From: Ross Zwisler <email@example.com> To: firstname.lastname@example.org Cc: Ross Zwisler <email@example.com>, "Anaczkowski, Lukasz" <firstname.lastname@example.org>, "Box, David E" <email@example.com>, "Kogut, Jaroslaw" <Jaroslaw.Kogut@intel.com>, "Koss, Marcin" <firstname.lastname@example.org>, "Koziej, Artur" <email@example.com>, "Lahtinen, Joonas" <firstname.lastname@example.org>, "Moore, Robert" <email@example.com>, "Nachimuthu, Murugasamy" <firstname.lastname@example.org>, "Odzioba, Lukasz" <email@example.com>, "Rafael J. Wysocki" <firstname.lastname@example.org>, "Rafael J. Wysocki" <email@example.com>, "Schmauss, Erik" <firstname.lastname@example.org>, "Verma, Vishal L" <email@example.com>, "Zheng, Lv" <firstname.lastname@example.org>, Andrew Morton <email@example.com>, Balbir Singh <firstname.lastname@example.org>, Brice Goglin <email@example.com>, Dan Williams <firstname.lastname@example.org>, Dave Hansen <email@example.com>, Jerome Glisse <firstname.lastname@example.org>, John Hubbard <email@example.com>, Len Brown <firstname.lastname@example.org>, Tim Chen <email@example.com>, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com Subject: [PATCH v3 0/3] create sysfs representation of ACPI HMAT Date: Wed, 13 Dec 2017 19:10:16 -0700 [thread overview] Message-ID: <firstname.lastname@example.org> (raw) This is the third revision of my patches adding a sysfs representation of the ACPI Heterogeneous Memory Attribute Table (HMAT). These patches are based on v4.15-rc3 and a working tree can be found here: https://git.kernel.org/pub/scm/linux/kernel/git/zwisler/linux.git/log/?h=hmat_v3 My goal is to get these patches merged for v4.16. Changes from previous version (https://lkml.org/lkml/2017/7/6/749): - Changed "HMEM" to "HMAT" and "hmem" to "hmat" throughout to make sure that this effort doesn't get confused with Jerome's HMM work and to make it clear that this enabling is tightly tied to the ACPI HMAT table. (John Hubbard) - Moved the link in the initiator (i.e. mem_init0/mem_tgt2) from pointing to the "mem_tgt2/local_init" attribute group to instead point at the mem_tgt2 target itself. (Brice Goglin) - Simplified the contents of both the initiators and the targets so that we just symlink to the NUMA node and don't duplicate information. For initiators this means that we no longer enumerate CPUs, and for targets this means that we don't provide physical address start and length information. All of this is already available in the NUMA node directory itself (i.e. /sys/devices/system/node/node0), and it already accounts for the fact that both multiple CPUs and multiple memory regions can be owned by a given NUMA node. Also removed some extra attributes (is_enabled, is_isolated) which I don't think are useful at this point in time. I have tested this against many different configs that I implemented using qemu. --- ==== Quick Summary ==== Platforms exist today which have multiple types of memory attached to a single CPU. These disparate memory ranges have some characteristics in common, such as CPU cache coherence, but they can have wide ranges of performance both in terms of latency and bandwidth. For example, consider a system that contains persistent memory, standard DDR memory and High Bandwidth Memory (HBM), all attached to the same CPU. There could potentially be an order of magnitude or more difference in performance between the slowest and fastest memory attached to that CPU. With the current Linux code NUMA nodes are CPU-centric, so all the memory attached to a given CPU will be lumped into the same NUMA node. This makes it very difficult for userspace applications to understand the performance of different memory ranges on a given CPU. We solve this issue by providing userspace with performance information on individual memory ranges. This performance information is exposed via sysfs: # grep . mem_tgt2/* mem_tgt2/local_init/* 2>/dev/null mem_tgt2/firmware_id:1 mem_tgt2/is_cached:0 mem_tgt2/local_init/read_bw_MBps:40960 mem_tgt2/local_init/read_lat_nsec:50 mem_tgt2/local_init/write_bw_MBps:40960 mem_tgt2/local_init/write_lat_nsec:50 This allows applications to easily find the memory that they want to use. We expect that the existing NUMA APIs will be enhanced to use this new information so that applications can continue to use them to select their desired memory. ==== Lots of Details ==== This patch set provides a sysfs representation of parts of the Heterogeneous Memory Attribute Table (HMAT), newly defined in ACPI 6.2. One major conceptual change in ACPI 6.2 related to this work