summaryrefslogtreecommitdiff
path: root/mm/page_alloc.c
AgeCommit message (Collapse)AuthorFilesLines
2024-03-04mm, vmscan: prevent infinite loop for costly GFP_NOIO | __GFP_RETRY_MAYFAIL ↵Vlastimil Babka1-4/+6
allocations Sven reports an infinite loop in __alloc_pages_slowpath() for costly order __GFP_RETRY_MAYFAIL allocations that are also GFP_NOIO. Such combination can happen in a suspend/resume context where a GFP_KERNEL allocation can have __GFP_IO masked out via gfp_allowed_mask. Quoting Sven: 1. try to do a "costly" allocation (order > PAGE_ALLOC_COSTLY_ORDER) with __GFP_RETRY_MAYFAIL set. 2. page alloc's __alloc_pages_slowpath tries to get a page from the freelist. This fails because there is nothing free of that costly order. 3. page alloc tries to reclaim by calling __alloc_pages_direct_reclaim, which bails out because a zone is ready to be compacted; it pretends to have made a single page of progress. 4. page alloc tries to compact, but this always bails out early because __GFP_IO is not set (it's not passed by the snd allocator, and even if it were, we are suspending so the __GFP_IO flag would be cleared anyway). 5. page alloc believes reclaim progress was made (because of the pretense in item 3) and so it checks whether it should retry compaction. The compaction retry logic thinks it should try again, because: a) reclaim is needed because of the early bail-out in item 4 b) a zonelist is suitable for compaction 6. goto 2. indefinite stall. (end quote) The immediate root cause is confusing the COMPACT_SKIPPED returned from __alloc_pages_direct_compact() (step 4) due to lack of __GFP_IO to be indicating a lack of order-0 pages, and in step 5 evaluating that in should_compact_retry() as a reason to retry, before incrementing and limiting the number of retries. There are however other places that wrongly assume that compaction can happen while we lack __GFP_IO. To fix this, introduce gfp_compaction_allowed() to abstract the __GFP_IO evaluation and switch the open-coded test in try_to_compact_pages() to use it. Also use the new helper in: - compaction_ready(), which will make reclaim not bail out in step 3, so there's at least one attempt to actually reclaim, even if chances are small for a costly order - in_reclaim_compaction() which will make should_continue_reclaim() return false and we don't over-reclaim unnecessarily - in __alloc_pages_slowpath() to set a local variable can_compact, which is then used to avoid retrying reclaim/compaction for costly allocations (step 5) if we can't compact and also to skip the early compaction attempt that we do in some cases Link: https://lkml.kernel.org/r/20240221114357.13655-2-vbabka@suse.cz Fixes: 3250845d0526 ("Revert "mm, oom: prevent premature OOM killer invocation for high order request"") Signed-off-by: Vlastimil Babka <vbabka@suse.cz> Reported-by: Sven van Ashbrook <svenva@chromium.org> Closes: https://lore.kernel.org/all/CAG-rBihs_xMKb3wrMO1%2B-%2Bp4fowP9oy1pa_OTkfxBzPUVOZF%2Bg@mail.gmail.com/ Tested-by: Karthikeyan Ramasubramanian <kramasub@chromium.org> Cc: Brian Geffon <bgeffon@google.com> Cc: Curtis Malainey <cujomalainey@chromium.org> Cc: Jaroslav Kysela <perex@perex.cz> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Michal Hocko <mhocko@kernel.org> Cc: Takashi Iwai <tiwai@suse.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2024-01-11Merge tag 'net-next-6.8' of ↵Linus Torvalds1-0/+7
git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next Pull networking updates from Paolo Abeni: "The most interesting thing is probably the networking structs reorganization and a significant amount of changes is around self-tests. Core & protocols: - Analyze and reorganize core networking structs (socks, netdev, netns, mibs) to optimize cacheline consumption and set up build time warnings to safeguard against future header changes This improves TCP performances with many concurrent connections up to 40% - Add page-pool netlink-based introspection, exposing the memory usage and recycling stats. This helps indentify bad PP users and possible leaks - Refine TCP/DCCP source port selection to no longer favor even source port at connect() time when IP_LOCAL_PORT_RANGE is set. This lowers the time taken by connect() for hosts having many active connections to the same destination - Refactor the TCP bind conflict code, shrinking related socket structs - Refactor TCP SYN-Cookie handling, as a preparation step to allow arbitrary SYN-Cookie processing via eBPF - Tune optmem_max for 0-copy usage, increasing the default value to 128KB and namespecifying it - Allow coalescing for cloned skbs coming from page pools, improving RX performances with some common configurations - Reduce extension header parsing overhead at GRO time - Add bridge MDB bulk deletion support, allowing user-space to request the deletion of matching entries - Reorder nftables struct members, to keep data accessed by the datapath first - Introduce TC block ports tracking and use. This allows supporting multicast-like behavior at the TC layer - Remove UAPI support for retired TC qdiscs (dsmark, CBQ and ATM) and classifiers (RSVP and tcindex) - More data-race annotations - Extend the diag interface to dump TCP bound-only sockets - Conditional notification of events for TC qdisc class and actions - Support for WPAN dynamic associations with nearby devices, to form a sub-network using a specific PAN ID - Implement SMCv2.1 virtual ISM device support - Add support for Batman-avd mulicast packet type BPF: - Tons of verifier improvements: - BPF register bounds logic and range support along with a large test suite - log improvements - complete precision tracking support for register spills - track aligned STACK_ZERO cases as imprecise spilled registers. This improves the verifier "instructions processed" metric from single digit to 50-60% for some programs - support for user's global BPF subprogram arguments with few commonly requested annotations for a better developer experience - support tracking of BPF_JNE which helps cases when the compiler transforms (unsigned) "a > 0" into "if a == 0 goto xxx" and the like - several fixes - Add initial TX metadata implementation for AF_XDP with support in mlx5 and stmmac drivers. Two types of offloads are supported right now, that is, TX timestamp and TX checksum offload - Fix kCFI bugs in BPF all forms of indirect calls from BPF into kernel and from kernel into BPF work with CFI enabled. This allows BPF to work with CONFIG_FINEIBT=y - Change BPF