| Message ID | 20250225-topic-845_smmu_not_coherent-v1-1-98ca9d17471c@oss.qualcomm.com (mailing list archive) |
|---|---|
| State | Accepted |
| Headers | show |
| Series | Revert "arm64: dts: qcom: sdm845: Affirm IDR0.CCTW on apps_smmu" | expand |
On Tue, Feb 25, 2025 at 02:44:40PM +0100, Konrad Dybcio wrote: > From: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com> > > There are reports that the pagetable walker cache coherency is not a > given across the spectrum of SDM845/850 devices, leading to lock-ups > and resets. It works fine on some devices (like the Dragonboard 845c, > but not so much on the Lenovo Yoga C630). > > This unfortunately looks like a fluke in firmware development, where > likely somewhere in the vast hypervisor stack, a change to accommodate > for this was only introduced after the initial software release (which > often serves as a baseline for products). > > Revert the change to avoid additional guesswork around crashes. > > This reverts commit 6b31a9744b8726c69bb0af290f8475a368a4b805. > > Reported-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> > Closes: https://lore.kernel.org/linux-arm-msm/20250215-yoga-dma-coherent-v1-1-2419ee184a81@linaro.org/ > Fixes: 6b31a9744b87 ("arm64: dts: qcom: sdm845: Affirm IDR0.CCTW on apps_smmu") > Signed-off-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com> > --- > arch/arm64/boot/dts/qcom/sdm845.dtsi | 1 - > 1 file changed, 1 deletion(-) Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> Should we enable dma-coherent for RB3 only then?
On Tue, 25 Feb 2025 14:44:40 +0100, Konrad Dybcio wrote: > There are reports that the pagetable walker cache coherency is not a > given across the spectrum of SDM845/850 devices, leading to lock-ups > and resets. It works fine on some devices (like the Dragonboard 845c, > but not so much on the Lenovo Yoga C630). > > This unfortunately looks like a fluke in firmware development, where > likely somewhere in the vast hypervisor stack, a change to accommodate > for this was only introduced after the initial software release (which > often serves as a baseline for products). > > [...] Applied, thanks! [1/1] Revert "arm64: dts: qcom: sdm845: Affirm IDR0.CCTW on apps_smmu" commit: f00db31d235946853fb430de8c6aa1295efc8353 Best regards,
diff --git a/arch/arm64/boot/dts/qcom/sdm845.dtsi b/arch/arm64/boot/dts/qcom/sdm845.dtsi index e0ce804bb1a35047b092d4745988170738a3cd03..d0314cdf0b92fd282915e7184f88b52bf309b2c2 100644 --- a/arch/arm64/boot/dts/qcom/sdm845.dtsi +++ b/arch/arm64/boot/dts/qcom/sdm845.dtsi @@ -5163,7 +5163,6 @@ apps_smmu: iommu@15000000 { <GIC_SPI 341 IRQ_TYPE_LEVEL_HIGH>, <GIC_SPI 342 IRQ_TYPE_LEVEL_HIGH>, <GIC_SPI 343 IRQ_TYPE_LEVEL_HIGH>; - dma-coherent; }; anoc_1_tbu: tbu@150c5000 {