| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <202202170410.fCsMg7GT-lkp@intel.com>
Date: Thu, 17 Feb 2022 04:44:29 +0800
From: kernel test robot <lkp@...el.com>
To: David Howells <dhowells@...hat.com>
Cc: kbuild-all@...ts.01.org,
GNU/Weeb Mailing List <gwml@...r.gnuweeb.org>,
linux-kernel@...r.kernel.org
Subject: [ammarfaizi2-block:dhowells/linux-fs/netfs-maple 29/42]
fs/netfs/buffered_write.c:610:58: error: 'struct netfs_i_context' has no
member named 'cache'
tree: https://github.com/ammarfaizi2/linux-block dhowells/linux-fs/netfs-maple
head: 5cb7f190822d09757b30cd9539e57eef72552d1f
commit: 261cd621bd0477d43de460dea6c7bf7fa81824be [29/42] netfs: Implement buffered writes through netfs_file_write_iter()
config: powerpc64-randconfig-m031-20220216 (https://download.01.org/0day-ci/archive/20220217/202202170410.fCsMg7GT-lkp@intel.com/config)
compiler: powerpc64-linux-gcc (GCC) 11.2.0
reproduce (this is a W=1 build):
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# https://github.com/ammarfaizi2/linux-block/commit/261cd621bd0477d43de460dea6c7bf7fa81824be
git remote add ammarfaizi2-block https://github.com/ammarfaizi2/linux-block
git fetch --no-tags ammarfaizi2-block dhowells/linux-fs/netfs-maple
git checkout 261cd621bd0477d43de460dea6c7bf7fa81824be
# save the config file to linux build tree
mkdir build_dir
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-11.2.0 make.cross O=build_dir ARCH=powerpc SHELL=/bin/bash fs/
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@...el.com>
All errors (new ones prefixed by >>):
fs/netfs/buffered_write.c: In function 'netfs_perform_write':
>> fs/netfs/buffered_write.c:610:58: error: 'struct netfs_i_context' has no member named 'cache'
610 | fscache_update_cookie(ctx->cache, NULL, &pos);
| ^~
vim +610 fs/netfs/buffered_write.c
453
454 /*
455 * Write data into a prereserved region of the pagecache attached to a netfs
456 * inode.
457 */
458 static ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter)
459 {
460 struct netfs_dirty_region *spare_region = NULL;
461 struct file *file = iocb->ki_filp;
462 struct netfs_i_context *ctx = netfs_i_context(file_inode(file));
463 struct folio *folio;
464 enum netfs_handle_nonuptodate nupt;
465 ssize_t written = 0, ret;
466 loff_t i_size, pos = iocb->ki_pos;
467 bool always_fill = false;
468 bool locked = false;
469
470 MA_STATE(mas, &ctx->dirty_regions, pos / PAGE_SIZE,
471 (pos + iov_iter_count(iter) - 1) / PAGE_SIZE);
472
473 ret = ctx->ops->validate_for_write(file_inode(file), file);
474 if (ret < 0)
475 return ret;
476
477 do {
478 size_t plen;
479 size_t offset; /* Offset into pagecache folio */
480 size_t bytes; /* Bytes to write to folio */
481 size_t copied; /* Bytes copied from user */
482
483 folio = netfs_grab_folio_for_write(file->f_mapping,
484 pos / PAGE_SIZE,
485 iov_iter_count(iter));
486 if (!folio) {
487 ret = -ENOMEM;
488 goto out;
489 }
490
491 plen = folio_size(folio);
492 offset = pos - folio_file_pos(folio);
493 bytes = min_t(size_t, plen - offset, iov_iter_count(iter));
494 locked = true;
495
496 if (!folio_test_uptodate(folio)) {
497 folio_unlock(folio); /* Avoid deadlocking fault-in */
498 locked = false;
499 }
500
501 /* Bring in the user page that we will copy from _first_.
502 * Otherwise there's a nasty deadlock on copying from the same
503 * page as we're writing to, without it being marked
504 * up-to-date.
505 *
506 * Not only is this an optimisation, but it is also required to
507 * check that the address is actually valid, when atomic
508 * usercopies are used, below.
