pike issueshttps://git.lysator.liu.se/pikelang/pike/-/issues2010-04-22T20:06:43Zhttps://git.lysator.liu.se/pikelang/pike/-/issues/5072Segfault in pike 7.8.352 (when running roxen-5.0.403)2010-04-22T20:06:43ZPeter BortasSegfault in pike 7.8.352 (when running roxen-5.0.403)Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=5072
Reported by Anders Blomdell, Reglerteknik, LTH <anders.blomdell@control.lth.se>
```
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x727cb...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=5072
Reported by Anders Blomdell, Reglerteknik, LTH <anders.blomdell@control.lth.se>
```
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x727cb90 (LWP 3647)]
gc_zap_ext_weak_refs_in_objects () at /tmp/leif/Pike-v7.8.352/src/object.c:2303
warning: Source file is more recent than executable.
2303 if (o->refs)
(gdb) list
2298 {
2299 gc_mark_object_pos = first_object;
2300 while (gc_mark_object_pos != gc_internal_object && gc_ext_weak_refs) {
2301 struct object *o = gc_mark_object_pos;
2302 gc_mark_object_pos = o->next;
2303 if (o->refs)
2304 gc_mark_object_as_referenced(o);
2305 }
2306 gc_mark_discard_queue();
2307 }
(gdb) print o
No symbol "o" in current context.
(gdb) print gc_mark_object_pos
$1 = (struct object *) 0xa490cbc
(gdb) print *gc_mark_object_pos
$2 = {refs = 0, prog = 0x0, next = 0xa49017c, prev = 0xb774c434, storage = 0x0}
(gdb) print *gc_mark_object_pos->next
$3 = {refs = 0, prog = 0x0, next = 0xa490d20, prev = 0xa490cbc, storage = 0x0}
(gdb) print *gc_mark_object_pos->prev
$4 = {refs = 1, prog = 0x942337c, next = 0xa490cbc, prev = 0xb774c420, storage = 0x0}
(gdb) print first_object
$5 = (struct object *) 0xa7449b8
(gdb) print gc_internal_object
$6 = (struct object *) 0xb774be30
```
A. Can refs=1 and storage=0x0 be the problem
B. Is there a way to force gc to run, in order to invetsigate problem further.Pike 7.8https://git.lysator.liu.se/pikelang/pike/-/issues/4872Only integral part of a float is outputted when the fractional part is 0.2009-06-30T16:15:10ZPeter BortasOnly integral part of a float is outputted when the fractional part is 0.Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=4872
Reported by Mathias Södermark <mathias@roxen.com>
Consider the following code and its output:
<set variable="var.float1" expr="(float) 1.0" />
<set variable="var....Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=4872
Reported by Mathias Södermark <mathias@roxen.com>
Consider the following code and its output:
<set variable="var.float1" expr="(float) 1.0" />
<set variable="var.float2" value="1.0" type="float"/>
Float 1: &var.float1; <debug showvar="var.float1"/><br/>
Float 2: &var.float2; <debug showvar="var.float2"/><br/>
Output:
Float 1: 1
Float 2: 1
The problem is only in the representation. It would be nice if floats are written with at least one decimal to distinguish them from integers.Pike 7.8Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/4822ODBC module broken on Windows2009-05-11T13:38:04ZPeter BortasODBC module broken on WindowsImported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=4822
Reported by Marcus Wellhardh <wellhard@roxen.com>
The ODBC module does not work on Windows. I get the following backtrace for any query:
```
F:\Roxen CMS-5.0.303\se...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=4822
Reported by Marcus Wellhardh <wellhard@roxen.com>
The ODBC module does not work on Windows. I get the following backtrace for any query:
```
F:\Roxen CMS-5.0.303\server-5.0.303\pike\bin>pike
Pike v7.8 release 220 running Hilfe v3.5 (Incremental Pike Frontend)
> object db = Sql.Sql("odbc://test:test@sparq");
> array(mapping) res = db->query("select PFId from aspnet_UsersPF");
odbc->fetch_row(): SQLGetData() failed:
-1:22003:[Microsoft][SQL Native Client]Numeric value out of range
t:\5.0\build\pike.win32_x86\pike\src\modules\Odbc\odbc_result.c:728: Odbc.odbc_r
esult()->fetch_row()
F:/Roxen CMS-5.0.303/server-5.0.303/pike/lib/modules/Sql.pmod/Sql.pike:372: Sql.
