> in the meanwhile, Matt, could you help do some > more work to narrow down the problematic change?
I went the opposite direction and started cherry picking changes from 6.15.y, starting where ath12k diverged from 6.12.y in Sep 2024. I found stability pretty quick and was able to bisect down to a single commit where stability started. See branch https://github.com/mdmower/linux/commits/mdm-6.12.35-ath12k-1/ for the cherry pick history on top of v6.12.35. The commit where stability started is: wifi: ath12k: modify ath12k_mac_op_bss_info_changed() for MLO https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/drivers/net/wireless/ath/ath12k?h=linux-6.15.y&id=afbab6e4e88da68cca94cabfc1604d71db161d42 I know this isn't exactly what you needed (especially since the commit is part of a patch series, not standalone), but maybe it'll help? Logs with ath12k debugging enabled: 1. Last broken revision in my branch: wifi: ath12k: modify ath12k_get_arvif_iter() for MLO https://github.com/mdmower/linux/commit/841b7f4af08f9d80e6db862218d341402dfc9acf https://cmphys.com/ath12k/dmesg-6.12.35-cherrypicks-841b7f4af08f-ath12kdebug.log 2. First fixed revision in my branch: wifi: ath12k: modify ath12k_mac_op_bss_info_changed() for MLO https://github.com/mdmower/linux/commit/931abb9e838e376a53a82c0f638fb63a9d31e737 https://cmphys.com/ath12k/dmesg-6.12.35-cherrypicks-931abb9e838e-ath12kdebug.log I'll continue to test this branch to make sure it's not a false positive, but so far it's looking good.
