objlayout.c 20 KB
Newer Older
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
/*
 *  pNFS Objects layout driver high level definitions
 *
 *  Copyright (C) 2007 Panasas Inc. [year of first publication]
 *  All rights reserved.
 *
 *  Benny Halevy <bhalevy@panasas.com>
 *  Boaz Harrosh <bharrosh@panasas.com>
 *
 *  This program is free software; you can redistribute it and/or modify
 *  it under the terms of the GNU General Public License version 2
 *  See the file COPYING included with this distribution for more details.
 *
 *  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. Neither the name of the Panasas company nor the names of its
 *     contributors may be used to endorse or promote products derived
 *     from this software without specific prior written permission.
 *
 *  THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS 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 REGENTS OR 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.
 */

40 41 42
#include <linux/kmod.h>
#include <linux/moduleparam.h>
#include <linux/ratelimit.h>
43 44 45 46
#include <scsi/osd_initiator.h>
#include "objlayout.h"

#define NFSDBG_FACILITY         NFSDBG_PNFS_LD
47 48 49 50 51 52 53 54 55
/*
 * Create a objlayout layout structure for the given inode and return it.
 */
struct pnfs_layout_hdr *
objlayout_alloc_layout_hdr(struct inode *inode, gfp_t gfp_flags)
{
	struct objlayout *objlay;

	objlay = kzalloc(sizeof(struct objlayout), gfp_flags);
56 57 58 59
	if (!objlay)
		return NULL;
	spin_lock_init(&objlay->lock);
	INIT_LIST_HEAD(&objlay->err_list);
60 61 62 63 64 65 66 67 68 69 70 71 72 73
	dprintk("%s: Return %p\n", __func__, objlay);
	return &objlay->pnfs_layout;
}

/*
 * Free an objlayout layout structure
 */
void
objlayout_free_layout_hdr(struct pnfs_layout_hdr *lo)
{
	struct objlayout *objlay = OBJLAYOUT(lo);

	dprintk("%s: objlay %p\n", __func__, objlay);

74
	WARN_ON(!list_empty(&objlay->err_list));
75 76 77
	kfree(objlay);
}

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
/*
 * Unmarshall layout and store it in pnfslay.
 */
struct pnfs_layout_segment *
objlayout_alloc_lseg(struct pnfs_layout_hdr *pnfslay,
		     struct nfs4_layoutget_res *lgr,
		     gfp_t gfp_flags)
{
	int status = -ENOMEM;
	struct xdr_stream stream;
	struct xdr_buf buf = {
		.pages =  lgr->layoutp->pages,
		.page_len =  lgr->layoutp->len,
		.buflen =  lgr->layoutp->len,
		.len = lgr->layoutp->len,
	};
	struct page *scratch;
	struct pnfs_layout_segment *lseg;

	dprintk("%s: Begin pnfslay %p\n", __func__, pnfslay);

	scratch = alloc_page(gfp_flags);
	if (!scratch)
		goto err_nofree;

	xdr_init_decode(&stream, &buf, NULL);
	xdr_set_scratch_buffer(&stream, page_address(scratch), PAGE_SIZE);

	status = objio_alloc_lseg(&lseg, pnfslay, &lgr->range, &stream, gfp_flags);
	if (unlikely(status)) {
		dprintk("%s: objio_alloc_lseg Return err %d\n", __func__,
			status);
		goto err;
	}

	__free_page(scratch);

	dprintk("%s: Return %p\n", __func__, lseg);
	return lseg;

err:
	__free_page(scratch);
err_nofree:
	dprintk("%s: Err Return=>%d\n", __func__, status);
	return ERR_PTR(status);
}

