linux.git/fs/erofs/decompressor.c, branch v5.7.14

erofs: handle corrupted images whose decompressed size less than it'd be

2020-03-03T15:40:52+00:00

As Lasse pointed out, "Looking at fs/erofs/decompress.c,
the return value from LZ4_decompress_safe_partial is only
checked for negative value to catch errors. ... So if
I understood it correctly, if there is bad data whose
uncompressed size is much less than it should be, it can
leave part of the output buffer untouched and expose the
previous data as the file content. "

Let's fix it now.

Cc: Lasse Collin 
Fixes: 7fc45dbc938a ("staging: erofs: introduce generic decompression backend")
[ Gao Xiang: v5.3+, I will manually backport this to stable later. ]
Link: https://lore.kernel.org/r/20200226081008.86348-3-gaoxiang25@huawei.com
Reviewed-by: Chao Yu 
Signed-off-by: Gao Xiang

erofs: use LZ4_decompress_safe() for full decoding

2020-03-03T15:40:16+00:00

As Lasse pointed out, "EROFS uses LZ4_decompress_safe_partial
for both partial and full blocks. Thus when it is decoding a
full block, it doesn't know if the LZ4 decoder actually decoded
all the input. The real uncompressed size could be bigger than
the value stored in the file system metadata.

Using LZ4_decompress_safe instead of _safe_partial when
decompressing a full block would help to detect errors."

So it's reasonable to use _safe in case of potential corrupted
images and it might have some speed gain as well although
I didn't observe much difference.

Note that legacy compressor (< 5.3, no LZ4_0PADDING) could
encode extra data in a pcluster, which is excluded as well.

Cc: Lasse Collin 
Fixes: 0ffd71bcc3a0 ("staging: erofs: introduce LZ4 decompression inplace")
[ Gao Xiang: v5.3+, I will manually backport this to stable later. ]
Link: https://lore.kernel.org/r/20200226081008.86348-2-gaoxiang25@huawei.com
Reviewed-by: Chao Yu 
Signed-off-by: Gao Xiang

erofs: fix out-of-bound read for shifted uncompressed block

2020-01-11T10:29:19+00:00

rq->out[1] should be valid before accessing. Otherwise,
in very rare cases, out-of-bound dirty onstack rq->out[1]
can equal to *in and lead to unintended memmove behavior.

Link: https://lore.kernel.org/r/20200107022546.19432-1-gaoxiang25@huawei.com
Fixes: 7fc45dbc938a ("staging: erofs: introduce generic decompression backend")
Cc:  # 5.3+
Reviewed-by: Chao Yu 
Signed-off-by: Gao Xiang

erofs: get rid of __stagingpage_alloc helper

2019-11-24T02:57:48+00:00

Now open code is much cleaner due to iterative development.

Link: https://lore.kernel.org/r/20191124025217.12345-1-hsiangkao@aol.com
Reviewed-by: Chao Yu 
Signed-off-by: Gao Xiang

erofs: rename errln/infoln/debugln to erofs_{err, info, dbg}

2019-09-05T18:10:09+00:00

Add prefix "erofs_" to these functions and print
sb->s_id as a prefix to erofs_{err, info} so that
the user knows which file system is affected.

Reported-by: Christoph Hellwig 
Signed-off-by: Gao Xiang 
Link: https://lore.kernel.org/r/20190904020912.63925-23-gaoxiang25@huawei.com
Signed-off-by: Greg Kroah-Hartman

erofs: kill use_vmap module parameter

2019-09-05T18:10:09+00:00

As Christoph said [1],
"vm_map_ram is supposed to generally behave better.  So if
it doesn't please report that that to the arch maintainer
and linux-mm so that they can look into the issue.  Having
user make choices of deep down kernel internals is just
a horrible interface.

Please talk to maintainers of other bits of the kernel
if you see issues and / or need enhancements. "

Let's redo the previous conclusion and kill the vmap
approach.

[1] https://lore.kernel.org/r/20190830165533.GA10909@infradead.org/
Reported-by: Christoph Hellwig 
Signed-off-by: Gao Xiang 
Link: https://lore.kernel.org/r/20190904020912.63925-21-gaoxiang25@huawei.com
Signed-off-by: Greg Kroah-Hartman

erofs: add "erofs_" prefix for common and short functions

2019-09-05T18:10:08+00:00

Add erofs_ prefix to free_inode, alloc_inode, ...

Reported-by: Christoph Hellwig 
Signed-off-by: Gao Xiang 
Link: https://lore.kernel.org/r/20190904020912.63925-19-gaoxiang25@huawei.com
Signed-off-by: Greg Kroah-Hartman

erofs: use feature_incompat rather than requirements

2019-09-05T18:10:07+00:00

As Christoph said [1], "This is only cosmetic, why
not stick to feature_compat and feature_incompat?"

In my thought, requirements means "incompatible"
instead of "feature" though.

[1] https://lore.kernel.org/r/20190902125109.GA9826@infradead.org/
Reported-by: Christoph Hellwig 
Signed-off-by: Gao Xiang 
Link: https://lore.kernel.org/r/20190904020912.63925-7-gaoxiang25@huawei.com
Signed-off-by: Greg Kroah-Hartman

erofs: remove all likely/unlikely annotations

2019-08-30T07:02:02+00:00

As Dan Carpenter suggested [1], I have to remove
all erofs likely/unlikely annotations.

[1] https://lore.kernel.org/linux-fsdevel/20190829154346.GK23584@kadam/
Reported-by: Dan Carpenter 
Signed-off-by: Gao Xiang 
Link: https://lore.kernel.org/r/20190829163827.203274-1-gaoxiang25@huawei.com
Signed-off-by: Greg Kroah-Hartman

erofs: move erofs out of staging

2019-08-24T12:20:10+00:00

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 
Cc: Alexander Viro 
Cc: Andrew Morton 
Cc: Stephen Rothwell 
Cc: Theodore Ts'o 
Cc: Pavel Machek 
Cc: David Sterba 
Cc: Amir Goldstein 
Cc: Christoph Hellwig 
Cc: Darrick J . Wong 
Cc: Dave Chinner 
Cc: Jaegeuk Kim 
Cc: Jan Kara 
Cc: Richard Weinberger 
Cc: Linus Torvalds 
Cc: Chao Yu 
Cc: Miao Xie 
Cc: Li Guifu 
Cc: Fang Wei 
Signed-off-by: Gao Xiang 
Link: https://lore.kernel.org/r/20190822213659.5501-1-hsiangkao@aol.com
Signed-off-by: Greg Kroah-Hartman