File libjpeg-turbo.changes of Package libjpeg-turbo
453
1
-------------------------------------------------------------------
2
Thu Jun 3 08:40:46 UTC 2021 - pgajdos@suse.com
3
4
- security update
5
- added patches
6
fix CVE-2020-17541 [bsc#1186764], stack-based buffer overflow in the "transform" component
7
+ libjpeg-turbo-CVE-2020-17541.patch
8
9
-------------------------------------------------------------------
10
Mon Jun 8 11:49:53 UTC 2020 - pgajdos@suse.com
11
12
- security update
13
- added patches
14
fix CVE-2020-13790 [bsc#1172491], heap-based buffer over-read in get_rgb_row() in rdppm.c via a malformed PPM input file
15
+ libjpeg-turbo-CVE-2020-13790.patch
16
17
-------------------------------------------------------------------
18
Wed Nov 13 07:55:36 UTC 2019 - pgajdos@suse.com
19
20
- security update [bsc#1156402]
21
- modified patches
22
% libjpeg-turbo-CVE-2019-2201.patch (use upstream one)
23
24
-------------------------------------------------------------------
25
Tue Nov 12 09:33:44 UTC 2019 - pgajdos@suse.com
26
27
- security update
28
- added patches
29
CVE-2019-2201 [bsc#1156402]
30
+ libjpeg-turbo-CVE-2019-2201.patch
31
32
-------------------------------------------------------------------
33
Wed Mar 13 13:02:11 UTC 2019 - pgajdos@suse.com
34
35
- security update
36
- added patches
37
CVE-2018-14498 [bsc#1128712]
38
+ libjpeg-turbo-CVE-2018-14498.patch
39
40
-------------------------------------------------------------------
41
Mon Dec 3 10:25:36 UTC 2018 - Petr Gajdos <pgajdos@suse.com>
42
43
- asan_build: build ASAN included
44
- debug_build: build more suitable for debugging
45
- disable silent rules
46
47
-------------------------------------------------------------------
48
Tue Jun 19 13:44:00 UTC 2018 - pgajdos@suse.com
49
50
- security update:
51
* CVE-2018-1152 [bsc#1098155]
52
+ libjpeg-turbo-CVE-2018-1152.patch
53
54
-------------------------------------------------------------------
55
Tue Jun 12 13:34:11 UTC 2018 - pgajdos@suse.com
56
57
- security update:
58
* CVE-2018-11813 [bsc#1096209]
59
+ libjpeg-turbo-CVE-2018-11813.patch
60
* remove redundant libjpeg-turbo-CVE-2017-15232.patch
61
[bsc#1062937#c17]
62
63
-------------------------------------------------------------------
64
Mon Dec 18 13:29:30 UTC 2017 - pgajdos@suse.com
65
66
- - update to version 1.5.3
67
1. Fixed a NullPointerException in the TurboJPEG Java wrapper that occurred
68
when using the YUVImage constructor that creates an instance backed by separate
69
image planes and allocates memory for the image planes.
70
2. Fixed an issue whereby the Java version of TJUnitTest would fail when
71
testing BufferedImage encoding/decoding on big endian systems.
72
3. Fixed a segfault in djpeg that would occur if an output format other than
73
PPM/PGM was selected along with the `-crop` option. The `-crop` option now
74
works with the GIF and Targa formats as well (unfortunately, it cannot be made
75
to work with the BMP and RLE formats due to the fact that those output engines
76
write scanlines in bottom-up order.) djpeg will now exit gracefully if an
77
output format other than PPM/PGM, GIF, or Targa is selected along with the
78
`-crop` option.
79
4. Fixed an issue whereby `jpeg_skip_scanlines()` would segfault if color
80
quantization was enabled.
81
5. TJBench (both C and Java versions) will now display usage information if any
82
command-line argument is unrecognized. This prevents the program from silently
83
ignoring typos.
84
6. Fixed an access violation in tjbench.exe (Windows) that occurred when the
85
program was used to decompress an existing JPEG image.
86
7. Fixed an ArrayIndexOutOfBoundsException in the TJExample Java program that
87
occurred when attempting to decompress a JPEG image that had been compressed
88
with 4:1:1 chrominance subsampling.
89
8. Fixed an issue whereby, when using `jpeg_skip_scanlines()` to skip to the
90
end of a single-scan (non-progressive) image, subsequent calls to
91
`jpeg_consume_input()` would return `JPEG_SUSPENDED` rather than
92
`JPEG_REACHED_EOI`.
