From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mta0.migadu.com (out-152.mta0.migadu.com [91.218.175.152]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id DAB5448C8D2 for ; Thu, 10 Sep 2026 12:58:31 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=91.218.175.152 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789045114; cv=none; b=sWc+8xiya1TCEzlFEuGyh5e/bd4WbZSdg/yg7M57Mf1yIPR2ZrSMqeozzvP5mx5w78p2lrYxdDiJythi5dQHsgRSJegS6QRV3n1/zANXb2YeS6mKWhtoXhhgQ00snNSzSl7c3K8keD66j47zncvqRp1z2TQHEUssy0ymGh8yNwY= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789045114; c=relaxed/simple; bh=I2uISl30iiE2hWkMlVDyAM9KcNSWgXypIHrbdxO3ZQE=; h=From:To:Cc:Subject:Date:Message-ID:MIME-Version; b=heOSKldaGqK/2NG1YbYmWepBWq0WUOaCcQ1j0EJjyvLATY6ol1jm/5buJ0R++q1VHvM391U211i6RVLFv7dSXKocBhGCxDau2+Y8dao0DG63ArM2rnO0oAbYFQ0r5kly9T9TlWs8IG+NIFmy6pYW5xuAosdU+o9l48aAdnUmvsY= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.dev; spf=pass smtp.mailfrom=linux.dev; dkim=pass (1024-bit key) header.d=linux.dev header.i=@linux.dev header.b=uXbL2Ys5; arc=none smtp.client-ip=91.218.175.152 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.dev Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.dev Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux.dev header.i=@linux.dev header.b="uXbL2Ys5" X-Envelope-To: linux-kernel@vger.kernel.org DKIM-Signature: a=rsa-sha256; bh=I2uISl30iiE2hWkMlVDyAM9KcNSWgXypIHrbdxO3ZQE=; c=simple/simple; d=linux.dev; h=from:to:subject:date:message-id:mime-version:content-type; s=key1; t=1789045109; v=1; x=1789649909; b=uXbL2Ys54IY90aeQQrvZGggMxBDbHHtmNCgkZ1U6U6KU25zmCM/JecztcjNPJcboGZaRCvpC te6432VPi21DjMc0pEeeiLVAlQqEEIvpLd4+06xkxF1hYTxrpBqFbw5e/+MhVqVKKEpExh4WLAB Xzs7FA1bBIHAca8jsT6vphAU= X-Envelope-To: linux-kernel@vger.kernel.org Received: by smtp.migadu.com with ESMTPS id ba4b9a2b436ae150; Thu, 10 Sep 2026 12:58:29 +0000 X-Mizu-Trace-ID: ba4b9a2b436ae150 X-Migadu-Flow: FLOW_OUT From: Tao Cui To: tj@kernel.org, josef@toxicopanda.com, axboe@kernel.dk Cc: cgroups@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, bpf@vger.kernel.org, andrii@kernel.org, ast@kernel.org, daniel@iogearbox.net, linux-kselftest@vger.kernel.org, cui.tao@linux.dev, cuitao@kylinos.cn Subject: [RFC PATCH v2 0/5] blk-iocost: BPF struct_ops cost model Date: Thu, 10 Sep 2026 20:58:12 +0800 Message-ID: <20260910125817.223354-1-cui.tao@linux.dev> X-Mailer: git-send-email 2.43.0 Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit From: Tao Cui This is v2 of the RFC. It incorporates the feedback from the first review round: the attachment model, pricing ownership and per-cgroup state handling have all been reworked, and the interface, registration, dispatch and configuration changes are now folded into a single patch. Thanks for the detailed review. Why a pluggable model at all ---------------------------- When iocost landed in 2019, its commit message already promised that "a later patch will also allow using bpf progs for cost models", and the code has carried the split for it ever since: calc_vtime_cost() is a dispatcher whose only implementation is calc_vtime_cost_builtin(). Seven years later the builtin linear model is still the only one. This series fills that slot, following the TCP congestion control model registration pattern: builtin algorithms remain the default while new ones can be prototyped in BPF. The measured problems --------------------- The builtin model prices each IO with a binary sequential/random base picked by a single per-cgroup cursor and a 16MB seek threshold, plus a per-page cost. On a virtio-blk device with the HDD autop profile, a 4k IO costs ~24us when judged sequential and ~2.7ms when judged random, a 112x spread, so a wrong judgement becomes a wrong price. Three classes of mispricing, all measured: 1. Heuristic rigidity. Two legitimate sequential readers in one cgroup (a database with multiple tablespaces, a threaded backup) ping-pong the single cursor and are all priced random: a measured 89x overcharge collapses throughput under the same weight. Random IO within a hot window smaller than the 16MB threshold is priced sequential: measured 107x undercharge, an accounting escape for hotspot workloads. No setting of the six builtin parameters seems able to fix this: telling the streams apart requires per-IO state tracking, which looks like logic rather than coefficients. 2. Device nonlinearity. SLC-cache phases, SMR band placement and shared controllers (multiple NVMe namespaces multiplexing one device) make the real cost of an identical IO vary by an order of magnitude over time or across namespaces. A static 6-parameter linear model has no way to express that. 3. Unpriced operations. Flush and zone append fall through to a cost of zero and bypass throttling entirely (fixed in a separate series already posted), but the same pattern extends to device quirks the builtin model was never taught. Mispricing feeds directly into the control loop: vtime budgets, surplus donation and the vrate feedback all consume the model's output, so a wrong model can skew the whole controller. How --- A bound BPF model fully owns pricing for every IO on the device: both the bio charging path and the request-level sizing path consult it, there is no per-IO or per-path fallback to the builtin formula, and the model prices every operation including flushes. The builtin cursor is not exposed; a model is expected to track its own stream state. u64 calc_cost(u64 opf, u64 nbytes, sector_t sector, struct blkcg *blkcg, u64 model_flags) opf is the full bio->bi_opf (the operation must be extracted with a mask, and the REQ_* flag bits, including PREFLUSH/FUA, are part of it); model_flags carries iocost-specific metadata which