig/menu.c:547:19: note: ‘jump’ was declared here
  struct jump_key *jump;
                   ^
---
 scripts/kconfig/menu.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
scripts/kconfig/menu.c +1 -2
index c1d5320..23b1827 100644
@@ -544,7 +544,7 @@ static void get_prompt_str(struct gstr *r, struct property *prop,
 {
 	int i, j;
 	struct menu *submenu[8], *menu, *location = NULL;
-	struct jump_key *jump;
+	struct jump_key *jump = NULL;
 
 	str_printf(r, _("Prompt: %s\n"), _(prop->text));
 	menu = prop->menu->parent;
--

1.7.10.4

#2 lore raw

Benjamin, Madhavan, All,

On 2013-09-19 11:22 -0400, Benjamin Poirier spake thusly:

Quoted 19 lines · Madhavan Srinivasan
Quoted 4 lines

In file included from scripts/kconfig/zconf.tab.c:2537:0:
scripts/kconfig/menu.c: In function ‘get_symbol_str’:
scripts/kconfig/menu.c:586:18: warning: ‘jump’ may be used uninitialized in this function [-Wmaybe-uninitialized]

                   ^^^^^

from gcc(1):
"These warnings are made optional because GCC is not smart enough to see all the reasons why the code might be correct despite appearing to have an error."

I do not see this warning with gcc 4.7.2 or 4.8.1. Which version are you using?

The code does:

static void get_prompt_str(struct gstr *r, struct property *prop, [...]

Quoted 8 lines
	if (head && location) {
		jump = xmalloc(sizeof(struct jump_key));
[... different basic block ...]
			if (head && location && menu == location)
				jump->offset = r->len - 1;

If the second "if" is true, then the first one was true as well, and "jump" is initilized.

Exactly what I was going to say.

However:
- I believe we should strive for a warning-free code whenever possible
- while still getting help from gcc to pinpoint potential issues.

Clearly, gcc is wrong here. Setting jump to NULL will cause fault if we try to dereference it. Since this should never happen given the code as it currently is, I'm slightly in favour of acking this patch.

Any other reason not to apply it?

Quoted 21 lines
Quoted 21 lines
     jump->offset = r->len - 1;
                  ^
scripts/kconfig/menu.c:547:19: note: ‘jump’ was declared here
  struct jump_key *jump;
                   ^
---
 scripts/kconfig/menu.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
scripts/kconfig/menu.c +1 -1
index c1d5320..23b1827 100644
@@ -544,7 +544,7 @@ static void get_prompt_str(struct gstr *r, struct property *prop,
 {
 	int i, j;
 	struct menu *submenu[8], *menu, *location = NULL;
-	struct jump_key *jump;
+	struct jump_key *jump = NULL;
 
Quoted 5 lines
Quoted 5 lines
scripts/kconfig/menu.c +0 -1
 	str_printf(r, _("Prompt: %s\n"), _(prop->text));
 	menu = prop->menu->parent;
--

1.7.10.4

Regards,
Yann E. MORIN.

Signature · 6 lines
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'
#3 lore raw
Quoted 37 lines · Yann E. MORIN

Benjamin, Madhavan, All,

On 2013-09-19 11:22 -0400, Benjamin Poirier spake thusly:

Quoted 19 lines · Madhavan Srinivasan
Quoted 4 lines

In file included from scripts/kconfig/zconf.tab.c:2537:0:
scripts/kconfig/menu.c: In function ‘get_symbol_str’:
scripts/kconfig/menu.c:586:18: warning: ‘jump’ may be used uninitialized in this function [-Wmaybe-uninitialized]

                   ^^^^^

from gcc(1):
"These warnings are made optional because GCC is not smart enough to see all the reasons why the code might be correct despite appearing to have an error."

I do not see this warning with gcc 4.7.2 or 4.8.1. Which version are you using?

The code does:

static void get_prompt_str(struct gstr *r, struct property *prop, [...]

Quoted 8 lines
	if (head && location) {
		jump = xmalloc(sizeof(struct jump_key));
[... different basic block ...]
			if (head && location && menu == location)
				jump->offset = r->len - 1;

If the second "if" is true, then the first one was true as well, and "jump" is initilized.

Exactly what I was going to say.

However:
- I believe we should strive for a warning-free code whenever possible
- while still getting help from gcc to pinpoint potential issues.

Quoted 5 lines

Clearly, gcc is wrong here. Setting jump to NULL will cause fault if we try to dereference it. Since this should never happen given the code as it currently is, I'm slightly in favour of acking this patch.

Any other reason not to apply it?

Did you manage to get that warning as well? I didn't. If it's only from old versions of gcc I'd feel better leaving the warning there than masking it, in case a future code change does really introduce a use before initialize.

