Bug 436861 - x86_64: I/O-port access SIGSEGV
Summary: x86_64: I/O-port access SIGSEGV
Alias: None
Product: Fedora
Classification: Fedora
Component: kernel
Version: 8
Hardware: x86_64
OS: Linux
Target Milestone: ---
Assignee: Kernel Maintainer List
QA Contact: Fedora Extras Quality Assurance
Depends On:
TreeView+ depends on / blocked
Reported: 2008-03-10 20:00 UTC by Jan Kratochvil
Modified: 2008-03-13 07:34 UTC (History)
0 users

Clone Of:
Last Closed: 2008-03-13 07:34:06 UTC

Attachments (Terms of Use)
Testcase (463 bytes, text/plain)
2008-03-10 20:00 UTC, Jan Kratochvil
no flags Details
The real testcase. (275 bytes, text/plain)
2008-03-10 20:04 UTC, Jan Kratochvil
no flags Details

Description Jan Kratochvil 2008-03-10 20:00:08 UTC
Description of problem:
irda-utils findchip segfaults because `outb' segfaults despite previous
successful ioperm(2).

Version-Release number of selected component (if applicable):
kernel- (F8, broken)
but it work on: kernel-2.6.25-0.101.rc4.git3.fc9.x86_64 (Rawhide)

How reproducible:

Steps to Reproduce:
1. gcc -o outport outport.c -Wall -ggdb2

Actual results:
Segmentation fault

Expected results:

Additional info:
strace ./outport
ioperm(0x3f0, 0x1, 0x1)                 = 0
--- SIGSEGV (Segmentation fault) @ 0 (0) ---
+++ killed by SIGSEGV +++

# gdb ./outport
(gdb) r
Starting program: /tmp/outport 

Program received signal SIGSEGV, Segmentation fault.
0x000000000040059b in main () at outport.c:14
14        __asm__ __volatile__ ("outb %b0,%w1": :"a" (value), "Nd" (port));
(gdb) x/i $rip
0x40059b <main+83>:     out    %al,(%dx)
(gdb) p/x $rdx
$1 = 0x3f0

Comment 1 Jan Kratochvil 2008-03-10 20:00:08 UTC
Created attachment 297504 [details]

Comment 2 Jan Kratochvil 2008-03-10 20:04:53 UTC
Created attachment 297505 [details]
The real testcase.

The previous testcase was an unrelated code - a filename typo.

Comment 3 Chuck Ebbert 2008-03-12 22:53:34 UTC
I get PASS on, with an AMD processor.

Comment 4 Jan Kratochvil 2008-03-13 07:34:06 UTC
(In reply to comment #3)
> I get PASS on, with an AMD processor.

So it is probably fixed. (this is Intel T7200)

Note You need to log in before you can comment on or make changes to this bug.