Opcode module (and thus Safe) think block eval is a string eval #9999

p5pRT · 2009-12-01T13:20:28Z

Migrated from rt.perl.org#70970 (status was 'resolved')

Searchable as RT70970$

p5pRT · 2009-12-01T13:20:29Z

From @timbunce

$ perl -Mops=:default -e 'eval { ; }'
'eval "string"' trapped by operation mask at -e line 1.

A block eval (entertry opcode) is being treated as a string eval (entereval opcode).

This is a serious problem because most users of Opcode & Safe want to
restrict use of string eval (entereval opcode) but doing so prevents
them using any code/modules that use block eval!

This is true for current blead and at least as far back as 5.8.6.

The bug manifested for me as

$ perl -Mops=:default,require /Users/timbo/pg/perl598/lib/5.8.9/Carp.pm
'eval "string"' trapped by operation mask at /Users/timbo/pg/perl598/lib/5.8.9/Carp.pm line 33.

where line 33 is eval { require Carp::Heavy };

(Actually the bug manifested while extending PostgreSQL PL/Perl to support the
warnings pragma, but you don't want to know about that.)

(Note that this works:
$ perl -Mops=:default,require -e 'use Carp; warn $INC{"Carp.pm"}'
/Users/timbo/pg/perl598/lib/5.8.9/Carp.pm at -e line 1.
because Carp was already loaded via ops->Opcode->Carp.)

Using B::Concise shows an entertry opcode:

$ perl -MO=Concise -e 'eval { ; }'
5 <@> leave[1 ref] vKP/REFC ->(end)
1 <0> enter ->2
2 <;> nextstate(main 2 -e:1) v ->3
4 <@> leavetry vK ->5
3 <|> entertry(other->4) v ->6
6 <0> stub v ->4
-e syntax OK

Using perl -DtxT shows that the entertry started out as a entereval:

$ perl -DtxT -e 'eval { ; }' 2>&1
### 0:LEX_NORMAL/XSTATE "\n;"
### <== UNIOP(ival=op_entereval)

### 1:LEX_NORMAL/XTERMBLOCK "{ ; }\n"
### <== '{'
...

In toke.c we have:

case KEY_eval:
s = SKIPSPACE1(s);
PL_expect = (*s == '{') ? XTERMBLOCK : XTERM;
UNIBRACK(OP_ENTEREVAL);

It seems that toke.c should use OP_ENTERTRY instead of OP_ENTEREVAL when
it knows the eval is followed by a block (which it does).

Perl Info


Flags:
    category=core
    severity=high

Site configuration information for perl v5.8.6:

Configured by timbo at Thu May 25 17:26:37 IST 2006.

Summary of my perl5 (revision 5 version 8 subversion 6) configuration:
  Platform:
    osname=darwin, osvers=8.5.2, archname=darwin-thread-multi-2level
    uname='darwin foo 8.5.2 darwin kernel version 8.5.2: mon feb 13 16:31:48 pst 2006; root:xnu-792.8.37.obj~1release_i386 i386 i386 '
    config_args='-des -Dprefix=/usr/local/perl58-i -Doptimize=-g -Duseithreads -Dusemultiplicity'
    hint=recommended, useposix=true, d_sigaction=define
    usethreads=define use5005threads=undef useithreads=define usemultiplicity=define
    useperlio=define d_sfio=undef uselargefiles=define usesocks=undef
    use64bitint=undef use64bitall=undef uselongdouble=undef
    usemymalloc=n, bincompat5005=undef
  Compiler:
    cc='cc', ccflags ='-fno-common -DPERL_DARWIN -no-cpp-precomp -DDEBUGGING -fno-strict-aliasing -pipe -I/usr/local/include -I/opt/local/include',
    optimize='-g',
    cppflags='-no-cpp-precomp -fno-common -DPERL_DARWIN -no-cpp-precomp -DDEBUGGING -fno-strict-aliasing -pipe -I/usr/local/include -I/opt/local/include'
    ccversion='', gccversion='4.0.1 (Apple Computer, Inc. build 5250)', gccosandvers=''
    intsize=4, longsize=4, ptrsize=4, doublesize=8, byteorder=1234
    d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=16
    ivtype='long', ivsize=4, nvtype='double', nvsize=8, Off_t='off_t', lseeksize=8
    alignbytes=8, prototype=define
  Linker and Libraries:
    ld='env MACOSX_DEPLOYMENT_TARGET=10.3 cc', ldflags =' -L/usr/local/lib -L/opt/local/lib'
    libpth=/usr/local/lib /opt/local/lib /usr/lib
    libs=-lgdbm -ldbm -ldl -lm -lc
    perllibs=-ldl -lm -lc
    libc=/usr/lib/libc.dylib, so=dylib, useshrplib=false, libperl=libperl.a
    gnulibc_version=''
  Dynamic Linking:
    dlsrc=dl_dyld.xs, dlext=bundle, d_dlsymun=undef, ccdlflags=' '
    cccdlflags=' ', lddlflags=' -bundle -undefined dynamic_lookup -L/usr/local/lib -L/opt/local/lib'

