From nobody Thu Apr 2 17:16:03 2026 Received: from mail-pf1-f180.google.com (mail-pf1-f180.google.com [209.85.210.180]) (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 3811337CD2B for ; Fri, 27 Mar 2026 11:48:20 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.180 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1774612107; cv=none; b=sT78dyXFd96ZOhilIzDNXHGhtQor7vykX87nAGd4/CjooS6UxLZXBLk2s675XelVbkBpK6W7ucD59x3l9dI+xwL20tVWg9MCQV/0Uj2UudVbIBu3IaJX0la5W4Jt2m+M6a3qkE7+deL3Whn41/yvzkR7jCDdfakJP6Z+IQrOl38= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1774612107; c=relaxed/simple; bh=x/eBU39nWUXP7kFunJS4TuBHFF7qqLUdPek65MZLnBQ=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=DFUa8LldK/NgbWL9EzvTZD96KkFNYuggyvYJ7gIndlQAvXvFHiTx7Ny/US8gi27wAuuy+NGuQYrJYDHmZLLs4MqwgzGChJMIn5m89DS8CEaZsAqr3GHnVACu5TEvJBqA9cT0XyaSpmtb+xrm0mwT48TbHa64Vui8VQ++ZKRBoOs= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=CyFostus; arc=none smtp.client-ip=209.85.210.180 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="CyFostus" Received: by mail-pf1-f180.google.com with SMTP id d2e1a72fcca58-82c70e4654eso891502b3a.2 for ; Fri, 27 Mar 2026 04:48:20 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20251104; t=1774612100; x=1775216900; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=bVoIY6pTYZDOqm8IuntqRqHezyaa7BDdVZSXw3DwDtQ=; b=CyFostusHGbSQfRd6MV9cHQsmjHxDxh3i7sxpy7IdOrxNhvFrpFaCgNoLz/BBjsvXM gZg9nzE/rKMh2FgU8SYNmZ9rXoqw3zmpA6j3GdHw/DOpbu0PuNIFop3FH22jopT/NIlm x+hc7mRDZHbCZ/w1NOOY4e6/aYywFs8ez2sNQr9LVrcqk0iOjg0m54WcXJuH6wrr0uk2 1LE1fRO4EVhmrpizV8B5xzm+iWRcAAoVaCVlLaspUSMmHcACHy7QPsgRUiP//ped+ki8 wzsfFdvYYjG1NzNYxfSkjjpOk31Az7zLlLouqZ0kaQENwNc1LJBJzlldK+lJAdzD4XaC dVCg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1774612100; x=1775216900; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-gg:x-gm-message-state:from :to:cc:subject:date:message-id:reply-to; bh=bVoIY6pTYZDOqm8IuntqRqHezyaa7BDdVZSXw3DwDtQ=; b=cr4VEMbfmnAOd02QWeCZIwGyRBX8PsKIupXS3DjSv+pZfeNsUMXcIzLsMhBpMu5rrd SaVOABjz5MWjjJtlYRLYEbv+l1NupKfUXd3cCcJtU+L7+xN/v9jgkXuQcstlmouPemuI zrXrOXSGwqk4CJo+wksNA+a8Twev06AKLX5p6Jha5G7eYVz+ZZ6bfoOZd/6eggKBT63w 3cGJSkYRgicfqpyJY8lyoE44hKxUfGd3+SIQMQB2yLqkYznxLdBFRjOugHjgL47eou11 /2jssNEWzQsKR9+X1uNfnt0KZ6BxYvoSYt0HNh6n6SJTzqC18lWzD71hVEAv6fxoOSXg /VhA== X-Forwarded-Encrypted: i=1; AJvYcCUW7SrVVGwB5JxuF3EZz8oRJYYg2KVdoudXrC76EdJJdAxfrGvYcZZhHT1QzVVJbYBlY4hNi60MpDjXd8A=@vger.kernel.org X-Gm-Message-State: AOJu0YwmYt+Bc5GbI7YDzVIFFVI5xRJ8hT68VVJfWczbiCdmVybe1wjM 6fjH6tEpRCS0U4dVRF3/wQ2GXvmSO6iWRv8DGH8BHS9MMeOX//SMX9dq X-Gm-Gg: ATEYQzzl0/Lcrp050QfvmFSmbOoz4D58aNCouv8myNVrsnTZMyBCyf24t3H8RoPPEm8 wrNqrkIEuQ/CMlQ7CF5uAgyhQXu/ezo5q88VgV8QwRCuaSUq3V8EBofBSxfJPeux/XJszPUMMVR mKn2HBb6OB/px/lWf3WkTjLyzNgM326QHcjtOAkJKmera8PRjetzym52iW8CHcpzY3E1eLJ1010 pKi1lh0gAUbjshCG9uDR8enlvJAecD0x1FYiRzgoq/jFzknb6h+7wVHhzSU+qzsO6kne8fZh2hR EjkowxbWqnQo14gafkntZNIsVnnsnNZ1v2q913XcRbHo67xrG3VbAURGZmLEG8i9Mx5MYWCPYpe 6b1uvAq+jvaewK3ARilAVjxXzT0TkrRNLj62wDIL1nLoTmygf36ijrp5EScZflKqYM0DUJB7Xs4 kYEya2srEQV1g6/g7rUOtzye+DBUWrcViTOyZYeJc/avH/sUoby81zrQ== X-Received: by 2002:a05:6a00:2294:b0:82a:7ad8:75f2 with SMTP id d2e1a72fcca58-82c960d5574mr2414880b3a.56.1774612099844; Fri, 27 Mar 2026 04:48:19 -0700 (PDT) Received: from localhost.localdomain ([116.128.244.171]) by smtp.gmail.com with ESMTPSA id d2e1a72fcca58-82c7d3c289dsm5235572b3a.47.2026.03.27.04.48.09 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Fri, 27 Mar 2026 04:48:19 -0700 (PDT) From: Chengkaitao To: axboe@kernel.dk, pjw@kernel.org, palmer@dabbelt.com, aou@eecs.berkeley.edu, alex@ghiti.fr, ojeda@kernel.org, boqun@kernel.org, gary@garyguo.net, bjorn3_gh@protonmail.com, lossin@kernel.org, a.hindborg@kernel.org, aliceryhl@google.com, tmgross@umich.edu, dakr@kernel.org, nathan@kernel.org, nick.desaulniers+lkml@gmail.com, morbo@google.com, justinstitt@google.com Cc: linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-riscv@lists.infradead.org, rust-for-linux@vger.kernel.org, bpf@vger.kernel.org, llvm@lists.linux.dev, Kaitao Cheng Subject: [RFC 2/2] tools/ufq_iosched: add BPF example scheduler and build scaffolding Date: Fri, 27 Mar 2026 19:47:41 +0800 Message-ID: <20260327114741.91500-3-pilgrimtao@gmail.com> X-Mailer: git-send-email 2.50.1 In-Reply-To: <20260327114741.91500-1-pilgrimtao@gmail.com> References: <20260327114741.91500-1-pilgrimtao@gmail.