From: Andi Kleen <ak@muc.de>
To: discuss@x86-64.org, linux-ia64@vger.kernel.org,
linux-kernel@vger.kernel.org
Cc: lse-tech@projects.sourceforge.net
Subject: [ANNOUNCE] numactl-0.8 - user space NUMA policy interface released
Date: 19 Dec 2005 09:17:14 +0100
Date: Mon, 19 Dec 2005 09:17:14 +0100 [thread overview]
Message-ID: <20051219081714.GA33604@muc.de> (raw)
Sorry for the huge cc list. I will try to create a more focussed
list for numa policy user space in the future.
numactl-0.8 has been released. It consists of libnuma - which
is the prefered NUMA API to the NUMA policy system calls in the
Linux kernel -, numactl - a command line program for to
run programs with a specific NUMA policy and various tools.
The biggest new features compared to 0.6 (0.7 was skipped) are lots
of bug fixes, PPC support and support for parsing numa distances
from sysfs.
It is backwards binary compatible to earlier libnuma releases.
This version has everything I wanted to implement for numactl 1.0.
Unless some bad bugs come up it will be eventually renamed to
numactl-1.0
Please report all problems to ak@suse.de
ftp://ftp.suse.com/pub/people/ak/numa/numactl-0.8.tar.gz
36fb1f81d647d66cf9c119db3c215be4 numactl-0.8.tar.gz
Changes since 0.6.5:
0.7-pre1
- add test/regress2 and some fixes to test programs
- Fix DSO relocation patch for global variables
- Change nodeset sizes back to be binary compatible with SLES9
- Cosmetic changes to manpages (pointed out by Eric S. Raymond)
- Make numa_run_on_node etc. act on current thread only even on NPTL systems
(Dinakar Guniguntala)
- Make numa_no_nodes / numa_all_nodes const (Werner Almesberger)
- Fix up the warnings caused by above change
- Add numa_distance() on systems with ACPI
- remove some obsolete code
- add rdtsc for ppc64
- fix unsigned/unsigned long confusion in cpumasks (Matt Dobson)
- fix CPU_BYTES and rename CPU_WORDS to CPU_LONGS (Matt Dobson)
- Print node distances in numactl
0.8
- hardend numactl command line parsing against bad arguments in some cases
- remove cpumask/nodemask confusion which has become a FAQ:
--cpubind deprecated, added --cpunodebind and --physcpubind= options
print both in --show, old cpumask kept for compatibility
- Fix --show problems
- various fixes for bugs noted by Mike Stroyan (thanks!)
- install set_mempolicy manpage
- various smaller fixes
-Andi Kleen
reply other threads:[~2005-12-19 8:17 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=20051219081714.GA33604@muc.de \
--to=ak@muc.de \
--cc=discuss@x86-64.org \
--cc=linux-ia64@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=lse-tech@projects.sourceforge.net \
/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