| Message ID | 20220302183515.448334-1-caleb.connolly@linaro.org (mailing list archive) |
|---|---|
| State | Awaiting Upstream |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | [v2] bluetooth: hci_event: don't print an error on vendor events | expand |
| Context | Check | Description |
|---|---|---|
| netdev/tree_selection | success | Guessing tree name failed - patch did not apply |
Hi Caleb, > Since commit 3e54c5890c87 ("Bluetooth: hci_event: Use of a function table to handle HCI events"), > some devices see warnings being printed for vendor events, e.g. > > [ 75.806141] Bluetooth: hci0: setting up wcn399x > [ 75.948311] Bluetooth: hci0: unexpected event 0xff length: 14 > 0 > [ 75.955552] Bluetooth: hci0: QCA Product ID :0x0000000a > [ 75.961369] Bluetooth: hci0: QCA SOC Version :0x40010214 > [ 75.967417] Bluetooth: hci0: QCA ROM Version :0x00000201 > [ 75.973363] Bluetooth: hci0: QCA Patch Version:0x00000001 > [ 76.000289] Bluetooth: hci0: QCA controller version 0x02140201 > [ 76.006727] Bluetooth: hci0: QCA Downloading qca/crbtfw21.tlv > [ 76.986850] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.013574] Bluetooth: hci0: QCA Downloading qca/oneplus6/crnv21.bin > [ 77.024302] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.032681] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.040674] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.049251] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.057997] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.066320] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.075065] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.083073] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.091250] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.099417] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.110166] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.118672] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.127449] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.137190] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.146192] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.154242] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.163183] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.171202] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.179364] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.187259] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > [ 77.198451] Bluetooth: hci0: QCA setup on UART is completed > > Avoid printing the event length warning for vendor events, this reverts > to the previous behaviour where such warnings weren't printed. > > Fixes: 3e54c5890c87 ("Bluetooth: hci_event: Use of a function table to handle HCI events") > Signed-off-by: Caleb Connolly <caleb.connolly@linaro.org> > --- > Changes since v1: > * Don't return early! Vendor events still get parsed despite the > warning. I should have looked a little more closely at that... > --- > net/bluetooth/hci_event.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) patch has been applied to bluetooth-stable tree. Regards Marcel
Hi Marcel, Caleb, On Wed, Mar 2, 2022 at 11:20 AM Marcel Holtmann <marcel@holtmann.org> wrote: > > Hi Caleb, > > > Since commit 3e54c5890c87 ("Bluetooth: hci_event: Use of a function table to handle HCI events"), > > some devices see warnings being printed for vendor events, e.g. > > > > [ 75.806141] Bluetooth: hci0: setting up wcn399x > > [ 75.948311] Bluetooth: hci0: unexpected event 0xff length: 14 > 0 > > [ 75.955552] Bluetooth: hci0: QCA Product ID :0x0000000a > > [ 75.961369] Bluetooth: hci0: QCA SOC Version :0x40010214 > > [ 75.967417] Bluetooth: hci0: QCA ROM Version :0x00000201 > > [ 75.973363] Bluetooth: hci0: QCA Patch Version:0x00000001 > > [ 76.000289] Bluetooth: hci0: QCA controller version 0x02140201 > > [ 76.006727] Bluetooth: hci0: QCA Downloading qca/crbtfw21.tlv > > [ 76.986850] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.013574] Bluetooth: hci0: QCA Downloading qca/oneplus6/crnv21.bin > > [ 77.024302] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.032681] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.040674] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.049251] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.057997] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.066320] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.075065] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.083073] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.091250] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.099417] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.110166] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.118672] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.127449] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.137190] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.146192] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.154242] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.163183] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.171202] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.179364] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.187259] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 > > [ 77.198451] Bluetooth: hci0: QCA setup on UART is completed > > > > Avoid printing the event length warning for vendor events, this reverts > > to the previous behaviour where such warnings weren't printed. > > > > Fixes: 3e54c5890c87 ("Bluetooth: hci_event: Use of a function table to handle HCI events") > > Signed-off-by: Caleb Connolly <caleb.connolly@linaro.org> > > --- > > Changes since v1: > > * Don't return early! Vendor events still get parsed despite the > > warning. I should have looked a little more closely at that... > > --- > > net/bluetooth/hci_event.c | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > patch has been applied to bluetooth-stable tree. > > Regards > > Marcel I believe a proper fix has already been pushed to bluetooth-next: https://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next.git/commit/?id=314d8cd2787418c5ac6b02035c344644f47b292b HCI_EV_VENDOR shall be assumed to be variable length and that also uses bt_dev_warn_ratelimited to avoid spamming the logs in case it still fails.