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Kaitao Cheng Add ufq_iosched as a simple example for the UFQ block I/O scheduler, In the ufq_simple example, we implement the eBPF struct_ops hooks the kernel exposes so we can exercise and validate the behavior and stability of the kernel UFQ scheduling framework. The Makefile and directory layout are modeled after sched_ext. This mirrors the sched_ext examples pattern so developers can experiment with user-defined queueing policies on top of IOSCHED_UFQ. Signed-off-by: Kaitao Cheng --- tools/ufq_iosched/.gitignore | 2 + tools/ufq_iosched/Makefile | 262 +++++++++++ tools/ufq_iosched/README.md | 136 ++++++ .../include/bpf-compat/gnu/stubs.h | 12 + tools/ufq_iosched/include/ufq/common.bpf.h | 73 +++ tools/ufq_iosched/include/ufq/common.h | 91 ++++ tools/ufq_iosched/include/ufq/simple_stat.h | 21 + tools/ufq_iosched/ufq_simple.bpf.c | 445 ++++++++++++++++++ tools/ufq_iosched/ufq_simple.c | 118 +++++ 9 files changed, 1160 insertions(+) create mode 100644 tools/ufq_iosched/.gitignore create mode 100644 tools/ufq_iosched/Makefile create mode 100644 tools/ufq_iosched/README.md create mode 100644 tools/ufq_iosched/include/bpf-compat/gnu/stubs.h create mode 100644 tools/ufq_iosched/include/ufq/common.bpf.h create mode 100644 tools/ufq_iosched/include/ufq/common.h create mode 100644 tools/ufq_iosched/include/ufq/simple_stat.h create mode 100644 tools/ufq_iosched/ufq_simple.bpf.c create mode 100644 tools/ufq_iosched/ufq_simple.c diff --git a/tools/ufq_iosched/.gitignore b/tools/ufq_iosched/.gitignore new file mode 100644 index 000000000000..d6264fe1c8cd --- /dev/null +++ b/tools/ufq_iosched/.gitignore @@ -0,0 +1,2 @@ +tools/ +build/ diff --git a/tools/ufq_iosched/Makefile b/tools/ufq_iosched/Makefile new file mode 100644 index 000000000000..7dc37d9172aa --- /dev/null +++ b/tools/ufq_iosched/Makefile @@ -0,0 +1,262 @@ +# SPDX-License-Identifier: GPL-2.0 +# Copyright (c) 2026 KylinSoft Corporation. +# Copyright (c) 2026 Kaitao Cheng +include ../build/Build.include +include ../scripts/Makefile.arch +include ../scripts/Makefile.include + +all: all_targets + +ifneq ($(LLVM),) +ifneq ($(filter %/,$(LLVM)),) +LLVM_PREFIX :=3D $(LLVM) +else ifneq ($(filter -%,$(LLVM)),) +LLVM_SUFFIX :=3D $(LLVM) +endif + +CLANG_TARGET_FLAGS_arm :=3D arm-linux-gnueabi +CLANG_TARGET_FLAGS_arm64 :=3D aarch64-linux-gnu +CLANG_TARGET_FLAGS_hexagon :=3D hexagon-linux-musl +CLANG_TARGET_FLAGS_m68k :=3D m68k-linux-gnu +CLANG_TARGET_FLAGS_mips :=3D mipsel-linux-gnu +CLANG_TARGET_FLAGS_powerpc :=3D powerpc64le-linux-gnu +CLANG_TARGET_FLAGS_riscv :=3D riscv64-linux-gnu +CLANG_TARGET_FLAGS_s390 :=3D s390x-linux-gnu +CLANG_TARGET_FLAGS_x86 :=3D x86_64-linux-gnu +CLANG_TARGET_FLAGS :=3D $(CLANG_TARGET_FLAGS_$(ARCH)) + +ifeq ($(CROSS_COMPILE),) +ifeq ($(CLANG_TARGET_FLAGS),) +$(error Specify CROSS_COMPILE or add '--target=3D' option to lib.mk) +else +CLANG_FLAGS +=3D --target=3D$(CLANG_TARGET_FLAGS) +endif # CLANG_TARGET_FLAGS +else +CLANG_FLAGS +=3D --target=3D$(notdir $(CROSS_COMPILE:%-=3D%)) +endif # CROSS_COMPILE + +CC :=3D $(LLVM_PREFIX)clang$(LLVM_SUFFIX) $(CLANG_FLAGS) -fintegrated-as +else +CC :=3D $(CROSS_COMPILE)gcc +endif # LLVM + +CURDIR :=3D $(abspath .) +TOOLSDIR :=3D $(abspath ..) +LIBDIR :=3D $(TOOLSDIR)/lib +BPFDIR :=3D $(LIBDIR)/bpf +TOOLSINCDIR :=3D $(TOOLSDIR)/include +BPFTOOLDIR :=3D $(TOOLSDIR)/bpf/bpftool +APIDIR :=3D $(TOOLSINCDIR)/uapi +GENDIR :=3D $(abspath ../../include/generated) +GENHDR :=3D $(GENDIR)/autoconf.h + +ifeq ($(O),) +OUTPUT_DIR :=3D $(CURDIR)/build +else +OUTPUT_DIR :=3D $(O)/build +endif # O +OBJ_DIR :=3D $(OUTPUT_DIR)/obj +INCLUDE_DIR :=3D $(OUTPUT_DIR)/include +BPFOBJ_DIR :=3D $(OBJ_DIR)/libbpf +UFQOBJ_DIR :=3D $(OBJ_DIR)/ufq_iosched +BINDIR :=3D $(OUTPUT_DIR)/bin +BPFOBJ :=3D $(BPFOBJ_DIR)/libbpf.a +ifneq ($(CROSS_COMPILE),) +HOST_BUILD_DIR :=3D $(OBJ_DIR)/host/obj +HOST_OUTPUT_DIR :=3D $(OBJ_DIR)/host +HOST_INCLUDE_DIR :=3D $(HOST_OUTPUT_DIR)/include +else +HOST_BUILD_DIR :=3D $(OBJ_DIR) +HOST_OUTPUT_DIR :=3D $(OUTPUT_DIR) +HOST_INCLUDE_DIR :=3D $(INCLUDE_DIR) +endif +HOST_BPFOBJ :=3D $(HOST_BUILD_DIR)/libbpf/libbpf.a +RESOLVE_BTFIDS :=3D $(HOST_BUILD_DIR)/resolve_btfids/resolve_btfids +DEFAULT_BPFTOOL :=3D $(HOST_OUTPUT_DIR)/sbin/bpftool + +VMLINUX_BTF_PATHS ?=3D $(if $(O),$(O)/vmlinux) \ + $(if $(KBUILD_OUTPUT),$(KBUILD_OUTPUT)/vmlinux) \ + ../../vmlinux \ + /sys/kernel/btf/vmlinux \ + /boot/vmlinux-$(shell uname -r) +VMLINUX_BTF ?=3D $(abspath $(firstword $(wildcard $(VMLINUX_BTF_PATHS)))) +ifeq ($(VMLINUX_BTF),) +$(error Cannot find a vmlinux for VMLINUX_BTF at any of "$(VMLINUX_BTF_PAT= HS)") +endif + +BPFTOOL ?=3D $(DEFAULT_BPFTOOL) + +ifneq ($(wildcard $(GENHDR)),) + GENFLAGS :=3D -DHAVE_GENHDR +endif + +CFLAGS +=3D -g -O2 -rdynamic -pthread -Wall -Werror $(GENFLAGS) \ + -I$(INCLUDE_DIR) -I$(GENDIR) -I$(LIBDIR) \ + -I$(TOOLSINCDIR) -I$(APIDIR) -I$(CURDIR)/include + +# Silence some warnings when compiled with clang +ifneq ($(LLVM),) +CFLAGS +=3D -Wno-unused-command-line-argument +endif + +LDFLAGS +=3D -lelf -lz -lpthread + +IS_LITTLE_ENDIAN =3D $(shell $(CC) -dM -E - &1 \ + | sed -n '/<...> search starts here:/,/End of search list./{ s| \(/.*\)|-= idirafter \1|p }') \ +$(shell $(1) -dM -E - $@ +else + $(call msg,CP,,$@) + $(Q)cp "$(VMLINUX_H)" $@ +endif + +$(UFQOBJ_DIR)/%.bpf.o: %.bpf.c $(INCLUDE_DIR)/vmlinux.h include/ufq/*.h \ + | $(BPFOBJ) $(UFQOBJ_DIR) + $(call msg,CLNG-BPF,,$(notdir $@)) + $(Q)$(CLANG) $(BPF_CFLAGS) -target bpf -c $< -o $@ + +$(INCLUDE_DIR)/%.bpf.skel.h: $(UFQOBJ_DIR)/%.bpf.o $(INCLUDE_DIR)/vmlinux.