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
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
|
// Copyright 2018 Google LLC
//
// 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 linux
import (
"gvisor.googlesource.com/gvisor/pkg/abi/linux"
"gvisor.googlesource.com/gvisor/pkg/sentry/arch"
"gvisor.googlesource.com/gvisor/pkg/sentry/fs"
"gvisor.googlesource.com/gvisor/pkg/sentry/kernel"
"gvisor.googlesource.com/gvisor/pkg/sentry/kernel/kdefs"
"gvisor.googlesource.com/gvisor/pkg/sentry/usermem"
"gvisor.googlesource.com/gvisor/pkg/syserror"
"gvisor.googlesource.com/gvisor/pkg/waiter"
)
const (
// EventMaskRead contains events that can be triggerd on reads.
EventMaskRead = waiter.EventIn | waiter.EventHUp | waiter.EventErr
)
// Read implements linux syscall read(2). Note that we try to get a buffer that
// is exactly the size requested because some applications like qemu expect
// they can do large reads all at once. Bug for bug. Same for other read
// calls below.
func Read(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.SyscallControl, error) {
fd := kdefs.FD(args[0].Int())
addr := args[1].Pointer()
size := args[2].SizeT()
file := t.FDMap().GetFile(fd)
if file == nil {
return 0, nil, syserror.EBADF
}
defer file.DecRef()
// Check that the file is readable.
if !file.Flags().Read {
return 0, nil, syserror.EBADF
}
// Check that the size is legitimate.
si := int(size)
if si < 0 {
return 0, nil, syserror.EINVAL
}
// Get the destination of the read.
dst, err := t.SingleIOSequence(addr, si, usermem.IOOpts{
AddressSpaceActive: true,
})
if err != nil {
return 0, nil, err
}
n, err := readv(t, file, dst)
t.IOUsage().AccountReadSyscall(n)
return uintptr(n), nil, handleIOError(t, n != 0, err, kernel.ERESTARTSYS, "read", file)
}
// Pread64 implements linux syscall pread64(2).
func Pread64(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.SyscallControl, error) {
fd := kdefs.FD(args[0].Int())
addr := args[1].Pointer()
size := args[2].SizeT()
offset := args[3].Int64()
file := t.FDMap().GetFile(fd)
if file == nil {
return 0, nil, syserror.EBADF
}
defer file.DecRef()
// Check that the offset is legitimate.
if offset < 0 {
return 0, nil, syserror.EINVAL
}
// Is reading at an offset supported?
if !file.Flags().Pread {
return 0, nil, syserror.ESPIPE
}
// Check that the file is readable.
if !file.Flags().Read {
return 0, nil, syserror.EBADF
}
// Check that the size is legitimate.
si := int(size)
if si < 0 {
return 0, nil, syserror.EINVAL
}
// Get the destination of the read.
dst, err := t.SingleIOSequence(addr, si, usermem.IOOpts{
AddressSpaceActive: true,
})
if err != nil {
return 0, nil, err
}
n, err := preadv(t, file, dst, offset)
t.IOUsage().AccountReadSyscall(n)
return uintptr(n), nil, handleIOError(t, n != 0, err, kernel.ERESTARTSYS, "pread64", file)
}
// Readv implements linux syscall readv(2).
func Readv(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.SyscallControl, error) {
fd := kdefs.FD(args[0].Int())
addr := args[1].Pointer()
iovcnt := int(args[2].Int())
file := t.FDMap().GetFile(fd)
if file == nil {
return 0, nil, syserror.EBADF
}
defer file.DecRef()
// Check that the file is readable.
if !file.Flags().Read {
return 0, nil, syserror.EBADF
}
// Read the iovecs that specify the destination of the read.
dst, err := t.IovecsIOSequence(addr, iovcnt, usermem.IOOpts{
AddressSpaceActive: true,
})
if err != nil {
return 0, nil, err
}
n, err := readv(t, file, dst)
t.IOUsage().AccountReadSyscall(n)
return uintptr(n), nil, handleIOError(t, n != 0, err, kernel.ERESTARTSYS, "readv", file)
}
// Preadv implements linux syscall preadv(2).
func Preadv(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.SyscallControl, error) {
fd := kdefs.FD(args[0].Int())
addr := args[1].Pointer()
iovcnt := int(args[2].Int())
offset := args[3].Int64()
file := t.FDMap().GetFile(fd)
if file == nil {
return 0, nil, syserror.EBADF
}
defer file.DecRef()
// Check that the offset is legitimate.
if offset < 0 {
return 0, nil, syserror.EINVAL
}
// Is reading at an offset supported?
if !file.Flags().Pread {
return 0, nil, syserror.ESPIPE
}
// Check that the file is readable.
if !file.Flags().Read {
return 0, nil, syserror.EBADF
}
// Read the iovecs that specify the destination of the read.
dst, err := t.IovecsIOSequence(addr, iovcnt, usermem.IOOpts{
AddressSpaceActive: true,
})
if err != nil {
return 0, nil, err
}
n, err := preadv(t, file, dst, offset)
t.IOUsage().AccountReadSyscall(n)
return uintptr(n), nil, handleIOError(t, n != 0, err, kernel.ERESTARTSYS, "preadv", file)
}
// Preadv2 implements linux syscall preadv2(2).
