Blame view
fs/nfs/nfs4filelayout.c
26 KB
7ab672ce3 NFSv4.1: pnfs: fi... |
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 |
/* * Module for the pnfs nfs4 file layout driver. * Defines all I/O and Policy interface operations, plus code * to register itself with the pNFS client. * * Copyright (c) 2002 * The Regents of the University of Michigan * All Rights Reserved * * Dean Hildebrand <dhildebz@umich.edu> * * Permission is granted to use, copy, create derivative works, and * redistribute this software and such derivative works for any purpose, * so long as the name of the University of Michigan is not used in * any advertising or publicity pertaining to the use or distribution * of this software without specific, written prior authorization. If * the above copyright notice or any other identification of the * University of Michigan is included in any copy of any portion of * this software, then the disclaimer below must also be included. * * This software is provided as is, without representation or warranty * of any kind either express or implied, including without limitation * the implied warranties of merchantability, fitness for a particular * purpose, or noninfringement. The Regents of the University of * Michigan shall not be liable for any damages, including special, * indirect, incidental, or consequential damages, with respect to any * claim arising out of or in connection with the use of the software, * even if it has been or is hereafter advised of the possibility of * such damages. */ #include <linux/nfs_fs.h> |
19345cb29 NFSv4.1: file lay... |
33 |
#include <linux/nfs_page.h> |
143cb494c fs: add module.h ... |
34 |
#include <linux/module.h> |
16b374ca4 NFSv4.1: pnfs: fi... |
35 36 37 |
#include "internal.h" #include "nfs4filelayout.h" |
7ab672ce3 NFSv4.1: pnfs: fi... |
38 39 40 41 42 43 |
#define NFSDBG_FACILITY NFSDBG_PNFS_LD MODULE_LICENSE("GPL"); MODULE_AUTHOR("Dean Hildebrand <dhildebz@umich.edu>"); MODULE_DESCRIPTION("The NFSv4 file layout driver"); |
cbdabc7f8 NFSv4.1: filelayo... |
44 |
#define FILELAYOUT_POLL_RETRY_MAX (15*HZ) |
cfe7f4120 NFSv4.1: filelayo... |
45 46 47 48 49 |
static loff_t filelayout_get_dense_offset(struct nfs4_filelayout_segment *flseg, loff_t offset) { u32 stripe_width = flseg->stripe_unit * flseg->dsaddr->stripe_count; |
3476f114a nfs: fix a minor ... |
50 51 |
u64 stripe_no; u32 rem; |
cfe7f4120 NFSv4.1: filelayo... |
52 53 |
offset -= flseg->pattern_offset; |
3476f114a nfs: fix a minor ... |
54 55 |
stripe_no = div_u64(offset, stripe_width); div_u64_rem(offset, flseg->stripe_unit, &rem); |
cfe7f4120 NFSv4.1: filelayo... |
56 |
|
3476f114a nfs: fix a minor ... |
57 |
return stripe_no * flseg->stripe_unit + rem; |
cfe7f4120 NFSv4.1: filelayo... |
58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 |
} /* This function is used by the layout driver to calculate the * offset of the file on the dserver based on whether the * layout type is STRIPE_DENSE or STRIPE_SPARSE */ static loff_t filelayout_get_dserver_offset(struct pnfs_layout_segment *lseg, loff_t offset) { struct nfs4_filelayout_segment *flseg = FILELAYOUT_LSEG(lseg); switch (flseg->stripe_type) { case STRIPE_SPARSE: return offset; case STRIPE_DENSE: return filelayout_get_dense_offset(flseg, offset); } BUG(); } |
cbdabc7f8 NFSv4.1: filelayo... |
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 |
static int filelayout_async_handle_error(struct rpc_task *task, struct nfs4_state *state, struct nfs_client *clp, int *reset) { if (task->tk_status >= 0) return 0; *reset = 0; switch (task->tk_status) { case -NFS4ERR_BADSESSION: case -NFS4ERR_BADSLOT: case -NFS4ERR_BAD_HIGH_SLOT: case -NFS4ERR_DEADSESSION: case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: case -NFS4ERR_SEQ_FALSE_RETRY: case -NFS4ERR_SEQ_MISORDERED: dprintk("%s ERROR %d, Reset session. Exchangeid " "flags 0x%x ", __func__, task->tk_status, clp->cl_exchange_flags); nfs4_schedule_session_recovery(clp->cl_session); break; case -NFS4ERR_DELAY: case -NFS4ERR_GRACE: case -EKEYEXPIRED: rpc_delay(task, FILELAYOUT_POLL_RETRY_MAX); break; |
a8a4ae3a8 NFSv41: Resend on... |
108 109 |
case -NFS4ERR_RETRY_UNCACHED_REP: break; |
cbdabc7f8 NFSv4.1: filelayo... |
110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 |
