qualcommax: migrate wifi configuration (ath10k) device paths for 6.12… #19479

AgustinLorenzo · 2025-07-20T21:54:52Z

Migrate wifi configuration (ath10k) device paths for 6.12 kernel
The device tree PCIe host node name has been changed in the new
6.12 kernel[1]. Hence we have to update the wifi device path to
make sure it can work properly.

This script is based on:
target/linux/qualcommax/ipq807x/base-files/etc/hotplug.d/ieee80211/05-wifi-migrate

[1] https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=linux-6.12.y&id=07299ba2e7d98045e6b522f7c5b97f402b15bc82

FYI @robimarko @Ansuel

The device tree PCIe host node name has been changed in the new 6.12 kernel[1]. Hence we have to update the wifi device path to make sure it can work properly. This script is based on: target/linux/qualcommax/ipq807x/base-files/etc/hotplug.d/ieee80211/05-wifi-migrate [1] https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=linux-6.12.y&id=07299ba2e7d98045e6b522f7c5b97f402b15bc82 Signed-off-by: Shiji Yang <yangshiji66@outlook.com> Link: #18725 Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>

AgustinLorenzo · 2025-07-27T01:38:16Z

Please do not merge this PR. Tomorrow I will try to combine the 6.6 script for both kernel versions.

Djfe · 2025-07-27T09:56:49Z

you can always mark PRs as drafts if you don't want them to be merged, yet

qosmio · 2025-07-29T18:02:03Z

Would this be better as just updating 05-wifi-migrate?

--- a/target/linux/qualcommax/ipq807x/base-files/etc/hotplug.d/ieee80211/05-wifi-migrate
+++ b/target/linux/qualcommax/ipq807x/base-files/etc/hotplug.d/ieee80211/05-wifi-migrate
@@ -30,10 +30,14 @@ check_path()

        to=${path/soc\//soc@0\/}
+       to=${to/pci\//pcie\/}

        # Checks if kernel version is less than 6.6.0, if it is and the path is using the new format,
        # then path should be migrated to the old format. This would allow users on platforms with two partitions,
        # to test 6.1 and 6.6.
-       [ "$(get_linux_version)" -lt "606000" ] && to=${path/soc@0\//soc\/}
+       [ "$(get_linux_version)" -lt "606000" ] && to=${to/soc@0\//soc\/}
+
+       [ "$(get_linux_version)" -lt "612000" ] && to=${to/pcie\//pci\/}

        [ "$path" = "$to" ] || do_migrate_radio "$config" "$path" "$to"
 }

AgustinLorenzo · 2025-07-30T11:08:50Z

Would this be better as just updating 05-wifi-migrate?

--- a/target/linux/qualcommax/ipq807x/base-files/etc/hotplug.d/ieee80211/05-wifi-migrate
+++ b/target/linux/qualcommax/ipq807x/base-files/etc/hotplug.d/ieee80211/05-wifi-migrate
@@ -30,10 +30,14 @@ check_path()

        to=${path/soc\//soc@0\/}
+       to=${to/pci\//pcie\/}

        # Checks if kernel version is less than 6.6.0, if it is and the path is using the new format,
        # then path should be migrated to the old format. This would allow users on platforms with two partitions,
        # to test 6.1 and 6.6.
-       [ "$(get_linux_version)" -lt "606000" ] && to=${path/soc@0\//soc\/}
+       [ "$(get_linux_version)" -lt "606000" ] && to=${to/soc@0\//soc\/}
+
+       [ "$(get_linux_version)" -lt "612000" ] && to=${to/pcie\//pci\/}

        [ "$path" = "$to" ] || do_migrate_radio "$config" "$path" "$to"
 }

Hi @qosmio

Would also be a solution, do you prefer to modify my PR and give credit to yourself or open a new PR?

Regards, Agustin

qosmio · 2025-07-30T19:19:43Z

Not sure how to modify and push to the same PR. But you're welcome to just change it too.

