Param member of TIFFFaxTabEnt structure a uint16 to reduce size of
the binary. It happens that the Hylafax software uses the tables that
follow this typedef (TIFFFaxMainTable, TIFFFaxWhiteTable,
TIFFFaxBlackTable), also they are not in a public libtiff header.
Raised by Lee Howard.
Fixes http://bugzilla.maptools.org/show_bug.cgi?id=2636
limit the return number of inks to SamplesPerPixel, so that code that parses
ink names doesn't go past the end of the buffer.
Reported by Agostino Sarubbo.
Fixes http://bugzilla.maptools.org/show_bug.cgi?id=2599
Reported by Agostino Sarubbo.
instanciate compute ntrips as TIFFhowmany_32(td->td_imagelength, rowsperstrip),
instead of a logic based on the total size of data. Which is faulty is
the total size of data is not sufficient to fill the whole image, and thus
results in reading outside of the StripByCounts/StripOffsets arrays when
using TIFFReadScanline().
Reported by Agostino Sarubbo.
Fixes http://bugzilla.maptools.org/show_bug.cgi?id=2608.
* libtiff/tif_strip.c: revert the change in TIFFNumberOfStrips() done
for http://bugzilla.maptools.org/show_bug.cgi?id=2587 / CVE-2016-9273 since
the above change is a better fix that makes it unnecessary.
overflow on generation of PixarLog / LUV compressed files, with
ColorMap, TransferFunction attached and nasty plays with bitspersample.
The fix for LUV has not been tested, but suffers from the same kind
of issue of PixarLog.
Reported by Agostino Sarubbo.
Fixes http://bugzilla.maptools.org/show_bug.cgi?id=2604
64-bit compilation under Microsoft Visual Studio of the form
"Arithmetic overflow: 32-bit value is shifted, then cast to 64-bit
value. Results might not be an expected value.". Problem was
reported on November 16, 2016 on the tiff mailing list.
NULL pointer when values of tags with TIFF_SETGET_C16_ASCII / TIFF_SETGET_C32_ASCII
access are 0-byte arrays.
Fixes http://bugzilla.maptools.org/show_bug.cgi?id=2593 (regression introduced
by previous fix done on 2016-11-11 for CVE-2016-9297).
Reported by Henri Salo.
values of tags with TIFF_SETGET_C16_ASCII / TIFF_SETGET_C32_ASCII
access are null terminated, to avoid potential read outside buffer
in _TIFFPrintField().
Fixes http://bugzilla.maptools.org/show_bug.cgi?id=2590
have no TileOffsets/StripOffsets tag, when OJPEG compression is
disabled. Prevent null pointer dereference in TIFFReadRawStrip1()
and other functions that expect td_stripbytecount to be non NULL.
Fixes http://bugzilla.maptools.org/show_bug.cgi?id=2585
writeBufferToSeparateStrips(), writeBufferToContigTiles() and
writeBufferToSeparateTiles() that could cause heap buffer overflows.
Reported by Henri Salo from Nixu Corporation.
Fixes http://bugzilla.maptools.org/show_bug.cgi?id=2592
value when it is non-zero, instead of recomputing it. This is needed in
TIFF_STRIPCHOP mode where td_nstrips is modified. Fixes a read outsize of
array in tiffsplit (or other utilities using TIFFNumberOfStrips()).
Fixes http://bugzilla.maptools.org/show_bug.cgi?id=2587
Replace assertions by runtime checks to avoid assertions in debug mode,
or buffer overflows in release mode. Can happen when dealing with
unusual tile size like YCbCr with subsampling. Reported as MSVR 35105
by Axel Souchet & Vishal Chauhan from the MSRC Vulnerabilities & Mitigations
team.
SMaxSampleValue when they have been read and the value of
SamplesPerPixel is changed afterwards (like when reading a
OJPEG compressed image with a missing SamplesPerPixel tag,
and whose photometric is RGB or YCbCr, forcing SamplesPerPixel
being 3). Otherwise when rewriting the directory (for example
with tiffset, we will expect 3 values whereas the array had been
allocated with just one), thus causing a out of bound read access.
