qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Peter Maydell <peter.maydell@linaro.org>
To: qemu-devel@nongnu.org
Subject: [PULL 05/38] hw/arm/virt: Support CPU cluster on ARM virt machine
Date: Thu, 20 Jan 2022 12:35:57 +0000	[thread overview]
Message-ID: <20220120123630.267975-6-peter.maydell@linaro.org> (raw)
In-Reply-To: <20220120123630.267975-1-peter.maydell@linaro.org>

From: Yanan Wang <wangyanan55@huawei.com>

ARM64 machines like Kunpeng Family Server Chips have a level
of hardware topology in which a group of CPU cores share L3
cache tag or L2 cache. For example, Kunpeng 920 typically
has 6 or 8 clusters in each NUMA node (also represent range
of CPU die), and each cluster has 4 CPU cores. All clusters
share L3 cache data, but CPU cores in each cluster share a
local L3 tag.

Running a guest kernel with Cluster-Aware Scheduling on the
Hosts which have physical clusters, if we can design a vCPU
topology with cluster level for guest kernel and then have
a dedicated vCPU pinning, the guest will gain scheduling
performance improvement from cache affinity of CPU cluster.

So let's enable the support for this new parameter on ARM
virt machines. After this patch, we can define a 4-level
CPU hierarchy like: cpus=*,maxcpus=*,sockets=*,clusters=*,
cores=*,threads=*.

Signed-off-by: Yanan Wang <wangyanan55@huawei.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Message-id: 20220107083232.16256-2-wangyanan55@huawei.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 hw/arm/virt.c   |  1 +
 qemu-options.hx | 10 ++++++++++
 2 files changed, 11 insertions(+)

diff --git a/hw/arm/virt.c b/hw/arm/virt.c
index 84c2444fff6..688d8736909 100644
--- a/hw/arm/virt.c
+++ b/hw/arm/virt.c
@@ -2718,6 +2718,7 @@ static void virt_machine_class_init(ObjectClass *oc, void *data)
     hc->unplug_request = virt_machine_device_unplug_request_cb;
     hc->unplug = virt_machine_device_unplug_cb;
     mc->nvdimm_supported = true;
+    mc->smp_props.clusters_supported = true;
     mc->auto_enable_numa_with_memhp = true;
     mc->auto_enable_numa_with_memdev = true;
     mc->default_ram_id = "mach-virt.ram";
diff --git a/qemu-options.hx b/qemu-options.hx
index ec90505d84e..ba3ae6a42aa 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -277,6 +277,16 @@ SRST
 
         -smp 16,sockets=2,dies=2,cores=2,threads=2,maxcpus=16
 
+    The following sub-option defines a CPU topology hierarchy (2 sockets
+    totally on the machine, 2 clusters per socket, 2 cores per cluster,
+    2 threads per core) for ARM virt machines which support sockets/clusters
+    /cores/threads. Some members of the option can be omitted but their values
+    will be automatically computed:
+
+    ::
+
+        -smp 16,sockets=2,clusters=2,cores=2,threads=2,maxcpus=16
+
     Historically preference was given to the coarsest topology parameters
     when computing missing values (ie sockets preferred over cores, which
     were preferred over threads), however, this behaviour is considered
-- 
2.25.1



