summaryrefslogtreecommitdiffstats
path: root/drivers/staging/erofs
Commit message (Collapse)AuthorAgeFilesLines
* erofs: move erofs out of stagingGao Xiang2019-08-2422-7246/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | EROFS filesystem has been merged into linux-staging for a year. EROFS is designed to be a better solution of saving extra storage space with guaranteed end-to-end performance for read-only files with the help of reduced metadata, fixed-sized output compression and decompression inplace technologies. In the past year, EROFS was greatly improved by many people as a staging driver, self-tested, betaed by a large number of our internal users, successfully applied to almost all in-service HUAWEI smartphones as the part of EMUI 9.1 and proven to be stable enough to be moved out of staging. EROFS is a self-contained filesystem driver. Although there are still some TODOs to be more generic, we have a dedicated team actively keeping on working on EROFS in order to make it better with the evolution of Linux kernel as the other in-kernel filesystems. As Pavel suggested, it's better to do as one commit since git can do moves and all histories will be saved in this way. Let's promote it from staging and enhance it more actively as a "real" part of kernel for more wider scenarios! Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Stephen Rothwell <sfr@canb.auug.org.au> Cc: Theodore Ts'o <tytso@mit.edu> Cc: Pavel Machek <pavel@denx.de> Cc: David Sterba <dsterba@suse.cz> Cc: Amir Goldstein <amir73il@gmail.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Darrick J . Wong <darrick.wong@oracle.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Jaegeuk Kim <jaegeuk@kernel.org> Cc: Jan Kara <jack@suse.cz> Cc: Richard Weinberger <richard@nod.at> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Chao Yu <yuchao0@huawei.com> Cc: Miao Xie <miaoxie@huawei.com> Cc: Li Guifu <bluce.liguifu@huawei.com> Cc: Fang Wei <fangwei1@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190822213659.5501-1-hsiangkao@aol.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: detect potential multiref due to corrupted imagesGao Xiang2019-08-211-3/+17
| | | | | | | | | | | | | As reported by erofs-utils fuzzer, currently, multiref (ondisk deduplication) hasn't been supported for now, we should forbid it properly. Fixes: 3883a79abd02 ("staging: erofs: introduce VLE decompression support") Cc: <stable@vger.kernel.org> # 4.19+ Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Link: https://lore.kernel.org/r/20190821140152.229648-1-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: avoid loop in submit chainsGao Xiang2019-08-211-1/+15
| | | | | | | | | | | | | | | | | | As reported by erofs-utils fuzzer, 2 conditions can happen in corrupted images, which can cause unexpected behaviors. - access the same pcluster one more time; - access the tail end pcluster again, e.g. _ access again (will trigger tail merging) | 1 2 3 1 2 -> 1 2 3 1 |_ tail end of the chain \___/ (unexpected behavior) Let's detect and avoid them now. Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190821030908.40282-1-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: avoid endless loop of invalid lookback distance 0Gao Xiang2019-08-211-0/+6
| | | | | | | | | | | | | As reported by erofs-utils fuzzer, Lookback distance should be a positive number, so it should be actually looked back rather than spinning. Fixes: 02827e1796b3 ("staging: erofs: add erofs_map_blocks_iter") Cc: <stable@vger.kernel.org> # 4.19+ Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Link: https://lore.kernel.org/r/20190819103426.87579-7-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: add two missing erofs_workgroup_put for corrupted imagesGao Xiang2019-08-201-2/+4
| | | | | | | | | | | | | | | | | As reported by erofs-utils fuzzer, these error handling path will be entered to handle corrupted images. Lack of erofs_workgroup_puts will cause unmounting unsuccessfully. Fix these return values to EFSCORRUPTED as well. Fixes: 3883a79abd02 ("staging: erofs: introduce VLE decompression support") Cc: <stable@vger.kernel.org> # 4.19+ Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Link: https://lore.kernel.org/r/20190819103426.87579-4-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: cannot set EROFS_V_Z_INITED_BIT if fill_inode_lazy failsGao Xiang2019-08-201-2/+1
| | | | | | | | | | | | | | As reported by erofs-utils fuzzer, unsupported compressed clustersize will make fill_inode_lazy fail, for such case we cannot set EROFS_V_Z_INITED_BIT since we need return failure for each z_erofs_map_blocks_iter(). Fixes: 152a333a5895 ("staging: erofs: add compacted compression indexes support") Cc: <stable@vger.kernel.org> # 5.3+ Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Link: https://lore.kernel.org/r/20190819103426.87579-3-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: some compressed cluster should be submitted for corrupted imagesGao Xiang2019-08-201-6/+5