verifier logic to validate global subprograms lazily instead of unconditionally before the main program, so they can be guarded using BPF CO-RE techniques - Support uid/gid options when mounting bpffs - Add a new kfunc which acquires the associated cgroup of a task within a specific cgroup v1 hierarchy where the latter is identified by its id - Extend verifier to allow bpf_refcount_acquire() of a map value field obtained via direct load which is a use-case needed in sched_ext - Add BPF link_info support for uprobe multi link along with bpftool integration for the latter - Support for VLAN tag in XDP hints - Remove deprecated bpfilter kernel leftovers given the project is developed in user-space (https://github.com/facebook/bpfilter) Misc: - Support for parellel TC self-tests execution - Increase MPTCP self-tests coverage - Updated the bridge documentation, including several so-far undocumented features - Convert all the net self-tests to run in unique netns, to avoid random failures due to conflict and allow concurrent runs - Add TCP-AO self-tests - Add kunit tests for both cfg80211 and mac80211 - Autogenerate Netlink families documentation from YAML spec - Add yml-gen support for fixed headers and recursive nests, the tool can now generate user-space code for all genetlink families for which we have specs - A bunch of additional module descriptions fixes - Catch incorrect freeing of pages belonging to a page pool Driver API: - Rust abstractions for network PHY drivers; do not cover yet the full C API, but already allow implementing functional PHY drivers in rust - Introduce queue and NAPI support in the netdev Netlink interface, allowing complete access to the device <> NAPIs <> queues relationship - Introduce notifications filtering for devlink to allow control application scale to thousands of instances - Improve PHY validation, requesting rate matching information for each ethtool link mode supported by both the PHY and host - Add support for ethtool symmetric-xor RSS hash - ACPI based Wifi band RFI (WBRF) mitigation feature for the AMD platform - Expose pin fractional frequency offset value over new DPLL generic netlink attribute - Convert older drivers to platform remove callback returning void - Add support for PHY package MMD read/write New hardware / drivers: - Ethernet: - Octeon CN10K devices - Broadcom 5760X P7 - Qualcomm SM8550 SoC - Texas Instrument DP83TG720S PHY - Bluetooth: - IMC Networks Bluetooth radio Removed: - WiFi: - libertas 16-bit PCMCIA support - Atmel at76c50x drivers - HostAP ISA/PCMCIA style 802.11b driver - zd1201 802.11b USB dongles - Orinoco ISA/PCMCIA 802.11b driver - Aviator/Raytheon driver - Planet WL3501 driver - RNDIS USB 802.11b driver Driver updates: - Ethernet high-speed NICs: - Intel (100G, ice, idpf): - allow one by one port representors creation and removal - add temperature and clock information reporting - add get/set for ethtool's header split ringparam - add again FW logging - adds support switchdev hardware packet mirroring - iavf: implement symmetric-xor RSS hash - igc: add support for concurrent physical and free-running timers - i40e: increase the allowable descriptors - nVidia/Mellanox: - Preparation for Socket-Direct multi-dev netdev. That will allow in future releases combining multiple PFs devices attached to different NUMA nodes under the same netdev - Broadcom (bnxt): - TX completion handling improvements - add basic ntuple filter support - reduce MSIX vectors usage for MQPRIO offload - add VXLAN support, USO offload and TX coalesce completion for P7 - Marvell Octeon EP: - xmit-more support - add PF-VF mailbox support and use it for FW notifications for VFs - Wangxun (ngbe/txgbe): - implement ethtool functions to operate pause param, ring param, coalesce channel number and msglevel - Netronome/Corigine (nfp): - add flow-steering support - support UDP segmentation offload - Ethernet NICs embedded, slower, virtual: - Xilinx AXI: remove duplicate DMA code adopting the dma engine driver - stmmac: add support for HW-accelerated VLAN stripping - TI AM654x sw: add mqprio, frame preemption & coalescing - gve: add support for non-4k page sizes. - virtio-net: support dynamic coalescing moderation - nVidia/Mellanox Ethernet datacenter switches: - allow firmware upgrade without a reboot - more flexible support for bridge flooding via the compressed FID flooding mode - Ethernet embedded switches: - Microchip: - fine-tune flow control and speed configurations in KSZ8xxx - KSZ88X3: enable setting rmii reference - Renesas: - add jumbo frames support - Marvell: - 88E6xxx: add "eth-mac" and "rmon" stats support - Ethernet PHYs: - aquantia: add firmware load support - at803x: refactor the driver to simplify adding support for more chip variants - NXP C45 TJA11xx: Add MACsec offload support - Wifi: - MediaTek (mt76): - NVMEM EEPROM improvements - mt7996 Extremely High Throughput (EHT) improvements - mt7996 Wireless Ethernet Dispatcher (WED) support - mt7996 36-bit DMA support - Qualcomm (ath12k): - support for a single MSI vector - WCN7850: support AP mode - Intel (iwlwifi): - new debugfs file fw_dbg_clear - allow concurrent P2P operation on DFS channels - Bluetooth: - QCA2066: support HFP offload - ISO: more broadcast-related improvements - NXP: better recovery in case receiver/transmitter get out of sync" * tag 'net-next-6.8' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1714 commits) lan78xx: remove redundant statement in lan78xx_get_eee lan743x: remove redundant statement in lan743x_ethtool_get_eee bnxt_en: Fix RCU locking for ntuple filters in bnxt_rx_flow_steer() bnxt_en: Fix RCU locking for ntuple filters in bnxt_srxclsrldel() bnxt_en: Remove unneeded variable in bnxt_hwrm_clear_vnic_filter() tcp: Revert no longer abort SYN_SENT when receiving some ICMP Revert "mlx5 updates 2023-12-20" Revert "net: stmmac: Enable Per DMA Channel interrupt" ipvlan: Remove usage of the deprecated ida_simple_xx() API ipvlan: Fix a typo in a comment net/sched: Remove ipt action tests net: stmmac: Use interrupt mode INTM=1 for per channel irq net: stmmac: Add support for TX/RX channel interrupt net: stmmac: Make MSI interrupt routine generic dt-bindings: net: snps,dwmac: per channel irq net: phy: at803x: make read_status more generic net: phy: at803x: add support for cdt cross short test for qca808x net: phy: at803x: refactor qca808x cable test get status function net: phy: at803x: generalize cdt fault length function net: ethernet: cortina: Drop TSO support ...