/*
 * Free a layout segement
 */
void
objlayout_free_lseg(struct pnfs_layout_segment *lseg)
{
	dprintk("%s: freeing layout segment %p\n", __func__, lseg);

	if (unlikely(!lseg))
		return;

	objio_free_lseg(lseg);
}

139 140 141 142 143 144 145 146 147 148 149 150
/*
 * I/O Operations
 */
static inline u64
end_offset(u64 start, u64 len)
{
	u64 end;

	end = start + len;
	return end >= start ? end : NFS4_MAX_UINT64;
}

151
static void _fix_verify_io_params(struct pnfs_layout_segment *lseg,
152 153
			   struct page ***p_pages, unsigned *p_pgbase,
			   u64 offset, unsigned long count)
154 155 156 157 158 159 160
{
	u64 lseg_end_offset;

	BUG_ON(offset < lseg->pls_range.offset);
	lseg_end_offset = end_offset(lseg->pls_range.offset,
				     lseg->pls_range.length);
	BUG_ON(offset >= lseg_end_offset);
161
	WARN_ON(offset + count > lseg_end_offset);
162

163 164 165 166
	if (*p_pgbase > PAGE_SIZE) {
		dprintk("%s: pgbase(0x%x) > PAGE_SIZE\n", __func__, *p_pgbase);
		*p_pages += *p_pgbase >> PAGE_SHIFT;
		*p_pgbase &= ~PAGE_MASK;
167 168 169 170 171 172 173
	}
}

/*
 * I/O done common code
 */
static void
174
objlayout_iodone(struct objlayout_io_res *oir)
175
{
176 177
	if (likely(oir->status >= 0)) {
		objio_free_result(oir);
178
	} else {
179
		struct objlayout *objlay = oir->objlay;
180 181

		spin_lock(&objlay->lock);
182
		objlay->delta_space_valid = OBJ_DSU_INVALID;
183
		list_add(&objlay->err_list, &oir->err_list);
184 185 186 187 188 189 190 191 192 193 194
		spin_unlock(&objlay->lock);
	}
}

/*
 * objlayout_io_set_result - Set an osd_error code on a specific osd comp.
 *
 * The @index component IO failed (error returned from target). Register
 * the error for later reporting at layout-return.
 */
void
195
objlayout_io_set_result(struct objlayout_io_res *oir, unsigned index,
196 197 198
			struct pnfs_osd_objid *pooid, int osd_error,
			u64 offset, u64 length, bool is_write)
{
199
	struct pnfs_osd_ioerr *ioerr = &oir->ioerrs[index];
200

201
	BUG_ON(index >= oir->num_comps);
202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222
	if (osd_error) {
		ioerr->oer_component = *pooid;
		ioerr->oer_comp_offset = offset;
		ioerr->oer_comp_length = length;
		ioerr->oer_iswrite = is_write;
		ioerr->oer_errno = osd_error;

		dprintk("%s: err[%d]: errno=%d is_write=%d dev(%llx:%llx) "
			"par=0x%llx obj=0x%llx offset=0x%llx length=0x%llx\n",
			__func__, index, ioerr->oer_errno,
			ioerr->oer_iswrite,
			_DEVID_LO(&ioerr->oer_component.oid_device_id),
			_DEVID_HI(&ioerr->oer_component.oid_device_id),
			ioerr->oer_component.oid_partition_id,
			ioerr->oer_component.oid_object_id,
			ioerr->oer_comp_offset,
			ioerr->oer_comp_length);
	} else {
		/* User need not call if no error is reported */
		ioerr->oer_errno = 0;
	}
223 224 225 226 227 228 229 230 231
}

/* Function scheduled on rpc workqueue to call ->nfs_readlist_complete().
 * This is because the osd completion is called with ints-off from
 * the block layer
 */
static void _rpc_read_complete(struct work_struct *work)
{
	struct rpc_task *task;
232
	struct nfs_pgio_data *rdata;
233 234 235

	dprintk("%s enter\n", __func__);
	task = container_of(work, struct rpc_task, u.tk_work);
236
	rdata = container_of(task, struct nfs_pgio_data, task);
237 238 239 240 241

	pnfs_ld_read_done(rdata);
}

void
242
objlayout_read_done(struct objlayout_io_res *oir, ssize_t status, bool sync)
243
{
244
	struct nfs_pgio_data *rdata = oir->rpcdata;
245