93
9. `jpeg_crop_scanlines()` now works correctly when decompressing grayscale
94
JPEG images that were compressed with a sampling factor other than 1 (for
95
instance, with `cjpeg -grayscale -sample 2x2`).
96
97
-------------------------------------------------------------------
98
Thu Oct 12 10:59:03 UTC 2017 - pgajdos@suse.com
99
100
- security update:
101
* CVE-2017-15232 [bsc#1062937]
102
+ libjpeg-turbo-CVE-2017-15232.patch
103
104
-------------------------------------------------------------------
105
Thu Oct 12 10:22:05 UTC 2017 - pgajdos@suse.com
106
107
- Update to version 1.5.2
108
+ Fixed several memory leaks in the TurboJPEG API library that
109
could occur if the library was built with certain compilers
110
and optimization levels.
111
+ The libjpeg-turbo memory manager will now honor the
112
max_memory_to_use structure member in jpeg_memory_mgr,
113
which can be set to the maximum amount of memory (in bytes)
114
that libjpeg-turbo should use during decompression or
115
multi-pass (including progressive) compression. This limit
116
can also be set using the JPEGMEM environment variable or
117
using the -maxmemory switch in cjpeg/djpeg/jpegtran.
118
+ TJBench will now run each benchmark for 1 second prior to
119
starting the timer, in order to improve the consistency of
120
the results. Furthermore, the -warmup option is now used to
121
specify the amount of warmup time rather than the number of
122
warmup iterations.
123
+ Fixed an error (short jump is out of range) that occurred
124
when assembling the 32-bit x86 SIMD extensions with NASM
125
versions prior to 2.04.
126
+ Fixed a regression introduced by 1.5 beta1[11] that prevented
127
the Java version of TJBench from outputting any reference images
128
(the -nowrite switch was accidentally enabled by default.)
129
libjpeg-turbo should now build and run with full AltiVec SIMD
130
acceleration on PowerPC-based AmigaOS 4 and OpenBSD systems.
131
132
-------------------------------------------------------------------
133
Wed Jan 18 10:07:00 UTC 2017 - bwiedemann@suse.com
134
135
- set build date to enable reproducible builds
136
137
-------------------------------------------------------------------
138
Wed Sep 21 10:50:36 UTC 2016 - idonmez@suse.com
139
140
- Update to version 1.5.1 fate#324061
141
+ Fix for PowerPC platforms lacking AltiVec instructions
142
+ Fix ABI problem with clang/llvm on aarch64.
143
+ Fancy upsampling is now supported when decompressing JPEG
144
images that use 4:4:0 (h1v2) chroma subsampling.
145
+ If merged upsampling isn't SIMD-accelerated but YCbCr-to-RGB
146
conversion is, then libjpeg-turbo will now disable merged
147
upsampling when decompressing YCbCr JPEG images into RGB
148
or extended RGB output images. This significantly speeds up
149
the decompression of 4:2:0 and 4:2:2 JPEGs on ARM platforms
150
if fancy upsampling is not used
151
(for example, if the -nosmooth option to djpeg is specified.)
152
+ The TurboJPEG API will now decompress 4:2:2 and 4:4:0 JPEG
153
images with 2x2 luminance sampling factors and 2x1 or 1x2
154
chrominance sampling factors.
155
+ Fixed an unsigned integer overflow in the libjpeg memory manager.
156
+ Fixed additional negative left shifts and other issues reported
157
by the GCC and Clang undefined behavior sanitizers when
158
attempting to decompress specially-crafted malformed JPEG
159
images. None of these issues posed a security threat, but
160
removing the warnings makes it easier to detect actual
161
security issues, should they arise in the future.
162
+ Fixed an out-of-bounds array reference, introduced by
163
1.4.902 and detected by the Clang undefined behavior sanitizer,
164
that could be triggered by a specially-crafted malformed
165
JPEG image with more than four components. Because the
166
out-of-bounds reference was still within the same structure,
167
it was not known to pose a security threat, but removing
168
the warning makes it easier to detect actual security issues,
169
should they arise in the future.
170
171
-------------------------------------------------------------------
172
Wed Jun 8 07:53:26 UTC 2016 - idonmez@suse.com
173
174
- Update to version 1.5.0
175
+ Fixed an issue whereby a malformed motion-JPEG frame could
176
cause the "fast path" of libjpeg-turbo's Huffman decoder to
177
read from uninitialized memory.
178
+ Added libjpeg-turbo version and build information to the global
179
string table of the libjpeg and TurboJPEG API libraries.
