From: kernel test robot <lkp@intel.com>
To: Matthew Wilcox <willy@infradead.org>
Cc: oe-kbuild-all@lists.linux.dev
Subject: [willy-pagecache:set_ptes 9/38] mm/filemap.c:2709: undefined reference to `flush_dcache_folio'
Date: Sat, 8 Jul 2023 10:57:10 +0800 [thread overview]
Message-ID: <202307081040.pxdFGWsP-lkp@intel.com> (raw)
tree: git://git.infradead.org/users/willy/pagecache set_ptes
head: 72c34bd0b2a432506d992bdc1e3a89260461667d
commit: 440dacbb8de713a54c1f2c71e070576dee31424e [9/38] arm: Implement the new page table range API
config: arm-randconfig-r061-20230704 (https://download.01.org/0day-ci/archive/20230708/202307081040.pxdFGWsP-lkp@intel.com/config)
compiler: arm-linux-gnueabi-gcc (GCC) 12.3.0
reproduce: (https://download.01.org/0day-ci/archive/20230708/202307081040.pxdFGWsP-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202307081040.pxdFGWsP-lkp@intel.com/
All errors (new ones prefixed by >>):
arm-linux-gnueabi-ld: mm/filemap.o: in function `filemap_read':
>> mm/filemap.c:2709: undefined reference to `flush_dcache_folio'
arm-linux-gnueabi-ld: mm/filemap.o: in function `filemap_splice_read':
mm/filemap.c:2964: undefined reference to `flush_dcache_folio'
arm-linux-gnueabi-ld: fs/libfs.o: in function `simple_read_folio':
>> fs/libfs.c:789: undefined reference to `flush_dcache_folio'
arm-linux-gnueabi-ld: fs/remap_range.o: in function `vfs_dedupe_file_range_compare':
>> fs/remap_range.c:234: undefined reference to `flush_dcache_folio'
>> arm-linux-gnueabi-ld: fs/remap_range.c:235: undefined reference to `flush_dcache_folio'
arm-linux-gnueabi-ld: fs/netfs/buffered_read.o:fs/netfs/buffered_read.c:84: more undefined references to `flush_dcache_folio' follow
vim +2709 mm/filemap.c
5ccc944dce3df5 Matthew Wilcox (Oracle 2022-06-10 2611)
723ef24b9b379e Kent Overstreet 2020-12-14 2612 /**
87fa0f3eb267ee Christoph Hellwig 2021-02-24 2613 * filemap_read - Read data from the page cache.
87fa0f3eb267ee Christoph Hellwig 2021-02-24 2614 * @iocb: The iocb to read.
87fa0f3eb267ee Christoph Hellwig 2021-02-24 2615 * @iter: Destination for the data.
87fa0f3eb267ee Christoph Hellwig 2021-02-24 2616 * @already_read: Number of bytes already read by the caller.
723ef24b9b379e Kent Overstreet 2020-12-14 2617 *
87fa0f3eb267ee Christoph Hellwig 2021-02-24 2618 * Copies data from the page cache. If the data is not currently present,
7e0a126519b826 Matthew Wilcox (Oracle 2022-04-29 2619) * uses the readahead and read_folio address_space operations to fetch it.
723ef24b9b379e Kent Overstreet 2020-12-14 2620 *
87fa0f3eb267ee Christoph Hellwig 2021-02-24 2621 * Return: Total number of bytes copied, including those already read by
87fa0f3eb267ee Christoph Hellwig 2021-02-24 2622 * the caller. If an error happens before any bytes are copied, returns
87fa0f3eb267ee Christoph Hellwig 2021-02-24 2623 * a negative error number.