AgustinLorenzo · 2025-07-31T12:14:57Z

Not sure how to modify and push to the same PR. But you're welcome to just change it too.

Hi @qosmio,

Okay, I do it and give you credit for the change :)

I am also seeing with the change to kernel 6.12 that the router leds stay on even though the ports are not linked.

Today I will compile again and include this patch and see if everything is working correctly.

Regards, Agustin

AgustinLorenzo · 2025-07-31T21:03:11Z

Hi @qosmio,

I've applied your changes, I'm going to compile to see how it looks :)

AgustinLorenzo · 2025-08-01T00:48:46Z

The change proposed by @qosmio works correctly, and the LED failure was also corrected in the upstream in this commit: 1de1df1

Link to the build: https://github.com/AgustinLorenzo/openwrt/releases/tag/ipq807x-nsswifi-2025-08-01-0156

robimarko · 2025-08-01T10:55:53Z

Not sure how to modify and push to the same PR. But you're welcome to just change it too.

Hi @qosmio,

Okay, I do it and give you credit for the change :)

I am also seeing with the change to kernel 6.12 that the router leds stay on even though the ports are not linked.

Today I will compile again and include this patch and see if everything is working correctly.

Regards, Agustin

This was fixed via:
1de1df1

AgustinLorenzo · 2025-08-01T14:56:40Z

I just put it in the comment above, @robimarko can you approve the change?

robimarko · 2025-08-01T20:06:46Z

target/linux/qualcommax/ipq807x/base-files/etc/hotplug.d/ieee80211/05-wifi-migrate


 	# Checks if kernel version is less than 6.6.0, if it is and the path is using the new format,
 	# then path should be migrated to the old format. This would allow users on platforms with two partitions,
 	# to test 6.1 and 6.6.
-	[ "$(get_linux_version)" -lt "606000" ] && to=${path/soc@0\//soc\/}
+	[ "$(get_linux_version)" -lt "606000" ] && to=${to/soc@0\//soc\/}


Has this been tested on a device?

Device with 6.12, test yourself: https://github.com/AgustinLorenzo/openwrt/releases/tag/ipq807x-nsswifi-2025-08-01-0156

@robimarko

Migrating a kernel 6.1 config on 6.12

$ cp /etc/config/wireless /etc/config/wireless.old $ ACTION=add sh /etc/hotplug.d/ieee80211/05-wifi-migrate

--- a/etc/config/wireless.old +++ b/etc/config/wireless @@ -1,7 +1,7 @@ config wifi-device '5g_l' option type 'mac80211' - option path 'platform/soc/c000000.wifi' + option path 'platform/soc@0/c000000.wifi' option band '5g' option txpower '18' option country 'US' @@ -13,7 +13,7 @@ config wifi-device '5g_l' config wifi-device '2g' option type 'mac80211' - option path 'platform/soc/c000000.wifi+1' + option path 'platform/soc@0/c000000.wifi+1' option band '2g' option txpower '15' option country 'US' @@ -24,7 +24,7 @@ config wifi-device '2g' config wifi-device '5g_h' option type 'mac80211' - option path 'soc/10000000.pci/pci0001:00/0001:00:00.0/0001:01:00.0' + option path 'soc@0/10000000.pcie/pci0001:00/0001:00:00.0/0001:01:00.0' option band '5g' option txpower '30' option country 'US'

Coaxing kernel version to 6.6

--- a/rom/lib/functions/system.sh +++ b/lib/functions/system.sh @@ -318,6 +318,7 @@ dt_is_enabled() { get_linux_version() { local ver=$(uname -r) + ver=6.6.100 local minor=${ver%\.*} printf "%d%02d%03d" ${ver%%\.*} ${minor#*\.} ${ver##*\.} 2>/dev/null