2024-01-08mm, treewide: rename MAX_ORDER to MAX_PAGE_ORDERKirill A. Shutemov1-12/+12
commit 23baf831a32c ("mm, treewide: redefine MAX_ORDER sanely") has changed the definition of MAX_ORDER to be inclusive. This has caused issues with code that was not yet upstream and depended on the previous definition. To draw attention to the altered meaning of the define, rename MAX_ORDER to MAX_PAGE_ORDER. Link: https://lkml.kernel.org/r/20231228144704.14033-2-kirill.shutemov@linux.intel.com Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2024-01-08mm, treewide: introduce NR_PAGE_ORDERSKirill A. Shutemov1-7/+6
NR_PAGE_ORDERS defines the number of page orders supported by the page allocator, ranging from 0 to MAX_ORDER, MAX_ORDER + 1 in total. NR_PAGE_ORDERS assists in defining arrays of page orders and allows for more natural iteration over them. [kirill.shutemov@linux.intel.com: fixup for kerneldoc warning] Link: https://lkml.kernel.org/r/20240101111512.7empzyifq7kxtzk3@box Link: https://lkml.kernel.org/r/20231228144704.14033-1-kirill.shutemov@linux.intel.com Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Reviewed-by: Zi Yan <ziy@nvidia.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-12-29mm, kasan: use KASAN_TAG_KERNEL instead of 0xffAndrey Konovalov1-1/+1
Use the KASAN_TAG_KERNEL marco instead of open-coding 0xff in the mm code. This macro is provided by include/linux/kasan-tags.h, which does not include any other headers, so it's safe to include it into mm.h without causing circular include dependencies. Link: https://lkml.kernel.org/r/71db9087b0aebb6c4dccbc609cc0cd50621533c7.1703188911.git.andreyknvl@google.com Signed-off-by: Andrey Konovalov <andreyknvl@google.com> Cc: Alexander Potapenko <glider@google.com> Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com> Cc: Dmitry Vyukov <dvyukov@google.com> Cc: Marco Elver <elver@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-12-20mm: page_alloc: simplify __free_pages_ok()Yajun Deng1-8/+1
There is redundant code in __free_pages_ok(). Use free_one_page() simplify it. Link: https://lkml.kernel.org/r/20231216030503.2126130-1-yajun.deng@linux.dev Signed-off-by: Yajun Deng <yajun.deng@linux.dev> Reviewed-by: Matthew Wilcox (Oracle) <willy@infradead.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-12-10mm: page_alloc: unreserve highatomic page blocks before oomCharan Teja Kalla1-8/+8
__alloc_pages_direct_reclaim() is called from slowpath allocation where high atomic reserves can be unreserved after there is a progress in reclaim and yet no suitable page is found. Later should_reclaim_retry() gets called from slow path allocation to decide if the reclaim needs to be retried before OOM kill path is taken. should_reclaim_retry() checks the available(reclaimable + free pages) memory against the min wmark levels of a zone and returns: a) true, if it is above the min wmark so that slow path allocation will do the reclaim retries. b) false, thus slowpath allocation takes oom kill path. should_reclaim_retry() can also unreserves the high atomic reserves **but only after all the reclaim retries are exhausted.** In a case where there are almost none reclaimable memory and free pages contains mostly the high atomic reserves but allocation context can't use these high atomic reserves, makes the available memory below min wmark levels hence false is returned from should_reclaim_retry() leading the allocation request to take OOM kill path. This can turn into a early oom kill if high atomic reserves are holding lot of free memory and unreserving of them is not attempted. (early)OOM is encountered on a VM with the below state: [ 295.998653] Normal free:7728kB boost:0kB min:804kB low:1004kB high:1204kB reserved_highatomic:8192KB active_anon:4kB inactive_anon:0kB active_file:24kB inactive_file:24kB unevictable:1220kB writepending:0kB present:70732kB managed:49224kB mlocked:0kB bounce:0kB free_pcp:688kB local_pcp:492kB free_cma:0kB [ 295.998656] lowmem_reserve[]: 0 32 [ 295.998659] Normal: 508*4kB (UMEH) 241*8kB (UMEH) 143*16kB (UMEH) 33*32kB (UH) 7*64kB (UH) 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 7752kB Per above log, the free memory of ~7MB exist in the high atomic reserves is not freed up before falling back to oom kill path. Fix it by trying to unreserve the high atomic reserves in should_reclaim_retry() before __alloc_pages_direct_reclaim() can fallback to oom kill path. Link: https://lkml.kernel.org/r/1700823445-27531-1-git-send-email-quic_charante@quicinc.com Fixes: 0aaa29a56e4f ("mm, page_alloc: reserve pageblocks for high-order atomic allocations on demand") Signed-off-by: Charan Teja Kalla <quic_charante@quicinc.com> Reported-by: Chris Goldsworthy <quic_cgoldswo@quicinc.com> Suggested-by: Michal Hocko <mhocko@suse.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: David Rientjes <rientjes@google.com> Cc: Chris Goldsworthy <quic_cgoldswo@quicinc.com> Cc: David Hildenbrand <david@redhat.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Pavankumar Kondeti <quic_pkondeti@quicinc.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-12-10mm: page_alloc: enforce minimum zone size to do high atomic reservesCharan Teja Kalla1-1/+4
Highatomic reserves are set to roughly 1% of zone for maximum and a pageblock size for minimum. Encountered a system with the below configuration: Normal free:7728kB boost:0kB min:804kB low:1004kB high:1204kB reserved_highatomic:8192KB managed:49224kB On such systems, even a single pageblock makes highatomic reserves are set to ~8% of the zone memory. This high value can easily exert pressure on the zone. Per discussion with Michal and Mel, it is not much useful to reserve the memory for highatomic allocations on such small systems[1]. Since the minimum size for high atomic reserves is always going to be a pageblock size and if 1% of zone managed pages is going to be below pageblock size, don't reserve memory for high atomic allocations. Thanks Michal for this suggestion[2]. Since no memory is being reserved for high atomic allocations and if respective allocation failures are seen, this patch can be reverted. [1] https://lore.kernel.org/linux-mm/20231117161956.d3yjdxhhm4rhl7h2@techsingularity.net/ [2] https://lore.kernel.org/linux-mm/ZVYRJMUitykepLRy@tiehlicka/ Link: https://lkml.kernel.org/r/c3a2a48e2cfe08176a80eaf01c110deb9e918055.1700821416.git.quic_charante@quicinc.com Signed-off-by: Charan Teja Kalla <quic_charante@quicinc.com> Acked-by: David Rientjes <rientjes@google.com> Cc: David Hildenbrand <david@redhat.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Michal Hocko <mhocko@suse.com> Cc: Pavankumar Kondeti <quic_pkondeti@quicinc.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-12-10mm: page_alloc: correct high atomic reserve calculationsCharan Teja Kalla1-2/+3