default: dprintk("%s DS error. Retry through MDS %d ", __func__, task->tk_status); *reset = 1; break; } task->tk_status = 0; return -EAGAIN; } /* NFS_PROTO call done callback routines */ static int filelayout_read_done_cb(struct rpc_task *task, struct nfs_read_data *data) { |
cbdabc7f8 NFSv4.1: filelayo... |
126 127 128 129 130 131 132 133 134 135 136 |
int reset = 0; dprintk("%s DS read ", __func__); if (filelayout_async_handle_error(task, data->args.context->state, data->ds_clp, &reset) == -EAGAIN) { dprintk("%s calling restart ds_clp %p ds_clp->cl_session %p ", __func__, data->ds_clp, data->ds_clp->cl_session); if (reset) { |
1b0ae0687 pnfs: make _set_l... |
137 |
pnfs_set_lo_fail(data->lseg); |
cbdabc7f8 NFSv4.1: filelayo... |
138 |
nfs4_reset_read(task, data); |
cbdabc7f8 NFSv4.1: filelayo... |
139 |
} |
d00c5d438 NFS: Get rid of n... |
140 |
rpc_restart_call_prepare(task); |
cbdabc7f8 NFSv4.1: filelayo... |
141 142 143 144 145 |
return -EAGAIN; } return 0; } |
16b374ca4 NFSv4.1: pnfs: fi... |
146 |
/* |
863a3c6c6 NFSv4.1: layoutco... |
147 148 149 150 151 152 153 154 155 156 157 158 159 160 |
* We reference the rpc_cred of the first WRITE that triggers the need for * a LAYOUTCOMMIT, and use it to send the layoutcommit compound. * rfc5661 is not clear about which credential should be used. */ static void filelayout_set_layoutcommit(struct nfs_write_data *wdata) { if (FILELAYOUT_LSEG(wdata->lseg)->commit_through_mds || wdata->res.verf->committed == NFS_FILE_SYNC) return; pnfs_set_layoutcommit(wdata); dprintk("%s ionde %lu pls_end_pos %lu ", __func__, wdata->inode->i_ino, |
acff58805 pnfs: save layout... |
161 |
(unsigned long) NFS_I(wdata->inode)->layout->plh_lwb); |
863a3c6c6 NFSv4.1: layoutco... |
162 163 164 |
} /* |
dc70d7b31 NFSv4.1: filelayo... |
165 166 167 168 169 170 171 |
* Call ops for the async read/write cases * In the case of dense layouts, the offset needs to be reset to its * original value. */ static void filelayout_read_prepare(struct rpc_task *task, void *data) { struct nfs_read_data *rdata = (struct nfs_read_data *)data; |
cbdabc7f8 NFSv4.1: filelayo... |
172 |
rdata->read_done_cb = filelayout_read_done_cb; |
dc70d7b31 NFSv4.1: filelayo... |
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 |
if (nfs41_setup_sequence(rdata->ds_clp->cl_session, &rdata->args.seq_args, &rdata->res.seq_res, 0, task)) return; rpc_call_start(task); } static void filelayout_read_call_done(struct rpc_task *task, void *data) { struct nfs_read_data *rdata = (struct nfs_read_data *)data; dprintk("--> %s task->tk_status %d ", __func__, task->tk_status); /* Note this may cause RPC to be resent */ rdata->mds_ops->rpc_call_done(task, data); } static void filelayout_read_release(void *data) { struct nfs_read_data *rdata = (struct nfs_read_data *)data; rdata->mds_ops->rpc_release(data); } |
a69aef149 NFSv4.1: pnfs fil... |
198 199 200 201 202 203 204 |
static int filelayout_write_done_cb(struct rpc_task *task, struct nfs_write_data *data) { int reset = 0; if (filelayout_async_handle_error(task, data->args.context->state, data->ds_clp, &reset) == -EAGAIN) { |
a69aef149 NFSv4.1: pnfs fil... |
205 206 207 208 |
dprintk("%s calling restart ds_clp %p ds_clp->cl_session %p ", __func__, data->ds_clp, data->ds_clp->cl_session); if (reset) { |
1b0ae0687 pnfs: make _set_l... |
209 |
pnfs_set_lo_fail(data->lseg); |
a69aef149 NFSv4.1: pnfs fil... |
210 |
nfs4_reset_write(task, data); |
d00c5d438 NFS: Get rid of n... |
211 212 |
} rpc_restart_call_prepare(task); |
a69aef149 NFSv4.1: pnfs fil... |
213 214 |
return -EAGAIN; } |
863a3c6c6 NFSv4.1: layoutco... |
215 |
filelayout_set_layoutcommit(data); |
a69aef149 NFSv4.1: pnfs fil... |
216 217 |
return 0; } |
e0c2b3801 NFSv4.1: filelayo... |
218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 |
/* Fake up some data that will cause nfs_commit_release to retry the writes. */ static void prepare_to_resend_writes(struct nfs_write_data *data) { struct nfs_page *first = nfs_list_entry(data->pages.next); data->task.tk_status = 0; memcpy(data->verf.verifier, first->wb_verf.verifier, sizeof(first->wb_verf.verifier)); data->verf.verifier[0]++; /* ensure verifier mismatch */ } static int filelayout_commit_done_cb(struct rpc_task *task, struct nfs_write_data *data) { int reset = 0; if (filelayout_async_handle_error(task, data->args.context->state, data->ds_clp, &reset) == -EAGAIN) { dprintk("%s calling restart ds_clp %p ds_clp->cl_session %p ", __func__, data->ds_clp, data->ds_clp->cl_session); if (reset) { prepare_to_resend_writes(data); |