--- a/etc/config/wireless.old +++ b/etc/config/wireless @@ -1,7 +1,7 @@ config wifi-device '5g_l' option type 'mac80211' - option path 'platform/soc/c000000.wifi' + option path 'platform/soc@0/c000000.wifi' option band '5g' option txpower '18' option country 'US' @@ -13,7 +13,7 @@ config wifi-device '5g_l' config wifi-device '2g' option type 'mac80211' - option path 'platform/soc/c000000.wifi+1' + option path 'platform/soc@0/c000000.wifi+1' option band '2g' option txpower '15' option country 'US' @@ -24,7 +24,7 @@ config wifi-device '2g' config wifi-device '5g_h' option type 'mac80211' - option path 'soc/10000000.pci/pci0001:00/0001:00:00.0/0001:01:00.0' + option path 'soc@0/10000000.pci/pci0001:00/0001:00:00.0/0001:01:00.0' option band '5g' option txpower '30' option country 'US'

Kernel 6.1

--- a/rom/lib/functions/system.sh +++ b/lib/functions/system.sh @@ -318,6 +318,7 @@ dt_is_enabled() { get_linux_version() { local ver=$(uname -r) + ver=6.1.100 local minor=${ver%\.*} printf "%d%02d%03d" ${ver%%\.*} ${minor#*\.} ${ver##*\.} 2>/dev/null

Reverts back to 6.1 config

➤ ACTION=add sh /etc/hotplug.d/ieee80211/05-wifi-migrate ➤ sha /etc/config/wireless.old /etc/config/wireless bca7dd3e742b518acb1da2340f7362bac39613bbe5c56aa3e0d608e0069e5d0d /etc/config/wireless.old bca7dd3e742b518acb1da2340f7362bac39613bbe5c56aa3e0d608e0069e5d0d /etc/config/wireless

AgustinLorenzo · 2025-08-09T11:02:06Z

@robimarko any news from that?

nbd168 · 2025-08-09T14:58:33Z

If you're doing a migration anyway, please migrate to phy names, as done on the mediatek mt7622.
See https://github.com/openwrt/openwrt/blob/main/target/linux/mediatek/mt7622/base-files/etc/board.d/02_network#L42
That way if the path ever changes again, user configurations won't need another migration.

When this is set up properly, phys are automatically renamed to wl0, wl1, ... based on the path in board.json, and those names remain stable, since board.json can change across updates.

AgustinLorenzo · 2025-08-11T13:18:41Z

Hi @nbd168 and @robimarko

It could be an option but I think it would be better right now to approve this PR to correct this problem that happens after the migration to 6.12 (which has already been done in other devices) and for the future to do the PR to migrate it as you said.

Regards

AgustinLorenzo · 2025-08-27T04:59:10Z

@robimarko any news from that? It's a small change for fix an issue with the migration to 6.12

… kernel The device tree PCIe host node name has been changed in the new 6.12 kernel[1]. Hence we have to update the wifi device path to make sure it can work properly. [1] https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=linux-6.12.y&id=07299ba2e7d98045e6b522f7c5b97f402b15bc82 Signed-off-by: Agustin Lorenzo <agustin.lorenzo@thinco.es> Signed-off-by: Sean Khan <datapronix@protonmail.com> Link: openwrt#19479 Signed-off-by: Robert Marko <robimarko@gmail.com>

robimarko · 2025-08-27T07:50:14Z

Thanks! Rebased on top of main and merged!

