Craig A. Berry wrote:
At 8:45 AM +0100 8/4/06, Martin J. Evans wrote:
On 03-Aug-2006 Craig A. Berry wrote:
perl "-MExtUtils::Command" -e "rm_f" foo.[0-9][0-9]
%SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual
address=00000000006B4000, PC=000000007AE26F60, PS=0000001B
The good news is that this does not appear to happen in the current
development stream ("bleadperl"), so something got fixed. The bad
news is I don't know what got fixed when. If I have any luck
narrowing down a fix, I'll let you know.
Craig,
Thats for that - where can I get bleadperl?
Things have got worse for me here. My problems seem to resolve to:
1. I get a lot of file has time in the future warnings from mms for any
module I build with HP's binary and for building perl from source
itself. I've tried vmstar /date=none (makes no difference), double
checked my time, tried changing to utc (no difference).
I have:
$ show logi *time*
(LNM$PROCESS_TABLE)
(LNM$JOB_899B8380)
(LNM$GROUP_000001)
(LNM$SYSTEM_TABLE)
"SYS$LOCALTIME" = "SYS$SYSROOT:[SYS$ZONEINFO.SYSTEM]GB."
"SYS$TIMEZONE_DAYLIGHT_SAVING" = "1"
"SYS$TIMEZONE_DIFFERENTIAL" = "3600"
"SYS$TIMEZONE_NAME" = "BST"
"SYS$TIMEZONE_RULE" = "UTC0"
"TCPIP$BIND_TIMEOUT" = "...."
(LNM$SYSCLUSTER_TABLE)
(DECW$LOGICAL_NAMES)
When I do show time it reports the time (currently) 1 hour beyond local
time i.e. it is 8:19pm here and show time reports 9:19pm.
2. The problem that started this - namely not specifying /ods2 on the
vmstar line. This is sorted but I think a note in the vms readme on this
would be well worth while.
3. The access violation on a mms clean which you refer to above.
Might be fixed by bleadperl but I have not tried that yet.
4. Baring in mind your comments on C compilers I did not know what HP
build their binary with so I tried to build perl from source. I get a
number of mms time in the future messages (especially for .EXISTS files)
and it bombs out with:
MCR [---]miniperl.exe "-I[---.lib]" "-I[---.lib]"
[---.lib.ExtUtils]xsubpp -nol
inenumbers -typemap [---.lib.ExtUtils]typemap BYTE.xs >BYTE.C
CC/DECC /Include=[]/Standard=Relaxed_ANSI/Prefix=All/Obj=.obj
/NOANSI_ALIAS/floa
t=ieee/ieee=denorm/Define=("VERSION=""2.00""","XS_VERSION=""2.00""")/Include=([-
--],[-.Encode])/NoList BYTE.c
CC/DECC /Include=[]/Standard=Relaxed_ANSI/Prefix=All/Obj=.obj
/NOANSI_ALIAS/floa
t=ieee/ieee=denorm/Define=("VERSION=""2.00""","XS_VERSION=""2.00""")/Include=([-
--],[-.Encode])/NoList BYTE_T.c
If F$Search("[---.LIB.AUTO.ENCODE.BYTE]BYTE.OLB").eqs."" Then
Library/Object/Cre
ate [---.LIB.AUTO.ENCODE.BYTE]BYTE.OLB
Library/Object/Replace [---.LIB.AUTO.ENCODE.BYTE]BYTE.OLB BYTE.OBJ,
BYTE_T.OBJ
If F$TrnLNm("PerlShr").eqs."" Then Define/NoLog/User PerlShr
Sys$Share:PerlShr.e
xe
Link /NoTrace/NoMap
/Shareable=[---.LIB.AUTO.ENCODE.BYTE]PL_ENCODE__BYTE.EXE Byt
e.opt/Option,[---]perlshr_attr.opt/Option
Running Mkbootstrap for Encode::Byte ()
MCR DKA0:[NICK.MARTIN.PERL-5_8_8]miniperl.exe "-I[---.lib]"
"-I[---.lib]" "-MExt
Utils::Command" -e chmod 644 BYTE.BS
MCR DKA0:[NICK.MARTIN.PERL-5_8_8]miniperl.exe "-I[---.lib]"
"-I[---.lib]" "-MExt
Utils::Command" -e cp Byte.bs [---.LIB.AUTO.ENCODE.BYTE]BYTE.BS
MCR DKA0:[NICK.MARTIN.PERL-5_8_8]miniperl.exe "-I[---.lib]"
"-I[---.lib]" "-MExt
Utils::Command" -e chmod 644 [---.LIB.AUTO.ENCODE.BYTE]BYTE.BS
%MMS-F-ABORT, For target SUBDIRS, CLI returned abort status: %X10EE8180.
%MMS-F-ABORT, For target DYNEXT, CLI returned abort status: %X10EE8034.
So it still looks to me like a time issue but I cannot work it out right
now.
If you have any ideas I'd greatly appreciate it as I'm struggling with
this now.
Martin
