summaryrefslogtreecommitdiffstats
path: root/xlators/features/arbiter/src/arbiter.c
blob: fa2b3bf5a6c6c7a2907aeff2f641c32f1e92b679 (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
/*
  Copyright (c) 2015 Red Hat, Inc. <http://www.redhat.com>
  This file is part of GlusterFS.

  This file is licensed to you under your choice of the GNU Lesser
  General Public License, version 3 or any later version (LGPLv3 or
  later), or the GNU General Public License, version 2 (GPLv2), in all
  cases as published by the Free Software Foundation.
*/

#include "arbiter.h"
#include "arbiter-mem-types.h"
#include <glusterfs/glusterfs.h>
#include <glusterfs/xlator.h>
#include <glusterfs/logging.h>

static arbiter_inode_ctx_t *
__arbiter_inode_ctx_get(inode_t *inode, xlator_t *this)
{
    arbiter_inode_ctx_t *ctx = NULL;
    int ret = 0;
    uint64_t ctx_addr = 0;

    ret = __inode_ctx_get(inode, this, &ctx_addr);
    if (ret == 0) {
        ctx = (arbiter_inode_ctx_t *)(long)ctx_addr;
        goto out;
    }

    ctx = GF_CALLOC(1, sizeof(*ctx), gf_arbiter_mt_inode_ctx_t);
    if (!ctx)
        goto out;

    ret = __inode_ctx_put(inode, this, (uint64_t)(uintptr_t)ctx);
    if (ret) {
        GF_FREE(ctx);
        ctx = NULL;
        gf_log_callingfn(this->name, GF_LOG_ERROR,
                         "failed to "
                         "set the inode ctx (%s)",
                         uuid_utoa(inode->gfid));
    }
out:
    return ctx;
}

static arbiter_inode_ctx_t *
arbiter_inode_ctx_get(inode_t *inode, xlator_t *this)
{
    arbiter_inode_ctx_t *ctx = NULL;

    LOCK(&inode->lock);
    {
        ctx = __arbiter_inode_ctx_get(inode, this);
    }
    UNLOCK(&inode->lock);
    return ctx;
}

int32_t
arbiter_lookup_cbk(call_frame_t *frame, void *cookie, xlator_t *this,
                   int32_t op_ret, int32_t op_errno, inode_t *inode,
                   struct iatt *buf, dict_t *xdata, struct iatt *postparent)
{
    arbiter_inode_ctx_t *ctx = NULL;

    if (op_ret != 0)
        goto unwind;
    ctx = arbiter_inode_ctx_get(inode, this);
    if (!ctx) {
        op_ret = -1;
        op_errno = ENOMEM;
        goto unwind;
    }
    memcpy(&ctx->iattbuf, buf, sizeof(ctx->iattbuf));

unwind:
    STACK_UNWIND_STRICT(lookup, frame, op_ret, op_errno, inode, buf, xdata,
                        postparent);
    return 0;
}

int32_t
arbiter_lookup(call_frame_t *frame, xlator_t *this, loc_t *loc, dict_t *xdata)
{
    STACK_WIND(frame, arbiter_lookup_cbk, FIRST_CHILD(this),
               FIRST_CHILD(this)->fops->lookup, loc, xdata);
    return 0;
}

int32_t
arbiter_truncate(call_frame_t *frame, xlator_t *this, loc_t *loc, off_t offset,
                 dict_t *xdata)
{
    arbiter_inode_ctx_t *ctx = NULL;
    struct iatt *buf = NULL;
    int32_t op_ret = 0;
    int32_t op_errno = 0;

    ctx = arbiter_inode_ctx_get(loc->inode, this);
    if (!ctx) {
        op_ret = -1;
        op_errno = ENOMEM;
        goto unwind;
    }
    buf = &ctx->iattbuf;
unwind:
    STACK_UNWIND_STRICT(truncate, frame, op_ret, op_errno, buf, buf, NULL);
    return 0;
}

int32_t
arbiter_ftruncate(call_frame_t *frame, xlator_t *this, fd_t *fd, off_t offset,
                  dict_t *xdata)

