Repository navigation
Bleadperl v5.27.1-114-g9449f0d6ae breaks MLEHMANN/Glib-EV-2.02.tar.gz #16122
Description
Activity
From @andk
Once again, affected are -DDEBUGGING builds only.
bisect
commit 9449f0d
Author: David Mitchell <davem@iabyn.com>
Date: Mon Jul 3 13:43:27 2017 +0100save si_stack_hwm across JMPENV_PUSH
other distros also affected
MARKSMITH/HTTP-HeaderParser-XS-0.20.tar.gz
FRACTAL/IO-Vectored-0.110.tar.gz
SDPRICE/Linux-DVB-DVBT-Advert-0.04.tar.gz
SDPRICE/Linux-DVB-DVBT-TS-0.08.tar.gzcpantesters report
http://www.cpantesters.org/cpan/report/72a6d462-7362-11e7-a0cc-5c58e11ca5c1
perl -V
Summary of my perl5 (revision 5 version 27 subversion 2) configuration:
Commit id: 9449f0d
Platform:
osname=linux
osvers=4.9.0-2-amd64
archname=x86_64-linux
uname='linux k93msid 4.9.0-2-amd64 #1 smp debian 4.9.18-1 (2017-03-30) x86_64 gnulinux '
config_args='-Dprefix=/home/sand/src/perl/repoperls/installed-perls/host/k93msid/v5.27.1-114-g9449f0d6ae/d2d1 -Dmyhostname=k93msid -Dinstallusrbinperl=n -Uversiononly -Dusedevel -des -Ui_db -Dlibswanted=cl pthread socket inet nsl gdbm dbm malloc dl ld sun m crypt sec util c cposix posix ucb BSD gdbm_compat -Uuseithreads -Uuselongdouble -DDEBUGGING=both'
hint=recommended
useposix=true
d_sigaction=define
useithreads=undef
usemultiplicity=undef
use64bitint=define
use64bitall=define
uselongdouble=undef
usemymalloc=n
default_inc_excludes_dot=define
bincompat5005=undef
Compiler:
cc='cc'
ccflags ='-fwrapv -DDEBUGGING -fno-strict-aliasing -pipe -fstack-protector-strong -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64'
optimize='-O2 -g'
cppflags='-fwrapv -DDEBUGGING -fno-strict-aliasing -pipe -fstack-protector-strong -I/usr/local/include'
ccversion=''
gccversion='6.3.0 20170406'
gccosandvers=''
intsize=4
longsize=8
ptrsize=8
doublesize=8
byteorder=12345678
doublekind=3
d_longlong=define
longlongsize=8
d_longdbl=define
longdblsize=16
longdblkind=3
ivtype='long'
ivsize=8
nvtype='double'
nvsize=8
Off_t='off_t'
lseeksize=8
alignbytes=8
prototype=define
Linker and Libraries:
ld='cc'
ldflags =' -fstack-protector-strong -L/usr/local/lib'
libpth=/usr/local/lib /usr/lib/gcc/x86_64-linux-gnu/6/include-fixed /usr/include/x86_64-linux-gnu /usr/lib /lib/x86_64-linux-gnu /lib/../lib /usr/lib/x86_64-linux-gnu /usr/lib/../lib /lib
libs=-lpthread -lnsl -ldl -lm -lcrypt -lutil -lc
perllibs=-lpthread -lnsl -ldl -lm -lcrypt -lutil -lc
libc=libc-2.24.so
so=so
useshrplib=false
libperl=libperl.a
gnulibc_version='2.24'
Dynamic Linking:
dlsrc=dl_dlopen.xs
dlext=so
d_dlsymun=undef
ccdlflags='-Wl,-E'
cccdlflags='-fPIC'
lddlflags='-shared -O2 -g -L/usr/local/lib -fstack-protector-strong'Characteristics of this binary (from libperl):
Compile-time options:
DEBUGGING
HAS_TIMES
PERLIO_LAYERS
PERL_COPY_ON_WRITE
PERL_DONT_CREATE_GVSV
PERL_MALLOC_WRAP
PERL_OP_PARENT
PERL_PRESERVE_IVUV
PERL_USE_DEVEL
USE_64_BIT_ALL
USE_64_BIT_INT
