On 22/4/24 19:30, Jingbo Xu wrote:
On 4/22/24 5:12 PM, Gao Xiang wrote:
Hi Ian,
(+Cc Jingbo here).
On 2024/4/22 16:31, Ian Kent wrote:
I'm new to the list so Hi to all,
I'm working with a heavily patched 5.14 kernel and I've gathered
together patches to bring erofs
up to 5.19 and I'm trying to run the erofs and fscache tests from a
checkout of the 1.7.1 repo.
(branch experimental-tests-fscache) and I have a couple of fails I
can't quite work out so I'm
hoping for a little halp.
Thanks for your interest and provide the detailed infos.
I guess a modified 5.14 kernel may be originated from RHEL 9?
I have a plan to backport the latest EROFS to CentOS stream 9, but
currently I'm busy in internal stuffs, so it's still a bit delayed...
For those familiar with erofs development and history it may look like
some patches are missing
but they may have already been present in the source tree I'm working
with ... so just ask if
you spot anything ...
Current set of patches I have is (the patch name corresponds to commit
title, more or less):
+ erofs-clear-compacted_2b-if-compacted_4b_initial-gt-totalidx.patch
+ erofs-add-support-for-the-full-decompressed-length.patch
+ erofs-add-fiemap-support-with-iomap.patch
+
erofs-remove-the-mapping-parameter-from-erofs_try_to_free_cached_page.patch
+ erofs-directly-use-wrapper-erofs_page_is_managed-when-shrinking.patch
+ erofs-dax-support-for-non-tailpacking-regular-file.patch
+ erofs-fix-deadlock-when-shrink-erofs-slab.patch
+ erofs-remove-useless-cache-strategy-of-DELAYEDALLOC.patch
+ erofs-fix-unsafe-pagevec-reuse-of-hooked-pclusters.patch
+ erofs-remove-the-fast-path-of-per-CPU-buffer-decompression.patch
+ erofs-decouple-basic-mount-options-from-fs_context.patch
+ erofs-add-multiple-device-support.patch
+ erofs-get-compression-algorithms-directly-on-mapping.patch
+ erofs-introduce-the-secondary-compression-head.patch
+ erofs-introduce-readmore-decompression-strategy.patch
+ erofs-rename-some-generic-methods-in-decompressor.patch
+
libxz-Avoid-overlapping-memcpy-with-invalid-input-with-in-place-decompression.patch
+ libxz-Add-MicroLZMA-decoder.patch
+ erofs-lzma-compression-support.patch
+ erofs-get-rid-of-lru-usage.patch
+ erofs-dont-trigger-WARN-when-decompression-fails.patch
+ erofs-rename-lz4_0pading-to-zero_padding.patch
+ erofs-add-sysfs-interface.patch
+ erofs-add-sysfs-node-to-control-sync-decompression-strategy.patch
+ erofs-Replace-zero-length-array-with-flexible-array-member.patch
+ erofs-clean-up-erofs_map_blocks-tracepoints.patch
+ erofs-tidy-up-z_erofs_lz4_decompress.patch
+ erofs-introduce-z_erofs_fixup_insize.patch
+ erofs-support-unaligned-data-decompression.patch
+ erofs-support-inline-data-decompression.patch
+ erofs-add-on-disk-compressed-tail-packing-inline-support.patch
+ erofs-introduce-meta-buffer-operations.patch
+ erofs-use-meta-buffers-for-inode-operations.patch
+ erofs-use-meta-buffers-for-super-operations.patch
+ erofs-use-meta-buffers-for-xattr-operations.patch
+ erofs-use-meta-buffers-for-zmap-operations.patch
+ erofs-fix-fsdax-partition-offset-handling.patch
+ erofs-avoid-unnecessary-z_erofs-decompressqueue_work-declaration.patch
+ erofs-fix-small-compressed-files-inlining.patch
+ erofs-fix-ztailpacking-on-gt-4GiB-filesystems.patch
+ erofs-use-meta-buffers-for-erofs_read_superblock.patch
+ erofs-get-rid-of-struct-z_erofs_collector.patch
+ erofs-clean-up-preload_compressed_pages.patch
+ erofs-silence-warnings-related-to-impossible-m_plen.patch
+ erofs-clean-up-z_erofs_extent_lookback.patch
+ erofs-refine-managed-inode-stuffs.patch
+ erofs-add-sanity-check-0for-kobject-in-erofs_unregister_sysfs.patch
+ erofs-use-meta-buffers-for-reading-directories.patch
+ erofs-use-meta-buffers-for-inode-lookup.patch
+ erofs-rename-ctime-to-mtime.patch
+ erofs-Convert-from-invalidatepage-to-invalidate_folio.patch
+ erofs-fix-use-after-free-of-on-stack-io.patch
+ erofs-Convert-erofs-zdata-to-read_folio.patch
