dcpext0 behaves like dcp on M1* devices and can sleep after display
init. This has the advantage of not breaking macOS when starting with an
initialized display.
dcpext* are according to Apple's tech specs slightly more powerful than
dcp. They are advertised as 6K at 60Hz while dcp seems to be limited to
5K at 60Hz.
Signed-off-by: Janne Grunau <j@jannau.net>
Apperently just informative as display init on M2 Ultra in m1n1 worked
as expected despite passing '0' as die number.
Signed-off-by: Janne Grunau <j@jannau.net>
On M2* desktop systems the HDMI output is not managed by dcp/dcp0. This
allows more flexibility, for example dcp on the M2 Mac mini can be
routed to the USB-C/ThunderBolt ports.
For m1n1 this adds quite ab bit of complexity. In addition to the DCP
iboot endpoint/protocol two additional endpoints have to be started.
Only the dptx-port endpoint has a meaningful AP <=> DCP communication.
In addition we start the system endpoint to optionally enable more
verbose syslog messages (if RTKIT_SYSLOG is defined).
In addition the dptx-phy or lpdptx-phy has to be programmed. Those two
phy look mostly identical. MacOS seems to use a static DP configuration
(4 lanes, HBR3*) regardless of the connected display.
The added SMC gpio support is used to power the MCDP29xx DP to HDMI
converter on.
Config is unfortunately a mess since Apple can't name the device tree
nodes consistently. On M2 Ultra devices only the four dcpext*/dispext*
on each die are used. M2 Ultra support is untested.
* link rate might differ on t602x based devices for 8k60 support
Signed-off-by: Janne Grunau <j@jannau.net>
This needs to ensure that the data processing does not block on
message retrieval from the mailbox. Command processing must not block
and have to ensure that afk_epic_command() receives the reply to its
command.
DCP's dptx-port endpoint (endpoint 0x2a) can send notifies at any time
which needs to be processed in the right context.
Signed-off-by: Janne Grunau <j@jannau.net>
This stops DCP from killing our modeset if the connection cycles.
Also force a (potential) configure cycle if the display is external;
this makes sure updated stage2s will have a chance at fixing issues of
old stage1s. Modesetting is fast when it's the same mode as before.
Signed-off-by: Hector Martin <marcan@marcan.st>