Quoted 38 lines
Quoted 22 lines
Quoted 22 lines
     jump->offset = r->len - 1;
                  ^
scripts/kconfig/menu.c:547:19: note: ‘jump’ was declared here
  struct jump_key *jump;
                   ^
---
 scripts/kconfig/menu.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
scripts/kconfig/menu.c +1 -1
index c1d5320..23b1827 100644
@@ -544,7 +544,7 @@ static void get_prompt_str(struct gstr *r, struct property *prop,
 {
 	int i, j;
 	struct menu *submenu[8], *menu, *location = NULL;
-	struct jump_key *jump;
+	struct jump_key *jump = NULL;
 
Quoted 5 lines
Quoted 5 lines
scripts/kconfig/menu.c +0 -1
 	str_printf(r, _("Prompt: %s\n"), _(prop->text));
 	menu = prop->menu->parent;
--

1.7.10.4

Regards,
Yann E. MORIN.

Signature · 6 lines
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'
#4 lore raw

Benjamin, All,

On 2013-09-19 15:13 -0400, Benjamin Poirier spake thusly:

Quoted 42 lines · Yann E. MORIN
Quoted 41 lines

Benjamin, Madhavan, All,

On 2013-09-19 11:22 -0400, Benjamin Poirier spake thusly:

Quoted 19 lines · Madhavan Srinivasan
Quoted 4 lines

In file included from scripts/kconfig/zconf.tab.c:2537:0:
scripts/kconfig/menu.c: In function ‘get_symbol_str’:
scripts/kconfig/menu.c:586:18: warning: ‘jump’ may be used uninitialized in this function [-Wmaybe-uninitialized]

                   ^^^^^

from gcc(1):
"These warnings are made optional because GCC is not smart enough to see all the reasons why the code might be correct despite appearing to have an error."

I do not see this warning with gcc 4.7.2 or 4.8.1. Which version are you using?

The code does:

static void get_prompt_str(struct gstr *r, struct property *prop, [...]

Quoted 8 lines
	if (head && location) {
		jump = xmalloc(sizeof(struct jump_key));
[... different basic block ...]
			if (head && location && menu == location)
				jump->offset = r->len - 1;

If the second "if" is true, then the first one was true as well, and "jump" is initilized.

Exactly what I was going to say.

However:
- I believe we should strive for a warning-free code whenever possible
- while still getting help from gcc to pinpoint potential issues.

Clearly, gcc is wrong here. Setting jump to NULL will cause fault if we try to dereference it. Since this should never happen given the code as it currently is, I'm slightly in favour of acking this patch.

Quoted 6 lines
Quoted 1 line

Any other reason not to apply it?

Did you manage to get that warning as well? I didn't. If it's only from old versions of gcc I'd feel better leaving the warning there than masking it, in case a future code change does really introduce a use before initialize.

No, I was not able to reproduce it with either gcc-4.4.7, gcc-4.6.4 or gcc-4.7.3 on my Ubuntu 13.04. Neither gcc-4.5 nor 4.8 are packaged, so I could not test them.

Anyway, -Wmaybe-uninitialized is new with 4.7. So, the warning can't happen with gcc < 4.7.

Madhavan, what gcc version are you using?

Regards,
Yann E. MORIN.

Signature · 6 lines
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'
#5 lore raw
Quoted 52 lines · Yann E. MORIN

Benjamin, All,

On 2013-09-19 15:13 -0400, Benjamin Poirier spake thusly:

Quoted 42 lines · Yann E. MORIN
Quoted 41 lines

Benjamin, Madhavan, All,

On 2013-09-19 11:22 -0400, Benjamin Poirier spake thusly:

Quoted 19 lines · Madhavan Srinivasan
Quoted 4 lines

In file included from scripts/kconfig/zconf.tab.c:2537:0:
scripts/kconfig/menu.c: In function ‘get_symbol_str’:
scripts/kconfig/menu.c:586:18: warning: ‘jump’ may be used uninitialized in this function [-Wmaybe-uninitialized]

                   ^^^^^

from gcc(1):
"These warnings are made optional because GCC is not smart enough to see all the reasons why the code might be correct despite appearing to have an error."

I do not see this warning with gcc 4.7.2 or 4.8.1. Which version are you using?

The code does:

static void get_prompt_str(struct gstr *r, struct property *prop, [...]

Quoted 8 lines
	if (head && location) {
		jump = xmalloc(sizeof(struct jump_key));
[... different basic block ...]
			if (head && location && menu == location)
				jump->offset = r->len - 1;

If the second "if" is true, then the first one was true as well, and "jump" is initilized.

Exactly what I was going to say.

However:
- I believe we should strive for a warning-free code whenever possible
- while still getting help from gcc to pinpoint potential issues.

