Re: [RFC PATCH v4 00/13] module: core code clean up

[Christophe Leroy <christophe.leroy@csgroup.eu>]

Le 03/02/2022 à 01:20, Luis Chamberlain a écrit :
> On Sun, Jan 30, 2022 at 09:32:01PM +0000, Aaron Tomlin wrote:
>> Hi Luis,
>>
>> As per your suggestion [1], this is an attempt to refactor and split
>> optional code out of core module support code into separate components.
>> This version is based on branch mcgrof/modules-next since a97ac8cb24a3/or
>> modules-5.17-rc1. Please let me know your thoughts.
>>
>> Changes since v1 [2]:
> 

I have another comment: I think patch 5 should be dropped.

Having something behave based on a CONFIG_ARCH_HAS_SOMETHING item is 
wrong. It is not because a plateform selects 
CONFIG_ARCH_HAS_STRICT_MODULE_RWX that the module core should behave 
differentely than with other platforms as far as the user has not 
selected CONFIG_STRICT_MODULE_RWX.

And the topic here is wrong. It is a coincidence if making that stuff 
depend on CONFIG_ARCH_HAS_STRICT_MODULE_RWX works. This is just because 
the only architectures that do the module allocation without Exec flag 
are architectures that have also selected 
CONFIG_ARCH_HAS_STRICT_MODULE_RWX. But it should also work on other 
architectures.

I don't know exactly what was the motivation for commit 93651f80dcb6 
("modules: fix compile error if don't have strict module rwx") at the 
first place but it is just wrong and we should fix it.

module_enable_x() should work just fine regardless of 
CONFIG_ARCH_HAS_STRICT_MODULE_RWX.

Thanks
Christophe