is not part of bio->bi_opf, such as whether the cost calculation is for a merged request; the return value is vtime, clamped to 1 second of device time per IO. blkcg is passed so the model can key per-cgroup state; state stored in BPF_MAP_TYPE_CGRP_STORAGE follows the cgroup lifetime, and optional blkcg_online()/ blkcg_offline() callbacks mirror the css lifecycle for models which want eager setup or teardown. The registration and binding model follows the TCP congestion control model registration pattern: registering a struct_ops makes the model available by its name, while io.cost.model binds one registered model to a device with "model=" and restores the builtin model with "model=linear". Unregistering a model removes it from the registry so it can no longer be selected by name; devices already using the model keep using it until they are switched back to the builtin model, at which point the reference is released. Sleepable models are rejected at verification, since calc_cost() runs under RCU read lock. Patch overview: 1/5: the BPF struct_ops cost model support: Kconfig, ops definition, name registry, registration, io.cost.model binding, unified dispatch and verifier checks 2/5: selftest with the 2x example model (the full builtin linear HDD formula at double cost) plus a runner and the selftest kernel config entries 3/5: add an iocost_ioc_tick tracepoint emitting the per-period controller state, so model quality can be evaluated without drgn (existing events are state-change driven and silent in steady state) 4/5: a second example model which replaces the single-cursor sequentiality heuristic with per-cgroup multi-stream detection keyed by the cgroup, the first consumer of the state interface 5/5: document the model= binding in cgroup-v2.rst Does it work ------------ Mechanism, verified functionally (QEMU, virtio-blk with the HDD profile, same 8s sequential-read workload from a 1%-weight cgroup, builtin vs the 2x example model): - per-IO charge: 2882us -> 5722us, a factor of 1.985x; the completed IO count halves and total cost.usage is conserved, i.e. the model output drives both charging and budgeting - the same ratio held across four hosts and 4k/64k/1M block sizes in the v1 measurements (2.00-2.03x on flash-backed hosts, within 2% of 2x on a real HDD behind a loaded host); the charging measurement is consistent with the v1 mechanism test, while v2 additionally dispatches the request sizing path through the model - edge cases: binding an unknown model name fails with ENOENT; unregistering a bound model leaves the device correctly priced until it is switched back to the builtin model; the readback shows the bound model name; the selftest runner checks the write error and errno of every step, including the restoration Payoff, demonstrated with the multi-stream example model (4/5) on the same setup, 4k IOs at weight 1000, builtin vs the model: - two sequential readers in one cgroup: priced 1961us/op by builtin (both judged random by the single cursor) and 23us/op by the model (each stream keeps its own slot); the completed IO count rises by two orders of magnitude - random IO inside an 8M window: priced 24us/op by builtin (undercharge, an accounting escape) and 2607us/op by the model - single-stream sequential and whole-disk random pricing are unchanged, so the model fixes both directions of mispricing without introducing a new one Non-interference, measured on enterprise NVMe: no measurable overhead when the BPF model is not attached. --- Changes in v2: - struct_ops models are now registered by name and bound per device through io.cost.model; the previous ctrl=bpf selection mechanism, the system-wide single-instance limit and its mutex are gone - a bound model fully owns pricing: the return-0 delegation to the builtin formula is gone, the request-level sizing path dispatches to the model too, and the builtin cursor is no longer exposed - iocg_id is replaced by the blkcg kptr; per-cgroup state uses cgroup storage with its lifetime, plus optional blkcg_online/offline callbacks - the full bio->bi_opf including PREFLUSH/FUA is preserved in opf, while model_flags carries iocost-specific metadata - sleepable models are rejected in .check_member - Kconfig depends on DEBUG_INFO_BTF - tracepoint renamed to iocost_ioc_tick; the example model fixes the merged-bio stream advancement and the map exhaustion limitation (cgroup storage), the selftest checks real write errors and the selftest kernel config carries the new options - the interface, registration, dispatch and configuration changes are folded into one patch Tao Cui (5): blk-iocost: add BPF struct_ops cost model support selftests/bpf: add iocost cost model test blk-iocost: add iocost_ioc_tick tracepoint for per-period device summary selftests/bpf: add multi-stream sequentiality example model docs: cgroup-v2: document io.cost model= binding Documentation/admin-guide/cgroup-v2.rst | 11 + block/Kconfig | 9 + block/Makefile | 1 + block/blk-cgroup.c | 3 + block/blk-iocost-bpf.c | 252 ++++++++++++++++++ block/blk-iocost.c | 138 +++++++++- include/linux/blk-iocost.h | 82 ++++++ include/trace/events/iocost.h | 40 +++ tools/testing/selftests/bpf/config | 2 + .../selftests/bpf/prog_tests/iocost_model.c | 194 ++++++++++++++ .../selftests/bpf/progs/iocost_model.c | 117 ++++++++ tools/testing/selftests/bpf/progs/iocost_ms.c | 137 ++++++++++ 12 files changed, 979 insertions(+), 7 deletions(-) create mode 100644 block/blk-iocost-bpf.c create mode 100644 include/linux/blk-iocost.h create mode 100644 tools/testing/selftests/bpf/prog_tests/iocost_model.c create mode 100644 tools/testing/selftests/bpf/progs/iocost_model.c create mode 100644 tools/testing/selftests/bpf/progs/iocost_ms.c -- 2.43.0