246
	oir->status = rdata->task.tk_status = status;
247
	if (status >= 0)
248
		rdata->res.count = status;
249
	else
250
		rdata->header->pnfs_error = status;
251 252
	objlayout_iodone(oir);
	/* must not use oir after this point */
253

254 255 256
	dprintk("%s: Return status=%zd eof=%d sync=%d\n", __func__,
		status, rdata->res.eof, sync);

257 258 259 260 261 262 263 264 265 266 267 268
	if (sync)
		pnfs_ld_read_done(rdata);
	else {
		INIT_WORK(&rdata->task.u.tk_work, _rpc_read_complete);
		schedule_work(&rdata->task.u.tk_work);
	}
}

/*
 * Perform sync or async reads.
 */
enum pnfs_try_status
269
objlayout_read_pagelist(struct nfs_pgio_data *rdata)
270
{
271 272
	struct nfs_pgio_header *hdr = rdata->header;
	struct inode *inode = hdr->inode;
273 274
	loff_t offset = rdata->args.offset;
	size_t count = rdata->args.count;
275
	int err;
276 277
	loff_t eof;

278
	eof = i_size_read(inode);
279 280
	if (unlikely(offset + count > eof)) {
		if (offset >= eof) {
281
			err = 0;
282 283
			rdata->res.count = 0;
			rdata->res.eof = 1;
284
			/*FIXME: do we need to call pnfs_ld_read_done() */
285 286 287 288 289
			goto out;
		}
		count = eof - offset;
	}

290
	rdata->res.eof = (offset + count) >= eof;
291
	_fix_verify_io_params(hdr->lseg, &rdata->args.pages,
292 293
			      &rdata->args.pgbase,
			      rdata->args.offset, rdata->args.count);
294

295
	dprintk("%s: inode(%lx) offset 0x%llx count 0x%Zx eof=%d\n",
296
		__func__, inode->i_ino, offset, count, rdata->res.eof);
297

298
	err = objio_read_pagelist(rdata);
299
 out:
300
	if (unlikely(err)) {
301
		hdr->pnfs_error = err;
302 303 304
		dprintk("%s: Returned Error %d\n", __func__, err);
		return PNFS_NOT_ATTEMPTED;
	}
305 306 307 308 309 310 311 312 313 314
	return PNFS_ATTEMPTED;
}

/* Function scheduled on rpc workqueue to call ->nfs_writelist_complete().
 * This is because the osd completion is called with ints-off from
 * the block layer
 */
static void _rpc_write_complete(struct work_struct *work)
{
	struct rpc_task *task;
315
	struct nfs_pgio_data *wdata;
316 317 318

	dprintk("%s enter\n", __func__);
	task = container_of(work, struct rpc_task, u.tk_work);
319
	wdata = container_of(task, struct nfs_pgio_data, task);
320 321 322 323 324

	pnfs_ld_write_done(wdata);
}

void
325
objlayout_write_done(struct objlayout_io_res *oir, ssize_t status, bool sync)
326
{
327
	struct nfs_pgio_data *wdata = oir->rpcdata;
328

329
	oir->status = wdata->task.tk_status = status;
330 331
	if (status >= 0) {
		wdata->res.count = status;
332
		wdata->writeverf.committed = oir->committed;
333
	} else {
334
		wdata->header->pnfs_error = status;
335
	}
336
	objlayout_iodone(oir);
337 338 339
	/* must not use oir after this point */

	dprintk("%s: Return status %zd committed %d sync=%d\n", __func__,
340
		status, wdata->writeverf.committed, sync);
341 342 343 344 345 346 347 348 349 350 351 352 353

	if (sync)
		pnfs_ld_write_done(wdata);
	else {
		INIT_WORK(&wdata->task.u.tk_work, _rpc_write_complete);
		schedule_work(&wdata->task.u.tk_work);
	}
}

