Comment 2 for bug 681092
Revision history for this message
| Alastair Bridgewater (alastair-bridgewater) wrote | #2 |
© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
73416ba
(Get the code!)
Thank you!
What's happening is that the function is being inlined completely rather than producing a local call or explicit closure, thus not introducing a new stack frame (and thus a complete closure environment) but a different physenv is still involved, leading the IR2 conversion of the REF node to believe that an ANCESTOR-FRAME-REF is required, and for some reason it picks up the /calling/ frame to read from.