linux-kernel - Re: [PATCH v2 0/2] clocksource/meson6

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]

Message-ID: <5ff1a888-cce2-8fd8-a98e-14c3d7c33acf@linaro.org>
Date:   Sun, 18 Nov 2018 02:27:16 +0100
From:   Daniel Lezcano <daniel.lezcano@...aro.org>
To:     Martin Blumenstingl <martin.blumenstingl@...glemail.com>,
        linux-amlogic@...ts.infradead.org, linux-kernel@...r.kernel.org,
        tglx@...utronix.de
Cc:     linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH v2 0/2] clocksource/meson6_timer: implement ARM delay
 timer

On 15/11/2018 23:46, Martin Blumenstingl wrote:
> While trying to add support for the ARM TWD Timer and the ARM Global
> Timer on Meson8, Meson8b and Meson8m2 (ARM Cortex-A5 and Cortex-A9 SoCs)
> I did a review of the existing driver.
> Unfortunately I found it hard to review because the pre-processor
> #defines did not match the names from the public S805 datasheet. Thus
> patch #1 adjusts these. No functional changes here, this is just
> preparation work for patch #2.
> 
> Using the ARM Global Timer (drivers/clocksource/arm_global_timer.c)
> would have given us a timer-based delay implementation (so udelay() and
> friends would use the timer instead of using a loop-based delay
> implementation). Unfortunately we can't use the ARM Global Timer yet
> because it's input clock is derived from the CPU clock (which can change
> once we enable CPU frequency scaling on these SoCs, for which I will be
> sending patches in the near future).
> Amlogic's 3.10 kernel uses Timer E as delay timer which (with the
> current configuration) has a resolution of 1us. So patch #2 uses
> register_current_timer_delay() to register Timer E as ARM delay timer
> (which will be especially useful as we have to use udelay() when
> changing the CPU clocks during DVFS).
> 
> 
> Changes since v1 at [0]:
> - convert the enums for the input clock (meson6_timera_input_clock and
>   meson6_timere_input_clock) to simple #defines as these are register
>   values and not something driver-internal. All other register values
>   are #defines so it makes sense that these are #defines as well.
> 
> 
> [0] https://patchwork.kernel.org/cover/10658591/
> 
> 
> Martin Blumenstingl (2):
>   clocksource: meson6_timer: use register names from the datasheet
>   clocksource: meson6_timer: implement ARM delay timer
> 
>  drivers/clocksource/meson6_timer.c | 128 +++++++++++++++++++----------
>  1 file changed, 85 insertions(+), 43 deletions(-)

Both applied, thanks!


-- 
 <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog