On Fri, 10 Mar 2023 17:30:04 +0800 Zhenhua Huang <quic_zhenhuah@xxxxxxxxxxx> wrote: > Kfence only needs its pool to be mapped as page granularity, previous > judgement was a bit over protected. Decouple it from judgement and do > page granularity mapping for kfence pool only [1]. > > To implement this, also relocate the kfence pool allocation before the > linear mapping setting up, arm64_kfence_alloc_pool is to allocate phys > addr, __kfence_pool is to be set after linear mapping set up. Why make this change? What are the benefits? What are the user visible effects? > LINK: [1] https://lore.kernel.org/linux-arm-kernel/1675750519-1064-1-git-send-email-quic_zhenhuah@xxxxxxxxxxx/T/ Chasing the links indicates that "page-granular mapping costed more (2M per 1GB) memory". Please spell all this out in this patch's changelog. btw. this format: Link: https://lore.kernel.org/linux-arm-kernel/1675750519-1064-1-git-send-email-quic_zhenhuah@xxxxxxxxxxx/T/ [1] is conventional.
