Pike-devel February 2003

pike-devel@lists.lysator.liu.se

10 participants
67 discussions

Weird problem with Image.SVG in Pike 7.4/7.5
by Marek Habersack 12 Feb '03

12 Feb '03

Hey all, I'm having a problem with Image.SVG in that when trying to do indices(Image.SVG) then Pike hangs quite badly - the only way to stop it is to kill it. It happens only in the Debian packaged version for both Pike versions, the version compiled from the tarballs works fine. Until recently the Debian package was using the --traditional style of installation, but now it uses a modified --new-style one which in 90% resembles the default style with minor differences: - all the pike code is installed in /usr/share/pike/X.Y.Z/ - all the binary files (.o, pike, .so) are in /usr/lib/pike/X.Y.Z/ - the files in /usr/share/pike/X.Y.Z/ are symlinked to from the /usr/lib/pike/X.Y.Z/ so that the layout looks like that: $ pwd /usr/lib/pike/7.5.3/lib/modules ... lrwxrwxrwx 1 root root 59 2003-02-06 23:40 _Image_DWG.pmod -> ../../../../../share/pike/7.5.3/lib/modules/_Image_DWG.pmod -rw-r--r-- 1 root root 6062 2003-02-06 19:40 _Image_DWG.pmod.o -rw-r--r-- 1 root root 10908 2003-02-06 19:42 _Image_FreeType.so -rw-r--r-- 1 root root 39188 2003-02-06 19:42 _Image_GIF.so -rw-r--r-- 1 root root 32844 2003-02-06 19:42 _Image_JPEG.so drwxr-xr-x 2 root root 94 2003-02-06 23:40 _Image.pmod lrwxrwxrwx 1 root root 59 2003-02-06 23:40 _Image_PSD.pmod -> ../../../../../share/pike/7.5.3/lib/modules/_Image_PSD.pmod -rw-r--r-- 1 root root 20777 2003-02-06 19:40 _Image_PSD.pmod.o lrwxrwxrwx 1 root root 58 2003-02-06 23:40 _Image_PS.pmod -> ../../../../../share/pike/7.5.3/lib/modules/_Image_PS.pmod -rw-r--r-- 1 root root 8087 2003-02-06 19:40 _Image_PS.pmod.o -rw-r--r-- 1 root root 768976 2003-02-06 19:42 Image.so -rw-r--r-- 1 root root 8156 2003-02-06 19:42 _Image_SVG.so -rw-r--r-- 1 root root 2964 2003-02-06 19:42 _Image_TIFF.so -rw-r--r-- 1 root root 22540 2003-02-06 19:42 _Image_TTF.so lrwxrwxrwx 1 root root 59 2003-02-06 23:40 _Image_XCF.pmod -> ../../../../../share/pike/7.5.3/lib/modules/_Image_XCF.pmod -rw-r--r-- 1 root root 45024 2003-02-06 19:40 _Image_XCF.pmod.o -rw-r--r-- 1 root root 10756 2003-02-06 19:42 _Image_XFace.so lrwxrwxrwx 1 root root 59 2003-02-06 23:40 _Image_XPM.pmod -> ../../../../../share/pike/7.5.3/lib/modules/_Image_XPM.pmod -rw-r--r-- 1 root root 11477 2003-02-06 19:40 _Image_XPM.pmod.o lrwxrwxrwx 1 root root 53 2003-02-06 23:40 Java.pmod -> ../../../../../share/pike/7.5.3/lib/modules/Java.pmod -rw-r--r-- 1 root root 44761 2003-02-06 19:40 Java.pmod.o so, as far as pike is concerned, the entire tree is in /usr/lib/pike/X.Y.Z/, the module/include paths are: $ pike7.5 --show-paths master.pike...: /usr/lib/pike/7.5.3/lib/master.pike Module path...: /usr/local/share/pike7.5/site_pike /usr/local/share/pike/site_pike /usr/local/lib/pike7.5/site_pike /usr/local/lib/pike/site_pike /usr/lib/pike/7.5.3/lib/modules Include path..: /usr/local/include/pike7.5 /usr/local/include/pike /usr/lib/pike/7.5.3/lib/include Program path..: $ pike7.4 --show-paths master.pike...: /usr/lib/pike/7.4.13/lib/master.pike Module path...: /usr/local/share/pike7.4/site_pike /usr/local/share/pike/site_pike /usr/local/lib/pike7.4/site_pike /usr/local/lib/pike/site_pike /usr/lib/pike/7.4.13/lib/modules Include path..: /usr/local/include/pike7.4 /usr/local/include/pike /usr/lib/pike/7.4.13/lib/include Program path..: Running pike with -t gives the following result (output comes from pike 7.5): - master.pike: 688: 8298f90->low_findprog("/usr/lib/pike/7.5.3/lib/modules/_Image_SVG",".so",0,1) - master.pike: 516: 8298e58->lock(2) - Return: module.c:67()->Builtin()->mutex_key() - master.pike: 555: 8298f90->master_file_stat("/usr/lib/pike/7.5.3/lib/modules/_Image_SVG.so") - master.pike: 295: 8298f90->combine_path_with_cwd("/usr/lib/pike/7.5.3/lib/modules/_Image_SVG.so") - Return: "/usr/lib/pike/7.5.3/lib/modules/_Image_SVG.so" - master.pike: 297: 8298f90->dirname("/usr/lib/pike/7.5.3/lib/modules/_Image_SVG.so") - Return: "/usr/lib/pike/7.5.3/lib/modules" - master.pike: 325: 82988b8->create() - Return: Stat(-rw-r--r-- 8156b) - master.pike: 555: 82988b8->[](1) - Return: 8156 - master.pike: 560: 82988b8->[](3) - Return: 1044556949 the above lines are the very last ones in the output - after that pike is hung. Trying to call Image.SVG() renders the same effect - pike is hung. My next step will be to compile the .deb with full debugging info, but I thought that somebody might have a clue as to what's wrong. TIA, marek

