[1/7] mmc: sdio: store owner from modules with sdio_register_driver()

Commit Message

Krzysztof Kozlowski March 29, 2024, 5:24 p.m. UTC

Modules registering driver with sdio_register_driver() might
forget to set .owner field.  The field is used by some of other kernel
parts for reference counting (try_module_get()), so it is expected that
drivers will set it.

Solve the problem by moving this task away from the drivers to the core
code, just like we did for platform_driver in
commit 9447057eaff8 ("platform_device: use a macro instead of
platform_driver_register").

Since many drivers forget to set the .owner, this effectively will fix
them.  Examples of fixed drivers are: ath6kl, b43, btsdio.c, ks7010,
libertas, MediaTek WiFi drivers, Realtek WiFi drivers, rsi, siano,
wilc1000, wl1251 and more.

Signed-off-by: Krzysztof Kozlowski <[e-mail]>
---
 drivers/mmc/core/sdio_bus.c   | 9 ++++++---
 include/linux/mmc/sdio_func.h | 5 ++++-
 2 files changed, 10 insertions(+), 4 deletions(-)

Comments

[e-mail] March 29, 2024, 5:57 p.m. UTC | #1

This is automated email and please do not reply to this email!

Dear submitter,

Thank you for submitting the patches to the linux bluetooth mailing list.
This is a CI test results with your patch series:
PW Link:https://patchwork.kernel.org/project/bluetooth/list/?series=839853

---Test result---

Test Summary:
CheckPatch                    PASS      4.61 seconds
GitLint                       PASS      2.29 seconds
SubjectPrefix                 FAIL      1.06 seconds
BuildKernel                   PASS      29.05 seconds
CheckAllWarning               PASS      33.45 seconds
CheckSparse                   PASS      38.27 seconds
CheckSmatch                   FAIL      34.67 seconds
BuildKernel32                 PASS      28.48 seconds
TestRunnerSetup               PASS      514.63 seconds
TestRunner_l2cap-tester       PASS      17.94 seconds
TestRunner_iso-tester         PASS      29.71 seconds
TestRunner_bnep-tester        PASS      5.63 seconds
TestRunner_mgmt-tester        PASS      109.04 seconds
TestRunner_rfcomm-tester      PASS      7.30 seconds
TestRunner_sco-tester         PASS      14.92 seconds
TestRunner_ioctl-tester       PASS      7.76 seconds
TestRunner_mesh-tester        PASS      5.83 seconds
TestRunner_smp-tester         PASS      6.77 seconds
TestRunner_userchan-tester    PASS      4.96 seconds
IncrementalBuild              PASS      53.91 seconds

Details
##############################
Test: SubjectPrefix - FAIL
Desc: Check subject contains "Bluetooth" prefix
Output:
"Bluetooth: " prefix is not specified in the subject
"Bluetooth: " prefix is not specified in the subject
"Bluetooth: " prefix is not specified in the subject
"Bluetooth: " prefix is not specified in the subject
"Bluetooth: " prefix is not specified in the subject
"Bluetooth: " prefix is not specified in the subject
"Bluetooth: " prefix is not specified in the subject
##############################
Test: CheckSmatch - FAIL
Desc: Run smatch tool with source
Output:

Segmentation fault (core dumped)
make[4]: *** [scripts/Makefile.build:244: net/bluetooth/hci_core.o] Error 139
make[4]: *** Deleting file 'net/bluetooth/hci_core.o'
make[3]: *** [scripts/Makefile.build:485: net/bluetooth] Error 2
make[2]: *** [scripts/Makefile.build:485: net] Error 2
make[2]: *** Waiting for unfinished jobs....
Segmentation fault (core dumped)
make[4]: *** [scripts/Makefile.build:244: drivers/bluetooth/bcm203x.o] Error 139
make[4]: *** Deleting file 'drivers/bluetooth/bcm203x.o'
make[4]: *** Waiting for unfinished jobs....
make[3]: *** [scripts/Makefile.build:485: drivers/bluetooth] Error 2
make[2]: *** [scripts/Makefile.build:485: drivers] Error 2
make[1]: *** [/github/workspace/src/src/Makefile:1919: .] Error 2
make: *** [Makefile:240: __sub-make] Error 2


