Skip to content

patina-v6.1.0

Choose a tag to compare

View all tags
@github-actions github-actions released this 11 Aug 21:13
· 810 commits to refs/heads/main since this release
patina-v6.1.0
37043d3
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

What's Changed

  • FixedSizeBlockAllocator: Do not panic when an allocation fails due to not having enough resources. @Javagedes (#683)
    Change Details
      ## Description
    • Impacts functionality?
    • Impacts security?
    • Breaking change?
    • Includes tests?
    • Includes documentation?

    How This Was Tested

    SCT tests previously failing as descrived in #671 and #673 no longer assert

    Integration Instructions

    N/A

      </blockquote>
  <hr>
</details>
  • debugger: Add descriptions to custom monitor commands @cfernald (#670)
    Change Details
      ## Description

    Adds descriptions to custom monitor commands and cleans up the help print for the added commands.

    Removed the version command as it is redundant with ?

    • Impacts functionality?
    • Impacts security?
    • Breaking change?
    • Includes tests?
    • Includes documentation?

    How This Was Tested

    Tested on Q35

    Integration Instructions

    N/A

      </blockquote>
  <hr>
</details>
  • AdvLogger: Remove UB and improve parsing safety @cfernald (#570)
    Change Details
      ## Description

    Undefined behavior and safety improvements:

    • Removes the undefined behavior by moving the log data into a UnsafeCell.
    • More tightly binds buffer parsing by using data slice and ZeroCopy to perform data casting.

    General Cleanup:

    • Introduce the wrapper AdvancedLog type to abstract the entire log structure and routines.
    • Simplified to use LogEntry for all intput and output of entries for AdvancedLog
    • Allow for readonly version of log for parsing

    CLOSES: #544

    • Impacts functionality?
    • Impacts security?
    • Breaking change?
    • Includes tests?
    • Includes documentation?

    How This Was Tested

    • Tested on QEMU
    • Unit tests
    • Integration Test
    • Manual parsing test

    Integration Instructions

    N/A

      </blockquote>
  <hr>
</details>
  • patina\_debugger: Improve the info (monitor ?) response @cfernald (#607)
    Change Details
      ## Description

    Adds instruction address and clarifies the exception type info. This is invoked by calling !uefiext.info or !uefiext.monitor ? in windbgx.

    • Impacts functionality?
    • Impacts security?
    • Breaking change?
    • Includes tests?
    • Includes documentation?

    How This Was Tested

    Tested on Q35

    Integration Instructions

    N/A

      </blockquote>
  <hr>
</details>
  • patina\_sdk: Add size check to page allocation's into\_raw\_ptr @cfernald (#661)
    Change Details
      ## Description

    Other routines in the Memory Manager API will return None in the event that the type of the pointer does not fit into the allocated pages. This moves that logic down to the ptr so that all returnes will do size checks. A into_raw_ptr_unchecked could be added in the future if a truely unchecked version is needed.

    This change additionally removes the try_ prefixes as the panic versions no longer exist, and this makes the interfaces more consistent.

    CLOSES: #620

    • Impacts functionality?
    • Impacts security?
    • Breaking change?
    • Includes tests?
    • Includes documentation?

    How This Was Tested

    • Q35 with patina tests
    • Unit tests

    Integration Instructions

    Update usage to remote try_ prefix and handle option from into_raw_ptr

      </blockquote>
  <hr>
</details>
  • chore: Release v6.0.4 @makubacki (#660)
    Change Details
      ## Description

    Release v6.0.4

    • Impacts functionality?
    • Impacts security?
    • Breaking change?
    • Includes tests?
    • Includes documentation?

    How This Was Tested

    N/A

    Integration Instructions

    See release notes



🚀 Features & ✨ Enhancements

  • [REBASE \& FF] Debugger: Remove the global monitor buffer @cfernald (#584)
    Change Details
      ## Description

    With the newest Windbgx (2507+), the configuration for UEFI has been updated such
    that the GDB extension will wait until a monitor command is completed with
    an OK packet to signal that all responses have been sent. This means that buffering
    the entire response into a single buffer is no longer required and the message
    may be sent over multiple GDB packets. So the monitor buffer that is pre-allocated
    is no longer required.

