Attention is currently required from: Arthur Heymans, Benjamin Doron, Felix Held, Patrick Rudolph.
4 comments:
File Documentation/arch/x86/x86_64.md:
Patch Set #1, Line 3: experimental
I'd prefer to make x86_64 support default in the next releases.
That sounds great to me.
The large memory model causes the compiler to emit 64bit addressing
instructions, which are not only slower, but also increase code size.
The x86_64 code runs a bit faster, but takes longer to be fetched from SPI as the size is bigger. […]
So can change this to:
```
Page tables can be used to provide security benefits, such as by marking memory
as non-executable or removing it entirely. This could be useful for SMM: mark
regular DRAM as NX.
The large memory model causes the compiler to emit 64bit addressing
instructions, which increases code size. In theory, this is roughly
made up for by the faster execution of the x86_64 code.
```
Patch Set #1, Line 28: must generate page tables at build time
It's describing the current implementation, not a requirement.
Sure, Can we update the text to reflect that? Maybe:
```
For the current implementation, coreboot generates page tables at build time that configure 0-4GiB
as WB, using 2MiB-pages.
```
Patch Set #1, Line 140: Here's a list of known issues:
It's under "KVM enabled qemu", so yes it's referring to qemu only. Real hardware is fine.
Change to `Known issues for QEMU:`?
To view, visit change 79160. To unsubscribe, or for help writing mail filters, visit settings.