Re: [ANN] CHICKEN 6.0.0 rerlease candidate available
Erik Falor via Chicken-users <[email protected]> Wed, 13 May 2026 09:40:55 -0600
| Newsgroups | gmane.lisp.scheme.chicken |
|---|---|
| Message-ID | <CAKwJTQ4BgkLwQhaH-m4Y2U8atEEqSX3Wgj8QG2=wUgJt54xkRg@mail.gmail.com> |
--0000000000005d4cc00651b4d0cd Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Operating system: Slackware64-current Hardware platform: armv7l (Raspberry Pi 3B) C Compiler: GCC 11.2.0 Installation works?: yes Tests work?: yes Installation of eggs works?: no Everything works well until I run ' <some dir>/bin/chicken-install srfi-1' which ends with this error message: fetching srfi-1 building srfi-1 /home/fadein/.cache/chicken-install/srfi-1/srfi-1.build.sh: line 11: build/bin/chicken-do: No such file or directory Error: shell command terminated with nonzero exit code 32512 "sh /home/fadein/.cache/chicken-install/srfi-1/srfi-1.build.sh" Upon inspection, I find that this generated build script first updates PATH like this: PATH=3D'build/bin':$PATH I think this is a mistake; instead of 'build/bin' it should have the fully-qualified installation path. Below that, it attempts to launch 'build/bin/chicken-do'. If PATH were set up correctly, that could simply be replaced by 'chicken-do'. However, everyone else who gets to this point of the process seems to have been able to build and install eggs, so maybe something uniquely weird is happening on my system. Thoughts? -Erik On Tue, May 12, 2026 at 2:22=E2=80=AFPM Felix Winkelmann via Chicken-users = < [email protected]> wrote: > On Tue May 12, 2026 at 10:18 PM CEST, Pietro Cerutti wrote: > >>> > >>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D csc tests ... > >>> dyld[4526]: Library not loaded: libchicken.dylib > >>> Referenced from: <EF4A9C65-6DDF-368F-9DC5-455A3285DB34> > /Users/pietrocerutti/chicken-6.0.0rc3/csc > >>> Reason: tried: 'libchicken.dylib' (no such file), > '/System/Volumes/Preboot/Cryptexes/OSlibchicken.dylib' (no such file), > 'libchicken.dylib' (no such file), > '/Users/pietrocerutti/chicken-6.0.0rc3/tests/libchicken.dylib' (no such > file), > '/System/Volumes/Preboot/Cryptexes/OS/Users/pietrocerutti/chicken-6.0.0rc= 3/tests/libchicken.dylib' > (no such file), > '/Users/pietrocerutti/chicken-6.0.0rc3/tests/libchicken.dylib' (no such > file) > >>> > >> > >>"OSlibchicken.dylib" looks wrong. Can you use "otool" to dump the > runtime linker > >>path of the produced executable from that test? > > > > I think it's just trying.. it simply is not looking into the root dir o= f > > the checkout / build. > > > > That test is an interpreted test that calls csc. However, when it uses > > system* to call csc, the DYLD_LIBRARY_PATH env is not retained. > > Right, that's SIP dropping the env-var. Did you make "install" before > "check"? > > > cheers, > felix > > > --=20 Erik Falor --0000000000005d4cc00651b4d0cd Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable <div dir=3D"ltr">Operating system: Slackware64-current<br>Hardware platform= : armv7l (Raspberry Pi 3B)<br>C Compiler: GCC 11.2.0<br> Installation works?: yes<br> Tests work?