1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
|
// Copyright 2019 The gVisor Authors.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package ext
import (
"io"
"sync/atomic"
"gvisor.dev/gvisor/pkg/abi/linux"
"gvisor.dev/gvisor/pkg/sentry/context"
"gvisor.dev/gvisor/pkg/sentry/fs/ext/disklayout"
"gvisor.dev/gvisor/pkg/syserror"
)
// inode represents an ext inode.
//
// inode uses the same inheritance pattern that pkg/sentry/vfs structures use.
// This has been done to increase memory locality.
//
// Implementations:
// inode --
// |-- pipe
// |-- dir
// |-- symlink
// |-- regular--
// |-- extent file
// |-- block map file
// |-- inline file
type inode struct {
// refs is a reference count. refs is accessed using atomic memory operations.
refs int64
// inodeNum is the inode number of this inode on disk. This is used to
// identify inodes within the ext filesystem.
inodeNum uint32
// diskInode gives us access to the inode struct on disk. Immutable.
diskInode disklayout.Inode
// This is immutable. The first field of the implementations must have inode
// as the first field to ensure temporality.
impl interface{}
}
// incRef increments the inode ref count.
func (in *inode) incRef() {
atomic.AddInt64(&in.refs, 1)
}
// tryIncRef tries to increment the ref count. Returns true if successful.
func (in *inode) tryIncRef() bool {
for {
refs := atomic.LoadInt64(&in.refs)
if refs == 0 {
return false
}
if atomic.CompareAndSwapInt64(&in.refs, refs, refs+1) {
return true
}
}
}
// decRef decrements the inode ref count and releases the inode resources if
// the ref count hits 0.
//
// Preconditions: Must have locked fs.mu.
func (in *inode) decRef(fs *filesystem) {
if refs := atomic.AddInt64(&in.refs, -1); refs == 0 {
delete(fs.inodeCache, in.inodeNum)
} else if refs < 0 {
panic("ext.inode.decRef() called without holding a reference")
}
}
// newInode is the inode constructor. Reads the inode off disk. Identifies
// inodes based on the absolute inode number on disk.
//
// Preconditions: Must hold the mutex of the filesystem containing dev.
func newInode(ctx context.Context, dev io.ReadSeeker, sb disklayout.SuperBlock, bgs []disklayout.BlockGroup, inodeNum uint32) (*inode, error) {
if inodeNum == 0 {
panic("inode number 0 on ext filesystems is not possible")
}
inodeRecordSize := sb.InodeSize()
var diskInode disklayout.Inode
if inodeRecordSize == disklayout.OldInodeSize {
diskInode = &disklayout.InodeOld{}
} else {
diskInode = &disklayout.InodeNew{}
}
// Calculate where the inode is actually placed.
inodesPerGrp := sb.InodesPerGroup()
blkSize := sb.BlockSize()
inodeTableOff := bgs[getBGNum(inodeNum, inodesPerGrp)].InodeTable() * blkSize
inodeOff := inodeTableOff + uint64(uint32(inodeRecordSize)*getBGOff(inodeNum, inodesPerGrp))
if err := readFromDisk(dev, int64(inodeOff), diskInode); err != nil {
return nil, err
}
// Build the inode based on its type.
inode := inode{
refs: 1,
inodeNum: inodeNum,
diskInode: diskInode,
}
switch diskInode.Mode().FileType() {
case linux.ModeSymlink:
f, err := newSymlink(dev, blkSize, inode)
if err != nil {
return nil, err
}
return &f.inode, nil
case linux.ModeRegular:
f, err := newRegularFile(dev, blkSize, inode)
if err != nil {
return nil, err
}
return &f.inode, nil
case linux.ModeDirectory:
return &newDirectroy(inode).inode, nil
case linux.ModeNamedPipe:
return &newNamedPipe(ctx, inode).inode, nil
default:
// TODO(b/134676337): Return appropriate errors for sockets and devices.
return nil, syserror.EINVAL
}
}
// getBGNum returns the block group number that a given inode belongs to.
func getBGNum(inodeNum uint32, inodesPerGrp uint32) uint32 {
return (inodeNum - 1) / inodesPerGrp
}
// getBGOff returns the offset at which the given inode lives in the block
// group's inode table, i.e. the index of the inode in the inode table.
func getBGOff(inodeNum uint32, inodesPerGrp uint32) uint32 {
return (inodeNum - 1) % inodesPerGrp
}
|
