Thread (62 messages) 62 messages, 9 authors, 2020-11-09

Re: [PATCH v5 05/15] mm/frame-vector: Use FOLL_LONGTERM

From: Christoph Hellwig <hch@infradead.org>
Date: 2020-11-04 19:11:17
Also in: kvm, linux-media, linux-mm, linux-samsung-soc, lkml

On Wed, Nov 04, 2020 at 03:02:14PM -0400, Jason Gunthorpe wrote:
I saw that once and I really couldn't make sense of it..
What use is having futex's that only work on THP pages? Confused

CH said there was no case of HAVE_FAST_GUP !ARCH_HAS_PTE_SPECIAL, is
one hidden someplace then?
ARCH_HAS_PTE_SPECIAL is selected by:

arc
arm		(if LPAE)
arm64
mips		(if !(32BIT && CPU_HAS_RIXI))
powerpc
riscv
s390
sh
sparc		(if SPARC64)
x86

HAVE_FAST_GUP is supported by:

arm		(if LPAE)
arm64
mips
powerpc
s390
sh
x86

so mips would be a candidate.  But I think only in theory, as
the options selecting CPU_HAS_RIXI do not select
CPU_SUPPORTS_HUGEPAGES.  Adding the mips maintainer to double check
my logic.

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help