Bug 759285

Summary: Issue with intel 82580 copper ethernet devices (pci id 150E)
Product: [Fedora] Fedora Reporter: Patrick Bergamin <patrick.bergamin>
Component: kernelAssignee: Kernel Maintainer List <kernel-maint>
Status: CLOSED NOTABUG QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 16CC: gansalmon, itamar, jonathan, kernel-maint, madhu.chinakonda
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2011-12-05 15:56:27 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
dmesg
none
lspci output none

Description Patrick Bergamin 2011-12-01 21:45:55 UTC
Created attachment 539401 [details]
dmesg

Description of problem:

I have a network module containing 8 intel 82580 ethernet devices.  4 Fiber (150F) and 4 Copper (150E).  On a Fedora 16 (x86_64) install after a short while (under 30s) running traffic over one of the copper ports the traffic simply stops. I attempted to use the igb 3.2.10 driver from the intel driver download site but those won't even initialize the copper ports.

I believe the hardware is fine because Ubuntu 11.10 appears to have no issues with these same copper ports.  I tried both the in kernel igb driver and the intel igb 3.2.20 driver on Ubuntu.

Version-Release number of selected component (if applicable):

Fedora 16 kernel version 3.1.2-1.fc16.x86_64

How reproducible:

Always.

Steps to Reproduce:
1. Connect network cable to intel 82580 copper port 
2. Allow NetworkManager to obtain an ip address via dhcp
3. attempt to pass traffic the configured interface
  
Additional info:

I don't know if it is related or not but the intel igb 3.2.10 driver fails to initialize the copper ports.  I enabled debugging in this driver and it fails with these errors (from igb 3.2.20 driver):

[  657.020130] igb 0000:05:00.3: eth0: PBA No: FFFFFF-0FF
[  657.020132] igb 0000:05:00.3: LRO is disabled 
[  657.020135] igb 0000:05:00.3: Using MSI-X interrupts. 1 rx queue(s), 1 tx que
ue(s)
[  657.020155] igb 0000:06:00.0: enabling device (0000 -> 0002)
[  657.020166] igb 0000:06:00.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19
[  657.020185] igb 0000:06:00.0: setting latency timer to 64
[  657.020443] e1000_set_mac_type
[  657.020445] e1000_init_mac_ops_generic
[  657.020447] e1000_init_nvm_ops_generic
[  657.020448] e1000_init_function_pointers_82575
[  657.020450] e1000_init_mac_params_82575
[  657.020453] e1000_sgmii_uses_mdio_82575
[  657.020458] e1000_init_nvm_params_82575
[  657.020460] e1000_init_phy_params_82575
[  657.020463] e1000_reset_mdicnfg_82580
[  657.020464] e1000_read_nvm_spi
[  657.020466] e1000_acquire_nvm_82575
[  657.020467] e1000_acquire_swfw_sync_82575
[  657.020469] e1000_get_hw_semaphore_generic
[  658.693181] Driver can't access device - SMBI bit is set.
[  658.693185] NVM Read Error

Comment 1 Patrick Bergamin 2011-12-01 21:46:26 UTC
Created attachment 539402 [details]
lspci output

Comment 2 Patrick Bergamin 2011-12-02 00:59:18 UTC
Update(In reply to comment #0)

> 
> I don't know if it is related or not but the intel igb 3.2.10 driver fails to
> initialize the copper ports.  I enabled debugging in this driver and it fails
> with these errors (from igb 3.2.20 driver):
> 

The problem here was actually caused by manually removing the fedora 16 igb driver with 'rmmod igb'.  If the Fedora igb driver is disabled (via modprobe.d/blacklist.conf) and the intel igb 3.2.10 driver is loaded manually with insmod the copper ports are successfully initialized.  But this driver still exhibits the same behaviour as the fedora 16 igb driver.  Which is after a short while of processing traffic it just stops.

Additionally the ifconfig output shows lots of errors, carrier errors, etc.

Comment 3 Patrick Bergamin 2011-12-05 15:56:27 UTC
I tried new hardware and there aren't any problems with the new hardware.  This seems to be a hardware issue.  I'm not sure why things were working on Ubuntu.
Sorry for the noise.