linuxppc-dev.lists.ozlabs.org archive mirror
 help / color / mirror / Atom feed
From: Christophe Leroy <christophe.leroy@c-s.fr>
To: Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	Paul Mackerras <paulus@samba.org>,
	Michael Ellerman <mpe@ellerman.id.au>,
	scottwood@freescale.com, Jonathan Corbet <corbet@lwn.net>
Cc: linux-kernel@vger.kernel.org, linuxppc-dev@lists.ozlabs.org,
	Joakim Tjernlund <joakim.tjernlund@transmode.se>,
	linux-doc@vger.kernel.org
Subject: [PATCH 00/20] powerpc/8xx: Use large pages for RAM and IMMR and other improvments.
Date: Wed, 12 Aug 2015 15:40:52 +0200 (CEST)	[thread overview]
Message-ID: <cover.1439382631.git.christophe.leroy@c-s.fr> (raw)

The main purpose of this patchset is to dramatically reduce the time
spent in DTLB miss handler. This is achieved by:
1/ Mapping RAM with 8M pages
2/ Mapping IMMR with a fixed 512K page

On a live running system (VoIP gateway for Air Trafic Control), over
a 10 minutes period (with 277s idle), we get 87 millions DTLB misses
and approximatly 35 secondes are spent in DTLB handler.
This represents 5.8% of the overall time and even 10.8% of the
non-idle time.
Among those 87 millions DTLB misses, 15% are on user addresses and
85% are on kernel addresses. And within the kernel addresses, 93%
are on addresses from the linear address space and only 7% are on
addresses from the virtual address space.

Once the full patchset applied, the number of DTLB misses during the
period is reduced to 11.8 millions for a duration of 5.8s, which
represents 2% of the non-idle time.

This patch also includes other miscellaneous improvements:
1/ Handling of CPU6 ERRATA directly in mtspr() C macro to reduce code
specific to PPC8xx
2/ Rewrite of a few non critical ASM functions in C
3/ Removal of some unused items

See related patches for details


Christophe Leroy (20):
  powerpc/8xx: Save r3 all the time in DTLB miss handler
  powerpc/8xx: Map linear kernel RAM with 8M pages
  powerpc: Update documentation for noltlbs kernel parameter
  powerpc/8xx: move setup_initial_memory_limit() into 8xx_mmu.c
  powerpc/8xx: Fix vaddr for IMMR early remap
  powerpc32: iounmap() cannot vunmap() area mapped by TLBCAMs either
  powerpc32: refactor x_mapped_by_bats() and x_mapped_by_tlbcam()
    together
  powerpc/8xx: Map IMMR area with 512k page at a fixed address
  powerpc/8xx: show IMMR area in startup memory layout
  powerpc/8xx: CONFIG_PIN_TLB unneeded for CONFIG_PPC_EARLY_DEBUG_CPM
  powerpc/8xx: map 16M RAM at startup
  powerpc/8xx: move 8xx SPRN defines into reg_8xx.h and add some missing
    ones
  powerpc/8xx: Handle CPU6 ERRATA directly in mtspr() macro
  powerpc/8xx: remove special handling of CPU6 errata in set_dec()
  powerpc/8xx: rewrite set_context() in C
  powerpc/8xx: rewrite flush_instruction_cache() in C
  powerpc32: Remove clear_pages()
  powerpc32: Remove useless/wrong MMU:setio progress message
  powerpc/8xx: also use r3 in the ITLB miss in all situations
  powerpc32: remove ioremap_base

 Documentation/kernel-parameters.txt         |   2 +-
 arch/powerpc/Kconfig.debug                  |   2 -
 arch/powerpc/include/asm/mmu-8xx.h          |  26 ++--
 arch/powerpc/include/asm/page_32.h          |   3 +-
 arch/powerpc/include/asm/pgtable-ppc32.h    |   5 +
 arch/powerpc/include/asm/reg.h              |   2 +
 arch/powerpc/include/asm/reg_8xx.h          | 106 ++++++++++++++++
 arch/powerpc/include/asm/time.h             |   6 +-
 arch/powerpc/kernel/head_8xx.S              | 151 +++++++++-------------
 arch/powerpc/kernel/misc_32.S               |  17 +--
 arch/powerpc/kernel/ppc_ksyms_32.c          |   1 -
 arch/powerpc/mm/8xx_mmu.c                   | 190 ++++++++++++++++++++++++++++
 arch/powerpc/mm/Makefile                    |   1 +
 arch/powerpc/mm/init_32.c                   |  23 ----
 arch/powerpc/mm/mem.c                       |   4 +
 arch/powerpc/mm/mmu_decl.h                  |  16 +--
 arch/powerpc/mm/pgtable_32.c                |  54 +++++++-
 arch/powerpc/platforms/embedded6xx/mpc10x.h |   8 --
 18 files changed, 444 insertions(+), 173 deletions(-)
 create mode 100644 arch/powerpc/mm/8xx_mmu.c