+ erofs-Convert-to-release_folio.patch
+ erofs-do-not-prompt-for-risk-any-more-when-using-big-pcluster.patch
+ erofs-remove-obsoleted-comments.patch
+ erofs-refine-on-disk-definition-comments.patch
+ erofs-fix-buffer-copy-overflow-of-ztailpacking-feature.patch
+ erofs-make-filesystem-exportable.patch
+ erofs-support-idmapped-mounts.patch
+ cachefiles-document-on-demand-read-mode.patch
+ erofs-make-erofs_map_blocks-generally-available.patch
+ erofs-add-fscache-mode-check-helper.patch
+ erofs-register-fscache-volume.patch
+ erofs-add-fscache-context-helper-functions.patch
+ erofs-add-anonymous-inode-caching-metadata-for-data-blobs.patch
+ erofs-add-erofs_fscache_read_folios-helper.patch
+ erofs-register-fscache-context-for-primary-data-blob.patch
+ erofs-register-fscache-context-for-extra-data-blobs.patch
+ erofs-implement-fscache-based-metadata-read.patch
+ erofs-implement-fscache-based-data-read-for-non-inline-layout.patch
+ erofs-implement-fscache-based-data-read-for-inline-layout.patch
+ erofs-implement-fscache-based-data-readahead.patch
+ erofs-add-fsid-mount-option.patch
+
erofs-change-to-use-asynchronous-io-for-fscache-readpage_readahead.patch
+ erofs-scan-devices-from-device-table.patch
+ erofs-leave-compressed-inodes-unsupported-in-fscache-mode-for-now.patch
+ erofs-fix-crash-when-enable-tracepoint-cachefiles_prep_read.patch
+ erofs-get-rid-of-struct-z_erofs_collection.patch
+ erofs-get-rid-of-label-restart_now.patch
+ erofs-simplify-z_erofs_pcluster_readmore.patch
+ erofs-fix-backmost-member-of-z_erofs_decompress_frontend.patch
+ erofs-missing-hunks.patch
The last patch consists of what looks like a few hunks added by Linus
to complete a folio pull
request that came in at the same time as the 5.19 erofs merge request.
I know the list of
patches isn't very useful but it should give some idea of what I have
and maybe someone can
spot a missing patch or so.
Anyway, my failing tests are erofs/021, erofs/022, erofs/024 and
fscache/005.
I guess the following failure fails as expected:
erofs/021 -- uncompressed sub-page block sizes (esp. 512-byte block
sizes, since v6.4)
erofs/022 -- long xattr prefix (since v6.4)
erofs/024 -- deflate algorithm support (since v6.6)
So these failures can be skipped on your side, I think I need to modify
these tests for gracefully skipping ... That is also why all testcases
are marked as "experimental" :-)
I'm not quite sure why "fscache/005" fails, hopefully Jingbo could
help you on this.
"Test if system could recover from previous wrong behaved user daemon."
from fscache/005 testcase.
I just test fscache/005 on the latest linux master branch and it
succeeds as expected.
I'm not sure if it is because the code base changes between v5.14 and
v5.19, since cachefiles/fscache is totally reworked since v5.16.
I see the following patches supporting on demand mode for cachefiles are
not included in the patch list you gave, but I believe they already
exist in your source tree, otherwise all fscache related testcases will
fail.
Yes, it appears a number of cachefiles changes were back ported earlier by
Jeff Layton as part of a RHEL NFS update.
99302ebd3af7 cachefiles: document on-demand read mode
1519670e4fec cachefiles: add tracepoints for on-demand read mode
4e4f1788af0e cachefiles: enable on-demand read mode
9032b6e8589f cachefiles: implement on-demand read
324b954ac80c cachefiles: notify the user daemon when withdrawing cookie
d11b0b043b40 cachefiles: unbind cachefiles gracefully in on-demand mode
c8383054506c cachefiles: notify the user daemon when looking up cookie
a06fac1599c1 cachefiles: extract write routine
I had seen these, which is what lead me to Jeffs' work but I'll check
again, thanks.
Anyway is there any hint in the fscache/005.bad?
I am working on getting more info. out of scripts (in an ad-hoc manner)
but it's
proved to be a little difficult so far.
Anyway, thanks for getting back to me on this.
Ian