is that proximity domains no longer need to contain a processor. We can now have memory-only proximity domains, which means that we can now have memory-only Linux NUMA nodes. Here is an example configuration where we have a single processor, one range of regular memory and one range of HBM: +---------------+ +----------------+ | Processor | | Memory | | prox domain 0 +---+ prox domain 1 | | NUMA node 1 | | NUMA node 2 | +-------+-------+ +----------------+ | +-------+----------+ | HBM | | prox domain 2 | | NUMA node 0 | +------------------+ This gives us one initiator (the processor) and two targets (the two memory ranges). Each of these three has its own ACPI proximity domain and associated Linux NUMA node. Note also that while there is a 1:1 mapping from each proximity domain to each NUMA node, the numbers don't necessarily match up. Additionally we can have extra NUMA nodes that don't map back to ACPI proximity domains. The above configuration could also have the processor and one of the two memory ranges sharing a proximity domain and NUMA node, but for the purposes of the HMAT the two memory ranges will need to be separated. The overall goal of this series and of the HMAT is to allow users to identify memory using its performance characteristics. This is complicated by the amount of HMAT data that could be present in very large systems, so in this series we only surface performance information for local (initiator,target) pairings. The changelog for patch 5 discusses this in detail. Ross Zwisler (3): acpi: HMAT support in acpi_parse_entries_array() hmat: add heterogeneous memory sysfs support hmat: add performance attributes MAINTAINERS | 6 + drivers/acpi/Kconfig | 1 + drivers/acpi/Makefile | 1 + drivers/acpi/hmat/Kconfig | 7 + drivers/acpi/hmat/Makefile | 2 + drivers/acpi/hmat/core.c | 797 ++++++++++++++++++++++++++++++++++++ drivers/acpi/hmat/hmat.h | 64 +++ drivers/acpi/hmat/initiator.c | 43 ++ drivers/acpi/hmat/perf_attributes.c | 56 +++ drivers/acpi/hmat/target.c | 55 +++ drivers/acpi/tables.c | 52 ++- 11 files changed, 1073 insertions(+), 11 deletions(-) create mode 100644 drivers/acpi/hmat/Kconfig create mode 100644 drivers/acpi/hmat/Makefile create mode 100644 drivers/acpi/hmat/core.c create mode 100644 drivers/acpi/hmat/hmat.h create mode 100644 drivers/acpi/hmat/initiator.c create mode 100644 drivers/acpi/hmat/perf_attributes.c create mode 100644 drivers/acpi/hmat/target.c -- 2.14.3
next reply other threads:[~2017-12-14 2:10 UTC|newest] Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top 2017-12-14 2:10 Ross Zwisler [this message] 2017-12-14 2:10 ` [PATCH v3 1/3] acpi: HMAT support in acpi_parse_entries_array() Ross Zwisler 2017-12-15 0:49 ` Rafael J. Wysocki 2017-12-15 1:10 ` Dan Williams 2017-12-16 1:53 ` Rafael J. Wysocki 2017-12-16 1:57 ` Dan Williams 2017-12-16 2:15 ` Rafael J. Wysocki 2017-12-14 2:10 ` [PATCH v3 2/3] hmat: add heterogeneous memory sysfs support Ross Zwisler 2017-12-15 0:52 ` Rafael J. Wysocki 2017-12-15 20:53 ` Ross Zwisler 2017-12-14 2:10 ` [PATCH v3 3/3] hmat: add performance attributes Ross Zwisler 2017-12-14 13:00 ` [PATCH v3 0/3] create sysfs representation of ACPI HMAT Michal Hocko 2017-12-18 20:35 ` Ross Zwisler 2017-12-20 16:41 ` Ross Zwisler 2017-12-21 13:18 ` Michal Hocko 2017-12-20 18:19 ` Matthew Wilcox 2017-12-20 20:22 ` Dave Hansen 2017-12-20 21:16 ` Matthew Wilcox 2017-12-20 21:24 ` Ross Zwisler 2017-12-20 22:29 ` Dan Williams 2017-12-20 22:41 ` Ross Zwisler 2017-12-21 20:31 ` Brice Goglin 2017-12-22 22:53 ` Dan Williams 2017-12-22 23:22 ` Ross Zwisler 2017-12-22 23:57 ` Dan Williams 2017-12-23 1:14 ` Rafael J. Wysocki 2017-12-27 9:10 ` Brice Goglin 2017-12-30 6:58 ` Matthew Wilcox 2017-12-30 9:19 ` Brice Goglin 2017-12-20 21:13 ` Ross Zwisler 2017-12-21 1:41 ` Elliott, Robert (Persistent Memory) 2017-12-22 21:46 ` Ross Zwisler 2017-12-21 12:50 ` Michael Ellerman 2017-12-22 3:09 ` Anshuman Khandual 2017-12-22 10:31 ` Kogut, Jaroslaw 2017-12-22 14:37 ` Anshuman Khandual 2017-12-22 17:13 ` Dave Hansen 2017-12-23 5:14 ` Anshuman Khandual 2017-12-22 22:13 ` Ross Zwisler 2017-12-23 6:56 ` Anshuman Khandual 2017-12-22 22:31 ` Ross Zwisler 2017-12-25 2:05 ` Liubo(OS Lab)
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 \ --email@example.com \ --firstname.lastname@example.org \ --cc=Jaroslaw.Kogut@intel.com \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --subject='Re: [PATCH v3 0/3] create sysfs representation of ACPI HMAT' \ /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
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).