Re: RCS will not compile om CentOS 8
"Knight, Dave" <[email protected]> Mon, 28 Sep 2020 18:31:18 -0400
| Newsgroups | gmane.comp.version-control.rcs.bugs |
|---|---|
| Message-ID | <CAOmkcGsROFWe5JmJLNCn5601UT776=67HZ9CYKYqDqy01dfQnQ@mail.gmail.com> |
--000000000000ca782d05b0673b40 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Thanks for the patch! This patch fixes the compile problem, but when I run "make check", test t810 fails (this may be unrelated to the fix): ... PASS: t803 PASS: t804 PASS: t805 FAIL: t810 PASS: t900 PASS: t999 =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 1 of 51 tests failed Please report to [email protected] =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 make[2]: *** [Makefile:1502: check-TESTS] Error 1 make[2]: Leaving directory '/home/dmk/RCS/rcs-5.9.4-fix/rcs-5.9.4/tests' make[1]: *** [Makefile:1626: check-am] Error 2 make[1]: Leaving directory '/home/dmk/RCS/rcs-5.9.4-fix/rcs-5.9.4/tests' make: *** [Makefile:1392: check-recursive] Error 1 dmk@del1:~/RCS/rcs-5.9.4-fix/rcs-5.9.4$ Perhaps the code for test t810 is informative: # t810 --- ci -l -d -T FILE <(C) boilerplate> . $srcdir/common split_std_out_err no . $srcdir/common-d ## # Check that =E2=80=98ci -l -d -T FILE=E2=80=99 does not set # the mtime of FILE,v to the epoch. # This test derives from a 5.8 regression (and bugfix). ## must 'echo new line >> $w' must 'ci -mm -l -d -T $w' test $w -nt $v && problem "$w newer than $v" exit 0 # t810 ends here That said, I doubt that this failure will impact use of this build with /BriefCase as the problem for which it tests seems not to be among the things that /BriefCase does with the RCS ci command. Nevertheless, I hope this email is helpful. Dave Knight Applied-cs-inc.com (0wner) and author of the /BriefCase Revision Control and Release Management toolkit which uses the RCS engine. -- In your time of need, may your piece be with you! The trouble with many of today's college graduates is not that they're ignorant, it's that they've been taught so much that just isn't so! (with apologies to Ronald Reagan). On Mon, Sep 28, 2020 at 5:33 PM Paul Eggert <[email protected]> wrote: > On 9/28/20 12:51 PM, Bill P wrote: > > > b-anchor.c:99:1: error: expected '=3D', ',', ';', 'asm' or '__attribute= __' > > before '{' token > > > > { > > This is due RCS 5.9.4's misuse of _Noreturn. The attached patch should > work > around the bug (but is good for GCC-compatible compilers only; something > fancier > is needed for other compilers). > > Looks like it's time for a new RCS release, if only to fix this porting > glitch. > --000000000000ca782d05b0673b40 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable <div dir=3D"ltr"> Thanks for the patch!<div><br>This patch fixes the compil= e problem, but when I run "make check",=C2=A0<div>test t810 fails= (this may be unrelated to the fix):<div><br></div><div>...</div><div>PASS:= t803<br>PASS: t804<br>PASS: t805<br>FAIL: t810<br>PASS: t900<br>PASS: t999= <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<br>1 of 51 tests failed<br>Please report to <a = href=3D"mailto:[email protected]" target=3D"_blank">[email protected]</a><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<br>make[2]: *** [Makefile:1502: check-TESTS] Error 1<= br>make[2]: Leaving directory '/home/dmk/RCS/rcs-5.9.4-fix/rcs-5.9.4/te= sts'<br>make[1]: *** [Makefile:1626: check-am] Error 2<br>make[1]: Leav= ing directory '/home/dmk/RCS/rcs-5.9.4-fix/rcs-5.9.4/tests'<br>make= : *** [Makefile:1392: check-recursive] Error 1<br>dmk@del1:~/RCS/rcs-5.9.4-= fix/rcs-5.9.4$<br><div><br><div>Perhaps the code for test t810 is informati= ve:</div><div><br></div><div># t810 --- ci -l -d -T FILE<br></div><div><br>= </div><div><(C) boilerplate></div><div><br>. $srcdir/common<br>split_= std_out_err no<br>. $srcdir/common-d<br><br>##<br># Check that =E2=80=98ci = -l -d -T FILE=E2=80=99 does not set<br># the mtime of FILE,v to the epoch.<= br># This test derives from a 5.8 regression (and bugfix).<br>##<br><br>mus= t 'echo new line >> $w'<br>must 'ci -mm -l -d -T $w'<= br>test $w -nt $v && problem "$w newer than $v"<br><br>ex= it 0<br><br># t810 ends here<br></div><div><br></div><div>That said, I doub= t that this failure will impact use of this build with /BriefCase as the pr= oblem for which it tests=C2=A0</div><div>seems not to be among the things t= hat /BriefCase does with the RCS ci command.</div><div><br></div><div>Never= theless, I hope this email is helpful.</div><div><br></div><div>Dave Knight= </div><div>Applied-cs-inc.com (0wner) and author of the=C2=A0</div><div>/Br= iefCase Revision=C2=A0Control and Release Management toolkit=C2=A0</div><di= v>which uses the RCS engine.</div><div><br></div><div><div><div dir=3D"ltr"= data-smartmail=3D"gmail_signature"><div dir=3D"ltr"><div><div dir=3D"ltr">= <div><div dir=3D"ltr"><div><div dir=3D"ltr"><div><div dir=3D"ltr"><div><div= dir=3D"ltr"><div><div dir=3D"ltr"><div><div dir=3D"ltr"><div><div dir=3D"l= tr"><div><div dir=3D"ltr"><div><div dir=3D"ltr"><div><div dir=3D"ltr"><div>= <div dir=3D"ltr"><div><div dir=3D"ltr"><div><div dir=3D"ltr"><div><div dir= =3D"ltr"><div><div dir=3D"ltr"><div><div dir=3D"ltr"><div><div dir=3D"ltr">= <div><div dir=3D"ltr"><div><div dir=3D"ltr"><div> <div> --<br><span><br>In your time of need, may your piece be with you!</span></d= iv><div><span><br></span></div><div>The trouble with many of today's co= llege graduates is not that they're</div><div> ignorant, it's that = they've been taught so much that just isn't so! <br></div><div><br>= </div><div>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 (with apologies to Ronald Reagan).<br></div>= <div><br></div></div></div></div></div></div></div></div></div></div></div>= </div></div></div></div></div></div></div></div></div></div></div></div></d= iv></div></div></div></div></div></div></div></div></div></div></div></div>= </div></div></div></div></div></div></div></div></div><br></div></div></div= ></div></div></div><br><div class=3D"gmail_quote"><div dir=3D"ltr" class=3D= "gmail_attr">On Mon, Sep 28, 2020 at 5:33 PM Paul Eggert <<a href=3D"mai= lto:[email protected]" target=3D"_blank">[email protected]</a>> wrote:= <br></div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8= ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">On 9/28/20 12:5= 1 PM, Bill P wrote:<br> <br> > b-anchor.c:99:1: error: expected '=3D', ',', ';= 9;, 'asm' or '__attribute__'<br> > before '{' token<br> > <br> > {<br> <br> This is due RCS 5.9.4's misuse of _Noreturn. The attached patch should = work <br> around the bug (but is good for GCC-compatible compilers only; something fa= ncier <br> is needed for other compilers).<br> <br> Looks like it's time for a new RCS release, if only to fix this porting= glitch.<br> </blockquote></div> --000000000000ca782d05b0673b40--