1b0ae0687 pnfs: make _set_l... |
241 |
pnfs_set_lo_fail(data->lseg); |
e0c2b3801 NFSv4.1: filelayo... |
242 |
} else |
d00c5d438 NFS: Get rid of n... |
243 |
rpc_restart_call_prepare(task); |
e0c2b3801 NFSv4.1: filelayo... |
244 245 246 247 248 |
return -EAGAIN; } return 0; } |
a69aef149 NFSv4.1: pnfs fil... |
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 |
static void filelayout_write_prepare(struct rpc_task *task, void *data) { struct nfs_write_data *wdata = (struct nfs_write_data *)data; if (nfs41_setup_sequence(wdata->ds_clp->cl_session, &wdata->args.seq_args, &wdata->res.seq_res, 0, task)) return; rpc_call_start(task); } static void filelayout_write_call_done(struct rpc_task *task, void *data) { struct nfs_write_data *wdata = (struct nfs_write_data *)data; /* Note this may cause RPC to be resent */ wdata->mds_ops->rpc_call_done(task, data); } static void filelayout_write_release(void *data) { struct nfs_write_data *wdata = (struct nfs_write_data *)data; wdata->mds_ops->rpc_release(data); } |
e0c2b3801 NFSv4.1: filelayo... |
275 276 277 278 279 280 281 282 283 |
static void filelayout_commit_release(void *data) { struct nfs_write_data *wdata = (struct nfs_write_data *)data; nfs_commit_release_pages(wdata); if (atomic_dec_and_test(&NFS_I(wdata->inode)->commits_outstanding)) nfs_commit_clear_lock(NFS_I(wdata->inode)); nfs_commitdata_release(wdata); } |
dc70d7b31 NFSv4.1: filelayo... |
284 285 286 287 288 |
struct rpc_call_ops filelayout_read_call_ops = { .rpc_call_prepare = filelayout_read_prepare, .rpc_call_done = filelayout_read_call_done, .rpc_release = filelayout_read_release, }; |
a69aef149 NFSv4.1: pnfs fil... |
289 290 291 292 293 |
struct rpc_call_ops filelayout_write_call_ops = { .rpc_call_prepare = filelayout_write_prepare, .rpc_call_done = filelayout_write_call_done, .rpc_release = filelayout_write_release, }; |
e0c2b3801 NFSv4.1: filelayo... |
294 295 296 297 298 |
struct rpc_call_ops filelayout_commit_call_ops = { .rpc_call_prepare = filelayout_write_prepare, .rpc_call_done = filelayout_write_call_done, .rpc_release = filelayout_commit_release, }; |
dc70d7b31 NFSv4.1: filelayo... |
299 300 301 302 303 304 305 306 307 308 309 310 311 312 |
static enum pnfs_try_status filelayout_read_pagelist(struct nfs_read_data *data) { struct pnfs_layout_segment *lseg = data->lseg; struct nfs4_pnfs_ds *ds; loff_t offset = data->args.offset; u32 j, idx; struct nfs_fh *fh; int status; dprintk("--> %s ino %lu pgbase %u req %Zu@%llu ", __func__, data->inode->i_ino, data->args.pgbase, (size_t)data->args.count, offset); |
c47abcf8f NFSv4.1: do not u... |
313 314 |
if (test_bit(NFS_DEVICEID_INVALID, &FILELAYOUT_DEVID_NODE(lseg)->flags)) return PNFS_NOT_ATTEMPTED; |
dc70d7b31 NFSv4.1: filelayo... |
315 316 317 318 319 |
/* Retrieve the correct rpc_client for the byte range */ j = nfs4_fl_calc_j_index(lseg, offset); idx = nfs4_fl_calc_ds_index(lseg, j); ds = nfs4_fl_prepare_ds(lseg, idx); if (!ds) { |
568e8c494 NFSv4.1: turn off... |
320 321 322 |
/* Either layout fh index faulty, or ds connect failed */ set_bit(lo_fail_bit(IOMODE_RW), &lseg->pls_layout->plh_flags); set_bit(lo_fail_bit(IOMODE_READ), &lseg->pls_layout->plh_flags); |
dc70d7b31 NFSv4.1: filelayo... |
323 324 |
return PNFS_NOT_ATTEMPTED; } |
c9895cb69 NFS: pnfs IPv6 su... |
325 326 |
dprintk("%s USE DS: %s ", __func__, ds->ds_remotestr); |
dc70d7b31 NFSv4.1: filelayo... |
327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 |
/* No multipath support. Use first DS */ data->ds_clp = ds->ds_clp; fh = nfs4_fl_select_ds_fh(lseg, j); if (fh) data->args.fh = fh; data->args.offset = filelayout_get_dserver_offset(lseg, offset); data->mds_offset = offset; /* Perform an asynchronous read to ds */ status = nfs_initiate_read(data, ds->ds_clp->cl_rpcclient, &filelayout_read_call_ops); BUG_ON(status != 0); return PNFS_ATTEMPTED; } |
a69aef149 NFSv4.1: pnfs fil... |
343 |
/* Perform async writes. */ |
0382b7440 NFSv4.1: implemen... |
344 345 346 |
