[PATCH v1] driver: bluetooth: hci_qca: Disable auto_off when BT_EN is pulled up by hw

[PATCH v1] driver: bluetooth: hci_qca: Disable auto_off when BT_EN is pulled up by hw

[Shuai Zhang]

If the BT SoC BT_EN is controlled by hardware, disable the AUTO_OFF
feature. Otherwise, BT will close the HCI layer except for the UART
after firmware download. However, the SoC remains active. If the SoC
sends a packet to the Host after firmware download, the Host cannot
respond since the HCI layer is closed, which will cause the firmware
to enter an incorrect state.

Signed-off-by: Shuai Zhang <quic_shuaz@quicinc.com>
---
 drivers/bluetooth/hci_qca.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index a2dc39c00..0cb70a3cf 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -2483,6 +2483,15 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 		set_bit(HCI_QUIRK_NON_PERSISTENT_SETUP, &hdev->quirks);
 		hdev->shutdown = qca_power_off;
 	}
+	/* If the BT SoC BT_EN is controlled by hardware, disable the AUTO_OFF
+	 * feature. Otherwise, BT will close the HCI layer except for
+	 * the UART after firmware download. However, the SoC remains active.
+	 * If the SoC sends a packet to the Host after firmware download,
+	 * the Host cannot respond since the HCI layer is closed, which will
+	 * cause the firmware to enter an incorrect state.
+	 */
+	else
+		hci_dev_clear_flag(hdev, HCI_AUTO_OFF);
 
 	if (data) {
 		/* Wideband speech support must be set per driver since it can't
-- 
2.34.1

RE: [v1] driver: bluetooth: hci_qca: Disable auto_off when BT_EN is pulled up by hw

[bluez.test.bot]

[-- Attachment #1: Type: text/plain, Size: 2116 bytes --]

This is automated email and please do not reply to this email!

Dear submitter,

Thank you for submitting the patches to the linux bluetooth mailing list.
This is a CI test results with your patch series:
PW Link:https://patchwork.kernel.org/project/bluetooth/list/?series=969791

---Test result---

Test Summary:
CheckPatch                    PENDING   0.32 seconds
GitLint                       PENDING   0.33 seconds
SubjectPrefix                 FAIL      0.37 seconds
BuildKernel                   PASS      25.00 seconds
CheckAllWarning               PASS      27.07 seconds
CheckSparse                   PASS      30.82 seconds
BuildKernel32                 PASS      24.33 seconds
TestRunnerSetup               PASS      463.98 seconds
TestRunner_l2cap-tester       PASS      25.26 seconds
TestRunner_iso-tester         PASS      38.25 seconds
TestRunner_bnep-tester        PASS      5.90 seconds
TestRunner_mgmt-tester        FAIL      135.62 seconds
TestRunner_rfcomm-tester      PASS      9.28 seconds
TestRunner_sco-tester         PASS      15.39 seconds
TestRunner_ioctl-tester       PASS      9.89 seconds
TestRunner_mesh-tester        PASS      7.39 seconds
TestRunner_smp-tester         PASS      8.52 seconds
TestRunner_userchan-tester    PASS      6.16 seconds
IncrementalBuild              PENDING   0.52 seconds

Details
##############################
Test: CheckPatch - PENDING
Desc: Run checkpatch.pl script
Output:

##############################
Test: GitLint - PENDING
Desc: Run gitlint
Output:

##############################
Test: SubjectPrefix - FAIL
Desc: Check subject contains "Bluetooth" prefix
Output:
"Bluetooth: " prefix is not specified in the subject
##############################
Test: TestRunner_mgmt-tester - FAIL
Desc: Run mgmt-tester with test-runner
Output:
Total: 490, Passed: 485 (99.0%), Failed: 1, Not Run: 4

Failed Test Cases
LL Privacy - Add Device 3 (AL is full)               Failed       0.250 seconds
##############################
Test: IncrementalBuild - PENDING
Desc: Incremental build with the patches in the series
Output:



---
Regards,
Linux Bluetooth

Re: [PATCH v1] driver: bluetooth: hci_qca: Disable auto_off when BT_EN is pulled up by hw

[Luiz Augusto von Dentz]

Hi Shuai,

On Mon, Jun 9, 2025 at 7:00 AM Shuai Zhang <quic_shuaz@quicinc.com> wrote:
>
> If the BT SoC BT_EN is controlled by hardware, disable the AUTO_OFF
> feature. Otherwise, BT will close the HCI layer except for the UART
> after firmware download. However, the SoC remains active. If the SoC
> sends a packet to the Host after firmware download, the Host cannot
> respond since the HCI layer is closed, which will cause the firmware
> to enter an incorrect state.
>
> Signed-off-by: Shuai Zhang <quic_shuaz@quicinc.com>
> ---
>  drivers/bluetooth/hci_qca.c | 9 +++++++++
>  1 file changed, 9 insertions(+)
>
> diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
> index a2dc39c00..0cb70a3cf 100644
> --- a/drivers/bluetooth/hci_qca.c
> +++ b/drivers/bluetooth/hci_qca.c
> @@ -2483,6 +2483,15 @@ static int qca_serdev_probe(struct serdev_device *serdev)
>                 set_bit(HCI_QUIRK_NON_PERSISTENT_SETUP, &hdev->quirks);
>                 hdev->shutdown = qca_power_off;
>         }
> +       /* If the BT SoC BT_EN is controlled by hardware, disable the AUTO_OFF
> +        * feature. Otherwise, BT will close the HCI layer except for
> +        * the UART after firmware download. However, the SoC remains active.
> +        * If the SoC sends a packet to the Host after firmware download,
> +        * the Host cannot respond since the HCI layer is closed, which will
> +        * cause the firmware to enter an incorrect state.
> +        */
> +       else
> +               hci_dev_clear_flag(hdev, HCI_AUTO_OFF);

This sounds like a bug imo, the driver shouldn't be messing up with
HCI_AUTO_OFF, if the packet mentioned above is required as part of the
setup stage then it shall be handled via setup callback otherwise the
stack will consider the setup as completed and proceed with auto off,
etc.

>         if (data) {
>                 /* Wideband speech support must be set per driver since it can't
> --
> 2.34.1
>
>


-- 
Luiz Augusto von Dentz