509 */
510 if (unlikely(fault_in_iov_iter_readable(iter, bytes))) {
511 ret = -EFAULT;
512 goto error_folio;
513 }
514
515 if (fatal_signal_pending(current)) {
516 ret = -EINTR;
517 goto error_folio;
518 }
519
520 if (!locked) {
521 ret = folio_lock_killable(folio);
522 if (ret < 0)
523 goto error_folio;
524 }
525
526 redo_prefetch:
527 /* See if we need to prefetch the area we're going to modify.
528 * We need to do this before we get a lock on the folio in case
529 * there's more than one writer competing for the same cache
530 * block.
531 */
532 nupt = netfs_handle_nonuptodate_folio(ctx, file, folio,
533 offset, bytes, always_fill);
534 _debug("nupt %u", nupt);
535 switch (nupt) {
536 case NETFS_JUST_PREFETCH:
537 ret = netfs_prefetch_for_write(file, folio, bytes);
538 if (ret < 0) {
539 _debug("prefetch = %zd", ret);
540 goto error_folio;
541 }
542 nupt = NETFS_FOLIO_IS_UPTODATE;
543 fallthrough;
544 case NETFS_FOLIO_IS_UPTODATE:
545 break;
546 case NETFS_MODIFY_AND_CLEAR:
547 zero_user_segment(&folio->page, 0, offset);
548 break;
549 case NETFS_WHOLE_FOLIO_MODIFY:
550 break;
551 }
552
553 /* Preallocate the space we need in the dirty region list. */
554 ret = mas_expected_entries(&mas, 1);
555 if (ret < 0)
556 goto error_folio;
557
558 if (!spare_region) {
559 spare_region = netfs_alloc_dirty_region();
560 if (IS_ERR(spare_region)) {
561 ret = PTR_ERR(spare_region);
562 spare_region = NULL;
563 goto error_folio;
564 }
565 }
566
567 if (mapping_writably_mapped(folio_file_mapping(folio)))
568 flush_dcache_folio(folio);
569 copied = copy_folio_from_iter_atomic(folio, offset, bytes, iter);
570 flush_dcache_folio(folio);
571
572 /* Deal with a (partially) failed copy */
573 if (!folio_test_uptodate(folio)) {
574 if (copied == 0) {
575 ret = -EFAULT;
576 goto error_folio;
577 }
578 if (copied < bytes) {
579 iov_iter_revert(iter, copied);
580 always_fill = true;
581 goto redo_prefetch;
582 }
583 switch (nupt) {
584 case NETFS_JUST_PREFETCH:
585 case NETFS_FOLIO_IS_UPTODATE:
586 /* We have the folio locked, so it really ought
587 * to be uptodate.
588 */
589 WARN(true, "Locked folio %lx became non-uptodate\n",
590 folio_index(folio));
591 ret = -EIO;
592 goto error_folio;
593 case NETFS_MODIFY_AND_CLEAR:
594 zero_user_segment(&folio->page, offset + copied, plen);
595 fallthrough;
596 case NETFS_WHOLE_FOLIO_MODIFY:
597 folio_mark_uptodate(folio);
598 break;
599 }
600 }
601
602 /* Update the inode size if we moved the EOF marker */
603 pos += copied;
604 i_size = i_size_read(file_inode(file));
605 if (pos > i_size) {
606 if (ctx->ops->update_i_size) {
607 ctx->ops->update_i_size(file, pos);
608 } else {
609 i_size_write(file_inode(file), pos);
> 610 fscache_update_cookie(ctx->cache, NULL, &pos);
611 }
612 }
613
614 netfs_commit_folio(ctx, file, &spare_region, &mas,
615 folio, offset, copied);
616
617 folio_mark_dirty(folio);
618 folio_unlock(folio);
619 folio_put(folio);
620 folio = NULL;
621
622 cond_resched();
623
624 written += copied;
625
626 balance_dirty_pages_ratelimited(file->f_mapping);
627 } while (iov_iter_count(iter));
628
629 out:
630 if (likely(written)) {
631 netfs_commit_region(ctx, &mas, iocb->ki_pos, written);
632
633 iocb->ki_pos += written;
634
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
Powered by blists - more mailing lists