From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tomi Valkeinen Date: Tue, 15 Nov 2016 11:46:10 +0000 Subject: Re: [PATCH v2] Fix loading of module radeonfb on PowerMac Message-Id: MIME-Version: 1 Content-Type: multipart/mixed; boundary="jm4563IB5xILfhwdtXWjFTvcWh01dKk9l" List-Id: References: <1479153557-20849-1-git-send-email-malat@debian.org> In-Reply-To: <1479153557-20849-1-git-send-email-malat@debian.org> To: Mathieu Malaterre , linux-fbdev@vger.kernel.org Cc: Lennart Sorensen , Jean-Christophe Plagniol-Villard , Benjamin Herrenschmidt , linuxppc-dev@lists.ozlabs.org --jm4563IB5xILfhwdtXWjFTvcWh01dKk9l Content-Type: multipart/mixed; boundary="iMCSaRpwaLpECA0SNONQadmAxDl9X9uuf"; protected-headers="v1" From: Tomi Valkeinen To: Mathieu Malaterre , linux-fbdev@vger.kernel.org Cc: Lennart Sorensen , Jean-Christophe Plagniol-Villard , Benjamin Herrenschmidt , linuxppc-dev@lists.ozlabs.org Message-ID: Subject: Re: [PATCH v2] Fix loading of module radeonfb on PowerMac References: <1479153557-20849-1-git-send-email-malat@debian.org> In-Reply-To: <1479153557-20849-1-git-send-email-malat@debian.org> --iMCSaRpwaLpECA0SNONQadmAxDl9X9uuf Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Hi, On 14/11/16 21:59, Mathieu Malaterre wrote: > When the linux kernel is build with (typical kernel ship with Debian > installer): >=20 > CONFIG_FB_OF=3Dy > CONFIG_VT_HW_CONSOLE_BINDING=3Dy > CONFIG_FB_RADEON=3Dm >=20 > The offb driver takes precedence over module radeonfb. It is then > impossible to load the module, error reported is: >=20 > [ 96.551486] radeonfb 0000:00:10.0: enabling device (0006 -> 0007) > [ 96.551526] radeonfb 0000:00:10.0: BAR 0: can't reserve [mem 0x98000= 000-0x9fffffff pref] > [ 96.551531] radeonfb (0000:00:10.0): cannot request region 0. > [ 96.551545] radeonfb: probe of 0000:00:10.0 failed with error -16 >=20 > This patch reproduce the behavior of the module radeon, so as to make i= t > possible to load radeonfb when offb is first loaded. >=20 > It should be noticed that `offb_destroy` is never called which explain = the > need to skip error detection on the radeon side. >=20 > Signed-off-by: Mathieu Malaterre > Link: https://bugs.debian.org/826629#57 > Link: https://bugzilla.kernel.org/show_bug.cgi?id=3D119741 > Suggested-by: Lennart Sorensen > --- Please always have a changelog in patch new revisions. For individual patches, you can insert the changes here. > drivers/video/fbdev/aty/radeon_base.c | 26 ++++++++++++++++++++++++++ > 1 file changed, 26 insertions(+) >=20 > diff --git a/drivers/video/fbdev/aty/radeon_base.c b/drivers/video/fbde= v/aty/radeon_base.c > index 218339a..84d634b 100644 > --- a/drivers/video/fbdev/aty/radeon_base.c > +++ b/drivers/video/fbdev/aty/radeon_base.c > @@ -2259,6 +2259,22 @@ static struct bin_attribute edid2_attr =3D { > .read =3D radeon_show_edid2, > }; > =20 > +static int radeon_kick_out_firmware_fb(struct