Re: [PATCH 1/2] debugfs: improve DEFINE_DEBUGFS_ATTRIBUTE for !CONFIG_DEBUGFS_FS

From: Nicolai Stange
Date: Thu Oct 13 2016 - 06:47:08 EST


Arnd Bergmann <arnd@xxxxxxxx> writes:

> On Thursday, October 13, 2016 11:59:54 AM CEST Nicolai Stange wrote:
>> >
>> > +ssize_t debugfs_attr_read(struct file *file, char __user *buf,
>> > + size_t len, loff_t *ppos);
>> > +ssize_t debugfs_attr_write(struct file *file, const char __user *buf,
>> > + size_t len, loff_t *ppos);
>> > +
>> > +#define DEFINE_DEBUGFS_ATTRIBUTE(__fops, __get, __set, __fmt) \
>> > +static int __fops ## _open(struct inode *inode, struct file *file) \
>> > +{ \
>> > + __simple_attr_check_format(__fmt, 0ull); \
>> > + return simple_attr_open(inode, file, __get, __set, __fmt); \
>> > +} \
>> > +static const struct file_operations __fops = { \
>> > + .owner = THIS_MODULE, \
>> > + .open = __fops ## _open, \
>> > + .release = simple_attr_release, \
>> > + .read = debugfs_attr_read, \
>> > + .write = debugfs_attr_write, \
>>
>> This depends on GCC dead code elimination to always work for this
>> situation, otherwise we'd get undefined references to
>> debugfs_attr_read/write(), right?
>
> Correct.
>
>> In order to avoid having to test your patch against all those older
>> versions of GCC, can we have a safety net here and define some dummy
>> debugfs_attr_read/write() for the !CONFIG_DEBUGFS case?
>
> The question of dead-code elimination in older gcc versions comes up
> occasionally, and I think all versions that are able to build the
> kernel these days get this right all the time, otherwise any code
> using IS_ENABLED() helpers to control the calling of external interfaces
> would be broken.
>
> We could probably use that macro here if you think that's better
> and do:
>
> static const struct file_operations __fops = {
> .owner = THIS_MODULE,
> .open = IS_ENABLED(CONFIG_DEBUGFS_FS) ? __fops ## _open : NULL,
> ...
>
>> If nothing else, it would IMHO make the !CONFIG_DEBUGFS case more
>> understandable because one had not to figure out that this actually
>> relies on dead code elimination to work.
>
> Sure, that's fine. Can you do the new version of that patch with
> the change then?

I'd be happy to (won't be able to do this before tomorrow though).

Thanks,

Nicolai