778129 - GLContextProviderEGL.cpp fails to compile on mingw-w64

Status: RESOLVED FIXED

(Core :: Graphics, defect)

Description

[Jacek Caban]

Attachment: fix v1.0

Because on win64 sizeof(long) != sizeof(void*), GCC is strict about preventing pointer to any int with smaller size to avoid common mistakes. Thus, the cast must be made more carefully.

Comment 1

[Jacek Caban]

James, ping

Comment 2

[James Willcox (:snorp) ([e-mail]) (he/him)]

(In reply to Jacek Caban from comment #0)
> Created attachment 646557 [details] [diff] [review]
> fix v1.0
> 
> Because on win64 sizeof(long) != sizeof(void*), GCC is strict about
> preventing pointer to any int with smaller size to avoid common mistakes.
> Thus, the cast must be made more carefully.

Sorry for the late reply. Slipped off my radar somehow.

Anyway, I think the sizeof(long) != sizeof(void*) thing is true elsewhere but the cast works. This cast to uintptr_t looks incorrect to me, considering that's not what GLuint is. Can you paste the error you get? Surely it's just a warning?

Bringing in some other folks who probably have an opinion...

Comment 3

[Jacek Caban]

Yes, it's an error and it looks like this:

mozilla-central/gfx/gl/GLContextProviderEGL.cpp:938:34: error: cast from ‘void*’ to ‘GLuint {aka unsigned int}’ loses precision [-fpermissive]

I've fixed a few such cases in the codebase, and it's nothing really special here... The trick is to always cast pointers to integers of the same size. Although GLUint is not the same as uintptr_t, it's not a problem, because there will be an implicit cast. So instead of one cast that both changes pointer to integer and loses precision, we have an explicit cast that does safe pointer to integer cast and then implicit cast (in assignment expression) that loses integer precision.

Comment 4

[James Willcox (:snorp) ([e-mail]) (he/him)]

Comment on attachment 646557 [details] [diff] [review]
fix v1.0

Alright, I'm convinced :)

Comment 5

[James Willcox (:snorp) ([e-mail]) (he/him)]

Do you need me to check this in for you?

Comment 6

[Jacek Caban]

Thanks for the review. I will commit it myself.

Comment 7

[Jacek Caban]

https://hg.mozilla.org/integration/mozilla-inbound/rev/111ee5107308

Comment 8

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/111ee5107308