USE_LARGE_FILES
USE_LOCALE
USE_LOCALE_COLLATE
USE_LOCALE_CTYPE
USE_LOCALE_NUMERIC
USE_LOCALE_TIME
USE_PERLIO
USE_PERL_ATOF
Built under linux
Compiled at Aug 20 2017 19:39:36
@INC:
/home/sand/src/perl/repoperls/installed-perls/host/k93msid/v5.27.1-114-g9449f0d6ae/d2d1/lib/site_perl/5.27.2/x86_64-linux
/home/sand/src/perl/repoperls/installed-perls/host/k93msid/v5.27.1-114-g9449f0d6ae/d2d1/lib/site_perl/5.27.2
/home/sand/src/perl/repoperls/installed-perls/host/k93msid/v5.27.1-114-g9449f0d6ae/d2d1/lib/5.27.2/x86_64-linux
/home/sand/src/perl/repoperls/installed-perls/host/k93msid/v5.27.1-114-g9449f0d6ae/d2d1/lib/5.27.2--
andreasFrom @jkeenan
On Mon, 21 Aug 2017 20:39:56 GMT, andreas.koenig.7os6VVqR@franz.ak.mind.de wrote:
Once again, affected are -DDEBUGGING builds only.
bisect
------
commit 9449f0d
Author: David Mitchell <davem@iabyn.com>
Date: Mon Jul 3 13:43:27 2017 +0100save si_stack_hwm across JMPENV_PUSH
other distros also affected
---------------------------
MARKSMITH/HTTP-HeaderParser-XS-0.20.tar.gz
FRACTAL/IO-Vectored-0.110.tar.gz
SDPRICE/Linux-DVB-DVBT-Advert-0.04.tar.gz
SDPRICE/Linux-DVB-DVBT-TS-0.08.tar.gzcpantesters report
------------------
http://www.cpantesters.org/cpan/report/72a6d462-7362-11e7-a0cc-
5c58e11ca5c1Examples of failures:
#####
[HTTP-HeaderParser-XS-0.20] 547 $/testing/blead/bin/prove -I/testing/blead/lib -vb t/10-basic.t
t/10-basic.t ..
1..39
HTTPHeaders.c: loadable library and perl binaries are mismatched (got handshake key 0xd100000, needed 0xd180000)
Dubious, test returned 1 (wstat 256, 0x100)
Failed 39/39 subtestsTest Summary Report
t/10-basic.t (Wstat: 256 Tests: 0 Failed: 0)
Non-zero exit status: 1
Parse errors: Bad plan. You planned 39 tests but ran 0.
Files=1, Tests=0, 0 wallclock secs ( 0.02 usr 0.00 sys + 0.03 cusr 0.00 csys = 0.05 CPU)
Result: FAIL
#####
[IO-Vectored-0.110] 515 $/testing/blead/bin/prove -I/testing/blead/lib -vb t/*.t
t/exceptions.t ..
1..6
Vectored.c: loadable library and perl binaries are mismatched (got handshake key 0xd100000, needed 0xd180000)
Dubious, test returned 1 (wstat 256, 0x100)
Failed 6/6 subtests
t/normal.t ......
1..4
Vectored.c: loadable library and perl binaries are mismatched (got handshake key 0xd100000, needed 0xd180000)
Dubious, test returned 1 (wstat 256, 0x100)
Failed 4/4 subtests
t/numeric.t .....
1..3
Vectored.c: loadable library and perl binaries are mismatched (got handshake key 0xd100000, needed 0xd180000)
Dubious, test returned 1 (wstat 256, 0x100)
Failed 3/3 subtestsTest Summary Report
t/exceptions.t (Wstat: 256 Tests: 0 Failed: 0)
Non-zero exit status: 1
Parse errors: Bad plan. You planned 6 tests but ran 0.
t/normal.t (Wstat: 256 Tests: 0 Failed: 0)
Non-zero exit status: 1
Parse errors: Bad plan. You planned 4 tests but ran 0.
t/numeric.t (Wstat: 256 Tests: 0 Failed: 0)
Non-zero exit status: 1
Parse errors: Bad plan. You planned 3 tests but ran 0.