{
    arbiter_inode_ctx_t *ctx = NULL;
    struct iatt *buf = NULL;
    int32_t op_ret = 0;
    int32_t op_errno = 0;

    ctx = arbiter_inode_ctx_get(fd->inode, this);
    if (!ctx) {
        op_ret = -1;
        op_errno = ENOMEM;
        goto unwind;
    }
    buf = &ctx->iattbuf;
unwind:
    STACK_UNWIND_STRICT(ftruncate, frame, op_ret, op_errno, buf, buf, NULL);
    return 0;
}

dict_t *
arbiter_fill_writev_xdata(fd_t *fd, dict_t *xdata, xlator_t *this)
{
    dict_t *rsp_xdata = NULL;
    int32_t ret = 0;
    int is_append = 1;

    if (!fd || !fd->inode || gf_uuid_is_null(fd->inode->gfid)) {
        goto out;
    }

    if (!xdata)
        goto out;

    rsp_xdata = dict_new();
    if (!rsp_xdata)
        goto out;

    if (dict_get(xdata, GLUSTERFS_OPEN_FD_COUNT)) {
        ret = dict_set_uint32(rsp_xdata, GLUSTERFS_OPEN_FD_COUNT,
                              fd->inode->fd_count);
        if (ret < 0) {
            gf_msg_debug(this->name, 0,
                         "Failed to set dict value"
                         " for GLUSTERFS_OPEN_FD_COUNT");
        }
    }
    if (dict_get(xdata, GLUSTERFS_WRITE_IS_APPEND)) {
        ret = dict_set_uint32(rsp_xdata, GLUSTERFS_WRITE_IS_APPEND, is_append);
        if (ret < 0) {
            gf_msg_debug(this->name, 0,
                         "Failed to set dict value"
                         " for GLUSTERFS_WRITE_IS_APPEND");
        }
    }
out:
    return rsp_xdata;
}

int32_t
arbiter_writev(call_frame_t *frame, xlator_t *this, fd_t *fd,
               struct iovec *vector, int32_t count, off_t off, uint32_t flags,
               struct iobref *iobref, dict_t *xdata)
{
    arbiter_inode_ctx_t *ctx = NULL;
    struct iatt *buf = NULL;
    dict_t *rsp_xdata = NULL;
    int op_ret = 0;
    int op_errno = 0;

    ctx = arbiter_inode_ctx_get(fd->inode, this);
    if (!ctx) {
        op_ret = -1;
        op_errno = ENOMEM;
        goto unwind;
    }
    buf = &ctx->iattbuf;
    op_ret = iov_length(vector, count);
    rsp_xdata = arbiter_fill_writev_xdata(fd, xdata, this);
unwind:
    STACK_UNWIND_STRICT(writev, frame, op_ret, op_errno, buf, buf, rsp_xdata);
    if (rsp_xdata)
        dict_unref(rsp_xdata);
    return 0;
}

int32_t
arbiter_fallocate(call_frame_t *frame, xlator_t *this, fd_t *fd,
                  int32_t keep_size, off_t offset, size_t len, dict_t *xdata)
{
    arbiter_inode_ctx_t *ctx = NULL;
    struct iatt *buf = NULL;
    int op_ret = 0;
    int op_errno = 0;

    ctx = arbiter_inode_ctx_get(fd->inode, this);
    if (!ctx) {
        op_ret = -1;
        op_errno = ENOMEM;
        goto unwind;
    }
    buf = &ctx->iattbuf;
unwind:
    STACK_UNWIND_STRICT(fallocate, frame, op_ret, op_errno, buf, buf, NULL);
    return 0;
}

int32_t
arbiter_discard(call_frame_t *frame, xlator_t *this, fd_t *fd, off_t offset,
                size_t len, dict_t *xdata)
{
    arbiter_inode_ctx_t *ctx = NULL;
    struct iatt *buf = NULL;
    int op_ret = 0;
    int op_errno = 0;