static enum pnfs_try_status filelayout_write_pagelist(struct nfs_write_data *data, int sync) { |
a69aef149 NFSv4.1: pnfs fil... |
347 348 349 350 351 352 |
struct pnfs_layout_segment *lseg = data->lseg; struct nfs4_pnfs_ds *ds; loff_t offset = data->args.offset; u32 j, idx; struct nfs_fh *fh; int status; |
c47abcf8f NFSv4.1: do not u... |
353 354 |
if (test_bit(NFS_DEVICEID_INVALID, &FILELAYOUT_DEVID_NODE(lseg)->flags)) return PNFS_NOT_ATTEMPTED; |
a69aef149 NFSv4.1: pnfs fil... |
355 356 357 358 359 360 361 362 363 364 365 |
/* Retrieve the correct rpc_client for the byte range */ j = nfs4_fl_calc_j_index(lseg, offset); idx = nfs4_fl_calc_ds_index(lseg, j); ds = nfs4_fl_prepare_ds(lseg, idx); if (!ds) { printk(KERN_ERR "%s: prepare_ds failed, use MDS ", __func__); set_bit(lo_fail_bit(IOMODE_RW), &lseg->pls_layout->plh_flags); set_bit(lo_fail_bit(IOMODE_READ), &lseg->pls_layout->plh_flags); return PNFS_NOT_ATTEMPTED; } |
c9895cb69 NFS: pnfs IPv6 su... |
366 367 |
dprintk("%s ino %lu sync %d req %Zu@%llu DS: %s ", __func__, |
a69aef149 NFSv4.1: pnfs fil... |
368 |
data->inode->i_ino, sync, (size_t) data->args.count, offset, |
c9895cb69 NFS: pnfs IPv6 su... |
369 |
ds->ds_remotestr); |
a69aef149 NFSv4.1: pnfs fil... |
370 |
|
a69aef149 NFSv4.1: pnfs fil... |
371 372 373 374 375 376 377 378 379 380 |
data->write_done_cb = filelayout_write_done_cb; data->ds_clp = ds->ds_clp; fh = nfs4_fl_select_ds_fh(lseg, j); if (fh) data->args.fh = fh; /* * Get the file offset on the dserver. Set the write offset to * this offset and save the original offset. */ data->args.offset = filelayout_get_dserver_offset(lseg, offset); |
a69aef149 NFSv4.1: pnfs fil... |
381 382 383 384 385 386 |
/* Perform an asynchronous write */ status = nfs_initiate_write(data, ds->ds_clp->cl_rpcclient, &filelayout_write_call_ops, sync); BUG_ON(status != 0); return PNFS_ATTEMPTED; |
0382b7440 NFSv4.1: implemen... |
387 |
} |
dc70d7b31 NFSv4.1: filelayo... |
388 |
/* |
16b374ca4 NFSv4.1: pnfs: fi... |
389 390 391 392 393 394 395 396 397 398 399 |
* filelayout_check_layout() * * Make sure layout segment parameters are sane WRT the device. * At this point no generic layer initialization of the lseg has occurred, * and nothing has been added to the layout_hdr cache. * */ static int filelayout_check_layout(struct pnfs_layout_hdr *lo, struct nfs4_filelayout_segment *fl, struct nfs4_layoutget_res *lgr, |
a75b9df9d NFSv4.1: Ensure t... |
400 401 |
struct nfs4_deviceid *id, gfp_t gfp_flags) |
16b374ca4 NFSv4.1: pnfs: fi... |
402 |
{ |
a1eaecbc4 NFSv4.1: make dev... |
403 |
struct nfs4_deviceid_node *d; |
16b374ca4 NFSv4.1: pnfs: fi... |
404 405 |
struct nfs4_file_layout_dsaddr *dsaddr; int status = -EINVAL; |
b7edfaa19 pnfs: add prefix ... |
406 |
struct nfs_server *nfss = NFS_SERVER(lo->plh_inode); |
16b374ca4 NFSv4.1: pnfs: fi... |
407 408 409 |
dprintk("--> %s ", __func__); |
7c24d9489 NFSv4.1: File lay... |
410 411 412 413 414 415 416 417 |
/* FIXME: remove this check when layout segment support is added */ if (lgr->range.offset != 0 || lgr->range.length != NFS4_MAX_UINT64) { dprintk("%s Only whole file layouts supported. Use MDS i/o ", __func__); goto out; } |
16b374ca4 NFSv4.1: pnfs: fi... |
418 |
if (fl->pattern_offset > lgr->range.offset) { |
3b6445a6f NFSv4.1: fix typo... |
419 420 |
dprintk("%s pattern_offset %lld too large ", |
16b374ca4 NFSv4.1: pnfs: fi... |
421 422 423 |
__func__, fl->pattern_offset); goto out; } |
75247affd NFSv4.1: reject z... |
424 425 426 |
if (!fl->stripe_unit || fl->stripe_unit % PAGE_SIZE) { dprintk("%s Invalid stripe unit (%u) ", |
16b374ca4 NFSv4.1: pnfs: fi... |
427 428 429 430 431 |
__func__, fl->stripe_unit); goto out; } /* find and reference the deviceid */ |
35c8bb543 NFSv4.1: use layo... |
432 433 |
d = nfs4_find_get_deviceid(NFS_SERVER(lo->plh_inode)->pnfs_curr_ld, NFS_SERVER(lo->plh_inode)->nfs_client, id); |
a1eaecbc4 NFSv4.1: make dev... |
434 |
if (d == NULL) { |
a75b9df9d NFSv4.1: Ensure t... |
435 |
dsaddr = get_device_info(lo->plh_inode, id, gfp_flags); |
16b374ca4 NFSv4.1: pnfs: fi... |
436 437 |
if (dsaddr == NULL) goto out; |
a1eaecbc4 NFSv4.1: make dev... |
438 439 |
} else dsaddr = container_of(d, struct nfs4_file_layout_dsaddr, id_node); |
c47abcf8f NFSv4.1: do not u... |
440 441 442 |
