linux-kernel - Re: [PATCH V3 04/12] misc: xilinx_sdfec: Add open, close and ioctl

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 for Android: free password hash cracker in your pocket

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

Date:   Sat, 4 May 2019 16:41:28 +0200
From:   Greg KH <gregkh@...uxfoundation.org>
To:     Arnd Bergmann <arnd@...db.de>
Cc:     Dragan Cvetic <dragan.cvetic@...inx.com>,
        Michal Simek <michal.simek@...inx.com>,
        Linux ARM <linux-arm-kernel@...ts.infradead.org>,
        Rob Herring <robh+dt@...nel.org>,
        Mark Rutland <mark.rutland@....com>,
        DTML <devicetree@...r.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Derek Kiernan <derek.kiernan@...inx.com>
Subject: Re: [PATCH V3 04/12] misc: xilinx_sdfec: Add open, close and ioctl

On Sat, May 04, 2019 at 10:35:02AM -0400, Arnd Bergmann wrote:
> On Thu, May 2, 2019 at 1:23 PM Greg KH <gregkh@...uxfoundation.org> wrote:
> >
> > On Sat, Apr 27, 2019 at 11:04:58PM +0100, Dragan Cvetic wrote:
> > > Add char device interface per DT node present and support
> > > file operations:
> > > - open(),
> > > - close(),
> > > - unlocked_ioctl(),
> > > - compat_ioctl().
> >
> > Why do you need compat_ioctl() at all?  Any "new" driver should never
> > need it.  Just create your structures properly.
> 
> The function he added was the version that is needed when the structures
> are compatible. I submitted a series to add a generic 'compat_ptr_ioctl'
> implementation that would save a few lines here doing the same thing,
> but it's not merged yet.
> 
> Generally speaking, every driver that has a .ioctl() function should also
> have a .compat_ioctl(), and ideally it should be exactly this trivial
> version.

Ok, for some reason I thought if there was no need for a compat ioctl
(i.e. no pointer mess), then no need for a callback at all.

thanks,

greg k-h