pkg/subsystem/linux: add the basic subsystem extraction code

4 files changed, 278 insertions, 1 deletions

diff --git a/pkg/subsystem/entity/entities.go b/pkg/subsystem/entity/entities.go
index 6d028e391..3bcdf8e93 100644
--- a/pkg/subsystem/entity/entities.go
+++ b/pkg/subsystem/entity/entities.go
@@ -4,7 +4,11 @@
 package entity
 
 type Subsystem struct {
-	PathRules []PathRule
+	Name        string
+	PathRules   []PathRule
+	Syscalls    []string
+	Lists       []string
+	Maintainers []string
 }
 
 // PathRule describes the part of the directory tree belonging to a single subsystem.
@@ -13,3 +17,7 @@ type PathRule struct {
 	// ExcludeRegexps are tested before IncludeRegexp.
 	ExcludeRegexp string
 }
+
+func (pr *PathRule) IsEmpty() bool {
+	return pr.IncludeRegexp == "" && pr.ExcludeRegexp == ""
+}
diff --git a/pkg/subsystem/linux/rules.go b/pkg/subsystem/linux/rules.go
new file mode 100644
index 000000000..2df641c5e
--- /dev/null
+++ b/pkg/subsystem/linux/rules.go
@@ -0,0 +1,13 @@
+// Copyright 2023 syzkaller project authors. All rights reserved.
+// Use of this source code is governed by Apache 2 LICENSE that can be found in the LICENSE file.
+
+package linux
+
+type customRules struct {
+	// The mapping between a Linux subsystem name and its system calls.
+	subsystemCalls map[string][]string
+}
+
+var (
+	linuxSubsystemRules = &customRules{}
+)
diff --git a/pkg/subsystem/linux/subsystems.go b/pkg/subsystem/linux/subsystems.go
new file mode 100644
index 000000000..b5ffa924c
--- /dev/null
+++ b/pkg/subsystem/linux/subsystems.go
@@ -0,0 +1,124 @@
+// Copyright 2023 syzkaller project authors. All rights reserved.
+// Use of this source code is governed by Apache 2 LICENSE that can be found in the LICENSE file.
+
+package linux
+
+import (
+	"fmt"
+	"io/fs"
+	"os"
+	"sort"
+
+	"github.com/google/syzkaller/pkg/subsystem/entity"
+)
+
+func ListFromRepo(repo string) ([]*entity.Subsystem, error) {
+	return listFromRepoInner(os.DirFS(repo), linuxSubsystemRules)
+}
+
+// listFromRepoInner allows for better testing.
+func listFromRepoInner(root fs.FS, rules *customRules) ([]*entity.Subsystem, error) {
+	records, err := getMaintainers(root)
+	if err != nil {
+		return nil, err
+	}
+	ctx := &linuxCtx{
+		root:       root,
+		rawRecords: records,
+		extraRules: rules,
+	}
+	ctx.groupByList()
+	list, err := ctx.getSubsystems()
+	if err != nil {
+		return nil, err
+	}
+	// Sort subsystems by name to keep output consistent.
+	sort.Slice(list, func(i, j int) bool { return list[i].Name < list[j].Name })
+	// Sort path rules to keep output consistent.
+	for _, entity := range list {
+		sort.Slice(entity.PathRules, func(i, j int) bool {
+			a, b := entity.PathRules[i], entity.PathRules[j]
+			if a.IncludeRegexp != b.IncludeRegexp {
+				return a.IncludeRegexp < b.IncludeRegexp
+			}
+			return a.ExcludeRegexp < b.ExcludeRegexp
+		})
+	}
+	return list, nil
+}
+
+type linuxCtx struct {
+	root       fs.FS
+	rawRecords []*maintainersRecord
+	extraRules *customRules
+}
+
+type subsystemCandidate struct {
+	records     []*maintainersRecord
+	commonEmail string
+}
+
+func (ctx *linuxCtx) groupByList() []*subsystemCandidate {
+	perList := make(map[string][]*maintainersRecord)
+	for _, record := range ctx.rawRecords {
+		for _, list := range record.lists {
+			perList[list] = append(perList[list], record)
+		}
+	}
+	ret := []*subsystemCandidate{}
+	for email, list := range perList {
+		ret = append(ret, &subsystemCandidate{
+			commonEmail: email,
+			records:     list,
+		})
+	}
+	return ret
+}
+
+func (ctx *linuxCtx) getSubsystems() ([]*entity.Subsystem, error) {
+	ret := []*entity.Subsystem{}
+	for _, raw := range ctx.groupByList() {
+		s := &entity.Subsystem{}
+		mergeRawRecords(s, raw.records)
+		ret = append(ret, s)
+	}
+	return ret, nil
+}
+
+func mergeRawRecords(subsystem *entity.Subsystem, records []*maintainersRecord) {
+	unique := func(list []string) []string {
+		m := make(map[string]struct{})
+		for _, s := range list {
+			m[s] = struct{}{}
+		}
+		ret := []string{}
+		for s := range m {
+			ret = append(ret, s)
+		}
+		return ret
+	}
+	var lists, maintainers []string
+	for _, record := range records {
+		rule := record.ToPathRule()
+		if !rule.IsEmpty() {
+			subsystem.PathRules = append(subsystem.PathRules, rule)
+		}
