This reverts commit 7dcd3e014aa7faeeaf4047190b22d8a19a0db696.
Qualcomm Bluetooth controllers like WCN6855 do not have persistent storage for the Bluetooth address and must therefore start as unconfigured to allow the user to set a valid address unless one has been provided by the boot firmware in the devicetree.
A recent change snuck into v6.8-rc7 and incorrectly started marking the default (non-unique) address as valid. This specifically also breaks the Bluetooth setup for some user of the Lenovo ThinkPad X13s.
Note that this is the second time Qualcomm breaks the driver this way and that this was fixed last year by commit 6945795bc81a ("Bluetooth: fix use-bdaddr-property quirk"), which also has some further details.
Fixes: 7dcd3e014aa7 ("Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT") Cc: stable@vger.kernel.org # 6.8 Cc: Janaki Ramaiah Thota quic_janathot@quicinc.com Signed-off-by: Johan Hovold johan+linaro@kernel.org --- drivers/bluetooth/hci_qca.c | 13 +------------ 1 file changed, 1 insertion(+), 12 deletions(-)
diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c index edd2a81b4d5e..f989c05f8177 100644 --- a/drivers/bluetooth/hci_qca.c +++ b/drivers/bluetooth/hci_qca.c @@ -7,7 +7,6 @@ * * Copyright (C) 2007 Texas Instruments, Inc. * Copyright (c) 2010, 2012, 2018 The Linux Foundation. All rights reserved. - * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved. * * Acknowledgements: * This file is based on hci_ll.c, which was... @@ -1904,17 +1903,7 @@ static int qca_setup(struct hci_uart *hu) case QCA_WCN6750: case QCA_WCN6855: case QCA_WCN7850: - - /* Set BDA quirk bit for reading BDA value from fwnode property - * only if that property exist in DT. - */ - if (fwnode_property_present(dev_fwnode(hdev->dev.parent), "local-bd-address")) { - set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks); - bt_dev_info(hdev, "setting quirk bit to read BDA from fwnode later"); - } else { - bt_dev_dbg(hdev, "local-bd-address` is not present in the devicetree so not setting quirk bit for BDA"); - } - + set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks); hci_set_aosp_capable(hdev);
ret = qca_read_soc_version(hdev, &ver, soc_type);
Hi Johan,
On Thu, Mar 14, 2024 at 4:44 AM Johan Hovold johan+linaro@kernel.org wrote:
This reverts commit 7dcd3e014aa7faeeaf4047190b22d8a19a0db696.
Qualcomm Bluetooth controllers like WCN6855 do not have persistent storage for the Bluetooth address and must therefore start as unconfigured to allow the user to set a valid address unless one has been provided by the boot firmware in the devicetree.
A recent change snuck into v6.8-rc7 and incorrectly started marking the default (non-unique) address as valid. This specifically also breaks the Bluetooth setup for some user of the Lenovo ThinkPad X13s.
Note that this is the second time Qualcomm breaks the driver this way and that this was fixed last year by commit 6945795bc81a ("Bluetooth: fix use-bdaddr-property quirk"), which also has some further details.
Fixes: 7dcd3e014aa7 ("Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT") Cc: stable@vger.kernel.org # 6.8 Cc: Janaki Ramaiah Thota quic_janathot@quicinc.com Signed-off-by: Johan Hovold johan+linaro@kernel.org
Well I guess I will need to start asking for evidence that this works on regular Linux distros then, because it looks like that is not the environment Janaki and others Qualcomm folks are testing with.
What I probably would consider as evidence is bluetoothd logs showing that the controller has been configured correctly or perhaps there is a simpler way?
drivers/bluetooth/hci_qca.c | 13 +------------ 1 file changed, 1 insertion(+), 12 deletions(-)
diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c index edd2a81b4d5e..f989c05f8177 100644 --- a/drivers/bluetooth/hci_qca.c +++ b/drivers/bluetooth/hci_qca.c @@ -7,7 +7,6 @@
- Copyright (C) 2007 Texas Instruments, Inc.
- Copyright (c) 2010, 2012, 2018 The Linux Foundation. All rights reserved.
- Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved.
- Acknowledgements:
- This file is based on hci_ll.c, which was...
@@ -1904,17 +1903,7 @@ static int qca_setup(struct hci_uart *hu) case QCA_WCN6750: case QCA_WCN6855: case QCA_WCN7850:
/* Set BDA quirk bit for reading BDA value from fwnode property* only if that property exist in DT.*/if (fwnode_property_present(dev_fwnode(hdev->dev.parent), "local-bd-address")) {set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks);bt_dev_info(hdev, "setting quirk bit to read BDA from fwnode later");} else {bt_dev_dbg(hdev, "local-bd-address` is not present in the devicetree so not setting quirk bit for BDA");}
set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks); hci_set_aosp_capable(hdev); ret = qca_read_soc_version(hdev, &ver, soc_type);-- 2.43.2
On Thu, Mar 14, 2024 at 10:30:36AM -0400, Luiz Augusto von Dentz wrote:
On Thu, Mar 14, 2024 at 4:44 AM Johan Hovold johan+linaro@kernel.org wrote:
This reverts commit 7dcd3e014aa7faeeaf4047190b22d8a19a0db696.
Qualcomm Bluetooth controllers like WCN6855 do not have persistent storage for the Bluetooth address and must therefore start as unconfigured to allow the user to set a valid address unless one has been provided by the boot firmware in the devicetree.
A recent change snuck into v6.8-rc7 and incorrectly started marking the default (non-unique) address as valid. This specifically also breaks the Bluetooth setup for some user of the Lenovo ThinkPad X13s.
Note that this is the second time Qualcomm breaks the driver this way and that this was fixed last year by commit 6945795bc81a ("Bluetooth: fix use-bdaddr-property quirk"), which also has some further details.
Fixes: 7dcd3e014aa7 ("Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT") Cc: stable@vger.kernel.org # 6.8 Cc: Janaki Ramaiah Thota quic_janathot@quicinc.com Signed-off-by: Johan Hovold johan+linaro@kernel.org
Well I guess I will need to start asking for evidence that this works on regular Linux distros then, because it looks like that is not the environment Janaki and others Qualcomm folks are testing with.
What I probably would consider as evidence is bluetoothd logs showing that the controller has been configured correctly or perhaps there is a simpler way?
Well, in this case we actually want the controller to remain unconfigured (e.g. to avoid having every user of the X13s unknowingly use the same default address).
I'm not sure why Qualcomm insists on breaking these quirks, but I guess they just haven't understood why they exist. It's of course convenient to be able to use the default address during development without first having to provide an address, but that's not a valid reason to break the driver.
From what I hear the Qualcomm developers only care about Android and I believe they have some out-of-tree hack for retrieving the device address directly from the rootfs.
For the X13s, and as I think I've mentioned before, we have been trying to get Qualcomm to tell us how to access the assigned addresses that are stored in some secure world storage so that we can set it directly from the driver. But until we figure that out, users will need to continue setting the address manually.
Johan
Bluetooth Maintainers, what's...
On 14.03.24 16:07, Johan Hovold wrote:
On Thu, Mar 14, 2024 at 10:30:36AM -0400, Luiz Augusto von Dentz wrote:
On Thu, Mar 14, 2024 at 4:44 AM Johan Hovold johan+linaro@kernel.org wrote:
This reverts commit 7dcd3e014aa7faeeaf4047190b22d8a19a0db696.