723ef24b9b379e Kent Overstreet 2020-12-14 2624 */
87fa0f3eb267ee Christoph Hellwig 2021-02-24 2625 ssize_t filemap_read(struct kiocb *iocb, struct iov_iter *iter,
87fa0f3eb267ee Christoph Hellwig 2021-02-24 2626 ssize_t already_read)
723ef24b9b379e Kent Overstreet 2020-12-14 2627 {
723ef24b9b379e Kent Overstreet 2020-12-14 2628 struct file *filp = iocb->ki_filp;
06c0444290cecf Kent Overstreet 2020-12-14 2629 struct file_ra_state *ra = &filp->f_ra;
723ef24b9b379e Kent Overstreet 2020-12-14 2630 struct address_space *mapping = filp->f_mapping;
723ef24b9b379e Kent Overstreet 2020-12-14 2631 struct inode *inode = mapping->host;
25d6a23e8d2808 Matthew Wilcox (Oracle 2021-12-06 2632) struct folio_batch fbatch;
ff993ba130009b Matthew Wilcox (Oracle 2021-02-24 2633) int i, error = 0;
06c0444290cecf Kent Overstreet 2020-12-14 2634 bool writably_mapped;
06c0444290cecf Kent Overstreet 2020-12-14 2635 loff_t isize, end_offset;
881a8706c13e41 Haibo Li 2023-06-28 2636 loff_t last_pos = ra->prev_pos;
723ef24b9b379e Kent Overstreet 2020-12-14 2637
723ef24b9b379e Kent Overstreet 2020-12-14 2638 if (unlikely(iocb->ki_pos >= inode->i_sb->s_maxbytes))
723ef24b9b379e Kent Overstreet 2020-12-14 2639 return 0;
3644e2d2dda78e Kent Overstreet 2020-12-18 2640 if (unlikely(!iov_iter_count(iter)))
3644e2d2dda78e Kent Overstreet 2020-12-18 2641 return 0;
3644e2d2dda78e Kent Overstreet 2020-12-18 2642
723ef24b9b379e Kent Overstreet 2020-12-14 2643 iov_iter_truncate(iter, inode->i_sb->s_maxbytes);
25d6a23e8d2808 Matthew Wilcox (Oracle 2021-12-06 2644) folio_batch_init(&fbatch);
723ef24b9b379e Kent Overstreet 2020-12-14 2645
06c0444290cecf Kent Overstreet 2020-12-14 2646 do {
06c0444290cecf Kent Overstreet 2020-12-14 2647 cond_resched();
723ef24b9b379e Kent Overstreet 2020-12-14 2648
^1da177e4c3f41 Linus Torvalds 2005-04-16 2649 /*
723ef24b9b379e Kent Overstreet 2020-12-14 2650 * If we've already successfully copied some data, then we
723ef24b9b379e Kent Overstreet 2020-12-14 2651 * can no longer safely return -EIOCBQUEUED. Hence mark
723ef24b9b379e Kent Overstreet 2020-12-14 2652 * an async read NOWAIT at that point.
^1da177e4c3f41 Linus Torvalds 2005-04-16 2653 */
87fa0f3eb267ee Christoph Hellwig 2021-02-24 2654 if ((iocb->ki_flags & IOCB_WAITQ) && already_read)
723ef24b9b379e Kent Overstreet 2020-12-14 2655 iocb->ki_flags |= IOCB_NOWAIT;
^1da177e4c3f41 Linus Torvalds 2005-04-16 2656
8c8387ee3f55a3 David Howells 2021-11-05 2657 if (unlikely(iocb->ki_pos >= i_size_read(inode)))
8c8387ee3f55a3 David Howells 2021-11-05 2658 break;
8c8387ee3f55a3 David Howells 2021-11-05 2659
3fc40265ae2b48 David Howells 2023-05-22 2660 error = filemap_get_pages(iocb, iter->count, &fbatch, false);
ff993ba130009b Matthew Wilcox (Oracle 2021-02-24 2661) if (error < 0)
06c0444290cecf Kent Overstreet 2020-12-14 2662 break;
^1da177e4c3f41 Linus Torvalds 2005-04-16 2663
^1da177e4c3f41 Linus Torvalds 2005-04-16 2664 /*
06c0444290cecf Kent Overstreet 2020-12-14 2665 * i_size must be checked after we know the pages are Uptodate.