+		lists = append(lists, record.lists...)
+		maintainers = append(maintainers, record.maintainers...)
+	}
+	subsystem.Lists = unique(lists)
+	// But there's a risk that we collect too many unrelated maintainers, so
+	// let's only merge them if there are no lists.
+	if len(records) <= 1 {
+		subsystem.Maintainers = unique(maintainers)
+	}
+}
+
+func getMaintainers(root fs.FS) ([]*maintainersRecord, error) {
+	f, err := root.Open("MAINTAINERS")
+	if err != nil {
+		return nil, fmt.Errorf("failed to open the MAINTAINERS file: %w", err)
+	}
+	defer f.Close()
+	return parseLinuxMaintainers(f)
+}
diff --git a/pkg/subsystem/linux/subsystems_test.go b/pkg/subsystem/linux/subsystems_test.go
new file mode 100644
index 000000000..f5699f26e
--- /dev/null
+++ b/pkg/subsystem/linux/subsystems_test.go
@@ -0,0 +1,132 @@
+// Copyright 2023 syzkaller project authors. All rights reserved.
+// Use of this source code is governed by Apache 2 LICENSE that can be found in the LICENSE file.
+
+package linux
+
+import (
+	"io/fs"
+	"testing"
+	"testing/fstest"
+
+	"github.com/google/syzkaller/pkg/subsystem/entity"
+	"github.com/stretchr/testify/assert"
+)
+
+func TestGroupLinuxSubsystems(t *testing.T) {
+	subsystems, err := listFromRepoInner(
+		prepareTestLinuxRepo(t, []byte(testMaintainers)),
+		nil)
+	if err != nil {
+		t.Fatal(err)
+	}
+	// The regexps used for matching rules may change later, so let's not compare them here.
+	for _, s := range subsystems {
+		s.PathRules = nil
+	}
+	expected := []*entity.Subsystem{
+		{
+			Lists:       []string{"linux-fsdevel@vger.kernel.org"},
+			Maintainers: []string{"email_vfs@email.com"},
+		},
+		{
+			Lists:       []string{"linux-ext4@vger.kernel.org"},
+			Maintainers: []string{"email_ext4@email.com", "email_ext4_2@email.com"},
+		},
+		{
+			Lists: []string{"linux-mm@kvack.org"},
+		},
+		{
+			Lists:       []string{"linux-kernel@vger.kernel.org"},
+			Maintainers: []string{"email_rest@email.com"},
+		},
+	}
+	assert.ElementsMatch(t, subsystems, expected)
+}
+
+func prepareTestLinuxRepo(t *testing.T, maintainers []byte) fs.FS {
+	return fstest.MapFS{
+		`fs/ext4/fsync.c`:        {},
+		`fs/ext4/mmp.c`:          {},
+		`fs/freevxfs/vxfs_olt.c`: {},
+		`fs/file.c`:              {},
+		`fs/internal.h`:          {},
+		`include/linux/fs.h`:     {},
+		`include/linux/mm.h`:     {},
+		`include/net/ah.h`:       {},
+		`mm/memory.c`:            {},
+		`mm/shmem.c`:             {},
+		`MAINTAINERS`:            {Data: maintainers},
+	}
+}
+
+var (
+	testMaintainers = `
+Maintainers List
+----------------
+
+.. note:: When reading this list, please look for the most precise areas
+          first. When adding to this list, please keep the entries in
+          alphabetical order.
+
+FILESYSTEMS (VFS and infrastructure)
+M:	Developer <email_vfs@email.com>
+L:	linux-fsdevel@vger.kernel.org
+S:	Maintained
+F:	fs/*
+F:	include/linux/fs.h
+F:	include/linux/fs_types.h
+F:	include/uapi/linux/fs.h
+F:	include/uapi/linux/openat2.h
+
+EXT4 FILE SYSTEM
+M:	Developer <email_ext4@email.com>
+M:	Developer <email_ext4_2@email.com>
+L:	linux-ext4@vger.kernel.org
+S:	Maintained
+W:	http://ext4.wiki.kernel.org
+Q:	http://patchwork.ozlabs.org/project/linux-ext4/list/
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4.git
+F:	Documentation/filesystems/ext4/
+F:	fs/ext4/
+F:	include/trace/events/ext4.h
+
+FREEVXFS FILESYSTEM
+M:	Developer <email_vxfs@email.com>
+S:	Maintained
+W:	ftp://ftp.openlinux.org/pub/people/hch/vxfs
+F:	fs/freevxfs/
+
+MEMORY MANAGEMENT
+M:	Developer <email_mm@email.com>
+L:	linux-mm@kvack.org
+S:	Maintained
+W:	http://www.linux-mm.org
+T:	git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
+T:	quilt git://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new
+F:	include/linux/gfp.h
+F:	include/linux/gfp_types.h
+F:	include/linux/memory_hotplug.h
+F:	include/linux/mm.h
+F:	include/linux/mmzone.h
+F:	include/linux/pagewalk.h
+F:	include/linux/vmalloc.h
+F:	mm/
+F:	tools/testing/selftests/vm/
+
+TMPFS (SHMEM FILESYSTEM)
+M:	Developer <email_shmem@email.com>
+L:	linux-mm@kvack.org
+S:	Maintained
+F:	include/linux/shmem_fs.h
+F:	mm/shmem.c
+
+THE REST
+M:	Developer <email_rest@email.com>
+L:	linux-kernel@vger.kernel.org
+S:	Buried alive in reporters
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
+F:	*
+F:	*/
+
+`
+)