---
Regards,
Linux Bluetooth

Francesco Dolcini March 31, 2024, 8:37 p.m. UTC | #2

Hello Krzysztof,

On Fri, Mar 29, 2024 at 06:24:31PM +0100, Krzysztof Kozlowski wrote:
> Modules registering driver with sdio_register_driver() might
> forget to set .owner field.  The field is used by some of other kernel
                             ^^ double space here

> parts for reference counting (try_module_get()), so it is expected that
> drivers will set it.
> 
> Solve the problem by moving this task away from the drivers to the core
> code, just like we did for platform_driver in
> commit 9447057eaff8 ("platform_device: use a macro instead of
> platform_driver_register").
> 
> Since many drivers forget to set the .owner, this effectively will fix
> them.  Examples of fixed drivers are: ath6kl, b43, btsdio.c, ks7010,
       ^^ and here

> libertas, MediaTek WiFi drivers, Realtek WiFi drivers, rsi, siano,
> wilc1000, wl1251 and more.
> 
> Signed-off-by: Krzysztof Kozlowski <[e-mail]>

My nit comments are irrelevant, but given you did the same twice I felt
like letting you know.

Reviewed-by: Francesco Dolcini <[e-mail]>

Krzysztof Kozlowski April 1, 2024, 9:44 a.m. UTC | #3

On 31/03/2024 22:37, Francesco Dolcini wrote:
> Hello Krzysztof,
> 
> On Fri, Mar 29, 2024 at 06:24:31PM +0100, Krzysztof Kozlowski wrote:
>> Modules registering driver with sdio_register_driver() might
>> forget to set .owner field.  The field is used by some of other kernel
>                              ^^ double space here
> 
>> parts for reference counting (try_module_get()), so it is expected that
>> drivers will set it.
>>
>> Solve the problem by moving this task away from the drivers to the core
>> code, just like we did for platform_driver in
>> commit 9447057eaff8 ("platform_device: use a macro instead of
>> platform_driver_register").
>>
>> Since many drivers forget to set the .owner, this effectively will fix
>> them.  Examples of fixed drivers are: ath6kl, b43, btsdio.c, ks7010,
>        ^^ and here

Which is correct :). Double space is on purpose.

> 
>> libertas, MediaTek WiFi drivers, Realtek WiFi drivers, rsi, siano,
>> wilc1000, wl1251 and more.
>>
>> Signed-off-by: Krzysztof Kozlowski <[e-mail]>
> 
> My nit comments are irrelevant, but given you did the same twice I felt
> like letting you know.
> 
> Reviewed-by: Francesco Dolcini <[e-mail]>
> 

Best regards,
Krzysztof

Patch

diff --git a/drivers/mmc/core/sdio_bus.c b/drivers/mmc/core/sdio_bus.c
index 71d885fbc228..c5fdfe2325f8 100644
--- a/drivers/mmc/core/sdio_bus.c
+++ b/drivers/mmc/core/sdio_bus.c
@@ -265,16 +265,19 @@  void sdio_unregister_bus(void)
 }
 
 /**
- *	sdio_register_driver - register a function driver
+ *	__sdio_register_driver - register a function driver
  *	@drv: SDIO function driver
+ *	@owner: owning module/driver
  */
-int sdio_register_driver(struct sdio_driver *drv)
+int __sdio_register_driver(struct sdio_driver *drv, struct module *owner)
 {
 	drv->drv.name = drv->name;
 	drv->drv.bus = &sdio_bus_type;
+	drv->drv.owner = owner;
+
 	return driver_register(&drv->drv);
 }
-EXPORT_SYMBOL_GPL(sdio_register_driver);
+EXPORT_SYMBOL_GPL(__sdio_register_driver);
 
 /**
  *	sdio_unregister_driver - unregister a function driver
diff --git a/include/linux/mmc/sdio_func.h b/include/linux/mmc/sdio_func.h
index 478855b8e406..fed1f5f4a8d3 100644
--- a/include/linux/mmc/sdio_func.h
+++ b/include/linux/mmc/sdio_func.h
@@ -106,7 +106,10 @@  struct sdio_driver {
 	.class = (dev_class), \
 	.vendor = SDIO_ANY_ID, .device = SDIO_ANY_ID
 
-extern int sdio_register_driver(struct sdio_driver *);
+/* use a macro to avoid include chaining to get THIS_MODULE */
+#define sdio_register_driver(drv) \
+	__sdio_register_driver(drv, THIS_MODULE)
+extern int __sdio_register_driver(struct sdio_driver *, struct module *);
 extern void sdio_unregister_driver(struct sdio_driver *);
 
 /**