[asterisk-bugs] [Zaptel 0005375]: [patch] Changes to PCI driver init
noreply at bugs.digium.com
noreply at bugs.digium.com
Sat Jun 7 11:05:21 CDT 2008
A NOTE has been added to this issue.
======================================================================
http://bugs.digium.com/view.php?id=5375
======================================================================
Reported By: reformed
Assigned To: kpfleming
======================================================================
Project: Zaptel
Issue ID: 5375
Category: wct4xxp
Reproducibility: always
Severity: minor
Priority: normal
Status: closed
Zaptel Version: CVS HEAD
SVN Branch (only for SVN checkouts, not tarball releases): N/A
SVN Revision (number only!):
Disclaimer on File?: N/A
Request Review:
Resolution: fixed
Fixed in Version:
======================================================================
Date Submitted: 10-03-2005 22:32 CDT
Last Modified: 06-07-2008 11:05 CDT
======================================================================
Summary: [patch] Changes to PCI driver init
Description:
After purchasing TE406P card we were experiencing problems with interrupt
generation by the card on the system with following spec's:
Motherboard: ASUS P5WD2 Premium (Intel 955X Chipset)
CPU: Pentium D 830 (Dual core)
HDD: SATA Raid
Kernel: 2.6.13 (of ftp.kernel.org)
/proc/interrupts:
CPU0 CPU1
0: 3201 161935181 IO-APIC-edge timer
1: 250 0 IO-APIC-edge i8042
7: 0 0 IO-APIC-edge parport0
9: 0 0 IO-APIC-level acpi
14: 14 0 IO-APIC-edge ide0
50: 75 0 IO-APIC-level HDA Intel, uhci_hcd:usb4
58: 0 0 IO-APIC-level uhci_hcd:usb1, ehci_hcd:usb5
66: 0 0 IO-APIC-level uhci_hcd:usb2
74: 0 0 IO-APIC-level uhci_hcd:usb3
82: 139256 11439569 PCI-MSI eth2
209: 7 0 IO-APIC-level wct4xxp
225: 4190 2295584 PCI-MSI libata
233: 17031628 0 IO-APIC-level ohci1394, SysKonnect SK-98xx
NMI: 0 0
LOC: 161942886 161942906
ERR: 0
MIS: 0
Whenever the module is loaded no interrupts are being generated.
A few interrupts were generated but probably by the init code.
It took a little while to trace where the problem was, but it appears that
pci_module_init did not initialise the card properly.
After a little googling around we found that 2.6.X kernel is moving from
pci_module_init to pci_register_driver.
2.6.XX is required by most modern motherboards or peripherals, perhaps
there is misconception that Digium cards have interrupt problems because of
this issue?
After applying the attached patched we have never experienced the
interrupt problem again.
To be sure we have switched between the patched and un-patched wct4xxp
modules multiple times (with reboots and without) with perfect results.
The system has been running now for 3 weeks in production routing over 40K
of calls in that between PRI and PABX (Asterisk is between PRI and PABX
using TE406P) without missing a beat.
======================================================================
----------------------------------------------------------------------
svnbot - 06-07-08 11:05
----------------------------------------------------------------------
Repository: dahdi
Revision: 791
U trunk/pciradio.c
U trunk/tor2.c
U trunk/wcfxo.c
U trunk/wct1xxp.c
U trunk/wct4xxp.c
U trunk/wctdm.c
U trunk/wcte11xp.c
U trunk/zaptel.h
------------------------------------------------------------------------
r791 | kpfleming | 2008-06-07 11:05:18 -0500 (Sat, 07 Jun 2008) | 2 lines
use pci_register_driver() on recent kernels (issue
http://bugs.digium.com/view.php?id=5375, compatibility
fix only, no functional difference)
------------------------------------------------------------------------
http://svn.digium.com/view/dahdi?view=rev&revision=791
Issue History
Date Modified Username Field Change
======================================================================
06-07-08 11:05 svnbot Note Added: 0088073
======================================================================
More information about the asterisk-bugs
mailing list