New ECN: Test points and 0R resistors for camera interface

[Rask Ingemann Lambertsen-2]

Index: docs/ecn/STATUS
===================================================================
--- docs/ecn/STATUS (revision 5626)
+++ docs/ecn/STATUS (working copy)
@@ -39,6 +39,7 @@
 0037    Execute Route ROUT1/LOUT1 back to LINE1/LINE2 input
 0038    Discuss Loudspeaker EMI filter
 0039 Discuss Insert 0R resistors into SPI signals
+0040 Discuss Test points and 0R resistors for camera interface
 
 Open ECNs
 ---------
@@ -63,3 +64,4 @@
 0037    AUDIO
 0038    AUDIO
 0039 CPU, LCM
+0040 CPU

--- /dev/null 2009-09-06 17:58:44.681774387 +0200
+++ docs/ecn/ecn0040.txt 2009-09-11 03:01:38.000000000 +0200
@@ -0,0 +1,24 @@
+Test points and 0R resistors for camera interface
+
+
+We should allow for experimentation with the camera interface. The
+24xx/64xx camera interface would be the obvious choice of interface to
+connect a camera to, but a few details need verification:
+
+The SC32442B manual specifies a short VSYNC pulse (high) between lines.
+Several camera modules instead keep VSYNC high during lines. It is not clear
+from the manual if simply inverting VSYNC polarity is enough to make it
+work.
+
+The SC32442B manual isn't quite clear on how to receive e.g. JPEG compressed
+data as offered by some cameras. E.g. will we need a separate VSYNC
+interrupt pin to deal with variable frame data lengths?
+
+The change is to route unused pins to test points and insert 0R into the
+rest.
+
+Impact: Little if any. The signals currently at the camera port are less
+than 1 Hz or not used at all.
+
+
+Author: Rask Ingemann Lambertsen <[hidden email]>


--
Rask Ingemann Lambertsen
Danish law requires addresses in e-mail to be logged and stored for a year

_______________________________________________
gta02-core mailing list
[hidden email]
https://lists.openmoko.org/mailman/listinfo/gta02-core

[Rask Ingemann Lambertsen-2]

On Fri, Sep 11, 2009 at 03:14:32AM +0200, Rask Ingemann Lambertsen wrote:
>
> --- /dev/null 2009-09-06 17:58:44.681774387 +0200
> +++ docs/ecn/ecn0040.txt 2009-09-11 03:01:38.000000000 +0200
> @@ -0,0 +1,24 @@
> +Test points and 0R resistors for camera interface

   I checked it in as ECN0041 with the current status of the pins added:

CAMDATA0/GPJ0 Insert 0R (BT_PIO5).
CAMDATA1/GPJ1 Add test point.
...
CAMDATA6/GPJ6 Add test point.
CAMDATA7/GPJ7 Insert 0R (WLAN_GPIO0).
CAMPCLK/GPJ8 Add test point.
CAMVSYNC/GPJ9 Add test point.
CAMHREF/GPJ10 Insert 0R (WLAN_DATA2).
CAMCLKOUT/GPJ11 Has 0R and test point (CHIP_PWD).
CAMRESET/GPJ12 Insert 0R (WLAN_nRESET).

CAMRESET is not really a special function, any GPIO could be used for the
camera reset signal.

--
Rask Ingemann Lambertsen
Danish law requires addresses in e-mail to be logged and stored for a year

_______________________________________________
gta02-core mailing list
[hidden email]
https://lists.openmoko.org/mailman/listinfo/gta02-core

[Werner Almesberger]

Rask Ingemann Lambertsen wrote:
> > +Test points and 0R resistors for camera interface

How about test points under the LCM ? There's plenty of space in
that area and we can keep traces short.

Regarding whether VSYNC has to be a short pulse: no idea. I also
looked in the 6410 docs, and they don't say anything that suggests
that a long pulse will work.

Regarding variable-sized (JPEG) frames: the interface is clearly
not designed for this. So I guess, once more, only trial and error
will tell :)

- Werner

_______________________________________________
gta02-core mailing list
[hidden email]
https://lists.openmoko.org/mailman/listinfo/gta02-core