linux-kernel - Re: [PATCH v3 2/4] bus: fsl-mc: add fsl-mc userspace support

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]

Date:   Tue, 20 Nov 2018 17:58:07 +0100
From:   "gregkh@...uxfoundation.org" <gregkh@...uxfoundation.org>
To:     Ioana Ciornei <ioana.ciornei@....com>
Cc:     Laurentiu Tudor <laurentiu.tudor@....com>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "netdev-owner@...r.kernel.org" <netdev-owner@...r.kernel.org>,
        Ioana Ciocoi Radulescu <ruxandra.radulescu@....com>,
        Horia Geanta <horia.geanta@....com>,
        Leo Li <leoyang.li@....com>
Subject: Re: [PATCH v3 2/4] bus: fsl-mc: add fsl-mc userspace support

On Tue, Nov 20, 2018 at 04:51:07PM +0000, Ioana Ciornei wrote:
> > On Tue, Nov 20, 2018 at 03:39:45PM +0000, Ioana Ciornei wrote:
> > > +
> > > +			error = fsl_mc_uapi_create_device_file(mc_bus);
> > > +			if (error < 0) {
> > > +				error = -EPROBE_DEFER;
> > > +				goto error_cleanup_msi_domain;
> > > +			}
> > 
> > What error could occur here that would be somehow fixed if you defer and try
> > this all again?  Shouldn't you just return the error given (as something really went
> > wrong here, it's not a "something else needs to start up first issue), and abort?
> > 
> 
> It's really a "something else needs to start up first" type of issue. The fsl_mc_uapi_create_device_file fails only if the misc_register fails and this could really happen when the misc char driver is not yet initialized.
> 
> Initially, I have used the register_chrdev API and you suggested to use a misc device and just defer probe if it's not yet initialized.

Ok, fair enough, looks good to me, thanks for answering.

greg k-h