From: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
To: Kamal Wadhwa <kamal.wadhwa@oss.qualcomm.com>,
Sebastian Reichel <sre@kernel.org>
Cc: linux-arm-msm@vger.kernel.org, linux-pm@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] power: supply: qcom_battmgr: Add multi-port USB-C power supply support
Date: Thu, 3 Sep 2026 09:13:58 +0200 [thread overview]
Message-ID: <be340fda-a8d6-40ab-a073-5acce519a90f@oss.qualcomm.com> (raw)
In-Reply-To: <20260806-b4-battmgr-multiport-usb-v2-1-1a6dd1e06cc2@oss.qualcomm.com>
On 8/6/26 5:06 PM, Kamal Wadhwa wrote:
> Extend the qcom_battmgr driver to report up to MAX_USB_PORTS (3)
> USB-C power supply ports on the X1E80100 & Glymur platform, which
> exposes more than one charger port to firmware.
[...]
> #define BATTMGR_USB_PROPERTY_GET 0x32
> #define BATTMGR_USB_PROPERTY_SET 0x33
> +#define BATTMGR_USB2_PROPERTY_GET 0xC0
> +#define BATTMGR_USB2_PROPERTY_SET 0xC1
> +#define BATTMGR_USB3_PROPERTY_GET 0xC2
> +#define BATTMGR_USB3_PROPERTY_SET 0xC3
nit: let's keep the defines lowercase
[...]
> +static int qcom_battmgr_usb_x1e80100_update(struct qcom_battmgr *battmgr,
> + enum power_supply_property psp)
> +{
> + unsigned int prop;
> + int ret;
> +
> + if (psp >= ARRAY_SIZE(x1e80100_usb_prop_map))
> + return -EINVAL;
> +
> + prop = x1e80100_usb_prop_map[psp];
> +
> + mutex_lock(&battmgr->lock);
> + ret = qcom_battmgr_request_property(battmgr, BATTMGR_USB_PROPERTY_GET, prop, 0);
> + mutex_unlock(&battmgr->lock);
> +
> + return ret;
> +}
I'd rather you pass the array and its size to this function rather than
wholly duplicating it
[...]
> +static int qcom_battmgr_usb_x1e80100_get_property(struct power_supply *psy,
> + enum power_supply_property psp,
> + union power_supply_propval *val)
Likewise there's no reason to duplicate this one
[...]
> +static int qcom_battmgr_usb2_x1e80100_update(struct qcom_battmgr *battmgr,
> + enum power_supply_property psp)
We can add another parameter for the property and save some duplication
as well
[...]
> +static int qcom_battmgr_usb2_get_property(struct power_supply *psy,
> + enum power_supply_property psp,
> + union power_supply_propval *val)
> +{
This function is also wholly duplicated, we can extract out the common
part
[...]
> static const enum power_supply_property sc8280xp_usb_props[] = {
> POWER_SUPPLY_PROP_ONLINE,
> + POWER_SUPPLY_PROP_VOLTAGE_NOW,
> + POWER_SUPPLY_PROP_VOLTAGE_MAX,
> + POWER_SUPPLY_PROP_CURRENT_NOW,
> + POWER_SUPPLY_PROP_CURRENT_MAX,
> + POWER_SUPPLY_PROP_INPUT_CURRENT_LIMIT,
> + POWER_SUPPLY_PROP_USB_TYPE,
> };
Hm, no one noticed this list was very short for a long time..
[...]
> + case BATTMGR_USB_PROPERTY_GET:
> + property = le32_to_cpu(resp->intval.property);
> + if (payload_len != sizeof(resp->intval)) {
> + dev_warn(battmgr->dev,
> + "invalid payload length for %#x request: %zd\n",
> + property, payload_len);
> + battmgr->error = -ENODATA;
> + return;
> + }
We can do something like:
if (opcode == BATTMGR_USB2_PROPERTY_GET)
usb_info = &battmgr->usb2;
else if (opcode == BATTMGR_USB3_PROPERTY_GET)
usb_info = &battmgr->usb3;
else
usb_info = &battmgr->usb;
and then have a shared case for all 3 in the switch-statement
[...]
> +static char *qcom_battmgr_battery[] = { "battery" };
> +
> static void qcom_battmgr_enable_worker(struct work_struct *work)
> {
> struct qcom_battmgr *battmgr = container_of(work, struct qcom_battmgr, enable_work);
> @@ -1591,11 +1989,53 @@ static void qcom_battmgr_enable_worker(struct work_struct *work)
> .hdr.type = cpu_to_le32(PMIC_GLINK_NOTIFY),
> .hdr.opcode = cpu_to_le32(BATTMGR_REQUEST_NOTIFICATION),
> };
> + struct power_supply *psy;
> int ret;
> + int num_ports_fw = 0;
>
> ret = qcom_battmgr_request(battmgr, &req, sizeof(req));
> if (ret)
> dev_err(battmgr->dev, "failed to request power notifications\n");
> +
> + if (battmgr->variant == QCOM_BATTMGR_X1E80100) {
> + mutex_lock(&battmgr->lock);
> + ret = qcom_battmgr_request_property(battmgr, BATTMGR_USB_PROPERTY_GET,
> + USB_NUM_PORTS, 0);
> + mutex_unlock(&battmgr->lock);
> + if (ret < 0) {
> + dev_dbg(battmgr->dev, "Failed to read USB_NUM_PORTS from SoCCP, rc=%d\n",
> + ret);
The pmic-glink drivers are supposed not to care what's on the other end,
let's just drop the "from SoCCP" part.
Also, the existing way this driver prints errors is:
"A problem occured: %d\n", without a 'rc=' prefix
Konrad
next prev parent reply other threads:[~2026-09-03 7:14 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-08-06 15:06 Kamal Wadhwa
2026-09-03 7:13 ` Konrad Dybcio [this message]
2026-09-03 7:15 ` Konrad Dybcio
2026-09-03 7:44 ` Konrad Dybcio
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=be340fda-a8d6-40ab-a073-5acce519a90f@oss.qualcomm.com \
--to=konrad.dybcio@oss.qualcomm.com \
--cc=kamal.wadhwa@oss.qualcomm.com \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=sre@kernel.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®