Received: from redirect.mail.gandi.net (relay14.mail.gandi.net [217.70.178.234]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id ACD6D611C for ; Tue, 4 Jul 2023 13:03:02 +0000 (UTC) Received: from spool.mail.gandi.net (spool4.mail.gandi.net [217.70.178.213]) by relay14.mail.gandi.net (Postfix) with ESMTPS id 8CEB91A0004 for ; Tue, 4 Jul 2023 13:03:00 +0000 (UTC) Received: from smtp2-g21.free.fr (smtp2-g21.free.fr [212.27.42.2]) by spool.mail.gandi.net (Postfix) with ESMTPS id 62E747803E7 for ; Tue, 4 Jul 2023 13:02:59 +0000 (UTC) Received: from mail.thom.fr.eu.org (unknown [IPv6:2a01:e0a:2aa:1dd0::2]) by smtp2-g21.free.fr (Postfix) with ESMTPS id 37246200597; Tue, 4 Jul 2023 15:02:58 +0200 (CEST) Received: from localhost (localhost [127.0.0.1]) by mail.thom.fr.eu.org (8.15.2/8.15.2/Debian-22) with ESMTP id 364D2uRB2790287; Tue, 4 Jul 2023 15:02:56 +0200 From: =?utf-8?q?Fran=C3=A7ois_Legal?= In-Reply-To: Content-Type: text/plain; charset="utf-8" X-Forward: 195.68.6.102, 165.225.20.106 Date: Tue, 04 Jul 2023 15:02:56 +0200 Cc: xenomai@xenomai.org To: "Jan Kiszka" Precedence: bulk X-Mailing-List: xenomai@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Message-ID: <3b7514-64a41880-35-28cbfc80@10274948> Subject: =?utf-8?q?Re=3A?= [PATCH v2] Make macb/gem RT driver to be built against linux =?utf-8?q?>=3D?= 5.0 kernels User-Agent: SOGoMail 5.0.1 Content-Transfer-Encoding: quoted-printable X-Virus-Scanned: clamav-milter 0.103.8 at tls-srv-03 X-Virus-Status: Clean Authentication-Results: spool.mail.gandi.net; dkim=none; dmarc=none; spf=pass (spool.mail.gandi.net: domain of francois.legal@thom.fr.eu.org designates 212.27.42.2 as permitted sender) smtp.mailfrom=francois.legal@thom.fr.eu.org Le Mardi, Juillet 04, 2023 14:46 CEST, Jan Kiszka a =C3=A9crit: > On 04.07.23 14:24, Fran=C3=A7ois Legal wrote: > > Le Mardi, Juillet 04, 2023 11:25 CEST, Jan Kiszka a =C3=A9crit: > > > >> On 29.06.23 22:04, Fran=C3=A7ois Legal wrote: > >>> PHY layer in upstream has changed a little after 5.0 in the way a= utonegociation > >>> capabilities are encoded. > >>> MACB platform data tend to disappear after 5.2 upstream. > >>> Plus, ioctl prototype has changed in xenomai. RTMACB driver does = not build > >>> anymore. > >>> > >>> Signed-off-by: Fran=C3=A7ois LEGAL > >>> --- > >>> kernel/drivers/net/drivers/macb.c | 60 ++++++++++++++++++----= ------ > >>> kernel/drivers/net/drivers/rt=5Fmacb.h | 2 +- > >>> 2 files changed, 40 insertions(+), 22 deletions(-) > >>> > >>> diff --git a/kernel/drivers/net/drivers/macb.c b/kernel/drivers/n= et/drivers/macb.c > >>> index 77957d906..0648afa3b 100644 > >>> --- a/kernel/drivers/net/drivers/macb.c > >>> +++ b/kernel/drivers/net/drivers/macb.c > >>> @@ -132,6 +132,7 @@ void rtmacb=5Fget=5Fhwaddr(struct macb *bp) > >>> bottom =3D macb=5For=5Fgem=5Freadl(bp, SA1B + i * 8); > >>> top =3D macb=5For=5Fgem=5Freadl(bp, SA1T + i * 8); > >>> > >>> +#if LINUX=5FVERSION=5FCODE < KERNEL=5FVERSION(5, 2, 0) > >> > >> Thanks for the patch, but master does not support kernels older th= an > >> 5.10 (dovetail) anymore. > >> > >> Is this patch needed for stable/v3.2.x or can it wait for v3.3? In= any > >> case, we need a master patch first, then we can discuss whether to= > >> backport it as well. > > > > Well, I started patching 3.2 then submitted it against master. I ca= n submit for 3.2 as well. I believe this is my target version. > > > >> > >> What we also need is enabling of the driver for CI to catch such i= ssues > >> earlier. See .gitlab-ci.yml and all those driver configs there. > > > > I don't really know about this. > > Do you want me to submit a patch for this file ? > > Would it be something like > > - ./scripts/config -m XENO=5FDRIVERS=5FNET=5FDRV=5FE1000=5FNEW= > > - ./scripts/config -m XENO=5FDRIVERS=5FNET=5FDRV=5FRT2500 > > + - ./scripts/config -m XENO=5FDRIVERS=5FNET=5FDRV=5FMACB > > - ./scripts/config -m XENO=5FDRIVERS=5FNET=5FFEC > > - ./scripts/config -m XENO=5FDRIVERS=5FNET=5FADDON=5FRTCAP > > > > Yes, if that is sufficient to actually build it (check if it has > dependencies that may not be fulfilled in a defconfig). Thing is, XENO=5FDRIVERS=5FNET=5FDRV=5FMACB depends on ARM. I'm not sur= e how this can be achieved in your CI build system Fran=C3=A7ois > > Jan > > -- > Siemens AG, Technology > Competence Center Embedded Linux >