[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <861s74s6dc.wl-marc.zyngier@arm.com>
Date: Thu, 29 Nov 2018 07:14:07 +0000
From: Marc Zyngier <marc.zyngier@....com>
To: Bjorn Andersson <bjorn.andersson@...aro.org>
Cc: Srinivas Kandagatla <srinivas.kandagatla@...aro.org>,
sudeep.holla@....com, tglx@...utronix.de, jason@...edaemon.net,
linux-kernel@...r.kernel.org, linux-arm-msm@...r.kernel.org,
rnayak@...eaurora.org, sboyd@...nel.org, srini@...nel.org,
nicolas.dechesne@...aro.org, ctatlor97@...il.com, vkoul@...nel.org
Subject: Re: [RFC PATCH v3 3/4] irqchip: gic-v3: Add quirk for msm8996 secured registers
Bjorn,
On Wed, 28 Nov 2018 23:33:27 +0000,
Bjorn Andersson <bjorn.andersson@...aro.org> wrote:
>
> On Mon 12 Nov 02:33 PST 2018, Srinivas Kandagatla wrote:
>
> > Access to GICR_WAKER is restricted on msm8996 SoC in Hypervisor.
> > Its been more than 2+ years of wait for this to be fixed, which has
> > no hopes to be fixed. This change was introduced for the "lead device"
> > on msm8996 platform. It looks like all publicly available msm8996 and
> > other qcom SoCs have this implementation.
> >
> > So add a quirk to not access this register on msm8996.
> >
> > With this quirk MSM8996 can at least boot out of mainline,
> > which can help community to work with boards based on MSM8996 and other
> > SoCs with have this restrictions. This Quirk is based on device tree
> > compatible string.
> >
> > Without this patch Qualcomm DB820c board reboots when GICR_WAKER
> > is accessed.
> >
>
> Hi Srinivas,
>
> Given that the change request introducing this restriction looks quite
> generic and that we've heard reports of other shipping platforms having
> the same restriction, could we make this slightly more generic?
s/restriction/gross bug/
> I.e. dropping the MSM8996 from the flag name and perhaps use a bool
> property in devicetree to toggle the flag, instead of adding compatibles
> as we figure out which Qualcomm platforms has this restriction.
If we are to introduce a property, then it ought to be an erratum
number (similar to what we do for the arch timer). Srinivas wasn't
able to obtain one despite the defect having been known for years, so
I guess this is out.
Thanks,
M.
--
Jazz is not dead, it just smell funny.
Powered by blists - more mailing lists