lkml.org 
[lkml]   [2020]   [Jun]   [22]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH v2 0/3] Preventing job distribution to isolated CPUs
From
Date

On 6/22/20 7:45 PM, Nitesh Narayan Lal wrote:
>
> Testing
> =======
> * Patch 1:
> Fix for cpumask_local_spread() is tested by creating VFs, loading
> iavf module and by adding a tracepoint to confirm that only housekeeping
> CPUs are picked when an appropriate profile is set up and all remaining
> CPUs when no CPU isolation is configured.
>
> * Patch 2:
> To test the PCI fix, I hotplugged a virtio-net-pci from qemu console
> and forced its addition to a specific node to trigger the code path that
> includes the proposed fix and verified that only housekeeping CPUs
> are included via tracepoint.
>
> * Patch 3:
> To test the fix in store_rps_map(), I tried configuring an isolated
> CPU by writing to /sys/class/net/en*/queues/rx*/rps_cpus which
> resulted in 'write error: Invalid argument' error. For the case
> where a non-isolated CPU is writing in rps_cpus the above operation
> succeeded without any error.
>
>
> Changes from v1:
> ===============
> - Included the suggestions made by Bjorn Helgaas in the commit messages.
> - Included the 'Reviewed-by' and 'Acked-by' received for Patch-2.
>
> [1] https://patchwork.ozlabs.org/project/netdev/patch/51102eebe62336c6a4e584c7a503553b9f90e01c.camel@marvell.com/
>
> Alex Belits (3):
> lib: Restrict cpumask_local_spread to houskeeping CPUs
> PCI: Restrict probe functions to housekeeping CPUs
> net: Restrict receive packets queuing to housekeeping CPUs
>
> drivers/pci/pci-driver.c | 5 ++++-
> lib/cpumask.c | 43 +++++++++++++++++++++++-----------------
> net/core/net-sysfs.c | 10 +++++++++-
> 3 files changed, 38 insertions(+), 20 deletions(-)
>
> --
>

Hi,

It seems that the cover email got messed up while I was sending the patches.
I am putting my intended cover-email below for now. I can send a v3 with proper
cover-email if needed. The reason, I am not sending it right now, is that if I
get some comments in my patches I will prefer including them as well in my
v3 posting.


"
This patch-set is originated from one of the patches that have been
posted earlier as a part of "Task_isolation" mode [1] patch series
by Alex Belits <abelits@marvell.com>. There are only a couple of
changes that I am proposing in this patch-set compared to what Alex
has posted earlier.


Context
=======
On a broad level, all three patches that are included in this patch
set are meant to improve the driver/library to respect isolated
CPUs by not pinning any job on it. Not doing so could impact
the latency values in RT use-cases.


Patches
=======
* Patch1:
  The first patch is meant to make cpumask_local_spread()
  aware of the isolated CPUs. It ensures that the CPUs that
  are returned by this API only includes housekeeping CPUs.

* Patch2:
  This patch ensures that a probe function that is called
  using work_on_cpu() doesn't run any task on an isolated CPU.

* Patch3:
  This patch makes store_rps_map() aware of the isolated
  CPUs so that rps don't queue any jobs on an isolated CPU.


Proposed Changes
================
To fix the above-mentioned issues Alex has used housekeeping_cpumask().
The only changes that I am proposing here are:
- Removing the dependency on CONFIG_TASK_ISOLATION that was proposed by
  Alex. As it should be safe to rely on housekeeping_cpumask()
  even when we don't have any isolated CPUs and we want
  to fall back to using all available CPUs in any of the above scenarios.
- Using both HK_FLAG_DOMAIN and HK_FLAG_WQ in all three patches, this is
  because we would want the above fixes not only when we have isolcpus but
  also with something like systemd's CPU affinity.


Testing
=======
* Patch 1:
  Fix for cpumask_local_spread() is tested by creating VFs, loading
  iavf module and by adding a tracepoint to confirm that only housekeeping
  CPUs are picked when an appropriate profile is set up and all remaining
  CPUs when no CPU isolation is configured.

* Patch 2:
  To test the PCI fix, I hotplugged a virtio-net-pci from qemu console
  and forced its addition to a specific node to trigger the code path that
  includes the proposed fix and verified that only housekeeping CPUs
  are included via tracepoint.

* Patch 3:
  To test the fix in store_rps_map(), I tried configuring an isolated
  CPU by writing to /sys/class/net/en*/queues/rx*/rps_cpus which
  resulted in 'write error: Invalid argument' error. For the case
  where a non-isolated CPU is writing in rps_cpus the above operation
  succeeded without any error.


Changes from v1: [2]
===============
- Included the suggestions made by Bjorn Helgaas in the commit message.
- Included the 'Reviewed-by' and 'Acked-by' received for Patch-2.

[1]
https://patchwork.ozlabs.org/project/netdev/patch/51102eebe62336c6a4e584c7a503553b9f90e01c.camel@marvell.com/
[2]
https://patchwork.ozlabs.org/project/linux-pci/cover/20200610161226.424337-1-nitesh@redhat.com/

Alex Belits (3):
  lib: Restrict cpumask_local_spread to houskeeping CPUs
  PCI: Restrict probe functions to housekeeping CPUs
  net: Restrict receive packets queuing to housekeeping CPUs

 drivers/pci/pci-driver.c |  5 ++++-
 lib/cpumask.c            | 43 +++++++++++++++++++++++-----------------
 net/core/net-sysfs.c     | 10 +++++++++-
 3 files changed, 38 insertions(+), 20 deletions(-)

--
"

--
Thanks
Nitesh

[unhandled content-type:application/pgp-signature]
\
 
 \ /
  Last update: 2020-06-23 03:05    [W:0.112 / U:1.256 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site