Qualcomm Bluetooth controllers like WCN6855 do not have persistent storage for the Bluetooth address and must therefore start as unconfigured to allow the user to set a valid address unless one has been provided by the boot firmware in the devicetree.
A recent change snuck into v6.8-rc7 and incorrectly started marking the default (non-unique) address as valid. This specifically also breaks the Bluetooth setup for some user of the Lenovo ThinkPad X13s.
Note that this is the second time Qualcomm breaks the driver this way and that this was fixed last year by commit 6945795bc81a ("Bluetooth: fix use-bdaddr-property quirk"), which also has some further details.
Fixes: 7dcd3e014aa7 ("Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT") Cc: stable@vger.kernel.org # 6.8 Cc: Janaki Ramaiah Thota quic_janathot@quicinc.com Signed-off-by: Johan Hovold johan+linaro@kernel.org
Well I guess I will need to start asking for evidence that this works on regular Linux distros then, because it looks like that is not the environment Janaki and others Qualcomm folks are testing with.
What I probably would consider as evidence is bluetoothd logs showing that the controller has been configured correctly or perhaps there is a simpler way?
Well, in this case we actually want the controller to remain unconfigured (e.g. to avoid having every user of the X13s unknowingly use the same default address).
I'm not sure why Qualcomm insists on breaking these quirks, but I guess they just haven't understood why they exist. It's of course convenient to be able to use the default address during development without first having to provide an address, but that's not a valid reason to break the driver.
From what I hear the Qualcomm developers only care about Android and I believe they have some out-of-tree hack for retrieving the device address directly from the rootfs.
For the X13s, and as I think I've mentioned before, we have been trying to get Qualcomm to tell us how to access the assigned addresses that are stored in some secure world storage so that we can set it directly from the driver. But until we figure that out, users will need to continue setting the address manually.
...the plan forward here? This to me sounds like a case where a quick revert is the right (interim?) solution, but nevertheless nothing happened for ~10 days now afaics. Or am I missing something?
Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat) -- Everything you wanna know about Linux kernel regression tracking: https://linux-regtracking.leemhuis.info/about/#tldr If I did something stupid, please tell me, as explained on that page.
#regzbot poke
Hi Johan,
On Mon, Mar 25, 2024 at 9:57 AM Linux regression tracking (Thorsten Leemhuis) regressions@leemhuis.info wrote:
Bluetooth Maintainers, what's...
On 14.03.24 16:07, Johan Hovold wrote:
On Thu, Mar 14, 2024 at 10:30:36AM -0400, Luiz Augusto von Dentz wrote:
On Thu, Mar 14, 2024 at 4:44 AM Johan Hovold johan+linaro@kernel.org wrote:
This reverts commit 7dcd3e014aa7faeeaf4047190b22d8a19a0db696.
Qualcomm Bluetooth controllers like WCN6855 do not have persistent storage for the Bluetooth address and must therefore start as unconfigured to allow the user to set a valid address unless one has been provided by the boot firmware in the devicetree.
A recent change snuck into v6.8-rc7 and incorrectly started marking the default (non-unique) address as valid. This specifically also breaks the Bluetooth setup for some user of the Lenovo ThinkPad X13s.
Note that this is the second time Qualcomm breaks the driver this way and that this was fixed last year by commit 6945795bc81a ("Bluetooth: fix use-bdaddr-property quirk"), which also has some further details.
Fixes: 7dcd3e014aa7 ("Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT") Cc: stable@vger.kernel.org # 6.8 Cc: Janaki Ramaiah Thota quic_janathot@quicinc.com Signed-off-by: Johan Hovold johan+linaro@kernel.org
Well I guess I will need to start asking for evidence that this works on regular Linux distros then, because it looks like that is not the environment Janaki and others Qualcomm folks are testing with.
What I probably would consider as evidence is bluetoothd logs showing that the controller has been configured correctly or perhaps there is a simpler way?
Well, in this case we actually want the controller to remain unconfigured (e.g. to avoid having every user of the X13s unknowingly use the same default address).
I'm not sure why Qualcomm insists on breaking these quirks, but I guess they just haven't understood why they exist. It's of course convenient to be able to use the default address during development without first having to provide an address, but that's not a valid reason to break the driver.
From what I hear the Qualcomm developers only care about Android and I believe they have some out-of-tree hack for retrieving the device address directly from the rootfs.
For the X13s, and as I think I've mentioned before, we have been trying to get Qualcomm to tell us how to access the assigned addresses that are stored in some secure world storage so that we can set it directly from the driver. But until we figure that out, users will need to continue setting the address manually.
...the plan forward here? This to me sounds like a case where a quick revert is the right (interim?) solution, but nevertheless nothing happened for ~10 days now afaics. Or am I missing something?
Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
Everything you wanna know about Linux kernel regression tracking: https://linux-regtracking.leemhuis.info/about/#tldr If I did something stupid, please tell me, as explained on that page.
#regzbot poke
I guess the following is the latest version:
https://patchwork.kernel.org/project/bluetooth/list/?series=836664
Or are you working on a v5?
On Mon, Mar 25, 2024 at 01:10:13PM -0400, Luiz Augusto von Dentz wrote:
On Mon, Mar 25, 2024 at 9:57 AM Linux regression tracking (Thorsten Leemhuis) regressions@leemhuis.info wrote:
On 14.03.24 16:07, Johan Hovold wrote:
On Thu, Mar 14, 2024 at 10:30:36AM -0400, Luiz Augusto von Dentz wrote:
On Thu, Mar 14, 2024 at 4:44 AM Johan Hovold johan+linaro@kernel.org wrote:
This reverts commit 7dcd3e014aa7faeeaf4047190b22d8a19a0db696.
Qualcomm Bluetooth controllers like WCN6855 do not have persistent storage for the Bluetooth address and must therefore start as unconfigured to allow the user to set a valid address unless one has been provided by the boot firmware in the devicetree.
A recent change snuck into v6.8-rc7 and incorrectly started marking the default (non-unique) address as valid. This specifically also breaks the Bluetooth setup for some user of the Lenovo ThinkPad X13s.
Note that this is the second time Qualcomm breaks the driver this way and that this was fixed last year by commit 6945795bc81a ("Bluetooth: fix use-bdaddr-property quirk"), which also has some further details.
Fixes: 7dcd3e014aa7 ("Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT") Cc: stable@vger.kernel.org # 6.8 Cc: Janaki Ramaiah Thota quic_janathot@quicinc.com Signed-off-by: Johan Hovold johan+linaro@kernel.org
...the plan forward here? This to me sounds like a case where a quick revert is the right (interim?) solution, but nevertheless nothing happened for ~10 days now afaics. Or am I missing something?
I guess the following is the latest version:
https://patchwork.kernel.org/project/bluetooth/list/?series=836664
Or are you working on a v5?
This patch (revert) fixes a separate issue than the series you link to above, but it is also a prerequisite for that series.
v4 is indeed the latest version, and it has been acked by Rob and Bjorn so you can take it all through the Bluetooth tree. Just remember to apply this patch (the revert) first.
Johan
Hi Johan
On Mon, Mar 25, 2024 at 1:24 PM Johan Hovold johan@kernel.org wrote:
On Mon, Mar 25, 2024 at 01:10:13PM -0400, Luiz Augusto von Dentz wrote:
On Mon, Mar 25, 2024 at 9:57 AM Linux regression tracking (Thorsten Leemhuis) regressions@leemhuis.info wrote:
On 14.03.24 16:07, Johan Hovold wrote:
On Thu, Mar 14, 2024 at 10:30:36AM -0400, Luiz Augusto von Dentz wrote:
On Thu, Mar 14, 2024 at 4:44 AM Johan Hovold johan+linaro@kernel.org wrote:
This reverts commit 7dcd3e014aa7faeeaf4047190b22d8a19a0db696.