180
+ Fixed a couple of issues in the PPM reader that would cause
181
buffer overruns in cjpeg if one of the values in a binary
182
PPM/PGM input file exceeded the maximum value defined in the
183
file's header. libjpeg-turbo 1.4.2 already included a similar
184
fix for ASCII PPM/PGM files. Note that these issues were not
185
security bugs, since they were confined to the cjpeg program
186
and did not affect any of the libjpeg-turbo libraries.
187
+ Fixed an issue whereby attempting to decompress a JPEG file with
188
a corrupt header using the tjDecompressToYUV2() function would
189
cause the function to abort without returning an error and,
190
under certain circumstances, corrupt the stack. This only
191
occurred if tjDecompressToYUV2() was called prior to calling
192
tjDecompressHeader3(), or if the return value from
193
tjDecompressHeader3() was ignored (both cases represent
194
incorrect usage of the TurboJPEG API.)
195
+ The jpeg_stdio_src(), jpeg_mem_src(), jpeg_stdio_dest(),
196
and jpeg_mem_dest() functions in the libjpeg API will now
197
throw an error if a source/destination manager has already
198
been assigned to the compress or decompress object by a
199
different function or by the calling program.
200
201
-------------------------------------------------------------------
202
Thu Oct 8 07:53:53 UTC 2015 - idonmez@suse.com
203
204
- Update to version 1.4.2
205
+ Crash fixes
206
+ clang compatibility fixes
207
+ See the included ChangeLog.txt for the details
208
- Drop libjpeg-turbo-1.4.0-int32.patch, not needed anymore.
209
- Drop libjpeg-turbo-remove-test.patch, fixed upstream.
210
211
-------------------------------------------------------------------
212
Thu Mar 5 11:43:27 UTC 2015 - jengelh@inai.de
213
214
- Remove useless same-name provides. Use download URLs not
215
dependent on directory structure.
216
217
-------------------------------------------------------------------
218
Mon Mar 2 16:22:23 UTC 2015 - normand@linux.vnet.ibm.com
219
220
- Remove float tests with new libjpeg-turbo-remove-test.patch
221
same as Fedora bug 1161585 related to upstream issue
222
https://sourceforge.net/p/libjpeg-turbo/bugs/83/
223
224
-------------------------------------------------------------------
225
Sat Jan 10 00:39:57 UTC 2015 - p.drouand@gmail.com
226
227
- Update to version 1.4.0
228
+ Fixed a build issue on OS X PowerPC platforms (md5cmp failed to build
229
because OS X does not provide the le32toh() and htole32() functions.)
230
+ The non-SIMD RGB565 color conversion code did not work correctly on big
231
endian machines. This has been fixed.
232
+ Fixed an issue in tjPlaneSizeYUV() whereby it would erroneously return 1
233
instead of -1 if componentID was > 0 and subsamp was TJSAMP_GRAY.
234
+ Fixed an issue in tjBufSizeYUV2() wherby it would erroneously return 0
235
instead of -1 if width was < 1.
236
+ The Huffman encoder now uses clz and bsr instructions for bit counting on
237
ARM64 platforms (see 1.4 beta1 [5].)
238
+ The close() method in the TJCompressor and TJDecompressor Java classes is
239
now idempotent. Previously, that method would call the native tjDestroy()
240
function even if the TurboJPEG instance had already been destroyed. This
241
caused an exception to be thrown during finalization, if the close() method had
242
already been called. The exception was caught, but it was still an expensive
243
operation.
244
+ The TurboJPEG API previously generated an error ("Could not determine
245
subsampling type for JPEG image") when attempting to decompress grayscale JPEG
246
images that were compressed with a sampling factor other than 1 (for instance,
247
with 'cjpeg -grayscale -sample 2x2'). Subsampling technically has no meaning
248
with grayscale JPEGs, and thus the horizontal and vertical sampling factors
249
for such images are ignored by the decompressor. However, the TurboJPEG API
250
was being too rigid and was expecting the sampling factors to be equal to 1
251
before it treated the image as a grayscale JPEG.
252
+ cjpeg, djpeg, and jpegtran now accept an argument of -version, which will
253
print the library version and exit.
254
+ Referring to 1.4 beta1 [15], another extremely rare circumstance was
255
discovered under which the Huffman encoder's local buffer can be overrun
256
when a buffered destination manager is being used and an
257
extremely-high-frequency block (basically junk image data) is being encoded.
258
Even though the Huffman local buffer was increased from 128 bytes to 136 bytes
259
to address the previous issue, the new issue caused even the larger buffer to
260
be overrun. Further analysis reveals that, in the absolute worst case (such as
261
setting alternating AC coefficients to 32767 and -32768 in the JPEG scanning
262
order), the Huffman encoder can produce encoded blocks that approach double the
263
size of the unencoded blocks. Thus, the Huffman local buffer was increased to
264
256 bytes, which should prevent any such issue from re-occurring in the future.
