mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Zhongqiu Han <zhongqiu.han@oss.qualcomm.com>
To: Sumit Gupta <sumitg@nvidia.com>,
	rafael@kernel.org, viresh.kumar@linaro.org, robh@kernel.org,
	krzk+dt@kernel.org, conor+dt@kernel.org, treding@nvidia.com,
	jonathanh@nvidia.com, linux-pm@vger.kernel.org,
	linux-tegra@vger.kernel.org, linux-kernel@vger.kernel.org,
	devicetree@vger.kernel.org
Cc: bbasu@nvidia.com, Zhongqiu Han <zhongqiu.han@oss.qualcomm.com>
Subject: Re: [PATCH v2 8/9] cpufreq: tegra194: Add support for Tegra264
Date: Wed, 18 Nov 2026 11:47:27 +0800	[thread overview]
Message-ID: <6452ab41-2d7c-4995-9f6f-1e8207767cdb@oss.qualcomm.com> (raw)
In-Reply-To: <20261117154954.625126-9-sumitg@nvidia.com>

On 11/17/2026 11:49 PM, Sumit Gupta wrote:
> Add cpufreq support for Tegra264. It requests and reads the CPU
> frequency differently from earlier SoCs.
> 
> To request a frequency, the driver writes NDIV to a core_freq_req
> register in the CCPLEX register space. Two clusters share a clock and
> its register, so they always run at the same frequency. The driver
> exposes each pair as one performance domain and one cpufreq policy.
> Each CPU's "performance-domains" property gives the domain index, which
> selects the register. If the CCPLEX node declares
> "#performance-domain-cells", probe fails if any CPU lacks the property.
> 
> To read the frequency, the driver samples the per-CPU 64-bit AMU
> counters through system registers. The AMU constant counter is the
> reference. It ticks at the system counter rate, so the driver takes the
> reference rate from CNTFRQ instead of the fixed 408 MHz clock. Probe
> fails if CONFIG_ARM64_AMU_EXTN is disabled or a CPU lacks AMU.
> 
> Signed-off-by: Sumit Gupta <sumitg@nvidia.com>
> ---
>   drivers/cpufreq/Kconfig.arm        |   4 +-
>   drivers/cpufreq/tegra194-cpufreq.c | 161 ++++++++++++++++++++++++++++-
>   2 files changed, 158 insertions(+), 7 deletions(-)
> +}
> +
> +/*
> + * The AMU counters are per-CPU system registers, so this must run on the CPU
> + * it samples. tegra194_calculate_speed() queues it there with queue_work_on().
> + */
> +static void tegra264_sample_counters(struct tegra194_cpufreq_data *data,
> +				     struct tegra_cpu_ctr *c)
> +{
> +	c->coreclk_cnt = read_sysreg_s(SYS_AMEVCNTR0_CORE_EL0);
> +	c->refclk_cnt = read_sysreg_s(SYS_AMEVCNTR0_CONST_EL0);


The core cycle counter and the constant counter are read separately
here, with interrupts enabled. An interrupt, or a preemption after it,
between the two reads skews the pair, and the resulting error in the
computed frequency becomes more significant as the gap increases. How
about reading both with local_irq_save(), like arm64 does in
amu_read_core_const_ctrs()?





-- 
Thx and BRs,
Zhongqiu Han

  reply	other threads:[~2026-11-18  3:47 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-17 15:49 [PATCH v2 0/9] cpufreq: tegra194: add Tegra264 support Sumit Gupta
2026-11-17 15:49 ` [PATCH v2 1/9] cpufreq: tegra194: move MPIDR read to CPU hotplug callback Sumit Gupta
2026-11-17 15:49 ` [PATCH v2 2/9] cpufreq: tegra194: remove duplicate counter sampling loops Sumit Gupta
2026-11-17 15:49 ` [PATCH v2 3/9] cpufreq: tegra194: add tegra_counter_delta helper Sumit Gupta
2026-11-17 15:49 ` [PATCH v2 4/9] cpufreq: tegra194: add per-SoC counter mask Sumit Gupta
2026-11-17 15:49 ` [PATCH v2 5/9] cpufreq: tegra194: split MMIO mapping from ACTMON offset Sumit Gupta
2026-11-17 15:49 ` [PATCH v2 6/9] dt-bindings: arm: nvidia: Document the Tegra264 CCPLEX cluster Sumit Gupta
2026-11-17 15:49 ` [PATCH v2 7/9] cpufreq: tegra194: derive frequency domain from device tree Sumit Gupta
2026-11-17 15:49 ` [PATCH v2 8/9] cpufreq: tegra194: Add support for Tegra264 Sumit Gupta
2026-11-18  3:47   ` Zhongqiu Han [this message]
2026-11-17 15:49 ` [PATCH v2 9/9] arm64: tegra: Add CCPLEX cluster node " Sumit Gupta
2026-11-18  3:55 ` [PATCH v2 0/9] cpufreq: tegra194: add Tegra264 support Zhongqiu Han

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=6452ab41-2d7c-4995-9f6f-1e8207767cdb@oss.qualcomm.com \
    --to=zhongqiu.han@oss.qualcomm.com \
    --cc=bbasu@nvidia.com \
    --cc=conor+dt@kernel.org \
    --cc=devicetree@vger.kernel.org \
    --cc=jonathanh@nvidia.com \
    --cc=krzk+dt@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=linux-tegra@vger.kernel.org \
    --cc=rafael@kernel.org \
    --cc=robh@kernel.org \
    --cc=sumitg@nvidia.com \
    --cc=treding@nvidia.com \
    --cc=viresh.kumar@linaro.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®