Change in ...coreboot[master]: cpu/x86/pae/pgtbl: Add memset with PAE

[Patrick Rudolph (Code Review)]

Patrick Rudolph has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/31549 )

Change subject: cpu/x86/pae/pgtbl: Add memset with PAE
......................................................................

Patch Set 6:

(4 comments)

https://review.coreboot.org/#/c/31549/5/Documentation/security/memory_cleari... 
File Documentation/security/memory_clearing.md:

https://review.coreboot.org/#/c/31549/5/Documentation/security/memory_cleari... 
PS5, Line 40: ## Architecture specific implementations
            : 
            : * [x86_32 architecture](../arch/x86/index.md)
> that doesn't look helpful to me as it points to generic x86 docs.

Done

https://review.coreboot.org/#/c/31549/5/src/cpu/x86/pae/pgtbl.c 
File src/cpu/x86/pae/pgtbl.c:

https://review.coreboot.org/#/c/31549/5/src/cpu/x86/pae/pgtbl.c@114 
PS5, Line 114: scratch2
> vmem_addr?

Done

https://review.coreboot.org/#/c/31549/5/src/cpu/x86/pae/pgtbl.c@117 
PS5, Line 117: scratch
> pgtbl_buf? also it's size should be stated somewhere.

Done

https://review.coreboot.org/#/c/31549/5/src/cpu/x86/pae/pgtbl.c@135 
PS5, Line 135: assert(scratch);
             : 	assert(IS_ALIGNED((uintptr_t)scratch, 4096));
             : 	assert(IS_ALIGNED((uintptr_t)scratch2, 2 * MiB));
             : 	assert(((uintptr_t)dest > (uintptr_t)scratch2 + 2 * MiB) ||
             : 	       ((uintptr_t)dest < (uintptr_t)scratch2));
             : 	assert(((uintptr_t)dest < (uintptr_t)scratch) ||
             : 	      ((uintptr_t)dest > (uintptr_t)scratch + sizeof(struct pg_table)));
> we don't always enforce asserts, so this may be unreliable.

Should I just return with an error?