[ANDROID KERNEL PATCH 5/7] Update Kernel core files
Sean McNeil
sean at mcneil.com
Thu Jan 29 13:56:10 CET 2009
Also, I don't understand this staging area at all. You don't have
drivers, etc. in staging when you've polluted the main code like
kernel/printk.c. The staging stuff is to isolate things and it certainly
isn't isolated when you do that.
Also, the Android drivers are not going into drivers/android anymore. In
staging, they shouldn't be in staging/android. They should be in
staging/misc.
Finally, all the power management stuff goes into kernel/power. It
doesn't belong in drivers/staging.
I actually have everything patched correctly here with the exception of
the drivers going into drivers/misc directly, but Michael seems to be
showing good initiative in getting things into the kernel. It would be
great if it could be done correctly, however, and placed in the
appropriate place. i.e. kernel/power and either drivers/misc or
drivers/staging/misc.
Andy Green wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Somebody in the thread at some point said:
> | Update kernel core files to support android suspend/resume. These
> | change are conditioned by ifdef
>
> Hi -
>
> Here's a bug that breaks build without
> CONFIG_ANDROID_CONSOLE_EARLYSUSPEND.. should it be #ifdef not #ifndef?
>
> | --- a/kernel/printk.c
> | +++ b/kernel/printk.c
> | @@ -82,7 +82,9 @@
> EXPORT_SYMBOL(printk_emergency_debug_spew_send_string);
> | * driver system.
> | */
> | static DECLARE_MUTEX(console_sem);
> | +#ifdef CONFIG_ANDROID_CONSOLE_EARLYSUSPEND
> | static DECLARE_MUTEX(secondary_console_sem);
> | +#endif
>
> | +#ifndef CONFIG_ANDROID_CONSOLE_EARLYSUSPEND
> | if (console_suspended) {
> | down(&secondary_console_sem);
> | return;
> | }
> | +#endif
>
> - -Andy
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.9 (GNU/Linux)
> Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org
>
> iEYEARECAAYFAkmBpLEACgkQOjLpvpq7dMqdjwCfbubhUyZzIvaxG9Ro6utWQPRN
> bpcAn1ZRYMggm0gAdrEbQsZnrU7sAth4
> =B3o0
> -----END PGP SIGNATURE-----
>
More information about the openmoko-kernel
mailing list