= h $(BPFTOOL) + $(eval sched=3D$(notdir $@)) + $(call msg,GEN-SKEL,,$(sched)) + $(Q)$(BPFTOOL) gen object $(<:.o=3D.linked1.o) $< + $(Q)$(BPFTOOL) gen object $(<:.o=3D.linked2.o) $(<:.o=3D.linked1.o) + $(Q)$(BPFTOOL) gen object $(<:.o=3D.linked3.o) $(<:.o=3D.linked2.o) + $(Q)diff $(<:.o=3D.linked2.o) $(<:.o=3D.linked3.o) + $(Q)$(BPFTOOL) gen skeleton $(<:.o=3D.linked3.o) name $(subst .bpf.skel.h= ,,$(sched)) > $@ + $(Q)$(BPFTOOL) gen subskeleton $(<:.o=3D.linked3.o) name $(subst .bpf.ske= l.h,,$(sched)) > $(@:.skel.h=3D.subskel.h) + +UFQ_COMMON_DEPS :=3D include/ufq/common.h include/ufq/simple_stat.h | $(BI= NDIR) + +c-sched-targets =3D ufq_simple + +$(addprefix $(BINDIR)/,$(c-sched-targets)): \ + $(BINDIR)/%: \ + $(filter-out %.bpf.c,%.c) \ + $(INCLUDE_DIR)/%.bpf.skel.h \ + $(UFQ_COMMON_DEPS) + $(eval sched=3D$(notdir $@)) + $(CC) $(CFLAGS) -c $(sched).c -o $(UFQOBJ_DIR)/$(sched).o + $(CC) -o $@ $(UFQOBJ_DIR)/$(sched).o $(BPFOBJ) $(LDFLAGS) + +$(c-sched-targets): %: $(BINDIR)/% + +install: all + $(Q)mkdir -p $(DESTDIR)/usr/local/bin/ + $(Q)cp $(BINDIR)/* $(DESTDIR)/usr/local/bin/ + +clean: + rm -rf $(OUTPUT_DIR) $(HOST_OUTPUT_DIR) + rm -f *.o *.bpf.o *.bpf.skel.h *.bpf.subskel.h + rm -f $(c-sched-targets) + +help: + @echo 'Building targets' + @echo '=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D' + @echo '' + @echo ' all - Compile all schedulers' + @echo '' + @echo 'Alternatively, you may compile individual schedulers:' + @echo '' + @printf ' %s\n' $(c-sched-targets) + @echo '' + @echo 'For any scheduler build target, you may specify an alternative' + @echo 'build output path with the O=3D environment variable. For exampl= e:' + @echo '' + @echo ' O=3D/tmp/ufq_iosched make all' + @echo '' + @echo 'will compile all schedulers, and emit the build artifacts to' + @echo '/tmp/ufq_iosched/build.' + @echo '' + @echo '' + @echo 'Installing targets' + @echo '=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D' + @echo '' + @echo ' install - Compile and install all schedulers to /usr/bin.' + @echo ' You may specify the DESTDIR=3D environment variable' + @echo ' to indicate a prefix for /usr/bin. For example:' + @echo '' + @echo ' DESTDIR=3D/tmp/ufq_iosched make install' + @echo '' + @echo ' will build the schedulers in CWD/build, and' + @echo ' install the schedulers to /tmp/ufq_iosched/usr/bin.' + @echo '' + @echo '' + @echo 'Cleaning targets' + @echo '=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D' + @echo '' + @echo ' clean - Remove all generated files' + +all_targets: $(c-sched-targets) + +.PHONY: all all_targets $(c-sched-targets) clean help + +# delete failed targets +.DELETE_ON_ERROR: + +# keep intermediate (.bpf.skel.h, .bpf.o, etc) targets +.SECONDARY: diff --git a/tools/ufq_iosched/README.md b/tools/ufq_iosched/README.md new file mode 100644 index 000000000000..d831bae7a326 --- /dev/null +++ b/tools/ufq_iosched/README.md @@ -0,0 +1,136 @@ +UFQ IOSCHED EXAMPLE SCHEDULERS +=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D + +# Introduction + +This directory contains a simple example of the ufq IO scheduler. It is me= ant +to illustrate the different kinds of IO schedulers you can build with ufq; +new schedulers will be added as the project evolves across releases. + +# Compiling the examples + +There are a few toolchain dependencies for compiling the example scheduler= s. + +## Toolchain dependencies + +1. clang >=3D 16.0.0 + +The schedulers are BPF programs, and therefore must be compiled with clang= . gcc +is actively working on adding a BPF backend compiler as well, but are still +missing some features such as BTF type tags which are necessary for using +kptrs. + +2. pahole >=3D 1.25 + +You may need pahole in order to generate BTF from DWARF. + +3. rust >=3D 1.70.0 + +Rust schedulers uses features present in the rust toolchain >=3D 1.70.0. Y= ou +should be able to use the stable build from rustup, but if that doesn't +work, try using the rustup nightly build. + +There are other requirements as well, such as make, but these are the main= / +non-trivial ones. + +## Compiling the kernel + +In order to run a ufq scheduler, you'll have to run a kernel compiled +with the patches in this repository, and with a minimum set of necessary +Kconfig options: + +``` +CONFIG_BPF=3Dy +CONFIG_IOSCHED_UFQ=3Dy +CONFIG_BPF_SYSCALL=3Dy +CONFIG_BPF_JIT=3Dy +CONFIG_DEBUG_INFO_BTF=3Dy +``` + +It's also recommended that you also include the following Kconfig options: + +``` +CONFIG_BPF_JIT_ALWAYS_ON=3Dy +CONFIG_BPF_JIT_DEFAULT_ON=3Dy +CONFIG_PAHOLE_HAS_SPLIT_BTF=3Dy +CONFIG_PAHOLE_HAS_BTF_TAG=3Dy +``` + +There is a `Kconfig` file in this directory whose contents you can append = to +your local `.config` file, as long as there are no conflicts with any exis= ting +options in the file. + +## Getting a vmlinux.h file + +You may notice that most of the example schedulers include a "vmlinux.h" f= ile. +This is a large, auto-generated header file that contains all of the types +defined in some vmlinux binary that