From: "Nils Büchner" <nils@buechner.me>
To: Daniel Lezcano <daniel.lezcano@kernel.org>,
"Rafael J . Wysocki" <rafael@kernel.org>,
Matthias Brugger <matthias.bgg@gmail.com>,
AngeloGioacchino Del Regno
<angelogioacchino.delregno@collabora.com>
Cc: "Zhang Rui" <rui.zhang@intel.com>,
"Lukasz Luba" <lukasz.luba@arm.com>,
linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-arm-kernel@lists.infradead.org,
linux-mediatek@lists.infradead.org,
"Nils Büchner" <nils@buechner.me>
Subject: [PATCH 1/2] thermal/drivers/mediatek/auxadc: Register per-sensor thermal zones
Date: Tue, 6 Oct 2026 04:54:09 +0200 [thread overview]
Message-ID: <20261006025410.506443-2-nils@buechner.me> (raw)
In-Reply-To: <20261006025410.506443-1-nils@buechner.me>
The driver registers only thermal zone 0, which reports the hottest of
all sensors. The MT8183 devicetree also describes a zone per sensor
(soc, gpu, md1, cpu-little, cpu-big, abb as sensor ids 1-6), and
mtk-svs looks up the "gpu-thermal" zone for its GPU banks. Without it
SVS fails to probe on MT8183:
SVSB_GPU: cannot get "gpu-thermal" thermal zone
mtk-svs 1100bc00.svs: error -ENODEV: svs bank resource setup fail
so the SoC never gets its calibrated, lower voltages.
Additionally register every sensor that has a devicetree zone, using
sensor id = sensor number + 1, read from that sensor's own measurement
register. Sensors without a zone are skipped, so SoCs whose devicetrees
only describe zone 0 are unaffected.
Runtime-tested on an Acer Chromebook Spin 311 (MT8183, juniper) with a
7.0-based kernel (the hardware is not set up to boot mainline);
build-tested on mainline with W=1. On the device: all six
sensor zones register and read plausible values (31-34 C idle, close
to the board sensors), and mtk-svs finds gpu-thermal.
The change and this changelog were drafted with an AI coding assistant
during a debugging session on the hardware; they were reviewed and
tested by the submitter.
Assisted-by: Claude Opus 5.5
Signed-off-by: Nils Büchner <nils@buechner.me>
---
drivers/thermal/mediatek/auxadc_thermal.c | 80 ++++++++++++++++++++++-
1 file changed, 79 insertions(+), 1 deletion(-)
diff --git a/drivers/thermal/mediatek/auxadc_thermal.c b/drivers/thermal/mediatek/auxadc_thermal.c
index 9ee2e7283..59c8eaf87 100644
--- a/drivers/thermal/mediatek/auxadc_thermal.c
+++ b/drivers/thermal/mediatek/auxadc_thermal.c
@@ -308,6 +308,13 @@ struct mtk_thermal_bank {
int id;
};
+/* A single sensor, exposed as its own thermal zone (DT sensor id = sensor + 1) */
+struct mtk_thermal_sensor {
+ struct mtk_thermal *mt;
+ int bank;
+ int index;
+};
+
struct mtk_thermal_data {
s32 num_banks;
s32 num_sensors;
@@ -870,6 +877,77 @@ static const struct thermal_zone_device_ops mtk_thermal_ops = {
.get_temp = mtk_read_temp,
};
+static int mtk_read_sensor_temp(struct thermal_zone_device *tz, int *temperature)
+{
+ struct mtk_thermal_sensor *sensor = thermal_zone_device_priv(tz);
+ struct mtk_thermal *mt = sensor->mt;
+ const struct mtk_thermal_data *conf = mt->conf;
+ struct mtk_thermal_bank *bank = &mt->banks[sensor->bank];
+ int temp;
+ u32 raw;
+
+ mtk_thermal_get_bank(bank);
+ raw = readl(mt->thermal_base + conf->msr[sensor->index]);
+ temp = mt->raw_to_mcelsius(mt, conf->bank_data[sensor->bank].sensors[sensor->index],
+ raw);
+ mtk_thermal_put_bank(bank);
+
+ if (!mtk_thermal_temp_is_valid(temp))
+ return -EAGAIN;
+
+ *temperature = temp;
+
+ return 0;
+}
+
+static const struct thermal_zone_device_ops mtk_thermal_sensor_ops = {
+ .get_temp = mtk_read_sensor_temp,
+};
+
+/*
+ * Zone 0 reports the hottest sensor. Also register every individual sensor
+ * the devicetree describes (sensor id = sensor number + 1): MT8183 SVS looks
+ * up its GPU bank temperature through the "gpu-thermal" zone and cannot
+ * probe without it. Sensors without a devicetree zone are skipped.