Patch series "mm: page_alloc: fixes for high atomic reserve caluculations", v3. The state of the system where the issue exposed shown in oom kill logs: [ 295.998653] Normal free:7728kB boost:0kB min:804kB low:1004kB high:1204kB reserved_highatomic:8192KB active_anon:4kB inactive_anon:0kB active_file:24kB inactive_file:24kB unevictable:1220kB writepending:0kB present:70732kB managed:49224kB mlocked:0kB bounce:0kB free_pcp:688kBlocal_pcp:492kB free_cma:0kB [ 295.998656] lowmem_reserve[]: 0 32 [ 295.998659] Normal: 508*4kB (UMEH) 241*8kB (UMEH) 143*16kB (UMEH) 33*32kB (UH) 7*64kB (UH) 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 7752kB From the above, it is seen that ~16MB of memory reserved for high atomic reserves against the expectation of 1% reserves which is fixed in the 1st patch. Don't reserve the high atomic page blocks if 1% of zone memory size is below a pageblock size. This patch (of 2): reserve_highatomic_pageblock() aims to reserve the 1% of the managed pages of a zone, which is used for the high order atomic allocations. It uses the below calculation to reserve: static void reserve_highatomic_pageblock(struct page *page, ....) { ....... max_managed = (zone_managed_pages(zone) / 100) + pageblock_nr_pages; if (zone->nr_reserved_highatomic >= max_managed) goto out; zone->nr_reserved_highatomic += pageblock_nr_pages; set_pageblock_migratetype(page, MIGRATE_HIGHATOMIC); move_freepages_block(zone, page, MIGRATE_HIGHATOMIC, NULL); out: .... } Since we are always appending the 1% of zone managed pages count to pageblock_nr_pages, the minimum it is turning into 2 pageblocks as the nr_reserved_highatomic is incremented/decremented in pageblock sizes. Encountered a system(actually a VM running on the Linux kernel) with the below zone configuration: Normal free:7728kB boost:0kB min:804kB low:1004kB high:1204kB reserved_highatomic:8192KB managed:49224kB The existing calculations making it to reserve the 8MB(with pageblock size of 4MB) i.e. 16% of the zone managed memory. Reserving such high amount of memory can easily exert memory pressure in the system thus may lead into unnecessary reclaims till unreserving of high atomic reserves. Since high atomic reserves are managed in pageblock size granules, as MIGRATE_HIGHATOMIC is set for such pageblock, fix the calculations for high atomic reserves as, minimum is pageblock size , maximum is approximately 1% of the zone managed pages. Link: https://lkml.kernel.org/r/cover.1700821416.git.quic_charante@quicinc.com Link: https://lkml.kernel.org/r/1660034138397b82a0a8b6ae51cbe96bd583d89e.1700821416.git.quic_charante@quicinc.com Signed-off-by: Charan Teja Kalla <quic_charante@quicinc.com> Acked-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: David Rientjes <rientjes@google.com> Cc: David Hildenbrand <david@redhat.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@suse.com> Cc: Pavankumar Kondeti <quic_pkondeti@quicinc.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-12-10mm/page_alloc: dedupe some memcg uncharging logicBrendan Jackman1-8/+4
The duplication makes it seem like some work is required before uncharging in the !PageHWPoison case. But it isn't, so we can simplify the code a little. Note the PageMemcgKmem check is redundant, but I've left it in as it avoids an unnecessary function call. Link: https://lkml.kernel.org/r/20231108164920.3401565-1-jackmanb@google.com Signed-off-by: Brendan Jackman <jackmanb@google.com> Reviewed-by: Yosry Ahmed <yosryahmed@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Muchun Song <muchun.song@linux.dev> Cc: Roman Gushchin <roman.gushchin@linux.dev> Cc: Shakeel Butt <shakeelb@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-11-26mm/page_pool: catch page_pool memory leaksJesper Dangaard Brouer1-0/+7
Pages belonging to a page_pool (PP) instance must be freed through the PP APIs in-order to correctly release any DMA mappings and release refcnt on the DMA device when freeing PP instance. When PP release a page (page_pool_release_page) the page->pp_magic value is cleared. This patch detect a leaked PP page in free_page_is_bad() via unexpected state of page->pp_magic value being PP_SIGNATURE. We choose to report and treat it as a bad page. It would be possible to release the page via returning it to the PP instance as the page->pp pointer is likely still valid. Notice this code is only activated when either compiled with CONFIG_DEBUG_VM or boot cmdline debug_pagealloc=on, and CONFIG_PAGE_POOL. Reduced example output of leak with PP_SIGNATURE = dead000000000040: BUG: Bad page state in process swapper/4 pfn:141fa6 page:000000006dbf8062 refcount:0 mapcount:0 mapping:0000000000000000 index:0x141fa6000 pfn:0x141fa6 flags: 0x2fffff80000000(node=0|zone=2|lastcpupid=0x1fffff) page_type: 0xffffffff() raw: 002fffff80000000 dead000000000040 ffff88814888a000 0000000000000000 raw: 0000000141fa6000 0000000000000001 00000000ffffffff 0000000000000000 page dumped because: page_pool leak [...] Call Trace: <IRQ> dump_stack_lvl+0x32/0x50 bad_page+0x70/0xf0 