4 13

Weird problem with Image.SVG in Pike 7.4/7.5
by David Hedbor ＠ Pike developers forum 12 Feb '03

12 Feb '03

Naturally, make that s/gcc/gdb/

1 0

Search bug ?
by David Gourdelier 12 Feb '03

12 Feb '03

Hi list, Am I missing something or is it a bug ? : > search(({ 1, 2, 3 }), 5); Compiler Error: 1:Bad argument 2 to search. Compiler Error: 1:Expected: function(string, string | int, void | int : int) | function(array((0=mixed)), 0, void | int : int) | function(mapping((1=mixed):(2=mixed)), 2, void | 1 : 1) | !(function(!array, mixed, void | int : mixed) | function(!mapping, mixed, void | int : mixed)) & function(mapping | array, mixed, void | int : zero) Compiler Error: 1:Got : function(array(int(1..3)), int(5..5) : void | mixed) > search(({ 1, 2, 3 }), 1); (1) Result: 0 > search(({ 1, 2, 3 }), -8); (2) Result: -1 Search returns an error when I search for something which is greater than the greater element of the array. It does this with either Pike 7.2 or 7.4. Thank you for your answers. -- David Gourdelier

4 3

Implicit vs. explicit type casting with Pike
by Marek Habersack 11 Feb '03

11 Feb '03

Hey all, Consider the following code: class settable { private mapping data = ([]); mixed [](string what) { if (data[what]) return data[what]; return ([])[0]; } mixed []=(string what, mixed contents) { data[what] = contents; return contents; } mixed ->(string what) { return [](what); } mixed ->=(string what, mixed contents) { return []=(what, contents); } array _indices() { return indices(data); } array _values() { return values(data); } mixed cast(string rtype) { switch(rtype) { case "mapping": write("cast called - they want a mapping\n"); return data; default: return throw(({"Unsupported type conversion for the object", backtrace()})); } } }; int main() { object s = settable(); s["test"] = 1; s->test1 = 2; s->test2 = ([]); write("settable: %O\n%O\n", indices(s), values(s)); m_delete((mapping)s, "test1"); return 0; } The above program works as expected, but if I remove the explicit cast, m_delete will throw an error: Bad argument 1 to m_delete(). Expected object with _m_delete Unknown program: m_delete(ptest()->settable(),"test1") ptest.pike:59: ptest()->main() Which is incorrect as per the m_delete synopsis: mixed m_delete(object|mapping map, mixed index) Two questions I have are: 1. is there a way to define the default implicit cast for an object, so that the above code would invoke the cast() method when m_delete is called without the explicit mapping cast, or 2. is there any way to install a hook for an object to intercept index deletion in the above code (without the cast)? TIA, marek

