Bugzilla – Full Text Bug Listing
| Summary: | [junit] Failures in Debug JUnit tests | ||
|---|---|---|---|
| Product: | [Tools] CDT | Reporter: | Marc Khouzam <marc.khouzam> |
| Component: | cdt-debug-dsf-gdb | Assignee: | Marc Khouzam <marc.khouzam> |
| Status: | RESOLVED FIXED | QA Contact: | Marc Khouzam <marc.khouzam> |
| Severity: | normal | ||
| Priority: | P3 | CC: | alvaro.sanchez-leon, cdtdoug, malaperle, marc.dumais, pawel.1.piech |
| Version: | 8.2 | ||
| Target Milestone: | 8.2 | ||
| Hardware: | PC | ||
| OS: | Linux | ||
| Whiteboard: | |||
|
Description
Marc Khouzam
I've pushed a fix for the GDBPatternMatchingExpressionsTest tests to Gerrit: https://git.eclipse.org/r/#/c/13292/ Those tests all pass on 64bit now. I'll make sure I didn't break them on 32bit and will commit after. It will help get less failures when we run them for the RC2 testing. The change I first pushed was wrong from some reason. I've pushed an update that is actually smaller. Some of the other failures I'm seeing are not due to 64bit, so I'm changing the title of the bug I pushed another patch to fix the Memory test failures: https://git.eclipse.org/r/13331 The first failure was because we now know the endianess of a memory byte but the test was not updated while the second failure was because different GDB versions gave different error messages for memory write failures. I've tested that all Memory JUnit tests pass from GDB 6.6 to GDB 7.6 on 64bit. I've also testing with GDB 7.6 on 32bit. I will commit. Fixed the register tests with: https://git.eclipse.org/r/13349 That was a 64bit issue where we used Integer instead of Long to parse a register value. I tested on both 64bit and 32bit and with older versions of GDB. I will commit. I've pushed another patch to Gerrit that fixes all the MIExpressions tests: https://git.eclipse.org/r/13522 The testsuite all passes for me with GDB 7.5 and 7.6. There is one failure in the Sessionless_suite when run with -ea, but we'll get that fixed soon. I'll commit to master and mark the JUnit tests as passed on the wiki page for our testing effort. . (In reply to comment #6) > There is one failure in the Sessionless_suite when run with -ea, but we'll > get that fixed soon. After discussions, we agreed that a JUnit tests should not trigger an assertion on purpose, so that we can always run with -ea. The following commit removes the test that was causing the problem: http://git.eclipse.org/c/cdt/org.eclipse.cdt.git/commit/?id=20a079bafafa6ed9050b43cd683be3f631a59476 *** cdt git genie on behalf of Marc Khouzam ***
Bug 409186 - Remove test that was causing assertion.
Change-Id: Ia4336dc239c2193259eee3350d67c2a4f0255c06
[*] http://git.eclipse.org/c/cdt/org.eclipse.cdt.git/commit/?id=20a079bafafa6ed9050b43cd683be3f631a59476
|