The Windows version of test 10054 does something strange with the string to be compiled.
Process.popen(RUNPIKE+" -e write(\"test\");") results in 1:'test' undefined. Compilation failed. z:/home/peter/hack/xenofarm/client/pike/vmwarexp.bortas.org/buildtmp/Pike7.3-20020924-090136/build/microsoft-windows-xp-5.1.2600-x86/test-install/pike/7.3.56/lib/master.pike:254: master()->compile_string("mixed create(int argc, array(string) argv,array(string) env) {write(test);;}",0,0) ^^^^^^
I have already committed a fix to that test. What happened was that the command
pike -e write("test");
was executed, and the M$ crt0.o interpred this as pike being invoked with the two arguments
-e write(test);
just like a UNIX shell would have.
/ Marcus Comstedt (ACROSS) (Hail Ilpalazzo!)
Previous text:
2002-09-24 13:31: Subject: Test 10054 (shift 1) (CRNL) failed.
The Windows version of test 10054 does something strange with the string to be compiled.
Process.popen(RUNPIKE+" -e write(\"test\");") results in 1:'test' undefined. Compilation failed. z:/home/peter/hack/xenofarm/client/pike/vmwarexp.bortas.org/buildtmp/Pike7.3-20020924-090136/build/microsoft-windows-xp-5.1.2600-x86/test-install/pike/7.3.56/lib/master.pike:254: master()->compile_string("mixed create(int argc, array(string) argv,array(string) env) {write(test);;}",0,0) ^^^^^^
/ Peter Bortas
M$s ctr0.o doesn't dby any chance interpret "'"s like in UNIX so we can unify that test? I'll test it when I get home, but since the test has a separate NT case to begin with I suppose there is a reason.
/ Peter Bortas
Previous text:
2002-09-24 13:45: Subject: Test 10054 (shift 1) (CRNL) failed.
I have already committed a fix to that test. What happened was that the command
pike -e write("test");
was executed, and the M$ crt0.o interpred this as pike being invoked with the two arguments
-e write(test);
just like a UNIX shell would have.
/ Marcus Comstedt (ACROSS) (Hail Ilpalazzo!)
I changed it to "write("test");", which should be interpreted correctly if I recall correctly. ' has no special interpretation on NT.
/ Marcus Comstedt (ACROSS) (Hail Ilpalazzo!)
Previous text:
2002-09-24 13:48: Subject: Test 10054 (shift 1) (CRNL) failed.
M$s ctr0.o doesn't dby any chance interpret "'"s like in UNIX so we can unify that test? I'll test it when I get home, but since the test has a separate NT case to begin with I suppose there is a reason.
/ Peter Bortas
No problem. Can I do that via make somehow? make xenofarm -DDLDEBUG?
/ Peter Bortas
Previous text:
2002-09-24 15:08: Subject: Test 10054 (shift 1) (CRNL) failed.
It seems like my fix for dynamic Gmp failed though. Would you mind compiling it with the DLDEBUG define in dlopen.c active?
/ Marcus Comstedt (ACROSS) (Hail Ilpalazzo!)
Very well, I'll do a manual build. Barring any file access problems you should be able to do one youself. Just copy ~peter/.init_ntrc and run /mnt/media/win32-pike/nt-tools/init_nt
/ Peter Bortas
Previous text:
2002-09-24 15:52: Subject: Test 10054 (shift 1) (CRNL) failed.
No.
/ Marcus Comstedt (ACROSS) (Hail Ilpalazzo!)
Well, if you're making one I'm not going to trouble the machine with a fourth parallel build. :-) (Xenofarm should start two builds to r54 about now...)
/ Marcus Comstedt (ACROSS) (Hail Ilpalazzo!)
Previous text:
2002-09-24 16:01: Subject: Test 10054 (shift 1) (CRNL) failed.
Very well, I'll do a manual build. Barring any file access problems you should be able to do one youself. Just copy ~peter/.init_ntrc and run /mnt/media/win32-pike/nt-tools/init_nt
/ Peter Bortas
Only 3. The two Xenofarm builds are sequencial. ;-)
/ Peter Bortas
Previous text:
2002-09-24 16:02: Subject: Test 10054 (shift 1) (CRNL) failed.
Well, if you're making one I'm not going to trouble the machine with a fourth parallel build. :-) (Xenofarm should start two builds to r54 about now...)
/ Marcus Comstedt (ACROSS) (Hail Ilpalazzo!)
There should be a build with testname "DLDEBUG" uploaded to the server now. But for some reason it doen't show. Either check the xenofarm incoming dir on pelix or copy aphrael:/tmp/xenofarm_result.tar.gz
/ Peter Bortas
Previous text:
2002-09-24 16:01: Subject: Test 10054 (shift 1) (CRNL) failed.
Very well, I'll do a manual build. Barring any file access problems you should be able to do one youself. Just copy ~peter/.init_ntrc and run /mnt/media/win32-pike/nt-tools/init_nt
/ Peter Bortas
pike-devel@lists.lysator.liu.se
-
Marcus Comstedt (ACROSS) (Hail Ilpalazzo!) ＠ Pike (-) developers forum -
Peter Bortas ＠ Pike developers forum