Revisions of libopenmpt
Dominique Leuenberger (dimstar_suse)
accepted
request 1188976
from
Mia Herkt (mia)
(revision 49)
- Update to 0.7.9: * [Sec] Potential division by 0 when seeking in the module with seek.sync_samples enabled (r21167). * [Change] The work-around for https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115049 has been changed from forcing -O1 on GCC 14 to setting -fno-ipa-ra on all GCC versions on non-ELF platforms. * MOD: Allow sample swapping to work when swapping from a non-looping, stopped sample back to a looping sample (fixes MOD.energy). * DBM: Import second sustain point in case the first sustain point is not set, or if it has a lower index than the first. * DBM: When several instruments referenced the same sample with different properties (volume, loop points, etc.), only one set of properties was imported (fixes DBM.Supernova). * DBM: Prioritize effects more correctly when the same effect is encountered in both effect columns of a cell (fixes DBM.143_Gnoj). * DBM: Don’t import offset effects when there’s a tone portmento next to them. * DBM: A few IT-specific playback quirks are disabled for more accurate playback (e.g. in “Are You Flying With Me?” by Jazzcat). * DIGI: Sample play direction was reset if adjacent channel contained a Note Cut note. * AMF: When running out of sample slots, file reading became be misaligned because the sample name was not skipped. * MED: Command 0F was not imported. * MED: Upper frequency limits should be more accurate now. * MED: Channel panning is now only applied in MMD2 files if the
Ana Guerrero (anag+factory)
accepted
request 1179592
from
Mia Herkt (mia)
(revision 48)
- Update to 0.7.8: * [Sec] Potential heap out-of-bounds read with malformed Dynamic Studio DSm files. * MED: Additional command pages can now be read. * MED: Echo DSP is now emulated. * AMS: Default channel panning was broken, using all-center panning instead of MOD-style panning. * STP: Correctly import volume slides where both nibbles are set. * MDL: Pitch slide depth was wrong since libopenmpt 0.6.0. * DTM: Fix pitch slide depth for samples with finetune. * DSm: Don’t import as Amiga module (and thus allow Amig a resampler to be enabled), as this PC format can easily exceed the Amiga frequency range. * DSm: Only the first 48 samples were ever read. * STM / STX: Don’t set channel panning. ScreamTracker 2 only outputs mono sound. * PSM: File header probing was stricter than actual file loader, rejecting manipulated files that would otherwise load fine.
Ana Guerrero (anag+factory)
accepted
request 1176837
from
Mia Herkt (mia)
(revision 47)
- Update to 0.7.7: * IT: Disable a few more compatibility flags for older SchismTracker builds. * IT: Halve the output volume of files saved with ChibiTracker, as its mixer is about half as loud as ours. * S3M: In mono mode, the ratio between sample and OPL volume was incorrect. * S3M: Detect files saved with PlayerPRO. * XM: Detect files saved with PlayerPRO. Improved detection of files saved with ModPlug Tracker. * XM: Files with impossibly small pattern headers are now rejected, like FastTracker 2 does. * STK: Loosen SoundTracker file rejection heuristics a bit to allow loading of files with malformed song titles and overly long samples. Other heuristics has been tightened a bit. * MOD: Change VBlank timing heuristic to always assume CIA timing when long samples unsupported by original ProTracker) are found. * MOD: Disable NoiseTracker quirks if empty sample slots have a default volume other than 0. * MOD: 7-bit panning detection may have returned incorrect results for MODs with a channel count other than 4. * MED: Improve handling of default tempo, panning and channel volume in files containing more than one song. * MO3: Reduced maximum allowed music data (not samples) size from 2 GiB to 512 MiB. * XMF: Header validation is now stricter. * MDL: Slightly more accurate import of pattern command 9 (enabling envelopes instead of setting envelope position). * ULT: Support the highest three octaves.
