10af2c2ddb
Merge branch 'master' into add-mapimage
2019-08-11 17:30:10 +01:00
e16fa657eb
Revert "Revert "try disabling IM for oss-fuzz""
...
This reverts commit 263f69d581
2019-08-11 12:41:38 +01:00
df9986489a
Revert "Revert "fix build with IM disabled""
...
This reverts commit 4dbf2167d0
2019-08-11 12:32:46 +01:00
263f69d581
Revert "try disabling IM for oss-fuzz"
...
This reverts commit faafd4eee0
2019-08-11 12:31:58 +01:00
4dbf2167d0
Revert "fix build with IM disabled"
...
This reverts commit 88b3aa8f2b
2019-08-11 12:31:39 +01:00
88b3aa8f2b
fix build with IM disabled
2019-08-10 13:37:31 +01:00
faafd4eee0
try disabling IM for oss-fuzz
...
libvips has to use ReadImage() when it opens an image via MagickCore,
and this can allocate very large amounts of memory, killing the fuzzer.
IM is already part of oss-fuzz, so we are not gaining much by
replicating that work. This patch attempts to disable IM for fuzzing tests.
2019-08-10 12:58:21 +01:00
d81efda782
better blocking of large input images in the fuzzer
...
things like 65494x5 pixel PNGs were being allowed through and caused
timeouts
2019-08-09 16:45:08 +01:00
765a416ca7
block huge tile_size in tiff load
...
fuzzed tiffs can give huge and mysterious values for tile_size
2019-08-09 16:32:35 +01:00
ccea473b0e
fix logic for page_height and n_pages getters
2019-08-09 15:46:56 +01:00
8b1817ddd0
fix a couple of clang warnings
2019-08-09 15:22:56 +01:00
7ef0ae826f
try suppressing libexif ubsan again
2019-08-09 13:49:15 +01:00
7c94d09ab6
revert ubsan supp file
...
nope, fails on travis
2019-08-09 13:15:54 +01:00
e75320bb0d
try enabling ubsan suppressions in travis
2019-08-09 13:03:14 +01:00
41ecc696a7
disable asan and ubsan supp files for now
...
getting strange parse errors on asan
2019-08-09 12:28:15 +01:00
cf94c2f71e
more suppressions to a subdir
...
since we now have four of them
2019-08-09 11:43:13 +01:00
5a3b1d7f91
break ubsan and asan suppressions out
...
they need to be in separate files
2019-08-09 11:40:51 +01:00
6eaa0bc4c4
try to suppress libexif errors from fuzz
2019-08-09 10:47:45 +01:00
d6baf157cf
ignore buffer overflow in magick
...
for lsan testing
2019-08-09 10:22:47 +01:00
cdee0feb93
Merge branch '8.8'
2019-08-09 10:12:55 +01:00
aafc2c7948
add changelog note for profiles.c fix
2019-08-09 10:11:45 +01:00
31f7421b11
Merge branch 'msvc-friendly-profiles' into 8.8
2019-08-09 10:10:49 +01:00
988f9594f4
Merge branch '8.8'
2019-08-08 16:11:02 +01:00
ff58c67e33
add locks for pdfium load
...
We used to lock within documents, ie. we did not allow two threads to
work on the same file.
However pdfium is not threadsafe in any way, and this is not supported,
see:
https://groups.google.com/forum/#!msg/pdfium/kyIdh_J4csg/K1LvfPiHDwAJ
This patch adds locks around pdfium calls.
see:
https://github.com/libvips/libvips/issues/1380
https://github.com/libvips/libvips/issues/1275
2019-08-08 15:54:16 +01:00
d6c8879f3d
NOte min size for find_trim
...
See https://github.com/libvips/libvips/issues/1392
2019-08-08 14:55:19 +01:00
d2d63f4ac4
revise profile_load
...
seems to work!
2019-08-08 13:23:28 +01:00
9cc72ea1c1
msvc dies with string literals larger than 64kb
...
use arrays instead of tseing literals
generates the literals, profile_load.c needs to be adjusted still
2019-08-08 11:29:39 +01:00
43d678c145
Merge pull request #1385 from DarthSim/feature/optimize_gif
...
Add GIF optimization options to …
2019-08-07 22:32:48 +01:00
c7e0c073ca
Add `optimize_gif_frames` and `optimize_gif_transparency` options to `vips_magicksave`
2019-08-08 02:52:54 +06:00
def0e73449
Merge branch '8.8'
2019-08-07 20:29:37 +01:00
675c150500
fix build with IM7
2019-08-07 20:29:19 +01:00
1a659e3860
clarify comment
2019-08-07 17:13:25 +01:00
7abe129ac4
don't invert signed ints in tiffload
2019-08-07 17:08:18 +01:00
7250b169d4
Merge branch '8.8'
2019-08-07 16:19:23 +01:00
f36927e78d
fix build against graphicsmagick
...
- add tests for more features missing in GM
- simplify the IM/GM configure tests
- revise the test suite to pass with GM and IM
2019-08-07 16:17:27 +01:00
5ddc154a81
prevent int overflow for int pixels in shrinkh
...
by using double for an accumulator
2019-08-06 17:00:53 +01:00
6ba34e479e
better shrinkv for int32 types
...
use double as the sum type to prevent int overflow
2019-08-06 16:55:10 +01:00
25b6f2fcf1
Merge branch 'revise-pthread'
2019-08-06 16:25:16 +01:00
fb92783659
musl is now 128kb, so make the comment vague
...
36676c59fc (commitcomment-34589167)
2019-08-06 15:53:38 +01:00
5649f6d865
update for "make dist"
2019-08-06 15:33:40 +01:00
36676c59fc
small polish to pthread discovery
...
- use AC_CHECK_FUNC() not AC_CHECK_FUNCS()
- safer names for save vars
2019-08-06 14:56:05 +01:00
98f3bf78a5
fix doc spelling of "optimise"
2019-08-06 14:46:43 +01:00
8b76606a23
Merge pull request #1387 from omira-sch/fuzz-corpus
...
Add oss-fuzz fixed testcases for regression testing
2019-08-06 14:13:58 +01:00
4304ef35ee
add already fixed testcases to fuzz corpus
2019-08-06 13:37:41 +02:00
7c8af4e78a
remove stale file
2019-08-06 13:37:41 +02:00
2211565eee
fix .gitignore
2019-08-06 13:37:41 +02:00
fb0069528d
Merge pull request #1386 from omira-sch/fuzz-improve-coverage
...
add smaller images suitable for fuzzing
2019-08-06 12:18:51 +01:00
e14a9b9221
add small images suitable for fuzzing
2019-08-06 12:29:34 +02:00
bdc61d5675
layout
2019-08-05 13:53:00 +01:00
0d2c14c967
try to improve pthread discovery
...
we were not using all the flags we should when testing for
pthread_setattr_default_np() ... place the test line after GTHREAD and
use those flags for linking
see https://github.com/libvips/libvips/issues/1384
thanks elfring
2019-08-05 13:35:47 +01:00
John Cupitt
DarthSim
Oscar Mira