| Message ID | 1381134884-5816-17-git-send-email-treding@nvidia.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On 10/07/2013 02:34 AM, Thierry Reding wrote: > From: Mikko Perttunen <mperttunen@nvidia.com> > > The Tegra114 display controller is backwards-compatible with previous > generations of the Tegra SoC. No code changes are required. If the HW is backwards-compatible, then there's no need to add extra compatible values to the driver; just write the following in the DT, and it'll just work: compatible = "nvidia,tegra114-dc", "nvidia,tegra30-dc";
On Fri, Oct 11, 2013 at 04:14:27PM -0600, Stephen Warren wrote: > On 10/07/2013 02:34 AM, Thierry Reding wrote: > > From: Mikko Perttunen <mperttunen@nvidia.com> > > > > The Tegra114 display controller is backwards-compatible with previous > > generations of the Tegra SoC. No code changes are required. > > If the HW is backwards-compatible, then there's no need to add extra > compatible values to the driver; just write the following in the DT, and > it'll just work: > > compatible = "nvidia,tegra114-dc", "nvidia,tegra30-dc"; One reason why I thought it might be useful to still include this, even though unnecessary, was to match it to the host1x_drm_subdevs table. We can probably remove the entry from there as well, though. As far as I can tell, the same holds for Tegra30, which is also backwards compatible with Tegra20 but the DTS doesn't contain the Tegra20 compatible. So to keep ABI compatibility we'll need to keep the nvidia,tegra30-dc in the driver's match tables, but I could still update the DTS to include the nvidia,tegra20-dc for correctness. Does that make sense? Thierry
On 10/12/2013 05:32 AM, Thierry Reding wrote: > On Fri, Oct 11, 2013 at 04:14:27PM -0600, Stephen Warren wrote: >> On 10/07/2013 02:34 AM, Thierry Reding wrote: >>> From: Mikko Perttunen <mperttunen@nvidia.com> >>> >>> The Tegra114 display controller is backwards-compatible with >>> previous generations of the Tegra SoC. No code changes are >>> required. >> >> If the HW is backwards-compatible, then there's no need to add >> extra compatible values to the driver; just write the following >> in the DT, and it'll just work: >> >> compatible = "nvidia,tegra114-dc", "nvidia,tegra30-dc"; > > One reason why I thought it might be useful to still include this, > even though unnecessary, was to match it to the host1x_drm_subdevs > table. We can probably remove the entry from there as well, > though. > > As far as I can tell, the same holds for Tegra30, which is also > backwards compatible with Tegra20 but the DTS doesn't contain the > Tegra20 compatible. So to keep ABI compatibility we'll need to keep > the nvidia,tegra30-dc in the driver's match tables, but I could > still update the DTS to include the nvidia,tegra20-dc for > correctness. > > Does that make sense? Yes, fixing the Tegra30 .dtsi file sounds like a good idea.
On Mon, Oct 14, 2013 at 12:05:18PM -0600, Stephen Warren wrote: > On 10/12/2013 05:32 AM, Thierry Reding wrote: > > On Fri, Oct 11, 2013 at 04:14:27PM -0600, Stephen Warren wrote: > >> On 10/07/2013 02:34 AM, Thierry Reding wrote: > >>> From: Mikko Perttunen <mperttunen@nvidia.com> > >>> > >>> The Tegra114 display controller is backwards-compatible with > >>> previous generations of the Tegra SoC. No code changes are > >>> required. > >> > >> If the HW is backwards-compatible, then there's no need to add > >> extra compatible values to the driver; just write the following > >> in the DT, and it'll just work: > >> > >> compatible = "nvidia,tegra114-dc", "nvidia,tegra30-dc"; > > > > One reason why I thought it might be useful to still include this, > > even though unnecessary, was to match it to the host1x_drm_subdevs > > table. We can probably remove the entry from there as well, > > though. > > > > As far as I can tell, the same holds for Tegra30, which is also > > backwards compatible with Tegra20 but the DTS doesn't contain the > > Tegra20 compatible. So to keep ABI compatibility we'll need to keep > > the nvidia,tegra30-dc in the driver's match tables, but I could > > still update the DTS to include the nvidia,tegra20-dc for > > correctness. > > > > Does that make sense? > > Yes, fixing the Tegra30 .dtsi file sounds like a good idea. Done. Thierry
diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c index 71cbda6..e8a0f5f 100644 --- a/drivers/gpu/drm/tegra/dc.c +++ b/drivers/gpu/drm/tegra/dc.c @@ -1183,6 +1183,7 @@ static int tegra_dc_remove(struct platform_device *pdev) } static struct of_device_id tegra_dc_of_match[] = { + { .compatible = "nvidia,tegra114-dc", }, { .compatible = "nvidia,tegra30-dc", }, { .compatible = "nvidia,tegra20-dc", }, { }, diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c index f9def3c..e958960 100644 --- a/drivers/gpu/drm/tegra/drm.c +++ b/drivers/gpu/drm/tegra/drm.c @@ -489,6 +489,7 @@ static const struct of_device_id host1x_drm_subdevs[] = { { .compatible = "nvidia,tegra30-dc", }, { .compatible = "nvidia,tegra30-hdmi", }, { .compatible = "nvidia,tegra30-gr2d", }, + { .compatible = "nvidia,tegra114-dc", }, { /* sentinel */ } };