Sql()->res_obj_to_array(Odbc.odbc_result())
F:/Roxen CMS-5.0.303/server-5.0.303/pike/lib/modules/Sql.pmod/Sql.pike:500: Sql.
Sql()->query("select PFId from aspnet_UsersPF")
```
Regression testing:
Roxen 4.5.146 (pike 7.4.464) works
Roxen 4.5.230 (pike 7.4.526) crashes
Roxen 4.5.307 (pike 7.4.592) gives this backtrace
Tested on krokodil.roxen.com.Pike 7.8Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/4649BACKPORT: sendfile truncation issue (Was: Files exceeding proto cache size ar...2009-04-16T14:11:39ZPeter BortasBACKPORT: sendfile truncation issue (Was: Files exceeding proto cache size are corrupted)Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=4649
Reported by Jonas Wallden <jonasw@roxen.com>
Requesting a file whose size is bigger than the protocol cache threshold will not be sent completely.
Both Safari and Fi...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=4649
Reported by Jonas Wallden <jonasw@roxen.com>
Requesting a file whose size is bigger than the protocol cache threshold will not be sent completely.
Both Safari and Firefox will stall and later give up with bytes missing at the end of the file (a few
hundred missing bytes is the normal case).
The same test works fine in 4.5.Pike 7.8Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/4631Parsing problems in XMLRPC module (and in the xml parser)2013-08-28T15:12:22ZPeter BortasParsing problems in XMLRPC module (and in the xml parser)Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=4631
Reported by Manifest0 , <demanuel@gmail.com>
The xmlrpc can't process the opensubtitles xml response.
```
Steps to reproduce:
Protocols.XMLRPC.Client cliente = Prot...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=4631
Reported by Manifest0 , <demanuel@gmail.com>
The xmlrpc can't process the opensubtitles xml response.
```
Steps to reproduce:
Protocols.XMLRPC.Client cliente = Protocols.XMLRPC.Client("http://
www.opensubtitles.org/xml-rpc");
string auth_token = cliente["LogIn"]("","","en","python")[0]["token"];
array(mapping(string:string)) map =({});
mapping(string:string) params = ([]);
params["sublanguageid"]="eng";
params["moviehash"]="8c7419a68bd85bd8";
params["moviebytesize"]="733700096";
map+=({params});
cliente["SearchSubtitles"](auth_token,map);
```
Error:
Bad arguments.
Unknown program: `*(0,"")
Other info:
1- It seems that the bug is both in the xml parser and in the xmlrpc module
2- You can see the xml response by using the following code:
```
Protocols.XMLRPC.Client cliente = Protocols.XMLRPC.Client("http://
www.opensubtitles.org/xml-rpc");
string auth_token = cliente["LogIn"]("","","en","python")[0]["token"];
array(mapping(string:string)) map =({});
mapping(string:string) params = ([]);
params["sublanguageid"]="eng";
params["moviehash"]="8c7419a68bd85bd8";
params["moviebytesize"]="733700096";
map+=({params});
object c=Protocols.HTTP.do_method("POST", cliente->url, 0,
(["Content-Type":"text/xml"]), 0,
Protocols.XMLRPC.encode_call("SearchSubtitles", ({ auth_token, map })));
string result=c->data();
write(result);
```
I hope o didn't forget anything.
