On 9/13/05, <b class="gmail_sendername">Ross Paterson</b> &lt;<a href="mailto:ross@monk.galois.com">ross@monk.galois.com</a>&gt; wrote:<div><span class="gmail_quote"></span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
ross&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;2005/09/13 08:28:53 PDT<br><br>&nbsp;&nbsp;Modified files:<br>&nbsp;&nbsp;&nbsp;&nbsp;src&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;builtin.c<br>&nbsp;&nbsp;Log:<br>&nbsp;&nbsp;Another tilt at the Windows DEP problem: make the thunk_data RWX<br>&nbsp;&nbsp;(though actually only the code part needs it).
</blockquote><div><br>
&nbsp;This seems to fix the problem. Now the ffi/Callback, ffi/Sin, and
ffi/Sparc tests pass with DEP enabled. The only tests that fail now are
static/mod154.hs. rts/read.hs, rts/arith.hs (known problems),
ffi/FileIO.hs (the test is broken: it does I/O on its own source file
and assumes that end-of-line is always only one byte), and libs/dirTest
(which I think is just making bad assumptions about how file
permissions work).<br>
<br>
- Brian<br>
</div></div>