summaryrefslogtreecommitdiff
path: root/network_io/win32/sendrecv.c
blob: 2299263483fa36f3cae06066863a99365aee4821 (plain)
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
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
/* ====================================================================
 * The Apache Software License, Version 1.1
 *
 * Copyright (c) 2000-2002 The Apache Software Foundation.  All rights
 * reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 *
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 *
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in
 *    the documentation and/or other materials provided with the
 *    distribution.
 *
 * 3. The end-user documentation included with the redistribution,
 *    if any, must include the following acknowledgment:
 *       "This product includes software developed by the
 *        Apache Software Foundation (http://www.apache.org/)."
 *    Alternately, this acknowledgment may appear in the software itself,
 *    if and wherever such third-party acknowledgments normally appear.
 *
 * 4. The names "Apache" and "Apache Software Foundation" must
 *    not be used to endorse or promote products derived from this
 *    software without prior written permission. For written
 *    permission, please contact apache@apache.org.
 *
 * 5. Products derived from this software may not be called "Apache",
 *    nor may "Apache" appear in their name, without prior written
 *    permission of the Apache Software Foundation.
 *
 * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
 * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
 * DISCLAIMED.  IN NO EVENT SHALL THE APACHE SOFTWARE FOUNDATION OR
 * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
 * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
 * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
 * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
 * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 * ====================================================================
 *
 * This software consists of voluntary contributions made by many
 * individuals on behalf of the Apache Software Foundation.  For more
 * information on the Apache Software Foundation, please see
 * <http://www.apache.org/>.
 */

#include "networkio.h"
#include "apr_errno.h"
#include "apr_general.h"
#include "apr_network_io.h"
#include "apr_lib.h"
#include "fileio.h"
#if APR_HAVE_TIME_H
#include <time.h>
#endif

/* MAX_SEGMENT_SIZE is the maximum amount of data that will be sent to a client
 * in one call of TransmitFile. This number must be small enough to give the 
 * slowest client time to receive the data before the socket timeout triggers.
 * The same problem can exist with apr_send(). In that case, we rely on the
 * application to adjust socket timeouts and max send segment sizes appropriately.
 * For example, Apache will in most cases call apr_send() with less than 8193 
 * bytes.
 */
#define MAX_SEGMENT_SIZE 65536
#define WSABUF_ON_STACK 50

APR_DECLARE(apr_status_t) apr_send(apr_socket_t *sock, const char *buf,
                                   apr_size_t *len)
{
    apr_ssize_t rv;
    WSABUF wsaData;
    int lasterror;
    DWORD dwBytes = 0;

    wsaData.len = *len;
    wsaData.buf = (char*) buf;

#ifndef _WIN32_WCE
    rv = WSASend(sock->socketdes, &wsaData, 1, &dwBytes, 0, NULL, NULL);
#else
    rv = send(sock->socketdes, wsaData.buf, wsaData.len, 0);
    dwBytes = rv;
#endif
    if (rv == SOCKET_ERROR) {
        lasterror = apr_get_netos_error();
        return lasterror;
    }

    *len = dwBytes;

    return APR_SUCCESS;
}


APR_DECLARE(apr_status_t) apr_recv(apr_socket_t *sock, char *buf,
                                   apr_size_t *len) 
{
    apr_ssize_t rv;
    WSABUF wsaData;
    int lasterror;
    DWORD dwBytes = 0;
    DWORD flags = 0;

    wsaData.len = *len;
    wsaData.buf = (char*) buf;

#ifndef _WIN32_WCE
    rv = WSARecv(sock->socketdes, &wsaData, 1, &dwBytes, &flags, NULL, NULL);
#else
    rv = recv(sock->socketdes, wsaData.buf, wsaData.len, 0);
    dwBytes = rv;
#endif
    if (rv == SOCKET_ERROR) {
        lasterror = apr_get_netos_error();
        *len = 0;
        return lasterror;
    }

    *len = dwBytes;
    return dwBytes == 0 ? APR_EOF : APR_SUCCESS;
}


APR_DECLARE(apr_status_t) apr_sendv(apr_socket_t *sock,
                                    const struct iovec *vec,
                                    apr_int32_t nvec, apr_size_t *nbytes)
{
    apr_status_t rc = APR_SUCCESS;
    apr_ssize_t rv;
    int i;
    DWORD dwBytes = 0;
    WSABUF wsabuf[WSABUF_ON_STACK];
    LPWSABUF pWsaBuf = wsabuf;

    if (nvec > WSABUF_ON_STACK) {
        pWsaBuf = (LPWSABUF) malloc(sizeof(WSABUF) * nvec);
        if (!pWsaBuf)
            return APR_ENOMEM;
    }
    for (i = 0; i < nvec; i++) {
        pWsaBuf[i].buf = vec[i].iov_base;
        pWsaBuf[i].len = vec[i].iov_len;
    }
#ifndef _WIN32_WCE
    rv = WSASend(sock->socketdes, pWsaBuf, nvec, &dwBytes, 0, NULL, NULL);
    if (rv == SOCKET_ERROR) {
        rc = apr_get_netos_error();
    }
#else
    for (i = 0; i < nvec; i++) {
        rv = send(sock->socketdes, pWsaBuf[i].buf, pWsaBuf[i].len, 0);
        if (rv == SOCKET_ERROR) {
            rc = apr_get_netos_error();
            break;
        }
        dwBytes += rv;
    }
#endif
    if (nvec > WSABUF_ON_STACK) 
        free(pWsaBuf);