/* Found deviceid is being reaped */ if (test_bit(NFS_DEVICEID_INVALID, &dsaddr->id_node.flags)) goto out_put; |
16b374ca4 NFSv4.1: pnfs: fi... |
443 |
fl->dsaddr = dsaddr; |
e414966b8 NFS: Remove no-op... |
444 445 446 |
if (fl->first_stripe_index >= dsaddr->stripe_count) { dprintk("%s Bad first_stripe_index %u ", |
16b374ca4 NFSv4.1: pnfs: fi... |
447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 |
__func__, fl->first_stripe_index); goto out_put; } if ((fl->stripe_type == STRIPE_SPARSE && fl->num_fh > 1 && fl->num_fh != dsaddr->ds_num) || (fl->stripe_type == STRIPE_DENSE && fl->num_fh != dsaddr->stripe_count)) { dprintk("%s num_fh %u not valid for given packing ", __func__, fl->num_fh); goto out_put; } if (fl->stripe_unit % nfss->rsize || fl->stripe_unit % nfss->wsize) { dprintk("%s Stripe unit (%u) not aligned with rsize %u " "wsize %u ", __func__, fl->stripe_unit, nfss->rsize, nfss->wsize); } status = 0; out: dprintk("--> %s returns %d ", __func__, status); return status; out_put: |
ea8eecdd1 NFSv4.1 move devi... |
474 |
nfs4_fl_put_deviceid(dsaddr); |
16b374ca4 NFSv4.1: pnfs: fi... |
475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 |
goto out; } static void filelayout_free_fh_array(struct nfs4_filelayout_segment *fl) { int i; for (i = 0; i < fl->num_fh; i++) { if (!fl->fh_array[i]) break; kfree(fl->fh_array[i]); } kfree(fl->fh_array); fl->fh_array = NULL; } static void _filelayout_free_lseg(struct nfs4_filelayout_segment *fl) { filelayout_free_fh_array(fl); kfree(fl); } static int filelayout_decode_layout(struct pnfs_layout_hdr *flo, struct nfs4_filelayout_segment *fl, struct nfs4_layoutget_res *lgr, |
a75b9df9d NFSv4.1: Ensure t... |
502 503 |
struct nfs4_deviceid *id, gfp_t gfp_flags) |
16b374ca4 NFSv4.1: pnfs: fi... |
504 |
{ |
35124a099 Cleanup XDR parsi... |
505 |
struct xdr_stream stream; |
f7da7a129 SUNRPC: introduce... |
506 |
struct xdr_buf buf; |
35124a099 Cleanup XDR parsi... |
507 508 |
struct page *scratch; __be32 *p; |
16b374ca4 NFSv4.1: pnfs: fi... |
509 510 511 512 513 |
uint32_t nfl_util; int i; dprintk("%s: set_layout_map Begin ", __func__); |
a75b9df9d NFSv4.1: Ensure t... |
514 |
scratch = alloc_page(gfp_flags); |
35124a099 Cleanup XDR parsi... |
515 516 |
if (!scratch) return -ENOMEM; |
f7da7a129 SUNRPC: introduce... |
517 |
xdr_init_decode_pages(&stream, &buf, lgr->layoutp->pages, lgr->layoutp->len); |
35124a099 Cleanup XDR parsi... |
518 519 520 521 522 523 524 |
xdr_set_scratch_buffer(&stream, page_address(scratch), PAGE_SIZE); /* 20 = ufl_util (4), first_stripe_index (4), pattern_offset (8), * num_fh (4) */ p = xdr_inline_decode(&stream, NFS4_DEVICEID4_SIZE + 20); if (unlikely(!p)) goto out_err; |
16b374ca4 NFSv4.1: pnfs: fi... |
525 526 |
memcpy(id, p, sizeof(*id)); p += XDR_QUADLEN(NFS4_DEVICEID4_SIZE); |
a1eaecbc4 NFSv4.1: make dev... |
527 |
nfs4_print_deviceid(id); |
16b374ca4 NFSv4.1: pnfs: fi... |
528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 |
nfl_util = be32_to_cpup(p++); if (nfl_util & NFL4_UFLG_COMMIT_THRU_MDS) fl->commit_through_mds = 1; if (nfl_util & NFL4_UFLG_DENSE) fl->stripe_type = STRIPE_DENSE; else fl->stripe_type = STRIPE_SPARSE; fl->stripe_unit = nfl_util & ~NFL4_UFLG_MASK; fl->first_stripe_index = be32_to_cpup(p++); p = xdr_decode_hyper(p, &fl->pattern_offset); fl->num_fh = be32_to_cpup(p++); dprintk("%s: nfl_util 0x%X num_fh %u fsi %u po %llu ", __func__, nfl_util, fl->num_fh, fl->first_stripe_index, fl->pattern_offset); |
cec765cf5 NFSv4.1: allow ze... |
546 547 |
/* Note that a zero value for num_fh is legal for STRIPE_SPARSE. * Futher checking is done in filelayout_check_layout */ |
e414966b8 NFS: Remove no-op... |
548 |
if (fl->num_fh > |
cec765cf5 NFSv4.1: allow ze... |
549 |
max(NFS4_PNFS_MAX_STRIPE_CNT, NFS4_PNFS_MAX_MULTI_CNT)) |
35124a099 Cleanup XDR parsi... |
550 |
goto out_err; |
cec765cf5 NFSv4.1: allow ze... |
551 552 553 554 555 556 |
if (fl->num_fh > 0) { fl->fh_array = kzalloc(fl->num_fh * sizeof(struct nfs_fh *), gfp_flags); if (!fl->fh_array) goto out_err; } |
16b374ca4 NFSv4.1: pnfs: fi... |
557 558 559 |
for (i = 0; i < fl->num_fh; i++) { /* Do we want to use a mempool here? */ |
a75b9df9d NFSv4.1: Ensure t... |
560 |
fl->fh_array[i] = kmalloc(sizeof(struct nfs_fh), gfp_flags); |
35124a099 Cleanup XDR parsi... |
