netdev - Re: [PATCH ethtool-next] sff-8636: report LOL / LOS / Tx Fault

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20230612084837.6c5b5ca9@kernel.org>
Date: Mon, 12 Jun 2023 08:48:37 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: Ido Schimmel <idosch@...dia.com>
Cc: mkubecek@...e.cz, danieller@...dia.com, netdev@...r.kernel.org,
 vladyslavt@...dia.com, linux@...linux.org.uk, andrew@...n.ch
Subject: Re: [PATCH ethtool-next] sff-8636: report LOL / LOS / Tx Fault

On Sun, 11 Jun 2023 10:39:27 +0300 Ido Schimmel wrote:
> > I was asked if "Linux reports" this information.
> > I'm not very confident on the merits, I'm going by the spec.
> > If it makes sense I'll send a similar patch for CMIS.  
> 
> I have two AOC connected to each other. When both are up:
> 
> # ethtool -m swp13 | grep "Rx loss of signal" -A 4
>         Rx loss of signal                         : None
>         Tx loss of signal                         : None
>         Rx loss of lock                           : None
>         Tx loss of lock                           : None
>         Tx adaptive eq fault                      : None
> 
> When I bring the other side down:
> 
> # ip link set dev swp14 down
> # ethtool -m swp13 | grep "Rx loss of signal" -A 4
>         Rx loss of signal                         : [ Yes, Yes, Yes, Yes ]
>         Tx loss of signal                         : None
>         Rx loss of lock                           : [ Yes, Yes, Yes, Yes ]
>         Tx loss of lock                           : None
>         Tx adaptive eq fault                      : None
> 
> When I bring the interface itself down:
> 
> # ip link set dev swp13 down
> # ethtool -m swp13 | grep "Rx loss of signal" -A 4
>         Rx loss of signal                         : [ Yes, Yes, Yes, Yes ]
>         Tx loss of signal                         : [ Yes, Yes, Yes, Yes ]
>         Rx loss of lock                           : [ Yes, Yes, Yes, Yes ]
>         Tx loss of lock                           : [ Yes, Yes, Yes, Yes ]
>         Tx adaptive eq fault                      : [ Yes, Yes, Yes, Yes ]
> 
> And I don't see these fields on PC:
> 
> # ethtool -m swp3 | grep "Rx loss of signal" -A 4

IOW works as expected, am I interpreting this correctly?

> > diff --git a/qsfp.h b/qsfp.h
> > index aabf09fdc623..b4a0ffe06da1 100644
> > --- a/qsfp.h
> > +++ b/qsfp.h
> > @@ -55,6 +55,8 @@
> >  #define	 SFF8636_TX2_FAULT_AW	(1 << 1)
> >  #define	 SFF8636_TX1_FAULT_AW	(1 << 0)
> >  
> > +#define	SFF8636_LOL_AW_OFFSET	0x05
> > +
> >  /* Module Monitor Interrupt Flags - 6-8 */
> >  #define	SFF8636_TEMP_AW_OFFSET	0x06
> >  #define	 SFF8636_TEMP_HALARM_STATUS		(1 << 7)
> > @@ -525,9 +527,15 @@
> >  /*  56h-5Fh reserved */
> >  
> >  #define	 SFF8636_OPTION_2_OFFSET	0xC1
> > +/* Tx input equalizers auto-adaptive */
> > +#define	  SFF8636_O2_TX_EQ_AUTO		(1 << 3)
> >  /* Rx output amplitude */
> >  #define	  SFF8636_O2_RX_OUTPUT_AMP	(1 << 0)
> >  #define	 SFF8636_OPTION_3_OFFSET	0xC2
> > +/* Rx CDR Loss of Lock */
> > +#define	  SFF8636_O3_RX_LOL		(1 << 5)
> > +/* Tx CDR Loss of Lock */
> > +#define	  SFF8636_O3_TX_LOL		(1 << 4)  
> 
> I'm looking at revision 2.10a and bit 4 is Rx while bit 5 is Tx.

Ugh, you're right, thanks.