was compiled with +[BTF](https://docs.kernel.org/bpf/btf.html) (i.e. with the BTF-related Kco= nfig +options specified above). + +The header file is created using `bpftool`, by passing it a vmlinux binary +compiled with BTF as follows: + +```bash +$ bpftool btf dump file /path/to/vmlinux format c > vmlinux.h +``` + +`bpftool` analyzes all of the BTF encodings in the binary, and produces a +header file that can be included by BPF programs to access those types. F= or +example, using vmlinux.h allows a scheduler to access fields defined direc= tly +in vmlinux + +The scheduler build system will generate this vmlinux.h file as part of the +scheduler build pipeline. It looks for a vmlinux file in the following +dependency order: + +1. If the O=3D environment variable is defined, at `$O/vmlinux` +2. If the KBUILD_OUTPUT=3D environment variable is defined, at + `$KBUILD_OUTPUT/vmlinux` +3. At `../../vmlinux` (i.e. at the root of the kernel tree where you're + compiling the schedulers) +3. `/sys/kernel/btf/vmlinux` +4. `/boot/vmlinux-$(uname -r)` + +In other words, if you have compiled a kernel in your local repo, its vmli= nux +file will be used to generate vmlinux.h. Otherwise, it will be the vmlinux= of +the kernel you're currently running on. This means that if you're running = on a +kernel with ufq support, you may not need to compile a local kernel at +all. + +### Aside on CO-RE + +One of the cooler features of BPF is that it supports +[CO-RE](https://nakryiko.com/posts/bpf-core-reference-guide/) (Compile Onc= e Run +Everywhere). This feature allows you to reference fields inside of structs= with +types defined internal to the kernel, and not have to recompile if you loa= d the +BPF program on a different kernel with the field at a different offset. + +## Compiling the schedulers + +Once you have your toolchain setup, and a vmlinux that can be used to gene= rate +a full vmlinux.h file, you can compile the schedulers using `make`: + +```bash +$ make -j($nproc) +``` + +## ufq_simple + +A simple IO scheduler that provides an example of a minimal ufq scheduler. +Populates commonly used kernel-exposed BPF interfaces for testing the UFQ +scheduler framework in the kernel. + +### llvm: [OFF] + +You may see the following output when building the schedulers: + +``` +Auto-detecting system features: +... clang-bpf-co-re: [ on ] +... llvm: [ OFF ] +... libcap: [ on ] +... libbfd: [ on ] +``` + +Seeing `llvm: [ OFF ]` here is not an issue. You can safely ignore. diff --git a/tools/ufq_iosched/include/bpf-compat/gnu/stubs.h b/tools/ufq_i= osched/include/bpf-compat/gnu/stubs.h new file mode 100644 index 000000000000..f200ac0f6ccd --- /dev/null +++ b/tools/ufq_iosched/include/bpf-compat/gnu/stubs.h @@ -0,0 +1,12 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Dummy gnu/stubs.h. clang can end up including /usr/include/gnu/stubs.h = when + * compiling BPF files although its content doesn't play any role. The fil= e in + * turn includes stubs-64.h or stubs-32.h depending on whether __x86_64__ = is + * defined. When compiling a BPF source, __x86_64__ isn't set and thus + * stubs-32.h is selected. However, the file is not there if the system do= esn't + * have 32bit glibc devel package installed leading to a build failure. + * + * The problem is worked around by making this file available in the inclu= de + * search paths before the system one when building BPF. + */ diff --git a/tools/ufq_iosched/include/ufq/common.bpf.h b/tools/ufq_iosched= /include/ufq/common.bpf.h new file mode 100644 index 000000000000..96832d317464 --- /dev/null +++ b/tools/ufq_iosched/include/ufq/common.bpf.h @@ -0,0 +1,73 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2026 KylinSoft Corporation. + * Copyright (c) 2026 Kaitao Cheng + */ +#ifndef __UFQ_COMMON_BPF_H +#define __UFQ_COMMON_BPF_H + +#ifdef LSP +#define __bpf__ +#include "../vmlinux/vmlinux.h" +#else +#include "vmlinux.h" +#endif + +#include +#include +#include +#include +#include "simple_stat.h" + +#define BPF_STRUCT_OPS(name, args...) \ + SEC("struct_ops/" #name) BPF_PROG(name, ##args) + +/* Define struct ufq_iosched_ops for .struct_ops.link in the BPF object */ +#define UFQ_OPS_DEFINE(__name, ...) \ + SEC(".struct_ops.link") \ + struct ufq_iosched_ops __name =3D { \ + __VA_ARGS__, \ + } + +/* list and rbtree */ +#define __contains(name, node) __attribute__((btf_decl_tag("contains:" #na= me ":" #node))) + +struct request *bpf_request_acquire(struct request *rq) __ksym; +bool bpf_request_put(struct request *rq) __ksym; +void bpf_request_release(struct request *rq) __ksym; + +void *bpf_obj_new_impl(__u64 local_type_id, void *meta) __ksym; +void bpf_obj_drop_impl(void *kptr, void *meta) __ksym; + +#define bpf_obj_new(type) ((type *)bpf_obj_new_impl(bpf_core_type_id_local= (type), NULL)) +#define bpf_obj_drop(kptr) bpf_obj_drop_impl(kptr, NULL) + +int bpf_list_push_front_impl(struct bpf_list_head *head, + struct bpf_list_node *node, + void *meta, __u64 off) __ksym; +#define