/*
 * Perform sync or async writes.
 */
enum pnfs_try_status
354
objlayout_write_pagelist(struct nfs_pgio_data *wdata,
355 356
			 int how)
{
357
	struct nfs_pgio_header *hdr = wdata->header;
358
	int err;
359

360
	_fix_verify_io_params(hdr->lseg, &wdata->args.pages,
361 362
			      &wdata->args.pgbase,
			      wdata->args.offset, wdata->args.count);
363

364
	err = objio_write_pagelist(wdata, how);
365
	if (unlikely(err)) {
366
		hdr->pnfs_error = err;
367 368 369
		dprintk("%s: Returned Error %d\n", __func__, err);
		return PNFS_NOT_ATTEMPTED;
	}
370 371 372
	return PNFS_ATTEMPTED;
}

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
void
objlayout_encode_layoutcommit(struct pnfs_layout_hdr *pnfslay,
			      struct xdr_stream *xdr,
			      const struct nfs4_layoutcommit_args *args)
{
	struct objlayout *objlay = OBJLAYOUT(pnfslay);
	struct pnfs_osd_layoutupdate lou;
	__be32 *start;

	dprintk("%s: Begin\n", __func__);

	spin_lock(&objlay->lock);
	lou.dsu_valid = (objlay->delta_space_valid == OBJ_DSU_VALID);
	lou.dsu_delta = objlay->delta_space_used;
	objlay->delta_space_used = 0;
	objlay->delta_space_valid = OBJ_DSU_INIT;
	lou.olu_ioerr_flag = !list_empty(&objlay->err_list);
	spin_unlock(&objlay->lock);

	start = xdr_reserve_space(xdr, 4);

	BUG_ON(pnfs_osd_xdr_encode_layoutupdate(xdr, &lou));

	*start = cpu_to_be32((xdr->p - start - 1) * 4);

	dprintk("%s: Return delta_space_used %lld err %d\n", __func__,
		lou.dsu_delta, lou.olu_ioerr_flag);
}

402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475
static int
err_prio(u32 oer_errno)
{
	switch (oer_errno) {
	case 0:
		return 0;

	case PNFS_OSD_ERR_RESOURCE:
		return OSD_ERR_PRI_RESOURCE;
	case PNFS_OSD_ERR_BAD_CRED:
		return OSD_ERR_PRI_BAD_CRED;
	case PNFS_OSD_ERR_NO_ACCESS:
		return OSD_ERR_PRI_NO_ACCESS;
	case PNFS_OSD_ERR_UNREACHABLE:
		return OSD_ERR_PRI_UNREACHABLE;
	case PNFS_OSD_ERR_NOT_FOUND:
		return OSD_ERR_PRI_NOT_FOUND;
	case PNFS_OSD_ERR_NO_SPACE:
		return OSD_ERR_PRI_NO_SPACE;
	default:
		WARN_ON(1);
		/* fallthrough */
	case PNFS_OSD_ERR_EIO:
		return OSD_ERR_PRI_EIO;
	}
}

static void
merge_ioerr(struct pnfs_osd_ioerr *dest_err,
	    const struct pnfs_osd_ioerr *src_err)
{
	u64 dest_end, src_end;

	if (!dest_err->oer_errno) {
		*dest_err = *src_err;
		/* accumulated device must be blank */
		memset(&dest_err->oer_component.oid_device_id, 0,
			sizeof(dest_err->oer_component.oid_device_id));

		return;
	}

	if (dest_err->oer_component.oid_partition_id !=
				src_err->oer_component.oid_partition_id)
		dest_err->oer_component.oid_partition_id = 0;

	if (dest_err->oer_component.oid_object_id !=
				src_err->oer_component.oid_object_id)
		dest_err->oer_component.oid_object_id = 0;

	if (dest_err->oer_comp_offset > src_err->oer_comp_offset)
		dest_err->oer_comp_offset = src_err->oer_comp_offset;

