04 Apr, 2014
1 commit
-
Include appropriate header file include/linux/decompress/inflate.h in
lib/decompress_inflate.c because it has prototype declaration of
function defined in lib/decompress_inflate.c.Also, fix the guard around the header file
include/linux/decompress/inflate.h to use a more unique guard symbol.
This avoids conflict with the INFLATE_H defined by
zlib_inflate/inflate.h.This eliminates the following warning in lib/decompress_inflate.c:
lib/decompress_inflate.c:35:17: warning: no previous prototype for `gunzip' [-Wmissing-prototypes]
Signed-off-by: Rashika Kheria
Reviewed-by: Josh Triplett
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
12 Sep, 2013
1 commit
-
When decompressing into memory, the output buffer length is set to some
arbitrarily high value (0x7fffffff) to indicate the output is, virtually,
unlimited in size.The problem with this is that some platforms have their physical memory at
high physical addresses (0x80000000 or more), and that the output buffer
address and its "unlimited" length cannot be added without overflowing.
An example of this can be found in inflate_fast():/* next_out is the output buffer address */
out = strm->next_out - OFF;
/* avail_out is the output buffer size. end will overflow if the output
* address is >= 0x80000104 */
end = out + (strm->avail_out - 257);This has huge consequences on the performance of kernel decompression,
since the following exit condition of inflate_fast() will be always true:} while (in < last && out < end);
Indeed, "end" has overflowed and is now always lower than "out". As a
result, inflate_fast() will return after processing one single byte of
input data, and will thus need to be called an unreasonably high number of
times. This probably went unnoticed because kernel decompression is fast
enough even with this issue.Nonetheless, adjusting the output buffer length in such a way that the
above pointer arithmetic never overflows results in a kernel decompression
that is about 3 times faster on affected machines.Signed-off-by: Alexandre Courbot
Tested-by: Jon Medhurst
Cc: Stephen Warren
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
14 Jan, 2011
4 commits
-
Check for end of the input buffer when skipping over the filename field in
the .gz file header.Signed-off-by: Lasse Collin
Cc: "H. Peter Anvin"
Cc: Alain Knaff
Cc: Albin Tonnerre
Cc: Phillip Lougher
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Currently users of mm.h need to include to use the macros
malloc() and free() provided by mm.h. This fixes it.Signed-off-by: Lasse Collin
Cc: "H. Peter Anvin"
Cc: Alain Knaff
Cc: Albin Tonnerre
Cc: Phillip Lougher
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
set_error_fn() has become a useless complication after c1e7c3ae59
("bzip2/lzma/gzip: pre-boot malloc doesn't return NULL on failure") fixed
the use of error() in malloc(). Only decompress_unlzma.c had some use for
it and that was easy to change too.This also gets rid of the static function pointer "error", which
should have been marked as __initdata.Signed-off-by: Lasse Collin
Cc: "H. Peter Anvin"
Cc: Alain Knaff
Cc: Albin Tonnerre
Cc: Phillip Lougher
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Signed-off-by: Lasse Collin
Cc: "H. Peter Anvin"
Cc: Alain Knaff
Cc: Albin Tonnerre
Cc: Phillip Lougher
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
24 Sep, 2009
1 commit
-
If the lzma/gzip decompressors are called with insufficient input data
(len > 0 & fill = NULL), they will attempt to call the fill function to
obtain more data, leading to a kernel oops.Signed-off-by: Phillip Lougher
Cc: "H. Peter Anvin"
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
08 Aug, 2009
2 commits
-
These includes were added by 079effb6933f34b9b1b67b08bd4fd7fb672d16ef
("kmemtrace, kbuild: fix slab.h dependency problem in
lib/decompress_inflate.c") to fix the build when using kmemtrace. However
this is not necessary when used to create a compressed kernel, and
actually creates issues (brings a lot of things unavailable in the
decompression environment), so don't include it if STATIC is defined.Signed-off-by: Albin Tonnerre
Cc: Sam Ravnborg
Cc: Russell King
Cc: Ingo Molnar
Cc: Thomas Gleixner
Cc: "H. Peter Anvin"
Cc: Pekka Enberg
Cc: Eduard - Gabriel Munteanu
Cc: Phillip Lougher
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Fix and improve comments in decompress/generic.h that describe the
decompressor API. Also remove an unused definition, and rename INBUF_LEN
in lib/decompress_inflate.c to conform to bzip2/lzma naming.Signed-off-by: Phillip Lougher
Cc: "H. Peter Anvin"
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
03 Apr, 2009
1 commit
-
Impact: fix build
lib/decompress_inflate.c depends on slab.h without including it:
CC lib/decompress_inflate.o
lib/decompress_inflate.c: In function ‘gunzip’:
lib/decompress_inflate.c:45: error: implicit declaration of function ‘kmalloc’
lib/decompress_inflate.c:45: warning: assignment makes pointer from integer without a cast
lib/decompress_inflate.c:57: warning: assignment makes pointer from integer without a cast
lib/decompress_inflate.c:65: warning: assignment makes pointer from integer without a cast
lib/decompress_inflate.c:71: warning: assignment makes pointer from integer without a cast
lib/decompress_inflate.c:154: error: implicit declaration of function ‘kfree’
make[1]: *** [lib/decompress_inflate.o] Error 1
make: *** [lib/] Error 2It gets included implicitly currently - but this will not be the
case with upcoming kmemtrace changes.Signed-off-by: Pekka Enberg
Cc: Eduard - Gabriel Munteanu
LKML-Reference:
Signed-off-by: Ingo Molnar
09 Jan, 2009
2 commits
-
Impact: Bug fix
Fix gunzip uncompression, so that it also works with files with
embedded filenames that are larger than one block.Signed-off-by: Alain Knaff
Signed-off-by: H. Peter Anvin -
Impact: Cleanup
Fix constant 0x8100 /* 32K */; according to Alain the value 0x8100 was
left over test code to test misalignment, the correct value is indeed
0x8000 == 32K.Signed-off-by: H. Peter Anvin
05 Jan, 2009
1 commit
-
Impact: Replaces inflate.c with a wrapper around zlib_inflate; new library code
This is the first part of the bzip2/lzma patch
The bzip patch is based on an idea by Christian Ludwig, includes support for
compressing the kernel with bzip2 or lzma rather than gzip. Both
compressors give smaller sizes than gzip. Lzma's decompresses faster
than bzip2.It also supports ramdisks and initramfs' compressed using these two
compressors.The functionality has been successfully used for a couple of years by
the udpcast projectThis version applies to "tip" kernel 2.6.28
This part contains:
- changed inflate.c to accomodate rest of patch
- implementation of bzip2 compression (not used at this stage yet)
- implementation of lzma compression (not used at this stage yet)
- Makefile routines to support bzip2 and lzma kernel compressionSigned-off-by: Alain Knaff
Signed-off-by: H. Peter Anvin