netdev - RE: [Intel-wired-lan] [PATCH iwl-net v3 2/6] igc: Lengthen the hardware retry time to prevent timeouts

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: <MW4PR11MB698491CB1DA8179D870F787FC2532@MW4PR11MB6984.namprd11.prod.outlook.com>
Date: Wed, 6 Nov 2024 23:53:11 +0000
From: "Hall, Christopher S" <christopher.s.hall@...el.com>
To: Paul Menzel <pmenzel@...gen.mpg.de>
CC: "intel-wired-lan@...ts.osuosl.org" <intel-wired-lan@...ts.osuosl.org>,
	"Zage, David" <david.zage@...el.com>, "Gomes, Vinicius"
	<vinicius.gomes@...el.com>, "netdev@...r.kernel.org"
	<netdev@...r.kernel.org>, "Cadore Cataldo, Rodrigo"
	<rodrigo.cadore@...coustics.com>, "Vinschen, Corinna" <vinschen@...hat.com>,
	Michal Swiatkowski <michal.swiatkowski@...ux.intel.com>, "Bar Gabay, MorX"
	<morx.bar.gabay@...el.com>, "Dahan, AvigailX" <avigailx.dahan@...el.com>
Subject: RE: [Intel-wired-lan] [PATCH iwl-net v3 2/6] igc: Lengthen the
 hardware retry time to prevent timeouts

Hi Paul,

> From: Paul Menzel <pmenzel@...gen.mpg.de>
> Sent: Wednesday, November 06, 2024 3:14 PM


> Subject: Re: [Intel-wired-lan] [PATCH iwl-net v3 2/6] igc: Lengthen the
> hardware retry time to prevent timeouts
> 
> Dear Christopher,
> 
> 
> Thank you for the patch.
> 
> I’d use the more specific summary/title below:

Will do.

> 
> igc: Lengthen hardware retry time to 4 μs to prevent timeouts
> 
> Am 06.11.24 um 19:47 schrieb Christopher S M Hall:
> > Lengthen the hardware retry timer to four microseconds.
> >
> > The i225/i226 hardware retries if it receives an inappropriate response
> > from the upstream device. If the device retries too quickly, the root
> > port does not respond.
> 
> Any idea why? Is it documented somewhere?

I do not. Theoretically, 1 us should work, but it does not. It could be a root
port problem or an issue with i225/i226 NIC. I am not able to directly observe
the state of either. 4 us has worked in all my testing I am comfortable with
that value. 2 us also works, but given the limited hardware at my disposal
I doubled the value to 4 us to be safe. PTM is not time critical. Typically,
software initiates a transaction between 8 and 32 times per second. There
is no performance impact for PTM or any other function of the card. The
timeout occurs rarely, but if the retry time is too short the PTM state
machine does not recover.

> > The issue can be reproduced with the following:
> >
> > $ sudo phc2sys -R 1000 -O 0 -i tsn0 -m
> >
> > Note: 1000 Hz (-R 1000) is unrealistically large, but provides a way to
> > quickly reproduce the issue.
> >
> > PHC2SYS exits with:
> >
> > "ioctl PTP_OFFSET_PRECISE: Connection timed out" when the PTM
> transaction
> >    fails
> 
> Why four microseconds, and not some other value?

See above.

> > Fixes: 6b8aa753a9f9 ("igc: Decrease PTM short interval from 10 us to 1 us")
> >
> > -#define IGC_PTM_SHORT_CYC_DEFAULT	1   /* Default short cycle
> interval */
> > +#define IGC_PTM_SHORT_CYC_DEFAULT	4   /* Default short cycle
> interval */

> Kind regards,
> 
> Paul

Thanks,
Chris