From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from mout.gmx.net ([212.227.15.19]) by bombadil.infradead.org with esmtps (Exim 4.92.3 #3 (Red Hat Linux)) id 1iX01P-00073y-3S for barebox@lists.infradead.org; Tue, 19 Nov 2019 09:42:08 +0000 References: <20190819133847.17015-1-s.hauer@pengutronix.de> <20190819133847.17015-7-s.hauer@pengutronix.de> <27e33ef9-bb89-dcd0-9829-6dd3036eeae9@pengutronix.de> From: Oleksij Rempel Message-ID: <98fca2ef-9c7b-0570-ef4d-11ff92b33fa4@rempel-privat.de> Date: Tue, 19 Nov 2019 10:41:55 +0100 MIME-Version: 1.0 In-Reply-To: <27e33ef9-bb89-dcd0-9829-6dd3036eeae9@pengutronix.de> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============2329080311293809803==" Sender: "barebox" Errors-To: barebox-bounces+u.kleine-koenig=pengutronix.de@lists.infradead.org Subject: Re: [PATCH 6/8] ARM: Add generic device tree 2nd stage support To: Ahmad Fatoum , barebox@lists.infradead.org, Sascha Hauer This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --===============2329080311293809803== Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="cCzus8UMWtAQaeFUuqHkE3OHnunUpQ8qm" This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --cCzus8UMWtAQaeFUuqHkE3OHnunUpQ8qm Content-Type: multipart/mixed; boundary="an74WCwCbjXvyMTIOeOisF2xCbH0h5hYl" --an74WCwCbjXvyMTIOeOisF2xCbH0h5hYl Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: quoted-printable Am 19.11.19 um 09:26 schrieb Ahmad Fatoum: > Hello Sascha, >=20 > On 8/19/19 3:38 PM, Sascha Hauer wrote: >> +void dt_2nd_aarch64(void *fdt) >> +{ >> + unsigned long image_start =3D (unsigned long)_text + global_variable= _offset(); >> + >> + arm_setup_stack(image_start); >=20 > Shouldn't we rather place the stack somewhere beyond the end of the bar= ebox image > instead of assuming that we can access the memory before the start?=20 >=20 > I am asking because I am testing use of the generic ARM board as defaul= t > second stage image for the multi-image AT91s. The board specific entry = point > generates only the PBL, which runs in SRAM to do DRAM setup and chainlo= ads > the generic dt barebox from MMC to start of SDRAM and then calls it alo= ng > with the dtb. Hm.. Is it possible that you can get an exception vector in the star of S= DRAM? > It works so far and I like that I now can have the same second stage ba= rebox > for all multi-image targets. (And one could even "falcon"-boot that way= ). > What do you think?> >> +ENTRY_FUNCTION(start_dt_2nd, r0, r1, r2) >> +{ >> + unsigned long image_start =3D (unsigned long)_text + global_variable= _offset(); >> + >> + arm_setup_stack(image_start); >=20 >=20 > Ditto. >=20 > Cheers > Ahmad >=20 --=20 Regards, Oleksij --an74WCwCbjXvyMTIOeOisF2xCbH0h5hYl-- --cCzus8UMWtAQaeFUuqHkE3OHnunUpQ8qm Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- iQEzBAEBCAAdFiEEpENFL0P3hvQ7p0DDdQOiSHVI77QFAl3TuOMACgkQdQOiSHVI 77S2HQf+MS3EldlPyfYPxanPO0yU6y5SllUM6izSKBeJXhXPIF20jKv6E4mwTsYV qiTsPU9OSwrWz4cepKoJV+p1OCPFE79JiLZQbWCegf5gBfBQuGW/HgvFZVxlUUzz nusR3i/GFjAX97EYF6gD1kp1BUFD6g+2xGp4SseVkFZzWGH0qJRfQQDBHobQMyHe NXOjSC5kMQP7jg9HhG6Qi0d3Lq27eYLgF/zFuCRvu3EgjqADHADbeCeWzgOW4+3x URYBdoRDnMDI3EQHG9XSIRfKNXvpq0N/IscpU2U5ze3fwIJwE6StAk5Ex7u/4WNc gCx8qaTC0Cs4f+FkX5iTIm9xk/b1ew== =iAb3 -----END PGP SIGNATURE----- --cCzus8UMWtAQaeFUuqHkE3OHnunUpQ8qm-- --===============2329080311293809803== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ barebox mailing list barebox@lists.infradead.org http://lists.infradead.org/mailman/listinfo/barebox --===============2329080311293809803==--