Ana Guerrero (anag+factory)
accepted
request 1163533
from
Mia Herkt (mia)
(revision 46)
- Update to 0.7.6: * [Sec] Potential heap out-of-bounds read or write past sample end with malformed sustain loops in SymMOD files. * MED: Transposed samples were playing too low in some files (e.g. mix94.mmd1). * OKT: Some files with garbage at the end were rejected (e.g. katharsis - piano lesson.okta). * Compressor DMO: It was possible that the plugin would not behave as intended at mix rates above 500 kHz. * Avoid re-allocating the loop state map contents on every playthrough of the module. - Run test suite during package build
Ana Guerrero (anag+factory)
accepted
request 1158797
from
Mia Herkt (mia)
(revision 45)
- Update to 0.7.5: * [Sec] Null-pointer write (32bit platforms) or excessive memory allocation (64bit platforms) when reading close to 4GiB of data from unseekable files.. * [Sec] Write buffer overflow when reading unseekable files close to 4GiB in size. * [Sec] Possible out-of-memory (32bit platforms) or excessive memory allocation (64bit platforms) when reading malformed data from unseekable files. * [Sec] DMF: Possible null-pointer write or excessive memory allocation when reading DMF files. * IT: In the previous version, Zxx macros in IT files made with older MPT versions were no longer working. * There was a periodic click when playing a module using the Chorus or Flanger DMO plugin at a mix rate exceeding ~136.5 kHz. * An older bugfix for undefined behaviour in the Distortion DMO plugin was incorrect, causing the distorted sound to be different in some situations. * xmp-openmpt: Metadata retrievel for playlist items was broken. - Changes in 0.7.4: * [Bug] Makefile: libopenmpt 0.7.3 broke running the test suite for Emscripten builds. * openmpt123: openmpt123 now uses a narrower layout on terminal windows with a width of less than 72 characters. * Setting all possible load_skip flags resulted in nothing being loaded at all, instead of just not loading the selected module parts. * When playing all subsongs, set_position_seconds didn’t always calculate the correct subsong to jump to.
Ana Guerrero (anag+factory)
accepted
request 1123699
from
Mia Herkt (mia)
(revision 44)
- Update to 0.7.3: * Fix: set_pitch_factor no longer affected OPL instruments since version 0.6.0.
Dominique Leuenberger (dimstar_suse)
accepted
request 1095930
from
Danilo Spinella (dspinella)
(revision 43)
Dominique Leuenberger (dimstar_suse)
accepted
request 1080266
from
Danilo Spinella (dspinella)
(revision 42)
Dominique Leuenberger (dimstar_suse)
accepted
request 1075084
from
Mia Herkt (mia)
(revision 41)
- Update to 0.6.9 * [Bug] An exception could be thrown during rendering when trying to access the release node of an empty envelope. * The fix for the OPL cutoff bug introduced in libopenmpt 0.6.7 was incomplete. * ULT: Offset commands exceeding 65535 samples were sometimes not imported correctly even if there was room for them. * After seeking with seek.sync_samples=1, the filter settings of playing notes were not updated since libopenmpt 0.6.7. * Loading of and seeking inside (malformed) modules with thousands of short sub-songs has been sped up.
Dominique Leuenberger (dimstar_suse)
accepted
request 1063428
from
Takashi Iwai (tiwai)
(revision 40)
Dominique Leuenberger (dimstar_suse)
accepted
request 1056956
from
Mia Herkt (mia)
(revision 39)
- Update to 0.6.7 * IT: In sample mode, portamento to a different sample turns off the filter if cutoff / resonance was previously 127 / 0. * S3M Detect files saved with Graoumf Tracker instead of claiming they were made with OpenMPT 4.47. * S3M: Pattern loop state was not propagated anymore since libopenmpt 0.6.0, leading to wrong song length calculation and SB0 + SBx being located on different channels not working properly anymore.
Dominique Leuenberger (dimstar_suse)
accepted
request 1007099
from
Mia Herkt (mia)
(revision 38)
- Update to 0.6.6 * [Sec] Possible crash when playing manipulated IT / MPTM files with a T00 command. * MTM: In MultiTracker, setting speed and tempo are mutually exclusive commands. Still, some MultiTracker modules were made to be played with external players such as DMP, so they actually rely on “standard” speed / tempo behaviour. Decide which behaviour to use by checking of speed and tempo commands are found on the same row. * MTM: Ignore sample loops if the loop end is <= 2. * Echo DMO: Migrate left and right delay values in modules made with OpenMPT versions between 1.27.01.00 and 1.30.05.00 to the correct interpretation. - Changes in 0.6.5: * [Bug] Makefile now also uses CC, CXX, LD, and AR when set as environment variables. * [New] Makefile now uses PKG_CONFIG as path to pkg-config. * DBM: Sample middle-C frequencies were slightly off. laffik_-_marynarze.dbm sounds much more bearable now. * DBM: Fixed pattern break parameter interpretation. * Echo DMO: Setting the delay parameter to 0 yielded a 2-second echo delay instead of 1ms. * Digi Booster Echo: Setting the delay parameter to 0 yielded a 510ms echo delay instead of approximately 334ms.