Qualcomm Bluetooth controllers like WCN6855 do not have persistent storage for the Bluetooth address and must therefore start as unconfigured to allow the user to set a valid address unless one has been provided by the boot firmware in the devicetree.
A recent change snuck into v6.8-rc7 and incorrectly started marking the default (non-unique) address as valid. This specifically also breaks the Bluetooth setup for some user of the Lenovo ThinkPad X13s.
Note that this is the second time Qualcomm breaks the driver this way and that this was fixed last year by commit 6945795bc81a ("Bluetooth: fix use-bdaddr-property quirk"), which also has some further details.
Fixes: 7dcd3e014aa7 ("Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT") Cc: stable@vger.kernel.org # 6.8 Cc: Janaki Ramaiah Thota quic_janathot@quicinc.com Signed-off-by: Johan Hovold johan+linaro@kernel.org
...the plan forward here? This to me sounds like a case where a quick revert is the right (interim?) solution, but nevertheless nothing happened for ~10 days now afaics. Or am I missing something?
I guess the following is the latest version:
https://patchwork.kernel.org/project/bluetooth/list/?series=836664
Or are you working on a v5?
This patch (revert) fixes a separate issue than the series you link to above, but it is also a prerequisite for that series.
v4 is indeed the latest version, and it has been acked by Rob and Bjorn so you can take it all through the Bluetooth tree. Just remember to apply this patch (the revert) first.
Doesn't seem to apply cleanly:
Applying: Bluetooth: qca: fix device-address endianness error: patch failed: drivers/bluetooth/hci_qca.c:1904 error: drivers/bluetooth/hci_qca.c: patch does not apply Patch failed at 0004 Bluetooth: qca: fix device-address endianness
Johan
On Mon, Mar 25, 2024 at 03:39:03PM -0400, Luiz Augusto von Dentz wrote:
On Mon, Mar 25, 2024 at 1:24 PM Johan Hovold johan@kernel.org wrote:
On Mon, Mar 25, 2024 at 01:10:13PM -0400, Luiz Augusto von Dentz wrote:
I guess the following is the latest version:
https://patchwork.kernel.org/project/bluetooth/list/?series=836664
Or are you working on a v5?
This patch (revert) fixes a separate issue than the series you link to above, but it is also a prerequisite for that series.
v4 is indeed the latest version, and it has been acked by Rob and Bjorn so you can take it all through the Bluetooth tree. Just remember to apply this patch (the revert) first.
Doesn't seem to apply cleanly:
Applying: Bluetooth: qca: fix device-address endianness error: patch failed: drivers/bluetooth/hci_qca.c:1904 error: drivers/bluetooth/hci_qca.c: patch does not apply Patch failed at 0004 Bluetooth: qca: fix device-address endianness
Did you apply this patch (the revert) before trying to apply the series?
Johan
Hi Johan,
On Mon, Mar 25, 2024 at 3:48 PM Johan Hovold johan@kernel.org wrote:
On Mon, Mar 25, 2024 at 03:39:03PM -0400, Luiz Augusto von Dentz wrote:
On Mon, Mar 25, 2024 at 1:24 PM Johan Hovold johan@kernel.org wrote:
On Mon, Mar 25, 2024 at 01:10:13PM -0400, Luiz Augusto von Dentz wrote:
I guess the following is the latest version:
https://patchwork.kernel.org/project/bluetooth/list/?series=836664
Or are you working on a v5?
This patch (revert) fixes a separate issue than the series you link to above, but it is also a prerequisite for that series.
v4 is indeed the latest version, and it has been acked by Rob and Bjorn so you can take it all through the Bluetooth tree. Just remember to apply this patch (the revert) first.
Doesn't seem to apply cleanly:
Applying: Bluetooth: qca: fix device-address endianness error: patch failed: drivers/bluetooth/hci_qca.c:1904 error: drivers/bluetooth/hci_qca.c: patch does not apply Patch failed at 0004 Bluetooth: qca: fix device-address endianness
Did you apply this patch (the revert) before trying to apply the series?
Probably needs rebasing:
Applying: Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT" error: drivers/bluetooth/hci_qca.c: does not match index Patch failed at 0001 Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT"
Johan
On Mon, Mar 25, 2024 at 04:07:03PM -0400, Luiz Augusto von Dentz wrote:
On Mon, Mar 25, 2024 at 3:48 PM Johan Hovold johan@kernel.org wrote:
On Mon, Mar 25, 2024 at 03:39:03PM -0400, Luiz Augusto von Dentz wrote:
On Mon, Mar 25, 2024 at 1:24 PM Johan Hovold johan@kernel.org wrote:
On Mon, Mar 25, 2024 at 01:10:13PM -0400, Luiz Augusto von Dentz wrote:
I guess the following is the latest version:
https://patchwork.kernel.org/project/bluetooth/list/?series=836664
Or are you working on a v5?
This patch (revert) fixes a separate issue than the series you link to above, but it is also a prerequisite for that series.
v4 is indeed the latest version, and it has been acked by Rob and Bjorn so you can take it all through the Bluetooth tree. Just remember to apply this patch (the revert) first.
Doesn't seem to apply cleanly:
Applying: Bluetooth: qca: fix device-address endianness error: patch failed: drivers/bluetooth/hci_qca.c:1904 error: drivers/bluetooth/hci_qca.c: patch does not apply Patch failed at 0004 Bluetooth: qca: fix device-address endianness
Did you apply this patch (the revert) before trying to apply the series?
Probably needs rebasing:
Applying: Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT" error: drivers/bluetooth/hci_qca.c: does not match index Patch failed at 0001 Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT"
I just verified that it applies cleanly to 6.9-rc1.
$ git checkout tmp v6.9-rc1 $ b4 am -sl ZgHVFjAZ1uqEiUa2@hovoldconsulting.com ... $ git am ./20240314_johan_linaro_revert_bluetooth_hci_qca_set_bda_quirk_bit_if_fwnode_exists_in_dt.mbx Applying: Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT" $ b4 am -sl 20240320075554.8178-2-johan+linaro@kernel.org ... $ git am ./v4_20240320_johan_linaro_bluetooth_qca_fix_device_address_endianness.mbx Applying: dt-bindings: bluetooth: add 'qcom,local-bd-address-broken' Applying: arm64: dts: qcom: sc7180-trogdor: mark bluetooth address as broken Applying: Bluetooth: add quirk for broken address properties Applying: Bluetooth: qca: fix device-address endianness
Do you have anything else in your tree which may interfere? What tree is that exactly?
Johan
Hi Johan,
On Mon, Mar 25, 2024 at 4:14 PM Johan Hovold johan@kernel.org wrote:
On Mon, Mar 25, 2024 at 04:07:03PM -0400, Luiz Augusto von Dentz wrote:
On Mon, Mar 25, 2024 at 3:48 PM Johan Hovold johan@kernel.org wrote:
On Mon, Mar 25, 2024 at 03:39:03PM -0400, Luiz Augusto von Dentz wrote:
On Mon, Mar 25, 2024 at 1:24 PM Johan Hovold johan@kernel.org wrote:
On Mon, Mar 25, 2024 at 01:10:13PM -0400, Luiz Augusto von Dentz wrote:
I guess the following is the latest version:
https://patchwork.kernel.org/project/bluetooth/list/?series=836664
Or are you working on a v5?