Best regards!Pike 7.6Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/4537Array gets emptied if modified in function2009-04-16T14:11:39ZPeter BortasArray gets emptied if modified in functionImported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=4537
Reported by @zino
Pike 7.7.45 from CVS as of a few hours ago:
```
% cat bug_array.pike
void fiddle(array a)
{
a += ({ 2 });
}
void main()
{
array a = ({});...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=4537
Reported by @zino
Pike 7.7.45 from CVS as of a few hours ago:
```
% cat bug_array.pike
void fiddle(array a)
{
a += ({ 2 });
}
void main()
{
array a = ({});
fiddle( a );
if(!sizeof(a))
werror("Array is broken: %O\n", a);
else
werror("Array is ok: %O\n", a);
}
% ~/bin/pike77-2/bin/pike bug_array.pike
Array is broken: ({ })
```
(Initially reported by Tim Johansson, Opera Software)Pike 7.8Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/4174(file << "foo") << "bar" don't compile on pike 7.6 (work fine with 7.2) (See ...2009-04-16T14:11:39ZPeter Bortas(file << "foo") << "bar" don't compile on pike 7.6 (work fine with 7.2) (See also #2996, #4156)Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=4174
Reported by Andrew Shirrayev, <andrews@gate.ort.spb.ru>
```
~$ pike7.2
Pike v7.2 release 580 running Hilfe v2.0 (Incremental Pike Frontend)
> (Stdio.stdout << "BAR!\...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=4174
Reported by Andrew Shirrayev, <andrews@gate.ort.spb.ru>
```
~$ pike7.2
Pike v7.2 release 580 running Hilfe v2.0 (Incremental Pike Frontend)
> (Stdio.stdout << "BAR!\n") << "FOO!\n";
BAR!
FOO!
Result: object
Terminal closed.
~$ pike7.6
Pike v7.6 release 75 running Hilfe v3.5 (Incremental Pike Frontend)
> (Stdio.stdout << "BAR!\n") << "FOO!\n";
Compiler Error: 1:Bad argument 2 to `<<.
Compiler Error: 1:Expected: !function(!object ... : mixed) &
(function(mixed, object : mixed) | function(object, mixed : mixed)) |
function(int, int : int)
Compiler Error: 1:Got : function(mixed, string : void | mixed)
Terminal closed.
```Pike 7.6Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/4052ORA-03106: fatal two-task communication protocol error with Oracle 9i+2015-11-05T17:43:41ZPeter BortasORA-03106: fatal two-task communication protocol error with Oracle 9i+Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=4052
Reported by Peter J. Holzer, WSR <hjp@wsr.ac.at>
```
Pike v7.4 release 340 Copyright &#65533; 1994-2004 Linköping University
(part of the Roxen 4.0.425 distribution)...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=4052
Reported by Peter J. Holzer, WSR <hjp@wsr.ac.at>
```
Pike v7.4 release 340 Copyright � 1994-2004 Linköping University
(part of the Roxen 4.0.425 distribution):
```
When pike is built against Oracle 9.2.0.1 or Oracle 10.2, some (but not all)
queries on an Oracle 9.2.0.6 database result in an ORA-03106 error after
the first row of the result set is returned. The problem always occurs for
the same queries, but I haven't found a pattern, which queries are affected.
When pike is built against Oracle 8.1.7 OR the database is Oracle 8.1.7,
the problem doesn't occur - both the database and the libraries need to be
9i or later to show the problem.
To reproduce:
```
In sqlplus:
set autocommit on
create table foo (KST0 int, kst1 int, name varchar2(80), aktuell number(1));
insert into foo values(1,2, 'Test', 1);
insert into foo values(3, 4, 'fasel', 0);
```
In Roxen Admin interface:
SELECT KST0, KST1, NAME, AKTUELL FROM FOO => Run Query
Works.
