Ticket #418 (closed bug: worksforme)

Opened 6 years ago

Last modified 6 years ago

memmove segfault in I/O buffering

Reported by: allison Owned by: whiteknight
Priority: normal Milestone: 1.2
Component: none Version: trunk
Severity: medium Keywords:
Cc: Language:
Patch status: Platform: linux

Description

Preparing for the release...

t/pmc/io.t test 23 fails on x86/32 Ubuntu 7.10

t/pmc/io.t .. 12/43
#   Failed test 'I/O buffering'
#   in t/pmc/io.t at line 434.
# Exited with error code: 139
# Received:
# Segmentation fault (core dumped)
#
# Expected:
# Successful
#

Output from gdb:

(gdb) run t/pmc/io_23.pir
Starting program: /home/allison/projects/svn/parrot/parrot t/pmc/io_23.pir
[Thread debugging using libthread_db enabled]
[New Thread -1226504000 (LWP 31357)]
warning: Lowest section in /usr/lib/libicudata.so.34 is .hash at 00000094

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread -1226504000 (LWP 31357)]
0xb6fa9e1c in memmove () from /lib/tls/i686/cmov/libc.so.6
(gdb) bt
#0  0xb6fa9e1c in memmove () from /lib/tls/i686/cmov/libc.so.6
#1  0xb7d68832 in Parrot_io_write_buffer (interp=0x804f040, filehandle=0x80f0fe8, s=0x80cfa3c)
    at src/io/buffer.c:663
#2  0xb7e2fbca in Parrot_FileHandle_nci_puts (interp=0x804f040, pmc=0x80bf6e8) at ./src/pmc/filehandle.pmc:517
#3  0xb7da49db in Parrot_NCI_invoke (interp=0x804f040, pmc=0x80bf6e8, next=0x0) at ./src/pmc/nci.pmc:330
#4  0xb7c12147 in Parrot_PCCINVOKE (interp=0x804f040, pmc=0x80f0fe8, method_name=0x8070cec, 
    signature=0xb7efb68c "S->I") at src/call/pcc.c:2848
#5  0xb7d66efc in Parrot_io_putps (interp=0x804f040, pmc=0x80f0fe8, s=0x80cfa3c) at src/io/api.c:458
#6  0xb7bb5a59 in Parrot_print_p_s (cur_opcode=0x81213f0, interp=0x804f040) at src/ops/io.ops:326
#7  0xb7c53cf9 in runops_slow_core (interp=0x804f040, pc=0x81213f0) at src/runops_cores.c:461
#8  0xb7c18660 in runops_int (interp=0x804f040, offset=0) at src/interpreter.c:980
#9  0xb7c19291 in runops (interp=0x804f040, offs=0) at src/call/ops.c:107

Change History

Changed 6 years ago by allison

Test skipped for 1.0 release in r37526, to be unskipped immediately after.

Changed 6 years ago by whiteknight

  • owner set to whiteknight
  • status changed from new to assigned

I just tried this test here (Ubuntu 9.04, x86-64), and it passes without problem. Can I get some kind of verification from people on other systems that this is or is not still an error?

Changed 6 years ago by allison

  • status changed from assigned to closed
  • resolution set to worksforme

I would test it out on the box that had the original failure, but I recently upgraded it to 9.04. That version of Ubuntu is so old, I can't even create a chroot for it, because the packages have been removed from the archives.

Testing on x86/32 Ubuntu 9.04 passes. I unskipped the test in r38872. We don't need to support software that the developers won't even support.

Note: See TracTickets for help on using tickets.