Re: [PATCH v2 1/7] xfs: Add xfs_is_{i,io,mmap}locked functions

[Christoph Hellwig <hch@infradead.org>]

> +static inline bool
> +__xfs_is_ilocked(
> +	struct rw_semaphore	*rwsem,
> +	bool			shared,
> +	bool			excl)

This calling conventions seems odd.  In other places like
lockdep we just have a bool excl.  This means we might get a false
positive when the lock is held exclusive but only shared is asserted,
but given that the low-level helpers can't give better information
that isn't going to hurt.

Also I'd name this function xfs_rwsem_is_locked, as there is nothing
inode specific here.

I also agree that this function needs good comments explaining the
rationale.

> +bool
> +xfs_is_ilocked(
> +	struct xfs_inode	*ip,
> +	int			lock_flags)
> +{
> +	return __xfs_is_ilocked(&ip->i_lock.mr_lock,
> +			(lock_flags & XFS_ILOCK_SHARED),
> +			(lock_flags & XFS_ILOCK_EXCL));
> +}
> +
> +bool
> +xfs_is_mmaplocked(
> +	struct xfs_inode	*ip,
> +	int			lock_flags)
> +{
> +	return __xfs_is_ilocked(&ip->i_mmaplock.mr_lock,
> +			(lock_flags & XFS_MMAPLOCK_SHARED),
> +			(lock_flags & XFS_MMAPLOCK_EXCL));
> +}
> +
> +bool
> +xfs_is_iolocked(
> +	struct xfs_inode	*ip,
> +	int			lock_flags)
> +{
> +	return __xfs_is_ilocked(&VFS_I(ip)->i_rwsem,
> +			(lock_flags & XFS_IOLOCK_SHARED),
> +			(lock_flags & XFS_IOLOCK_EXCL));
> +}
>  #endif

What is the benefit of these helpers?  xfs_isilocked can just
call __xfs_is_ilocked / xfs_rwsem_is_locked directly.