```
again in sqlplus:
delete from foo where kst0=3;
insert into foo values(0, 2, 'Administratives', 1);
```
Click again on Run Query in Roxen:
KST0 KST1 NAME AKTUELL
0 2 Administratives 1
While running SELECT KST0, KST1, NAME, AKTUELL FROM FOO :
OCIStmtFetch:code=-1:ORA-03106: fatal two-task communication protocol error
Query took 0.001s, 1 rows in the replyPike 7.4Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/3916Pike 7.6 and 7.7 doesn't compile on FreeBSD 5.4-RELEASE2009-04-16T14:11:39ZPeter BortasPike 7.6 and 7.7 doesn't compile on FreeBSD 5.4-RELEASEImported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3916
Reported by Xavier Beaudouin, Xavier Beaudouin <kiwi@oav.net>
On FreeBSD 5.4-RELEASE (as well on 5.3) compiler fails with :
```
Making GL
gmake[5]: Entering directo...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3916
Reported by Xavier Beaudouin, Xavier Beaudouin <kiwi@oav.net>
On FreeBSD 5.4-RELEASE (as well on 5.3) compiler fails with :
```
Making GL
gmake[5]: Entering directory `/home/kiwi/xenoclient/pike-7.6/glopy.fr.tiscali.com/buildtmp/
Pike-v7.6-snapshot/build/freebsd-5.4-release-i386/post_modules/GL'
gmake[6]: Entering directory `/home/kiwi/xenoclient/pike-7.6/glopy.fr.tiscali.com/buildtmp/
Pike-v7.6-snapshot/build/freebsd-5.4-release-i386/post_modules/GL'
Compiling /home/kiwi/xenoclient/pike-7.6/glopy.fr.tiscali.com/buildtmp/Pike-v7.6-snapshot/
src/post_modules/GL/top.c
/home/kiwi/xenoclient/pike-7.6/glopy.fr.tiscali.com/buildtmp/Pike-v7.6-snapshot/build/
freebsd-5.4-release-i386/pike -DNOT_INSTALLED -DPRECOMPILED_SEARCH_MORE -m/home/
kiwi/xenoclient/pike-7.6/glopy.fr.tiscali.com/buildtmp/Pike-v7.6-snapshot/build/freebsd-5.4-
release-i386/master.pike /home/kiwi/xenoclient/pike-7.6/glopy.fr.tiscali.com/buildtmp/Pike-
v7.6-snapshot/src/post_modules/GL/gen.pike < /home/kiwi/xenoclient/pike-7.6/
glopy.fr.tiscali.com/buildtmp/Pike-v7.6-snapshot/src/post_modules/GL/auto.c.in > auto.c
Fatal error 'Spinlock called when not threaded.' at line 87 in file /usr/src/lib/libpthread/thread/
thr_spinlock.c (errno = 2)
Abort trap (core dumped)
gmake[6]: *** [auto.c] Error 134
```
This bug is also present on pike 7.7 as well.
Please see xenofarm for glopy.fr.tiscali.com (warning this machine will be dead Friday 13th
because I quit Tiscali as job).
http://pike.ida.liu.se/generated/pikefarm/7.6/271_201/makelog.html
http://pike.ida.liu.se/generated/pikefarm/7.7/900_161/makelog.htmlPike 7.6Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/3812Sort is broken2009-04-16T14:11:39ZPeter BortasSort is brokenImported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3812
Reported by Fredrik Noring, Roxen Internet Software <noring@roxen.com>
The following works in Pike 7.2 but fails in Pike 7.4:
```
int main()
{
array a = ({ /* 94 ...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3812
Reported by Fredrik Noring, Roxen Internet Software <noring@roxen.com>
The following works in Pike 7.2 but fails in Pike 7.4:
```
int main()
{
array a = ({ /* 94 elements */
90839040,
96352256,
80316416,
88322048,
73676800,
77709312,
101908480,
72744960,
54609920,
251849736,
62978048,
106463232,
820152320,
111521792,
43429888,
72683520,
99155968,
106059776,
69969920,
131769352,
34936832,
54810624,
90839040,
94869504,
94869504,
106463232,
56076288,
106059776,
99418112,
115529728,
100892672,
64229376,
100124672,
110780416,
101908480,
290721792,
58802176,
111521792,
68894720,
58322944,
93726720,
95713280,
79521792,
157540352,
69969920,
115529728,
108609536,
72300544,
72744960,
88031232,
95713280,
120080384,
110376960,
63164416,
86994944,
103079936,
68622336,
54810624,
60518400,
103079936,
107485184,
58411008,
366772224,
72769536,
58322944,
291342336,
97054720,
64915456,
68485120,
72769536,
99155968,
4792706056,
121389056,
60237824,
56006656,
60237824,
60518400,
68485120,
108003328,
88322048,
58802176,
108072960,
195770368,
93726720,
88031232,
73601024,
110780416,
54609920,
53366784,
142299136,
70117376,
64229376,
85239808,
53366784
});
a = map((array(string))a, Gmp.mpz, 10);
array b = (array(string))a;
sort(a, b);
write("%O\n", a);
return 0;
}
```Pike 7.4https://git.lysator.liu.se/pikelang/pike/-/issues/3618BACKPORT: Type error segfault.2009-04-16T14:11:39ZPeter BortasBACKPORT: Type error segfault.Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3618
Reported by Martin Nilsson, IDA <nilsson@pike.ida.liu.se>
```
The code
Stdio.File o = Stdio.FakeFile("");
```
generates
```
Program received signal SIGSEGV, Segmen...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3618
Reported by Martin Nilsson, IDA <nilsson@pike.ida.liu.se>
```
The code
Stdio.File o = Stdio.FakeFile("");
```
generates
```
Program received signal SIGSEGV, Segmentation fault.
