Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Thu, 30 Mar 2023 08:30:17 +0200
Message-ID: <87wn2yn43q.wl-tiwai@suse.de>
From:   Takashi Iwai <tiwai@suse.de>
To:     Jakub Kicinski <kuba@kernel.org>
Cc:     Takashi Iwai <tiwai@suse.de>,
        Sasha Neftin <sasha.neftin@intel.com>,
        Bagas Sanjaya <bagasdotme@gmail.com>,
        Paul Menzel <pmenzel@molgen.mpg.de>,
        Jesse Brandeburg <jesse.brandeburg@intel.com>,
        Tony Nguyen <anthony.l.nguyen@intel.com>,
        netdev@vger.kernel.org, intel-wired-lan@lists.osuosl.org,
        regressions@lists.linux.dev, linux-kernel@vger.kernel.org
Subject: Re: [Intel-wired-lan] [REGRESSION] e1000e probe/link detection fails since 6.2 kernel
In-Reply-To: <20230329121232.7873ad95@kernel.org>
References: <87jzz13v7i.wl-tiwai@suse.de>
        <652a9a96-f499-f31f-2a55-3c80b6ac9c75@molgen.mpg.de>
        <ZCP5jOTNypwG4xK6@debian.me>
        <87a5zwosd7.wl-tiwai@suse.de>
        <20230329121232.7873ad95@kernel.org>
User-Agent: Wanderlust/2.15.9 (Almost Unreal) Emacs/27.2 Mule/6.0
MIME-Version: 1.0 (generated by SEMI-EPG 1.14.7 - "Harue")
Content-Type: text/plain; charset=ISO-2022-JP
Precedence: bulk

On Wed, 29 Mar 2023 21:12:32 +0200,
Jakub Kicinski wrote:
> 
> On Wed, 29 Mar 2023 10:48:36 +0200 Takashi Iwai wrote:
> > On Wed, 29 Mar 2023 10:40:44 +0200,
> > Bagas Sanjaya wrote:
> > > 
> > > On Tue, Mar 28, 2023 at 04:39:01PM +0200, Paul Menzel wrote:  
> > > > Does openSUSE Tumbleweed make it easy to bisect the regression at least on
> > > > “rc level”? It be great if narrow it more down, so we know it for example
> > > > regressed in 6.2-rc7.
> > > >   
> > > 
> > > Alternatively, can you do bisection using kernel sources from Linus's
> > > tree (git required)?  
> > 
> > That'll be a last resort, if no one has idea at all :)
> 
> I had a quick look yesterday, there's only ~6 or so commits to e1000e.
> Should be a fairly quick bisection, hopefully?

*IFF* it's an e1000e-specific bug, right?

Through a quick glance, the only significant change in e1000e is the
commit 1060707e3809
    ptp: introduce helpers to adjust by scaled parts per million

Others are only for MTP/ADP and new devices, which must be irrelevant.
The tracing must be irrelevant, and the kmap change must be OK.

Can 1060707e3809 be the cause of such a bug?


thanks,

Takashi