[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <3401475f-bdbd-4b3f-8961-d3d747dc82f1@t-8ch.de>
Date: Sun, 26 Jan 2025 18:20:42 +0100
From: Thomas Weißschuh <linux@...ssschuh.net>
To: Arnd Bergmann <arnd@...db.de>
Cc: Anna-Maria Gleixner <anna-maria@...utronix.de>,
Frederic Weisbecker <frederic@...nel.org>, Thomas Gleixner <tglx@...utronix.de>,
linux-kernel@...r.kernel.org, Richard Cochran <richardcochran@...il.com>,
Cyrill Gorcunov <gorcunov@...il.com>
Subject: Re: [PATCH] posix-clock: Remove duplicate ioctl handler
Hi Arnd,
On 2025-01-25 15:28:29+0100, Arnd Bergmann wrote:
> On Sat, Jan 25, 2025, at 11:20, Thomas Weißschuh wrote:
> > static int posix_clock_open(struct inode *inode, struct file *fp)
> > {
> > int err;
> > @@ -174,7 +154,7 @@ static const struct file_operations
> > posix_clock_file_operations = {
> > .open = posix_clock_open,
> > .release = posix_clock_release,
> > #ifdef CONFIG_COMPAT
> > - .compat_ioctl = posix_clock_compat_ioctl,
> > + .compat_ioctl = posix_clock_ioctl,
> > #endif
>
> I would remove the #ifdef/#endif as well here. I would also suggest
> using compat_ptr_ioctl(), so on s390 the pointers get converted
> correctly, even if it clears the upper bit on the integer arguments.
Using compat_ptr_ioctl() here will break the logic from the closely
related fix to the PTP ioctl handler [0].
There is no other posix_clock_file_operations implementer than PTP.
I can drop the #ifdef.
Thomas
[0] https://lore.kernel.org/lkml/20250125-posix-clock-compat_ioctl-v2-1-11c865c500eb@weissschuh.net/
Powered by blists - more mailing lists