Re: Lilypond 2.23.0 crashes on layout

[Jonas Hahnfeld]

Am Dienstag, dem 16.02.2021 um 15:44 +0100 schrieb Arusekk:
> Hello lilypond crew!
> 
> Reporting the bug here, since I saw on the bug link that you prefer mail 
> over GitLab issues.
> 
> When trying to process any file with \layout{} directive, lilypond 
> crashes (Segmentation fault / address not mapped).
> 
> I was able to reproduce it on current git master 
> (2c9361d3d2d9e9bdfca626d01076c94fb81e2e88) as well.
> 
> My distro: Gentoo Linux
> 
> Installed (possibly relevant) library versions:
> 
> [IP-] [ ] dev-scheme/guile-2.2.7:12/2.2-1
> 
> [IP-] [ ] x11-libs/pango-1.42.4-r2:0
> 
> [IP-] [ ] dev-libs/boehm-gc-8.0.4:0

Usual disclaimer that LilyPond doesn't fully support Guile 2.2 (yet),
so you'll definitely see worse performance (mostly slower startup) and
might run into occasional issues, like this one. Guile 1.8 should be
the safer choice for now (if you still have it in Gentoo).

> typescript.txt contains colored output (made with `script` utility, 
> contains ANSI escapes, view best with cat or less -R) from lilypond 
> being run under AddressSanitizer, which suggests that there is a 
> use-after-free bug, while processing the example from snk6.ly; it also 
> fails to process files as simple as `\score{c \layout{}}`, though 
> providing arguably less helpful, but distinctly different output 
> (attached as basic.txt).

That said, I've seen this issue before during my tests with Guile 2.2
(Translators being collected while still referenced somewhere), but
never had a stack trace to investigate. So it's already great to have
that, but do I read this correctly that you can reliably reproduce the
issue on your system? I'm asking because I can't on mine (Arch Linux)
and in fact, your file works just fine for me. So finding out how to
trigger this kind of bugs more reliably would be a huge step.
Regarding AddressSanitizer, did you just compile LilyPond with it or
all dependencies? I didn't attempt so far because I thought it wouldn't
work, but maybe I should try?

Jonas

> 
> I was not able to create such a minimal input file that would trigger 
> the helpful trace.
> 
> Off-topic, I believe I transcribed the piece to sound pretty accurately,
> 
> but I have not seen the layout yet (because of the bug), so please don't 
> tell me that it renders ugly.
> 
> Million thanks for such an awesome piece of software anyway :)
> 
> Please let me know if you need more info, or if the bug should be 
> reported to one of the libraries instead.

(It could of course be an issue in one of the libraries, especially the
stack trace in Pango looks odd, but it's equally possible that it's a
heap corruption bug that may show in the least expected places...)

signature.asc
Description: This is a digitally signed message part