0x42070664 in _IO_default_xsputn () from /lib/i686/libc.so.6
(gdb) bt
#0 0x42070664 in _IO_default_xsputn () from /lib/i686/libc.so.6
#1 0x42047e92 in vfprintf () from /lib/i686/libc.so.6
#2 0x4206aa54 in vsnprintf () from /lib/i686/libc.so.6
#3 0x080f61f0 in va_yyerror (fmt=0x42135980 "", args=0x42135980 "")
at /home/nilsson/Pike/7.5/src/program.c:5850
#4 0x080f6218 in my_yyerror (fmt=0x42135980 "")
at /home/nilsson/Pike/7.5/src/program.c:5858
#5 0x080d4d53 in yyexplain_nonmatching_types (type_a=0x4213920c,
type_b=Cannot access memory at address 0xbf7ffffc
)
at /home/nilsson/Pike/7.5/src/pike_types.c:4479
#6 0x080f930b in yyexplain_not_compatible (a=0x831697c, b=0x8447078, flags=0)
at /home/nilsson/Pike/7.5/src/program.c:7789
#7 0x080f930b in yyexplain_not_compatible (a=0x831697c, b=0x8447078, flags=0)
at /home/nilsson/Pike/7.5/src/program.c:7789
#8 0x080f930b in yyexplain_not_compatible (a=0x831697c, b=0x8447078, flags=0)
```
...Pike 7.4Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/3577Mysql library glue uses incorrect memory management2009-04-16T14:11:39ZPeter BortasMysql library glue uses incorrect memory managementImported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3577
Reported by Stephen R. van den Berg, Cubic Circle <srb@cuci.nl>
This bug has been discussed at length on the Pike mailinglist.
It was confirmed by the designer of th...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3577
Reported by Stephen R. van den Berg, Cubic Circle <srb@cuci.nl>
This bug has been discussed at length on the Pike mailinglist.
It was confirmed by the designer of the sql-library-interface that this
patch indeed fixes a flaw in the implementation of the mysql-glue.
It was also noted that quite some implementations (wrongly) seem to depend
on the misguided old behaviour of the lib.
Consensus was that there might be a need to have a bug-compatibility flag
to reinstate the old behaviour.Pike 7.8Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/3507Global #defines don't propagate in version compat.2009-04-16T14:11:39ZPeter BortasGlobal #defines don't propagate in version compat.Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3507
Reported by @grubba
```
$ pike -DFOOBAR=GAZONK
Pike v7.5 release 12 running Hilfe v3.5 (Incremental Pike Frontend)
> FOOBAR;
Compiler Error: 1:Undefined identifier "...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3507
Reported by @grubba
```
$ pike -DFOOBAR=GAZONK
Pike v7.5 release 12 running Hilfe v3.5 (Incremental Pike Frontend)
> FOOBAR;
Compiler Error: 1:Undefined identifier "GAZONK".
Terminal closed.
$ pike -DFOOBAR=GAZONK -V7.4
Pike v7.5 release 12 running Hilfe v3.5 (Incremental Pike Frontend)
> FOOBAR;
Compiler Error: 1:Undefined identifier "FOOBAR".
Terminal closed.
