mbox series

[0/1,SRU,B/master] Fix a regression in lan78xx driver

Message ID 1539271111-24862-1-git-send-email-paolo.pisati@canonical.com
Headers show
Series Fix a regression in lan78xx driver | expand

Message

Paolo Pisati Oct. 11, 2018, 3:18 p.m. UTC
BugLink: https://bugs.launchpad.net/bugs/1797406

Impact:

Since the release of Ubuntu-raspi2_4.15.0-1022.24, the ethernet port of
the RaspberryPi 3B+ has stopped working: no phy interrupts are generated
when a cable is inserted, so it's like the ethernet port is always
disconnected.

The problem lies in commit 72eff2505735 ("lan78xx: Connect phy early"),
that was backported into Bionic/master (and then percolated down into
Bionic/raspi2):

"
Commit 92571a1aae40 ("lan78xx: Connect phy early") moves the PHY
initialisation into lan78xx_probe, but lan78xx_open subsequently calls
lan78xx_reset. As well as forcing a second round of link negotiation,
this reset frequently prevents the phy interrupt from being generated
(even though the link is up), rendering the interface unusable.

Fix this issue by removing the lan78xx_reset call from lan78xx_open.
"

Fix:

Apply the attached patch and recompile

How to test:

Boot a patched kernel and checks if the ethernet port is working

Regression potential:

None, the attached patch is a fix for the aforementioned patch, and
fixes exactly this issue - it's upstream already and was backported into
gregkh's stable branches 4.17+.

The bug shows up on a Raspberry board, but since the troublesome commit
originated in Bionic/master, the fix applys there (and then propagates
through a rebase).

Phil Elwell (1):
  lan78xx: Don't reset the interface on open

 drivers/net/usb/lan78xx.c | 4 ----
 1 file changed, 4 deletions(-)

Comments

Khalid Elmously Oct. 22, 2018, 7:27 a.m. UTC | #1
On 2018-10-11 17:18:30 , Paolo Pisati wrote:
> BugLink: https://bugs.launchpad.net/bugs/1797406
> 
> Impact:
> 
> Since the release of Ubuntu-raspi2_4.15.0-1022.24, the ethernet port of
> the RaspberryPi 3B+ has stopped working: no phy interrupts are generated
> when a cable is inserted, so it's like the ethernet port is always
> disconnected.
> 
> The problem lies in commit 72eff2505735 ("lan78xx: Connect phy early"),
> that was backported into Bionic/master (and then percolated down into
> Bionic/raspi2):
> 
> "
> Commit 92571a1aae40 ("lan78xx: Connect phy early") moves the PHY
> initialisation into lan78xx_probe, but lan78xx_open subsequently calls
> lan78xx_reset. As well as forcing a second round of link negotiation,
> this reset frequently prevents the phy interrupt from being generated
> (even though the link is up), rendering the interface unusable.
> 
> Fix this issue by removing the lan78xx_reset call from lan78xx_open.
> "
> 
> Fix:
> 
> Apply the attached patch and recompile
> 
> How to test:
> 
> Boot a patched kernel and checks if the ethernet port is working
> 
> Regression potential:
> 
> None, the attached patch is a fix for the aforementioned patch, and
> fixes exactly this issue - it's upstream already and was backported into
> gregkh's stable branches 4.17+.
> 
> The bug shows up on a Raspberry board, but since the troublesome commit
> originated in Bionic/master, the fix applys there (and then propagates
> through a rebase).
> 
> Phil Elwell (1):
>   lan78xx: Don't reset the interface on open
> 
>  drivers/net/usb/lan78xx.c | 4 ----
>  1 file changed, 4 deletions(-)
> 
> -- 
> 2.17.1
> 
> 
> -- 
> kernel-team mailing list
> kernel-team@lists.ubuntu.com
> https://lists.ubuntu.com/mailman/listinfo/kernel-team