free_unref_page_prepare+0x263/0x430 free_unref_page+0x34/0x130 mlx5e_free_rx_mpwqe+0x190/0x1c0 [mlx5_core] mlx5e_post_rx_mpwqes+0x1ac/0x280 [mlx5_core] mlx5e_napi_poll+0x12b/0x710 [mlx5_core] ? skb_free_head+0x4f/0x90 __napi_poll+0x2b/0x1c0 net_rx_action+0x27b/0x360 The advantage is the Call Trace directly points to the function leaking the PP page, which in this case is an on purpose bug introduced into the mlx5 driver to test this code change. Currently PP will periodically in page_pool_release_retry() printk warning "stalled pool shutdown" which cannot be directly corrolated to leaking and might as well be a false positive due to SKBs being stuck on a socket for an extended period. After this patch we should be able to remove this printk. Signed-off-by: Jesper Dangaard Brouer <hawk@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-25mm: add page_rmappable_folio() wrapperHugh Dickins1-6/+2
folio_prep_large_rmappable() is being used repeatedly along with a conversion from page to folio, a check non-NULL, a check order > 1: wrap it all up into struct folio *page_rmappable_folio(struct page *). Link: https://lkml.kernel.org/r/8d92c6cf-eebe-748-e29c-c8ab224c741@google.com Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Christoph Lameter <cl@linux.com> Cc: David Hildenbrand <david@redhat.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: "Huang, Ying" <ying.huang@intel.com> Cc: Kefeng Wang <wangkefeng.wang@huawei.com> Cc: Matthew Wilcox (Oracle) <willy@infradead.org> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Michal Hocko <mhocko@suse.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Cc: Nhat Pham <nphamcs@gmail.com> Cc: Sidhartha Kumar <sidhartha.kumar@oracle.com> Cc: Suren Baghdasaryan <surenb@google.com> Cc: Tejun heo <tj@kernel.org> Cc: Vishal Moola (Oracle) <vishal.moola@gmail.com> Cc: Yang Shi <shy828301@gmail.com> Cc: Yosry Ahmed <yosryahmed@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-10-25mm: page_alloc: check the order of compound page even when the order is zeroHyesoo Yu1-3/+3
For compound pages, the head sets the PG_head flag and the tail sets the compound_head to indicate the head page. If a user allocates a compound page and frees it with a different order, the compound page information will not be properly initialized. To detect this problem, compound_order(page) and the order argument are compared, but this is not checked when the order argument is zero. That error should be checked regardless of the order. Link: https://lkml.kernel.org/r/20231023083217.1866451-1-hyesoo.yu@samsung.com Signed-off-by: Hyesoo Yu <hyesoo.yu@samsung.com> Reviewed-by: Vishal Moola (Oracle) <vishal.moola@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-10-25mm: page_alloc: skip memoryless nodes entirelyQi Zheng1-2/+5
Patch series "handle memoryless nodes more appropriately", v3. Currently, in the process of initialization or offline memory, memoryless nodes will still be built into the fallback list of itself or other nodes. This is not what we expected, so this patch series removes memoryless nodes from the fallback list entirely. This patch (of 2): In find_next_best_node(), we skipped the memoryless nodes when building the zonelists of other normal nodes (N_NORMAL), but did not skip the memoryless node itself when building the zonelist. This will cause it to be traversed at runtime. For example, say we have node0 and node1, node0 is memoryless node, then the fallback order of node0 and node1 as follows: [ 0.153005] Fallback order for Node 0: 0 1 [ 0.153564] Fallback order for Node 1: 1 After this patch, we skip memoryless node0 entirely, then the fallback order of node0 and node1 as follows: [ 0.155236] Fallback order for Node 0: 1 [ 0.155806] Fallback order for Node 1: 1 So it becomes completely invisible, which will reduce runtime overhead. And in this way, we will not try to allocate pages from memoryless node0, then the panic mentioned in [1] will also be fixed. Even though this problem has been solved by dropping the NODE_MIN_SIZE constrain in x86 [2], it would be better to fix it in core MM as well. [1]. https://lore.kernel.org/all/20230212110305.93670-1-zhengqi.arch@bytedance.com/ [2]. https://lore.kernel.org/all/20231017062215.171670-1-rppt@kernel.org/ [zhengqi.arch@bytedance.com: update comment, per Ingo] Link: https://lkml.kernel.org/r/7300fc00a057eefeb9a68c8ad28171c3f0ce66ce.1697799303.git.zhengqi.arch@bytedance.com Link: https://lkml.kernel.org/r/cover.1697799303.git.zhengqi.arch@bytedance.com Link: https://lkml.kernel.org/r/cover.1697711415.git.zhengqi.arch@bytedance.com Link: https://lkml.kernel.org/r/157013e978468241de4a4c05d5337a44638ecb0e.1697711415.git.zhengqi.arch@bytedance.com Signed-off-by: Qi Zheng <zhengqi.arch@bytedance.com> Acked-by: David Hildenbrand <david@redhat.com> Acked-by: Ingo Molnar <mingo@kernel.org> Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com> Cc: "Huang, Ying" <ying.huang@intel.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Matthew Wilcox (Oracle) <willy@infradead.org> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Michal Hocko <mhocko@suse.com> Cc: Mike Rapoport <rppt@kernel.org> Cc: Oscar Salvador <osalvador@suse.de> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-10-25mm, pcp: reduce detecting time of consecutive high order page freeingHuang Ying1-12/+15