This patch (revert) fixes a separate issue than the series you link to above, but it is also a prerequisite for that series.
v4 is indeed the latest version, and it has been acked by Rob and Bjorn so you can take it all through the Bluetooth tree. Just remember to apply this patch (the revert) first.
Doesn't seem to apply cleanly:
Applying: Bluetooth: qca: fix device-address endianness error: patch failed: drivers/bluetooth/hci_qca.c:1904 error: drivers/bluetooth/hci_qca.c: patch does not apply Patch failed at 0004 Bluetooth: qca: fix device-address endianness
Did you apply this patch (the revert) before trying to apply the series?
Probably needs rebasing:
Applying: Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT" error: drivers/bluetooth/hci_qca.c: does not match index Patch failed at 0001 Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT"
I just verified that it applies cleanly to 6.9-rc1.
$ git checkout tmp v6.9-rc1 $ b4 am -sl ZgHVFjAZ1uqEiUa2@hovoldconsulting.com ... $ git am ./20240314_johan_linaro_revert_bluetooth_hci_qca_set_bda_quirk_bit_if_fwnode_exists_in_dt.mbx Applying: Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT" $ b4 am -sl 20240320075554.8178-2-johan+linaro@kernel.org ... $ git am ./v4_20240320_johan_linaro_bluetooth_qca_fix_device_address_endianness.mbx Applying: dt-bindings: bluetooth: add 'qcom,local-bd-address-broken' Applying: arm64: dts: qcom: sc7180-trogdor: mark bluetooth address as broken Applying: Bluetooth: add quirk for broken address properties Applying: Bluetooth: qca: fix device-address endiannessDo you have anything else in your tree which may interfere? What tree is that exactly?
bluetooth-next tree, why would it be anything other than that? All the CI automation is done on bluetooth-next and if you are asking to be done via bluetooth tree which is based on the latest rc that is not how things works here, we usually first apply to bluetooth-next and in case it needs to be backported then it later done via pull-request.
Johan
On Mon, Mar 25, 2024 at 04:31:53PM -0400, Luiz Augusto von Dentz wrote:
On Mon, Mar 25, 2024 at 4:14 PM Johan Hovold johan@kernel.org wrote:
On Mon, Mar 25, 2024 at 04:07:03PM -0400, Luiz Augusto von Dentz wrote:
Probably needs rebasing:
Applying: Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT" error: drivers/bluetooth/hci_qca.c: does not match index Patch failed at 0001 Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT"
I just verified that it applies cleanly to 6.9-rc1.
$ git checkout tmp v6.9-rc1 $ b4 am -sl ZgHVFjAZ1uqEiUa2@hovoldconsulting.com ... $ git am ./20240314_johan_linaro_revert_bluetooth_hci_qca_set_bda_quirk_bit_if_fwnode_exists_in_dt.mbx Applying: Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT" $ b4 am -sl 20240320075554.8178-2-johan+linaro@kernel.org ... $ git am ./v4_20240320_johan_linaro_bluetooth_qca_fix_device_address_endianness.mbx Applying: dt-bindings: bluetooth: add 'qcom,local-bd-address-broken' Applying: arm64: dts: qcom: sc7180-trogdor: mark bluetooth address as broken Applying: Bluetooth: add quirk for broken address properties Applying: Bluetooth: qca: fix device-address endiannessDo you have anything else in your tree which may interfere? What tree is that exactly?
bluetooth-next tree, why would it be anything other than that?
I ask because I did not see anything in either the bluetooth or bluetooth-next tree which should interfere.
And I just verified that by applying the revert followed by the series to bluetooth-next. In that order it applies just fine, as expected.
All the CI automation is done on bluetooth-next and if you are asking to be done via bluetooth tree which is based on the latest rc that is not how things works here, we usually first apply to bluetooth-next and in case it needs to be backported then it later done via pull-request.
The revert fixes a regression in 6.7-rc7 and should get to Linus as soon as possible and I assume you have some way to get fixes into mainline for the current development cycle.
The series fixes a critical bug in the Qualcomm driver and should similarly get into mainline as soon as possible to avoid having people unknowingly start relying on the broken behaviour (reversed address). The bug in this case is older, but since the bug is severe and we're only at rc1, I don't think this one should wait for 6.10 either.
Johan
Hi Johan,
On Tue, Mar 26, 2024 at 3:09 AM Johan Hovold johan@kernel.org wrote:
On Mon, Mar 25, 2024 at 04:31:53PM -0400, Luiz Augusto von Dentz wrote:
On Mon, Mar 25, 2024 at 4:14 PM Johan Hovold johan@kernel.org wrote:
On Mon, Mar 25, 2024 at 04:07:03PM -0400, Luiz Augusto von Dentz wrote:
Probably needs rebasing:
Applying: Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT" error: drivers/bluetooth/hci_qca.c: does not match index Patch failed at 0001 Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT"
I just verified that it applies cleanly to 6.9-rc1.
$ git checkout tmp v6.9-rc1 $ b4 am -sl ZgHVFjAZ1uqEiUa2@hovoldconsulting.com ... $ git am ./20240314_johan_linaro_revert_bluetooth_hci_qca_set_bda_quirk_bit_if_fwnode_exists_in_dt.mbx Applying: Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT" $ b4 am -sl 20240320075554.8178-2-johan+linaro@kernel.org ... $ git am ./v4_20240320_johan_linaro_bluetooth_qca_fix_device_address_endianness.mbx Applying: dt-bindings: bluetooth: add 'qcom,local-bd-address-broken' Applying: arm64: dts: qcom: sc7180-trogdor: mark bluetooth address as broken Applying: Bluetooth: add quirk for broken address properties Applying: Bluetooth: qca: fix device-address endiannessDo you have anything else in your tree which may interfere? What tree is that exactly?
bluetooth-next tree, why would it be anything other than that?
I ask because I did not see anything in either the bluetooth or bluetooth-next tree which should interfere.
And I just verified that by applying the revert followed by the series to bluetooth-next. In that order it applies just fine, as expected.
All the CI automation is done on bluetooth-next and if you are asking to be done via bluetooth tree which is based on the latest rc that is not how things works here, we usually first apply to bluetooth-next and in case it needs to be backported then it later done via pull-request.
The revert fixes a regression in 6.7-rc7 and should get to Linus as soon as possible and I assume you have some way to get fixes into mainline for the current development cycle.
Yeah I will send it later today to be included in the next rc release and since it is marked for stable that shall trigger the process of backporting it.
The series fixes a critical bug in the Qualcomm driver and should similarly get into mainline as soon as possible to avoid having people unknowingly start relying on the broken behaviour (reversed address). The bug in this case is older, but since the bug is severe and we're only at rc1, I don't think this one should wait for 6.10 either.
The revert is now pushed, I had to apply it manually though since it didn't apply cleanly, that said the other set still don't apply:
Applying: Bluetooth: qca: fix device-address endianness error: patch failed: drivers/bluetooth/btqca.c:826 error: drivers/bluetooth/btqca.c: patch does not apply error: drivers/bluetooth/hci_qca.c: does not match index Patch failed at 0004 Bluetooth: qca: fix device-address endianness
So please rebase and send a v5.
