Re: [PATCH v3 5/6] btrfs: implement shutdown ioctl

[Qu Wenruo <quwenruo.btrfs@gmx.com>]

在 2025/7/4 15:55, Johannes Thumshirn 写道:
> On 04.07.25 01:43, Qu Wenruo wrote:
>>    long btrfs_ioctl(struct file *file, unsigned int
>>    		cmd, unsigned long arg)
>>    {
>> @@ -5201,6 +5231,8 @@ long btrfs_ioctl(struct file *file, unsigned int
>>    	struct btrfs_fs_info *fs_info = inode_to_fs_info(inode);
>>    	struct btrfs_root *root = BTRFS_I(inode)->root;
>>    	void __user *argp = (void __user *)arg;
>> +	/* If @arg is just an unsigned long value. */
>> +	unsigned long flags;
>>    
>>    	switch (cmd) {
>>    	case FS_IOC_GETVERSION:
>> @@ -5349,6 +5381,14 @@ long btrfs_ioctl(struct file *file, unsigned int
>>    #endif
>>    	case BTRFS_IOC_SUBVOL_SYNC_WAIT:
>>    		return btrfs_ioctl_subvol_sync(fs_info, argp);
>> +#ifdef CONFIG_BTRFS_EXPERIMENTAL
>> +	case BTRFS_IOC_SHUTDOWN:
>> +		if (!capable(CAP_SYS_ADMIN))
>> +			return -EPERM;
>> +		if (get_user(flags, (__u32 __user *)arg))
>> +			return -EFAULT;
>> +		return btrfs_emergency_shutdown(fs_info, flags);
>> +#endif
> 
> With that you'll get buildbot complaints if CONFIG_BTRFS_EXPERIMENTAL=n
> because flags is unused.
> 
> I'd probably put the get_user(flags, ...) into
> btrfs_emergency_shutdown() to silence the compiler.

That sounds a lot better than all my alternatives (a dedicated {} block 
to put @flags into, get_user() at the beginning of btrfs_ioctl())

Will go your path when merging the patches.

Thanks,
Qu