Reminder: If you are responding to Leonard, he's not on the ksh-integration-discuss alias, so you'll need to cc: him directly.
Roland Mainz wrote: > Leonard Li wrote: > >>Roland Mainz wrote On 05/29/06 18:55,: >> >>>Leonard Li wrote: >>> >>>>We are working on removing the internal memory map locking limitation. >>>>Before this is done, the wrapper/implementation method is still required. >>> >>>Do you have any idea why "dbx -check access" fails with "RTC: program >>>wrote over private RTC data -- disabling Error Checking file: memerror.c >>>line: 297" ? >>>I am pretty much sure that the ksh93 code did not write over unallocated >>>areas... >> >>This error is rare. > > > ksh93 seems to trigger this error for each invocation (which means we > have a reproduceable testcase... ;-/ ) ... ;-( > > >>It means an "write to unallocated" error occurs and >>it tries to write to rtc internal memory. One possible explanation is >>RTC's memory map is out of syn. Does this error happen if the standard >>heap management routines (in libc.so) are used? > > > AFAIK the standard heap management cannot be simply "dropped in" since > there is AFAIK some extra functionality which cannot easily be emulated > (for example read libast/common/vmalloc/vmbest.c&co.)... ;-( > > ---- > > Bye, > Roland >
