Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20210624082911.5d013e8c@canb.auug.org.au>
In-Reply-To: <20210624082911.5d013e8c@canb.auug.org.au>
From:   Marcin Wojtas <mw@semihalf.com>
Date:   Thu, 24 Jun 2021 00:46:48 +0200
Message-ID: <CAPv3WKfiL+sR+iK_BjGKDhtNgjoxKEPv49bU1X9_7+v+ytdR1w@mail.gmail.com>
Subject: Re: linux-next: build failure after merge of the net-next tree
To:     Stephen Rothwell <sfr@canb.auug.org.au>
Cc:     David Miller <davem@davemloft.net>,
        Networking <netdev@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Linux Next Mailing List <linux-next@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Hi Stephen,

czw., 24 cze 2021 o 00:29 Stephen Rothwell <sfr@canb.auug.org.au> napisa=C5=
=82(a):
>
> Hi all,
>
> Today's linux-next build (x86_64 modules_install) failed like this:
>
> depmod: ../tools/depmod.c:1792: depmod_report_cycles_from_root: Assertion=
 `is < stack_size' failed.
>
> Caused by commit
>
> 62a6ef6a996f ("net: mdiobus: Introduce fwnode_mdbiobus_register()")
>
> (I bisected to there and tested the commit before.)
>
> The actual build is an x86_64 allmodconfig, followed by a
> modules_install.  This happens in my cross build environment as well as
> a native build.
>
> $ gcc --version
> gcc (Debian 10.2.1-6) 10.2.1 20210110
> $ ld --version
> GNU ld (GNU Binutils for Debian) 2.35.2
> $ /sbin/depmod --version
> kmod version 28
> -ZSTD +XZ -ZLIB +LIBCRYPTO -EXPERIMENTAL
>
> I have no idea why that commit should caused this failure.

Thank you for letting us know. Not sure if related, but I just found
out that this code won't compile for the !CONFIG_FWNODE_MDIO. Below
one-liner fixes it:

--- a/include/linux/fwnode_mdio.h
+++ b/include/linux/fwnode_mdio.h
@@ -40,7 +40,7 @@ static inline int fwnode_mdiobus_register(struct mii_bus =
*bus,
         * This way, we don't have to keep compat bits around in drivers.
         */

-       return mdiobus_register(mdio);
+       return mdiobus_register(bus);
 }
 #endif

I'm curious if this is the case. Tomorrow I'll resubmit with above, so
I'd appreciate recheck.

Thanks,
Marcin