<span class=3D"Asgive ng" style=3D"border-style:none;background:= none">: yes</span><br><div>Installation of eggs works?: no<br></div><div><b= r></div><div>Everything works well until I run ' <some dir>/bin/c= hicken-install srfi-1' which ends with this error message:</div><div><b= r></div><div>fetching srfi-1<br>building srfi-1<br>/home/fadein/.cache/chic= ken-install/srfi-1/<a href=3D"http://srfi-1.build.sh">srfi-1.build.sh</a>: = line 11: build/bin/chicken-do: No such file or directory<br><br>Error: shel= l command terminated with nonzero exit code<br>32512<br>"sh /home/fade= in/.cache/chicken-install/srfi-1/<a href=3D"http://srfi-1.build.sh">srfi-1.= build.sh</a>"</div><div><br></div><div>Upon inspection, I find that th= is generated build script first updates PATH like this:</div><div><br></div= ><div>PATH=3D'build/bin':$PATH</div><div><br></div><div>I think thi= s is a mistake; instead of 'build/bin' it should=C2=A0<span class= =3D"Asgive ng" style=3D"border-style:none;background:none">have</span> the = fully-qualified installation path. <span class=3D"Asgive ng" style=3D"borde= r-style:none;background:none">Below that, it</span> attempts to launch=C2= =A0'build/bin/chicken-do'. If PATH were set up correctly, <span cla= ss=3D"Asgive ng" style=3D"border-style:none;background:none">that</span> co= uld simply be replaced by 'chicken-do'.</div><div><br></div><div>Ho= wever, everyone else who <span class=3D"Asgive ng" style=3D"border-style:no= ne;background:none">gets to</span> this point of the process seems <span cl= ass=3D"Asgive ng" style=3D"border-style:none;background:none">to have been = </span>able to build and install eggs, so maybe something uniquely weird is= happening on my system. Thoughts?</div><div><br></div><div>-Erik</div></di= v><br><div class=3D"gmail_quote"><div dir=3D"ltr" class=3D"gmail_attr">On T= ue, May 12, 2026 at 2:22=E2=80=AFPM Felix Winkelmann via Chicken-users <= <a href=3D"mailto:[email protected]" target=3D"_blank">chicken-users= @nongnu.org</a>> wrote:<br></div><blockquote class=3D"gmail_quote" style= =3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding= -left:1ex">On Tue May 12, 2026 at 10:18 PM CEST, Pietro Cerutti wrote:<br> >>><br> >>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D csc tests ...<= br> >>> dyld[4526]: Library not loaded: libchicken.dylib<br> >>>=C2=A0 =C2=A0 Referenced from: <EF4A9C65-6DDF-368F-9DC5-455A= 3285DB34> /Users/pietrocerutti/chicken-6.0.0rc3/csc<br> >>>=C2=A0 =C2=A0 Reason: tried: 'libchicken.dylib' (no suc= h file), '/System/Volumes/Preboot/Cryptexes/OSlibchicken.dylib' (no= such file), 'libchicken.dylib' (no such file), '/Users/pietroc= erutti/chicken-6.0.0rc3/tests/libchicken.dylib' (no such file), '/S= ystem/Volumes/Preboot/Cryptexes/OS/Users/pietrocerutti/chicken-6.0.0rc3/tes= ts/libchicken.dylib' (no such file), '/Users/pietrocerutti/chicken-= 6.0.0rc3/tests/libchicken.dylib' (no such file)<br> >>><br> >><br> >>"OSlibchicken.dylib" looks wrong. Can you use "otool= " to dump the runtime linker<br> >>path of the produced executable from that test?<br> ><br> > I think it's just trying.. it simply is not looking into the root = dir of <br> > the checkout / build.<br> ><br> > That test is an interpreted test that calls csc. However, when it uses= <br> > system* to call csc, the DYLD_LIBRARY_PATH env is not retained.<br> <br> Right, that's SIP dropping the env-var. Did you make "install"= ; before "check"?<br> <br> <br> cheers,<br> felix<br> <br> <br> </blockquote></div><div><br clear=3D"all"></div><br><span class=3D"gmail_si= gnature_prefix">-- </span><br><div dir=3D"ltr" class=3D"gmail_signature"><d= iv dir=3D"ltr">Erik Falor<br></div></div> --0000000000005d4cc00651b4d0cd--