Parse attributes in extended VTL0 APIs by femiatl · Pull Request #653 · microsoft/litebox

femiatl · 2026-02-09T19:49:51Z

VTL0 sends data to the secure kernel during and after initial boot. Use the new attributes parameter to extend load_kdata, allowing it to process data after boot and use the improved format for sending data from VTL0. Small, simple data buffers like certificates can be sent efficiently with the same APIs used for larger aggregated data like module info.

The new data format is leveraged to reduce the size of the kernel string table passed from ~10MB (all of rodata) to ~250KB.

The validate_module and validate_kexec APIs are extended to use the new attributes.

sangho2 · 2026-02-10T17:45:15Z

@femiatl Could you please resolve the conflicts such that CI can work on this PR.

VTL0 sends data to the secure kernel during and after initial boot. Use the new attributes parameter to extend load_kdata, allowing it to process data after boot and use the improved format for sending data from VTL0. Small, simple data buffers like certificates can be sent efficiently with the same APIs used for larger aggregated data like module info. The new data format is leveraged to reduce the size of the kernel string table passed from ~10MB (all of rodata) to ~250KB. The validate_module and validate_kexec APIs are extended to use the new attributes.

github-actions · 2026-02-11T02:56:04Z

🤖 SemverChecks 🤖 ⚠️ Potential breaking API changes detected ⚠️

Click for details

--- failure function_parameter_count_changed: pub fn parameter count changed ---

Description:
A publicly-visible function now takes a different number of parameters.
        ref: https://doc.rust-lang.org/cargo/reference/semver.html#fn-change-arity
       impl: https://github.com/obi1kenobi/cargo-semver-checks/tree/v0.46.0/src/lints/function_parameter_count_changed.ron

Failed in:
  litebox_platform_lvbs::mshv::vsm::mshv_vsm_protect_memory now takes 3 parameters instead of 2, in /home/runner/work/litebox/litebox/litebox_platform_lvbs/src/mshv/vsm.rs:259
  litebox_platform_lvbs::mshv::vsm::mshv_vsm_load_kdata now takes 3 parameters instead of 2, in /home/runner/work/litebox/litebox/litebox_platform_lvbs/src/mshv/vsm.rs:336

--- failure inherent_method_missing: pub method removed or renamed ---

Description:
A publicly-visible method or associated fn is no longer available under its prior name. It may have been renamed or removed entirely.
        ref: https://doc.rust-lang.org/cargo/reference/semver.html#item-remove
       impl: https://github.com/obi1kenobi/cargo-semver-checks/tree/v0.46.0/src/lints/inherent_method_missing.ron

Failed in:
  Symbol::from_bytes, previously in file /home/runner/work/litebox/litebox/target/semver-checks/git-main/67aa94c6a750cd2791fe9edab62e6b29b9624493/litebox_platform_lvbs/src/mshv/vsm.rs:1778

--- failure method_parameter_count_changed: pub method parameter count changed ---

Description:
A publicly-visible method now takes a different number of parameters, not counting the receiver (self) parameter.
        ref: https://doc.rust-lang.org/cargo/reference/semver.html#fn-change-arity
       impl: https://github.com/obi1kenobi/cargo-semver-checks/tree/v0.46.0/src/lints/method_parameter_count_changed.ron

Failed in:
  litebox_platform_lvbs::mshv::vsm::SymbolTable::build_from_container now takes 2 parameters instead of 4, in /home/runner/work/litebox/litebox/litebox_platform_lvbs/src/mshv/vsm.rs:1844

sangho2 · 2026-02-13T15:18:49Z

It seems that this change is based on new ABI to simplify data transfer. Good idea! As @tgopinath-microsoft noted in #652, this cannot be merged yet.

One overall feedback or requested change: This PR needs some detailed comments to explain new ABI and data structures. Also, some functions are with old comments (e.g., these functions no longer get nranges), and some other functions are without any comments.

litebox_platform_lvbs/src/mshv/heki_data.rs

sangho2 · 2026-02-13T16:29:56Z

litebox_platform_lvbs/src/mshv/vsm.rs

@@ -254,12 +256,14 @@ pub fn mshv_vsm_end_of_boot() -> i64 {

 /// VSM function for protecting certain memory ranges (e.g., kernel text, data, heap).
 /// `pa` and `nranges` specify a memory area containing the information about the memory ranges to protect.


nranges is no longer used.

I updated the function parameter name, as well as the documentation for the changed functions

sangho2 · 2026-02-13T16:31:38Z

litebox_platform_lvbs/src/mshv/vsm.rs

 /// `pa` and `nranges` specify a memory area containing the information about the memory ranges to protect.
-pub fn mshv_vsm_protect_memory(pa: u64, nranges: u64) -> Result<i64, VsmError> {
+pub fn mshv_vsm_protect_memory(pa: u64, va: u64, attr: u64) -> Result<i64, VsmError> {
+    if attr == 0 {


Fine for now, but breaking ABI change like this should be gracefully handled in the future.

Agreed. Let me update the PR with some doc for breaking API as well

litebox_platform_lvbs/src/mshv/heki_data.rs

litebox_platform_lvbs/src/mshv/vsm.rs

sangho2 · 2026-02-13T16:47:18Z

litebox_platform_lvbs/src/mshv/vsm.rs

-        }
-    }
+    let attr = HekiDataAttr::from_bytes(attr.to_le_bytes());
+    let data_type = attr.dtype_or_err().unwrap_or(HekiKdataType::Unknown);


If this is unknown data, better to return an error here.

Do we want to remove the Unknown option from the kdata enum as well? Otherwise we still have to handle it in a match somewhere

femiatl requested review from athvu, karissanchez, sangho2 and tgopinath-microsoft February 9, 2026 19:49

femiatl self-assigned this Feb 9, 2026

femiatl force-pushed the femia/lvbs/pagewalk branch 2 times, most recently from 5d880ea to 54e3a52 Compare February 11, 2026 02:45

femiatl force-pushed the femia/lvbs/pagewalk branch from 54e3a52 to d4626b2 Compare February 11, 2026 02:49

sangho2 reviewed Feb 13, 2026

View reviewed changes

		@@ -254,12 +256,14 @@ pub fn mshv_vsm_end_of_boot() -> i64 {

		/// VSM function for protecting certain memory ranges (e.g., kernel text, data, heap).
		/// `pa` and `nranges` specify a memory area containing the information about the memory ranges to protect.

Comments

Conversation

femiatl commented Feb 9, 2026

Uh oh!

sangho2 commented Feb 10, 2026

Uh oh!

github-actions bot commented Feb 11, 2026

Uh oh!

sangho2 commented Feb 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

sangho2 Feb 13, 2026

Choose a reason for hiding this comment

Uh oh!

femiatl Feb 21, 2026

Choose a reason for hiding this comment

Uh oh!

sangho2 Feb 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

femiatl Feb 13, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

sangho2 Feb 13, 2026

Choose a reason for hiding this comment

Uh oh!

femiatl Feb 19, 2026

Choose a reason for hiding this comment

Uh oh!

femiatl Feb 21, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

sangho2 commented Feb 13, 2026 •

edited

Loading

sangho2 Feb 13, 2026 •

edited

Loading