Pike-devel

pike-devel@lists.lysator.liu.se

3044 discussions

New 7.6 release ???
by Xavier Beaudouin 29 Jun '05

29 Jun '05

Hello, Is there any projects to release a new 7.6 ? Several bugs seems to have being fixed. Is it possible that release master can release a new release for pike 7.6 ? Many thanks, /Xavier

2 2

Nettle + OpenBSD 3.5 / sparc64
by Xavier Beaudouin 29 Jun '05

29 Jun '05

Hello, Can Nettle being fixed ont OpenBSD 3.5 or disable all assembler usage on this specific Operating System ? gcc -I. -DHAVE_CONFIG_H -g -O2 -ggdb3 -Wall -W -Wmissing-prototypes -Wmissing-declarations -Wstrict-prototypes -Wpointer-arith -Wbad-function-cast -Wnested-externs -fpic -MT aes.o -MT aes.o.d -MD -MP -MF aes.o.d -fpic -c aes.s aes.s: Assembler messages: aes.s:273: Error: relocation overflow aes.s:300: Error: relocation overflow aes.s:327: Error: relocation overflow aes.s:397: Error: relocation overflow aes.s:430: Error: relocation overflow aes.s:463: Error: relocation overflow gmake[1]: *** [aes.o] Error 1 gmake[1]: Leaving directory `/home/kiwi/src/nettle-1.12' gmake: *** [all] Error 2 kiwi@red-pill:~/src/nettle-1.12$ gcc -v Reading specs from /usr/lib/gcc-lib/sparc64-unknown-openbsd3.5/3.3.2/specs Configured with: Thread model: single gcc version 3.3.2 (propolice) kiwi@red-pill:~/src/nettle-1.12$ uname -a OpenBSD red-pill.oav.net 3.5 GENERIC#123 sparc64 Many thanks, /Xavier PS: note that using directly ./configure --diable-assembler doesn't seem to really desactivate assembler usage on the machine red-pill (see pikefarm)

1 0

int->pow(float)
by Tobias Josefowitz 22 Jun '05

22 Jun '05

int->pow(float) gives same result as int->pow(floor(float)). > 4->pow(2.9); (2) Result: 16 this is not restricted to Hilfe.

2 6

Pike 7.6 release
by Bill Welliver 21 Jun '05

21 Jun '05

Hi, What are the chances of getting a new Pike 7.6 release out in the next month or so? It's been quite a while (almost/over a year?) and there are some good bug fixes that would be nice to get out. Actually, I have a patch to backend in 7.7 that could probably stand to be backported (I'm not 100% sure the problem exists in 7.6 though)... I'm not completely familiar with the procedures involved with a new release, but would be willing to help out. Any thoughts? Bill

1 0

Calendar woes
by Peter Bortas ＠ Pike developers forum 15 Jun '05

15 Jun '05

Oh well, change the day too.

1 0

Relocatable modules
by Jonas Walld�n ＠ Pike developers forum 06 Jun '05

06 Jun '05

What is the status of PIKE_MODULE_RELOC? In the 7.4 master it's disabled with the comment "currently broken" but not in 7.6 or 7.7. Still, it's being used in 7.4 on Windows, right? I tried enabling it (--with-relocatable-dumped-modules) in a test build of 7.4 on Darwin but didn't succeed which wasn't surprising given the comment above. However, not even in a shiny new 7.7 build could I get it working: ceylon:pike-reloc $ ./moved77/pike/7.7.21/bin/pike "Couldn't load master program from /tmp/pike-reloc/install77/pike/7.7.21/lib/master.pike.\n" Oh well, at least providing the paths explicitly should work: ceylon:pike-reloc $ ./moved77/pike/7.7.21/bin/pike -m /tmp/pike-reloc/moved77/pike/7.7.21/lib/master.pike -M /tmp/pike-reloc/moved77/pike/7.7.21/lib/modules/ -P /tmp/pike-reloc/moved77/pike/7.7.21/lib/modules/ -I /tmp/pike-reloc/moved77/pike/7.7.21/include/ Could not resolv Getopt. (Perhaps the installed pike tree has been moved.) /tmp/pike-reloc/moved77/pike/7.7.21/lib/master.pike:2388: master()->__lambda_65601_2_line_2385("Getopt",UNDEFINED) What am I missing here? Also, a related question is what the whole point of the Master Cookie is? Overcome limitations that one cannot find the path of the currently executing process? At least on Darwin that can be solved with some system calls such as NSModuleForSymbol(), and maybe other systems have similar calls nowadays?

4 10

Relocatable modules
by Marcus Comstedt (ACROSS) (Hail Ilpalazzo!) ＠ Pike (-) developers forum 05 Jun '05

05 Jun '05

Of course, the string ${PIKE_MODULE_PATH} should also be changed to something else in this case.

1 0

Kernel version in build directory names
by Martin Stjernholm, Roxen IS ＠ Pike developers forum 22 May '05

22 May '05

I'd like to remove the -r part from the uname call that creates the build directory names. That means that one would get build directories like this build/linux-x86_64 instead of build/linux-2.6.10-5-amd64-k8-x86_64 The reason is that I find it a bit annoying to have to discard all my pike build dirs every time I install a new kernel. It also doesn't seem relevant to depend on the kernel version since it afaik doesn't affect the build environment. It'd be more apt to depend on the libc version (or actually the version of every lib that pike builds against). Thoughts?

6 9

webmaster?
by Martin Bähr 17 May '05

17 May '05

hi, who is watching the webmaster(a)pike.ida.liu.se address? i got a complaint from someone claiming he had sent mail there but had not received a reaction yet. does it make sense to extend the alias to all people who have write access to the website? greetings, martin.

1 0

Bump version ?
by Xavier Beaudouin 09 May '05

09 May '05

Hello, Seems according to pikefarm that 7.6 has lot of green (excepted some case, but it is usual). Can pike maintainer bump version and release a new pike if possible (This should help to have a new release for Caudium Group and help us to get a stable pike version). Many thanks, /Xavier

1 0

Jump to page:

2026

2025

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