On Tue, Mar 26, 2024 at 10:17:13AM -0400, Luiz Augusto von Dentz wrote:
On Tue, Mar 26, 2024 at 3:09 AM Johan Hovold johan@kernel.org wrote:
On Mon, Mar 25, 2024 at 04:31:53PM -0400, Luiz Augusto von Dentz wrote:
On Mon, Mar 25, 2024 at 4:14 PM Johan Hovold johan@kernel.org wrote:
I just verified that it applies cleanly to 6.9-rc1.
$ git checkout tmp v6.9-rc1 $ b4 am -sl ZgHVFjAZ1uqEiUa2@hovoldconsulting.com ... $ git am ./20240314_johan_linaro_revert_bluetooth_hci_qca_set_bda_quirk_bit_if_fwnode_exists_in_dt.mbx Applying: Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT" $ b4 am -sl 20240320075554.8178-2-johan+linaro@kernel.org ... $ git am ./v4_20240320_johan_linaro_bluetooth_qca_fix_device_address_endianness.mbx Applying: dt-bindings: bluetooth: add 'qcom,local-bd-address-broken' Applying: arm64: dts: qcom: sc7180-trogdor: mark bluetooth address as broken Applying: Bluetooth: add quirk for broken address properties Applying: Bluetooth: qca: fix device-address endiannessDo you have anything else in your tree which may interfere? What tree is that exactly?
bluetooth-next tree, why would it be anything other than that?
I ask because I did not see anything in either the bluetooth or bluetooth-next tree which should interfere.
And I just verified that by applying the revert followed by the series to bluetooth-next. In that order it applies just fine, as expected.
All the CI automation is done on bluetooth-next and if you are asking to be done via bluetooth tree which is based on the latest rc that is not how things works here, we usually first apply to bluetooth-next and in case it needs to be backported then it later done via pull-request.
The revert fixes a regression in 6.7-rc7 and should get to Linus as soon as possible and I assume you have some way to get fixes into mainline for the current development cycle.
Yeah I will send it later today to be included in the next rc release and since it is marked for stable that shall trigger the process of backporting it.
The series fixes a critical bug in the Qualcomm driver and should similarly get into mainline as soon as possible to avoid having people unknowingly start relying on the broken behaviour (reversed address). The bug in this case is older, but since the bug is severe and we're only at rc1, I don't think this one should wait for 6.10 either.
The revert is now pushed, I had to apply it manually though since it didn't apply cleanly, that said the other set still don't apply:
I have no idea what you did here but you need to drop that commit immediately, it's completely messed up:
https://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next.git...
This is the revert:
https://lore.kernel.org/lkml/20240314084412.1127-1-johan+linaro@kernel.org/
and as you can see your commit bears no resemblance to the revert (instead it includes code from the below series which obviously then fails to apply):
Applying: Bluetooth: qca: fix device-address endianness error: patch failed: drivers/bluetooth/btqca.c:826 error: drivers/bluetooth/btqca.c: patch does not apply error: drivers/bluetooth/hci_qca.c: does not match index Patch failed at 0004 Bluetooth: qca: fix device-address endianness
So please rebase and send a v5.
There should be no need to resend as the problem is clearly on your end.
Just drop whatever you applied and start fresh:
$ b4 am -sl ZgHVFjAZ1uqEiUa2@hovoldconsulting.com ... $ git am ./20240314_johan_linaro_revert_bluetooth_hci_qca_set_bda_quirk_bit_if_fwnode_exists_in_dt.mbx Applying: Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT"
This is the revert from this thread.
$ b4 am -sl 20240320075554.8178-2-johan+linaro@kernel.org ... $ git am ./v4_20240320_johan_linaro_bluetooth_qca_fix_device_address_endianness.mbx Applying: dt-bindings: bluetooth: add 'qcom,local-bd-address-broken' Applying: arm64: dts: qcom: sc7180-trogdor: mark bluetooth address as broken Applying: Bluetooth: add quirk for broken address properties Applying: Bluetooth: qca: fix device-address endianness
And this is the follow-on series that depends on the revert.
Johan
Hi Luiz,
On Tue, Mar 26, 2024 at 04:18:16PM +0100, Johan Hovold wrote:
On Tue, Mar 26, 2024 at 10:17:13AM -0400, Luiz Augusto von Dentz wrote:
On Tue, Mar 26, 2024 at 3:09 AM Johan Hovold johan@kernel.org wrote:
On Mon, Mar 25, 2024 at 04:31:53PM -0400, Luiz Augusto von Dentz wrote:
All the CI automation is done on bluetooth-next and if you are asking to be done via bluetooth tree which is based on the latest rc that is not how things works here, we usually first apply to bluetooth-next and in case it needs to be backported then it later done via pull-request.
The revert fixes a regression in 6.7-rc7 and should get to Linus as soon as possible and I assume you have some way to get fixes into mainline for the current development cycle.
Yeah I will send it later today to be included in the next rc release and since it is marked for stable that shall trigger the process of backporting it.
The series fixes a critical bug in the Qualcomm driver and should similarly get into mainline as soon as possible to avoid having people unknowingly start relying on the broken behaviour (reversed address). The bug in this case is older, but since the bug is severe and we're only at rc1, I don't think this one should wait for 6.10 either.
I just double checked the bluetooth-next branch and everything looks good now (revert + endianness fix series). Thanks!
Did I understand you correctly that you'll be able to get all five commits into 6.9 during this development cycle (e.g. 6.9-rc2)?
Johan
Hi Johan,
On Tue, Mar 26, 2024 at 12:20 PM Johan Hovold johan@kernel.org wrote:
Hi Luiz,
On Tue, Mar 26, 2024 at 04:18:16PM +0100, Johan Hovold wrote:
On Tue, Mar 26, 2024 at 10:17:13AM -0400, Luiz Augusto von Dentz wrote:
On Tue, Mar 26, 2024 at 3:09 AM Johan Hovold johan@kernel.org wrote:
On Mon, Mar 25, 2024 at 04:31:53PM -0400, Luiz Augusto von Dentz wrote:
All the CI automation is done on bluetooth-next and if you are asking to be done via bluetooth tree which is based on the latest rc that is not how things works here, we usually first apply to bluetooth-next and in case it needs to be backported then it later done via pull-request.
The revert fixes a regression in 6.7-rc7 and should get to Linus as soon as possible and I assume you have some way to get fixes into mainline for the current development cycle.
Yeah I will send it later today to be included in the next rc release and since it is marked for stable that shall trigger the process of backporting it.
The series fixes a critical bug in the Qualcomm driver and should similarly get into mainline as soon as possible to avoid having people unknowingly start relying on the broken behaviour (reversed address). The bug in this case is older, but since the bug is severe and we're only at rc1, I don't think this one should wait for 6.10 either.
I just double checked the bluetooth-next branch and everything looks good now (revert + endianness fix series). Thanks!
Did I understand you correctly that you'll be able to get all five commits into 6.9 during this development cycle (e.g. 6.9-rc2)?
Yep, I will be preparing a pull-request with them later this week, there are some other fixes that I want to get in as well.
On Tue, Mar 26, 2024 at 12:58:12PM -0400, Luiz Augusto von Dentz wrote:
On Tue, Mar 26, 2024 at 12:20 PM Johan Hovold johan@kernel.org wrote:
Did I understand you correctly that you'll be able to get all five commits into 6.9 during this development cycle (e.g. 6.9-rc2)?
Yep, I will be preparing a pull-request with them later this week, there are some other fixes that I want to get in as well.
Perfect, thanks!
Johan
On Thu, 14 Mar 2024 09:44:12 +0100 Johan Hovold johan+linaro@kernel.org wrote:
This reverts commit 7dcd3e014aa7faeeaf4047190b22d8a19a0db696.