Dominique Leuenberger (dimstar_suse)
accepted
request 982866
from
Mia Herkt (mia)
(revision 37)
- Update to 0.6.4 * Fix: Ping-pong sample loops sometimes stopped playing at the end of the loop.
Dominique Leuenberger (dimstar_suse)
accepted
request 978626
from
Takashi Iwai (tiwai)
(revision 36)
Dominique Leuenberger (dimstar_suse)
accepted
request 950104
from
Mia Herkt (mia)
(revision 34)
- Update to 0.6.1: * [Bug] Possible hang with malformed DMF, DSM, MED, MUS, OKT and SymMOD files containing 65536 or more patterns when destroying the module. * [Bug] Avoid NaNs and infinite values with custom tunings and in the I3DL2Reverb plugin. * MIDI macros are now evaluated when seeking. * The letter “z” is now evaluated in fixed MIDI macros (Z80…ZFF) the same way as in Impulse Tracker. * MOD: Loosened VBlank timing heuristics so that “frame of mind” by Dascon plays correctly. * MOD: Validate the contents of “hidden” patterns beyond the end of the order list when the file size matches the expected size when only taken “official” patterns into account. This fixes Shofixti Ditty.mod from Star Control 2 while keeping other (partly broken) modules working. * MED: Command 20 (reverse sample) is now only applied when it’s next to a note. * S3M: Introducing the “Send OPL key-off when triggering notes” compatibility setting broke retrigger for OPL notes again (they retriggered rather than not retriggering). * S3M: Retriggering a note no longer resets its pitch after a portamento. * S3M: Partially implement retrigger behaviour for stopped notes in SoundBlaster mode: Like in IT, it is not possible to retrigger a sample that has already stopped playing. * DIGI: Improve compatibility with E3x reverse sample command. * DSym: Tempos < 32 were treated as tempo slides. * SymMOD: Key-off command was not implemented properly. - Changes in 0.6.0:
Dominique Leuenberger (dimstar_suse)
accepted
request 942189
from
Mia Herkt (mia)
(revision 33)
- Update to 0.5.14: * [Sec] Possible out-of-bounds read in Chorus plugin with NaN plugin parameters. Most implementations of the “fast math” compiler optimizations will prevent this crash but it is not guaranteed. (r16096) * [Bug] Fixed undefined behaviour with custom tunings found with ubsan. * OPL: Improved channel allocation strategy when there are lots of faded-out notes. * MOD: Digital Tracker MODs have four unknown bytes right after the magic bytes which seem to be ignored even by Digital Tracker itself. Just skip over them. * The logic when to turn off the resonant filter was broken in some edge cases since libopenmpt 0.5.1. * IMF: Implemented XE1-XE3 commands, which disable individual envelopes. Command Nxy (cutoff slide + resonance) is now also partially supported (only resonance for now). * IMF: Filter envelopes were upside down. * MTM: Omitting pattern loading through the load.skip_patterns ctl caused sample data to be corrupted. * S3M: Ignore O00 commands in files created with Scream Tracker 3.00 and 3.01, as this command only gained effect memory in version 3.03. * STM: Use S3M-like sample swap behaviour. * XM: Disable arpeggio quirk for XMs made with Skale Tracker. Fixes KAPTENFL.XM.
Dominique Leuenberger (dimstar_suse)
accepted
request 935240
from
Danilo Spinella (dspinella)
(revision 32)
Dominique Leuenberger (dimstar_suse)
accepted
request 927424
from
Takashi Iwai (tiwai)
(revision 31)
Dominique Leuenberger (dimstar_suse)
accepted
request 913906
from
Mia Herkt (mia)
(revision 30)
- Update to 0.5.11: * [Sec] Possible crash with malformed modules when trying to access non-existent plugin slots FX251-FX255. * [Sec] Possible read beyond sample start after swapping to a sample with loop points set but not loop enabled. * [Sec] Fixed various possible crashes with malformed MMCMP files. * [Sec] MED: Possible read past end of sequence name (stack-allocated, so relatively unlikely to result in a crash). * Fixed excessive memory usage with files claiming to have an extremely high rows per beat count while also using tempo swing. Maximum rows per beat are now limited to 65536. * STP: Avoid creating thousands of patterns when loading malformed files even though no more pattern data can be read.
Displaying revisions 1 - 20 of 49