All of lore.kernel.org
 help / color / mirror / Atom feed
From: Pavel Tatashin <pasha.tatashin@soleen.com>
To: catalin.marinas@arm.com, will.deacon@arm.com,
	akpm@linux-foundation.org, rppt@linux.vnet.ibm.com,
	mhocko@suse.com, ard.biesheuvel@linaro.org,
	andrew.murray@arm.com, james.morse@arm.com, marc.zyngier@arm.com,
	sboyd@kernel.org, linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org, pasha.tatashin@soleen.com
Subject: [PATCH v3 0/3] Early boot time stamps for arm64
Date: Wed, 26 Dec 2018 11:45:06 -0500	[thread overview]
Message-ID: <20181226164509.22916-1-pasha.tatashin@soleen.com> (raw)

Changelog:
v2 - v3
	Addressed comments from Will Deacon: split into three
	patches, removed hardcoded assumption from vdso, don't
	assign arch_timer_read_counter during early boot.
v1 - v2
	Addressed comments from Marc Zyngier

I made early boot time stamps available for SPARC and X86.

x86:
https://lore.kernel.org/lkml/20180719205545.16512-1-pasha.tatashin@oracle.com

sparc:
https://www.spinics.net/lists/sparclinux/msg18063.html

As discussed at plumbers, I would like to add the same for arm64. The
implementation does not have to be perfect, and should work only when early
clock is easy to determine. arm64 defines a clock register, and thus makes
it easy, but on some platforms frequency register is broken, so if it is
not known, simply don't initialize clock early.

dmesg before:
https://paste.ubuntu.com/p/3pJ5kgJHyN

dmesg after:
https://paste.ubuntu.com/p/RHKGVd9nSM

As seen from the above with base smp_init is finished after 0.47s:
[    0.464585] SMP: Total of 8 processors activated.

But, in reality, 3.2s is missing which is a quiet long considering this is
only 60G domain.


Pavel Tatashin (3):
  arm_arch_timer: add macro for timer nbits
  arm64: vdso: Use ARCH_TIMER_NBITS to calculate mask
  arm64: Early boot time stamps

 arch/arm64/kernel/asm-offsets.c       |  1 +
 arch/arm64/kernel/setup.c             | 25 +++++++++++++++++++++++++
 arch/arm64/kernel/vdso/gettimeofday.S |  3 +--
 drivers/clocksource/arm_arch_timer.c  |  8 ++++----
 include/clocksource/arm_arch_timer.h  |  3 +++
 5 files changed, 34 insertions(+), 6 deletions(-)

-- 
2.20.1


WARNING: multiple messages have this Message-ID (diff)
From: Pavel Tatashin <pasha.tatashin@soleen.com>
To: catalin.marinas@arm.com, will.deacon@arm.com,
	akpm@linux-foundation.org, rppt@linux.vnet.ibm.com,
	mhocko@suse.com, ard.biesheuvel@linaro.org,
	andrew.murray@arm.com, james.morse@arm.com, marc.zyngier@arm.com,
	sboyd@kernel.org, linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org, pasha.tatashin@soleen.com
Subject: [PATCH v3 0/3] Early boot time stamps for arm64
Date: Wed, 26 Dec 2018 11:45:06 -0500	[thread overview]
Message-ID: <20181226164509.22916-1-pasha.tatashin@soleen.com> (raw)

Changelog:
v2 - v3
	Addressed comments from Will Deacon: split into three
	patches, removed hardcoded assumption from vdso, don't
	assign arch_timer_read_counter during early boot.
v1 - v2
	Addressed comments from Marc Zyngier

I made early boot time stamps available for SPARC and X86.

x86:
https://lore.kernel.org/lkml/20180719205545.16512-1-pasha.tatashin@oracle.com

sparc:
https://www.spinics.net/lists/sparclinux/msg18063.html

As discussed at plumbers, I would like to add the same for arm64. The
implementation does not have to be perfect, and should work only when early
clock is easy to determine. arm64 defines a clock register, and thus makes
it easy, but on some platforms frequency register is broken, so if it is
not known, simply don't initialize clock early.

dmesg before:
https://paste.ubuntu.com/p/3pJ5kgJHyN

dmesg after:
https://paste.ubuntu.com/p/RHKGVd9nSM

As seen from the above with base smp_init is finished after 0.47s:
[    0.464585] SMP: Total of 8 processors activated.

But, in reality, 3.2s is missing which is a quiet long considering this is
only 60G domain.


Pavel Tatashin (3):
  arm_arch_timer: add macro for timer nbits
  arm64: vdso: Use ARCH_TIMER_NBITS to calculate mask
  arm64: Early boot time stamps

 arch/arm64/kernel/asm-offsets.c       |  1 +
 arch/arm64/kernel/setup.c             | 25 +++++++++++++++++++++++++
 arch/arm64/kernel/vdso/gettimeofday.S |  3 +--
 drivers/clocksource/arm_arch_timer.c  |  8 ++++----
 include/clocksource/arm_arch_timer.h  |  3 +++
 5 files changed, 34 insertions(+), 6 deletions(-)

-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

             reply	other threads:[~2018-12-26 16:45 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-12-26 16:45 Pavel Tatashin [this message]
2018-12-26 16:45 ` [PATCH v3 0/3] Early boot time stamps for arm64 Pavel Tatashin
2018-12-26 16:45 ` [PATCH v3 1/3] arm_arch_timer: add macro for timer nbits Pavel Tatashin
2018-12-26 16:45   ` Pavel Tatashin
2018-12-26 16:45 ` [PATCH v3 2/3] arm64: vdso: Use ARCH_TIMER_NBITS to calculate mask Pavel Tatashin
2018-12-26 16:45   ` Pavel Tatashin
2018-12-26 16:45 ` [PATCH v3 3/3] arm64: Early boot time stamps Pavel Tatashin
2018-12-26 16:45   ` Pavel Tatashin
2019-01-03 10:51   ` Marc Zyngier
2019-01-03 10:51     ` Marc Zyngier
2019-01-03 19:58     ` Pavel Tatashin
2019-01-03 19:58       ` Pavel Tatashin
2019-01-04 15:39       ` Marc Zyngier
2019-01-04 15:39         ` Marc Zyngier
2019-01-04 16:23         ` Pavel Tatashin
2019-01-04 16:23           ` Pavel Tatashin
2019-01-04 16:49           ` Marc Zyngier
2019-01-04 16:49             ` Marc Zyngier
2019-01-04 20:54             ` Pavel Tatashin
2019-01-04 20:54               ` Pavel Tatashin

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=20181226164509.22916-1-pasha.tatashin@soleen.com \
    --to=pasha.tatashin@soleen.com \
    --cc=akpm@linux-foundation.org \
    --cc=andrew.murray@arm.com \
    --cc=ard.biesheuvel@linaro.org \
    --cc=catalin.marinas@arm.com \
    --cc=james.morse@arm.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=marc.zyngier@arm.com \
    --cc=mhocko@suse.com \
    --cc=rppt@linux.vnet.ibm.com \
    --cc=sboyd@kernel.org \
    --cc=will.deacon@arm.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 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.