In current PCP auto-tuning design, if the number of pages allocated is much more than that of pages freed on a CPU, the PCP high may become the maximal value even if the allocating/freeing depth is small, for example, in the sender of network workloads. If a CPU was used as sender originally, then it is used as receiver after context switching, we need to fill the whole PCP with maximal high before triggering PCP draining for consecutive high order freeing. This will hurt the performance of some network workloads. To solve the issue, in this patch, we will track the consecutive page freeing with a counter in stead of relying on PCP draining. So, we can detect consecutive page freeing much earlier. On a 2-socket Intel server with 128 logical CPU, we tested SCTP_STREAM_MANY test case of netperf test suite with 64-pair processes. With the patch, the network bandwidth improves 5.0%. This restores the performance drop caused by PCP auto-tuning. Link: https://lkml.kernel.org/r/20231016053002.756205-10-ying.huang@intel.com Signed-off-by: "Huang, Ying" <ying.huang@intel.com> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: David Hildenbrand <david@redhat.com> Cc: Johannes Weiner <jweiner@redhat.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Pavel Tatashin <pasha.tatashin@soleen.com> Cc: Matthew Wilcox <willy@infradead.org> Cc: Christoph Lameter <cl@linux.com> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: Sudeep Holla <sudeep.holla@arm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-10-25mm, pcp: decrease PCP high if free pages < high watermarkHuang Ying1-2/+31
One target of PCP is to minimize pages in PCP if the system free pages is too few. To reach that target, when page reclaiming is active for the zone (ZONE_RECLAIM_ACTIVE), we will stop increasing PCP high in allocating path, decrease PCP high and free some pages in freeing path. But this may be too late because the background page reclaiming may introduce latency for some workloads. So, in this patch, during page allocation we will detect whether the number of free pages of the zone is below high watermark. If so, we will stop increasing PCP high in allocating path, decrease PCP high and free some pages in freeing path. With this, we can reduce the possibility of the premature background page reclaiming caused by too large PCP. The high watermark checking is done in allocating path to reduce the overhead in hotter freeing path. Link: https://lkml.kernel.org/r/20231016053002.756205-9-ying.huang@intel.com Signed-off-by: "Huang, Ying" <ying.huang@intel.com> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: David Hildenbrand <david@redhat.com> Cc: Johannes Weiner <jweiner@redhat.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Pavel Tatashin <pasha.tatashin@soleen.com> Cc: Matthew Wilcox <willy@infradead.org> Cc: Christoph Lameter <cl@linux.com> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: Sudeep Holla <sudeep.holla@arm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-10-25mm: tune PCP high automaticallyHuang Ying1-25/+94
The target to tune PCP high automatically is as follows, - Minimize allocation/freeing from/to shared zone - Minimize idle pages in PCP - Minimize pages in PCP if the system free pages is too few To reach these target, a tuning algorithm as follows is designed, - When we refill PCP via allocating from the zone, increase PCP high. Because if we had larger PCP, we could avoid to allocate from the zone. - In periodic vmstat updating kworker (via refresh_cpu_vm_stats()), decrease PCP high to try to free possible idle PCP pages. - When page reclaiming is active for the zone, stop increasing PCP high in allocating path, decrease PCP high and free some pages in freeing path. So, the PCP high can be tuned to the page allocating/freeing depth of workloads eventually. One issue of the algorithm is that if the number of pages allocated is much more than that of pages freed on a CPU, the PCP high may become the maximal value even if the allocating/freeing depth is small. But this isn't a severe issue, because there are no idle pages in this case. One alternative choice is to increase PCP high when we drain PCP via trying to free pages to the zone, but don't increase PCP high during PCP refilling. This can avoid the issue above. But if the number of pages allocated is much less than that of pages freed on a CPU, there will be many idle pages in PCP and it is hard to free these idle pages. 1/8 (>> 3) of PCP high will be decreased periodically. The value 1/8 is kind of arbitrary. Just to make sure that the idle PCP pages will be freed eventually. On a 2-socket Intel server with 224 logical CPU, we run 8 kbuild instances in parallel (each with `make -j 28`) in 8 cgroup. This simulates the kbuild server that is used by 0-Day kbuild service. With the patch, the build time decreases 3.5%. The cycles% of the spinlock contention (mostly for zone lock) decreases from 11.0% to 0.5%. The number of PCP draining for high order pages freeing (free_high) decreases 65.6%. The number of pages allocated from zone (instead of from PCP) decreases 83.9%. Link: https://lkml.kernel.org/r/20231016053002.756205-8-ying.huang@intel.com Signed-off-by: "Huang, Ying" <ying.huang@intel.com> Suggested-by: Mel Gorman <mgorman@techsingularity.net> Suggested-by: Michal Hocko <mhocko@suse.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: David Hildenbrand <david@redhat.com> Cc: Johannes Weiner <jweiner@redhat.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Pavel Tatashin <pasha.tatashin@soleen.com> Cc: Matthew Wilcox <willy@infradead.org> Cc: Christoph Lameter <cl@linux.com> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: Sudeep Holla <sudeep.holla@arm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-10-25mm: add framework for PCP high auto-tuningHuang Ying1-25/+46