Locally applied patches:
    


@INC for perl v5.8.6:
    /usr/local/perl58-i/lib/5.8.6/darwin-thread-multi-2level
    /usr/local/perl58-i/lib/5.8.6
    /usr/local/perl58-i/lib/site_perl/5.8.6/darwin-thread-multi-2level
    /usr/local/perl58-i/lib/site_perl/5.8.6
    /usr/local/perl58-i/lib/site_perl
    .


Environment for perl v5.8.6:
    HOME=/Users/timbo
    LANG=en_IE.UTF-8
    LANGUAGE (unset)
    LC_ALL=en_IE.UTF-8
    LD_LIBRARY_PATH (unset)
    LOGDIR (unset)
    PERL_BADLANG (unset)
    SHELL=/bin/bash

p5pRT · 2009-12-16T14:51:13Z

From @timbunce

Any chance someone could take a look at this for 5.12?

I've tried but rapidly went out of my depth (assertion failures etc etc).

Tim.

On Tue, Dec 01, 2009 at 05:20:29AM -0800, Tim Bunce wrote:

$ perl \-Mops=&#8203;:default \-e 'eval \{ ; \}'
'eval "string"' trapped by operation mask at \-e line 1\.
A block eval (entertry opcode) is being treated as a string eval (entereval opcode).

This is a serious problem because most users of Opcode & Safe want to
restrict use of string eval (entereval opcode) but doing so prevents
them using any code/modules that use block eval!

This is true for current blead and at least as far back as 5.8.6.

The bug manifested for me as
$ perl \-Mops=&#8203;:default\,require /Users/timbo/pg/perl598/lib/5\.8\.9/Carp\.pm
'eval "string"' trapped by operation mask at /Users/timbo/pg/perl598/lib/5\.8\.9/Carp\.pm line 33\.
where line 33 is eval { require Carp::Heavy };

(Actually the bug manifested while extending PostgreSQL PL/Perl to support the
warnings pragma, but you don't want to know about that.)

(Note that this works:
$ perl -Mops=:default,require -e 'use Carp; warn $INC{"Carp.pm"}'
/Users/timbo/pg/perl598/lib/5.8.9/Carp.pm at -e line 1.
because Carp was already loaded via ops->Opcode->Carp.)

Using B::Concise shows an entertry opcode:
$ perl \-MO=Concise \-e 'eval \{ ; \}'
5  \<@&#8203;> leave\[1 ref\] vKP/REFC \->$end$
1     \<0> enter \->2
2     \<;> nextstate$main 2 \-e&#8203;:1$ v \->3
4     \<@&#8203;> leavetry vK \->5
3        \<|> entertry$other\->4$ v \->6
6        \<0> stub v \->4
\-e syntax OK
Using perl -DtxT shows that the entertry started out as a entereval:
$ perl \-DtxT \-e 'eval \{ ; \}' 2>&1
\#\#\# 0&#8203;:LEX\_NORMAL/XSTATE "\\n;"
\#\#\# \<== UNIOP$ival=op\_entereval$

\#\#\# 1&#8203;:LEX\_NORMAL/XTERMBLOCK "\{ ; \}\\n"
\#\#\# \<== '\{'
\.\.\.
In toke.c we have:
    case KEY\_eval&#8203;:
        s = SKIPSPACE1$s$; 
        PL\_expect = $\*s == '\{'$ ? XTERMBLOCK : XTERM;
        UNIBRACK$OP\_ENTEREVAL$;
It seems that toke.c should use OP_ENTERTRY instead of OP_ENTEREVAL when
it knows the eval is followed by a block (which it does).

[Please do not change anything below this line]
-----------------------------------------------------------------
---
Flags:
category=core
severity=high
---
Site configuration information for perl v5.8.6:

Configured by timbo at Thu May 25 17:26:37 IST 2006.

