View Issue Details [ Jump to Notes ] | [ Issue History ] [ Print ] | ||||||||||||
ID | Project | Category | View Status | Date Submitted | Last Update | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
0000182 | mercury | Bug | public | 2011-02-02 14:00 | 2011-02-02 14:00 | ||||||||
Reporter | pbone | ||||||||||||
Assigned To | |||||||||||||
Priority | normal | Severity | minor | Reproducibility | have not tried | ||||||||
Status | new | Resolution | open | ||||||||||
Product Version | |||||||||||||
Target Version | Fixed in Version | ||||||||||||
Summary | 0000182: Segfaults could be reported better. | ||||||||||||
Description | In low-level C grades it's possible to detect if a segfault is caused by a stack-overrun, in which case the runtime system can report that it is either definitely or definitely-not due to unbounded/deep recursion. There appears to be code to do this but I haven't seen it work. If we can fix this then we should fix the default message in low-level C grades. The default message says something like. "...this is probably due to unbounded tail recursion...", we shouldn't suggest this if we know the segfault is not due to a stack overrun since this only confuses users, making them blame Mercury when it's more likely to be a problem with foreign code they're linking with. | ||||||||||||
Tags | No tags attached. | ||||||||||||
Attached Files |
|
Issue History | |||
Date Modified | Username | Field | Change |
---|---|---|---|
2011-02-02 14:00 | pbone | New Issue |