Ticket #695 (closed patch: fixed)

Opened 6 years ago

Last modified 5 years ago

subtest 3 in t/dynoplibs/myops.t segfaults on darwin

Reported by: urkle Owned by:
Priority: normal Milestone: 1.4
Component: core Version: 1.2.0
Severity: medium Keywords:
Cc: Language:
Patch status: new Platform:

Description

subtest 3 in the t/dynoplibs/myops.t segfault on darwin (Mac OS X Leopard 10.5.7).

Here is the stack trace.

0   libSystem.B.dylib             	0x92dbfe42 __kill + 10
1   libSystem.B.dylib             	0x92e3223a raise + 26
2   libSystem.B.dylib             	0x92e3e679 abort + 73
3   myops_ops.bundle              	0x0073235b Parrot_hcf + 11 (myops.ops:56)
4   libparrot.dylib               	0x004745f4 runops_slow_core + 260 (cores.c:462)
5   libparrot.dylib               	0x004737f6 runops_int + 422 (main.c:986)
6   libparrot.dylib               	0x004211d0 runops + 240 (ops.c:111)
7   libparrot.dylib               	0x00421489 runops_args + 649 (ops.c:257)
8   libparrot.dylib               	0x004222da Parrot_runops_fromc_args + 186 (ops.c:325)
9   libparrot.dylib               	0x003fef3e Parrot_runcode + 750 (embed.c:1010)
10  libparrot.dylib               	0x005e6743 imcc_run_pbc + 323 (main.c:807)
11  libparrot.dylib               	0x005e73c9 imcc_run + 873 (main.c:1099)
12  parrot                        	0x00001a79 main + 185 (main.c:61)
13  parrot                        	0x00001986 start + 54

Attachments

eliminate_hcf.txt Download (1.2 KB) - added by doughera 6 years ago.

Change History

Changed 6 years ago by doughera

  • patch set to new

Yes, it segfaults everywhere, and produces a core dump under many conditions. It is, as far as I can tell, a completely pointless opcode and test. (The opcode is 'hcf', which is supposed to stand for "Halt and Catch Fire".)

Generating core dumps from this opcode has produced numerous tickets over the years as users helpfully try to diagnose why parrot is dumping core. Every time this happens, someone wastes time debugging the core dump and reporting it, and someone else wastes time trying to explain why the test is there. See RT #55040 for a recent go-round on this issue.

The attached patch actually removes the opcode and the corresponding test.

Changed 6 years ago by doughera

Changed 6 years ago by doughera

  • component changed from none to core
  • type changed from bug to patch

Changed 6 years ago by urkle

HAHA.. ok... I was wondering that myself. as the opcode did seem kind of useless (and the commit logs surrounding it). Essentially it could be considered a successful test, a parrot did halt and catch fire:-D..

Changed 5 years ago by coke

  • status changed from new to closed
  • resolution set to fixed

The hcf opcode (and associated test) is now removed, as per TT #450.

Changed 5 years ago by coke

  • milestone set to 1.4
Note: See TracTickets for help on using tickets.