Anil Sasidharan wrote:
Hi Dirk,
The m5-rc14 release of android kernel source has updated
the binder module as one of the "misc" modules (along with other
Android specific "misc" modules like power, logger etc.). So you won't
find the binder entry in /proc/devices, instead you will have to cat
/proc/misc to view the minor number of binder. Hence there is no
significance for the /dev/binder device node and you can remove that
entry from your file system. The ".config" file that was used to
configure my kernel is attached herewith. This config file does not
have an entry for "CONFIG_BINDER" as it is obsolete and got a new
configuration entry "CONFIG_ANDROID_BINDER_IPC".
Hope this will help you to overcome the problem.
Yes, many thanks (to Raymond as well who gave similiar hint)! This
fixed the issue with /dev/binder! Seems that I was confused regarding
the differences in web description of (older) binder interface and
changes in m5-rc14.
I updated wiki page [1] accordingly.
Looking at strace runtime output, everything seems to be fine now. But
... ;)
But seems that I'm still missing the last (?) step before success:
I still have this screen blanking issue. At boot I get penguin logo, I
can "echo Hello world > /dev/tty0", ts_calibrate works. Everything
looks fine. But just around
writev(3, [{"\4", 1}, {"SurfaceFlinger\0", 15}, {"SurfaceFlinger is
starting\0", 27}],3
from strace runtime output the screen is *blanked* (no Android string,
no red eye, just switched to black with backlight still on). Similiar
like when the screensaver is activated after ~10min.
I wonder if on OSK there is a simliar issue like Raymond had in [2].
Looking into Raymond's fix for this in his patch [3] I liked to apply
it to OSK fb driver as well and give it a try. But I can't locate the
corresponding lines and wonder that I don't have a omap24xxfb.c (in
2.6.23-omap1 from git).
Any ideas regarding this screen blanking issue? Or is it possible that
because of only 32MB everything looks fine, no error messages, but
simply no output to the screen?
Many thanks
Dirk
[1] http://elinux.org/Android_on_OMAP
[2] http://marc.info/?l=linux-omap&m=120399654528241&w=2
[3] drivers/video/omap/omap24xxfb.c in Raymond's
android_m5r14_vs_omap2430_2621.patch:
-- cut --
Index: drivers/video/omap/omap24xxfb.c
===================================================================
--- drivers/video/omap/omap24xxfb.c (revision 4)
+++ drivers/video/omap/omap24xxfb.c (working copy)
@@ -228,7 +228,7 @@
break;
}
v->xres_virtual = v->xres;
- v->yres_virtual = v->yres;
+ v->yres_virtual = v->yres * 2;
}
return v;
}
@@ -1154,6 +1154,8 @@
wait_for_reg_sync(SCHEDULE_WAIT, oinfo->timeout);
}
+ schedule_timeout_interruptible(HZ/50);
+
DBGLEAVE;
return ret;
}
-- cut --
Great!
As it seems that I still have some problems with /dev/binder and my
display settings: Do you kindly like to send me:
- Output of your
cat /proc/devices | grep binder
ls -la /dev/binder
- Your kernel .config
?
Many thanks
Dirk
--
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
