linux-kernel - Re: [Patch v5 5/8] firmware: qcom: scm: Convert to streaming DMA APIS

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, 24 May 2016 22:37:31 -0500
From:	Andy Gross <andy.gross@...aro.org>
To:	Kevin Hilman <khilman@...nel.org>
Cc:	linux-arm-msm <linux-arm-msm@...r.kernel.org>,
	linux-arm-kernel <linux-arm-kernel@...ts.infradead.org>,
	lkml <linux-kernel@...r.kernel.org>,
	Bjorn Andersson <bjorn.andersson@...aro.org>,
	Stephen Boyd <sboyd@...eaurora.org>,
	devicetree <devicetree@...r.kernel.org>,
	jilai wang <jilaiw@...eaurora.org>
Subject: Re: [Patch v5 5/8] firmware: qcom: scm: Convert to streaming DMA APIS

On Mon, May 23, 2016 at 04:02:06PM -0500, Andy Gross wrote:
> On 23 May 2016 at 14:26, Kevin Hilman <khilman@...nel.org> wrote:
> > Hi Andy,
> >
> > On Thu, May 12, 2016 at 8:46 PM, Andy Gross <andy.gross@...aro.org> wrote:
> >> This patch converts the Qualcomm SCM driver to use the streaming DMA APIs
> >> for communication buffers.
> >>
> >> Signed-off-by: Andy Gross <andy.gross@...aro.org>
> >
> > This patch has landed in linux-next in the form of commit a551c3dbd689
> > (firmware: qcom: scm: Convert to streaming DMA APIS), and kernelci.org
> > found some boot breakage in next-20160523 on apq8064[1] which was
> > bisected down to this commit.
> >
> > I reverted this commit on top of next-20160523 and it no longer
> > builds, so I didn't validate if things boot again with this patch
> > reverted.
> 
> Ouch I missed this failure.  I'll investigate and get it fixed.

So the root cause was the fact that the DFAB clock required by the SCM is an RPM
clock.  That support isn't present yet in the kernel, so SCM probe fails.

The core clock is really only accessed so that we can bump the clock on it up to
the max for performance.  As such, I'll make it optional in the platform code.

This does bring up the issue of probe defer causing issues with the spm driver,
as it calls set_warm_boot_addr.  That may have to be addressed, but is probably
a problem best fixed in the spm.

Regards,

Andy