Re: [PATCH v3 1/2] usb: gadget: goku_udc: move debug output to debugfs

[Cheng Lingfei <chenglingfei@foxmail.com>]

On 8/26/2026 5:27 PM, Greg Kroah-Hartman wrote:
> On Wed, Aug 26, 2026 at 05:15:06PM +0800, Cheng Lingfei wrote:
>> goku_udc exposes diagnostic state through /proc/driver/udc. The single
>> global proc entry collides when more than one controller is probed and
>> triggers a proc registration warning.
>>
>> The data is intended only for debugging and is not a userspace ABI. Move it
>> under the USB debugfs root and use the PCI device name for a per-device
>> directory. Create the file only after the UDC has been registered.
>>
>> Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
>> Link: https://lore.kernel.org/2026082616-glue-atlas-0cfd@gregkh
>> Assisted-by: Codex:gpt-5.6
>> Signed-off-by: Cheng Lingfei <chenglingfei@foxmail.com>
>> ---
>>   drivers/usb/gadget/udc/goku_udc.c | 47 +++++++++++++++++++++++++++------------
>>   drivers/usb/gadget/udc/goku_udc.h |  5 ++++-
>>   2 files changed, 37 insertions(+), 15 deletions(-)
>>
>> diff --git a/drivers/usb/gadget/udc/goku_udc.c b/drivers/usb/gadget/udc/goku_udc.c
>> index ac2a984c2f87..5ad8633f522b 100644
>> --- a/drivers/usb/gadget/udc/goku_udc.c
>> +++ b/drivers/usb/gadget/udc/goku_udc.c
>> @@ -20,6 +20,7 @@
>>   // #define	VERBOSE		/* extra debug messages (success too) */
>>   // #define	USB_TRACE	/* packet-level success messages */
>>   
>> +#include <linux/debugfs.h>
>>   #include <linux/kernel.h>
>>   #include <linux/module.h>
>>   #include <linux/pci.h>
>> @@ -30,9 +31,9 @@
>>   #include <linux/timer.h>
>>   #include <linux/list.h>
>>   #include <linux/interrupt.h>
>> -#include <linux/proc_fs.h>
>>   #include <linux/seq_file.h>
>>   #include <linux/device.h>
>> +#include <linux/usb.h>
>>   #include <linux/usb/ch9.h>
>>   #include <linux/usb/gadget.h>
>>   #include <linux/prefetch.h>
>> @@ -1050,9 +1051,7 @@ static inline const char *dmastr(void)
>>   		return "(dma IN)";
>>   }
>>   
>> -#ifdef CONFIG_USB_GADGET_DEBUG_FILES
>> -
>> -static const char proc_node_name [] = "driver/udc";
>> +#ifdef CONFIG_USB_GADGET_DEBUG_FS
> 
> Why do you need this config option check anymore?

Thank you for your review. This driver follows the existing UDC driver 
pattern (e.g., gr_udc.c), but the additional config option check is 
unnecessary here. The debugfs subsystem already provides stub functions 
when CONFIG_DEBUG_FS is disabled: debugfs_create_dir() returns 
ERR_PTR(-ENODEV), and the removal helpers are safe no-ops.

I'll remove the CONFIG_USB_GADGET_DEBUG_FS conditional from this driver 
in the next revision.

I also noticed that several other UDC drivers still rely on this Kconfig 
option. Some of those checks may be redundant for the same reason, 
though in certain drivers the option also guards debug-only state or 
code. Would it be worthwhile to audit the remaining users in a separate 
follow-up series and, if none require compile-time gating, consider 
removing CONFIG_USB_GADGET_DEBUG_FS entirely?

-- 
Best regards,
Cheng Lingfei