On Fri, May 8, 2020 at 7:40 AM Joerg Roedel <joro@xxxxxxxxxx> wrote: > > Hi, > > after the recent issue with vmalloc and tracing code[1] on x86 and a > long history of previous issues related to the vmalloc_sync_mappings() > interface, I thought the time has come to remove it. Please > see [2], [3], and [4] for some other issues in the past. > > The patches are based on v5.7-rc4 and add tracking of page-table > directory changes to the vmalloc and ioremap code. Depending on which > page-table levels changes have been made, a new per-arch function is > called: arch_sync_kernel_mappings(). > > On x86-64 with 4-level paging, this function will not be called more > than 64 times in a systems runtime (because vmalloc-space takes 64 PGD > entries which are only populated, but never cleared). What's the maximum on other system types? It might make more sense to take the memory hit and pre-populate all the tables at boot so we never have to sync them.
