linux-kernel - Re: [PATCH v3 06/15] arm64: dts: qcom: sc8280xp: Fix the base addresses of LLCC banks

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]

Date:   Wed, 21 Dec 2022 11:25:26 +0530
From:   Manivannan Sadhasivam <manivannan.sadhasivam@...aro.org>
To:     Borislav Petkov <bp@...en8.de>
Cc:     Steev Klimaszewski <steev@...i.org>, andersson@...nel.org,
        robh+dt@...nel.org, krzysztof.kozlowski+dt@...aro.org,
        tony.luck@...el.com, quic_saipraka@...cinc.com,
        konrad.dybcio@...aro.org, linux-arm-msm@...r.kernel.org,
        linux-kernel@...r.kernel.org, james.morse@....com,
        mchehab@...nel.org, rric@...nel.org, linux-edac@...r.kernel.org,
        quic_ppareek@...cinc.com, luca.weiss@...rphone.com,
        ahalaney@...hat.com
Subject: Re: [PATCH v3 06/15] arm64: dts: qcom: sc8280xp: Fix the base
 addresses of LLCC banks

On Wed, Dec 21, 2022 at 12:55:03AM +0100, Borislav Petkov wrote:
> On Tue, Dec 20, 2022 at 03:22:07PM +0530, Manivannan Sadhasivam wrote:
> > This is a genuine use-after-free bug that happens because the edac core frees
> > the memory assigned to "llcc_driv_data" pointer that gets passed as "pvt_info".
> > 
> > Here, the LLCC driver is one creating the "qcom_llcc_edac" platform device and
> > also allocating memory for "llcc_driv_data". But since during qcom_edac driver
> > removal, we are just unregistering the driver and the platform device still
> > stays around, the edac driver is not supposed to free any memory associated
> > with the platform device.
> 
> If you mean
> 
> __edac_device_free_ctl_info()
> 
> it is very well supposed to free it as it allocates it in
> edac_device_alloc_ctl_info().
> 
> If qcom_llcc_edac_probe() simply goes and assigns something of its own
> to edev_ctl->pvt_info, then that driver gets to keep the pieces ofc.
> 

Right. It is the issue of the qcom driver from the start.

Thanks,
Mani

> -- 
> Regards/Gruss,
>     Boris.
> 
> https://people.kernel.org/tglx/notes-about-netiquette

-- 
மணிவண்ணன் சதாசிவம்