// TODO: Implement RWF_HIPRI functionality.
func Preadv2(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.SyscallControl, error) {
// While the syscall is
// preadv2(int fd, struct iovec* iov, int iov_cnt, off_t offset, int flags)
// the linux internal call
// (https://elixir.bootlin.com/linux/v4.18/source/fs/read_write.c#L1248)
// splits the offset argument into a high/low value for compatibility with
// 32-bit architectures. The flags argument is the 5th argument.
fd := kdefs.FD(args[0].Int())
addr := args[1].Pointer()
iovcnt := int(args[2].Int())
offset := args[3].Int64()
flags := int(args[5].Int())
file := t.FDMap().GetFile(fd)
if file == nil {
return 0, nil, syserror.EBADF
}
defer file.DecRef()
// Check that the offset is legitimate.
if offset < -1 {
return 0, nil, syserror.EINVAL
}
// Is reading at an offset supported?
if offset > -1 && !file.Flags().Pread {
return 0, nil, syserror.ESPIPE
}
// Check that the file is readable.
if !file.Flags().Read {
return 0, nil, syserror.EBADF
}
// Check flags field.
if flags&^linux.RWF_VALID != 0 {
return 0, nil, syserror.EOPNOTSUPP
}
// Read the iovecs that specify the destination of the read.
dst, err := t.IovecsIOSequence(addr, iovcnt, usermem.IOOpts{
AddressSpaceActive: true,
})
if err != nil {
return 0, nil, err
}
// If preadv2 is called with an offset of -1, readv is called.
if offset == -1 {
n, err := readv(t, file, dst)
t.IOUsage().AccountReadSyscall(n)
return uintptr(n), nil, handleIOError(t, n != 0, err, kernel.ERESTARTSYS, "preadv2", file)
}
n, err := preadv(t, file, dst, offset)
t.IOUsage().AccountReadSyscall(n)
return uintptr(n), nil, handleIOError(t, n != 0, err, kernel.ERESTARTSYS, "preadv2", file)
}
func readv(t *kernel.Task, f *fs.File, dst usermem.IOSequence) (int64, error) {
n, err := f.Readv(t, dst)
if err != syserror.ErrWouldBlock || f.Flags().NonBlocking {
if n > 0 {
// Queue notification if we read anything.
f.Dirent.InotifyEvent(linux.IN_ACCESS, 0)
}
return n, err
}
// Register for notifications.
w, ch := waiter.NewChannelEntry(nil)
f.EventRegister(&w, EventMaskRead)
total := n
for {
// Shorten dst to reflect bytes previously read.
dst = dst.DropFirst64(n)
// Issue the request and break out if it completes with anything
// other than "would block".
n, err = f.Readv(t, dst)
total += n
if err != syserror.ErrWouldBlock {
break
}
// Wait for a notification that we should retry.
if err = t.Block(ch); err != nil {
break
}
}
f.EventUnregister(&w)
if total > 0 {
// Queue notification if we read anything.
f.Dirent.InotifyEvent(linux.IN_ACCESS, 0)
}
return total, err
}
func preadv(t *kernel.Task, f *fs.File, dst usermem.IOSequence, offset int64) (int64, error) {
n, err := f.Preadv(t, dst, offset)
if err != syserror.ErrWouldBlock || f.Flags().NonBlocking {
if n > 0 {
// Queue notification if we read anything.
f.Dirent.InotifyEvent(linux.IN_ACCESS, 0)
}
return n, err
}
// Register for notifications.
w, ch := waiter.NewChannelEntry(nil)
f.EventRegister(&w, EventMaskRead)
total := n
for {
// Shorten dst to reflect bytes previously read.
dst = dst.DropFirst64(n)
// Issue the request and break out if it completes with anything
// other than "would block".
n, err = f.Preadv(t, dst, offset+total)
total += n
if err != syserror.ErrWouldBlock {
break
}
// Wait for a notification that we should retry.
if err = t.Block(ch); err != nil {
break
}
}
f.EventUnregister(&w)
if total > 0 {
// Queue notification if we read anything.
f.Dirent.InotifyEvent(linux.IN_ACCESS, 0)
}
return total, err
}
|