12 64

sscanf %O
by Mirar ＠ Pike developers forum 11 Feb '03

11 Feb '03

I started to write a sscanf %O, which I believe we have discussed previously, to match sprintf %O (and generic Pike constants). So far, it understands integers, strings and character constants: > array_sscanf("17 \"hepp\" 'x' gurka","%O%O%O%s"); (3) Result: ({ /* 4 elements */ 17, "hepp", 120, " gurka" })

5 14

gcc/icc
by Mirar ＠ Pike developers forum 10 Feb '03

10 Feb '03

I downloaded icc-7.0 and ran a comparison. test gcc icc Pike start overhead..... 0.001s 0.001s Ackermann............... 0.660s 0.646s Append array............ 0.470s (1063830/s) 0.534s (935551) Append mapping.......... 2.890s (3460/s) 2.785s (3591) Append multiset......... 0.476s (20997/s) 0.432s (23148) Array & String Juggling. 0.687s 0.604s Read binary INT16....... 0.283s (3537736/s) 0.257s (3892944) Read binary INT32....... 0.177s (2820513/s) 0.150s (3342618) Read binary INT128...... 0.895s (11173/s) 0.857s (11673) Clone null-object....... 0.228s (1313869/s) 0.177s (1691176) Clone object............ 0.429s (699153/s) 0.369s (812641) Compile................. 1.080s (22352 lines/s) 0.893s (27022) Compile & Exec.......... 0.904s (665265 lines/s) 0.738s (814537) GC...................... 0.587s 0.578s Insert in mapping....... 0.443s (1128668/s) 0.446s (1121076) Insert in multiset...... 0.880s (568182/s) 0.777s (643777) Matrix multiplication... 0.410s 0.402s Loops Nested (local).... 0.332s (50561473 iters/s) 0.487s (34473732) Loops Nested (global)... 0.528s (31788409 iters/s) 0.723s (23209587) Loops Recursed.......... 1.383s (758464 iters/s) 0.504s (2078675) Note that icc function calls seems to be much faster (Loops Recursed), and that the speed is comparable even though the icc version is compiled without machine code. (Both are with 64-bit float and int.) Has anyone looked into how difficult it would be to get icc to use the machine code stuff?

10 36

Extending Autodocs ?
by Xavier Beaudouin 10 Feb '03

10 Feb '03

Hello, Since we are speaking about autodocs, I've got a clue : Is there any way to extend it ? Eg adding autodoc for other things like tag / containers / entities / modules for example Caudium or Roxen ? Pike autodocs is good for code, but we (The Caudium Group) wishes to use it to autodocs as well the others part of Caudium/Camas project using same tools for documentation. Is there any way to add such "plug-ins" / "extension" to autodocs ? /Xavier

2 1

CPU_TIME_IS_THREAD_LOCAL
by Marcus Comstedt (ACROSS) (Hail Ilpalazzo!) ＠ Pike (-) developers forum 10 Feb '03

10 Feb '03

The use of #elif rather than #else in this code + #if CPU_TIME_IS_THREAD_LOCAL == YES + OBJ2THREAD(Pike_interpreter.thread_id)->auto_gc_time += last_gc_time; + #elif CPU_TIME_IS_THREAD_LOCAL == NO + auto_gc_time += last_gc_time; + #endif seems to suggest that setting CPU_TIME_IS_THREAD_LOCAL to something other than `YES' or `NO' is valid, and results in no gc time accounting at all. Is that correct? In that case I suspect grubba's fix for the no-treads case is incomplete. If not, why are there two tests here?

3 5

autodoc
by Peta, jo det �r jag ＠ Pike developers forum 10 Feb '03

10 Feb '03

Is there anyway to compile the documentation for only a module instead of the entire tree?

3 2

Pike threads benchmark
by Marek Habersack 10 Feb '03

10 Feb '03

Is there anything like that? I want to compare the performance of pike with various thread libraries and before writing any test, I'd like to learn whether anything like that exists. TIA, marek

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

Pike-devel February 2003