The "one binary" decision isn't relevant here, is it? It would make more sense to implement 5-level paging within the base EDK2 architecture. This would allow that feature to be tested in

Continue my comments from here. It is because of the *one binary* so that we have to include all the drivers in the firmware binary. Some of the DXE drivers are not called in TDX, but they're

Leif and Bob: Thanks for your test and review. I create PR https://github.com/tianocore/edk2/pull/1690 for this patch. Thanks Liming to data. "); ");

Continue my comments from here. CFV is populated in post build. We can provide such python scripts to do the SB keys enrollment. As I mentioned above, SB keys are enrolled in post build phase. We

Reminder: TianoCore Design Meeting - APAC/NAMO When: 06/11/2021 9:30am to 10:30am (UTC+08:00) Asia/Shanghai Where: Microsoft Teams Organizer: Ray Ni ray.ni@... View

Hi Liming, Note - this does not only affect GCC49: it also affects at least GCC5 and CLANG38. This patch resolves the issue. Tested-by: Leif Lindholm <leif@...> (GCC49, GCC5,

Fix typographical error in the OvmfPkg/README by correcting the 'an' mistyped as 'and'. Cc: Ard Biesheuvel <ardb+tianocore@...> Cc: Jordan Justen <jordan.l.justen@...> Cc:

Reviewed-by: Bob Feng <bob.c.feng@...>

Ray: When convert the image to ACPI data, only data section is required. So, we can update GenFw tool not do this check when convert ACPI data. I just send the patch to update GenFw tool. This

Cool! Thank you so much!! It answers my question that why CI cannot detect such failure because it's using GCC5+. Leif, Can you verify that this can resolve your GCC49 issue? Thanks, Ray

Leif, OVMF Bhyve GCC build succeeds without GenFW assertion issue. I dumped the ELF image generated for OvmfPkg/Bhyve/AcpiTables/AcpiTables. Below output shows that .text does have "AX" set: --- $

With GCC49 tool chain, aslc file is compiled into elf image. But, its text section has no CODE attribute. So, it can't be detected by new GenFw tool.For this type file. its text section is not

Leif, Sorry to hear that. It seems a CI gap that doesn't capture such errors. It looks like the logic update to detect .text section doesn't work in your case. I am trying to build the OVMF Bhyve

No. Actually add the braces. Again, namespaces. And here. And here. Etc. Please add some vendor prefix to all of these. Vendor prefix. / Leif

Contents of ACPI tables not reviewed. Creation modes on many of these files are incorrect (as commented on cover letter). Address this, and you can have: Acked-by: Leif Lindholm

This should be a central function. If we have spectacularly failed up until this point, please do this separately. OK, this has officially gone too far. Please add proper vendor prefixes to anything

Again - on your head be it to start symbol names with Acpi and hope for no clashes. But given that: Reviewed-by: Leif Lindholm <leif@...>

Reviewed-by: Leif Lindholm <leif@...> / Leif

This isn't obvious enough. I'm a bear of very little brain, and I could possibly figure out what's happening here, given time, but please put a proper description in commit message (and possibly

This looks like something that would be a good target to commonalise between platforms in the future. But for now: Reviewed-by: Leif Lindholm <leif@...> / Leif