Fix two building errors when we build the kernel in the non-src folder #3215
Conversation
If we build the kernel with "-O=$non-src-folder", this driver will introdcue a building error because of the header's location. Signed-off-by: Hui Wang <[email protected]>
If we build the kernel with "-O=$non-src-folder", this driver will introdcue a building error because of the header's location. Signed-off-by: Hui Wang <[email protected]>
|
@kbingham This reverses your commit - any comments? |
|
@pelwell Ack on this. It looks like my earlier fix was squashed in, so perhaps it got squashed or merged with some other change that affected things here. Anyway, I can confirm Jason's patch compiles VC_SM where as without Jason's patch - there is a failure to build the VC_SM code. I have not tested the rtl8192cu. |
|
Thanks. I'll take the 8192 change on trust. |
|
Well, to put your mind at ease, as I had the build environment all set up I've added CONFIG_RTL8192CU=y and it built successfully. Out of interest, why is this driver included in the RPi kernel? Is there a specific use case needed to support this driver? If not - I'd perhaps recommend dropping it! |
|
|
When using 4.9 kernel we switched to mainline driver and got a lot of complaints about unreliable wireless. We currently build both and blacklist one so users can switch. It is possible the driver has improved in the meantime, but I'm not too optimistic. |
It is because the building process can't find the needed header files. And this already introduced building errors on the branches of 5.2.y and 5.3.y for RPi4, and I guess all branches need these two fixes, it is safe to merge these two patches to base branch then it will apply to all branches automatically?