All of lore.kernel.org
 help / color / mirror / Atom feed
From: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
To: Shrikanth Hegde <sshegde@linux.vnet.ibm.com>
Cc: Mark Rutland <mark.rutland@arm.com>,
	Valentin Schneider <vschneid@redhat.com>,
	Peter Zijlstra <peterz@infradead.org>,
	linux-kernel@vger.kernel.org, Nicholas Piggin <npiggin@gmail.com>,
	Rohan McLure <rmclure@linux.ibm.com>,
	linuxppc-dev <linuxppc-dev@lists.ozlabs.org>,
	Josh Poimboeuf <jpoimboe@kernel.org>
Subject: Re: [PATCH v2 2/6] powerpc/smp: Enable Asym packing for cores on shared processor
Date: Fri, 20 Oct 2023 11:14:45 +0530	[thread overview]
Message-ID: <20231020054445.GL2194132@linux.vnet.ibm.com> (raw)
In-Reply-To: <45a14ebb-91e7-489d-ad5d-6d39a48bc1f5@linux.vnet.ibm.com>

* Shrikanth Hegde <sshegde@linux.vnet.ibm.com> [2023-10-19 21:26:56]:

> 
> 
> On 10/18/23 10:07 PM, Srikar Dronamraju wrote:
> > If there are shared processor LPARs, underlying Hypervisor can have more
> > virtual cores to handle than actual physical cores.
> > 
> > Starting with Power 9, a core has 2 nearly independent thread groups.
> > On a shared processors LPARs, it helps to pack threads to lesser number
> > of cores so that the overall system performance and utilization
> > improves. PowerVM schedules at a core level. Hence packing to fewer
> > cores helps.
> > 
> > For example: Lets says there are two 8-core Shared LPARs that are
> > actually sharing a 8 Core shared physical pool, each running 8 threads
> > each. Then Consolidating 8 threads to 4 cores on each LPAR would help
> > them to perform better. This is because each of the LPAR will get
> > 100% time to run applications and there will no switching required by
> > the Hypervisor.
> > 
> > To achieve this, enable SD_ASYM_PACKING flag at CACHE, MC and DIE level.
> 
> This would have a conflict with tip/master. 
> DIE has been renamed to PKG and Both changelog and code below should 
> change DIE to PKG. 

Once the changes are part of powerpc/merge, will rebase and accomodate the
changes from DIE to PKG.


-- 
Thanks and Regards
Srikar Dronamraju

WARNING: multiple messages have this Message-ID (diff)
From: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
To: Shrikanth Hegde <sshegde@linux.vnet.ibm.com>
Cc: Mark Rutland <mark.rutland@arm.com>,
	Valentin Schneider <vschneid@redhat.com>,
	Peter Zijlstra <peterz@infradead.org>,
	linux-kernel@vger.kernel.org,
	Rohan McLure <rmclure@linux.ibm.com>,
	Nicholas Piggin <npiggin@gmail.com>,
	linuxppc-dev <linuxppc-dev@lists.ozlabs.org>,
	Josh Poimboeuf <jpoimboe@kernel.org>,
	Michael Ellerman <mpe@ellerman.id.au>
Subject: Re: [PATCH v2 2/6] powerpc/smp: Enable Asym packing for cores on shared processor
Date: Fri, 20 Oct 2023 11:14:45 +0530	[thread overview]
Message-ID: <20231020054445.GL2194132@linux.vnet.ibm.com> (raw)
In-Reply-To: <45a14ebb-91e7-489d-ad5d-6d39a48bc1f5@linux.vnet.ibm.com>

* Shrikanth Hegde <sshegde@linux.vnet.ibm.com> [2023-10-19 21:26:56]:

> 
> 
> On 10/18/23 10:07 PM, Srikar Dronamraju wrote:
> > If there are shared processor LPARs, underlying Hypervisor can have more
> > virtual cores to handle than actual physical cores.
> > 
> > Starting with Power 9, a core has 2 nearly independent thread groups.
> > On a shared processors LPARs, it helps to pack threads to lesser number
> > of cores so that the overall system performance and utilization
> > improves. PowerVM schedules at a core level. Hence packing to fewer
> > cores helps.
> > 
> > For example: Lets says there are two 8-core Shared LPARs that are
> > actually sharing a 8 Core shared physical pool, each running 8 threads
> > each. Then Consolidating 8 threads to 4 cores on each LPAR would help
> > them to perform better. This is because each of the LPAR will get
> > 100% time to run applications and there will no switching required by
> > the Hypervisor.
> > 
> > To achieve this, enable SD_ASYM_PACKING flag at CACHE, MC and DIE level.
> 
> This would have a conflict with tip/master. 
> DIE has been renamed to PKG and Both changelog and code below should 
> change DIE to PKG. 