Hi Luiz, >>> Since commit 3e54c5890c87 ("Bluetooth: hci_event: Use of a function table to handle HCI events"), >>> some devices see warnings being printed for vendor events, e.g. >>> >>> [ 75.806141] Bluetooth: hci0: setting up wcn399x >>> [ 75.948311] Bluetooth: hci0: unexpected event 0xff length: 14 > 0 >>> [ 75.955552] Bluetooth: hci0: QCA Product ID :0x0000000a >>> [ 75.961369] Bluetooth: hci0: QCA SOC Version :0x40010214 >>> [ 75.967417] Bluetooth: hci0: QCA ROM Version :0x00000201 >>> [ 75.973363] Bluetooth: hci0: QCA Patch Version:0x00000001 >>> [ 76.000289] Bluetooth: hci0: QCA controller version 0x02140201 >>> [ 76.006727] Bluetooth: hci0: QCA Downloading qca/crbtfw21.tlv >>> [ 76.986850] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.013574] Bluetooth: hci0: QCA Downloading qca/oneplus6/crnv21.bin >>> [ 77.024302] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.032681] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.040674] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.049251] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.057997] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.066320] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.075065] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.083073] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.091250] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.099417] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.110166] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.118672] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.127449] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.137190] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.146192] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.154242] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.163183] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.171202] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.179364] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.187259] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 >>> [ 77.198451] Bluetooth: hci0: QCA setup on UART is completed >>> >>> Avoid printing the event length warning for vendor events, this reverts >>> to the previous behaviour where such warnings weren't printed. >>> >>> Fixes: 3e54c5890c87 ("Bluetooth: hci_event: Use of a function table to handle HCI events") >>> Signed-off-by: Caleb Connolly <caleb.connolly@linaro.org> >>> --- >>> Changes since v1: >>> * Don't return early! Vendor events still get parsed despite the >>> warning. I should have looked a little more closely at that... >>> --- >>> net/bluetooth/hci_event.c | 2 +- >>> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> patch has been applied to bluetooth-stable tree. >> > I believe a proper fix has already been pushed to bluetooth-next: > > https://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next.git/commit/?id=314d8cd2787418c5ac6b02035c344644f47b292b > > HCI_EV_VENDOR shall be assumed to be variable length and that also > uses bt_dev_warn_ratelimited to avoid spamming the logs in case it > still fails. ok, I reverted the patch and lets this go via net-next tree then. Stable can pick this up if it really becomes a larger problem. Regards Marcel
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index fc30f4c03d29..9b7c9ab77971 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -6822,7 +6822,7 @@ static void hci_event_func(struct hci_dev *hdev, u8 event, struct sk_buff *skb, * possible to partially parse the event so leave to callback to * decide if that is acceptable. */ - if (skb->len > ev->max_len) + if (skb->len > ev->max_len && event != HCI_EV_VENDOR) bt_dev_warn(hdev, "unexpected event 0x%2.2x length: %u > %u", event, skb->len, ev->max_len);
Since commit 3e54c5890c87 ("Bluetooth: hci_event: Use of a function table to handle HCI events"), some devices see warnings being printed for vendor events, e.g. [ 75.806141] Bluetooth: hci0: setting up wcn399x [ 75.948311] Bluetooth: hci0: unexpected event 0xff length: 14 > 0 [ 75.955552] Bluetooth: hci0: QCA Product ID :0x0000000a [ 75.961369] Bluetooth: hci0: QCA SOC Version :0x40010214 [ 75.967417] Bluetooth: hci0: QCA ROM Version :0x00000201 [ 75.973363] Bluetooth: hci0: QCA Patch Version:0x00000001 [ 76.000289] Bluetooth: hci0: QCA controller version 0x02140201 [ 76.006727] Bluetooth: hci0: QCA Downloading qca/crbtfw21.tlv [ 76.986850] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.013574] Bluetooth: hci0: QCA Downloading qca/oneplus6/crnv21.bin [ 77.024302] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.032681] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.040674] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.049251] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.057997] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.066320] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.075065] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.083073] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.091250] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.099417] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.110166] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.118672] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.127449] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.137190] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.146192] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.154242] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.163183] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.171202] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.179364] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.187259] Bluetooth: hci0: unexpected event 0xff length: 3 > 0 [ 77.198451] Bluetooth: hci0: QCA setup on UART is completed Avoid printing the event length warning for vendor events, this reverts to the previous behaviour where such warnings weren't printed. Fixes: 3e54c5890c87 ("Bluetooth: hci_event: Use of a function table to handle HCI events") Signed-off-by: Caleb Connolly <caleb.connolly@linaro.org> --- Changes since v1: * Don't return early! Vendor events still get parsed despite the warning. I should have looked a little more closely at that... --- net/bluetooth/hci_event.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)