linux-kernel - Re: [PATCH 0/6] 32-bit Meson: add the ARM TWD and Global Timers

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:   Sun, 2 Dec 2018 22:12:21 +0100
From:   Martin Blumenstingl <martin.blumenstingl@...glemail.com>
To:     khilman@...libre.com
Cc:     linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
        linux-amlogic@...ts.infradead.org, carlo@...one.org
Subject: Re: [PATCH 0/6] 32-bit Meson: add the ARM TWD and Global Timers

Hi Kevin,

On Fri, Nov 23, 2018 at 8:53 PM Martin Blumenstingl
<martin.blumenstingl@...glemail.com> wrote:
>
> The 32-bit Meson SoCs use Cortex-A9 or Cortex-A5 cores. These come
> with the ARM TWD ("Timer Watchdog") which contains a timer and a
> watchdog as well as the ARM Global Timer.
>
> This enables the corresponding configs for the 32-bit Meson target.
> Additionally this adds and enables the ARM TWD timer. The Global
> Timer is added but currently disabled because it's clock input is
> the PERIPH clock which is derived from the CPU clock. Thus the rate
> of the PERIPH clock will change when changing the CPU frequency.
> Unfortunately the Global Timer driver doesn't handle clocks with
> changing rates yet (unlike the TWD timer), thus we keep it disabled
> for now.
>
> The whole series is inspired by an almost 3 year old patch from
> Carlo: [0]
>
>
> Dependencies:
> - I build this on top of my other series "ARM: dts: meson: add the
>   timer interrupts and clocks" from [1]
this is already merged into your v4.21/dt branch

> - CLKID_PERIPH requires updated clock driver headers. Neil provided
>   a tag which includes the updated headers: [2]
this is still a dependency which you could easily pull in

> - There is no runtime dependency on the PERIPH clock as we don't
>   have CPU frequency scaling support enabled yet. In case the TWD
>   timer driver can't find the clock it falls back to auto-detecting
>   the clock rate at boot time. This is safe as long as we don't have
>   .dts patches in place which allow changing the CPU clock rate. Once
>   we enable CPU frequency scaling support for the PERIPH clock becomes
>   mandatory so the TWD timer driver knows about changes to the PERIPH
>   clock (which is derived from the CPU clock).
and there's still not a hard runtime dependency until you apply [3]
"ARM: dts: enable CPU frequency scaling on Meson8/Meson8b"

> Martin Blumenstingl (6):
>   ARM: meson: select HAVE_ARM_TWD and ARM_GLOBAL_TIMER
>   ARM: dts: meson: group the Cortex-A5 / Cortex-A9 peripherals
>   ARM: dts: meson8: add the ARM TWD timer
>   ARM: dts: meson8: add the Cortex-A9 global timer
>   ARM: dts: meson8b: add the ARM TWD timer
>   ARM: dts: meson8b: add the Cortex-A5 global timer
>
>  arch/arm/boot/dts/meson.dtsi   | 24 ++++++++++++++++--------
>  arch/arm/boot/dts/meson8.dtsi  | 32 +++++++++++++++++++++++++++-----
>  arch/arm/boot/dts/meson8b.dtsi | 32 +++++++++++++++++++++++++++-----
>  arch/arm/mach-meson/Kconfig    |  2 ++
>  4 files changed, 72 insertions(+), 18 deletions(-)
if you plan to send another pull-request to the arm-soc tree then
please consider including this series.
it fixes some harmless (but still noisy) warnings during boot which
also also seen by Odroid-C1 in your KernelCI lab:
  Clockevents: could not switch to one-shot mode:
   dummy_timer is not functional.
  Clockevents: could not switch to one-shot mode:
   dummy_timer is not functional.
  Clockevents: could not switch to one-shot mode:
   dummy_timer is not functional.
  Clockevents: could not switch to one-shot mode:
   dummy_timer is not functional.
  Could not switch to high resolution mode on CPU 3
  Could not switch to high resolution mode on CPU 2
  Could not switch to high resolution mode on CPU 0
  Could not switch to high resolution mode on CPU 1


Regards
Martin


[0] https://patchwork.kernel.org/patch/7797581/
[1] https://patchwork.kernel.org/cover/10687005/
[2] http://lists.infradead.org/pipermail/linux-amlogic/2018-November/009136.html
[3] https://patchwork.kernel.org/cover/10705475/