Skip to content

S
swiftshader
Switch branch/tag
  1. 12 May, 2021 2 commits
    • Work around MemorySanitizer scalar SSE false positives · c4d054c6
      Nicolas Capens authored 
      Scalar SSE instructions only use the lowest scalar of an SSE vector
register, but MemorySanitizer doesn't recognize some of them so it will
check the entire 128-bit operand for uninitialized bits.

This change makes sure the other elements of the vector gets zero-
initialized instead of leaving it undefined. This affects Round, Trunc,
Frac, Ceil, Floor, Sqrt, and RcpSqrt scalar operations.

Note that this workaround results in MemorySanitizer marking the entire
output vector to have a well-defined value, which isn't actually the
case in non-MSan builds. Fortunately, a scalar can't be cast into a
vector (unlike our 'emulated' small vectors), so we just have to make
sure to immediately extract the scalar from the intrinsic's result.

Bug: b/172238865
Change-Id: If68388e476ac9e27e2de33ddf2efab4124540c7a
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/54269
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Tested-by: 's avatarNicolas Capens <nicolascapens@google.com>
Reviewed-by: 's avatarAlexis Hétu <sugoi@google.com>
    • Work around MemorySanitizer movmsk false positives · b162fcf9
      Nicolas Capens authored 
      MemorySanitizer does not recognize the x86 movmskps and pmovmskb
instructions. In such cases it falls back to checking all input operands
for uninitialized bits. This causes false positives specifically with
Reactor's Byte8 type which is often loaded from memory, leaving the
upper 64-bit of an SSE register undefined.

This change masks out the unused parts of the input vector, leaving just
the sign bits.

Bug: b/172238865
Change-Id: I50c921a7ff8a4ebdba89136bb82c9f46ccdc3769
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/54268
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Tested-by: 's avatarNicolas Capens <nicolascapens@google.com>
Reviewed-by: 's avatarAlexis Hétu <sugoi@google.com>
  3. 11 May, 2021 1 commit
  5. 10 May, 2021 3 commits
  7. 07 May, 2021 1 commit
  9. 06 May, 2021 2 commits
  11. 05 May, 2021 4 commits
  13. 28 Apr, 2021 1 commit
  15. 22 Apr, 2021 1 commit
    • Format switch statements consistently · 112faf44
      Nicolas Capens authored 
      Each non-fallthrough case should end with a break (or return). If a
scope is needed because local variables are defined, it should end
before this break.

This avoids bugs such as:

switch(i)
{
case 0:
    if(c)
    {
        // lots
        // of
        // code

        // Easy to misread as always breaking instead of conditionally
        // falling through due to not spotting the if(c).
        break;
    }

case 1:
    // ...
}

The new scope should also be indented. It makes it easier to spot where
each case ends and where the switch ends. This is achieved by setting
IndentCaseBlocks to true.

Lastly, the case labels themselves should not be indented. Like goto
labels they mark where in the code to jump to, and the code itself is
already indented within the switch block.

Bug: b/144825072
Change-Id: I9a130d1d234795f53b5872e411f1315f56a0e908
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/39551
Commit-Queue: Nicolas Capens <nicolascapens@google.com>
Tested-by: 's avatarNicolas Capens <nicolascapens@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Reviewed-by: 's avatarAlexis Hétu <sugoi@google.com>
  17. 20 Apr, 2021 2 commits
  19. 16 Apr, 2021 1 commit
  21. 14 Apr, 2021 1 commit
  23. 13 Apr, 2021 2 commits
  25. 09 Apr, 2021 1 commit
  27. 29 Mar, 2021 1 commit
    • Automate installing the API library for testing · 4c0f888d
      Nicolas Capens authored 
      To run dEQP tests we previously either had to install the Vulkan Loader
and configure it to find the SwiftShader Vulkan ICD, or manually copy
over the library.

This change introduces the SWIFTSHADER_VULKAN_API_LIBRARY_INSTALL_PATH
environment variable, which can be set to a path where the API library
should be copied to. When used with dEQP it ensures we always test with
the last build, instead of the path provided in the JSON file used by
the Vulkan loader.

Note this is only meant as a convenience for SwiftShader development.
It shouldn't be used as a substitute for installing the ICD for
system-wide usage. The latter should probably use CMAKE_INSTALL_PREFIX
instead, but there are no known use cases yet for installing SwiftShader
as the system-level driver, through the CMake build.

Bug: b/149486511
Change-Id: I5ec9d669967ba0c05a763263605ee0ad5cfd505e
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/53768
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Tested-by: 's avatarNicolas Capens <nicolascapens@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Reviewed-by: 's avatarAlexis Hétu <sugoi@google.com>
  29. 26 Mar, 2021 1 commit
  31. 25 Mar, 2021 2 commits
  33. 24 Mar, 2021 1 commit
  35. 17 Mar, 2021 1 commit
  37. 16 Mar, 2021 1 commit
    • Fix multiview renderpass queries · a0aeb64e
      Nicolas Capens authored 
      The spec states that "If queries are used while executing a render pass
instance that has multiview enabled, the query uses N consecutive query
indices in the query pool (starting at query) where N is the number of
bits set in the view mask in the subpass the query is used in. How the
numerical results of the query are distributed among the queries is
implementation-dependent. For example, some implementations may write
each view’s results to a distinct query, while other implementations
may write the total result to the first query and write zero to the
other queries."

This wasn't working properly for occlusion queries, but we missed it
because dEQP doesn't provide coverage for this case if timestamp queries
are not supported. This change fixes it for both query types.

Note that for occlusion queries we write the total result to the first
query, and set the remaining query results to 0, as explicitly allowed
by the spec.

For timestamp queries each query result is a valid timestamp (note that
the spec also allows for only the first query to be a timestamp, while
the remaining ones are zero).

Bug: b/142643809
Change-Id: I531248c2822c5f68d8636d4a3d153082c48cac1d
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/53668
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Tested-by: 's avatarNicolas Capens <nicolascapens@google.com>
Reviewed-by: 's avatarAlexis Hétu <sugoi@google.com>
  39. 15 Mar, 2021 1 commit
  41. 13 Mar, 2021 1 commit
  43. 12 Mar, 2021 5 commits
  45. 11 Mar, 2021 2 commits
  47. 10 Mar, 2021 1 commit
  49. 08 Mar, 2021 1 commit