forwarded 985129 https://github.com/musescore/MuseScore/pull/7728
tags 985129 - moreinfo
tags 985129 + confirmed upstream pending
affects 985129 musescore
outlook 985129 a patch has been found and musescore{2,3} will be uploaded soon
thanks
@Fabian since you were the driving force behind SF3 integration
into FluidSynth itself, you might wish to have a look at the
patch as well.
Paul Menzel dixit:
> (gdb) print *s
> $2 = {_valid = true, sf = 0x563075eb55a0, start = 0, end = 0,
> loopstart = 29362, loopend = 38401, samplerate = 44100,
> origpitch = 63, pitchadj = -8, sampletype = 17, data = 0x0,
Good news: With sf3convert, I was able to identify this sample:
<Sample name="Oboe D#4">
<start>25705425</start>
<end>25743841</end>
<loopstart>29362</loopstart>
<loopend>38401</loopend>
<samplerate>44100</samplerate>
<origpitch>63</origpitch>
<pitchadj>-8</pitchadj>
<sampletype>1</sampletype>
</Sample>
(Ignore that sampletype is 1 here.)
Bad news: the sample and the soundfont is fine, so this probably
was a temporary hiccup, I/O or decompression-related… or maybe
even caused by threads interjecting… the last commit related to
mutex locking around soundfont loading is present, though… no idea.
More good news: I have a patch, also forwarded upstream, which I
believe fixes this bug, but also others. I’m not uploading this any
more tonight but later but if you want to test it already I provide
it at: https://people.debian.org/~tg/musescore3_3.2.3+dfsg2-11_amd64.deb
bye,
//mirabilos
--
22:20⎜<asarch> The crazy that persists in his craziness becomes a master
22:21⎜<asarch> And the distance between the craziness and geniality is
only measured by the success 18:35⎜<asarch> "Psychotics are consistently
inconsistent. The essence of sanity is to be inconsistently inconsistent