bpf_list_push_front(head, node) bpf_list_push_front_impl(head, nod= e, NULL, 0) + +int bpf_list_push_back_impl(struct bpf_list_head *head, + struct bpf_list_node *node, + void *meta, __u64 off) __ksym; +#define bpf_list_push_back(head, node) bpf_list_push_back_impl(head, node,= NULL, 0) + +struct bpf_list_node *bpf_list_pop_front(struct bpf_list_head *head) __ksy= m; +struct bpf_list_node *bpf_list_pop_back(struct bpf_list_head *head) __ksym; +bool bpf_list_empty(struct bpf_list_head *head) __ksym; +struct bpf_list_node *bpf_list_del(struct bpf_list_head *head, + struct bpf_list_node *node) __ksym; + +struct bpf_rb_node *bpf_rbtree_remove(struct bpf_rb_root *root, + struct bpf_rb_node *node) __ksym; +int bpf_rbtree_add_impl(struct bpf_rb_root *root, struct bpf_rb_node *node, + bool (less)(struct bpf_rb_node *a, const struct bpf_rb_node *b), + void *meta, __u64 off) __ksym; +#define bpf_rbtree_add(head, node, less) bpf_rbtree_add_impl(head, node, l= ess, NULL, 0) + +struct bpf_rb_node *bpf_rbtree_first(struct bpf_rb_root *root) __ksym; + +void *bpf_refcount_acquire_impl(void *kptr, void *meta) __ksym; +#define bpf_refcount_acquire(kptr) bpf_refcount_acquire_impl(kptr, NULL) + +#endif /* __UFQ_COMMON_BPF_H */ diff --git a/tools/ufq_iosched/include/ufq/common.h b/tools/ufq_iosched/inc= lude/ufq/common.h new file mode 100644 index 000000000000..c537c9e34056 --- /dev/null +++ b/tools/ufq_iosched/include/ufq/common.h @@ -0,0 +1,91 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2026 KylinSoft Corporation. + * Copyright (c) 2026 Kaitao Cheng + */ +#ifndef __UFQ_IOSCHED_COMMON_H +#define __UFQ_IOSCHED_COMMON_H + +#ifdef __KERNEL__ +#error "Should not be included by BPF programs" +#endif + +#include +#include +#include +#include +#include +#include +#include "simple_stat.h" + +typedef uint8_t u8; +typedef uint16_t u16; +typedef uint32_t u32; +typedef uint64_t u64; +typedef int8_t s8; +typedef int16_t s16; +typedef int32_t s32; +typedef int64_t s64; + +#define UFQ_ERR(__fmt, ...) \ + do { \ + fprintf(stderr, "[UFQ_ERR] %s:%d", __FILE__, __LINE__); \ + if (errno) \ + fprintf(stderr, " (%s)\n", strerror(errno)); \ + else \ + fprintf(stderr, "\n"); \ + fprintf(stderr, __fmt __VA_OPT__(,) __VA_ARGS__); \ + fprintf(stderr, "\n"); \ + \ + exit(EXIT_FAILURE); \ + } while (0) + +#define UFQ_ERR_IF(__cond, __fmt, ...) \ + do { \ + if (__cond) \ + UFQ_ERR((__fmt) __VA_OPT__(,) __VA_ARGS__); \ + } while (0) + +/* + * struct ufq_iosched_ops can grow over time. With common.bpf.h::UFQ_OPS_D= EFINE() + * and UFQ_OPS_LOAD()/UFQ_OPS_ATTACH(), libbpf performs struct_ops attachm= ent. + */ +#define UFQ_OPS_OPEN(__ops_name, __ufq_name) ({ \ + struct __ufq_name *__skel; \ + \ + __skel =3D __ufq_name##__open(); \ + UFQ_ERR_IF(!__skel, "Could not open " #__ufq_name); \ + (void)__skel->maps.__ops_name; \ + __skel; \ +}) + +#define UFQ_OPS_LOAD(__skel, __ops_name, __ufq_name) ({ \ + (void)(__skel)->maps.__ops_name; \ + UFQ_ERR_IF(__ufq_name##__load((__skel)), "Failed to load skel"); \ +}) + +/* + * New versions of bpftool emit additional link placeholders for BPF maps, + * and set up BPF skeleton so libbpf can auto-attach BPF maps (v1.5+). Old + * libbpf ignores those links. Disable autoattach on newer libbpf to avoid + * attaching twice when we attach struct_ops explicitly. + */ +#if LIBBPF_MAJOR_VERSION > 1 || \ + (LIBBPF_MAJOR_VERSION =3D=3D 1 && LIBBPF_MINOR_VERSION >=3D 5) +#define __UFQ_OPS_DISABLE_AUTOATTACH(__skel, __ops_name) \ + bpf_map__set_autoattach((__skel)->maps.__ops_name, false) +#else +#define __UFQ_OPS_DISABLE_AUTOATTACH(__skel, __ops_name) do {} while (0) +#endif + +#define UFQ_OPS_ATTACH(__skel, __ops_name, __ufq_name) ({ \ + struct bpf_link *__link; \ + __UFQ_OPS_DISABLE_AUTOATTACH(__skel, __ops_name); \ + UFQ_ERR_IF(__ufq_name##__attach((__skel)), "Failed to attach skel"); \ + __link =3D bpf_map__attach_struct_ops((__skel)->maps.__ops_name); \ + UFQ_ERR_IF(!__link, "Failed to attach struct_ops"); \ + __link; \ +}) + +#endif /* __UFQ_IOSCHED_COMMON_H */ + diff --git a/tools/ufq_iosched/include/ufq/simple_stat.h b/tools/ufq_iosche= d/include/ufq/simple_stat.h new file mode 100644 index 000000000000..286d5999a457 --- /dev/null +++ b/tools/ufq_iosched/include/ufq/simple_stat.h @@ -0,0 +1,21 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2026 KylinSoft Corporation. + * Copyright (c) 2026 Kaitao Cheng + */ +#ifndef __UFQ_SIMPLE_STAT_H +#define __UFQ_SIMPLE_STAT_H + +enum ufq_simp_stat_index { + UFQ_SIMP_INSERT_CNT, + UFQ_SIMP_INSERT_SIZE, + UFQ_SIMP_DISPATCH_CNT, + UFQ_SIMP_DISPATCH_SIZE, + UFQ_SIMP_MERGE_CNT, + UFQ_SIMP_MERGE_SIZE, + UFQ_SIMP_FINISH_CNT, + UFQ_SIMP_FINISH_SIZE, + UFQ_SIMP_STAT_MAX, +}; + +#endif /* __UFQ_SIMPLE_STAT_H */ diff --git a/tools/ufq_iosched/ufq_simple.bpf.c b/tools/ufq_iosched/ufq_sim= ple.bpf.c new file mode 100644 index 000000000000..63edd7772600 --- /dev/null +++ b/tools/ufq_iosched/ufq_simple.bpf.c @@ -0,0 +1,445 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2026 