265
+ The new tjPlaneSizeYUV(), tjPlaneWidth(), and tjPlaneHeight() functions
266
were not actually usable on any platform except OS X and Windows, because
267
those functions were not included in the libturbojpeg mapfile. This has been
268
fixed.
269
+ Restored the JPP(), JMETHOD(), and FAR macros in the libjpeg-turbo header
270
files. The JPP() and JMETHOD() macros were originally implemented in libjpeg
271
as a way of supporting non-ANSI compilers that lacked support for prototype
272
parameters. libjpeg-turbo has never supported such compilers, but some
273
software packages still use the macros to define their own prototypes.
274
Similarly, libjpeg-turbo has never supported MS-DOS and other platforms that
275
have far symbols, but some software packages still use the FAR macro. A pretty
276
good argument can be made that this is a bad practice on the part of the
277
software in question, but since this affects more than one package, it's just
278
easier to fix it here.
279
+ Fixed issues that were preventing the ARM 64-bit SIMD code from compiling
280
for iOS, and included an ARMv8 architecture in all of the binaries installed by
281
the "official" libjpeg-turbo SDK for OS X.
282
- Adapt patches to upstream changes
283
libjpeg-ocloexec.patch > libjpeg-1.4.0-ocloexec.patch
284
libjpeg-turbo-1.3.0-int32.patch > libjpeg-turbo-1.4.0-int32.patch
285
- Remove libjpeg-turbo-CVE-2014-9092.patch; fixed on upstream release
286
- Bump tminor to 1
287
288
-------------------------------------------------------------------
289
Thu Nov 27 09:50:00 UTC 2014 - pgajdos@suse.com
290
291
- security update CVE-2014-9092 [bnc#906761]
292
* added libjpeg-turbo-CVE-2014-9092.patch
293
294
-------------------------------------------------------------------
295
Wed Oct 15 11:39:09 UTC 2014 - olaf@aepfle.de
296
297
- Obsolete jpeg version 6b, 8.0.1 and 8.0.2 to avoid conflicts
298
during zypper dup
299
300
-------------------------------------------------------------------
301
Mon Mar 31 09:29:11 UTC 2014 - pgajdos@suse.com
302
303
- update to 1.3.1:
304
* Fixed a bug whereby attempting to encode a progressive JPEG
305
with arithmetic entropy coding (by passing arguments of
306
-progressive -arithmetic to cjpeg or jpegtran, for instance)
307
would result in an error, "Requested feature was omitted at
308
compile time".
309
* Fixed a couple of issues whereby malformed JPEG images would
310
cause libjpeg-turbo to use uninitialized memory during
311
decompression.
312
* Fixed an error ("Buffer passed to JPEG library is too small")
313
that occurred when calling the TurboJPEG YUV encoding function
314
with a very small (< 5x5) source image, and added a unit test
315
to check for this error.
316
* etc. see ChangeLog.txt
317
318
-------------------------------------------------------------------
319
Thu Feb 20 21:16:13 UTC 2014 - tchvatal@suse.com
320
321
- Provide right name for turbo library 32bit generating.
322
323
-------------------------------------------------------------------
324
Fri Feb 7 13:55:46 UTC 2014 - tchvatal@suse.com
325
326
- Require libturbojpeg in the devel package.
327
Otherwise we just have dangling symlink there.
328
329
-------------------------------------------------------------------
330
Tue Sep 10 12:42:45 UTC 2013 - pgajdos@suse.com
331
332
- update do 1.3.0:
333
* Fixed a Huffman encoder bug that prevented I/O suspension from
334
working properly.
335
* Added support for additional scaling factors (3/8, 5/8, 3/4,
336
7/8, 9/8, 5/4, 11/8, 3/2, 13/8, 7/4, 15/8, and 2) when
337
decompressing. Note that the IDCT will not be SIMD-accelerated
338
when using any of these new scaling factors.
339
* The tjDecompressToYUV() function now supports the TJFLAG_FASTDCT
340
flag.
341
* cjpeg can now be used to generate JPEG files with the RGB
342
colorspace (feature ported from jpeg-8d.)
343
* etc. see ChangeLog.txt
344
345
-------------------------------------------------------------------
346
Mon May 27 18:21:55 UTC 2013 - crrodriguez@opensuse.org
347
348
- Build with full RELRO as this library is exposed to
349
possible malicious images.
