From: steve.capper@linaro.org (Steve Capper)
To: linux-arm-kernel@lists.infradead.org
Subject: [RFC PATCH 0/6] Huge pages for short descriptors on ARM
Date: Fri, 13 Dec 2013 19:05:40 +0000 [thread overview]
Message-ID: <1386961546-10061-1-git-send-email-steve.capper@linaro.org> (raw)
Hello,
This RFC is another attempt at bringing HugeTLB pages and Transparent
Huge Pages (THP) to ARM on short descriptors.
Since my last resend in October I have rebased the code and simplified
it quite a lot. Rather than translate pte representations from software
bits to hardware bits and from hardware bits to software bits, I use
huge_pte_... analogues of the pte functions to deal with huge pages
directly.
There is one small bit of translation that takes place to populate an
appropriate pgprot_t value for the VMA containing the huge page. Once
we have that pgprot_t, we can manipulate huge ptes/pmds as normal with
the bit and modify funcs.
I have tested this series on an Arndale board running 3.13-rc3 with the
cache flushing fixes mentioned at:
http://lists.infradead.org/pipermail/linux-arm-kernel/2013-December/219081.html
The libhugetlbfs checks, LTP and some custom THP PROT_NONE tests were
used to test this series.
To go forward with this series I would really like to know if anyone is
interested in huge pages for short-descriptors on ARM, and whether or
not these patches work well for those people?
As always, I would also appreciate any comments, critique or flames :-).
Cheers,
--
Steve
Steve Capper (6):
mm: hugetlb: Introduce huge_pte_page and huge_pte_present
arm: mm: Adjust the parameters for __sync_icache_dcache
arm: mm: Make mmu_gather aware of huge pages
arm: mm: Compute pgprot values for huge page sections
arm: mm: HugeTLB support for non-LPAE systems
arm: mm: Add Transparent HugePage support for non-LPAE
arch/arm/Kconfig | 4 +-
arch/arm/include/asm/hugetlb-2level.h | 121 ++++++++++++++++++++++++++++++
arch/arm/include/asm/hugetlb-3level.h | 6 ++
arch/arm/include/asm/hugetlb.h | 10 +--
arch/arm/include/asm/pgtable-2level.h | 137 ++++++++++++++++++++++++++++++++--
arch/arm/include/asm/pgtable-3level.h | 6 ++
arch/arm/include/asm/pgtable.h | 9 +--
arch/arm/include/asm/tlb.h | 10 ++-
arch/arm/kernel/head.S | 10 ++-
arch/arm/mm/fault.c | 13 ----
arch/arm/mm/flush.c | 9 +--
arch/arm/mm/fsr-2level.c | 4 +-
arch/arm/mm/hugetlbpage.c | 2 +-
arch/arm/mm/mmu.c | 52 +++++++++++++
include/linux/hugetlb.h | 8 ++
mm/hugetlb.c | 18 ++---
16 files changed, 368 insertions(+), 51 deletions(-)
create mode 100644 arch/arm/include/asm/hugetlb-2level.h
--
1.8.1.4
next reply other threads:[~2013-12-13 19:05 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-12-13 19:05 Steve Capper [this message]
2013-12-13 19:05 ` [RFC PATCH 1/6] mm: hugetlb: Introduce huge_pte_page and huge_pte_present Steve Capper
2013-12-13 19:05 ` [RFC PATCH 2/6] arm: mm: Adjust the parameters for __sync_icache_dcache Steve Capper
2013-12-13 19:05 ` [RFC PATCH 3/6] arm: mm: Make mmu_gather aware of huge pages Steve Capper
2013-12-13 19:05 ` [RFC PATCH 4/6] arm: mm: Compute pgprot values for huge page sections Steve Capper
2013-12-13 19:05 ` [RFC PATCH 5/6] arm: mm: HugeTLB support for non-LPAE systems Steve Capper
2013-12-13 19:05 ` [RFC PATCH 6/6] arm: mm: Add Transparent HugePage support for non-LPAE Steve Capper
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1386961546-10061-1-git-send-email-steve.capper@linaro.org \
--to=steve.capper@linaro.org \
--cc=linux-arm-kernel@lists.infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).