Skip to contentSkip to conversation

Patch to include/linux/kernel.h breaks 3rd party modules.

PATCH Valdis Kletnieks 5 participants 5 messages Dec 21, 2016 – Dec 22, 2016

lore
Share & tools
lore
#0 lore raw

Yes, I know that usually out-of-tree modules are on their own. However, this one may require a rethink..

(Sorry for not catching this sooner, I hadn't tried to deal with the affected module since this patch hit linux-next in next-20161128)

commit 7fd8329ba502ef76dd91db561c7aed696b2c7720

Author: Petr Mladek <[email protected]>
Date:   Wed Sep 21 13:47:22 2016 +0200

taint/module: Clean up global and module taint flags handling

Contains this chunk:

include/linux/kernel.h +7 -0
@@ -506,6 +506,15 @@ extern enum system_states {
 #define TAINT_UNSIGNED_MODULE          13
 #define TAINT_SOFTLOCKUP               14
 #define TAINT_LIVEPATCH                        15
+#define TAINT_FLAGS_COUNT              16
+
+struct taint_flag {
+       char true;      /* character printed when tainted */
+       char false;     /* character printed when not tainted */
+       bool module;    /* also show as a per-module taint flag */
+};
 

and hilarity ensues when an out-of-tree module has this:

# ifndef true
#  define true  (1)
# endif
# ifndef false
#  define false (0)
# endif

My proposed fix: change true/false to tainted/untainted. If this is agreeable, I'll code and submit the fix.

#1 lore raw

+++ Valdis Kletnieks [21/12/16 15:42 -0500]:

Quoted 37 lines

Yes, I know that usually out-of-tree modules are on their own. However, this one may require a rethink..

(Sorry for not catching this sooner, I hadn't tried to deal with the affected module since this patch hit linux-next in next-20161128)

commit 7fd8329ba502ef76dd91db561c7aed696b2c7720

Author: Petr Mladek <[email protected]>
Date:   Wed Sep 21 13:47:22 2016 +0200

taint/module: Clean up global and module taint flags handling

Contains this chunk:

include/linux/kernel.h +7 -0
@@ -506,6 +506,15 @@ extern enum system_states {
 #define TAINT_UNSIGNED_MODULE          13
 #define TAINT_SOFTLOCKUP               14
 #define TAINT_LIVEPATCH                        15
+#define TAINT_FLAGS_COUNT              16
+
+struct taint_flag {
+       char true;      /* character printed when tainted */
+       char false;     /* character printed when not tainted */
+       bool module;    /* also show as a per-module taint flag */
+};
 

and hilarity ensues when an out-of-tree module has this:

# ifndef true
#  define true  (1)
# endif
# ifndef false
#  define false (0)
# endif
Quoted 2 lines

My proposed fix: change true/false to tainted/untainted. If this is agreeable, I'll code and submit the fix.

Sure, that's fine with me.

Jessica

#2 lore raw
In reply to Valdis Kletnieks
Quoted 37 lines · Valdis Kletnieks

Yes, I know that usually out-of-tree modules are on their own. However, this one may require a rethink..

(Sorry for not catching this sooner, I hadn't tried to deal with the affected module since this patch hit linux-next in next-20161128)

commit 7fd8329ba502ef76dd91db561c7aed696b2c7720

Author: Petr Mladek <[email protected]>
Date:   Wed Sep 21 13:47:22 2016 +0200

taint/module: Clean up global and module taint flags handling

Contains this chunk:

include/linux/kernel.h +7 -0
@@ -506,6 +506,15 @@ extern enum system_states {
 #define TAINT_UNSIGNED_MODULE          13
 #define TAINT_SOFTLOCKUP               14
 #define TAINT_LIVEPATCH                        15
+#define TAINT_FLAGS_COUNT              16
+
+struct taint_flag {
+       char true;      /* character printed when tainted */
+       char false;     /* character printed when not tainted */
+       bool module;    /* also show as a per-module taint flag */
+};
 

and hilarity ensues when an out-of-tree module has this:

# ifndef true
#  define true  (1)
# endif
# ifndef false
#  define false (0)
# endif
Quoted 2 lines

My proposed fix: change true/false to tainted/untainted. If this is agreeable, I'll code and submit the fix.

That certainly makes sense, but this kind of macros really ought to be killed off. In-tree there are only two such places - arch/powerpc/boot/types.h (no idea what's the environment there, but seeing that it starts with #include <stdbool.h>...) and fs/cifs/smbencrypt.c.

BTW, looking at arch/powerpc/boot/types.h... How does it manage to survive, anyway? gcc stdbool.h has #define bool _Bool, so that typedef int bool; in there would turn into typedef int _Bool, which should *not* be accepted by any C compiler. Or does it define __stdcplusplus somehow? Confused...

#3 lore raw
In reply to Valdis Kletnieks
Quoted 5 lines · Valdis Kletnieks

Yes, I know that usually out-of-tree modules are on their own. However, this one may require a rethink..

(Sorry for not catching this sooner, I hadn't tried to deal with the affected module since this patch hit linux-next in next-20161128)

So fix your out of tree module.

#4 lore raw
In reply to Valdis Kletnieks
Quoted 37 lines · Valdis Kletnieks

Yes, I know that usually out-of-tree modules are on their own. However, this one may require a rethink..

(Sorry for not catching this sooner, I hadn't tried to deal with the affected module since this patch hit linux-next in next-20161128)

commit 7fd8329ba502ef76dd91db561c7aed696b2c7720

Author: Petr Mladek <[email protected]>
Date:   Wed Sep 21 13:47:22 2016 +0200

taint/module: Clean up global and module taint flags handling

Contains this chunk:

include/linux/kernel.h +7 -0
@@ -506,6 +506,15 @@ extern enum system_states {
 #define TAINT_UNSIGNED_MODULE          13
 #define TAINT_SOFTLOCKUP               14
 #define TAINT_LIVEPATCH                        15
+#define TAINT_FLAGS_COUNT              16
+
+struct taint_flag {
+       char true;      /* character printed when tainted */
+       char false;     /* character printed when not tainted */
+       bool module;    /* also show as a per-module taint flag */
+};
 

and hilarity ensues when an out-of-tree module has this:

# ifndef true
#  define true  (1)
# endif
# ifndef false
#  define false (0)
# endif
Quoted 2 lines

My proposed fix: change true/false to tainted/untainted. If this is agreeable, I'll code and submit the fix.

Great catch! I did not have a good feeling about the names. But I did not found this problem and kept them to reduce changes in the code.

If we change it, I would go even further and make the purpose clear, e.g. use char_tainted/char_untainted. The names feel like booleans wihtout the char_ prefix.

Best Regards,
Petr