    *nbytes = dwBytes;
    return rc;
}


APR_DECLARE(apr_status_t) apr_sendto(apr_socket_t *sock, apr_sockaddr_t *where,
                                     apr_int32_t flags, const char *buf, 
                                     apr_size_t *len)
{
    apr_ssize_t rv;

    rv = sendto(sock->socketdes, buf, (*len), flags, 
                (const struct sockaddr*)&where->sa, 
                where->salen);
    if (rv == SOCKET_ERROR) {
        *len = 0;
        return apr_get_netos_error();
    }

    *len = rv;
    return APR_SUCCESS;
}


APR_DECLARE(apr_status_t) apr_recvfrom(apr_sockaddr_t *from, 
                                       apr_socket_t *sock,
                                       apr_int32_t flags, 
                                       char *buf, apr_size_t *len)
{
    apr_ssize_t rv;

    rv = recvfrom(sock->socketdes, buf, (*len), flags, 
                  (struct sockaddr*)&from->sa, &from->salen);
    if (rv == SOCKET_ERROR) {
        (*len) = 0;
        return apr_get_netos_error();
    }
    (*len) = rv;
    if (rv == 0 && sock->type == SOCK_STREAM)
        return APR_EOF;

    return APR_SUCCESS;
}


static void collapse_iovec(char **buf, int *len, struct iovec *iovec, int numvec, apr_pool_t *p)
{
    int ptr = 0;

    if (numvec == 1) {
        *buf = iovec[0].iov_base;
        *len = iovec[0].iov_len;
    }
    else {
        int i;
        for (i = 0; i < numvec; i++) {
            *len += iovec[i].iov_len;
        }

        *buf = apr_palloc(p, *len); /* Should this be a malloc? */

        for (i = 0; i < numvec; i++) {
            memcpy((char*)*buf + ptr, iovec[i].iov_base, iovec[i].iov_len);
            ptr += iovec[i].iov_len;
        }
    }
}


#if APR_HAS_SENDFILE
/*
 *#define WAIT_FOR_EVENT
 * Note: Waiting for the socket directly is much faster than creating a seperate
 * wait event. There are a couple of dangerous aspects to waiting directly 
 * for the socket. First, we should not wait on the socket if concurrent threads
 * can wait-on/signal the same socket. This shouldn't be happening with Apache since 
 * a socket is uniquely tied to a thread. This will change when we begin using 
 * async I/O with completion ports on the socket. 
 */

/*
 * apr_status_t apr_sendfile(apr_socket_t *, apr_file_t *, apr_hdtr_t *, 
 *                         apr_off_t *, apr_size_t *, apr_int32_t flags)
 *    Send a file from an open file descriptor to a socket, along with 
 *    optional headers and trailers
 * arg 1) The socket to which we're writing
 * arg 2) The open file from which to read
 * arg 3) A structure containing the headers and trailers to send
 * arg 4) Offset into the file where we should begin writing
 * arg 5) Number of bytes to send out of the file
 * arg 6) APR flags that are mapped to OS specific flags
 */
APR_DECLARE(apr_status_t) apr_sendfile(apr_socket_t *sock, apr_file_t *file,
                                       apr_hdtr_t *hdtr, apr_off_t *offset,
                                       apr_size_t *len, apr_int32_t flags) 
{
    apr_status_t status = APR_SUCCESS;
    apr_ssize_t rv;
    apr_off_t curoff = *offset;
    DWORD dwFlags = 0;
    DWORD nbytes;
    OVERLAPPED overlapped;
    TRANSMIT_FILE_BUFFERS tfb, *ptfb = NULL;
    int ptr = 0;
    int bytes_to_send;   /* Bytes to send out of the file (not including headers) */
    int disconnected = 0;
    HANDLE wait_event;

    if (apr_os_level < APR_WIN_NT) {
        return APR_ENOTIMPL;
    }

    /* Initialize the overlapped structure */
    memset(&overlapped,'\0', sizeof(overlapped));
#ifdef WAIT_FOR_EVENT
    wait_event = overlapped.hEvent = CreateEvent(NULL, FALSE, FALSE, NULL);
#else
    wait_event = (HANDLE) sock->socketdes;
#endif

    /* Use len to keep track of number of total bytes sent (including headers) */
    bytes_to_send = *len;
    *len = 0;

