[opensuse-kernel] questionable MFD sub-options
As it seems like that subsystem isn't really getting its Kconfig right, I wonder whether what is currently getting enabled as built-in shouldn't be disabled instead. The list of things appears to vary between configs, but the net effect imo should never be MFD_CORE=y (i.e. any option that results in that should be a candidate for disabling, along with - as was done in other areas - options that can't possibly occur on systems we target, e.g. EZX_PCAP described to be needed for certain phones only, yet enabled across the board). Thoughts? Jan -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
Hi Jan, Le jeudi 06 septembre 2012 à 07:37 +0100, Jan Beulich a écrit :
As it seems like that subsystem isn't really getting its Kconfig right, I wonder whether what is currently getting enabled as built-in shouldn't be disabled instead. The list of things appears to vary between configs, but the net effect imo should never be MFD_CORE=y (i.e. any option that results in that should be a candidate for disabling, along with - as was done in other areas - options that can't possibly occur on systems we target, e.g. EZX_PCAP described to be needed for certain phones only, yet enabled across the board).
I completely agree. For some reason, many MFD drivers can't be built as modules. I complained about that a few times in the past but nobody cared. I think it is a reasonable assumption that any driver which can only be built-in is not suitable for a generic kernel configuration and should be disabled. I think all these drivers are for devices only found on embedded boards. If anyone misses one of these drivers, let them complain and provide a patch fixing the driver in question. Thanks, -- Jean Delvare Suse L3 -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 9/6/12 2:37 AM, Jan Beulich wrote:
As it seems like that subsystem isn't really getting its Kconfig right, I wonder whether what is currently getting enabled as built-in shouldn't be disabled instead. The list of things appears to vary between configs, but the net effect imo should never be MFD_CORE=y (i.e. any option that results in that should be a candidate for disabling, along with - as was done in other areas - options that can't possibly occur on systems we target, e.g. EZX_PCAP described to be needed for certain phones only, yet enabled across the board).
Thoughts?
Jan
Agreed.
commit ff91de486aec5f4cf0cd263e3bea7dba722093c7
Author: Jeff Mahoney
On 11.09.12 at 03:44, Jeff Mahoney
wrote: On 9/6/12 2:37 AM, Jan Beulich wrote: As it seems like that subsystem isn't really getting its Kconfig right, I wonder whether what is currently getting enabled as built-in shouldn't be disabled instead. The list of things appears to vary between configs, but the net effect imo should never be MFD_CORE=y (i.e. any option that results in that should be a candidate for disabling, along with - as was done in other areas - options that can't possibly occur on systems we target, e.g. EZX_PCAP described to be needed for certain phones only, yet enabled across the board). Thoughts?
Agreed.
commit ff91de486aec5f4cf0cd263e3bea7dba722093c7 Author: Jeff Mahoney
Date: Mon Sep 10 21:39:18 2012 -0400 config: disabled MFD drivers that aren't buildable as modules
This allows MFD_CORE to be modular on all arch/flavors except for ARM and the non-PAE flavors on i386, since it's used with the OLPC XO-1.
But you didn't apply/push this yet, did you? Jan -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 9/11/12 2:51 AM, Jan Beulich wrote:
On 11.09.12 at 03:44, Jeff Mahoney
wrote: On 9/6/12 2:37 AM, Jan Beulich wrote: As it seems like that subsystem isn't really getting its Kconfig right, I wonder whether what is currently getting enabled as built-in shouldn't be disabled instead. The list of things appears to vary between configs, but the net effect imo should never be MFD_CORE=y (i.e. any option that results in that should be a candidate for disabling, along with - as was done in other areas - options that can't possibly occur on systems we target, e.g. EZX_PCAP described to be needed for certain phones only, yet enabled across the board). Thoughts?
Agreed.
commit ff91de486aec5f4cf0cd263e3bea7dba722093c7 Author: Jeff Mahoney
Date: Mon Sep 10 21:39:18 2012 -0400 config: disabled MFD drivers that aren't buildable as modules
This allows MFD_CORE to be modular on all arch/flavors except for ARM and the non-PAE flavors on i386, since it's used with the OLPC XO-1.
But you didn't apply/push this yet, did you?
Looks like I did indeed forget to push. Thanks. - -Jeff - -- Jeff Mahoney SUSE Labs -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.18 (Darwin) Comment: GPGTools - http://gpgtools.org Comment: Using GnuPG with Mozilla - http://www.enigmail.net/ iQIcBAEBAgAGBQJQT0anAAoJEB57S2MheeWyVtYQAKe6BVhuSkublzIzEsswESq5 CDpjIoY9OvVoHNW2+7fB4AvbhkLLW/4A0VpgzfHpG57UkmZOb6tgGFppUxg+6XrS R/pqDU1mPBd3rezAUEccX+3k2vXHiXuB9UMDq52/0Q8H8/mHGENueGzF5bpKdT1P GAvK0uPBibplTgtFUxYQ70jnZR6Iyr7Ua+1KBVTLuvgUZkaCOaXizybQxqHVNmt+ SZbbciDvOAS5mZMvYlltV7lfnXCmIPpWoY0NBQCa/eHsytZoLCfNAwClZxD8Gwe3 F03AZ3jPmHQLJ4nRsPHq3fHjxfIDFHKT8RTc3zs6aqcvRalvY80UmMreQV5S/bag E1ZgA9rYwS2NMqLl0iHdF7iIe8mtvPAinywwSAXS/AcG/3hkll9bwVOmFf9c4H64 Rzt24Az2olclz+kaDF4Hd+p02rinhiiBkqPrw72Mv8b9nukgP38zQ9W8q5MvLVEy i7bmES8UcJkdiXgB6g1cYYGKSMmSOdsUsADoCZUIxGE0wieiq+2Go1RkYYE3dOXl WlN9KAzP40+UzYXEGuII3VEWEKcK72qbusxw8a+RPBOLhbVCrY47prktRu1w6Nzg caEUf8Oc5YfH69VtrWly760erD4AfpcBSuGVIa470KjwaETe1yzQW3/aCbRmjQqU NmqtGqTDgN0YHhHLu23j =ym3x -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
participants (3)
-
Jan Beulich
-
Jean Delvare
-
Jeff Mahoney