On Tue, Apr 30, 2024 at 12:24:40AM +0100, Bryan O'Donoghue wrote: > On 29/04/2024 15:01, Marc Gonzalez wrote: > > Work around missing MSA_READY indicator in ath10k driver > > (apply work-around for all msm8998 devices) > > > > CHANGELOG v3 > > - Add a paragraph in binding commit to explain why we use > > a DT property instead of a firmware feature bit. > > - Warn if the "no_msa_ready_indicator" property is true, > > but we actually receive the indicator. > > > > Marc Gonzalez (3): > > dt-bindings: net: wireless: ath10k: add qcom,no-msa-ready-indicator prop > > wifi: ath10k: do not always wait for MSA_READY indicator > > arm64: dts: qcom: msm8998: set qcom,no-msa-ready-indicator for wifi > > > > Documentation/devicetree/bindings/net/wireless/qcom,ath10k.yaml | 5 +++++ > > arch/arm64/boot/dts/qcom/msm8998.dtsi | 1 + > > drivers/net/wireless/ath/ath10k/qmi.c | 11 +++++++++++ > > drivers/net/wireless/ath/ath10k/qmi.h | 1 + > > 4 files changed, 18 insertions(+) > > > > I wonder if you could infer the workaround based on firmware version, > instead of kernel passed flag ? > It's been a while, but I attempted this for the similar workaround for SDM845 RB3 et al. I vaguely remember concluding that the different devices I worked on, had firmware from different branches without any suitable version scheme to use for such comparison - which is why we have "qcom,snoc-host-cap-8bit-quirk;" in those nodes (which apparently is not documented in the yaml). I'd also rather see us ask Marc spending time on some more fruitful exercise... Regards, Bjorn