KylinSoft Corporation. + * Copyright (c) 2026 Kaitao Cheng + */ +#include + +char _license[] SEC("license") =3D "GPL"; + +#define UFQ_DISK_SUM 20 +#define BLK_MQ_INSERT_AT_HEAD 0x01 +#define REQ_OP_MASK ((1 << 8) - 1) +#define SECTOR_SHIFT 9 + +struct { + __uint(type, BPF_MAP_TYPE_PERCPU_ARRAY); + __uint(key_size, sizeof(u32)); + __uint(value_size, sizeof(u64)); + __uint(max_entries, UFQ_SIMP_STAT_MAX); +} stats SEC(".maps"); + +enum ufq_simp_data_dir { + UFQ_SIMP_READ, + UFQ_SIMP_WRITE, + UFQ_SIMP_DIR_COUNT +}; + +struct queue_list_node { + struct bpf_list_node node; + struct request __kptr * req; +}; + +struct sort_tree_node { + struct bpf_refcount ref; + struct bpf_rb_node rb_node; + struct bpf_list_node list_node; + u64 key; + struct request __kptr * req; +}; + +struct ufq_simple_data { + struct bpf_spin_lock lock; + struct bpf_rb_root sort_tree_read __contains(sort_tree_node, rb_node); + struct bpf_rb_root sort_tree_write __contains(sort_tree_node, rb_node); + struct bpf_list_head dispatch __contains(queue_list_node, node); + struct bpf_list_head fifo_list __contains(sort_tree_node, list_node); +}; + +struct { + __uint(type, BPF_MAP_TYPE_HASH); + __uint(max_entries, UFQ_DISK_SUM); + __type(key, s32); + __type(value, struct ufq_simple_data); +} ufq_map SEC(".maps"); + +static void stat_add(u32 idx, u32 val) +{ + u64 *cnt_p =3D bpf_map_lookup_elem(&stats, &idx); + + if (cnt_p) + (*cnt_p) +=3D val; +} + +static void stat_sub(u32 idx, u32 val) +{ + u64 *cnt_p =3D bpf_map_lookup_elem(&stats, &idx); + + if (cnt_p) + (*cnt_p) -=3D val; +} + +static bool sort_tree_less(struct bpf_rb_node *a, const struct bpf_rb_node= *b) +{ + struct sort_tree_node *node_a, *node_b; + + node_a =3D container_of(a, struct sort_tree_node, rb_node); + node_b =3D container_of(b, struct sort_tree_node, rb_node); + + return node_a->key < node_b->key; +} + +static struct ufq_simple_data *dd_init_sched(struct request_queue *q) +{ + struct ufq_simple_data ufq_sd =3D {}, *ufq_sp; + int ret, id =3D q->id; + + bpf_printk("ufq_simple init sched!"); + ret =3D bpf_map_update_elem(&ufq_map, &id, &ufq_sd, BPF_NOEXIST); + if (ret) { + bpf_printk("ufq_simple/init_sched: update ufq_map err %d", ret); + return NULL; + } + + ufq_sp =3D bpf_map_lookup_elem(&ufq_map, &id); + if (!ufq_sp) { + bpf_printk("ufq_simple/init_sched: lookup queue id %d in ufq_map failed"= , id); + return NULL; + } + + return ufq_sp; +} + +int BPF_STRUCT_OPS(ufq_simple_init_sched, struct request_queue *q) +{ + if (dd_init_sched(q)) + return 0; + else + return -EPERM; +} + +int BPF_STRUCT_OPS(ufq_simple_exit_sched, struct request_queue *q) +{ + int id =3D q->id; + + bpf_printk("ufq_simple exit sched!"); + bpf_map_delete_elem(&ufq_map, &id); + return 0; +} + +int BPF_STRUCT_OPS(ufq_simple_insert_req, struct request_queue *q, + struct request *rq, blk_insert_t flags, + struct list_head *freeq) +{ + struct ufq_simple_data *ufq_sd; + struct queue_list_node *qnode; + struct sort_tree_node *snode, *lnode; + int id =3D q->id, ret =3D 0; + struct request *acquired, *old; + enum ufq_simp_data_dir dir =3D ((rq->cmd_flags & REQ_OP_MASK) & 1) ? + UFQ_SIMP_WRITE : UFQ_SIMP_READ; + + ufq_sd =3D bpf_map_lookup_elem(&ufq_map, &id); + if (!ufq_sd) { + ufq_sd =3D dd_init_sched(q); + if (!ufq_sd) { + bpf_printk("ufq_simple/insert_req: dd_init_sched failed"); + return -EPERM; + } + } + + if (flags & BLK_MQ_INSERT_AT_HEAD) { + /* create queue_list_node */ + qnode =3D bpf_obj_new(typeof(*qnode)); + if (!qnode) { + bpf_printk("ufq_simple/insert_req: qnode alloc failed"); + return -ENOMEM; + } + + acquired =3D bpf_request_acquire(rq); + if (!acquired) { + bpf_obj_drop(qnode); + bpf_printk("ufq_simple/head-insert_req: request_acquire failed"); + return -EPERM; + } + + /* Set request for queue_list_node */ + old =3D bpf_kptr_xchg(&qnode->req, acquired); + if (old) + bpf_request_release(old); + + /* Add queue_list_node to dispatch list */ + bpf_spin_lock(&ufq_sd->lock); + ret =3D bpf_list_push_back(&ufq_sd->dispatch, &qnode->node); + bpf_spin_unlock(&ufq_sd->lock); + } else { + /* create sort_tree_node */ + snode =3D bpf_obj_new(typeof(*snode)); + if (!snode) { + bpf_printk("ufq_simple/insert_req: sort_tree_node alloc failed"); + return -ENOMEM; + } + + /* Use request's starting sector as sort key */ + snode->key =3D rq->__sector; + + /* + * Acquire request reference again for sort_tree_node (each node + * needs independent reference) + */ + acquired =3D bpf_request_acquire(rq); + if (!acquired) { + bpf_obj_drop(snode); + bpf_printk("ufq_simple/insert_req: bpf_request_acquire failed"); + return -EPERM; + } + + /* Set request for sort_tree_node */ + old =3D bpf_kptr_xchg(&snode->req, acquired); + if (old) + bpf_request_release(old); + + /* Add sort_tree_node to red-black tree and list_node to fifo_list */ + bpf_spin_lock(&ufq_sd->lock); + if (dir =3D=3D UFQ_SIMP_READ) + bpf_rbtree_add(&ufq_sd->sort_tree_read, &snode->rb_node, sort_tree_less= ); + else + bpf_rbtree_add(&ufq_sd->sort_tree_write, &snode->rb_node, sort_tree_les= s); + + /* Acquire