All of lore.kernel.org
 help / color / mirror / Atom feed
From: Nicholas Piggin <npiggin@gmail.com>
To: linuxppc-dev@lists.ozlabs.org
Cc: Nicholas Piggin <npiggin@gmail.com>
Subject: [PATCH 00/13] idle and soft-irq improvements and POWER9 idle optimisation
Date: Sun,  6 Aug 2017 03:02:28 +1000	[thread overview]
Message-ID: <20170805170241.22966-1-npiggin@gmail.com> (raw)

The first 9 patches are a bunch of independent small cleanups
and improvements I've collected. The next 4 are improvements to
POWER9 idle state entry and exit. They significantly change how
we enter idle, now in virtual mode and without ptesync. EC=0 idle
is significantly simplified from there, and does not save anything.

This runs fine on a DD1 test machine, but the hardware has some
problem with performance all over the place, so I'm not able to get
anything meaningful there yet.

Thanks,
Nick

Nicholas Piggin (13):
  powerpc/64s: masked interrupt avoid branch
  powerpc/64s: masked interrupt avoid instruction
  powerpc/64s: masked interrupt returns to kernel so avoid r13 restore
  powerpc/64: cleanup __check_irq_replay
  powerpc/64s: irq replay merge HV and non-HV paths for doorbell replay
  powerpc/64s: irq replay external use the HV handler in HV mode on
    POWER9
  powerpc/64: remove redundant instruction in interrupt replay
  powerpc/64s: irq replay remove spurious irq reason
  powerpc/64: runlatch CTRL[RUN] set optimisation
  powerpc/64s: idle simplify KVM idle on POWER9
  powerpc/64s: idle POWER9 can execute stop without ptesync
  powerpc/64s: idle POWER9 can execute stop in virtual mode
  powerpc/64s: idle ESL=0 stop can avoid all save/restore overhead

 arch/powerpc/include/asm/kvm_book3s_asm.h |  4 +++
 arch/powerpc/kernel/entry_64.S            |  7 +---
 arch/powerpc/kernel/exceptions-64s.S      | 19 +++++-----
 arch/powerpc/kernel/idle_book3s.S         | 60 ++++++++++++++-----------------
 arch/powerpc/kernel/irq.c                 | 47 ++++++++++++------------
 arch/powerpc/kernel/process.c             | 35 +++++++++++++-----
 arch/powerpc/kvm/book3s_hv.c              | 37 ++++++++++++++++---
 arch/powerpc/kvm/book3s_hv_rmhandlers.S   | 32 ++++++++++-------
 8 files changed, 143 insertions(+), 98 deletions(-)

-- 
2.11.0

             reply	other threads:[~2017-08-05 17:03 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-08-05 17:02 Nicholas Piggin [this message]
2017-08-05 17:02 ` [PATCH 01/13] powerpc/64s: masked interrupt avoid branch Nicholas Piggin
2017-08-05 17:02 ` [PATCH 02/13] powerpc/64s: masked interrupt avoid instruction Nicholas Piggin
2017-08-05 17:02 ` [PATCH 03/13] powerpc/64s: masked interrupt returns to kernel so avoid r13 restore Nicholas Piggin
2017-08-05 17:02 ` [PATCH 04/13] powerpc/64: cleanup __check_irq_replay Nicholas Piggin
2017-08-05 17:02 ` [PATCH 05/13] powerpc/64s: irq replay merge HV and non-HV paths for doorbell replay Nicholas Piggin
2017-08-05 17:02 ` [PATCH 06/13] powerpc/64s: irq replay external use the HV handler in HV mode on POWER9 Nicholas Piggin
2017-08-05 17:02 ` [PATCH 07/13] powerpc/64: remove redundant instruction in interrupt replay Nicholas Piggin
2017-08-05 17:02 ` [PATCH 08/13] powerpc/64s: irq replay remove spurious irq reason Nicholas Piggin
2017-08-05 23:00   ` Benjamin Herrenschmidt
2017-08-06  0:51     ` Nicholas Piggin
2017-08-10 13:12       ` Michael Ellerman
2017-08-05 17:02 ` [PATCH 09/13] powerpc/64: runlatch CTRL[RUN] set optimisation Nicholas Piggin
2017-08-05 17:02 ` [PATCH 10/13] powerpc/64s: idle simplify KVM idle on POWER9 Nicholas Piggin
2017-08-05 17:02   ` Nicholas Piggin
2017-08-08 10:36   ` Gautham R Shenoy
2017-08-08 10:48     ` Gautham R Shenoy
2017-08-08 12:42     ` Nicholas Piggin
2017-08-08 12:42       ` Nicholas Piggin
2017-08-10  6:24       ` Gautham R Shenoy
2017-08-10  6:36         ` Gautham R Shenoy
2017-08-10 13:14   ` Michael Ellerman
2017-08-10 13:14     ` Michael Ellerman
2017-08-10 13:53     ` Nicholas Piggin
2017-08-10 13:53       ` Nicholas Piggin
2017-08-05 17:02 ` [PATCH 11/13] powerpc/64s: idle POWER9 can execute stop without ptesync Nicholas Piggin
2017-08-08 10:37   ` Gautham R Shenoy
2017-08-10 13:15   ` Michael Ellerman
2017-08-10 14:03     ` Nicholas Piggin
2017-08-05 17:02 ` [PATCH 12/13] powerpc/64s: idle POWER9 can execute stop in virtual mode Nicholas Piggin
2017-08-05 22:57   ` Benjamin Herrenschmidt
2017-08-08 10:45   ` Gautham R Shenoy
2017-08-05 17:02 ` [PATCH 13/13] powerpc/64s: idle ESL=0 stop can avoid all save/restore overhead Nicholas Piggin
2017-08-05 17:11   ` Nicholas Piggin
2017-08-08 10:49   ` Gautham R Shenoy

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=20170805170241.22966-1-npiggin@gmail.com \
    --to=npiggin@gmail.com \
    --cc=linuxppc-dev@lists.ozlabs.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.