| | | | | | | | | | | | | | | | | | | As reported by erofs_utils fuzzer, a logical page can belong to at most 2 compressed clusters, if one compressed cluster is corrupted, but the other has been ready in submitting chain. The chain needs to submit anyway in order to keep the page working properly (page unlocked with PG_error set, PG_uptodate not set). Let's fix it now. Fixes: 3883a79abd02 ("staging: erofs: introduce VLE decompression support") Cc: <stable@vger.kernel.org> # 4.19+ Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Link: https://lore.kernel.org/r/20190819103426.87579-2-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: fix an error handling in erofs_readdir()Gao Xiang2019-08-201-2/+9
| | | | | | | | | | | | | | | | | | | | | | | Richard observed a forever loop of erofs_read_raw_page() [1] which can be generated by forcely setting ->u.i_blkaddr to 0xdeadbeef (as my understanding block layer can handle access beyond end of device correctly). After digging into that, it seems the problem is highly related with directories and then I found the root cause is an improper error handling in erofs_readdir(). Let's fix it now. [1] https://lore.kernel.org/r/1163995781.68824.1566084358245.JavaMail.zimbra@nod.at/ Reported-by: Richard Weinberger <richard@nod.at> Fixes: 3aa8ec716e52 ("staging: erofs: add directory operations") Cc: <stable@vger.kernel.org> # 4.19+ Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190818125457.25906-1-hsiangkao@aol.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: refuse to mount images with malformed volume nameGao Xiang2019-08-201-2/+7
| | | | | | | | | | | | | | | | As Richard reminder [1], A valid volume name should be ended in NIL terminator within the length of volume_name. Since this field currently isn't really used, let's fix it to avoid potential bugs in the future. [1] https://lore.kernel.org/r/1133002215.69049.1566119033047.JavaMail.zimbra@nod.at/ Reported-by: Richard Weinberger <richard@nod.at> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Link: https://lore.kernel.org/r/20190818102824.22330-1-hsiangkao@aol.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: use common file type conversionGao Xiang2019-08-183-28/+7
| | | | | | | | | | | | | | | Deduplicate the EROFS file type conversion implementation and remove EROFS_FT_* definitions since it's the same as defined by POSIX, let's follow ext2 as Linus pointed out [1] commit e10892189428 ("ext2: use common file type conversion"). [1] https://lore.kernel.org/r/CAHk-=wiUs+b=iVKM3mVooXgVk7cmmC67KTmnAuL0cd_cMMVAKw@mail.gmail.com/ Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Link: https://lore.kernel.org/r/20190816071142.8633-1-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: correct all misused ENOTSUPPGao Xiang2019-08-155-11/+11
| | | | | | | | | | | | | As Chao pointed out [1], ENOTSUPP is used for NFS protocol only, we should use EOPNOTSUPP instead... [1] https://lore.kernel.org/lkml/108ee2f9-75dd-b8ab-8da7-b81c17bafbf6@huawei.com/ Reported-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Link: https://lore.kernel.org/r/20190814103705.60698-3-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: differentiate unsupported on-disk formatGao Xiang2019-08-152-5/+5
| | | | | | | | | | For some specific fields, use EOPNOTSUPP instead of EIO for values which look sane but aren't supported right now. Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190814103705.60698-2-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: introduce EFSCORRUPTED and more logsGao Xiang2019-08-157-20/+36
| | | | | | | | | | | | | | | | | Previously, EROFS uses EIO to indicate that filesystem is corrupted as well. However, as Pavel said [1], other filesystems tend to use EUCLEAN(EFSCORRUPTED) instead, let's follow what others do right now. Also, add some more prints to the syslog. [1] https://lore.kernel.org/lkml/20190813114821.GB11559@amd/ Suggested-by: Pavel Machek <pavel@denx.de> Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190814103705.60698-1-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: removing an extra call to iloc() in fill_inode()Pratik Shinde2019-08-151-3/+4
| | | | | | | | | | | in fill_inode() we call iloc() twice.Avoiding the extra call by storing the result. Signed-off-by: Pratik Shinde <pratikshinde320@gmail.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Cc: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190813203840.13782-1-pratikshinde320@gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: xattr.c: avoid BUG_ONGao Xiang2019-08-141-5/+9
| | | | | | | | | | | | | Kill all the remaining BUG_ON in EROFS: - one BUG_ON was used to detect xattr on-disk corruption, proper error handling should be added for it instead; - the other BUG_ONs are used to detect potential issues, use DBG_BUGON only in (eng) debugging version. Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Link: https://lore.kernel.org/r/20190813023054.73126-3-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: remove incomplete cleancacheGao Xiang2019-08-142-7/+0
| | | | | | | | | | | | cleancache was not fully implemented in EROFS. In addition, it's tend to remove the whole cleancache in related attempt [1]. [1] https://lore.kernel.org/linux-fsdevel/20190527103207.13287-3-jgross@suse.com/ Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Link: https://lore.kernel.org/r/20190813023054.73126-2-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: inline erofs_inode_is_data_compressed()Gao Xiang2019-08-141-1/+1