+ */
+static int mtk_thermal_register_sensors(struct mtk_thermal *mt)
+{
+ const struct mtk_thermal_data *conf = mt->conf;
+ struct thermal_zone_device *tzdev;
+ struct mtk_thermal_sensor *sensor;
+ int id, b, i;
+
+ for (id = 1; id <= conf->num_sensors; id++) {
+ sensor = NULL;
+ for (b = 0; b < conf->num_banks && !sensor; b++) {
+ for (i = 0; i < conf->bank_data[b].num_sensors; i++) {
+ if (conf->bank_data[b].sensors[i] != id - 1)
+ continue;
+ sensor = devm_kzalloc(mt->dev, sizeof(*sensor), GFP_KERNEL);
+ if (!sensor)
+ return -ENOMEM;
+ sensor->mt = mt;
+ sensor->bank = b;
+ sensor->index = i;
+ break;
+ }
+ }
+ if (!sensor)
+ continue;
+
+ tzdev = devm_thermal_of_zone_register(mt->dev, id, sensor,
+ &mtk_thermal_sensor_ops);
+ if (IS_ERR(tzdev)) {
+ if (PTR_ERR(tzdev) == -ENODEV)
+ continue;
+ return dev_err_probe(mt->dev, PTR_ERR(tzdev),
+ "failed to register sensor %d\n", id);
+ }
+ }
+
+ return 0;
+}
+
static void mtk_thermal_init_bank(struct mtk_thermal *mt, int num,
u32 apmixed_phys_base, u32 auxadc_phys_base,
int ctrl_id)
@@ -1294,7 +1372,7 @@ static int mtk_thermal_probe(struct platform_device *pdev)
if (ret)
dev_warn(&pdev->dev, "error in thermal_add_hwmon_sysfs");
- return 0;
+ return mtk_thermal_register_sensors(mt);
}
static struct platform_driver mtk_thermal_driver = {
--
2.53.0
next prev parent reply other threads:[~2026-10-06 2:54 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-10-06 2:54 [PATCH 0/2] mediatek: Make SVS work on MT8183 Nils Büchner
2026-10-06 2:54 ` Nils Büchner [this message]
2026-10-07 11:51 ` [PATCH 1/2] thermal/drivers/mediatek/auxadc: Register per-sensor thermal zones Daniel Lezcano
2026-10-06 2:54 ` [PATCH 2/2] soc: mediatek: mtk-svs: Don't fail init01 on a bank whose buck is not at vboot yet Nils Büchner
2026-10-06 8:32 ` Chen-Yu Tsai
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=20261006025410.506443-2-nils@buechner.me \
--to=nils@buechner.me \
--cc=angelogioacchino.delregno@collabora.com \
--cc=daniel.lezcano@kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mediatek@lists.infradead.org \
--cc=linux-pm@vger.kernel.org \
--cc=lukasz.luba@arm.com \
--cc=matthias.bgg@gmail.com \
--cc=rafael@kernel.org \
--cc=rui.zhang@intel.com \
/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