linux-kernel - Re: [PATCH] usb: typec: tipd: improve handling of failures in interrupt handlers

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] [day] [month] [year] [list]

Message-ID: <YmGFKdVAuG5JEfzg@kroah.com>
Date:   Thu, 21 Apr 2022 18:24:09 +0200
From:   Greg KH <gregkh@...uxfoundation.org>
To:     Tom Rix <trix@...hat.com>
Cc:     heikki.krogerus@...ux.intel.com, nathan@...nel.org,
        ndesaulniers@...gle.com, sven@...npeter.dev, alyssa@...enzweig.io,
        marcan@...can.st, martink@...teo.de, saranya.gopal@...el.com,
        linux-usb@...r.kernel.org, linux-kernel@...r.kernel.org,
        llvm@...ts.linux.dev
Subject: Re: [PATCH] usb: typec: tipd: improve handling of failures in
 interrupt handlers

On Sat, Apr 09, 2022 at 09:00:13AM -0400, Tom Rix wrote:
> clang static analysis reports this representative issue
> core.c:516:6: warning: Branch condition evaluates
>   to a garbage value
>   if (event)
>       ^~~~~
> 
> In cd321x_interrupt(), a successful call to
> tps6598x_read64() is the only way event is set,
> and if a failure happens the irq should not be
> reported as handled.

Please use the full 72 columns.

> 
> Instead of initializing event, rework the
> usage of ret by initializing it to IRQ_NONE
> and then setting it when event is known to
> be not zero.  This removes the if-statement
> before the return.

So the code today is correct, but clang is wrong?  We don't need to do
anything then...

> tps6598x_interrupt() is similar.

This line makes no sense, sorry.

greg k-h