06c0444290cecf Kent Overstreet 2020-12-14 2666 *
06c0444290cecf Kent Overstreet 2020-12-14 2667 * Checking i_size after the check allows us to calculate
06c0444290cecf Kent Overstreet 2020-12-14 2668 * the correct value for "nr", which means the zero-filled
06c0444290cecf Kent Overstreet 2020-12-14 2669 * part of the page is not copied back to userspace (unless
06c0444290cecf Kent Overstreet 2020-12-14 2670 * another truncate extends the file - this is desired though).
^1da177e4c3f41 Linus Torvalds 2005-04-16 2671 */
06c0444290cecf Kent Overstreet 2020-12-14 2672 isize = i_size_read(inode);
06c0444290cecf Kent Overstreet 2020-12-14 2673 if (unlikely(iocb->ki_pos >= isize))
25d6a23e8d2808 Matthew Wilcox (Oracle 2021-12-06 2674) goto put_folios;
06c0444290cecf Kent Overstreet 2020-12-14 2675 end_offset = min_t(loff_t, isize, iocb->ki_pos + iter->count);
723ef24b9b379e Kent Overstreet 2020-12-14 2676
06c0444290cecf Kent Overstreet 2020-12-14 2677 /*
06c0444290cecf Kent Overstreet 2020-12-14 2678 * Once we start copying data, we don't want to be touching any
06c0444290cecf Kent Overstreet 2020-12-14 2679 * cachelines that might be contended:
06c0444290cecf Kent Overstreet 2020-12-14 2680 */
06c0444290cecf Kent Overstreet 2020-12-14 2681 writably_mapped = mapping_writably_mapped(mapping);
06c0444290cecf Kent Overstreet 2020-12-14 2682
06c0444290cecf Kent Overstreet 2020-12-14 2683 /*
5ccc944dce3df5 Matthew Wilcox (Oracle 2022-06-10 2684) * When a read accesses the same folio several times, only
06c0444290cecf Kent Overstreet 2020-12-14 2685 * mark it as accessed the first time.
06c0444290cecf Kent Overstreet 2020-12-14 2686 */
881a8706c13e41 Haibo Li 2023-06-28 2687 if (!pos_same_folio(iocb->ki_pos, last_pos - 1,
5ccc944dce3df5 Matthew Wilcox (Oracle 2022-06-10 2688) fbatch.folios[0]))
25d6a23e8d2808 Matthew Wilcox (Oracle 2021-12-06 2689) folio_mark_accessed(fbatch.folios[0]);
06c0444290cecf Kent Overstreet 2020-12-14 2690
25d6a23e8d2808 Matthew Wilcox (Oracle 2021-12-06 2691) for (i = 0; i < folio_batch_count(&fbatch); i++) {
25d6a23e8d2808 Matthew Wilcox (Oracle 2021-12-06 2692) struct folio *folio = fbatch.folios[i];
d996fc7f615feb Matthew Wilcox (Oracle 2021-10-31 2693) size_t fsize = folio_size(folio);
d996fc7f615feb Matthew Wilcox (Oracle 2021-10-31 2694) size_t offset = iocb->ki_pos & (fsize - 1);
cbd59c48ae2bca Matthew Wilcox (Oracle 2021-02-24 2695) size_t bytes = min_t(loff_t, end_offset - iocb->ki_pos,
d996fc7f615feb Matthew Wilcox (Oracle 2021-10-31 2696) fsize - offset);
cbd59c48ae2bca Matthew Wilcox (Oracle 2021-02-24 2697) size_t copied;
06c0444290cecf Kent Overstreet 2020-12-14 2698
d996fc7f615feb Matthew Wilcox (Oracle 2021-10-31 2699) if (end_offset < folio_pos(folio))
cbd59c48ae2bca Matthew Wilcox (Oracle 2021-02-24 2700) break;
cbd59c48ae2bca Matthew Wilcox (Oracle 2021-02-24 2701) if (i > 0)
d996fc7f615feb Matthew Wilcox (Oracle 2021-10-31 2702) folio_mark_accessed(folio);
06c0444290cecf Kent Overstreet 2020-12-14 2703 /*
d996fc7f615feb Matthew Wilcox (Oracle 2021-10-31 2704) * If users can be writing to this folio using arbitrary
d996fc7f615feb Matthew Wilcox (Oracle 2021-10-31 2705) * virtual addresses, take care of potential aliasing
d996fc7f615feb Matthew Wilcox (Oracle 2021-10-31 2706) * before reading the folio on the kernel side.