561 562 563 564 565 566 |
if (!fl->fh_array[i]) goto out_err_free; p = xdr_inline_decode(&stream, 4); if (unlikely(!p)) goto out_err_free; |
16b374ca4 NFSv4.1: pnfs: fi... |
567 568 569 570 571 |
fl->fh_array[i]->size = be32_to_cpup(p++); if (sizeof(struct nfs_fh) < fl->fh_array[i]->size) { printk(KERN_ERR "Too big fh %d received %d ", i, fl->fh_array[i]->size); |
35124a099 Cleanup XDR parsi... |
572 |
goto out_err_free; |
16b374ca4 NFSv4.1: pnfs: fi... |
573 |
} |
35124a099 Cleanup XDR parsi... |
574 575 576 577 |
p = xdr_inline_decode(&stream, fl->fh_array[i]->size); if (unlikely(!p)) goto out_err_free; |
16b374ca4 NFSv4.1: pnfs: fi... |
578 |
memcpy(fl->fh_array[i]->data, p, fl->fh_array[i]->size); |
16b374ca4 NFSv4.1: pnfs: fi... |
579 580 581 582 |
dprintk("DEBUG: %s: fh len %d ", __func__, fl->fh_array[i]->size); } |
35124a099 Cleanup XDR parsi... |
583 |
__free_page(scratch); |
16b374ca4 NFSv4.1: pnfs: fi... |
584 |
return 0; |
35124a099 Cleanup XDR parsi... |
585 586 587 588 589 590 |
out_err_free: filelayout_free_fh_array(fl); out_err: __free_page(scratch); return -EIO; |
16b374ca4 NFSv4.1: pnfs: fi... |
591 |
} |
c879513e9 NFSv4.1: shift fi... |
592 593 594 595 596 597 598 599 |
static void filelayout_free_lseg(struct pnfs_layout_segment *lseg) { struct nfs4_filelayout_segment *fl = FILELAYOUT_LSEG(lseg); dprintk("--> %s ", __func__); nfs4_fl_put_deviceid(fl->dsaddr); |
425eb736c NFSv4.1: alloc an... |
600 |
kfree(fl->commit_buckets); |
c879513e9 NFSv4.1: shift fi... |
601 602 |
_filelayout_free_lseg(fl); } |
16b374ca4 NFSv4.1: pnfs: fi... |
603 604 |
static struct pnfs_layout_segment * filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid, |
a75b9df9d NFSv4.1: Ensure t... |
605 606 |
struct nfs4_layoutget_res *lgr, gfp_t gfp_flags) |
16b374ca4 NFSv4.1: pnfs: fi... |
607 608 609 610 611 612 613 |
{ struct nfs4_filelayout_segment *fl; int rc; struct nfs4_deviceid id; dprintk("--> %s ", __func__); |
a75b9df9d NFSv4.1: Ensure t... |
614 |
fl = kzalloc(sizeof(*fl), gfp_flags); |
16b374ca4 NFSv4.1: pnfs: fi... |
615 616 |
if (!fl) return NULL; |
a75b9df9d NFSv4.1: Ensure t... |
617 618 |
rc = filelayout_decode_layout(layoutid, fl, lgr, &id, gfp_flags); if (rc != 0 || filelayout_check_layout(layoutid, fl, lgr, &id, gfp_flags)) { |
16b374ca4 NFSv4.1: pnfs: fi... |
619 620 621 |
_filelayout_free_lseg(fl); return NULL; } |
425eb736c NFSv4.1: alloc an... |
622 623 624 625 626 627 628 629 630 631 632 |
/* This assumes there is only one IOMODE_RW lseg. What * we really want to do is have a layout_hdr level * dictionary of <multipath_list4, fh> keys, each * associated with a struct list_head, populated by calls * to filelayout_write_pagelist(). * */ if ((!fl->commit_through_mds) && (lgr->range.iomode == IOMODE_RW)) { int i; int size = (fl->stripe_type == STRIPE_SPARSE) ? fl->dsaddr->ds_num : fl->dsaddr->stripe_count; |
a75b9df9d NFSv4.1: Ensure t... |
633 |
fl->commit_buckets = kcalloc(size, sizeof(struct list_head), gfp_flags); |
425eb736c NFSv4.1: alloc an... |
634 635 636 637 638 639 640 641 |
if (!fl->commit_buckets) { filelayout_free_lseg(&fl->generic_hdr); return NULL; } fl->number_of_buckets = size; for (i = 0; i < size; i++) INIT_LIST_HEAD(&fl->commit_buckets[i]); } |
16b374ca4 NFSv4.1: pnfs: fi... |
642 643 |
return &fl->generic_hdr; } |
94ad1c80e NFSv4.1: coelesce... |
644 645 646 |
/* * filelayout_pg_test(). Called by nfs_can_coalesce_requests() * |
18ad0a9f2 NFSv4.1: change p... |
647 648 |
* return true : coalesce page * return false : don't coalesce page |
94ad1c80e NFSv4.1: coelesce... |
649 |
*/ |
1751c3638 NFS: Cleanup of t... |
650 |
static bool |
94ad1c80e NFSv4.1: coelesce... |
651 652 653 654 655 |
filelayout_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, struct nfs_page *req) { u64 p_stripe, r_stripe; u32 stripe_unit; |
19345cb29 NFSv4.1: file lay... |
656 657 658 |
if (!pnfs_generic_pg_test(pgio, prev, req) || !nfs_generic_pg_test(pgio, prev, req)) return false; |
89a58e32d NFSv4.1: use pnfs... |
659 |
|
94ad1c80e NFSv4.1: coelesce... |
660 661 662 663 664 665 666 667 668 |
p_stripe = (u64)prev->wb_index << PAGE_CACHE_SHIFT; r_stripe = (u64)req->wb_index << PAGE_CACHE_SHIFT; stripe_unit = FILELAYOUT_LSEG(pgio->pg_lseg)->stripe_unit; do_div(p_stripe, stripe_unit); do_div(r_stripe, stripe_unit); return (p_stripe == r_stripe); } |