Files=3, Tests=0, 0 wallclock secs ( 0.02 usr 0.01 sys + 0.10 cusr 0.00 csys = 0.13 CPU)
Result: FAIL
#####--
James E Keenan (jkeenan@cpan.org)The RT System itself - Status changed from 'new' to 'open'
From @iabyn
On Tue, Aug 22, 2017 at 06:35:32PM -0700, James E Keenan via RT wrote:
On Mon, 21 Aug 2017 20:39:56 GMT, andreas.koenig.7os6VVqR@franz.ak.mind.de wrote:
Once again, affected are -DDEBUGGING builds only.
bisect
------
commit 9449f0d
Author: David Mitchell <davem@iabyn.com>
Date: Mon Jul 3 13:43:27 2017 +0100save si_stack_hwm across JMPENV_PUSH
other distros also affected
---------------------------
MARKSMITH/HTTP-HeaderParser-XS-0.20.tar.gz
FRACTAL/IO-Vectored-0.110.tar.gz
SDPRICE/Linux-DVB-DVBT-Advert-0.04.tar.gz
SDPRICE/Linux-DVB-DVBT-TS-0.08.tar.gzExamples of failures:
#####
[HTTP-HeaderParser-XS-0.20] 547 $/testing/blead/bin/prove -I/testing/blead/lib -vb t/10-basic.t
t/10-basic.t ..
1..39
HTTPHeaders.c: loadable library and perl binaries are mismatched (got handshake key 0xd100000, needed 0xd180000)They should hopefully be fixed with the commit below which I've just
pushed. I've only tested IO-Vectored, but I'm hoping/assuming they're all
the same cause and fix.commit d2dccc0
Author: David Mitchell <davem@iabyn.com>
AuthorDate: Wed Aug 23 10:18:53 2017 +0100
Commit: David Mitchell <davem@iabyn.com>
CommitDate: Wed Aug 23 10:18:53 2017 +0100make 'struct jmpenv' size independent of DEBUGGING
RT #131942
My commit v5.27.1-114-g9449f0d added a field je_old_stack_hwm to
the jmpenv structure on debugging builds.
However, one field in the interpreter structure is a JMPENV (rather than
JMPENV*), so this can make the interpreter structure size vary on
debugging and non-debugging builds.
This is usually a no-no, since it breaks re.pm, which recompiles
the various re functions with DEBUGGING enabled, even on non-debugging
fields.
This was supposed to be handled by the extra '!defined DEBUGGING_RE_ONLY'
condition, and indeed all core XS modules including re.xs work ok.
However, it seems to have broken several CPAN modules. I still don't
understand why, but the easiest fix is to just always include the
debugging je_old_stack_hwm field in all builds.Affected files ...
M cop.hDifferences ...
Inline Patch
diff --git a/cop.h b/cop.h index f621cd2..34ebbc0 100644 --- a/cop.h +++ b/cop.h @@ -35,9 +35,7 @@ struct jmpenv { int je_ret; /* last exception thrown */ bool je_mustcatch; /* need to call longjmp()? */ U16 je_old_delaymagic; /* saved PL_delaymagic */ -#if defined DEBUGGING && !defined DEBUGGING_RE_ONLY SSize_t je_old_stack_hwm; -#endif }; typedef struct jmpenv JMPENV; --
Art is anything that has a label (especially if the label is "untitled 1")
From @jkeenan
On Wed, 23 Aug 2017 09:30:10 GMT, davem wrote:
On Tue, Aug 22, 2017 at 06:35:32PM -0700, James E Keenan via RT wrote:
On Mon, 21 Aug 2017 20:39:56 GMT,
andreas.koenig.7os6VVqR@franz.ak.mind.de wrote:Once again, affected are -DDEBUGGING builds only.
bisect
------
commit 9449f0d
Author: David Mitchell <davem@iabyn.com>
Date: Mon Jul 3 13:43:27 2017 +0100save si_stack_hwm across JMPENV_PUSH
other distros also affected
---------------------------
MARKSMITH/HTTP-HeaderParser-XS-0.20.tar.gz
FRACTAL/IO-Vectored-0.110.tar.gz
SDPRICE/Linux-DVB-DVBT-Advert-0.04.tar.gz
SDPRICE/Linux-DVB-DVBT-TS-0.08.tar.gzExamples of failures:
#####
[HTTP-HeaderParser-XS-0.20] 547 $/testing/blead/bin/prove/testing/blead/lib -vb t/10-basic.t
-I
t/10-basic.t ..
