Skip to contentSkip to conversation

kbuild: make module name conflict fatal error

PATCH In review Masahiro Yamada 4 participants 6 messages May 11, 2020 – Jun 9, 2020

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 kbuild: make module name conflict fatal error

I think all the warnings have been fixed by now. Make it a fatal error.

Check it before modpost because we need to stop building *.ko files. Also, pass modules.order via a script parameter.

---

 Makefile                 |  7 +++++--
 scripts/modules-check.sh | 16 +++++++++++++---
 2 files changed, 18 insertions(+), 5 deletions(-)
Makefile +5 -2
index 1162cb04860c..a2e4d0945500 100644
@@ -1328,9 +1328,12 @@ all: modules
 # using awk while concatenating to the final file.
 
 PHONY += modules
-modules: $(if $(KBUILD_BUILTIN),vmlinux) modules.order
+modules: $(if $(KBUILD_BUILTIN),vmlinux) modules_check
 	$(Q)$(MAKE) -f $(srctree)/scripts/Makefile.modpost
-	$(Q)$(CONFIG_SHELL) $(srctree)/scripts/modules-check.sh
+
+PHONY += modules_check
+modules_check: modules.order
+	$(Q)$(CONFIG_SHELL) $(srctree)/scripts/modules-check.sh $<
 
 modules.order: descend
 	$(Q)$(AWK) '!x[$$0]++' $(addsuffix /$@, $(build-dirs)) > $@
scripts/modules-check.sh +13 -3
index f51f446707b8..43de226071ae 100755
@@ -3,14 +3,24 @@
 
 set -e
 
+if [ $# != 1 ]; then
+	echo "Usage: $0 <modules.order>" >& 2
+	exit 1
+fi
+
+exit_code=0
+
 # Check uniqueness of module names
 check_same_name_modules()
 {
-	for m in $(sed 's:.*/::' modules.order | sort | uniq -d)
+	for m in $(sed 's:.*/::' $1 | sort | uniq -d)
 	do
-		echo "warning: same module names found:" >&2
+		echo "error: the following would cause module name conflict:" >&2
 		sed -n "/\/$m/s:^:  :p" modules.order >&2
+		exit_code=1
 	done
 }
 
-check_same_name_modules
+check_same_name_modules "$1"
+
+exit $exit_code
Signature · 1 line
2.25.1
#1 lore raw

Hi,

Quoted 2 lines · Masahiro Yamada

I think all the warnings have been fixed by now. Make it a fatal error.

Not entirely. With this patch in the tree, I get:

Building sparc64:allmodconfig ... failed
--------------
Error log:
error: the following would cause module name conflict:
drivers/char/adi.ko drivers/input/joystick/adi.ko
make[1]: *** [modules_check] Error 1
make[1]: *** Waiting for unfinished jobs....
make: *** [__sub-make] Error 2

Reverting this patch fixes the problem.

Guenter

#2 lore raw
Quoted 18 lines · Guenter Roeck

Hi,

Quoted 2 lines · Masahiro Yamada

I think all the warnings have been fixed by now. Make it a fatal error.

Not entirely. With this patch in the tree, I get:

Building sparc64:allmodconfig ... failed
--------------
Error log:
error: the following would cause module name conflict:
drivers/char/adi.ko drivers/input/joystick/adi.ko
make[1]: *** [modules_check] Error 1
make[1]: *** Waiting for unfinished jobs....
make: *** [__sub-make] Error 2

Quoted 1 line

Reverting this patch fixes the problem.

As it doesn't look like either of these drivers can be "auto-loaded" based on hardware detection, I don't know what to suggest as for renaming either of them.

Any ideas?

thanks,

greg k-h

#3 lore raw
Quoted 20 lines · Greg Kroah-Hartman
Quoted 18 lines · Guenter Roeck

Hi,

Quoted 2 lines · Masahiro Yamada

I think all the warnings have been fixed by now. Make it a fatal error.

Not entirely. With this patch in the tree, I get:

Building sparc64:allmodconfig ... failed
--------------
Error log:
error: the following would cause module name conflict:
drivers/char/adi.ko drivers/input/joystick/adi.ko
make[1]: *** [modules_check] Error 1
make[1]: *** Waiting for unfinished jobs....
make: *** [__sub-make] Error 2

Quoted 7 lines
Quoted 1 line

Reverting this patch fixes the problem.