reference count since the node is also added to fifo_list */ + lnode =3D bpf_refcount_acquire(snode); + if (!lnode) { + struct bpf_rb_root *tree =3D (dir =3D=3D UFQ_SIMP_READ) ? + &ufq_sd->sort_tree_read : &ufq_sd->sort_tree_write; + struct bpf_rb_node *rb_node; + + rb_node =3D bpf_rbtree_remove(tree, &snode->rb_node); + bpf_spin_unlock(&ufq_sd->lock); + if (rb_node) + bpf_obj_drop(container_of(rb_node, struct sort_tree_node, rb_node)); + bpf_printk("ufq_simple/insert_req: bpf_refcount_acquire failed"); + return -EPERM; + } + + ret =3D bpf_list_push_back(&ufq_sd->fifo_list, &lnode->list_node); + bpf_spin_unlock(&ufq_sd->lock); + } + + if (!ret) { + stat_add(UFQ_SIMP_INSERT_CNT, 1); + stat_add(UFQ_SIMP_INSERT_SIZE, rq->__data_len); + } + return ret; +} + +struct request *BPF_STRUCT_OPS(ufq_simple_dispatch_req, struct request_que= ue *q) +{ + struct request *rq =3D NULL; + struct bpf_list_node *list_node; + struct bpf_rb_node *rb_node =3D NULL; + struct queue_list_node *qnode; + struct sort_tree_node *snode, *lnode; + struct ufq_simple_data *ufq_sd; + int id =3D q->id; + + ufq_sd =3D bpf_map_lookup_elem(&ufq_map, &id); + if (!ufq_sd) { + bpf_printk("ufq_simple/dispatch_req: ufq_map lookup %d failed", id); + return NULL; + } + + bpf_spin_lock(&ufq_sd->lock); + list_node =3D bpf_list_pop_front(&ufq_sd->dispatch); + + if (list_node) { + qnode =3D container_of(list_node, struct queue_list_node, node); + rq =3D bpf_kptr_xchg(&qnode->req, NULL); + bpf_spin_unlock(&ufq_sd->lock); + bpf_obj_drop(qnode); + } else { + rb_node =3D bpf_rbtree_first(&ufq_sd->sort_tree_read); + if (rb_node) { + rb_node =3D bpf_rbtree_remove(&ufq_sd->sort_tree_read, rb_node); + } else { + rb_node =3D bpf_rbtree_first(&ufq_sd->sort_tree_write); + if (rb_node) + rb_node =3D bpf_rbtree_remove(&ufq_sd->sort_tree_write, rb_node); + } + + if (!rb_node) { + bpf_spin_unlock(&ufq_sd->lock); + goto out; + } + + snode =3D container_of(rb_node, struct sort_tree_node, rb_node); + + /* Get request from sort_tree_node (this will be returned) */ + rq =3D bpf_kptr_xchg(&snode->req, NULL); + + /* Remove list_node from fifo_list (must be done while holding lock) */ + list_node =3D bpf_list_del(&ufq_sd->fifo_list, &snode->list_node); + bpf_spin_unlock(&ufq_sd->lock); + + if (list_node) { + lnode =3D container_of(list_node, struct sort_tree_node, list_node); + bpf_obj_drop(lnode); + } + bpf_obj_drop(snode); + } + if (!rq) + bpf_printk("ufq_simple/dispatch_req: no request to dispatch"); + +out: + if (rq) { + stat_add(UFQ_SIMP_DISPATCH_CNT, 1); + stat_add(UFQ_SIMP_DISPATCH_SIZE, rq->__data_len); + } + + return rq; +} + +bool BPF_STRUCT_OPS(ufq_simple_has_req, struct request_queue *q, int rqs_c= ount) +{ + struct ufq_simple_data *ufq_sd; + bool has; + int id =3D q->id; + + ufq_sd =3D bpf_map_lookup_elem(&ufq_map, &id); + if (!ufq_sd) { + bpf_printk("ufq_simple/has_req: ufq_map lookup %d failed", id); + return false; + } + + bpf_spin_lock(&ufq_sd->lock); + has =3D !bpf_list_empty(&ufq_sd->dispatch) || + bpf_rbtree_root(&ufq_sd->sort_tree_read) || + bpf_rbtree_root(&ufq_sd->sort_tree_write); + bpf_spin_unlock(&ufq_sd->lock); + + return has; +} + +void BPF_STRUCT_OPS(ufq_simple_finish_req, struct request *rq) +{ + if (rq) { + stat_add(UFQ_SIMP_FINISH_CNT, 1); + stat_add(UFQ_SIMP_FINISH_SIZE, rq->__data_len); + bpf_request_put(rq); + } +} + +struct request *BPF_STRUCT_OPS(ufq_simple_next_req, struct request_queue *= q, + struct request *rq) +{ + return NULL; +} + +struct request *BPF_STRUCT_OPS(ufq_simple_former_req, struct request_queue= *q, + struct request *rq) +{ + return NULL; +} + +struct request *BPF_STRUCT_OPS(ufq_simple_merge_req, struct request_queue = *q, + struct request *rq, int *type) +{ + struct sort_tree_node *snode =3D NULL, *lnode =3D NULL; + sector_t rq_start, rq_end, other_start, other_end; + enum elv_merge mt =3D ELEVATOR_NO_MERGE; + struct bpf_list_node *list_node =3D NULL; + struct bpf_rb_node *rb_node =3D NULL; + struct ufq_simple_data *ufq_sd; + struct request *targ =3D NULL; + enum ufq_simp_data_dir dir; + struct bpf_rb_root *tree; + int id =3D q->id; + int count =3D 0; + + *type =3D ELEVATOR_NO_MERGE; + dir =3D ((rq->cmd_flags & REQ_OP_MASK) & 1) ? UFQ_SIMP_WRITE : UFQ_SIMP_R= EAD; + ufq_sd =3D bpf_map_lookup_elem(&ufq_map, &id); + if (!ufq_sd) + return NULL; + + /* Calculate current request position and end */ + rq_start =3D rq->__sector; + rq_end =3D rq_start + (rq->__data_len >> SECTOR_SHIFT); + + if (dir =3D=3D UFQ_SIMP_READ) + tree =3D &ufq_sd->sort_tree_read; + else + tree =3D &ufq_sd->sort_tree_write; + + bpf_spin_lock(&ufq_sd->lock); + rb_node =3D bpf_rbtree_root(tree); + if (!rb_node) { + bpf_spin_unlock(&ufq_sd->lock); + return NULL; + } + + while (mt =3D=3D ELEVATOR_NO_MERGE && rb_node && count < 100) { + count++; + snode =3D container_of(rb_node, struct sort_tree_node, rb_node); + targ =3D bpf_kptr_xchg(&snode->req, NULL); + if (!targ) + break; + + other_start =3D