Summary of my perl5 (revision 5 version 8 subversion 6) configuration:
Platform:
osname=darwin, osvers=8.5.2, archname=darwin-thread-multi-2level
uname='darwin foo 8.5.2 darwin kernel version 8.5.2: mon feb 13 16:31:48 pst 2006; root:xnu-792.8.37.obj~1release_i386 i386 i386 '
config_args='-des -Dprefix=/usr/local/perl58-i -Doptimize=-g -Duseithreads -Dusemultiplicity'
hint=recommended, useposix=true, d_sigaction=define
usethreads=define use5005threads=undef useithreads=define usemultiplicity=define
useperlio=define d_sfio=undef uselargefiles=define usesocks=undef
use64bitint=undef use64bitall=undef uselongdouble=undef
usemymalloc=n, bincompat5005=undef
Compiler:
cc='cc', ccflags ='-fno-common -DPERL_DARWIN -no-cpp-precomp -DDEBUGGING -fno-strict-aliasing -pipe -I/usr/local/include -I/opt/local/include',
optimize='-g',
cppflags='-no-cpp-precomp -fno-common -DPERL_DARWIN -no-cpp-precomp -DDEBUGGING -fno-strict-aliasing -pipe -I/usr/local/include -I/opt/local/include'
ccversion='', gccversion='4.0.1 (Apple Computer, Inc. build 5250)', gccosandvers=''
intsize=4, longsize=4, ptrsize=4, doublesize=8, byteorder=1234
d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=16
ivtype='long', ivsize=4, nvtype='double', nvsize=8, Off_t='off_t', lseeksize=8
alignbytes=8, prototype=define
Linker and Libraries:
ld='env MACOSX_DEPLOYMENT_TARGET=10.3 cc', ldflags =' -L/usr/local/lib -L/opt/local/lib'
libpth=/usr/local/lib /opt/local/lib /usr/lib
libs=-lgdbm -ldbm -ldl -lm -lc
perllibs=-ldl -lm -lc
libc=/usr/lib/libc.dylib, so=dylib, useshrplib=false, libperl=libperl.a
gnulibc_version=''
Dynamic Linking:
dlsrc=dl_dyld.xs, dlext=bundle, d_dlsymun=undef, ccdlflags=' '
cccdlflags=' ', lddlflags=' -bundle -undefined dynamic_lookup -L/usr/local/lib -L/opt/local/lib'

Locally applied patches:

---
@INC for perl v5.8.6:
/usr/local/perl58-i/lib/5.8.6/darwin-thread-multi-2level
/usr/local/perl58-i/lib/5.8.6
/usr/local/perl58-i/lib/site_perl/5.8.6/darwin-thread-multi-2level
/usr/local/perl58-i/lib/site_perl/5.8.6
/usr/local/perl58-i/lib/site_perl
.

---
Environment for perl v5.8.6:
HOME=/Users/timbo
LANG=en_IE.UTF-8
LANGUAGE (unset)
LC_ALL=en_IE.UTF-8
LD_LIBRARY_PATH (unset)
LOGDIR (unset)
PERL_BADLANG (unset)
SHELL=/bin/bash

p5pRT · 2009-12-20T11:13:26Z

From @rgarcia

2009/12/1 Tim Bunce <perlbug-followup@perl.org>:

$ perl -Mops=:default -e 'eval { ; }'
'eval "string"' trapped by operation mask at -e line 1.

A block eval (entertry opcode) is being treated as a string eval (entereval opcode).

This is a serious problem because most users of Opcode & Safe want to
restrict use of string eval (entereval opcode) but doing so prevents
them using any code/modules that use block eval!

This should be fixed by commit 32e2a35.
Note that this is a core bug, not a Safe/ops bug: it's not possible to
backport it to older perls.

p5pRT · 2009-12-20T11:13:29Z

The RT System itself - Status changed from 'new' to 'open'

p5pRT · 2009-12-20T11:16:28Z

@rgs - Status changed from 'open' to 'resolved'

p5pRT closed this as completed Dec 20, 2009

p5pRT added Severity High distro-darwin type-core labels Oct 18, 2019

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Opcode module (and thus Safe) think block eval is a string eval #9999

Opcode module (and thus Safe) think block eval is a string eval #9999

p5pRT commented Dec 1, 2009

p5pRT commented Dec 1, 2009

p5pRT commented Dec 16, 2009

p5pRT commented Dec 20, 2009

p5pRT commented Dec 20, 2009

p5pRT commented Dec 20, 2009

Opcode module (and thus Safe) think block eval is a string eval #9999

Opcode module (and thus Safe) think block eval is a string eval #9999

Comments

p5pRT commented Dec 1, 2009

p5pRT commented Dec 1, 2009

From @timbunce

p5pRT commented Dec 16, 2009

From @timbunce

p5pRT commented Dec 20, 2009

From @rgarcia

p5pRT commented Dec 20, 2009

p5pRT commented Dec 20, 2009