Qualcomm Bluetooth controllers like WCN6855 do not have persistent storage for the Bluetooth address and must therefore start as unconfigured to allow the user to set a valid address unless one has been provided by the boot firmware in the devicetree.
A recent change snuck into v6.8-rc7 and incorrectly started marking the default (non-unique) address as valid. This specifically also breaks the Bluetooth setup for some user of the Lenovo ThinkPad X13s.
Note that this is the second time Qualcomm breaks the driver this way and that this was fixed last year by commit 6945795bc81a ("Bluetooth: fix use-bdaddr-property quirk"), which also has some further details.
Fixes: 7dcd3e014aa7 ("Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT") Cc: stable@vger.kernel.org # 6.8 Cc: Janaki Ramaiah Thota quic_janathot@quicinc.com Signed-off-by: Johan Hovold johan+linaro@kernel.org
Thanks Johan, this revert does indeed fix Bluetooth for me on the X13s.
Reported-by: Clayton Craft clayton@craftyguy.net Tested-by: Clayton Craft clayton@craftyguy.net
-Clayton
Hello:
This patch was applied to bluetooth/bluetooth-next.git (master) by Luiz Augusto von Dentz luiz.von.dentz@intel.com:
On Thu, 14 Mar 2024 09:44:12 +0100 you wrote:
This reverts commit 7dcd3e014aa7faeeaf4047190b22d8a19a0db696.
Qualcomm Bluetooth controllers like WCN6855 do not have persistent storage for the Bluetooth address and must therefore start as unconfigured to allow the user to set a valid address unless one has been provided by the boot firmware in the devicetree.
[...]
Here is the summary with links: - Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT" https://git.kernel.org/bluetooth/bluetooth-next/c/ac0cf3552972
You are awesome, thank you!
Hi Johan, We made this change to configure the device which supports persistent memory for the BD-Address So to make device functional in both scenarios we are adding a new property in dts file to distinguish persistent and non-persistent support of BD Address and set HCI_QUIRK_USE_BDADDR_PROPERTY bit accordingly
On 3/26/2024 9:00 PM, patchwork-bot+bluetooth@kernel.org wrote:
Hello:
This patch was applied to bluetooth/bluetooth-next.git (master) by Luiz Augusto von Dentz luiz.von.dentz@intel.com:
On Thu, 14 Mar 2024 09:44:12 +0100 you wrote:
This reverts commit 7dcd3e014aa7faeeaf4047190b22d8a19a0db696.
Qualcomm Bluetooth controllers like WCN6855 do not have persistent storage for the Bluetooth address and must therefore start as unconfigured to allow the user to set a valid address unless one has been provided by the boot firmware in the devicetree.
[...]
Here is the summary with links:
- Revert "Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT" https://git.kernel.org/bluetooth/bluetooth-next/c/ac0cf3552972
You are awesome, thank you!
Thanks, Janaki Ram
[ Please wrap your emails at 72 columns or so. ]
On Thu, Mar 28, 2024 at 08:25:16PM +0530, Janaki Ramaiah Thota wrote:
We made this change to configure the device which supports persistent memory for the BD-Address
Can you say something more about which devices support persistent storage for the address? Is that all or just some of the chip variants?
So to make device functional in both scenarios we are adding a new property in dts file to distinguish persistent and non-persistent support of BD Address and set HCI_QUIRK_USE_BDADDR_PROPERTY bit accordingly
Depending on the answer to my questions above, you may be able to infer this from the compatible string and/or you can read out the address from the device and only set the quirk if it's set to the default address.
You should not need to add a new property for this.
Johan
On 3/28/2024 8:53 PM, Johan Hovold wrote: Hi Johan, Thanks for the valuable inputs.
[ Please wrap your emails at 72 columns or so. ]
Noted.
On Thu, Mar 28, 2024 at 08:25:16PM +0530, Janaki Ramaiah Thota wrote:
We made this change to configure the device which supports persistent memory for the BD-Address
Can you say something more about which devices support persistent storage for the address? Is that all or just some of the chip variants?
Most of the devices support persistent storage, and bd-address storage is chosen based on the OEM and Target.
So to make device functional in both scenarios we are adding a new property in dts file to distinguish persistent and non-persistent support of BD Address and set HCI_QUIRK_USE_BDADDR_PROPERTY bit accordingly
Depending on the answer to my questions above, you may be able to infer this from the compatible string and/or you can read out the address from the device and only set the quirk if it's set to the default address.
You should not need to add a new property for this.
Johan
As per my understanding, altering the compatible string may cause duplicate configuration, right ?
Thanks, JanakiRam
On Fri, Mar 29, 2024 at 12:55:40PM +0530, Janaki Ramaiah Thota wrote:
On 3/28/2024 8:53 PM, Johan Hovold wrote:
On Thu, Mar 28, 2024 at 08:25:16PM +0530, Janaki Ramaiah Thota wrote:
We made this change to configure the device which supports persistent memory for the BD-Address
Can you say something more about which devices support persistent storage for the address? Is that all or just some of the chip variants?
Most of the devices support persistent storage, and bd-address storage is chosen based on the OEM and Target.
Can you be more specific about which devices support it (or say which do not)?
Is the address stored in some external eeprom or similar which the OEM can choose to populate?
So to make device functional in both scenarios we are adding a new property in dts file to distinguish persistent and non-persistent support of BD Address and set HCI_QUIRK_USE_BDADDR_PROPERTY bit accordingly
Depending on the answer to my questions above, you may be able to infer this from the compatible string and/or you can read out the address from the device and only set the quirk if it's set to the default address.
You should not need to add a new property for this.
As per my understanding, altering the compatible string may cause duplicate configuration, right ?
If it's the same device and just a different configuration then we can't use the compatible string for this.
It seems we need a patch like the below to address this. But please provide some more details (e.g. answers to the questions above) so I can determine what the end result should look like.
Johan
From 9719effe80fcc17518131816fdfeb1824cfa08b6 Mon Sep 17 00:00:00 2001 From: Johan Hovold johan+linaro@kernel.org Date: Thu, 20 Apr 2023 14:10:55 +0200 Subject: [PATCH] Bluetooth: btqca: add invalid device address check
Some Qualcomm Bluetooth controllers lack persistent storage for the device address and therefore end up using the default address 00:00:00:00:5a:ad.
Apparently this depends on how the controller has been integrated so we can not use the device type alone to determine when the address is valid.
Instead read back the address during setup() and only set the HCI_QUIRK_USE_BDADDR_PROPERTY flag when needed.