```Pike 7.6Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/3465Mysql use mysql_store_result() instead of mysql_use_result()2009-04-16T14:11:39ZPeter BortasMysql use mysql_store_result() instead of mysql_use_result()Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3465
Reported by Stephen R. van den Berg, Cubic Circle <srb@cuci.nl>
The implementation of the Mysql.so module uses the mysql_store_result()
function to actually obtain t...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3465
Reported by Stephen R. van den Berg, Cubic Circle <srb@cuci.nl>
The implementation of the Mysql.so module uses the mysql_store_result()
function to actually obtain the result from the database. This will always
result in the full query being allocated in memory.
This is *not* good, of course.
The intention was that the fetch_row primitives actually do something
usefull. Merely using mysql_use_result() instead of mysqL_store_result()
will result in overall lower memory requirements and increased speed
(needless to say, since Roxen leans heavily on Mysql internally, it seems
rather silly that this was never corrected before).Pike 7.2Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/3441global. coredump2011-02-08T11:56:13ZPeter Bortasglobal. coredumpImported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3441
Reported by Johan H Sundström, IDA <jhs@pike.ida.liu.se>
This code produces a coredumpwith pike 7.4 at pelix:
```
int a;
int main()
{
int b;
b += global.a;
}
``...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3441
Reported by Johan H Sundström, IDA <jhs@pike.ida.liu.se>
This code produces a coredumpwith pike 7.4 at pelix:
```
int a;
int main()
{
int b;
b += global.a;
}
```
Substituting "+=" for "=" instead renders the error:
```
core.pike:5:Assigning a void expression.
Pike: Failed to compile script:
Compilation failed.
```Pike 7.8Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/3438Bad data to decode_value crashes pike2009-04-16T14:11:39ZPeter BortasBad data to decode_value crashes pikeImported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3438
Reported by Stefan Wallström <stewa@lysator.liu.se>
```
decode_value("¶ke0\1\2\6\7__edit_flag`h\6\0page`f1\6\n__sb_edit_area`&\6\7cge`fn\6e1eli_\5oxie&_ireeet`t`t\6x...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3438
Reported by Stefan Wallström <stewa@lysator.liu.se>
```
decode_value("¶ke0\1\2\6\7__edit_flag`h\6\0page`f1\6\n__sb_edit_area`&\6\7cge`fn\6e1eli_\5oxie&_ireeet`t`t\6xd\0soe`e1eli_`\6\0al`t`t\6x\6\6m0lmrl_f`r");
/home/stewa/Pike7.2/src/encode.c:1527: Fatal error:
error in type string.
Attempting to dump backlog (may fail)...
Backtrace at time of fatal:
Optimizer:0: ___Foo4711()
/usr/local/pike/7.2.462/lib/master.pike:208: master()->compile_string()
```Pike 7.2Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/3435Unexpected error from thread function: 62009-04-16T14:11:39ZPeter BortasUnexpected error from thread function: 6Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3435
Reported by Tomas Nilsson, exRoxen <tomas.nilsson@roxen.com>
Pike 7.2 on Windows XP fails with the messages below.
The error does not occur with sources from 2003-05...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3435
Reported by Tomas Nilsson, exRoxen <tomas.nilsson@roxen.com>
Pike 7.2 on Windows XP fails with the messages below.
The error does not occur with sources from 2003-05-06.
```
H:\tmp>pike72n
Pike v7.2 release 496 running Hilfe v2.0 (Incremental Pike Frontend)
> system.gethostbyname("pyton");
\codework\pike\72o\src\threads.c:38: Fatal error:
Unexpected error from thread function: 6
Attempting to dump backlog (may fail)...