	dest_end = end_offset(dest_err->oer_comp_offset,
			      dest_err->oer_comp_length);
	src_end =  end_offset(src_err->oer_comp_offset,
			      src_err->oer_comp_length);
	if (dest_end < src_end)
		dest_end = src_end;

	dest_err->oer_comp_length = dest_end - dest_err->oer_comp_offset;

	if ((src_err->oer_iswrite == dest_err->oer_iswrite) &&
	    (err_prio(src_err->oer_errno) > err_prio(dest_err->oer_errno))) {
			dest_err->oer_errno = src_err->oer_errno;
	} else if (src_err->oer_iswrite) {
		dest_err->oer_iswrite = true;
		dest_err->oer_errno = src_err->oer_errno;
	}
}

static void
encode_accumulated_error(struct objlayout *objlay, __be32 *p)
{
476
	struct objlayout_io_res *oir, *tmp;
477 478
	struct pnfs_osd_ioerr accumulated_err = {.oer_errno = 0};

479
	list_for_each_entry_safe(oir, tmp, &objlay->err_list, err_list) {
480 481
		unsigned i;

482 483
		for (i = 0; i < oir->num_comps; i++) {
			struct pnfs_osd_ioerr *ioerr = &oir->ioerrs[i];
484 485 486 487

			if (!ioerr->oer_errno)
				continue;

488 489 490
			printk(KERN_ERR "NFS: %s: err[%d]: errno=%d "
				"is_write=%d dev(%llx:%llx) par=0x%llx "
				"obj=0x%llx offset=0x%llx length=0x%llx\n",
491 492 493 494 495 496 497 498 499 500 501
				__func__, i, ioerr->oer_errno,
				ioerr->oer_iswrite,
				_DEVID_LO(&ioerr->oer_component.oid_device_id),
				_DEVID_HI(&ioerr->oer_component.oid_device_id),
				ioerr->oer_component.oid_partition_id,
				ioerr->oer_component.oid_object_id,
				ioerr->oer_comp_offset,
				ioerr->oer_comp_length);

			merge_ioerr(&accumulated_err, ioerr);
		}
502 503
		list_del(&oir->err_list);
		objio_free_result(oir);
504 505 506 507 508 509 510 511 512 513 514
	}

	pnfs_osd_xdr_encode_ioerr(p, &accumulated_err);
}

void
objlayout_encode_layoutreturn(struct pnfs_layout_hdr *pnfslay,
			      struct xdr_stream *xdr,
			      const struct nfs4_layoutreturn_args *args)
{
	struct objlayout *objlay = OBJLAYOUT(pnfslay);
515
	struct objlayout_io_res *oir, *tmp;
516 517 518 519 520 521 522 523
	__be32 *start;

	dprintk("%s: Begin\n", __func__);
	start = xdr_reserve_space(xdr, 4);
	BUG_ON(!start);

	spin_lock(&objlay->lock);

524
	list_for_each_entry_safe(oir, tmp, &objlay->err_list, err_list) {
525 526 527 528
		__be32 *last_xdr = NULL, *p;
		unsigned i;
		int res = 0;

529 530
		for (i = 0; i < oir->num_comps; i++) {
			struct pnfs_osd_ioerr *ioerr = &oir->ioerrs[i];
531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553

			if (!ioerr->oer_errno)
				continue;

			dprintk("%s: err[%d]: errno=%d is_write=%d "
				"dev(%llx:%llx) par=0x%llx obj=0x%llx "
				"offset=0x%llx length=0x%llx\n",
				__func__, i, ioerr->oer_errno,
				ioerr->oer_iswrite,
				_DEVID_LO(&ioerr->oer_component.oid_device_id),
				_DEVID_HI(&ioerr->oer_component.oid_device_id),
				ioerr->oer_component.oid_partition_id,
				ioerr->oer_component.oid_object_id,
				ioerr->oer_comp_offset,
				ioerr->oer_comp_length);

			p = pnfs_osd_xdr_ioerr_reserve_space(xdr);
			if (unlikely(!p)) {
				res = -E2BIG;
				break; /* accumulated_error */
			}

			last_xdr = p;
554
			pnfs_osd_xdr_encode_ioerr(p, &oir->ioerrs[i]);
555 556 557 558 559 560 561 562 563 564 565 566 567 568 569
		}

