paddr_t should be larger than 32 bits when CFG_WITH_LPAE is enabled

[glneo]

Like in the Linux kernel, when LPAE support is enabled the type for storing physical addresses should be a >=40 bit type (64 bit). 'paddr_t' is unconditionally typedef'd as 'uintptr_t' making it is 32 bits wide on 32 bit platforms even when LPAE is supported by the platform.

[jenswi-linaro]

Agree, it has always been the intent to have paddr_t as large as needed.
Patches are welcome! :-)

[glneo]

Was working on a patch for this, but ran into the issue that if we are a 32-bit core, when OPTEE casts from 'void *' to 'paddr_t' we get a build warning/error (cast from pointer to integer of different size), this is done in a large number of places. I'll find the time for some patches on this here in a little bit. :)

[jbech-linaro]

@glneo , have you been able to look at this? If yes, any updates? (I'm trying to clean up open issues).

[glneo]

No, sorry I haven't had any time to look further into this, a quick attempt at a fix lead to further complications as this is used in several places where I do not know what the correct behavior should be, I'll submit an attempt but cannot guarantee correctness.

[github-actions]

This issue has been marked as a stale issue because it has been open (more than) 30 days with no activity. Remove the stale label or add a comment saying that you would like to have the label removed otherwise this issue will automatically be closed in 5 days. Note, that you can always re-open a closed issue at any time.