Backtrace at time of fatal:
system: gethostbyname("pyton")
-:1: ___Foo4711()
h:/codework/pike/72o/lib/modules/Tools.pmod/Hilfe.pmod:112:
do_evaluate("mixed _
__Foo4711() { return (mixed)(system.gethostbyname(\"pyton\")); }\n",1)
h:/codework/pike/72o/lib/modules/Tools.pmod/Hilfe.pmod:624:
parse_statement("sys
tem.gethostbyname(\"pyton\");")
h:/codework/pike/72o/lib/modules/Tools.pmod/Hilfe.pmod:402: do_parse()
h:/codework/pike/72o/lib/modules/Tools.pmod/Hilfe.pmod:244:
add_buffer("system.g
ethostbyname(\"pyton\");\n")
h:/codework/pike/72o/lib/modules/Tools.pmod/Hilfe.pmod:644:
add_input_line("syst
em.gethostbyname(\"pyton\");\n")
h:/codework/pike/72o/lib/modules/Tools.pmod/Hilfe.pmod:724: create()
Hilfe: StdinHilfe()
```
H:\tmp>Pike 7.2Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/3415Probable code-generator bug caught in F_APPLY.2009-04-16T14:11:39ZPeter BortasProbable code-generator bug caught in F_APPLY.Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3415
Reported by Jeff Utter, <funk@softhome.net>
Hello, i'm using Caudium 1.3 w/ Pike 7.4.21 It seems theres a bug in
interpert.c that causes pike to segfault often.. her...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3415
Reported by Jeff Utter, <funk@softhome.net>
Hello, i'm using Caudium 1.3 w/ Pike 7.4.21 It seems theres a bug in
interpert.c that causes pike to segfault often.. here's the gdb output:
```
Program received signal SIGSEGV, Segmentation fault.
0x0806e0ca in low_mega_apply (type=APPLY_SVALUE, args=3, arg1=0x288d90a8,
arg2=0x0) at /root/pike-7.4.21/src/interpret.c:1569
1569 assign_svalue(save_sp,Pike_sp-1);
(gdb) bt
#0 0x0806e0ca in low_mega_apply (type=APPLY_SVALUE, args=3, arg1=0x288d90a8,
arg2=0x0) at /root/pike-7.4.21/src/interpret.c:1569
#1 0x0806c701 in opcode_F_APPLY (arg1=137568256) at interpret_functions.h:1841
#2 0x08a83e18 in ?? ()
#3 0x0806f78d in o_catch (pc=0x88ea711 "¡0X\037\b\203@\034\rÇ\004$¿")
at /root/pike-7.4.21/src/interpret.c:1757
#4 0x0806a436 in opcode_F_CATCH () at interpret_functions.h:1201
#5 0x088ea70d in ?? ()
#6 0x0806f6c0 in mega_apply (type=APPLY_STACK, args=2, arg1=0x0, arg2=0x0)
at /root/pike-7.4.21/src/interpret.c:1713
#7 0x0806f7e4 in f_call_function (args=2)
at /root/pike-7.4.21/src/interpret.c:1777
#8 0x080e7dda in new_thread_func (data=0x8a2cd30)
at /root/pike-7.4.21/src/threads.c:772
#9 0x281f148e in _thread_start () from /usr/lib/libc_r.so.5
```
I'm not much of a C coder at all.. hope this can help someone out.Pike 7.4Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/3278mixed x={} dumps core2009-04-16T14:11:39ZPeter Bortasmixed x={} dumps coreImported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3278
Reported by Martin Nilsson, IDA <nilsson@pike.ida.liu.se>
bin/pike --gdb -e "mixed x={}"
gives
```
-:1:parse error
Program received signal SIGSEGV, Segmentation f...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3278
Reported by Martin Nilsson, IDA <nilsson@pike.ida.liu.se>
bin/pike --gdb -e "mixed x={}"
gives
```
-:1:parse error
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 1024 (LWP 28011)]
recursive_add_call_arg (n=0x0, arg=0x82d1518)
at /home/nilsson/Pike/7.3/src/las.c:1714
1714 switch(n->token)
(gdb) bt
#0 recursive_add_call_arg (n=0x0, arg=0x82d1518)
at /home/nilsson/Pike/7.3/src/las.c:1714
#1 0x0806358c in yyparse () at /home/nilsson/Pike/7.3/src/language.yacc:1777
#2 0x080f11b5 in run_pass1 (c=0x82f0770)
at /home/nilsson/Pike/7.3/src/program.c:5466