| | | | | | | | | | As a helper in erofs_fs.h, erofs_inode_is_data_compressed() should be inlined. Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Link: https://lore.kernel.org/r/20190813023054.73126-1-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: update KconfigGao Xiang2019-08-021-26/+28
| | | | | | | | | | | | Keep in line with erofs-outofstaging patchset: - turn on CONFIG_EROFS_FS_ZIP by default; - turn on CONFIG_EROFS_FS_SECURITY by default suggested by David; - update Kconfig description. Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-23-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: update super.cGao Xiang2019-08-021-3/+2
| | | | | | | | | | | Keep in line with erofs-outofstaging patchset: - "Chao Yu" is most commonly used in Linux community; - quoted string split across lines. Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-22-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: tidy up utils.cGao Xiang2019-08-021-14/+13
| | | | | | | | | | | Keep in line with erofs-outofstaging patchset: - Update comments in erofs_try_to_release_workgroup; - code style cleanup. Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-21-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: turn cache strategies into mount optionsGao Xiang2019-08-027-140/+85
| | | | | | | | | | | | | | | | Kill all kconfig cache strategies and turn them into mount options "cache_strategy={disable|readahead|readaround}". As the first step, cached pages can still be usable after cache is disabled by remounting, and these pages will be fallen out over time, which can be refined in the later version if some requirement is needed. Update related document as well. Suggested-by: Theodore Ts'o <tytso@mit.edu> Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-20-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: remove clusterbits in sbiGao Xiang2019-08-023-16/+1
| | | | | | | | | | clustersize can now be set on per-file basis rather than per-filesystem basis. Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-19-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: tidy up decompression frontendGao Xiang2019-08-022-614/+465
| | | | | | | | | | | | | | | | | | | | | | | | | | | Although this patch has an amount of changes, it is hard to separate into smaller patches. Most changes are due to structure renaming for better understand and straightforward, z_erofs_vle_workgroup to z_erofs_pcluster since it represents a physical cluster; z_erofs_vle_work to z_erofs_collection since it represents a collection of logical pages; z_erofs_vle_work_builder to z_erofs_collector since it's used to fill z_erofs_{pcluster,collection}. struct z_erofs_vle_work_finder has no extra use compared with struct z_erofs_collector, delete it. FULL_LENGTH bit is integrated into .length of pcluster so that it can be updated with the corresponding length change in atomic. Minor, add comments for better description. Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-18-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: remove redundant braces in inode.cGao Xiang2019-08-021-10/+8
| | | | | | | | | | Remove redundant braces in inode.c since these are all single statements. Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-17-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: tidy up zpvec.hGao Xiang2019-08-022-19/+12
| | | | | | | | | | | | | - use shorter function names: z_erofs_pagevec_enqueue and z_erofs_pagevec_dequeue; - minor code cleanup. In order to keep in line with erofs-outofstaging patchset. Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-16-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: kill CONFIG_EROFS_FS_USE_VM_MAP_RAMGao Xiang2019-08-023-17/+17
| | | | | | | | | | | Turn into a module parameter ("use_vmap") as it can be set at runtime. Suggested-by: David Sterba <dsterba@suse.cz> Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-15-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: refine erofs_allocpage()Gao Xiang2019-08-024-16/+8
| | | | | | | | | | remove duplicated code in decompressor by introducing failable erofs_allocpage(). Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-14-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: drop __GFP_NOFAIL for managed inodeGao Xiang2019-08-021-2/+1
| | | | | | | | | | For historical reasons, __GFP_NOFAIL was set for managed inode. It's no need using that since EROFS can handle it properly. Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-13-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: kill all failure handling in fill_super()Gao Xiang2019-08-021-66/+52
| | | | | | | | | | | | | | | .kill_sb() will do that instead in order to remove duplicated code. Note that the initialzation of managed_cache is now moved after s_root is assigned since it's more preferred to iput() in .put_super() and all inodes should be evicted before the end of generic_shutdown_super(sb). Suggested-by: Al Viro <viro@zeniv.linux.org.uk> Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-12-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: kill sbi->dev_nameGao Xiang2019-08-022-49/+7