06c0444290cecf Kent Overstreet 2020-12-14 2707 */
d996fc7f615feb Matthew Wilcox (Oracle 2021-10-31 2708) if (writably_mapped)
d996fc7f615feb Matthew Wilcox (Oracle 2021-10-31 @2709) flush_dcache_folio(folio);
cbd59c48ae2bca Matthew Wilcox (Oracle 2021-02-24 2710)
d996fc7f615feb Matthew Wilcox (Oracle 2021-10-31 2711) copied = copy_folio_to_iter(folio, offset, bytes, iter);
06c0444290cecf Kent Overstreet 2020-12-14 2712
87fa0f3eb267ee Christoph Hellwig 2021-02-24 2713 already_read += copied;
06c0444290cecf Kent Overstreet 2020-12-14 2714 iocb->ki_pos += copied;
881a8706c13e41 Haibo Li 2023-06-28 2715 last_pos = iocb->ki_pos;
06c0444290cecf Kent Overstreet 2020-12-14 2716
06c0444290cecf Kent Overstreet 2020-12-14 2717 if (copied < bytes) {
06c0444290cecf Kent Overstreet 2020-12-14 2718 error = -EFAULT;
06c0444290cecf Kent Overstreet 2020-12-14 2719 break;
^1da177e4c3f41 Linus Torvalds 2005-04-16 2720 }
^1da177e4c3f41 Linus Torvalds 2005-04-16 2721 }
25d6a23e8d2808 Matthew Wilcox (Oracle 2021-12-06 2722) put_folios:
25d6a23e8d2808 Matthew Wilcox (Oracle 2021-12-06 2723) for (i = 0; i < folio_batch_count(&fbatch); i++)
25d6a23e8d2808 Matthew Wilcox (Oracle 2021-12-06 2724) folio_put(fbatch.folios[i]);
25d6a23e8d2808 Matthew Wilcox (Oracle 2021-12-06 2725) folio_batch_init(&fbatch);
06c0444290cecf Kent Overstreet 2020-12-14 2726 } while (iov_iter_count(iter) && iocb->ki_pos < isize && !error);
^1da177e4c3f41 Linus Torvalds 2005-04-16 2727
^1da177e4c3f41 Linus Torvalds 2005-04-16 2728 file_accessed(filp);
881a8706c13e41 Haibo Li 2023-06-28 2729 ra->prev_pos = last_pos;
87fa0f3eb267ee Christoph Hellwig 2021-02-24 2730 return already_read ? already_read : error;
^1da177e4c3f41 Linus Torvalds 2005-04-16 2731 }
87fa0f3eb267ee Christoph Hellwig 2021-02-24 2732 EXPORT_SYMBOL_GPL(filemap_read);
^1da177e4c3f41 Linus Torvalds 2005-04-16 2733
:::::: The code at line 2709 was first introduced by commit
:::::: d996fc7f615feb5986f67829e18a8d8400f41361 filemap: Convert filemap_read() to use a folio
:::::: TO: Matthew Wilcox (Oracle) <willy@infradead.org>
:::::: CC: Matthew Wilcox (Oracle) <willy@infradead.org>
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
reply other threads:[~2023-07-08 2:57 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=202307081040.pxdFGWsP-lkp@intel.com \
--to=lkp@intel.com \
--cc=oe-kbuild-all@lists.linux.dev \
--cc=willy@infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.