The page allocation performance requirements of different workloads are usually different. So, we need to tune PCP (per-CPU pageset) high to optimize the workload page allocation performance. Now, we have a system wide sysctl knob (percpu_pagelist_high_fraction) to tune PCP high by hand. But, it's hard to find out the best value by hand. And one global configuration may not work best for the different workloads that run on the same system. One solution to these issues is to tune PCP high of each CPU automatically. This patch adds the framework for PCP high auto-tuning. With it, pcp->high of each CPU will be changed automatically by tuning algorithm at runtime. The minimal high (pcp->high_min) is the original PCP high value calculated based on the low watermark pages. While the maximal high (pcp->high_max) is the PCP high value when percpu_pagelist_high_fraction sysctl knob is set to MIN_PERCPU_PAGELIST_HIGH_FRACTION. That is, the maximal pcp->high that can be set via sysctl knob by hand. It's possible that PCP high auto-tuning doesn't work well for some workloads. So, when PCP high is tuned by hand via the sysctl knob, the auto-tuning will be disabled. The PCP high set by hand will be used instead. This patch only adds the framework, so pcp->high will be set to pcp->high_min (original default) always. We will add actual auto-tuning algorithm in the following patches in the series. Link: https://lkml.kernel.org/r/20231016053002.756205-7-ying.huang@intel.com Signed-off-by: "Huang, Ying" <ying.huang@intel.com> Acked-by: Mel Gorman <mgorman@techsingularity.net> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: David Hildenbrand <david@redhat.com> Cc: Johannes Weiner <jweiner@redhat.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Pavel Tatashin <pasha.tatashin@soleen.com> Cc: Matthew Wilcox <willy@infradead.org> Cc: Christoph Lameter <cl@linux.com> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: Sudeep Holla <sudeep.holla@arm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-10-25mm, page_alloc: scale the number of pages that are batch allocatedHuang Ying1-10/+43
When a task is allocating a large number of order-0 pages, it may acquire the zone->lock multiple times allocating pages in batches. This may unnecessarily contend on the zone lock when allocating very large number of pages. This patch adapts the size of the batch based on the recent pattern to scale the batch size for subsequent allocations. On a 2-socket Intel server with 224 logical CPU, we run 8 kbuild instances in parallel (each with `make -j 28`) in 8 cgroup. This simulates the kbuild server that is used by 0-Day kbuild service. With the patch, the cycles% of the spinlock contention (mostly for zone lock) decreases from 12.6% to 11.0% (with PCP size == 367). Link: https://lkml.kernel.org/r/20231016053002.756205-6-ying.huang@intel.com Signed-off-by: "Huang, Ying" <ying.huang@intel.com> Suggested-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Mel Gorman <mgorman@techsingularity.net> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: David Hildenbrand <david@redhat.com> Cc: Johannes Weiner <jweiner@redhat.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Pavel Tatashin <pasha.tatashin@soleen.com> Cc: Matthew Wilcox <willy@infradead.org> Cc: Christoph Lameter <cl@linux.com> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: Sudeep Holla <sudeep.holla@arm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-10-25mm: restrict the pcp batch scale factor to avoid too long latencyHuang Ying1-1/+1
In page allocator, PCP (Per-CPU Pageset) is refilled and drained in batches to increase page allocation throughput, reduce page allocation/freeing latency per page, and reduce zone lock contention. But too large batch size will cause too long maximal allocation/freeing latency, which may punish arbitrary users. So the default batch size is chosen carefully (in zone_batchsize(), the value is 63 for zone > 1GB) to avoid that. In commit 3b12e7e97938 ("mm/page_alloc: scale the number of pages that are batch freed"), the batch size will be scaled for large number of page freeing to improve page freeing performance and reduce zone lock contention. Similar optimization can be used for large number of pages allocation too. To find out a suitable max batch scale factor (that is, max effective batch size), some tests and measurement on some machines were done as follows. A set of debug patches are implemented as follows, - Set PCP high to be 2 * batch to reduce the effect of PCP high - Disable free batch size scaling to get the raw performance. - The code with zone lock held is extracted from rmqueue_bulk() and free_pcppages_bulk() to 2 separate functions to make it easy to measure the function run time with ftrace function_graph tracer. - The batch size is hard coded to be 63 (default), 127, 255, 511, 1023, 2047, 4095. Then will-it-scale/page_fault1 is used to generate the page allocation/freeing workload. The page allocation/freeing throughput (page/s) is measured via will-it-scale. The page allocation/freeing average latency (alloc/free latency avg, in us) and allocation/freeing latency at 99 percentile (alloc/free latency 99%, in us) are measured with ftrace function_graph tracer. The test results are as follows, Sapphire Rapids Server ====================== Batch throughput free latency free latency alloc latency alloc latency page/s avg / us 99% / us avg / us 99% / us ----- ---------- ------------ ------------ ------------- ------------- 63 513633.4 2.33 3.57 2.67 6.83 127 517616.7 4.35 6.65 4.22 13.03 255 520822.8 8.29 13.32 7.52 25.24 511 524122.0 15.79 23.42 14.02 49.35 1023 525980.5 30.25 44.19 25.36 94.88 2047 526793.6 59.39 84.50 45.22 140.81 Ice Lake Server =============== Batch throughput free latency free latency alloc latency alloc latency page/s avg / us 99% / us avg / us 99% / us ----- ---------- ------------ ------------ ------------- ------------- 63 620210.3 2.21 3.68 2.02 4.35 127 627003.0 4.09 6.86 3.51 8.28 255 630777.5 7.70 13.50 6.17 15.97 511 633651.5 14.85 22.62 11.66 31.08 1023 637071.1 28.55 42.02 20.81 54.36 2047 638089.7 56.54 84.06 39.28 91.68 Cascade Lake Server =================== Batch throughput free latency free latency alloc latency alloc latency page/s avg / us 99% / us avg / us 99% / us ----- ---------- ------------ ------------ ------------- ------------- 63 404706.7 3.29 5.03 3.53 4.75 127 422475.2 6.12 9.09 6.36 8.76 255 411522.2 11.68 16.97 10.90 16.39 511 428124.1 22.54 31.28 19.86 32.25 1023 414718.4 43.39 