Benjamin Herrenschmidt <b...@kernel.crashing.org> writes: > On Wed, 2016-07-13 at 15:08 +0530, Aneesh Kumar K.V wrote: >> This switch most of the early feature check to use the non static key >> variant of the function. In later patches we will be switching >> cpu_has_feature and mmu_has_feature to use static keys and we can use >> them only after static key/jump label is initialized. Any check for >> feature before jump label init should be done using this new helper. > > I'm not sure about that. This is converting way way way way more > functions than is needed. Especially if Michael applies my series > there will be very little code run before the patching, really only the > MMU initialization.... >
But then all of them are __init functions and that help in: 1) Avoid adding those location information to the __jump_table section. (I assume that will also have an impact of vmlinux size ?) 2) Simple rule regarding when to use __cpu_has_feature and cpu_has_feature 3) No need to runtime patch lot of these __init code paths which will we throw out after boot. Do you see any draw back in making all the __init functions to use __cpu_has_feature() ? -aneesh _______________________________________________ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
