From: John Kacur <jkacur@redhat.com>
To: Tomas Glozar <tglozar@redhat.com>
Cc: linux-rt-users@vger.kernel.org
Subject: Re: [PATCH 3/6] rteval: Exclude isolcpus from kcompile by default
Date: Tue, 25 Jul 2023 09:40:52 -0400 (EDT) [thread overview]
Message-ID: <9f54b61a-50fc-b3f0-5cdb-9037d35f530@redhat.com> (raw)
In-Reply-To: <20230630091951.916865-4-tglozar@redhat.com>
On Fri, 30 Jun 2023, Tomas Glozar wrote:
> Allows correct calculation of make job count that does not include
> isolated CPUs, on which the loads won't be scheduled.
>
> Signed-off-by: Tomas Glozar <tglozar@redhat.com>
> ---
> rteval/modules/loads/kcompile.py | 8 ++++++--
> 1 file changed, 6 insertions(+), 2 deletions(-)
>
> diff --git a/rteval/modules/loads/kcompile.py b/rteval/modules/loads/kcompile.py
> index 35ee5cb..30c5965 100644
> --- a/rteval/modules/loads/kcompile.py
> +++ b/rteval/modules/loads/kcompile.py
> @@ -37,6 +37,7 @@ from rteval.systopology import CpuList, SysTopology
>
> expand_cpulist = CpuList.expand_cpulist
> compress_cpulist = CpuList.compress_cpulist
> +nonisolated_cpulist = CpuList.nonisolated_cpulist
>
> DEFAULT_KERNEL_PREFIX = "linux-6.1"
>
> @@ -55,17 +56,20 @@ class KBuildJob:
> if not os.path.isdir(self.objdir):
> os.mkdir(self.objdir)
>
> + # Exclude isolated CPUs if cpulist not set
> + cpus_available = len(nonisolated_cpulist(self.node.cpus.cpulist))
> +
> if os.path.exists('/usr/bin/numactl') and not cpulist:
> # Use numactl
> self.binder = f'numactl --cpunodebind {int(self.node)}'
> - self.jobs = self.calc_jobs_per_cpu() * len(self.node)
> + self.jobs = self.calc_jobs_per_cpu() * cpus_available
> elif cpulist:
> # Use taskset
> self.jobs = self.calc_jobs_per_cpu() * len(cpulist)
> self.binder = f'taskset -c {compress_cpulist(cpulist)}'
> else:
> # Without numactl calculate number of jobs from the node
> - self.jobs = self.calc_jobs_per_cpu() * len(self.node)
> + self.jobs = self.calc_jobs_per_cpu() * cpus_available
>
> self.runcmd = f"make O={self.objdir} -C {self.kdir} -j{self.jobs}"
> self.cleancmd = f"make O={self.objdir} -C {self.kdir} clean allmodconfig"
> --
> 2.41.0
>
>
Signed-off-by: John Kacur <jkacur@redhat.com>
next prev parent reply other threads:[~2023-07-25 13:41 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-06-30 9:19 [PATCH 0/6] rteval: Handle isolcpus correctly Tomas Glozar
2023-06-30 9:19 ` [PATCH 1/6] rteval: Detect isolcpus in systopology Tomas Glozar
2023-07-25 13:33 ` John Kacur
2023-07-25 13:36 ` John Kacur
2023-06-30 9:19 ` [PATCH 2/6] rteval: Report isolated CPUs Tomas Glozar
2023-07-25 13:38 ` John Kacur
2023-06-30 9:19 ` [PATCH 3/6] rteval: Exclude isolcpus from kcompile by default Tomas Glozar
2023-07-25 13:40 ` John Kacur [this message]
2023-06-30 9:19 ` [PATCH 4/6] rteval: Exclude isolcpus from stressng " Tomas Glozar
2023-07-25 13:42 ` John Kacur
2023-06-30 9:19 ` [PATCH 5/6] rteval: Fix CPU count calculation for hackbench Tomas Glozar
2023-07-25 13:44 ` John Kacur
2023-06-30 9:19 ` [PATCH 6/6] rteval: Exclude isolcpus from loads report Tomas Glozar
2023-07-25 14:32 ` John Kacur
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=9f54b61a-50fc-b3f0-5cdb-9037d35f530@redhat.com \
--to=jkacur@redhat.com \
--cc=linux-rt-users@vger.kernel.org \
--cc=tglozar@redhat.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).