		/* TODO: use xdr_write_pages */
		if (unlikely(res)) {
			/* no space for even one error descriptor */
			BUG_ON(!last_xdr);

			/* we've encountered a situation with lots and lots of
			 * errors and no space to encode them all. Use the last
			 * available slot to report the union of all the
			 * remaining errors.
			 */
			encode_accumulated_error(objlay, last_xdr);
			goto loop_done;
		}
570 571
		list_del(&oir->err_list);
		objio_free_result(oir);
572 573 574 575 576 577 578 579 580
	}
loop_done:
	spin_unlock(&objlay->lock);

	*start = cpu_to_be32((xdr->p - start - 1) * 4);
	dprintk("%s: Return\n", __func__);
}


581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615
/*
 * Get Device Info API for io engines
 */
struct objlayout_deviceinfo {
	struct page *page;
	struct pnfs_osd_deviceaddr da; /* This must be last */
};

/* Initialize and call nfs_getdeviceinfo, then decode and return a
 * "struct pnfs_osd_deviceaddr *" Eventually objlayout_put_deviceinfo()
 * should be called.
 */
int objlayout_get_deviceinfo(struct pnfs_layout_hdr *pnfslay,
	struct nfs4_deviceid *d_id, struct pnfs_osd_deviceaddr **deviceaddr,
	gfp_t gfp_flags)
{
	struct objlayout_deviceinfo *odi;
	struct pnfs_device pd;
	struct page *page, **pages;
	u32 *p;
	int err;

	page = alloc_page(gfp_flags);
	if (!page)
		return -ENOMEM;

	pages = &page;
	pd.pages = pages;

	memcpy(&pd.dev_id, d_id, sizeof(*d_id));
	pd.layout_type = LAYOUT_OSD2_OBJECTS;
	pd.pages = &page;
	pd.pgbase = 0;
	pd.pglen = PAGE_SIZE;
	pd.mincount = 0;
616
	pd.maxcount = PAGE_SIZE;
617

618 619
	err = nfs4_proc_getdeviceinfo(NFS_SERVER(pnfslay->plh_inode), &pd,
			pnfslay->plh_lc_cred);
620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648
	dprintk("%s nfs_getdeviceinfo returned %d\n", __func__, err);
	if (err)
		goto err_out;

	p = page_address(page);
	odi = kzalloc(sizeof(*odi), gfp_flags);
	if (!odi) {
		err = -ENOMEM;
		goto err_out;
	}
	pnfs_osd_xdr_decode_deviceaddr(&odi->da, p);
	odi->page = page;
	*deviceaddr = &odi->da;
	return 0;

err_out:
	__free_page(page);
	return err;
}

void objlayout_put_deviceinfo(struct pnfs_osd_deviceaddr *deviceaddr)
{
	struct objlayout_deviceinfo *odi = container_of(deviceaddr,
						struct objlayout_deviceinfo,
						da);

	__free_page(odi->page);
	kfree(odi);
}
649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779

enum {
	OBJLAYOUT_MAX_URI_LEN = 256, OBJLAYOUT_MAX_OSDNAME_LEN = 64,
	OBJLAYOUT_MAX_SYSID_HEX_LEN = OSD_SYSTEMID_LEN * 2 + 1,
	OSD_LOGIN_UPCALL_PATHLEN  = 256
};

static char osd_login_prog[OSD_LOGIN_UPCALL_PATHLEN] = "/sbin/osd_login";

module_param_string(osd_login_prog, osd_login_prog, sizeof(osd_login_prog),
		    0600);
MODULE_PARM_DESC(osd_login_prog, "Path to the osd_login upcall program");