Fixes http://bugzilla.maptools.org/show_bug.cgi?id=2500
(CVE-2014-8127, duplicate: CVE-2016-3658)
* libtiff/tif_write.c: avoid null pointer dereference on td_stripoffset
when writing directory, if FIELD_STRIPOFFSETS was artificially set
for a hack case in OJPEG case.
Fixes http://bugzilla.maptools.org/show_bug.cgi?id=2500
(CVE-2014-8127, duplicate: CVE-2016-3658)
compressed images. Reported by Tyler Bohan of Cisco Talos as
TALOS-CAN-0187 / CVE-2016-5652.
Also prevents writing 2 extra uninitialized bytes to the file stream.
t2p_readwrite_pdf_image_tile(), causing crash, when reading a
JPEG compressed image with TIFFTAG_JPEGTABLES length being one.
Reported as MSVR 35101 by Axel Souchet and Vishal Chauhan from
the MSRC Vulnerabilities & Mitigations team.
required tags. Found on test case of MSVR 35100.
* tools/tiffcrop.c: fix read of undefined buffer in
readContigStripsIntoBuffer() due to uint16 overflow. Probably not a
security issue but I can be wrong. Reported as MSVR 35100 by Axel
Souchet from the MSRC Vulnerabilities & Mitigations team.
read floating point images.
* libtiff/tif_predict.c (PredictorSetup): Enforce bits-per-sample
requirements of floating point predictor (3). Fixes CVE-2016-3622
"Divide By Zero in the tiff2rgba tool."
places where it isn't done currently, but it seems this patch is enough.
in heap or stack allocated buffers. Reported as MSVR 35093,
MSVR 35096 and MSVR 35097. Discovered by Axel Souchet and Vishal
Chauhan from the MSRC Vulnerabilities & Mitigations team.
* tools/tiff2pdf.c: fix out-of-bounds write vulnerabilities in
heap allocate buffer in t2p_process_jpeg_strip(). Reported as MSVR
35098. Discovered by Axel Souchet and Vishal Chauhan from the MSRC
Vulnerabilities & Mitigations team.
* libtiff/tif_pixarlog.c: fix out-of-bounds write vulnerabilities
in heap allocated buffers. Reported as MSVR 35094. Discovered by
Axel Souchet and Vishal Chauhan from the MSRC Vulnerabilities &
Mitigations team.
* libtiff/tif_write.c: fix issue in error code path of TIFFFlushData1()
that didn't reset the tif_rawcc and tif_rawcp members. I'm not
completely sure if that could happen in practice outside of the odd
behaviour of t2p_seekproc() of tiff2pdf). The report points that a
better fix could be to check the return value of TIFFFlushData1() in
places where it isn't done currently, but it seems this patch is enough.
Reported as MSVR 35095. Discovered by Axel Souchet & Vishal Chauhan &
Suha Can from the MSRC Vulnerabilities & Mitigations team.
if more input samples are provided than expected by PixarLogSetupEncode.
Idea based on libtiff-CVE-2016-3990.patch from
libtiff-4.0.3-25.el7_2.src.rpm by Nikola Forro, but with different and
simpler check. (bugzilla #2544)
invalid tests that rejected valid files. (bugzilla #2545)
buffer, when -b mode is enabled, that could result in out-of-bounds
write. Based initially on patch tiff-CVE-2016-3945.patch from
libtiff-4.0.3-25.el7_2.src.rpm by Nikola Forro, with correction for
invalid tests that rejected valid files.
on a tiled separate TIFF with more than 8 samples per pixel.
Reported by Kaixiang Zhang of the Cloud Security Team, Qihoo 360
(CVE-2016-5321, bugzilla #2558)
TIFFReadEncodedTile() directly use user provided buffer when
no compression (and other conditions) to save a memcpy().
* libtiff/tif_write.c: make TIFFWriteEncodedStrip() and
TIFFWriteEncodedTile() directly use user provided buffer when
no compression to save a memcpy().
PHOTOMETRIC_LOGL, there is only one sample per pixel. Avoid
potential invalid memory write on corrupted/unexpected images when
using the TIFFRGBAImageBegin() interface (reported by
Clay Wood)
(CVE-2016-587)
ras2tiff, sgi2tiff, sgisv, and ycbcr are completely removed from
the distribution. The libtiff tools rgb2ycbcr and thumbnail are
only built in the build tree for testing. Old files are put in
new 'archive' subdirectory of the source repository, but not in
distribution archives. These changes are made in order to lessen
the maintenance burden.
functions in non debug builds by replacing assert()s by regular if
checks (bugzilla #2522).
