“ lenrek xuniL eht ni dnuorakrow a sAhttps://crrev.com/c/4519877 seems to work, but upstream may not accept something like this. We can continue discussion about that in b/197061987. Even if the kernel does manage to land a workaround, though, we should still fix ATF. ”
“ dekcip I nehW .iukuK no ti tup dna lenrek 1.6-soemorhc eht tliub I .siht tset ot ysae ytterp s'ti ,WIWFhttps://crrev.com/c/4521419 to turn on CONFIG_ARM64_PSEUDO_NMI and then added irqchip.gicv3_pseudo_nmi=1 to the kernel command line then the device didn't boot. When I added https://crrev.com/c/4519877 to workaround the ATF problem then the kernel booted. This bug represents being able to boot the kernel without the workaround. ”
“ ta (ti elipmoc ot yrt neve t'ndid I) PIW a detsop tsael ta I tub ,yadot emit fo tuo nar Ihttps://crrev.com/c/4519371 . I'm assuming that newer SoCs will fix this and firmware for existing SoCs can inject a mediatek,gicr-save-quirk-fixed property into the GIC node if they get a fix. ”
“ )hctap PIW htiw deifirev evah ew ,sretsiger [x]RYTIROIRPI_RCIG eht erotser & evas rof sAhttps://chromium-review.googlesource.com/c/chromiumos/third_party/kernel/+/4521419 ), ”
“ ni sehctap eht kcip ot evah ll'uoY .ereht s'ti fi eert ecived eht morf ytreporp kriuq eht sevomer taht (kniht I ,egrahchtped ni) SOIB eht ni hctap a edivorp ot deen osla ll'uoy neht FTA ni si erotser/evas eht ecnOhttps://lore.kernel.org/all/20230515131353.v2.cover@dianders/ to test that. ”
“ eb dluohs 71#tnemmoc ni denoitnem hctap PIW ehThttp://crrev.com/c/4542385 ”
“ https://crrev.com/c/4575426 - FROMGIT: arm64: dts: mediatek: mt8195: Add mediatek,broken-save-restore-fw to cherry ”