… kernel The device tree PCIe host node name has been changed in the new 6.12 kernel[1]. Hence we have to update the wifi device path to make sure it can work properly. [1] https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=linux-6.12.y&id=07299ba2e7d98045e6b522f7c5b97f402b15bc82 Signed-off-by: Agustin Lorenzo <agustin.lorenzo@thinco.es> Signed-off-by: Sean Khan <datapronix@protonmail.com> Link: openwrt#19479 Signed-off-by: Robert Marko <robimarko@gmail.com>

… kernel The device tree PCIe host node name has been changed in the new 6.12 kernel[1]. Hence we have to update the wifi device path to make sure it can work properly. [1] https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=linux-6.12.y&id=07299ba2e7d98045e6b522f7c5b97f402b15bc82 Signed-off-by: Agustin Lorenzo <agustin.lorenzo@thinco.es> Signed-off-by: Sean Khan <datapronix@protonmail.com> Link: openwrt/openwrt#19479 Signed-off-by: Robert Marko <robimarko@gmail.com>

github-actions bot added the target/qualcommax pull request/issue for qualcommax target label Jul 20, 2025

AgustinLorenzo force-pushed the main_fix_ath10k_612 branch from 5da1084 to 740c35b Compare July 20, 2025 23:19

AgustinLorenzo marked this pull request as draft July 27, 2025 01:37

AgustinLorenzo force-pushed the main_fix_ath10k_612 branch from 740c35b to b90b402 Compare July 27, 2025 01:37

AgustinLorenzo force-pushed the main_fix_ath10k_612 branch 2 times, most recently from ddbe7f9 to 8c63b91 Compare July 31, 2025 21:02

AgustinLorenzo marked this pull request as ready for review July 31, 2025 21:02

AgustinLorenzo force-pushed the main_fix_ath10k_612 branch from 8c63b91 to f6d530f Compare August 1, 2025 14:56

robimarko reviewed Aug 1, 2025

View reviewed changes

AgustinLorenzo force-pushed the main_fix_ath10k_612 branch from f6d530f to 4fb3785 Compare August 9, 2025 11:02

AgustinLorenzo force-pushed the main_fix_ath10k_612 branch from 4fb3785 to 2beaf0e Compare August 11, 2025 13:16

AgustinLorenzo force-pushed the main_fix_ath10k_612 branch from 2beaf0e to f4ed230 Compare August 27, 2025 04:58

robimarko force-pushed the main_fix_ath10k_612 branch from f4ed230 to 21059cb Compare August 27, 2025 07:50

openwrt-bot merged commit 21059cb into openwrt:main Aug 27, 2025
2 checks passed

AgustinLorenzo deleted the main_fix_ath10k_612 branch August 31, 2025 12:35

Uh oh!

qualcommax: migrate wifi configuration (ath10k) device paths for 6.12… #19479

qualcommax: migrate wifi configuration (ath10k) device paths for 6.12… #19479

Conversation

AgustinLorenzo commented Jul 20, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

AgustinLorenzo commented Jul 27, 2025

Uh oh!

Djfe commented Jul 27, 2025

Uh oh!

qosmio commented Jul 29, 2025

Uh oh!

AgustinLorenzo commented Jul 30, 2025

Uh oh!

qosmio commented Jul 30, 2025

Uh oh!

AgustinLorenzo commented Jul 31, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

AgustinLorenzo commented Jul 31, 2025

Uh oh!

AgustinLorenzo commented Aug 1, 2025

Uh oh!

robimarko commented Aug 1, 2025

Uh oh!

AgustinLorenzo commented Aug 1, 2025

Uh oh!

robimarko Aug 1, 2025

Choose a reason for hiding this comment

Uh oh!

AgustinLorenzo Aug 1, 2025

Choose a reason for hiding this comment

Uh oh!

qosmio Aug 1, 2025

Choose a reason for hiding this comment

Uh oh!

AgustinLorenzo commented Aug 9, 2025

Uh oh!

nbd168 commented Aug 9, 2025

Uh oh!

AgustinLorenzo commented Aug 11, 2025

Uh oh!

AgustinLorenzo commented Aug 27, 2025

Uh oh!

Uh oh!

robimarko commented Aug 27, 2025

Uh oh!

Uh oh!

AgustinLorenzo commented Jul 20, 2025 •

edited

Loading

AgustinLorenzo commented Jul 31, 2025 •

edited

Loading