On Sat, 25 Mar 2000, David Brownell wrote:
> Could someone who knows how it's _supposed_ to be working please
> sumamrize the status of power management support and USB?
Alan posted an interim fix to this some time ago:
http://electricrain.com/lists/archive/linux-usb/2000/03/msg00628.html
I couldn't find the explanation why it was not applied but I'm sure Alan
can fill us in if I don't recall it correctly; disabling/enabling of
irqs should be handled by the generic PM code and not by the individual
drivers.
So the question to Alan, is somebody working on it and what kind of
changes are exactly needed to the PM layer? Btw. are my private mails to
you getting there at all?
> As a data point, I've got a laptop running 2.3.99-pre3 and I've
> seen these troublesome failure modes:
>
> - When screen blanking kicks in, typing on my USB
> keyboard doesn't wake things up. Gotta use the
> builtin keyboard to that, then go back to USB.
Odd, I'm at work so can not test this but as I see this the code to
unblank both is in the same place. Function handle_scancode does
effectively a show_console (a patch here makes it explicit):
diff -urN --exclude=*~ linux/drivers/char/keyboard.c
linux-2.3.99-pre3/drivers/char/keyboard.c
--- linux/drivers/char/keyboard.c Tue Feb 15 01:31:14 2000
+++ linux-2.3.99-pre3/drivers/char/keyboard.c Sun Mar 26 16:33:04 2000
@@ -207,8 +207,7 @@
pm_access(pm_kbd);
- do_poke_blanked_console = 1;
- tasklet_schedule(&console_tasklet);
+ show_console();
add_keyboard_randomness(scancode | up_flag);
tty = ttytab? ttytab[fg_console]: NULL;
Maybe you're experiencing some kind of BIOS initiated blanking which is
completely ignorant of the standard console blanking?
Maybe mouse events should trigger unblank as well. Or is this handled by
gpm, I don't know?
diff -urN --exclude=*~ linux/drivers/usb/mousedev.c
linux-2.3.99-pre3/drivers/usb/mousedev.c
--- linux/drivers/usb/mousedev.c Mon Mar 20 04:29:40 2000
+++ linux-2.3.99-pre3/drivers/usb/mousedev.c Sun Mar 26 16:39:43 2000
@@ -37,6 +37,7 @@
#include <linux/init.h>
#include <linux/input.h>
#include <linux/config.h>
+#include <linux/kbd_kern.h>
#ifndef CONFIG_MOUSEDEV_SCREEN_X
#define CONFIG_MOUSEDEV_SCREEN_X 1024
@@ -141,6 +142,7 @@
list = list->next;
}
+ show_console();
wake_up_interruptible(&mousedev->wait);
}
> - When I shutdown the machine and power it back up,
> USB doesn't seem to work at all. rmmod/modload
> of usb-ohci makes it come alive again.
Infact I got a nice oops.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
