>Details and reply can be found in the LiteSpeed Forums thread:
>
>http://litespeedtech.com/support/forum/showthread.php?t=1061
Well, I would not call that details; "changes to the storage of
errno"; well no.
"Wrong output in truss", I don't think so. Strange arguments to
write, more like it.
Where does the write(2,) call originate?
Why does it attempt to write 18MB part of which is not in allocated
memory?
This sound ssuspiciously like "the applicaiton stopped working in build X;
it worked before" - conjecture: "the 32 bit ABI has changed".
The conjecture is wrong as we don't really do that; but what could have
happened is one of:
- we inadvertedly changed the semantics
- the application did something wrong but got away with it until
there was a slight disturbance in the source which caused things
to be different.
Based on past experience, my money is firmly on #2.
Casper
_______________________________________________
opensolaris-code mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/opensolaris-code
