linux-kernel - Re: [PATCH v6 1/7] kernfs: Introduce hashed mutexes to replace global kernfs_open_file

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <YgqWeZrK6XAiICmo@slm.duckdns.org>
Date:   Mon, 14 Feb 2022 07:50:49 -1000
From:   Tejun Heo <tj@...nel.org>
To:     Imran Khan <imran.f.khan@...cle.com>
Cc:     gregkh@...uxfoundation.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v6 1/7] kernfs: Introduce hashed mutexes to replace
 global kernfs_open_file_mutex.

Hello,

On Mon, Feb 14, 2022 at 11:03:16PM +1100, Imran Khan wrote:
> +extern struct kernfs_global_locks *kernfs_locks;
> +
> +static inline struct mutex *kernfs_open_file_mutex_ptr(struct kernfs_node *kn)
> +{
> +	int idx = hash_ptr(kn, NR_KERNFS_LOCK_BITS);
> +
> +	return &kernfs_locks->open_file_mutex[idx].lock;
> +}
> +
> +static inline struct mutex *kernfs_open_file_mutex_lock(struct kernfs_node *kn)
> +{
> +	struct mutex *lock;
> +
> +	lock = kernfs_open_file_mutex_ptr(kn);
> +
> +	mutex_lock(lock);
> +
> +	return lock;
> +}

So, splitting patches this way doesn't really help. Because this patch
introduces code which isn't used and the second patch does all the
meaningful changes. It'd be better if the first patch introduces the
interface without changing the actual locking - ie. introduce and convert to
use kernfs_open_file_mutex*() but make it return the same old global mutex,
and then the second patch adds the hashed locks and updates
kernfs_open_file_mutex*() to actually return hashed locks. This way, the
meaningful changes are split into two patches which can be verified
independently.

Thanks.

-- 
tejun