    ctx = arbiter_inode_ctx_get(fd->inode, this);
    if (!ctx) {
        op_ret = -1;
        op_errno = ENOMEM;
        goto unwind;
    }
    buf = &ctx->iattbuf;
unwind:
    STACK_UNWIND_STRICT(discard, frame, op_ret, op_errno, buf, buf, NULL);
    return 0;
}

int32_t
arbiter_zerofill(call_frame_t *frame, xlator_t *this, fd_t *fd, off_t offset,
                 off_t len, dict_t *xdata)
{
    arbiter_inode_ctx_t *ctx = NULL;
    struct iatt *buf = NULL;
    int op_ret = 0;
    int op_errno = 0;

    ctx = arbiter_inode_ctx_get(fd->inode, this);
    if (!ctx) {
        op_ret = -1;
        op_errno = ENOMEM;
        goto unwind;
    }
    buf = &ctx->iattbuf;
unwind:
    STACK_UNWIND_STRICT(zerofill, frame, op_ret, op_errno, buf, buf, NULL);
    return 0;
}

static int32_t
arbiter_readv(call_frame_t *frame, xlator_t *this, fd_t *fd, size_t size,
              off_t offset, uint32_t flags, dict_t *xdata)
{
    STACK_UNWIND_STRICT(readv, frame, -1, ENOSYS, NULL, 0, NULL, NULL, NULL);
    return 0;
}

static int32_t
arbiter_seek(call_frame_t *frame, xlator_t *this, fd_t *fd, off_t offset,
             gf_seek_what_t what, dict_t *xdata)
{
    STACK_UNWIND_STRICT(seek, frame, -1, ENOSYS, 0, xdata);
    return 0;
}

int32_t
mem_acct_init(xlator_t *this)
{
    int ret = -1;

    ret = xlator_mem_acct_init(this, gf_arbiter_mt_end + 1);
    if (ret)
        gf_log(this->name, GF_LOG_ERROR,
               "Memory accounting "
               "initialization failed.");
    return ret;
}

int
reconfigure(xlator_t *this, dict_t *options)
{
    return 0;
}

int
arbiter_forget(xlator_t *this, inode_t *inode)
{
    arbiter_inode_ctx_t *ctx = NULL;
    uint64_t ctx_addr = 0;

    inode_ctx_del(inode, this, &ctx_addr);
    if (!ctx_addr)
        return 0;
    ctx = (arbiter_inode_ctx_t *)(long)ctx_addr;
    GF_FREE(ctx);
    return 0;
}

int32_t
init(xlator_t *this)
{
    if (!this->children || this->children->next) {
        gf_log(this->name, GF_LOG_ERROR,
               "'arbiter' not configured with exactly one child");
        return -1;
    }

    if (!this->parents)
        gf_log(this->name, GF_LOG_ERROR, "dangling volume. check volfile ");

    return 0;
}

void
fini(xlator_t *this)
{
    return;
}

struct xlator_fops fops = {
    .lookup = arbiter_lookup,

    /* Return success for these inode write FOPS without winding it down to
     * posix; this is needed for AFR write transaction logic to work.*/
    .truncate = arbiter_truncate,
    .writev = arbiter_writev,
    .ftruncate = arbiter_ftruncate,
    .fallocate = arbiter_fallocate,
    .discard = arbiter_discard,
    .zerofill = arbiter_zerofill,

    /* AFR is not expected to wind these inode read FOPS initiated by the
     * application to the arbiter brick. But in case a bug causes them
     * to be called, we return ENOSYS. */
    .readv = arbiter_readv,
    .seek = arbiter_seek,

    /* The following inode read FOPS initiated by the application are not
     * wound by AFR either but internal logic like  shd, glfsheal and
     * client side healing in AFR will send them for selfheal/ inode refresh
     * operations etc.,so we need to wind them down to posix:
     *
     * (f)stat, readdir(p), readlink, (f)getxattr.*/

    /* All other FOPs not listed here are safe to be wound down to posix.*/
};

struct xlator_cbks cbks = {
    .forget = arbiter_forget,
};

struct volume_options options[] = {
    {.key = {NULL}},
};