Excerpts from Edgecombe, Rick P's message of December 5, 2020 4:33 am: > On Fri, 2020-12-04 at 18:12 +1000, Nicholas Piggin wrote: >> Excerpts from Edgecombe, Rick P's message of December 1, 2020 6:21 >> am: >> > On Sun, 2020-11-29 at 01:25 +1000, Nicholas Piggin wrote: >> > > Support huge page vmalloc mappings. Config option >> > > HAVE_ARCH_HUGE_VMALLOC >> > > enables support on architectures that define HAVE_ARCH_HUGE_VMAP >> > > and >> > > supports PMD sized vmap mappings. >> > > >> > > vmalloc will attempt to allocate PMD-sized pages if allocating >> > > PMD >> > > size >> > > or larger, and fall back to small pages if that was unsuccessful. >> > > >> > > Allocations that do not use PAGE_KERNEL prot are not permitted to >> > > use >> > > huge pages, because not all callers expect this (e.g., module >> > > allocations vs strict module rwx). >> > >> > Several architectures (x86, arm64, others?) allocate modules >> > initially >> > with PAGE_KERNEL and so I think this test will not exclude module >> > allocations in those cases. >> >> Ah, thanks. I guess archs must additionally ensure that their >> PAGE_KERNEL allocations are suitable for huge page mappings before >> enabling the option. >> >> If there is interest from those archs to support this, I have an >> early (un-posted) patch that adds an explicit VM_HUGE flag that could >> override the pessemistic arch default. It's not much trouble to add >> this >> to the large system hash allocations. It's very out of date now but >> I >> can at least give what I have to anyone doing an arch support that >> wants it. > > Ahh, sorry, I totally missed that this was only enabled for powerpc. > > That patch might be useful for me actually. Or maybe a VM_NOHUGE, since > there are only a few places where executable vmallocs are created? I'm > not sure what the other issues are. Yeah good question, VM_HUGE might be safer but maybe it would be possible there's only a few problems that have to be annotated with VM_NOHUGE, good point. I'll dig it out and see. > I am endeavoring to have small module allocations share large pages, so > this infrastructure is a big help already. > https://lore.kernel.org/lkml/20201120202426.18009-1-rick.p.edgecombe@xxxxxxxxx/ Oh nice that's what I wanted to do next! We should try get this work for x86 as well then. Thanks, Nick