On Wednesday 29 August 2012 04:02 PM, Tomi Valkeinen wrote:
On Fri, 2012-08-24 at 18:21 +0530, Archit Taneja wrote:
+enum omap_dss_output_id {
+ OMAP_DSS_OUTPUT_DPI = 1 << 0,
+ OMAP_DSS_OUTPUT_DBI = 1 << 1,
+ OMAP_DSS_OUTPUT_SDI = 1 << 2,
+ OMAP_DSS_OUTPUT_DSI1 = 1 << 3,
+ OMAP_DSS_OUTPUT_VENC = 1 << 4,
+ OMAP_DSS_OUTPUT_DSI2 = 1 << 5,
+ OMAP_DSS_OUTPUT_HDMI = 1 << 6,
+};
I'm not sure about this. We already have enum omap_display_type. If you
need the instance number, you could have that as a separate int field.
Where do you need the output_id?
output_id is used to take care of situations where there our multiple
outputs of the same type, like DSI1 and DSI2. An enum helps when we
check if an overlay manager supports that output instance or not. For
ex, on OMAP4, LCD1 connects to DSI1 and not DSI2.
I add a func called dss_feat_get_supported_outputs(channel) later to
check for this. When setting a new output for a manager, we just do an
'&' to see if the output in question is in the mask of the manager's set
of supported outputs.
After thinking about this, I think we should remove the
omap_display_type and the supported_displays stuff. It doesn't really
work anymore, as we have more complex connections with omap4+. With a
quick glance to the code, I think it should be quite easy to remove it.
And we need something like your omap_dss_output_id. I'm just not sure
about the enum. But perhaps it's the easiest option for now, as some
kind of array of similar would be more complex to implement, and I'm not
sure if it really gives anything. You could move the VENC away from
between DSI1 and DSI2, though. I'm not sure why you put VENC in between
=).
I put the outputs in the order in which they were introduced in OMAPs,
I'm not sure why I did that, maybe I copied that approach after looking
at some other enum. I'll change it.
Archit
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