350
351
-------------------------------------------------------------------
352
Fri May 17 07:28:47 UTC 2013 - pgajdos@suse.com
353
354
- libjpeg62-turbo package created [bnc#807183]
355
356
-------------------------------------------------------------------
357
Mon Aug 13 17:05:35 UTC 2012 - dmueller@suse.com
358
359
- selfconflicts are not possible, remove it
360
361
-------------------------------------------------------------------
362
Mon Jul 23 11:33:56 UTC 2012 - pgajdos@suse.com
363
364
- Update to version 1.2.1:
365
* fixed heap overflow [bnc#771791]
366
367
-------------------------------------------------------------------
368
Thu Mar 22 13:25:09 UTC 2012 - idonmez@suse.com
369
370
- Update to version 1.2.0
371
* Fixed out-of-bounds read in SSE2 SIMD code
372
* Added a compile-time macro (LIBJPEG_TURBO_VERSION) that can
373
be used to check the version of libjpeg-turbo against which
374
an application was compiled.
375
* Added new RGBA/BGRA/ABGR/ARGB colorspace extension constants
376
* libjpeg-turbo will now correctly decompress erroneous
377
CMYK/YCCK JPEGs whose K component is assigned a component ID
378
of 1 instead of 4.
379
* Added SIMD routines for RGB-to-grayscale color conversion
380
* Improved the performance of the C color conversion routines
381
* Added a function to the TurboJPEG API that performs lossless
382
transforms.
383
* Added support for 4:4:0 (transposed 4:2:2) subsampling
384
385
-------------------------------------------------------------------
386
Sat Nov 19 20:38:03 UTC 2011 - coolo@suse.com
387
388
- add libtool as buildrequire to avoid implicit dependency
389
390
-------------------------------------------------------------------
391
Sat Nov 12 22:54:58 UTC 2011 - crrodriguez@opensuse.org
392
393
- Open all file descriptors with O_CLOEXEC, extended description
394
in the patch file.
395
396
-------------------------------------------------------------------
397
Thu May 19 14:44:56 CEST 2011 - pgajdos@suse.cz
398
399
- updated to 1.1.1:
400
* Fixed a 1-pixel error in row 0, column 21 of the luminance
401
plane generated by tjEncodeYUV().
402
* libjpeg-turbo's accelerated Huffman decoder previously
403
ignored unexpected markers found in the middle of the
404
JPEG data stream during decompression. It will now
405
hand off decoding of a particular block to the unaccelerated
406
Huffman decoder if an unexpected marker is found, so that
407
the unaccelerated Huffman decoder can generate an appropriate
408
warning.
409
* Fixed a bug in jpeg_read_coefficients() whereby it would
410
not initialize cinfo->image_width and cinfo->image_height
411
if libjpeg v7 or v8 emulation was enabled. This specifically
412
caused the jpegoptim program to fail if it was linked against
413
a version of libjpeg-turbo that was built with libjpeg v7 or
414
v8 emulation.
415
* Eliminated excessive I/O overhead that occurred when reading
416
BMP files in cjpeg.
417
*
418
419
-------------------------------------------------------------------
420
Thu Mar 3 13:53:18 CET 2011 - pgajdos@suse.cz
421
422
- updated to 1.1.0:
423
* Added further protections against invalid Huffman codes.
424
* Added an extended version of tjDecompressHeader().
425
* Added arithmetic encoding and decoding support.
426
* TurboJPEG/OSS can now compress from/decompress to
427
grayscale bitmaps.
428
* Added emulation of the libjpeg v7 and v8 APIs and ABIs.
429
* Added two new TurboJPEG API functions, tjEncodeYUV() and
430
tjDecompressToYUV().
431
* The TurboJPEG dynamic library now uses versioned symbols.
432
* Fixed visual artifacts in grayscale JPEG compression
433
caused by a typo in the RGB-to-chrominance lookup tables.
434
- see ChangeLog.txt for details.
435
- removed upstreamed jpegtran.patch and rh639672.patch
436
437
-------------------------------------------------------------------
438
Tue Dec 14 14:32:56 CET 2010 - pgajdos@suse.cz
439
440
- spec file cleanup
441
442
-------------------------------------------------------------------
443
Fri Dec 10 13:02:58 UTC 2010 - pgajdos@novell.com
444
445
- this jpeg version will be the default to the prejudice of jpeg8
446
from now on
447
448
-------------------------------------------------------------------
449
Sun Nov 7 12:53:26 UTC 2010 - prusnak@opensuse.org
450
451
- created package based on Fedora one (v 1.0.1)
452
453