Fixes: de79a9df1692 ("Bluetooth: btqcomsmd: use HCI_QUIRK_USE_BDADDR_PROPERTY") Fixes: e668eb1e1578 ("Bluetooth: hci_core: Don't stop BT if the BD address missing in dts") Fixes: 6945795bc81a ("Bluetooth: fix use-bdaddr-property quirk") Cc: stable@vger.kernel.org # 6.5 Signed-off-by: Johan Hovold johan+linaro@kernel.org --- drivers/bluetooth/btqca.c | 33 +++++++++++++++++++++++++++++++++ drivers/bluetooth/hci_qca.c | 2 -- 2 files changed, 33 insertions(+), 2 deletions(-)
diff --git a/drivers/bluetooth/btqca.c b/drivers/bluetooth/btqca.c index 19cfc342fc7b..15124157372c 100644 --- a/drivers/bluetooth/btqca.c +++ b/drivers/bluetooth/btqca.c @@ -15,6 +15,8 @@
#define VERSION "0.1"
+#define QCA_BDADDR_DEFAULT (&(bdaddr_t) {{ 0xad, 0x5a, 0x00, 0x00, 0x00, 0x00 }}) + int qca_read_soc_version(struct hci_dev *hdev, struct qca_btsoc_version *ver, enum qca_btsoc_type soc_type) { @@ -612,6 +614,35 @@ int qca_set_bdaddr_rome(struct hci_dev *hdev, const bdaddr_t *bdaddr) } EXPORT_SYMBOL_GPL(qca_set_bdaddr_rome);
+static void qca_check_bdaddr(struct hci_dev *hdev) +{ + struct hci_rp_read_bd_addr *bda; + struct sk_buff *skb; + int err; + + if (bacmp(&hdev->public_addr, BDADDR_ANY)) + return; + + skb = __hci_cmd_sync(hdev, HCI_OP_READ_BD_ADDR, 0, NULL, + HCI_INIT_TIMEOUT); + if (IS_ERR(skb)) { + err = PTR_ERR(skb); + bt_dev_err(hdev, "Failed to read device address (%d)", err); + return; + } + + if (skb->len != sizeof(*bda)) { + bt_dev_err(hdev, "Device address length mismatch"); + goto free_skb; + } + + bda = (struct hci_rp_read_bd_addr *)skb->data; + if (!bacmp(&bda->bdaddr, QCA_BDADDR_DEFAULT)) + set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks); +free_skb: + kfree_skb(skb); +} + static void qca_generate_hsp_nvm_name(char *fwname, size_t max_size, struct qca_btsoc_version ver, u8 rom_ver, u16 bid) { @@ -818,6 +849,8 @@ int qca_uart_setup(struct hci_dev *hdev, uint8_t baudrate, break; }
+ qca_check_bdaddr(hdev); + bt_dev_info(hdev, "QCA setup on UART is completed");
return 0; diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c index b266db18c8cc..b621a0a40ea4 100644 --- a/drivers/bluetooth/hci_qca.c +++ b/drivers/bluetooth/hci_qca.c @@ -1908,8 +1908,6 @@ static int qca_setup(struct hci_uart *hu) case QCA_WCN6750: case QCA_WCN6855: case QCA_WCN7850: - set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks); - qcadev = serdev_device_get_drvdata(hu->serdev); if (qcadev->bdaddr_property_broken) set_bit(HCI_QUIRK_BDADDR_PROPERTY_BROKEN, &hdev->quirks);
On 4/3/2024 5:54 PM, Johan Hovold wrote:
On Fri, Mar 29, 2024 at 12:55:40PM +0530, Janaki Ramaiah Thota wrote:
On 3/28/2024 8:53 PM, Johan Hovold wrote:
On Thu, Mar 28, 2024 at 08:25:16PM +0530, Janaki Ramaiah Thota wrote:
We made this change to configure the device which supports persistent memory for the BD-Address
Can you say something more about which devices support persistent storage for the address? Is that all or just some of the chip variants?
Most of the devices support persistent storage, and bd-address storage is chosen based on the OEM and Target.
Can you be more specific about which devices support it (or say which do not)?
We know below BT chipsets supports persistent storage(OTP) for BDA WCN7850, WCN6855, WCN6750
Is the address stored in some external eeprom or similar which the OEM can choose to populate?
This persistent storage is One Time Programmable (OTP) reserved memory which resides in the BT chip.
So to make device functional in both scenarios we are adding a new property in dts file to distinguish persistent and non-persistent support of BD Address and set HCI_QUIRK_USE_BDADDR_PROPERTY bit accordingly
Depending on the answer to my questions above, you may be able to infer this from the compatible string and/or you can read out the address from the device and only set the quirk if it's set to the default address.
You should not need to add a new property for this.
As per my understanding, altering the compatible string may cause duplicate configuration, right ?
Yes, we are correct.
If it's the same device and just a different configuration then we can't use the compatible string for this.
It seems we need a patch like the below to address this. But please provide some more details (e.g. answers to the questions above) so I can determine what the end result should look like.
Johan
From 9719effe80fcc17518131816fdfeb1824cfa08b6 Mon Sep 17 00:00:00 2001 From: Johan Hovold johan+linaro@kernel.org Date: Thu, 20 Apr 2023 14:10:55 +0200 Subject: [PATCH] Bluetooth: btqca: add invalid device address check
Some Qualcomm Bluetooth controllers lack persistent storage for the device address and therefore end up using the default address 00:00:00:00:5a:ad.
Apparently this depends on how the controller has been integrated so we can not use the device type alone to determine when the address is valid.
Instead read back the address during setup() and only set the HCI_QUIRK_USE_BDADDR_PROPERTY flag when needed.
Fixes: de79a9df1692 ("Bluetooth: btqcomsmd: use HCI_QUIRK_USE_BDADDR_PROPERTY") Fixes: e668eb1e1578 ("Bluetooth: hci_core: Don't stop BT if the BD address missing in dts") Fixes: 6945795bc81a ("Bluetooth: fix use-bdaddr-property quirk") Cc: stable@vger.kernel.org # 6.5 Signed-off-by: Johan Hovold johan+linaro@kernel.org
drivers/bluetooth/btqca.c | 33 +++++++++++++++++++++++++++++++++ drivers/bluetooth/hci_qca.c | 2 -- 2 files changed, 33 insertions(+), 2 deletions(-)
diff --git a/drivers/bluetooth/btqca.c b/drivers/bluetooth/btqca.c index 19cfc342fc7b..15124157372c 100644 --- a/drivers/bluetooth/btqca.c +++ b/drivers/bluetooth/btqca.c @@ -15,6 +15,8 @@ #define VERSION "0.1" +#define QCA_BDADDR_DEFAULT (&(bdaddr_t) {{ 0xad, 0x5a, 0x00, 0x00, 0x00, 0x00 }})
- int qca_read_soc_version(struct hci_dev *hdev, struct qca_btsoc_version *ver, enum qca_btsoc_type soc_type) {
@@ -612,6 +614,35 @@ int qca_set_bdaddr_rome(struct hci_dev *hdev, const bdaddr_t *bdaddr) } EXPORT_SYMBOL_GPL(qca_set_bdaddr_rome); +static void qca_check_bdaddr(struct hci_dev *hdev) +{
- struct hci_rp_read_bd_addr *bda;
- struct sk_buff *skb;
- int err;
- if (bacmp(&hdev->public_addr, BDADDR_ANY))
return;- skb = __hci_cmd_sync(hdev, HCI_OP_READ_BD_ADDR, 0, NULL,
HCI_INIT_TIMEOUT);- if (IS_ERR(skb)) {
err = PTR_ERR(skb);bt_dev_err(hdev, "Failed to read device address (%d)", err);return;- }
- if (skb->len != sizeof(*bda)) {
bt_dev_err(hdev, "Device address length mismatch");goto free_skb;- }
- bda = (struct hci_rp_read_bd_addr *)skb->data;
- if (!bacmp(&bda->bdaddr, QCA_BDADDR_DEFAULT))
set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks);+free_skb:
- kfree_skb(skb);
+}
- static void qca_generate_hsp_nvm_name(char *fwname, size_t max_size, struct qca_btsoc_version ver, u8 rom_ver, u16 bid) {
@@ -818,6 +849,8 @@ int qca_uart_setup(struct hci_dev *hdev, uint8_t baudrate, break; }
- qca_check_bdaddr(hdev);
- bt_dev_info(hdev, "QCA setup on UART is completed");
return 0; diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c index b266db18c8cc..b621a0a40ea4 100644 --- a/drivers/bluetooth/hci_qca.c +++ b/drivers/bluetooth/hci_qca.c @@ -1908,8 +1908,6 @@ static int qca_setup(struct hci_uart *hu) case QCA_WCN6750: case QCA_WCN6855: case QCA_WCN7850:
set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks);- qcadev = serdev_device_get_drvdata(hu->serdev); if (qcadev->bdaddr_property_broken) set_bit(HCI_QUIRK_BDADDR_PROPERTY_BROKEN, &hdev->quirks);
Thanks for the patch. This change looks fine and it will resolve the current OTP issue.
