Alan-Jowett
/
ubpf
/
21964759047
/
2
80%
master: 92%
master: 92%
|
Ran
|
Files
85
|
Run time
3s
|
Badge
Embed ▾
README BADGES
|
coverage: 77.23%. Remained the same
push
github
web-flowImplement MOVSX instruction support for CPU v4 (RFC 9669) (#760) * Implement MOVSX support for validation, interpreter and JIT Co-authored-by: Alan-Jowett <20480683+Alan-Jowett@users.noreply.github.com> * Fix x86_64 JIT MOVSX - force REX prefix for byte register access Co-authored-by: Alan-Jowett <20480683+Alan-Jowett@users.noreply.github.com> * Format code with clang-format Co-authored-by: Alan-Jowett <20480683+Alan-Jowett@users.noreply.github.com> * Revert "Format code with clang-format" This reverts commit 0f55b0593. * Add CPU version v4 flag for MOVSX and RFC9669 conformance tests Configure cmake/ctest to automatically run MOVSX and RFC9669 tests with --cpu_version v4 flag. This ensures these v4 instruction tests run with the correct CPU version instead of being skipped. Tests affected: - All movsx* tests (movsx832-reg, movsx1632-reg, movsx864-reg, movsx1664-reg, movsx3264-reg, rfc9669_movsx) - All rfc9669_* conformance tests (88 tests total) Co-authored-by: Alan-Jowett <20480683+Alan-Jowett@users.noreply.github.com> * Fix type conversion portability in MOVSX interpreter Add intermediate unsigned casts to make sign-extension behavior well-defined across all platforms. Converting uint64_t directly to int8_t/int16_t/int32_t is implementation-defined when the value doesn't fit. Cast through uint8_t/uint16_t/uint32_t first to ensure portable truncation before sign extension. Co-authored-by: Alan-Jowett <20480683+Alan-Jowett@users.noreply.github.com> --------- Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com> Co-authored-by: Alan-Jowett <20480683+Alan-Jowett@users.noreply.github.com>
3741 of 4844 relevant lines covered (77.23%)
374754.52 hits per line
