linux-kernel - Re: [PATCH] firmware: qcom_scm: disable clocks if qcom_scm_bw

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: Sat, 16 Mar 2024 12:58:01 -0500
From: Bjorn Andersson <andersson@...nel.org>
To: Konrad Dybcio <konrad.dybcio@...aro.org>
Cc: Gabor Juhos <j4g8y7@...il.com>, Sibi Sankar <quic_sibis@...cinc.com>, 
	linux-arm-msm@...r.kernel.org, linux-kernel@...r.kernel.org, stable@...r.kernel.org
Subject: Re: [PATCH] firmware: qcom_scm: disable clocks if
 qcom_scm_bw_enable() fails

On Wed, Mar 06, 2024 at 05:02:37PM +0100, Konrad Dybcio wrote:
> 
> 
> On 3/6/24 05:10, Elliot Berman wrote:
> > On Tue, Mar 05, 2024 at 10:15:19PM +0100, Konrad Dybcio wrote:
> > > 
> > > 
> > > On 3/4/24 14:14, Gabor Juhos wrote:
> > > > There are several functions which are calling qcom_scm_bw_enable()
> > > > then returns immediately if the call fails and leaves the clocks
> > > > enabled.
> > > > 
> > > > Change the code of these functions to disable clocks when the
> > > > qcom_scm_bw_enable() call fails. This also fixes a possible dma
> > > > buffer leak in the qcom_scm_pas_init_image() function.
> > > > 
> > > > Compile tested only due to lack of hardware with interconnect
> > > > support.
> > > > 
> > > > Cc: stable@...r.kernel.org
> > > > Fixes: 65b7ebda5028 ("firmware: qcom_scm: Add bw voting support to the SCM interface")
> > > > Signed-off-by: Gabor Juhos <j4g8y7@...il.com>
> > > > ---
> > > 
> > > Taking a closer look, is there any argument against simply
> > > putting the clk/bw en/dis calls in qcom_scm_call()?
> > 
> > We shouldn't do this because the clk/bw en/dis calls are only needed in
> > few SCM calls.
> 
> Then the argument list could be expanded with `bool require_resources`,
> or so still saving us a lot of boilerplate
> 

I don't think there's reason for making this more general, because I
think this is a problem specific to PAS - much related to Bartosz
special handling of shmbridge for these calls.

It would be very nice if someone could help document why this is.

Regards,
Bjorn