1..39
HTTPHeaders.c: loadable library and perl binaries are mismatched (got
handshake key 0xd100000, needed 0xd180000)They should hopefully be fixed with the commit below which I've just
pushed. I've only tested IO-Vectored, but I'm hoping/assuming they're
all
the same cause and fix.commit d2dccc0
Author: David Mitchell <davem@iabyn.com>
AuthorDate: Wed Aug 23 10:18:53 2017 +0100
Commit: David Mitchell <davem@iabyn.com>
CommitDate: Wed Aug 23 10:18:53 2017 +0100make 'struct jmpenv' size independent of DEBUGGING
RT #131942
My commit v5.27.1-114-g9449f0d added a field je_old_stack_hwm to
the jmpenv structure on debugging builds.However, one field in the interpreter structure is a JMPENV (rather
than
JMPENV*), so this can make the interpreter structure size vary on
debugging and non-debugging builds.This is usually a no-no, since it breaks re.pm, which recompiles
the various re functions with DEBUGGING enabled, even on non-debugging
fields.This was supposed to be handled by the extra '!defined
DEBUGGING_RE_ONLY'
condition, and indeed all core XS modules including re.xs work ok.However, it seems to have broken several CPAN modules. I still don't
understand why, but the easiest fix is to just always include the
debugging je_old_stack_hwm field in all builds.Affected files ...
M cop.hDifferences ...
diff --git a/cop.h b/cop.h
index f621cd2..34ebbc0 100644
--- a/cop.h
+++ b/cop.h
@@ -35,9 +35,7 @@ struct jmpenv {
int je_ret; /* last exception
thrown */
bool je_mustcatch; /* need to call longjmp()? */
U16 je_old_delaymagic; /* saved PL_delaymagic */
-#if defined DEBUGGING && !defined DEBUGGING_RE_ONLY
SSize_t je_old_stack_hwm;
-#endif
};typedef struct jmpenv JMPENV;
These all appear to be fixed.
#####
$ ./bin/perl -v | head -2 | tail -1
This is perl 5, version 27, subversion 4 (v5.27.4 (v5.27.3-14-gd2dccc0)) built for x86_64-linux$ ./bin/perl -Ilib -MHTTP::HeaderParser::XS -E 'say q|hello world|;'
hello world$ ./bin/perl -Ilib -MIO::Vectored -E 'say q|hello world|;'
hello world$ ./bin/perl -Ilib -MLinux::DVB::DVBT::Advert -E 'say q|hello world|;'
hello world$ ./bin/perl -Ilib -MLinux::DVB::DVBT::TS -E 'say q|hello world|;'
hello world$ ./bin/perl -Ilib -MEV -E 'say q|hello world|;'
hello world
#####The Glib::EV case was more complicated because the author has a Makefile.PL which requires responding to prompts. So when I first tried to install this via './bin/cpanm', I got a FAIL at the Configure stage. When I went into the .cpanm build directories, manually ran Makefile.PL and responded to the prompts, I was then able to call make, make test and make install.
Unless other problems appear, I believe this ticket can be closed.
Thank you very much.
--
James E Keenan (jkeenan@cpan.org)From @kentfredric
On 24 August 2017 at 05:09, James E Keenan via RT
<perlbug-followup@perl.org> wrote:The Glib::EV case was more complicated because the author has a Makefile.PL which requires responding to prompts. So when I first tried to install this via './bin/cpanm', I got a FAIL at the Configure stage. When I went into the .cpanm build directories, manually ran Makefile.PL and responded to the prompts, I was then able to call make, make test and make install.
cpanm --help | grep inter
--interactive Turns on interactive configure (required
for Task:: modules)This is *usually* sufficient to get around interactive prompting.
--
KentKENTNL - https://metacpan.org/author/KENTNL
Migrated from rt.perl.org#131942 (status was 'resolved')
Searchable as RT131942$