-- 
2.1.0

             reply	other threads:[~2015-08-12 13:40 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-08-12 13:40 Christophe Leroy [this message]
2015-08-12 13:40 ` [PATCH 01/20] powerpc/8xx: Save r3 all the time in DTLB miss handler Christophe Leroy
2015-08-12 13:40 ` [PATCH 02/20] powerpc/8xx: Map linear kernel RAM with 8M pages Christophe Leroy
2015-08-13  7:00   ` Segher Boessenkool
2015-08-12 13:40 ` [PATCH 03/20] powerpc: Update documentation for noltlbs kernel parameter Christophe Leroy
2015-08-12 13:41 ` [PATCH 04/20] powerpc/8xx: move setup_initial_memory_limit() into 8xx_mmu.c Christophe Leroy
2015-08-12 13:41 ` [PATCH 05/20] powerpc/8xx: Fix vaddr for IMMR early remap Christophe Leroy
2015-08-12 13:41 ` [PATCH 06/20] powerpc32: iounmap() cannot vunmap() area mapped by TLBCAMs either Christophe Leroy
2015-08-12 13:41 ` [PATCH 07/20] powerpc32: refactor x_mapped_by_bats() and x_mapped_by_tlbcam() together Christophe Leroy
2015-08-12 13:41 ` [PATCH 08/20] powerpc/8xx: Map IMMR area with 512k page at a fixed address Christophe Leroy
2015-08-12 13:41 ` [PATCH 09/20] powerpc/8xx: show IMMR area in startup memory layout Christophe Leroy
2015-08-12 13:41 ` [PATCH 10/20] powerpc/8xx: CONFIG_PIN_TLB unneeded for CONFIG_PPC_EARLY_DEBUG_CPM Christophe Leroy
2015-08-12 13:41 ` [PATCH 11/20] powerpc/8xx: map 16M RAM at startup Christophe Leroy
2015-08-12 13:41 ` [PATCH 12/20] powerpc/8xx: move 8xx SPRN defines into reg_8xx.h and add some missing ones Christophe Leroy
2015-08-12 13:41 ` [PATCH 13/20] powerpc/8xx: Handle CPU6 ERRATA directly in mtspr() macro Christophe Leroy
2015-08-12 13:41 ` [PATCH 14/20] powerpc/8xx: remove special handling of CPU6 errata in set_dec() Christophe Leroy
2015-08-12 13:41 ` [PATCH 15/20] powerpc/8xx: rewrite set_context() in C Christophe Leroy
2015-08-12 13:41 ` [PATCH 16/20] powerpc/8xx: rewrite flush_instruction_cache() " Christophe Leroy
2015-08-12 13:41 ` [PATCH 17/20] powerpc32: Remove clear_pages() Christophe Leroy
2015-08-12 13:41 ` [PATCH 18/20] powerpc32: Remove useless/wrong MMU:setio progress message Christophe Leroy
2015-08-12 13:41 ` [PATCH 19/20] powerpc/8xx: also use r3 in the ITLB miss in all situations Christophe Leroy
2015-08-12 13:41 ` [PATCH 20/20] powerpc32: remove ioremap_base Christophe Leroy

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=cover.1439382631.git.christophe.leroy@c-s.fr \
    --to=christophe.leroy@c-s.fr \
    --cc=benh@kernel.crashing.org \
    --cc=corbet@lwn.net \
    --cc=joakim.tjernlund@transmode.se \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=mpe@ellerman.id.au \
    --cc=paulus@samba.org \
    --cc=scottwood@freescale.com \
    /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).