As it doesn't look like either of these drivers can be "auto-loaded" based on hardware detection, I don't know what to suggest as for renaming either of them.

Any ideas?

I see zero chance of a kernel actually needing to provide both drivers, given that the hardware is 20 years apart and gameports are almost exclusive to x86 PCs. How about an ugly hack:

drivers/input/joystick/Kconfig +1 -0
index 940b744639c7..6f73f02059b5 100644
@@ -45,6 +45,7 @@ config JOYSTICK_A3D
 config JOYSTICK_ADI
        tristate "Logitech ADI digital joysticks and gamepads"
        select GAMEPORT
+       depends on ADI!=m # avoid module name conflict
        help
          Say Y here if you have a Logitech controller using the ADI
          protocol over the PC gameport.

Arnd

#4 lore raw
Quoted 47 lines · Arnd Bergmann
Quoted 20 lines · Greg Kroah-Hartman
Quoted 18 lines · Guenter Roeck

Hi,

Quoted 2 lines · Masahiro Yamada

I think all the warnings have been fixed by now. Make it a fatal error.

Not entirely. With this patch in the tree, I get:

Building sparc64:allmodconfig ... failed
--------------
Error log:
error: the following would cause module name conflict:
drivers/char/adi.ko drivers/input/joystick/adi.ko
make[1]: *** [modules_check] Error 1
make[1]: *** Waiting for unfinished jobs....
make: *** [__sub-make] Error 2

Quoted 7 lines
Quoted 1 line

Reverting this patch fixes the problem.

As it doesn't look like either of these drivers can be "auto-loaded" based on hardware detection, I don't know what to suggest as for renaming either of them.

Any ideas?

I see zero chance of a kernel actually needing to provide both drivers, given that the hardware is 20 years apart and gameports are almost exclusive to x86 PCs. How about an ugly hack:

drivers/input/joystick/Kconfig +1 -0
index 940b744639c7..6f73f02059b5 100644
@@ -45,6 +45,7 @@ config JOYSTICK_A3D
 config JOYSTICK_ADI
        tristate "Logitech ADI digital joysticks and gamepads"
        select GAMEPORT
+       depends on ADI!=m # avoid module name conflict
        help
          Say Y here if you have a Logitech controller using the ADI
          protocol over the PC gameport.
Quoted 1 line

Arnd

Looks sane to me, thanks!

greg k-h

#5 lore raw
In reply to Arnd Bergmann
Quoted 48 lines · Arnd Bergmann
Quoted 20 lines · Greg Kroah-Hartman
Quoted 18 lines · Guenter Roeck

Hi,

Quoted 2 lines · Masahiro Yamada

I think all the warnings have been fixed by now. Make it a fatal error.

Not entirely. With this patch in the tree, I get:

Building sparc64:allmodconfig ... failed
--------------
Error log:
error: the following would cause module name conflict:
drivers/char/adi.ko drivers/input/joystick/adi.ko
make[1]: *** [modules_check] Error 1
make[1]: *** Waiting for unfinished jobs....
make: *** [__sub-make] Error 2

Quoted 7 lines
Quoted 1 line

Reverting this patch fixes the problem.

As it doesn't look like either of these drivers can be "auto-loaded" based on hardware detection, I don't know what to suggest as for renaming either of them.

Any ideas?

I see zero chance of a kernel actually needing to provide both drivers, given that the hardware is 20 years apart and gameports are almost exclusive to x86 PCs. How about an ugly hack:

drivers/input/joystick/Kconfig +1 -0
index 940b744639c7..6f73f02059b5 100644
@@ -45,6 +45,7 @@ config JOYSTICK_A3D
 config JOYSTICK_ADI
        tristate "Logitech ADI digital joysticks and gamepads"
        select GAMEPORT
+       depends on ADI!=m # avoid module name conflict
        help
          Say Y here if you have a Logitech controller using the ADI
          protocol over the PC gameport.
Quoted 1 line

Arnd

I am OK with this.
Thank you.

Signature · 2 lines
Best Regards
Masahiro Yamada