From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from metis.ext.pengutronix.de ([2001:67c:670:201:290:27ff:fe1d:cc33]) by bombadil.infradead.org with esmtps (Exim 4.90_1 #2 (Red Hat Linux)) id 1gyBpS-0007cP-FY for barebox@lists.infradead.org; Mon, 25 Feb 2019 08:41:40 +0000 Date: Mon, 25 Feb 2019 09:41:36 +0100 From: Sascha Hauer Message-ID: <20190225084136.bbofwkzwnupxawbq@pengutronix.de> References: <20190221102531.25291-1-u.kleine-koenig@pengutronix.de> <20190222081139.wf775ytgd6zerdsv@pengutronix.de> <20190222212234.fzambuxfvhml7gjb@pengutronix.de> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <20190222212234.fzambuxfvhml7gjb@pengutronix.de> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-15" Content-Transfer-Encoding: quoted-printable Sender: "barebox" Errors-To: barebox-bounces+u.kleine-koenig=pengutronix.de@lists.infradead.org Subject: Re: [PATCH 1/2] imd: create new helper macro to add data from oftree to image meta data To: Uwe =?iso-8859-15?Q?Kleine-K=F6nig?= Cc: barebox@lists.infradead.org, "Thorsten K . Scherer" On Fri, Feb 22, 2019 at 10:22:34PM +0100, Uwe Kleine-K=F6nig wrote: > On Fri, Feb 22, 2019 at 09:11:39AM +0100, Sascha Hauer wrote: > > Hi Uwe, > > = > > On Thu, Feb 21, 2019 at 11:25:30AM +0100, Uwe Kleine-K=F6nig wrote: > > > Signed-off-by: Uwe Kleine-K=F6nig > > > --- > > > include/image-metadata.h | 4 ++++ > > > scripts/gen-dtb-s | 4 ++-- > > > 2 files changed, 6 insertions(+), 2 deletions(-) > > > = > > > diff --git a/include/image-metadata.h b/include/image-metadata.h > > > index 8739bffb92e1..5904d95acd37 100644 > > > --- a/include/image-metadata.h > > > +++ b/include/image-metadata.h > > > @@ -118,6 +118,10 @@ static inline void imd_used(const void *unused) > > > = > > > #define IMD_USED(_name) \ > > > imd_used(&__barebox_imd_##_name) > > > +#define IMD_USED_OF(_name) ({ \ > > > + extern char __barebox_imd_OF_ ## _name[]; \ > > > + imd_used(&__barebox_imd_OF_ ## _name); \ > > > + }) > > > = > > > #endif /* __BAREBOX__ */ > > > = > > > diff --git a/scripts/gen-dtb-s b/scripts/gen-dtb-s > > > index 2b44ffb361eb..c3144df851bf 100755 > > > --- a/scripts/gen-dtb-s > > > +++ b/scripts/gen-dtb-s > > > @@ -18,8 +18,8 @@ FDTGET=3Dscripts/dtc/fdtget > > > = > > > if [ "$imd" =3D "y" ]; then > > > echo ".section .barebox_imd_0.${name},\"a\"" > > > - echo ".global __imd_${name}_start" > > > - echo "__imd_${name}_start:" > > > + echo ".global __barebox_imd_OF_${name}" > > > + echo "__barebox_imd_OF_${name}:" > > = > > This breaks somewhere when the compressed variants of the dtbs are used. > > Building imx_v7_defconfig fails here with: > > = > > arch/arm/dts/built-in-pbl.o: In function `__dtb_z_imx6dl_wandboard_end': > > arch/arm/dts/pbl-vf610-zii-ssmb-spu3.dtb.o:(.dtbz.rodata.imx6dl_wandboa= rd+0x2340): undefined reference to `__imd_imx6dl_wandboard_start' > > OBJCOPYB images/start_imx6dl_nitrogen6x_1g.pblb > > arch/arm/dts/built-in-pbl.o: In function `__dtb_z_imx6q_wandboard_end': > > arch/arm/dts/pbl-vf610-zii-ssmb-spu3.dtb.o:(.dtbz.rodata.imx6q_wandboar= d+0x2320): undefined reference to `__imd_imx6q_wandboard_start' > = > The following makes it build again (on top of my patch): > = > diff --git a/scripts/gen-dtb-s b/scripts/gen-dtb-s > index c3144df851bf..90294c4840f4 100755 > --- a/scripts/gen-dtb-s > +++ b/scripts/gen-dtb-s > @@ -71,5 +71,5 @@ echo ".global __dtb_z_${name}_end" > echo ".balign STRUCT_ALIGNMENT" > = > if [ "$imd" =3D "y" ]; then > - echo ".word __imd_${name}_start" > + echo ".word __barebox_imd_OF_${name}" > fi > = > But I admit I don't understand why this is referenced here. Also I > wonder why imx_defconfig + CONFIG_MACH_PHYTEC_SOM_IMX6=3Dy doesn't fail. We are using -fdata-sections -ffunction-sections to throw away unused data and functions. In gen-dtb-s the lines you are changing in your fixups do not have a separate section, so it ends up in the section that has previously been set which here is: .section .dtbz.rodata.${name},\"a\" So when the wandboard pulls in the symbol __dtb_z_imx6q_wandboard_start then we must also have __imd_imx6q_wandboard_start. I wonder which purpose these lines have anyway. Nothing in the code seems to reference __imd_*_start, so I just removed the lines for now. Sascha -- = Pengutronix e.K. | | Industrial Linux Solutions | http://www.pengutronix.de/ | Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 | _______________________________________________ barebox mailing list barebox@lists.infradead.org http://lists.infradead.org/mailman/listinfo/barebox