| | | | | | | | | | | | | | | | | As Al said, "the only use of sbi->dev_name is debugging printks and all of those have sb->s_id available, with device name stored in there. Which makes the whole thing bloody weird". sbi->dev_name was used for our debugging use and it's better to just use s_id in community and delete the whole erofs_mount_private stuff. Suggested-by: Al Viro <viro@zeniv.linux.org.uk> Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-11-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: clean up shrinker stuffsGao Xiang2019-08-023-51/+56
| | | | | | | | | | | | | | | - rename erofs_register_super / erofs_unregister_super to erofs_shrinker_register / erofs_shrinker_unregister; - fold the only erofs_shrink_workstation external call to erofs_shrinker_unregister; - localize erofs_shrink_workstation; - localize erofs_shrinker_info by introducing erofs_init_shrinker and erofs_exit_shrinker. Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-10-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: kill CONFIG_EROFS_FS_IO_MAX_RETRIESGao Xiang2019-08-022-15/+2
| | | | | | | | | | | | | | CONFIG_EROFS_FS_IO_MAX_RETRIES seems a runtime setting and users have no idea about the change in behaviour. Let's remove the setting currently and could turn it into a module parameter if it's really needed. Suggested-by: David Sterba <dsterba@suse.cz> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-9-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: remove redundant #include "internal.h"Gao Xiang2019-08-022-2/+0
| | | | | | | | | Because #include "internal.h" is included in xattr.h Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-8-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: clean up internal.hGao Xiang2019-08-025-115/+92
| | | | | | | | | | | | | Tidy up relative order of variables / declarations in internal.h, move some local static functions out into other files and add tags at the end of #endif acrossing several lines. No logic change. Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-7-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: sunset erofs_workstn_{lock,unlock}Gao Xiang2019-08-022-8/+5
| | | | | | | | | | It was used for Linux backward compatibility, and no use for upstream kernel. Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-6-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: keep up erofs_fs.h with erofs-outofstaging patchsetGao Xiang2019-08-021-20/+19
| | | | | | | | | | | | The main change is to reserve all checksums except for superblock, since it's more useful to do block-based verity for read-only fs. Some comments change as well, which is minor. Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-5-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: fix dummy functions erofs_{get, list}xattrGao Xiang2019-08-021-7/+7
| | | | | | | | | | | | | dummy functions erofs_{get,list}xattr should be inlined without xattr enabled. Signed-off-by: Yue Hu <zbestahu@gmail.com> [ Gao Xiang : this patch was "staging: erofs: remove needless dummy functions of erofs_{get,list}xattr. "] Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-4-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: rename source files for better understandingGao Xiang2019-08-025-14/+14
| | | | | | | | | | Keep in line with erofs-outofstaging patchset as well, see https://lore.kernel.org/linux-fsdevel/20190725095658.155779-1-gaoxiang25@huawei.com/ Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-3-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: update source file headersGao Xiang2019-08-0220-80/+27
| | | | | | | | | | | - Use the correct style for all SPDX License Identifiers; - Get rid of the unnecessary license boilerplate; - Use "GPL-2.0-only" instead of "GPL-2.0" suggested-by Stephen. Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190731155752.210602-2-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: support bmapChao Yu2019-07-221-0/+33
| | | | | | | | | Add erofs_bmap() to support FIBMAP ioctl on flatmode inode. Reviewed-by: Gao Xiang <gaoxiang25@huawei.com> Signed-off-by: Chao Yu <yuchao0@huawei.com> Link: https://lore.kernel.org/r/20190716093256.108791-1-yuchao0@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: avoid opened loop codesChao Yu2019-07-221-8/+8
| | | | | | | | | Use __GFP_NOFAIL to avoid opened loop codes in z_erofs_vle_unzip(). Signed-off-by: Chao Yu <yuchao0@huawei.com> Reviewed-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190716094422.110805-1-yuchao0@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: Remove function erofs_kill_sb()Nishka Dasgupta2019-07-221-6/+1
| | | | | | | | | | | | Remove function erofs_kill_sb as all it does is call kill_block_super. Modify references to the former to point to the latter. Issue found with Coccinelle. Signed-off-by: Nishka Dasgupta <nishkadg.linux@gmail.com> Reviewed-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Link: https://lore.kernel.org/r/20190712071247.2357-1-nishkadg.linux@gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs:converting all 'unsigned' to 'unsigned int'Pratik Shinde2019-07-224-20/+23