Fix potential out-of-bound reads in case of short input data.
interface in case of unsupported values of SamplesPerPixel/ExtraSamples
for LogLUV / CIELab. Add explicit call to TIFFRGBAImageOK() in
TIFFRGBAImageBegin(). Fix CVE-2015-8665 reported by limingxing and
CVE-2015-8683 reported by zzf of Alibaba.
should return -1 in case of failure of tif_encodestrip() as documented
* libtiff/tif_dumpmode.c: DumpModeEncode() should return 0 in case of
failure so that the above mentionned functions detect the error.
Roger Leigh (via tiff mailing list on 2015-09-01) to fix issue
with BSD make and to make use of cmake in 'distcheck' target
conditional on if cmake is available.
Roger Leigh (via tiff mailing list on 2015-09-01).
CMake build is now included in 'distcheck' target.
Builds with CMake 2.8.9 and newer.
Tar is now resquested to use POSIX PAX format.
Roger Leigh (via tiff mailing list on 2015-08-31.
CMake reads all version information directly from configure.ac to
avoid duplication of values. This basically greps over the file
for the LIBTIFF_* variables, then translates them to the form
needed for cmake. This includes the release version and libtool
shared library version information.
Make shared/static library building configurable. Currently it
always builds shared libraries, with static libs having a _static
suffix (copying zlib, but it means it's got a non-standard name).
CMake has a -DBUILD_SHARED_LIBS=ON|OFF option to select one or the
other, which is now used instead. There's now a single "tiff"
target to build either shared or static as required, and all the
tests and tools are linked with this. Note: the Windows tests fail
when linked with a static libtiff (says: libtiff.dll not found).
Not really a regression since this was not tested up to this
point, and it's likely the unit tests haven't (ever?) been run on
Windows with a static libtiff, so there's some additional
portability issue here to address. Works fine on UNIX systems,
and fine on Windows with the default to build a DLL.
Add a missing file which wasn't being distributed, causing unit
tests to fail. Note that "find . -name '*.cmake'" lists all the
CMake files which need distributing in addition to all the
CMakeLists.txt files (which now are distributed).
or 32 bit integer, when byte swapping is needed, in
horizontal predictor (#2521). Also fixes decoding when there is
a single pixel to code (unlikely case...) and byte swapping is
involved.
add explicit masking with 0xff before casting
to unsigned char (make icc -check=conversions happy)
* libtiff/tif_predict.c: operate on unsigned datatypes when
computing/applying differences to avoid undefined behaviour of
signed types (C standard compliance)
by Roger Leigh posted to tiff list on Wed, 1 Jul 2015 15:58:20
+0100.
* cmake: Add extra warning flags. Patch #2 of 3 by Roger Leigh
posted to tiff list on Wed, 1 Jul 2015 15:58:20 +0100.
* cmake: Correct snprintf fallback for VS2015. Patch #1 of 3 by
Roger Leigh posted to tiff list on Wed, 1 Jul 2015 15:58:20 +0100.
libtiff mailing list on Mon, 22 Jun 2015 21:21:01 +0100. Several
corrections to ensure that the autotools build still works were
added by me. I have not yet tested the build using 'cmake' or
MSVC with 'nmake'.
add-on library.
* test/Makefile.am (JPEG_DEPENDENT_CHECK_PROG): raw_decode
requires JPEG support to compile. Use Automake conditional to
only include it when JPEG support is available.
definitions that configure produces, including for WIN64. Still
needs to be tested.
'lld' is not assured by the run-time DLLs and so GCC warns.
Add TIFF_SIZE_T and TIFF_SIZE_FORMAT to provide a type definition
and printf format specifier to deal with printing values of
'size_t' type. In particular, this was necessary for WIN64.
Added a configure test for if the system headers provide 'optarg'
(normal case) and block out the many explicit 'extern' statements
in the utilities. This was found to be necessary under Windows
when getopt is in a DLL and the symbols are already imported with
dllimport via standard header files.
