Bugzilla – Full Text Bug Listing
| Summary: | [LTTng] SIGSEGV in the C library when parsing a v2.6 trace on a 64bit Linux | ||||||
|---|---|---|---|---|---|---|---|
| Product: | z_Archived | Reporter: | Francois Chouinard <fchouinard> | ||||
| Component: | LinuxTools | Assignee: | Francois Chouinard <fchouinard> | ||||
| Status: | CLOSED WONTFIX | QA Contact: | Francois Chouinard <fchouinard> | ||||
| Severity: | major | ||||||
| Priority: | P3 | CC: | patrick.tasse | ||||
| Version: | unspecified | ||||||
| Target Milestone: | --- | ||||||
| Hardware: | PC | ||||||
| OS: | Linux | ||||||
| Whiteboard: | |||||||
| Bug Depends on: | |||||||
| Bug Blocks: | 321366 | ||||||
| Attachments: |
|
||||||
While this is unfortunate, it'll have to wait for a post-Helios maintenance release. This bug was for Legacy LTTng which is being removed in Linux Tools 2.0. |
Created attachment 171514 [details] Post-mortem dump Trying to parse a v2.6 trace on a 64bit Linux machine results in a segmentation fault and Eclipse dies. - The same trace works fine on a 32bits Linux. - The other trace formats (v2.3 and v2.5) work OK with 64bit Linux - The library itself works well 64bits LTTV. At first glance, there seems to be a glitch in the JNI component (JniParser) when processing the new trace format on a 64bit architecture.