| | | | | | | | | Fixed checkpatch warnings: converting all 'unsigned' to 'unsigned int' Signed-off-by: Pratik Shinde <pratikshinde320@gmail.com> Reviewed-by: Gao Xiang <gaoxiang25@huawei.com> Link: https://lore.kernel.org/r/20190715122127.13733-1-pratikshinde320@gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: fix LZ4 limited bounced page mis-reuseGao Xiang2019-07-031-22/+28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Like all lz77-based algrithms, lz4 has a dynamically populated ("sliding window") dictionary and the maximum lookback distance is 65535. Therefore the number of bounced pages could be limited by erofs based on this property. However, just now we observed some lz4 sequences in the extreme case cannot be decompressed correctly after this feature is enabled, the root causes after analysis are clear as follows: 1) max bounced pages should be 17 rather than 16 pages; 2) considering the following case, the broken implementation could reuse unsafely in advance (in other words, reuse it less than a safe distance), 0 1 2 ... 16 17 18 ... 33 34 b p b b note that the bounce page that we are concerned was allocated at 0, and it reused at 18 since page 17 exists, but it mis-reused at 34 in advance again, which causes decompress failure. This patch resolves the issue by introducing a bitmap to mark whether the page in the same position of last round is a bounced page or not, and a micro stack data structure to store all available bounced pages. Fixes: 7fc45dbc938a ("staging: erofs: introduce generic decompression backend") Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: Replace kzalloc(struct ..) with kzalloc(*ptr)Shobhit Kukreti2019-07-012-3/+2
| | | | | | | | | | Resolve checkpatch warning: Prefer kzalloc(sizeof(*ptr)...) over kzalloc(sizeof(struct ..) Signed-off-by: Shobhit Kukreti <shobhitkukreti@gmail.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Reviewed-by: Gao Xiang <gaoxiang25@huawei.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: don't check special inode layoutYue Hu2019-07-011-0/+1
| | | | | | | | | | | Currently, we will check if inode layout is compression or inline if the inode is special in fill_inode(). Also set ->i_mapping->a_ops for it. That is pointless since the both modes won't be set for special inode when creating EROFS filesystem image. So, let's avoid it. Signed-off-by: Yue Hu <huyue2@yulong.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: return the error value if fill_inline_data() failsYue Hu2019-07-011-2/+2
| | | | | | | | | | | | We should consider the error returned by fill_inline_data() when filling last page in fill_inode(). If not getting inode will be successful even though last page is bad. That is illogical. Also change -EAGAIN to 0 in fill_inline_data() to stand for successful filling. Signed-off-by: Yue Hu <huyue2@yulong.com> Reviewed-by: Gao Xiang <gaoxiang25@huawei.com> Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: integrate decompression inplaceGao Xiang2019-06-264-4/+16
| | | | | | | | | | | | | | Decompressor needs to know whether it's a partial or full decompression since only full decompression can be decompressed in-place. On kirin980 platform, sequential read is finally increased to 812MiB/s after decompression inplace is enabled. Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: switch to new decompression backendGao Xiang2019-06-265-280/+24
| | | | | | | | | | | | | | This patch integrates new decompression framework to erofs decompression path, and remove the old decompression implementation as well. On kirin980 platform, sequential read is slightly improved to 778MiB/s after the new decompression backend is used. Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* staging: erofs: introduce LZ4 decompression inplaceGao Xiang2019-06-263-5/+35
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | compressed data will be usually loaded into last pages of the extent (the last page for 4k) for in-place decompression (more specifically, in-place IO), as ilustration below, start of compressed logical extent | end of this logical extent | | ______v___________________________v________ ... | page 6 | page 7 | page 8 | page 9 | ... |__________|__________|__________|__________| . ^ . ^ . |compressed| . | data | . . . |< dstsize >|<margin>| oend iend op ip Therefore, it's possible to do decompression inplace (thus no memcpy at all) if the margin is sufficient and safe enough [1], and it can be implemented only for fixed-size output compression compared with fixed-size input compression. No memcpy for most of in-place IO (about 99% of enwik9) after decompression inplace is implemented and sequential read will be improved of course (see the following patches for test results). [1] https://github.com/lz4/lz4/commit/b17f578a919b7e6b078cede2d52be29dd48c8e8c https://github.com/lz4/lz4/commit/5997e139f53169fa3a1c1b4418d2452a90b01602 Reviewed-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Gao Xiang <gaoxiang25@huawei.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
generated by cgit v1.2.3-71-gd317 (git 2.46.0) at 2026-01-31 16:11:40 +0000