    /* Handle the goofy case of sending headers/trailers and a zero byte file */
    if (!bytes_to_send && hdtr) {
        if (hdtr->numheaders) {
            rv = apr_sendv(sock, hdtr->headers, hdtr->numheaders, &nbytes);
            if (rv != APR_SUCCESS)
                return rv;
            *len += nbytes;
        }
        if (hdtr->numtrailers) {
            rv = apr_sendv(sock, hdtr->trailers, hdtr->numtrailers, &nbytes);
            if (rv != APR_SUCCESS)
                return rv;
            *len += nbytes;
        }
        return APR_SUCCESS;
    }

    /* Collapse the headers into a single buffer */
    if (hdtr && hdtr->numheaders) {
        memset(&tfb, '\0', sizeof (tfb));
        ptfb = &tfb;
        collapse_iovec((char **)&ptfb->Head, &ptfb->HeadLength, hdtr->headers, 
                       hdtr->numheaders, sock->cntxt);
    }

    while (bytes_to_send) {
        if (bytes_to_send > MAX_SEGMENT_SIZE) {
            nbytes = MAX_SEGMENT_SIZE;
        }
        else {
            /* Last call to TransmitFile() */
            nbytes = bytes_to_send;
            /* Collapse the trailers into a single buffer */
            if (hdtr && hdtr->numtrailers) {
                memset(&tfb, '\0', sizeof (tfb));
                ptfb = &tfb;
                collapse_iovec((char**) &ptfb->Tail, &ptfb->TailLength, 
                               hdtr->trailers, hdtr->numtrailers, sock->cntxt);
            }
            /* Disconnect the socket after last send */
            if (flags & APR_SENDFILE_DISCONNECT_SOCKET) {
                dwFlags |= TF_REUSE_SOCKET;
                dwFlags |= TF_DISCONNECT;
                disconnected = 1;
            }
        }

        overlapped.Offset = (DWORD)(curoff);
#if APR_HAS_LARGE_FILES
        overlapped.OffsetHigh = (DWORD)(curoff >> 32);
#endif
        rv = TransmitFile(sock->socketdes,  /* socket */
                          file->filehand, /* open file descriptor of the file to be sent */
                          nbytes,         /* number of bytes to send. 0=send all */
                          0,              /* Number of bytes per send. 0=use default */
                          &overlapped,    /* OVERLAPPED structure */
                          ptfb,           /* header and trailer buffers */
                          dwFlags);       /* flags to control various aspects of TransmitFile */
        if (!rv) {
            status = apr_get_netos_error();
            if ((status == APR_FROM_OS_ERROR(ERROR_IO_PENDING)) ||
                (status == APR_FROM_OS_ERROR(WSA_IO_PENDING))) {
                rv = WaitForSingleObject(wait_event, 
                                         (DWORD)(sock->timeout >= 0 
                                                 ? sock->timeout_ms : INFINITE));
                if (rv == WAIT_OBJECT_0) {
                    status = APR_SUCCESS;
                    if (!disconnected) {
                        if (!GetOverlappedResult(wait_event, &overlapped, 
                                                 &nbytes, FALSE)) {
                            status = APR_FROM_OS_ERROR(GetLastError());
                        }
                        /* Ugly code alert: GetOverlappedResult returns
                         * a count of all bytes sent. This loop only
                         * tracks bytes sent out of the file.
                         */
                        else if (ptfb) {
                            nbytes -= (ptfb->HeadLength + ptfb->TailLength);
                        }
                    }
                }
                else if (rv == WAIT_TIMEOUT)
                    status = APR_FROM_OS_ERROR(WAIT_TIMEOUT);
                else if (rv == WAIT_ABANDONED) {
                    /* Hummm... WAIT_ABANDONDED is not an error code. It is
                     * a return specific to the Win32 WAIT functions that
                     * indicates that a thread exited while holding a
                     * mutex. Should consider triggering an assert
                     * to detect the condition...
                     */
                    status = APR_FROM_OS_ERROR(WAIT_TIMEOUT);
                }
                else
                    status = apr_get_os_error();
            }
        }
        if (status != APR_SUCCESS)
            break;

        bytes_to_send -= nbytes;
        curoff += nbytes;
        *len += nbytes;
        /* Adjust len for any headers/trailers sent */
        if (ptfb) {
            *len += (ptfb->HeadLength + ptfb->TailLength);
            ptfb = NULL;
        }
    }

    if (status == APR_SUCCESS) {
        /* Mark the socket as disconnected, but do not close it.
         * Note: The application must have stored the socket prior to making
         * the call to apr_sendfile in order to either reuse it or close it.
         */
        if (flags & APR_SENDFILE_DISCONNECT_SOCKET) {
            sock->disconnected = 1;
            sock->socketdes = INVALID_SOCKET;
        }
    }

#ifdef WAIT_FOR_EVENT
    CloseHandle(overlapped.hEvent);
#endif
    return status;
}
#endif