From: Prarit Bhargava <prarit@redhat.com>
To: linux-pm@vger.kernel.org
Cc: lenb@kernel.org, len.brown@intel.com, charles.rose@dell.com,
rafael@kernel.org, suravee.suthikulpanit@amd.com,
Prarit Bhargava <prarit@redhat.com>
Subject: [RESEND PATCH 0/8] turbostat: Fix AMD output by making turbostat aware of nodes
Date: Fri, 3 Nov 2017 08:25:42 -0400 [thread overview]
Message-ID: <20171103122550.13341-1-prarit@redhat.com> (raw)
3rd try on this ...
Still haven't heard anything after getting comments on v1... resending.
AMD Ryzen, etc., is gaining more traction. I'm seeing more reports of this
problem. Cc'ing Dell & AMD engineers
P.
---8<---
AMD family processors do not show all cores in the output of turbostat. This
occurs because AMD has multiple nodes per socket and enumerates cores
within each node from 0. For example, socket 0 may have two nodes (0 and 1)
and those nodes both have cores enumerated from 0 through 7. turbostat cannot
handle this configuration, and as a result only shows 1/2 the cores in its
output.
This patchset makes turbostate aware of nodes. It has been tested on
various AMD and Intel systems and no issues have been found.
v2: Fix overwriting of thread_id in 3/8. Add Node output to turbostat (patch
8/8). Various checkpatch.pl cleanups.
Prarit Bhargava (8):
turbostat: set max_num_cpus equal to the cpumask length
turbostat: Fix node and siblings lookup data
turbostat: Calculate additional node information for a package
turbostat: track thread ID in cpu_topology
turbostat: rename num_cores_per_pkg to num_cores_per_node
turbostat: remove num_ from cpu_topology struct
turbostat: add node information into turbostat calculations
turbostat: Add Node in output
tools/power/x86/turbostat/turbostat.c | 449 ++++++++++++++++++++++------------
1 file changed, 289 insertions(+), 160 deletions(-)
--
2.15.0.rc0.39.g2f0e14e64
next reply other threads:[~2017-11-03 12:26 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-11-03 12:25 Prarit Bhargava [this message]
2017-11-03 12:25 ` [RESEND PATCH 1/8] turbostat: set max_num_cpus equal to the cpumask length Prarit Bhargava
2017-11-03 12:25 ` [RESEND PATCH 2/8] turbostat: Fix node and siblings lookup data Prarit Bhargava
2017-12-08 22:47 ` Len Brown
2017-12-09 11:25 ` Prarit Bhargava
2018-01-03 12:58 ` Prarit Bhargava
2018-01-28 17:43 ` Len Brown
2018-01-28 23:16 ` Prarit Bhargava
2017-11-03 12:25 ` [RESEND PATCH 3/8] turbostat: Calculate additional node information for a package Prarit Bhargava
2017-11-03 12:25 ` [RESEND PATCH 4/8] turbostat: track thread ID in cpu_topology Prarit Bhargava
2017-11-03 12:25 ` [RESEND PATCH 5/8] turbostat: rename num_cores_per_pkg to num_cores_per_node Prarit Bhargava
2017-11-03 12:25 ` [RESEND PATCH 6/8] turbostat: remove num_ from cpu_topology struct Prarit Bhargava
2017-11-03 12:25 ` [RESEND PATCH 7/8] turbostat: add node information into turbostat calculations Prarit Bhargava
2017-11-03 12:25 ` [RESEND PATCH 8/8] turbostat: Add Node in output Prarit Bhargava
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=20171103122550.13341-1-prarit@redhat.com \
--to=prarit@redhat.com \
--cc=charles.rose@dell.com \
--cc=len.brown@intel.com \
--cc=lenb@kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=rafael@kernel.org \
--cc=suravee.suthikulpanit@amd.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).