    Additionally, this change introduces a MonitorBuffer structure that will be inserted
    on the stack. This temporary stack is used to reduce the number of packets that will
    be sent. Format commands in rust will result in a large number of small writes as it
    formats the various types of potentially character by character. This new stack buffer
    serves as an optimization to reduce those packets.

    The second change in this PR is to remove the use of an allocated formatted string for
    the GCD monitor command.

    • Impacts functionality?
    • Impacts security?
    • Breaking change?
    • Includes tests?
    • Includes documentation?

    How This Was Tested

    Tested on Q35

    Integration Instructions

    N/A

      </blockquote>
  <hr>
</details>

🐛 Bug Fixes

  • dxe\_core: Add capabilities to EFI memory map descriptors @cfernald (#675)
    Change Details
      ## Description

    This PR changes the behavior of efi_get_memory_map to return the capabilities and not the active attributes as this is required by the UEFI specification. Without this behavior failures can occur in Windows when the boot loader attempts to find allocatable memory that it wants to use uncached.

    image
    • Impacts functionality?
    • Impacts security?
    • Breaking change?
    • Includes tests?
    • Includes documentation?

    How This Was Tested

    • Q35
    • SBSA
    • aarch64 platform

    Integration Instructions

    N/A

      </blockquote>
  <hr>
</details>
  • core\_locate\_device\_path: return InvalidParameter if device path is null @Javagedes (#668)
    Change Details
      ## Description

    If core_locate_device_path is provided a nullptr as the device path, a null pointer exception is hit in remaining_device_path function, as it's safety requirements are not upheld. This commit updates the logic in core_locate_device_path to return invalid parameter if the device path provided is a null pointer.

    • Impacts functionality?
    • Impacts security?
    • Breaking change?
    • Includes tests?
    • Includes documentation?

    How This Was Tested

    SCTs not only stopped asserting, but now pass the previously asserting test case.

    Integration Instructions

    N/A



  • dxe\_core: Point to Reallocated Table in Dbg Image Info @os-d (#664)
    Change Details
      ## Description

    Currently, the debug image info table updates the Rust slice pointer to a newly reallocated table, but does not update the UEFI defined structure to point to the new table, as a result the old table gets freed and the table in memory (discovered by the debugger) points to garbage.

    This correctly points the in memory UEFI spec defined structure to point to the new table.

    • Impacts functionality?
    • Impacts security?
    • Breaking change?
    • Includes tests?
    • Includes documentation?

    How This Was Tested

    Tested on Q35 when the SCTs caused the image count to go above the initial table size. This resolved the issue of the debug extension not being able to load image symbols.

    Integration Instructions

    N/A.

      </blockquote>
  <hr>
</details>
  • patina\_dxe\_core: Prevent calling RaiseTPL on a TPL lower than current TPL. @Javagedes (#666)
    Change Details
      ## Description

    This commit drops the allocators variable, which raises the TPL to TPL_HIGH_LEVEL throughout the lock. Dropping this early is necessary because MemoryAttributesTable::install() attempts to lock the system tables, which attempts to raise the TPL to TPL_NOTIFY.

    • Impacts functionality?
    • Impacts security?
    • Breaking change?
    • Includes tests?
    • Includes documentation?

    How This Was Tested

    Validated that previously asserting SCT tests now succeed

    Integration Instructions

    N/A

      </blockquote>
  <hr>
</details>

📖 Documentation Updates

  • RFC: Move to cargo-llvm-cov for Code Coverage @os-d (#625)
    Change Details
      ## Description

    Status: FCP

    This introduces an RFC that proposes to move to cargo-llvm-cov for code coverage. Details in the RFC.

    • Impacts functionality?
    • Impacts security?
    • Breaking change?
    • Includes tests?
    • Includes documentation?

    How This Was Tested

    N/A.

    Integration Instructions

    N/A.

      </blockquote>
  <hr>
</details>

Full Changelog: patina-v6.0.4...v6.1.0

Contributors

cfernald, makubacki, and 2 other contributors
Assets 2
Loading