struct __auto_login {
	char uri[OBJLAYOUT_MAX_URI_LEN];
	char osdname[OBJLAYOUT_MAX_OSDNAME_LEN];
	char systemid_hex[OBJLAYOUT_MAX_SYSID_HEX_LEN];
};

static int __objlayout_upcall(struct __auto_login *login)
{
	static char *envp[] = { "HOME=/",
		"TERM=linux",
		"PATH=/sbin:/usr/sbin:/bin:/usr/bin",
		NULL
	};
	char *argv[8];
	int ret;

	if (unlikely(!osd_login_prog[0])) {
		dprintk("%s: osd_login_prog is disabled\n", __func__);
		return -EACCES;
	}

	dprintk("%s uri: %s\n", __func__, login->uri);
	dprintk("%s osdname %s\n", __func__, login->osdname);
	dprintk("%s systemid_hex %s\n", __func__, login->systemid_hex);

	argv[0] = (char *)osd_login_prog;
	argv[1] = "-u";
	argv[2] = login->uri;
	argv[3] = "-o";
	argv[4] = login->osdname;
	argv[5] = "-s";
	argv[6] = login->systemid_hex;
	argv[7] = NULL;

	ret = call_usermodehelper(argv[0], argv, envp, UMH_WAIT_PROC);
	/*
	 * Disable the upcall mechanism if we're getting an ENOENT or
	 * EACCES error. The admin can re-enable it on the fly by using
	 * sysfs to set the objlayoutdriver.osd_login_prog module parameter once
	 * the problem has been fixed.
	 */
	if (ret == -ENOENT || ret == -EACCES) {
		printk(KERN_ERR "PNFS-OBJ: %s was not found please set "
			"objlayoutdriver.osd_login_prog kernel parameter!\n",
			osd_login_prog);
		osd_login_prog[0] = '\0';
	}
	dprintk("%s %s return value: %d\n", __func__, osd_login_prog, ret);

	return ret;
}

/* Assume dest is all zeros */
static void __copy_nfsS_and_zero_terminate(struct nfs4_string s,
					   char *dest, int max_len,
					   const char *var_name)
{
	if (!s.len)
		return;

	if (s.len >= max_len) {
		pr_warn_ratelimited(
			"objlayout_autologin: %s: s.len(%d) >= max_len(%d)",
			var_name, s.len, max_len);
		s.len = max_len - 1; /* space for null terminator */
	}

	memcpy(dest, s.data, s.len);
}

/* Assume sysid is all zeros */
static void _sysid_2_hex(struct nfs4_string s,
		  char sysid[OBJLAYOUT_MAX_SYSID_HEX_LEN])
{
	int i;
	char *cur;

	if (!s.len)
		return;

	if (s.len != OSD_SYSTEMID_LEN) {
		pr_warn_ratelimited(
		    "objlayout_autologin: systemid_len(%d) != OSD_SYSTEMID_LEN",
		    s.len);
		if (s.len > OSD_SYSTEMID_LEN)
			s.len = OSD_SYSTEMID_LEN;
	}

	cur = sysid;
	for (i = 0; i < s.len; i++)
		cur = hex_byte_pack(cur, s.data[i]);
}

int objlayout_autologin(struct pnfs_osd_deviceaddr *deviceaddr)
{
	int rc;
	struct __auto_login login;

	if (!deviceaddr->oda_targetaddr.ota_netaddr.r_addr.len)
		return -ENODEV;

	memset(&login, 0, sizeof(login));
	__copy_nfsS_and_zero_terminate(
		deviceaddr->oda_targetaddr.ota_netaddr.r_addr,
		login.uri, sizeof(login.uri), "URI");

	__copy_nfsS_and_zero_terminate(
		deviceaddr->oda_osdname,
		login.osdname, sizeof(login.osdname), "OSDNAME");

	_sysid_2_hex(deviceaddr->oda_systemid, login.systemid_hex);

	rc = __objlayout_upcall(&login);
	if (rc > 0) /* script returns positive values */
		rc = -ENODEV;

	return rc;
}