Clearly, gcc is wrong here. Setting jump to NULL will cause fault if we try to dereference it. Since this should never happen given the code as it currently is, I'm slightly in favour of acking this patch.

Quoted 6 lines
Quoted 1 line

Any other reason not to apply it?

Did you manage to get that warning as well? I didn't. If it's only from old versions of gcc I'd feel better leaving the warning there than masking it, in case a future code change does really introduce a use before initialize.

Quoted 8 lines

No, I was not able to reproduce it with either gcc-4.4.7, gcc-4.6.4 or gcc-4.7.3 on my Ubuntu 13.04. Neither gcc-4.5 nor 4.8 are packaged, so I could not test them.

Anyway, -Wmaybe-uninitialized is new with 4.7. So, the warning can't happen with gcc < 4.7.

Madhavan, what gcc version are you using?

gcc i have is gcc version 4.8.0 20130526

Quoted 4 lines

Regards,
Yann E. MORIN.

#6 lore raw
Quoted 13 lines · Madhavan Srinivasan
Quoted 12 lines · Yann E. MORIN

Benjamin, All,

On 2013-09-19 15:13 -0400, Benjamin Poirier spake thusly:

Quoted 9 lines · Yann E. MORIN
Quoted 8 lines

Benjamin, Madhavan, All,

On 2013-09-19 11:22 -0400, Benjamin Poirier spake thusly:

Quoted 5 lines · Madhavan Srinivasan
Quoted 4 lines

In file included from scripts/kconfig/zconf.tab.c:2537:0:
scripts/kconfig/menu.c: In function ‘get_symbol_str’:
scripts/kconfig/menu.c:586:18: warning: ‘jump’ may be used uninitialized in this function [-Wmaybe-uninitialized]

[...]

Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines

I do not see this warning with gcc 4.7.2 or 4.8.1. Which version are you using?

[...]

Quoted 10 lines
Quoted 9 lines

No, I was not able to reproduce it with either gcc-4.4.7, gcc-4.6.4 or gcc-4.7.3 on my Ubuntu 13.04. Neither gcc-4.5 nor 4.8 are packaged, so I could not test them.

Anyway, -Wmaybe-uninitialized is new with 4.7. So, the warning can't happen with gcc < 4.7.

Madhavan, what gcc version are you using?

gcc i have is gcc version 4.8.0 20130526

In that case, I don't have a strong opinion whether the patch should be merged or not, but if it is merged, I think a comment should be added to the code or the changelog stating that this is to work around a deficiency observed in gcc 4.8.0.

#7 lore raw

Benjamin, Madhavan, All,

On 2013-09-20 13:58 -0400, Benjamin Poirier spake thusly:

Quoted 28 lines · Madhavan Srinivasan
Quoted 13 lines · Yann E. MORIN
Quoted 12 lines

Benjamin, All,

On 2013-09-19 15:13 -0400, Benjamin Poirier spake thusly:

Quoted 9 lines · Yann E. MORIN
Quoted 8 lines

Benjamin, Madhavan, All,

On 2013-09-19 11:22 -0400, Benjamin Poirier spake thusly:

Quoted 5 lines · Madhavan Srinivasan
Quoted 4 lines

In file included from scripts/kconfig/zconf.tab.c:2537:0:
scripts/kconfig/menu.c: In function ‘get_symbol_str’:
scripts/kconfig/menu.c:586:18: warning: ‘jump’ may be used uninitialized in this function [-Wmaybe-uninitialized]

[...]

Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines

I do not see this warning with gcc 4.7.2 or 4.8.1. Which version are you using?

[...]

Quoted 8 lines
Quoted 8 lines

No, I was not able to reproduce it with either gcc-4.4.7, gcc-4.6.4 or gcc-4.7.3 on my Ubuntu 13.04. Neither gcc-4.5 nor 4.8 are packaged, so I could not test them.

Anyway, -Wmaybe-uninitialized is new with 4.7. So, the warning can't happen with gcc < 4.7.

Quoted 7 lines
Quoted 2 lines
Quoted 1 line

Madhavan, what gcc version are you using?

gcc i have is gcc version 4.8.0 20130526

In that case, I don't have a strong opinion whether the patch should be merged or not, but if it is merged, I think a comment should be added to the code or the changelog stating that this is to work around a deficiency observed in gcc 4.8.0.

If it is /fixed/ in gcc 4.8.1 (as you imply since you tested it), I don't see a reason to apply this patch, in the end.

Rather, I would just put a comment in the code stating this is a spurious warning, and keep the code as-is.

Madhavan, care to resend your patch with such a comment instead of setting jump=NULL, please?

Regards,
Yann E. MORIN.

Signature · 6 lines
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'