62.52 40.00 66.33 2047 429848.7 86.64 120.34 71.14 106.08 Commet Lake Desktop =================== Batch throughput free latency free latency alloc latency alloc latency page/s avg / us 99% / us avg / us 99% / us ----- ---------- ------------ ------------ ------------- ------------- 63 795183.13 2.18 3.55 2.03 3.05 127 803067.85 3.91 6.56 3.85 5.52 255 812771.10 7.35 10.80 7.14 10.20 511 817723.48 14.17 27.54 13.43 30.31 1023 818870.19 27.72 40.10 27.89 46.28 Coffee Lake Desktop =================== Batch throughput free latency free latency alloc latency alloc latency page/s avg / us 99% / us avg / us 99% / us ----- ---------- ------------ ------------ ------------- ------------- 63 510542.8 3.13 4.40 2.48 3.43 127 514288.6 5.97 7.89 4.65 6.04 255 516889.7 11.86 15.58 8.96 12.55 511 519802.4 23.10 28.81 16.95 26.19 1023 520802.7 45.30 52.51 33.19 45.95 2047 519997.1 90.63 104.00 65.26 81.74 From the above data, to restrict the allocation/freeing latency to be less than 100 us in most times, the max batch scale factor needs to be less than or equal to 5. Although it is reasonable to use 5 as max batch scale factor for the systems tested, there are also slower systems. Where smaller value should be used to constrain the page allocation/freeing latency. So, in this patch, a new kconfig option (PCP_BATCH_SCALE_MAX) is added to set the max batch scale factor. Whose default value is 5, and users can reduce it when necessary. Link: https://lkml.kernel.org/r/20231016053002.756205-5-ying.huang@intel.com Signed-off-by: "Huang, Ying" <ying.huang@intel.com> Acked-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Mel Gorman <mgorman@techsingularity.net> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: David Hildenbrand <david@redhat.com> Cc: Johannes Weiner <jweiner@redhat.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Pavel Tatashin <pasha.tatashin@soleen.com> Cc: Matthew Wilcox <willy@infradead.org> Cc: Christoph Lameter <cl@linux.com> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: Sudeep Holla <sudeep.holla@arm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-10-25mm, pcp: reduce lock contention for draining high-order pagesHuang Ying1-1/+37
In commit f26b3fa04611 ("mm/page_alloc: limit number of high-order pages on PCP during bulk free"), the PCP (Per-CPU Pageset) will be drained when PCP is mostly used for high-order pages freeing to improve the cache-hot pages reusing between page allocating and freeing CPUs. On system with small per-CPU data cache slice, pages shouldn't be cached before draining to guarantee cache-hot. But on a system with large per-CPU data cache slice, some pages can be cached before draining to reduce zone lock contention. So, in this patch, instead of draining without any caching, "pcp->batch" pages will be cached in PCP before draining if the size of the per-CPU data cache slice is more than "3 * batch". In theory, if the size of per-CPU data cache slice is more than "2 * batch", we can reuse cache-hot pages between CPUs. But considering the other usage of cache (code, other data accessing, etc.), "3 * batch" is used. Note: "3 * batch" is chosen to make sure the optimization works on recent x86_64 server CPUs. If you want to increase it, please check whether it breaks the optimization. On a 2-socket Intel server with 128 logical CPU, with the patch, the network bandwidth of the UNIX (AF_UNIX) test case of lmbench test suite with 16-pair processes increase 70.5%. The cycles% of the spinlock contention (mostly for zone lock) decreases from 46.1% to 21.3%. The number of PCP draining for high order pages freeing (free_high) decreases 89.9%. The cache miss rate keeps 0.2%. Link: https://lkml.kernel.org/r/20231016053002.756205-4-ying.huang@intel.com Signed-off-by: "Huang, Ying" <ying.huang@intel.com> Acked-by: Mel Gorman <mgorman@techsingularity.net> Cc: Sudeep Holla <sudeep.holla@arm.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: David Hildenbrand <david@redhat.com> Cc: Johannes Weiner <jweiner@redhat.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Pavel Tatashin <pasha.tatashin@soleen.com> Cc: Matthew Wilcox <willy@infradead.org> Cc: Christoph Lameter <cl@linux.com> Cc: Arjan van de Ven <arjan@linux.intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-10-25mm, pcp: avoid to drain PCP when process exitHuang Ying1-3/+8
Patch series "mm: PCP high auto-tuning", v3. The page allocation performance requirements of different workloads are often different. So, we need to tune the PCP (Per-CPU Pageset) high on each CPU automatically to optimize the page allocation performance. The list of patches in series is as follows, [1/9] mm, pcp: avoid to drain PCP when process exit [2/9] cacheinfo: calculate per-CPU data cache size [3/9] mm, pcp: reduce lock contention for draining high-order pages [4/9] mm: restrict the pcp batch scale factor to avoid too long latency [5/9] mm, page_alloc: scale the number of pages that are batch allocated [6/9] mm: add framework for PCP high auto-tuning [7/9] mm: tune PCP high automatically [8/9] mm, pcp: decrease PCP high if free pages < high watermark [9/9] mm, pcp: reduce detecting time of consecutive high order page freeing Patch [1/9], [2/9], [3/9] optimize the PCP draining for consecutive high-order pages freeing. Patch [4/9], [5/9] optimize batch freeing and allocating. Patch [6/9], [7/9], [8/9] implement and optimize a PCP high auto-tuning method. Patch [9/9] optimize the PCP draining for consecutive high order page freeing based on PCP high auto-tuning. The test results for patches with performance impact are as follows, kbuild ====== On a 2-socket Intel server with 224 logical CPU, we run 8 kbuild instances in parallel (each with `make -j 28`) in 8 cgroup. This simulates the kbuild server that is used by 0-Day kbuild service. build time lock contend% free_high alloc_zone ---------- ---------- --------- ---------- base 100.