  parent reply	other threads:[~2022-01-20 16:23 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-20 12:35 [PULL 00/38] target-arm queue Peter Maydell
2022-01-20 12:35 ` [PULL 01/38] hw/arm/virt: KVM: Enable PAuth when supported by the host Peter Maydell
2022-01-20 12:35 ` [PULL 02/38] hw: Move MARVELL_88W8618 Kconfig from audio/ to arm/ Peter Maydell
2022-01-20 12:35 ` [PULL 03/38] hw/arm/musicpal: Fix coding style of code related to MV88W8618 device Peter Maydell
2022-01-20 12:35 ` [PULL 04/38] hw/net: Move MV88W8618 network device out of hw/arm/ directory Peter Maydell
2022-01-20 12:35 ` Peter Maydell [this message]
2022-01-20 12:35 ` [PULL 06/38] hw/arm/virt: Support cluster level in DT cpu-map Peter Maydell
2022-01-20 12:35 ` [PULL 07/38] hw/acpi/aml-build: Improve scalability of PPTT generation Peter Maydell
2022-01-20 12:36 ` [PULL 08/38] tests/acpi/bios-tables-test: Allow changes to virt/PPTT file Peter Maydell
2022-01-20 12:36 ` [PULL 09/38] hw/acpi/aml-build: Support cluster level in PPTT generation Peter Maydell
2022-01-20 12:36 ` [PULL 10/38] tests/acpi/bios-table-test: Update expected virt/PPTT file Peter Maydell
2022-01-20 12:36 ` [PULL 11/38] docs/can: convert to restructuredText Peter Maydell
2022-01-20 12:36 ` [PULL 12/38] virtio-mem: Correct default THP size for ARM64 Peter Maydell
2022-01-20 12:36 ` [PULL 13/38] hw/arm/virt: Support for virtio-mem-pci Peter Maydell
2022-01-20 12:36 ` [PULL 14/38] hw/intc/arm_gic: Implement read of GICC_IIDR Peter Maydell
2022-01-20 12:36 ` [PULL 15/38] hw/intc/arm_gic: Allow reset of the running priority Peter Maydell
2022-01-20 12:36 ` [PULL 16/38] hw/arm/virt: Add a control for the the highmem PCIe MMIO Peter Maydell
2022-01-20 12:36 ` [PULL 17/38] hw/arm/virt: Add a control for the the highmem redistributors Peter Maydell
2022-01-20 12:36 ` [PULL 18/38] hw/arm/virt: Honor highmem setting when computing the memory map Peter Maydell
2022-02-13  5:05   ` Akihiko Odaki
2022-02-13 10:22     ` Marc Zyngier
2022-02-13 10:45       ` Peter Maydell
2022-02-13 11:38         ` Akihiko Odaki
2022-02-13 12:57           ` Peter Maydell
2022-01-20 12:36 ` [PULL 19/38] hw/arm/virt: Use the PA range to compute " Peter Maydell
2022-01-20 12:36 ` [PULL 20/38] hw/arm/virt: Disable highmem devices that don't fit in the PA range Peter Maydell
2022-01-20 12:36 ` [PULL 21/38] hw/arm/virt: Drop superfluous checks against highmem Peter Maydell
2022-01-20 12:36 ` [PULL 22/38] hw/arm: kudo add lm75s behind bus 1 switch at 75 Peter Maydell
2022-01-20 12:36 ` [PULL 23/38] hw/misc/aspeed_i3c.c: Introduce a dummy AST2600 I3C model Peter Maydell
2022-01-20 12:36 ` [PULL 24/38] hw/arm/aspeed: Add the i3c device to the AST2600 SoC Peter Maydell
2022-01-20 12:36 ` [PULL 25/38] hw/intc/arm_gicv3_its: Fix event ID bounds checks Peter Maydell
2022-01-20 12:36 ` [PULL 26/38] hw/intc/arm_gicv3_its: Convert int ID check to num_intids convention Peter Maydell
2022-01-20 12:36 ` [PULL 27/38] hw/intc/arm_gicv3_its: Fix handling of process_its_cmd() return value Peter Maydell
2022-01-20 12:36 ` [PULL 28/38] hw/intc/arm_gicv3_its: Don't use data if reading command failed Peter Maydell
2022-01-20 12:36 ` [PULL 29/38] hw/intc/arm_gicv3_its: Use enum for return value of process_* functions Peter Maydell
2022-01-20 12:36 ` [PULL 30/38] hw/intc/arm_gicv3_its: Fix return codes in process_its_cmd() Peter Maydell
2022-01-20 12:36 ` [PULL 31/38] hw/intc/arm_gicv3_its: Refactor process_its_cmd() to reduce nesting Peter Maydell
2022-01-20 12:36 ` [PULL 32/38] hw/intc/arm_gicv3_its: Fix return codes in process_mapti() Peter Maydell
2022-01-20 12:36 ` [PULL 33/38] hw/intc/arm_gicv3_its: Fix return codes in process_mapc() Peter Maydell
2022-01-20 12:36 ` [PULL 34/38] hw/intc/arm_gicv3_its: Fix return codes in process_mapd() Peter Maydell
2022-01-20 12:36 ` [PULL 35/38] hw/intc/arm_gicv3_its: Factor out "find address of table entry" code Peter Maydell
2022-01-20 12:36 ` [PULL 36/38] hw/intc/arm_gicv3_its: Check indexes before use, not after Peter Maydell
2022-01-20 12:36 ` [PULL 37/38] hw/intc/arm_gicv3_its: Range-check ICID before indexing into collection table Peter Maydell
2022-01-20 12:36 ` [PULL 38/38] hw/intc/arm_gicv3: Check for !MEMTX_OK instead of MEMTX_ERROR Peter Maydell

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=20220120123630.267975-6-peter.maydell@linaro.org \
    --to=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.org \
    /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).