#3 0x080f17ed in compile (aprog=0x82f0650, ahandler=0x0, amajor=-1,
aminor=-1, atarget=0x0, aplaceholder=0x0)
at /home/nilsson/Pike/7.3/src/program.c:5731
#4 0x08122691 in f_compile (args=4)
at /home/nilsson/Pike/7.3/src/builtin_functions.c:3225
#5 0x080702d8 in opcode_F_CALL_BUILTIN (arg1=15)
at /home/nilsson/Pike/7.3/src/interpret_functions.h:2050
#6 0x083203f8 in ?? ()
#7 0x08072b00 in mega_apply (type=APPLY_LOW, args=2, arg1=0x82a0618,
arg2=0x52) at /home/nilsson/Pike/7.3/src/interpret.c:1431
#8 0x08073657 in apply (o=0x82a0618, fun=0x8198077 "_main", args=2)
at /home/nilsson/Pike/7.3/src/interpret.c:1734
#9 0x080ca5d1 in main (argc=4, argv=0xbffffa8c)
at /home/nilsson/Pike/7.3/src/main.c:726
#10 0x400c7336 in __libc_start_main (main=0x80c9c30 <main>, argc=4,
ubp_av=0xbffffa8c, init=0x805f96c <_init>, fini=0x81771d0 <_fini>,
rtld_fini=0x4000d2fc <_dl_fini>, stack_end=0xbffffa7c)
at ../sysdeps/generic/libc-start.c:129
```Pike 7.4Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbströmhttps://git.lysator.liu.se/pikelang/pike/-/issues/3269Crash in stringbuilder2009-04-16T14:11:39ZPeter BortasCrash in stringbuilderImported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3269
Reported by David Hedbor, Idonex / Real Networks <david@hedbor.org>
This crash bug occurs at such an early stage that pike can't run at all. I
see two different resu...Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3269
Reported by David Hedbor, Idonex / Real Networks <david@hedbor.org>
This crash bug occurs at such an early stage that pike can't run at all. I
see two different results depending on if I have assembler enabled or not
(I believe that these are the only differences).
```
at /home/neotron/pike/7.3/src/stralloc.c:2050
#1 0x0811783c in append_path_unix (s=0xbffff1a8, path=
{ptr = 0x82e6cc4
"/home/neotron/pike/build/7.3-nodebug-asm/lib/modules/Getopt.pmod", shift =
0}, len=64) at /home/neotron/pike/7.3/src/combine_path.h:73
#2 0x081179be in f_combine_path_unix (args=2)
at /home/neotron/pike/7.3/src/stralloc.h:89
#3 0x0806ecfb in opcode_F_CALL_BUILTIN_AND_RETURN (arg1=-1073745512)
at /home/neotron/pike/7.3/src/interpret_functions.h:2059
#4 0x0831a15a in ?? ()
#5 0x08071510 in mega_apply (type=3221221784, args=-1073745512,
arg1=0xbffff198, arg2=0xbffff198)
at /home/neotron/pike/7.3/src/interpret.c:1415
#6 0x08072139 in apply (o=0x402d7000,
fun=0xbffff198 "\200m.\b\220\034\017\b\001", args=65536)
at /home/neotron/pike/7.3/src/interpret.c:1718
#7 0x080c18fd in main (argc=7, argv=0xbffff3c4)
at /home/neotron/pike/7.3/src/main.c:726
#8 0x400b7082 in __libc_start_main () from /lib/i686/libc.so.6
(gdb) p *s
$1 = {s = 0xbffff198, malloced = 256, known_shift = 0}
(gdb) p *s->s
$2 = {refs = 137260416, size_shift = 135208080, len = 1, hval = 27,
next = 0xbffff198, str = ""}
and without assembler
Starting program: /home/neotron/pike/build/7.3-nodebug/pike
-m/home/neotron/pike/build/7.3-nodebug/master.pike
/home/neotron/pike/7.3/bin/test_pike.pike --no-watchdog -a -v -v
/home/neotron/pike/7.3/src/stralloc.c:155: Fatal error:
Breakpoint 1, debug_fatal (fmt=0x1 <Address 0x1 out of bounds>)
at /home/neotron/pike/7.3/src/error.c:371
371 if (in_fatal)
(gdb) cont
Continuing.
Illegal shift size!
Attempting to dump backlog (may fail)...
Unrecognized backtrace format: combine_path_with_cwd
```
Program received signal SIGABRT, Aborted.
Mandrake Cooker, gcc 3.2Pike 7.4Henrik (Grubba) GrubbströmHenrik (Grubba) Grubbström