7c24d9489 NFSv4.1: File lay... |
669 670 671 672 673 674 675 676 677 678 679 680 681 682 |
void filelayout_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) { BUG_ON(pgio->pg_lseg != NULL); pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, req->wb_context, 0, NFS4_MAX_UINT64, IOMODE_READ, GFP_KERNEL); /* If no lseg, fall back to read through mds */ if (pgio->pg_lseg == NULL) |
1f9453578 NFS: Clean up - s... |
683 |
nfs_pageio_reset_read_mds(pgio); |
7c24d9489 NFSv4.1: File lay... |
684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 |
} void filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) { BUG_ON(pgio->pg_lseg != NULL); pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, req->wb_context, 0, NFS4_MAX_UINT64, IOMODE_RW, GFP_NOFS); /* If no lseg, fall back to write through mds */ if (pgio->pg_lseg == NULL) |
1f9453578 NFS: Clean up - s... |
700 |
nfs_pageio_reset_write_mds(pgio); |
7c24d9489 NFSv4.1: File lay... |
701 |
} |
1751c3638 NFS: Cleanup of t... |
702 |
static const struct nfs_pageio_ops filelayout_pg_read_ops = { |
7c24d9489 NFSv4.1: File lay... |
703 |
.pg_init = filelayout_pg_init_read, |
1751c3638 NFS: Cleanup of t... |
704 |
.pg_test = filelayout_pg_test, |
493292ddc NFS: Move the pnf... |
705 |
.pg_doio = pnfs_generic_pg_readpages, |
1751c3638 NFS: Cleanup of t... |
706 707 708 |
}; static const struct nfs_pageio_ops filelayout_pg_write_ops = { |
7c24d9489 NFSv4.1: File lay... |
709 |
.pg_init = filelayout_pg_init_write, |
1751c3638 NFS: Cleanup of t... |
710 |
.pg_test = filelayout_pg_test, |
dce81290e NFS: Move the pnf... |
711 |
.pg_doio = pnfs_generic_pg_writepages, |
1751c3638 NFS: Cleanup of t... |
712 |
}; |
e0c2b3801 NFSv4.1: filelayo... |
713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 |
static bool filelayout_mark_pnfs_commit(struct pnfs_layout_segment *lseg) { return !FILELAYOUT_LSEG(lseg)->commit_through_mds; } static u32 select_bucket_index(struct nfs4_filelayout_segment *fl, u32 j) { if (fl->stripe_type == STRIPE_SPARSE) return nfs4_fl_calc_ds_index(&fl->generic_hdr, j); else return j; } struct list_head *filelayout_choose_commit_list(struct nfs_page *req) { struct pnfs_layout_segment *lseg = req->wb_commit_lseg; struct nfs4_filelayout_segment *fl = FILELAYOUT_LSEG(lseg); u32 i, j; struct list_head *list; /* Note that we are calling nfs4_fl_calc_j_index on each page * that ends up being committed to a data server. An attractive * alternative is to add a field to nfs_write_data and nfs_page * to store the value calculated in filelayout_write_pagelist * and just use that here. */ j = nfs4_fl_calc_j_index(lseg, (loff_t)req->wb_index << PAGE_CACHE_SHIFT); i = select_bucket_index(fl, j); list = &fl->commit_buckets[i]; if (list_empty(list)) { /* Non-empty buckets hold a reference on the lseg */ get_lseg(lseg); } return list; } static u32 calc_ds_index_from_commit(struct pnfs_layout_segment *lseg, u32 i) { struct nfs4_filelayout_segment *flseg = FILELAYOUT_LSEG(lseg); if (flseg->stripe_type == STRIPE_SPARSE) return i; else return nfs4_fl_calc_ds_index(lseg, i); } static struct nfs_fh * select_ds_fh_from_commit(struct pnfs_layout_segment *lseg, u32 i) { struct nfs4_filelayout_segment *flseg = FILELAYOUT_LSEG(lseg); if (flseg->stripe_type == STRIPE_SPARSE) { if (flseg->num_fh == 1) i = 0; else if (flseg->num_fh == 0) /* Use the MDS OPEN fh set in nfs_read_rpcsetup */ return NULL; } return flseg->fh_array[i]; } static int filelayout_initiate_commit(struct nfs_write_data *data, int how) { struct pnfs_layout_segment *lseg = data->lseg; struct nfs4_pnfs_ds *ds; u32 idx; struct nfs_fh *fh; idx = calc_ds_index_from_commit(lseg, data->ds_commit_index); ds = nfs4_fl_prepare_ds(lseg, idx); if (!ds) { printk(KERN_ERR "%s: prepare_ds failed, use MDS ", __func__); set_bit(lo_fail_bit(IOMODE_RW), &lseg->pls_layout->plh_flags); set_bit(lo_fail_bit(IOMODE_READ), &lseg->pls_layout->plh_flags); prepare_to_resend_writes(data); data->mds_ops->rpc_release(data); return -EAGAIN; } dprintk("%s ino %lu, how %d ", __func__, data->inode->i_ino, how); data->write_done_cb = filelayout_commit_done_cb; data->ds_clp = ds->ds_clp; fh = select_ds_fh_from_commit(lseg, data->ds_commit_index); if (fh) data->args.fh = fh; return nfs_initiate_commit(data, ds->ds_clp->cl_rpcclient, &filelayout_commit_call_ops, how); } /* * This is only useful while we are using whole file layouts. */ static struct pnfs_layout_segment *find_only_write_lseg(struct inode *inode) { struct pnfs_layout_segment *lseg, *rv = NULL; spin_lock(&inode->i_lock); list_for_each_entry(lseg, &NFS_I(inode)->layout->plh_segs, pls_list) if (lseg->pls_range.iomode == IOMODE_RW) rv = get_lseg(lseg); spin_unlock(&inode->i_lock); return rv; } static int alloc_ds_commits(struct inode *inode, struct list_head *list) { struct pnfs_layout_segment *lseg; struct nfs4_filelayout_segment *fl; struct nfs_write_data *data; int i, j; /* Won't need this when non-whole file layout segments are supported * instead we will use a pnfs_layout_hdr structure */ lseg = find_only_write_lseg(inode); if (!lseg) return 0; fl = FILELAYOUT_LSEG(lseg); for (i = 0; i < fl->number_of_buckets; i++) { if (list_empty(&fl->commit_buckets[i])) continue; data = nfs_commitdata_alloc(); if (!data) goto out_bad; data->ds_commit_index = i; data->lseg = lseg; list_add(&data->pages, list); } put_lseg(lseg); return 0; out_bad: for (j = i; j < fl->number_of_buckets; j++) { if (list_empty(&fl->commit_buckets[i])) continue; nfs_retry_commit(&fl->commit_buckets[i], lseg); put_lseg(lseg); /* associated with emptying bucket */ } put_lseg(lseg); /* Caller will clean up entries put on list */ return -ENOMEM; } /* This follows nfs_commit_list pretty closely */ static int filelayout_commit_pagelist(struct inode *inode, struct list_head *mds_pages, int how) { struct nfs_write_data *data, *tmp; LIST_HEAD(list); if (!list_empty(mds_pages)) { data = nfs_commitdata_alloc(); if (!data) goto out_bad; data->lseg = NULL; list_add(&data->pages, &list); } if (alloc_ds_commits(inode, &list)) goto out_bad; list_for_each_entry_safe(data, tmp, &list, pages) { list_del_init(&data->pages); atomic_inc(&NFS_I(inode)->commits_outstanding); if (!data->lseg) { nfs_init_commit(data, mds_pages, NULL); nfs_initiate_commit(data, NFS_CLIENT(inode), data->mds_ops, how); } else { nfs_init_commit(data, &FILELAYOUT_LSEG(data->lseg)->commit_buckets[data->ds_commit_index], data->lseg); filelayout_initiate_commit(data, how); } } return 0; out_bad: list_for_each_entry_safe(data, tmp, &list, pages) { nfs_retry_commit(&data->pages, data->lseg); list_del_init(&data->pages); nfs_commit_free(data); } nfs_retry_commit(mds_pages, NULL); nfs_commit_clear_lock(NFS_I(inode)); return -ENOMEM; } |
1775bc342 NFSv4.1: purge de... |
899 900 901 902 903 |
static void filelayout_free_deveiceid_node(struct nfs4_deviceid_node *d) { nfs4_fl_free_deviceid(container_of(d, struct nfs4_file_layout_dsaddr, id_node)); } |
7ab672ce3 NFSv4.1: pnfs: fi... |
904 |
static struct pnfs_layoutdriver_type filelayout_type = { |
ea8eecdd1 NFSv4.1 move devi... |
905 906 907 908 909 |
.id = LAYOUT_NFSV4_1_FILES, .name = "LAYOUT_NFSV4_1_FILES", .owner = THIS_MODULE, .alloc_lseg = filelayout_alloc_lseg, .free_lseg = filelayout_free_lseg, |
1751c3638 NFS: Cleanup of t... |
910 911 |
.pg_read_ops = &filelayout_pg_read_ops, .pg_write_ops = &filelayout_pg_write_ops, |
e0c2b3801 NFSv4.1: filelayo... |
912 913 914 |
.mark_pnfs_commit = filelayout_mark_pnfs_commit, .choose_commit_list = filelayout_choose_commit_list, .commit_pagelist = filelayout_commit_pagelist, |
dc70d7b31 NFSv4.1: filelayo... |
915 |
.read_pagelist = filelayout_read_pagelist, |
0382b7440 NFSv4.1: implemen... |
916 |
.write_pagelist = filelayout_write_pagelist, |
1775bc342 NFSv4.1: purge de... |
917 |
.free_deviceid_node = filelayout_free_deveiceid_node, |
7ab672ce3 NFSv4.1: pnfs: fi... |
918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 |
}; static int __init nfs4filelayout_init(void) { printk(KERN_INFO "%s: NFSv4 File Layout Driver Registering... ", __func__); return pnfs_register_layoutdriver(&filelayout_type); } static void __exit nfs4filelayout_exit(void) { printk(KERN_INFO "%s: NFSv4 File Layout Driver Unregistering... ", __func__); pnfs_unregister_layoutdriver(&filelayout_type); } |
f85ef69ce pnfs: simplify pn... |
935 |
MODULE_ALIAS("nfs-layouttype4-1"); |
7ab672ce3 NFSv4.1: pnfs: fi... |
936 937 |
module_init(nfs4filelayout_init); module_exit(nfs4filelayout_exit); |