linux-kernel - Re: [PATCH v6 01/12] driver: platform: Add helper for safer setting of driver

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:   Mon, 4 Apr 2022 13:14:22 +0300
From:   Andy Shevchenko <andy.shevchenko@...il.com>
To:     Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>
Cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        "Rafael J. Wysocki" <rafael@...nel.org>,
        Stuart Yoder <stuyoder@...il.com>,
        "K. Y. Srinivasan" <kys@...rosoft.com>,
        Haiyang Zhang <haiyangz@...rosoft.com>,
        Stephen Hemminger <sthemmin@...rosoft.com>,
        Wei Liu <wei.liu@...nel.org>, Dexuan Cui <decui@...rosoft.com>,
        Bjorn Helgaas <bhelgaas@...gle.com>,
        Bjorn Andersson <bjorn.andersson@...aro.org>,
        Mathieu Poirier <mathieu.poirier@...aro.org>,
        Vineeth Vijayan <vneethv@...ux.ibm.com>,
        Peter Oberparleiter <oberpar@...ux.ibm.com>,
        Heiko Carstens <hca@...ux.ibm.com>,
        Vasily Gorbik <gor@...ux.ibm.com>,
        Alexander Gordeev <agordeev@...ux.ibm.com>,
        Christian Borntraeger <borntraeger@...ux.ibm.com>,
        Sven Schnelle <svens@...ux.ibm.com>,
        Andy Gross <agross@...nel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        linux-clk <linux-clk@...r.kernel.org>,
        NXP Linux Team <linux-imx@....com>,
        linux-arm Mailing List <linux-arm-kernel@...ts.infradead.org>,
        Linux on Hyper-V List <linux-hyperv@...r.kernel.org>,
        linux-pci <linux-pci@...r.kernel.org>,
        linux-remoteproc@...r.kernel.org, linux-s390@...r.kernel.org,
        linux-arm-msm@...r.kernel.org,
        ALSA Development Mailing List <alsa-devel@...a-project.org>,
        linux-spi <linux-spi@...r.kernel.org>,
        virtualization@...ts.linux-foundation.org,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Rasmus Villemoes <linux@...musvillemoes.dk>
Subject: Re: [PATCH v6 01/12] driver: platform: Add helper for safer setting
 of driver_override

On Mon, Apr 4, 2022 at 12:34 PM Krzysztof Kozlowski
<krzysztof.kozlowski@...aro.org> wrote:
> On 04/04/2022 11:16, Andy Shevchenko wrote:
> > On Sun, Apr 3, 2022 at 9:38 PM Krzysztof Kozlowski
> > <krzysztof.kozlowski@...aro.org> wrote:

...

> >> +int driver_set_override(struct device *dev, const char **override,
> >> +                       const char *s, size_t len)
> >> +{
> >> +       const char *new, *old;
> >> +       char *cp;
> >
> >> +       if (!override || !s)
> >> +               return -EINVAL;
> >
> > Still not sure if we should distinguish (s == NULL && len == 0) from
> > (s != NULL && len == 0).
> > Supplying the latter seems confusing (yes, I see that in the old code). Perhaps
> > !s test, in case you want to leave it, should be also commented.
>
> The old semantics were focused on sysfs usage, so clearing is by passing
> an empty string. In the case of sysfs empty string is actually "\n". I
> intend to keep the semantics also for the in-kernel usage and in such
> case empty string can be also "".
>
> If I understand your comment correctly, you propose to change it to NULL
> for in-kernel usage, but that would change the semantics.

Yes. It's also possible to have a wrapper for sysfs use.

> > Another approach is to split above to two checks and move !s after !len.
>
> I don't follow why... The !override and !s are invalid uses. !len is a
> valid user for internal callers, just like "\n" is for sysfs.

I understand but always supplying s maybe an overhead for in-kernel usages.

In any case, it's not critical right now, just a remark that it can be modified.

> >> +       /*
> >> +        * The stored value will be used in sysfs show callback (sysfs_emit()),
> >> +        * which has a length limit of PAGE_SIZE and adds a trailing newline.
> >> +        * Thus we can store one character less to avoid truncation during sysfs
> >> +        * show.
> >> +        */
> >> +       if (len >= (PAGE_SIZE - 1))
> >> +               return -EINVAL;
> >
> > Perhaps explain the case in the comment here?
>
> You mean the case we discuss here (to clear override with "")? Sure.

Yep. Before the below check.

> >> +       if (!len) {
> >> +               device_lock(dev);
> >> +               old = *override;
> >> +               *override = NULL;
> >
> >> +               device_unlock(dev);
> >> +               goto out_free;
> >
> > You may deduplicate this one, by
> >
> >                goto out_unlock_free;
> >
> > But I understand your intention to keep lock-unlock in one place, so
> > perhaps dropping that label would be even better in this case and
> > keeping it
>
> Yes, exactly.
>
> >        kfree(old);
> >        return 0;
> >
> > here instead of goto.
>
> Slightly more code, but indeed maybe easier to follow. I'll do like this.


-- 
With Best Regards,
Andy Shevchenko