| Message ID | 20211014132206.19220-1-wangyanan55@huawei.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | hw/arm/virt: Introduce cpu topology support | expand |
Ping... for review of the newly added ACPI patches. On 2021/10/14 21:21, Yanan Wang wrote: > Hi, > > This is the latest v8 with update in patch #6 and #8. Now only one > generic reference file for PPTT is added in tests/data/acpi/virt. > Machiel and Igor, please help to have a look, thanks! And sorry for the name typo, definitely should be Michael of course. Thanks, Yanan > Description of this series: > Once the view of an accurate virtual cpu topology is provided to guest, > with a well-designed vCPU pinning to the pCPU we may get a huge benefit, > e.g., the scheduling performance improvement. See Dario Faggioli's > research and the related performance tests in [1] for reference. > > This patch series introduces cpu topology support for Arm platform. > Both cpu-map in DT and ACPI PPTT table are introduced to store the > topology information. And we only describe the topology information > to 6.2 and newer virt machines, considering compatibility. > > [1] https://kvmforum2020.sched.com/event/eE1y/virtual-topology-for-virtual-machines > -friend-or-foe-dario-faggioli-suse > > Series tested locally on Arm64 machines kunpeng920. > After booting a Linux guest with "-smp 16,sockets=4,cores=4,threads=1,maxcpus=16", > through lscpu we will see the information about CPU topology like: > Architecture: aarch64 > Byte Order: Little Endian > CPU(s): 16 > On-line CPU(s) list: 0-15 > Thread(s) per core: 1 > Core(s) per socket: 4 > Socket(s): 4 > NUMA node(s): 1 > Vendor ID: 0x48 > Model: 0 > Stepping: 0x1 > BogoMIPS: 200.00 > NUMA node0 CPU(s): 0-15 > > and with "-smp 16" we will see: > Architecture: aarch64 > Byte Order: Little Endian > CPU(s): 16 > On-line CPU(s) list: 0-15 > Thread(s) per core: 1 > Core(s) per socket: 16 > Socket(s): 1 > NUMA node(s): 1 > Vendor ID: 0x48 > Model: 0 > Stepping: 0x1 > BogoMIPS: 200.00 > NUMA node0 CPU(s): 0-15 > > Changelog: > v7->v8: > - rebased on top of master (commit e5b2333f24) > - only add one generic expected file for PPTT instead of four, which works > fine enough for now (patch #6 and #8 updated) > - v7: https://lore.kernel.org/qemu-devel/20211007030746.10420-1-wangyanan55@huawei.com/ > > v6->v7: > - rebased on top of master (commit ca61fa4b80) > - use newly introduced acpi_table_begin/acpi_table_end APIs to build PPTT > (patch #5 updated) > - add reference files for PPTT to fix broken bios-table-test for Aarch64 > virt machine (patch #6-#8 added) > - v6: https://lore.kernel.org/qemu-devel/20210824122016.144364-1-wangyanan55@huawei.com/ > > Andrew Jones (2): > hw/arm/virt: Add cpu-map to device tree > hw/acpi/aml-build: Add PPTT table > > Yanan Wang (6): > hw/arm/virt: Only describe cpu topology since virt-6.2 > device_tree: Add qemu_fdt_add_path > hw/acpi/aml-build: Add Processor hierarchy node structure > tests/data/acpi/virt: Add an empty expected file for PPTT > hw/arm/virt-acpi-build: Generate PPTT table > tests/data/acpi/virt: Update the empty expected file for PPTT > > hw/acpi/aml-build.c | 86 +++++++++++++++++++++++++++++++++++ > hw/arm/virt-acpi-build.c | 8 +++- > hw/arm/virt.c | 71 +++++++++++++++++++++++++---- > include/hw/acpi/aml-build.h | 7 +++ > include/hw/arm/virt.h | 4 +- > include/sysemu/device_tree.h | 1 + > softmmu/device_tree.c | 44 +++++++++++++++++- > tests/data/acpi/virt/PPTT | Bin 0 -> 76 bytes > 8 files changed, 207 insertions(+), 14 deletions(-) > create mode 100644 tests/data/acpi/virt/PPTT > > -- > 2.19.1 > > .