targ->__sector; + other_end =3D other_start + (targ->__data_len >> SECTOR_SHIFT); + + targ =3D bpf_kptr_xchg(&snode->req, targ); + if (targ) { + bpf_spin_unlock(&ufq_sd->lock); + bpf_request_release(targ); + return NULL; + } + + if (rq_start > other_end) + rb_node =3D bpf_rbtree_right(tree, rb_node); + else if (rq_end < other_start) + rb_node =3D bpf_rbtree_left(tree, rb_node); + else if (rq_end =3D=3D other_start) + mt =3D ELEVATOR_FRONT_MERGE; + else if (other_end =3D=3D rq_start) + mt =3D ELEVATOR_BACK_MERGE; + else + break; + + if (mt) { + rb_node =3D bpf_rbtree_remove(tree, rb_node); + if (rb_node) { + snode =3D container_of(rb_node, + struct sort_tree_node, rb_node); + targ =3D bpf_kptr_xchg(&snode->req, NULL); + + list_node =3D bpf_list_del(&ufq_sd->fifo_list, + &snode->list_node); + bpf_spin_unlock(&ufq_sd->lock); + if (targ) { + *type =3D mt; + stat_add(UFQ_SIMP_MERGE_CNT, 1); + stat_add(UFQ_SIMP_MERGE_SIZE, targ->__data_len); + stat_sub(UFQ_SIMP_INSERT_CNT, 1); + stat_sub(UFQ_SIMP_INSERT_SIZE, targ->__data_len); + } + + if (list_node) { + lnode =3D container_of(list_node, + struct sort_tree_node, list_node); + bpf_obj_drop(lnode); + } + + bpf_obj_drop(snode); + } else { + bpf_spin_unlock(&ufq_sd->lock); + *type =3D ELEVATOR_NO_MERGE; + } + return targ; + } + } + bpf_spin_unlock(&ufq_sd->lock); + + return NULL; +} + +UFQ_OPS_DEFINE(ufq_simple_ops, + .init_sched =3D (void *)ufq_simple_init_sched, + .exit_sched =3D (void *)ufq_simple_exit_sched, + .insert_req =3D (void *)ufq_simple_insert_req, + .dispatch_req =3D (void *)ufq_simple_dispatch_req, + .has_req =3D (void *)ufq_simple_has_req, + .finish_req =3D (void *)ufq_simple_finish_req, + .next_req =3D (void *)ufq_simple_next_req, + .former_req =3D (void *)ufq_simple_former_req, + .merge_req =3D (void *)ufq_simple_merge_req, + .name =3D "ufq_simple"); diff --git a/tools/ufq_iosched/ufq_simple.c b/tools/ufq_iosched/ufq_simple.c new file mode 100644 index 000000000000..3e043e475461 --- /dev/null +++ b/tools/ufq_iosched/ufq_simple.c @@ -0,0 +1,118 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2026 KylinSoft Corporation. + * Copyright (c) 2026 Kaitao Cheng + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include "ufq_simple.bpf.skel.h" + +const char help_fmt[] =3D +"A simple ufq scheduler.\n" +"\n" +"Usage: %s [-v] [-d] [-h]\n" +"\n" +" -v Print version\n" +" -d Print libbpf debug messages\n" +" -h Display this help and exit\n"; + +#define UFQ_SIMPLE_VERSION "0.1.0" +#define TIME_INTERVAL 3 +static bool verbose; +static volatile int exit_req; +__u64 old_stats[UFQ_SIMP_STAT_MAX]; + +static int libbpf_print_fn(enum libbpf_print_level level, const char *form= at, va_list args) +{ + if (level =3D=3D LIBBPF_DEBUG && !verbose) + return 0; + return vfprintf(stderr, format, args); +} + +static void sigint_handler(int simple) +{ + exit_req =3D 1; +} + +static void read_stats(struct ufq_simple *skel, __u64 *stats) +{ + int nr_cpus =3D libbpf_num_possible_cpus(); + __u64 cnts[UFQ_SIMP_STAT_MAX][nr_cpus]; + __u32 idx; + + memset(stats, 0, sizeof(stats[0]) * UFQ_SIMP_STAT_MAX); + + for (idx =3D 0; idx < UFQ_SIMP_STAT_MAX; idx++) { + int ret, cpu; + + ret =3D bpf_map_lookup_elem(bpf_map__fd(skel->maps.stats), + &idx, cnts[idx]); + if (ret < 0) + continue; + for (cpu =3D 0; cpu < nr_cpus; cpu++) + stats[idx] +=3D cnts[idx][cpu]; + } +} + +int main(int argc, char **argv) +{ + struct ufq_simple *skel; + struct bpf_link *link; + __u32 opt; + + libbpf_set_print(libbpf_print_fn); + signal(SIGINT, sigint_handler); + signal(SIGTERM, sigint_handler); + + skel =3D UFQ_OPS_OPEN(ufq_simple_ops, ufq_simple); + + while ((opt =3D getopt(argc, argv, "vdh")) !=3D -1) { + switch (opt) { + case 'v': + printf("ufq_simple version: %s\n", UFQ_SIMPLE_VERSION); + return 0; + case 'd': + verbose =3D true; + break; + default: + fprintf(stderr, help_fmt, basename(argv[0])); + return opt !=3D 'h'; + } + } + + UFQ_OPS_LOAD(skel, ufq_simple_ops, ufq_simple); + link =3D UFQ_OPS_ATTACH(skel, ufq_simple_ops, ufq_simple); + + printf("ufq_simple loop ...\n"); + while (!exit_req) { + __u64 stats[UFQ_SIMP_STAT_MAX]; + + printf("--------------------------------\n"); + read_stats(skel, stats); + printf("bps:%lluk iops:%llu\n", + (stats[UFQ_SIMP_FINISH_SIZE] - + old_stats[UFQ_SIMP_FINISH_SIZE]) / 1024 / TIME_INTERVAL, + (stats[UFQ_SIMP_FINISH_CNT] - + old_stats[UFQ_SIMP_FINISH_CNT]) / TIME_INTERVAL); + printf("(insert: cnt=3D%llu size=3D%llu) (merge: cnt=3D%llu size=3D%l= lu)\n", + stats[UFQ_SIMP_INSERT_CNT], stats[UFQ_SIMP_INSERT_SIZE], + stats[UFQ_SIMP_MERGE_CNT], stats[UFQ_SIMP_MERGE_SIZE]); + printf("(dispatch: cnt=3D%llu size=3D%llu) (finish: cnt=3D%llu size=3D%l= lu)\n", + stats[UFQ_SIMP_DISPATCH_CNT], stats[UFQ_SIMP_DISPATCH_SIZE], + stats[UFQ_SIMP_FINISH_CNT], stats[UFQ_SIMP_FINISH_SIZE]); + memcpy(old_stats, stats, sizeof(old_stats)); + sleep(TIME_INTERVAL); + } + + printf("ufq_simple loop exit ...\n"); + bpf_link__destroy(link); + ufq_simple__destroy(skel); + + return 0; +} --=20 2.43.0