-- Thanks, JanakiRam
Hi Johan, Are you planing to merge your below patch ?
On 4/5/2024 6:29 PM, Janaki Ramaiah Thota wrote:
On 4/3/2024 5:54 PM, Johan Hovold wrote:
On Fri, Mar 29, 2024 at 12:55:40PM +0530, Janaki Ramaiah Thota wrote:
On 3/28/2024 8:53 PM, Johan Hovold wrote:
On Thu, Mar 28, 2024 at 08:25:16PM +0530, Janaki Ramaiah Thota wrote:
We made this change to configure the device which supports persistent memory for the BD-Address
Can you say something more about which devices support persistent storage for the address? Is that all or just some of the chip variants?
Most of the devices support persistent storage, and bd-address storage is chosen based on the OEM and Target.
Can you be more specific about which devices support it (or say which do not)?
We know below BT chipsets supports persistent storage(OTP) for BDA WCN7850, WCN6855, WCN6750
Is the address stored in some external eeprom or similar which the OEM can choose to populate?
This persistent storage is One Time Programmable (OTP) reserved memory which resides in the BT chip.
So to make device functional in both scenarios we are adding a new property in dts file to distinguish persistent and non-persistent support of BD Address and set HCI_QUIRK_USE_BDADDR_PROPERTY bit accordingly
Depending on the answer to my questions above, you may be able to infer this from the compatible string and/or you can read out the address from the device and only set the quirk if it's set to the default address.
You should not need to add a new property for this.
As per my understanding, altering the compatible string may cause duplicate configuration, right ?
Yes, we are correct.
If it's the same device and just a different configuration then we can't use the compatible string for this.
It seems we need a patch like the below to address this. But please provide some more details (e.g. answers to the questions above) so I can determine what the end result should look like.
Johan
From 9719effe80fcc17518131816fdfeb1824cfa08b6 Mon Sep 17 00:00:00 2001 From: Johan Hovold johan+linaro@kernel.org Date: Thu, 20 Apr 2023 14:10:55 +0200 Subject: [PATCH] Bluetooth: btqca: add invalid device address check
Some Qualcomm Bluetooth controllers lack persistent storage for the device address and therefore end up using the default address 00:00:00:00:5a:ad.
Apparently this depends on how the controller has been integrated so we can not use the device type alone to determine when the address is valid.
Instead read back the address during setup() and only set the HCI_QUIRK_USE_BDADDR_PROPERTY flag when needed.
Fixes: de79a9df1692 ("Bluetooth: btqcomsmd: use HCI_QUIRK_USE_BDADDR_PROPERTY") Fixes: e668eb1e1578 ("Bluetooth: hci_core: Don't stop BT if the BD address missing in dts") Fixes: 6945795bc81a ("Bluetooth: fix use-bdaddr-property quirk") Cc: stable@vger.kernel.org # 6.5 Signed-off-by: Johan Hovold johan+linaro@kernel.org
drivers/bluetooth/btqca.c | 33 +++++++++++++++++++++++++++++++++ drivers/bluetooth/hci_qca.c | 2 -- 2 files changed, 33 insertions(+), 2 deletions(-)
diff --git a/drivers/bluetooth/btqca.c b/drivers/bluetooth/btqca.c index 19cfc342fc7b..15124157372c 100644 --- a/drivers/bluetooth/btqca.c +++ b/drivers/bluetooth/btqca.c @@ -15,6 +15,8 @@ #define VERSION "0.1" +#define QCA_BDADDR_DEFAULT (&(bdaddr_t) {{ 0xad, 0x5a, 0x00, 0x00, 0x00, 0x00 }})
int qca_read_soc_version(struct hci_dev *hdev, struct qca_btsoc_version *ver, enum qca_btsoc_type soc_type) { @@ -612,6 +614,35 @@ int qca_set_bdaddr_rome(struct hci_dev *hdev, const bdaddr_t *bdaddr) } EXPORT_SYMBOL_GPL(qca_set_bdaddr_rome); +static void qca_check_bdaddr(struct hci_dev *hdev) +{ + struct hci_rp_read_bd_addr *bda; + struct sk_buff *skb; + int err;
+ if (bacmp(&hdev->public_addr, BDADDR_ANY)) + return;
+ skb = __hci_cmd_sync(hdev, HCI_OP_READ_BD_ADDR, 0, NULL, + HCI_INIT_TIMEOUT); + if (IS_ERR(skb)) { + err = PTR_ERR(skb); + bt_dev_err(hdev, "Failed to read device address (%d)", err); + return; + }
+ if (skb->len != sizeof(*bda)) { + bt_dev_err(hdev, "Device address length mismatch"); + goto free_skb; + }
+ bda = (struct hci_rp_read_bd_addr *)skb->data; + if (!bacmp(&bda->bdaddr, QCA_BDADDR_DEFAULT)) + set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks); +free_skb: + kfree_skb(skb); +}
static void qca_generate_hsp_nvm_name(char *fwname, size_t max_size, struct qca_btsoc_version ver, u8 rom_ver, u16 bid) { @@ -818,6 +849,8 @@ int qca_uart_setup(struct hci_dev *hdev, uint8_t baudrate, break; } + qca_check_bdaddr(hdev);
bt_dev_info(hdev, "QCA setup on UART is completed"); return 0; diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c index b266db18c8cc..b621a0a40ea4 100644 --- a/drivers/bluetooth/hci_qca.c +++ b/drivers/bluetooth/hci_qca.c @@ -1908,8 +1908,6 @@ static int qca_setup(struct hci_uart *hu) case QCA_WCN6750: case QCA_WCN6855: case QCA_WCN7850: - set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks);
qcadev = serdev_device_get_drvdata(hu->serdev); if (qcadev->bdaddr_property_broken) set_bit(HCI_QUIRK_BDADDR_PROPERTY_BROKEN, &hdev->quirks);
Thanks for the patch. This change looks fine and it will resolve the current OTP issue.
-- Thanks, JanakiRam
Regards, Janaki Ram
On Mon, Apr 15, 2024 at 01:20:36PM +0200, Johan Hovold wrote:
On Mon, Apr 15, 2024 at 04:22:51PM +0530, Janaki Ramaiah Thota wrote:
Are you planing to merge your below patch ?
Yes, sorry about the delay. Was busy with other things last week.
I'll revisit and post it shortly.
For the record, I've now posted a fix for this here:
https://lore.kernel.org/r/20240416091509.19995-1-johan+linaro@kernel.org
Johan
linux-stable-mirror@lists.linaro.org
-
Clayton Craft -
Janaki Ramaiah Thota -
Johan Hovold -
Johan Hovold -
Linux regression tracking (Thorsten Leemhuis) -
Luiz Augusto von Dentz -
patchwork-bot+bluetooth＠kernel.org