pci_dev *pdev) > +{ > + struct apertures_struct *ap; > + > + ap =3D alloc_apertures(1); > + if (!ap) > + return -ENOMEM; > + > + ap->ranges[0].base =3D pci_resource_start(pdev, 0); > + ap->ranges[0].size =3D pci_resource_len(pdev, 0); > + > + remove_conflicting_framebuffers(ap, KBUILD_MODNAME, false); > + kfree(ap); > + > + return 0; > +} > =20 > static int radeonfb_pci_register(struct pci_dev *pdev, > const struct pci_device_id *ent) > @@ -2314,19 +2330,27 @@ static int radeonfb_pci_register(struct pci_dev= *pdev, > rinfo->fb_base_phys =3D pci_resource_start (pdev, 0); > rinfo->mmio_base_phys =3D pci_resource_start (pdev, 2); > =20 > + ret =3D radeon_kick_out_firmware_fb(pdev); > + if (ret) > + return ret; > + > /* request the mem regions */ > ret =3D pci_request_region(pdev, 0, "radeonfb framebuffer"); > if (ret < 0) { > printk( KERN_ERR "radeonfb (%s): cannot request region 0.\n", > pci_name(rinfo->pdev)); > +#ifndef CONFIG_PPC > goto err_release_fb; > +#endif If this is not a problem on PPC, the kernel shouldn't print an error eith= er. > } > =20 > ret =3D pci_request_region(pdev, 2, "radeonfb mmio"); > if (ret < 0) { > printk( KERN_ERR "radeonfb (%s): cannot request region 2.\n", > pci_name(rinfo->pdev)); > +#ifndef CONFIG_PPC > goto err_release_pci0; > +#endif Same here. > } > =20 > /* map the regions */ > @@ -2511,10 +2535,12 @@ static int radeonfb_pci_register(struct pci_dev= *pdev, > iounmap(rinfo->mmio_base); > err_release_pci2: > pci_release_region(pdev, 2); > +#ifndef CONFIG_PPC > err_release_pci0: > pci_release_region(pdev, 0); > err_release_fb: > framebuffer_release(info); > +#endif > err_disable: > err_out: > return ret; >=20 So I don't quite follow what's going on here. Why the CONFIG_PPC conditionals? Is this problem only with OFFB and only with PPC? And I think the code itself should have comments on this rather strange behavior: the driver fails to get HW resources, but decides to ignore the failure on PPC. Tomi --iMCSaRpwaLpECA0SNONQadmAxDl9X9uuf-- --jm4563IB5xILfhwdtXWjFTvcWh01dKk9l Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBCAAGBQJYKvWCAAoJEPo9qoy8lh71XiUP/jzJk6QXKvLGydV/QDUYF+aC mWpZBTJB6ABKJGZG081hWCzEYgy4CIA7Zt7vat94xnIh0GD05JRohXzgWpBwFTXv 250nYqO5ryatbbQmlLbvopSsy6T5W8YQX5P0SBx3GQDssaJ75ji4E/urYposuUig 2KGflFLV1rHw84GbfS2hdAhy/2yLgtoAK8j0WTm345th0chnnumC09SXJawqhzMH X0DrT0AmsOhmTxZsHvcA7oU2UAC7JHNcODXMk1Fy+DBnySBkTrjgZAuAWZCFJWf3 Ycivrqy6Ecv1vK/SC5RXE7Z62qgQWuDMulrAlNH0eOp2OS9gtvyRhg3yS8acq29a nuSlfxW9LnlSGDKXz8raAmdU1S1j1NN2K/6iuGygWClL8e15koPOojC8Dj9/T1UF FA9XQLGIMlcxsCzu7yv+7ukYf/t/Bsr9obS/TpPjhVNDj6ZARMVuiXZVRUM+9JPY rfgM+SiKDXboeLunlLMtOPSveI0Qd7DWPN+tIJGdbbfqEMjWUhuhNX1L61bttlxp 3qIApsKwK4fSWkfStgw76wwqDsCA3nt3tTNAR0dyvak6Zkd3kqYA18lozfRlYCAM cVzqfDyM8yqNofbmYra8BWZg6e069Q5HYEfprTDKYGeq/1i7TGIyJVw1Opibhd2J aOqO7Il3DXESG9bT/Gfe =RK1v -----END PGP SIGNATURE----- --jm4563IB5xILfhwdtXWjFTvcWh01dKk9l--