Once the changes are part of powerpc/merge, will rebase and accomodate the
changes from DIE to PKG.


-- 
Thanks and Regards
Srikar Dronamraju

  reply	other threads:[~2023-10-20  5:46 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-10-18 16:37 [PATCH v2 0/6] powerpc/smp: Shared processor sched optimizations Srikar Dronamraju
2023-10-18 16:37 ` [PATCH v2 1/6] powerpc/smp: Cache CPU has Asymmetric SMP Srikar Dronamraju
2023-10-18 16:37   ` Srikar Dronamraju
2023-10-19  4:33   ` Michael Ellerman
2023-10-19  4:33     ` Michael Ellerman
2023-10-20  9:09     ` Srikar Dronamraju
2023-10-20  9:09       ` Srikar Dronamraju
2023-10-18 16:37 ` [PATCH v2 2/6] powerpc/smp: Enable Asym packing for cores on shared processor Srikar Dronamraju
2023-10-18 16:37   ` Srikar Dronamraju
2023-10-19  4:38   ` Michael Ellerman
2023-10-19  4:38     ` Michael Ellerman
2023-10-19  7:48     ` Peter Zijlstra
2023-10-19  7:48       ` Peter Zijlstra
2023-10-19 11:50       ` Michael Ellerman
2023-10-19 11:50         ` Michael Ellerman
2023-10-19 12:54       ` Srikar Dronamraju
2023-10-19 12:54         ` Srikar Dronamraju
2023-10-19 15:56   ` Shrikanth Hegde
2023-10-19 15:56     ` Shrikanth Hegde
2023-10-20  5:44     ` Srikar Dronamraju [this message]
2023-10-20  5:44       ` Srikar Dronamraju
2023-10-18 16:37 ` [PATCH v2 3/6] powerpc/smp: Move shared_processor static key to smp.h Srikar Dronamraju
2023-10-18 16:37   ` Srikar Dronamraju
2023-10-19  4:41   ` Michael Ellerman
2023-10-19  4:41     ` Michael Ellerman
2023-10-19  4:41     ` Michael Ellerman
2023-10-19 13:08     ` Srikar Dronamraju
2023-10-19 13:08       ` Srikar Dronamraju
2023-10-20 10:44       ` Michael Ellerman
2023-10-20 10:44         ` Michael Ellerman
2023-10-20 10:44         ` Michael Ellerman
2023-10-19 10:30   ` Shrikanth Hegde
2023-10-19 10:30     ` Shrikanth Hegde
2023-10-18 16:37 ` [PATCH v2 4/6] powerpc/smp: Disable MC domain for shared processor Srikar Dronamraju
2023-10-18 16:37   ` Srikar Dronamraju
2023-10-19  4:48   ` Michael Ellerman
2023-10-19  4:48     ` Michael Ellerman
2023-10-19  7:50     ` Peter Zijlstra
2023-10-19  7:50       ` Peter Zijlstra
2023-10-19 13:23       ` Srikar Dronamraju
2023-10-19 13:23         ` Srikar Dronamraju
2023-10-19 13:16     ` Srikar Dronamraju
2023-10-19 13:16       ` Srikar Dronamraju
2023-10-18 16:37 ` [PATCH v2 5/6] powerpc/smp: Add read_mostly attribute Srikar Dronamraju
2023-10-18 16:37   ` Srikar Dronamraju
2023-10-19  4:49   ` Michael Ellerman
2023-10-19  4:49     ` Michael Ellerman
2023-10-19  7:51   ` Peter Zijlstra
2023-10-19  7:51     ` Peter Zijlstra
2023-10-19 12:56     ` Srikar Dronamraju
2023-10-19 12:56       ` Srikar Dronamraju
2023-10-18 16:37 ` [PATCH v2 6/6] powerpc/smp: Avoid asym packing within thread_group of a core Srikar Dronamraju
2023-10-18 16:37   ` Srikar Dronamraju

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=20231020054445.GL2194132@linux.vnet.ibm.com \
    --to=srikar@linux.vnet.ibm.com \
    --cc=jpoimboe@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=mark.rutland@arm.com \
    --cc=npiggin@gmail.com \
    --cc=peterz@infradead.org \
    --cc=rmclure@linux.ibm.com \
    --cc=sshegde@linux.vnet.ibm.com \
    --cc=vschneid@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 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.