Bug 27225594

Change-Id: Ice38ce8ca8ef157e24d309fe6ac9ce2f69e9e816
Reviewed-on: https://swiftshader-review.googlesource.com/5363Tested-by: Nicolas Capens <capn@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

Change-Id: I47447de635055f9fadf43a71f5423e8fcba884aa
Reviewed-on: https://swiftshader-review.googlesource.com/5358Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

Change-Id: If473b7d83e54ccdca710dffe353b539213ae2046
Reviewed-on: https://swiftshader-review.googlesource.com/5357Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

Implemented switch/case for glsl in OpenGL ES 3.0.
For simplicity, it is implemented as a loop without a condition,
so break statements work properly like so:

begin switch
  if(...) // 1st case
  ...
  else if(...) // other cases
  ...
  else // default case
  ...
end switch // Anchor point for break statements

All related dEQP tests pass, except 7 tests where vertex shaders
contain a switch or a loop within another switch. These 7 failures
have only about 5% of bad pixel and seem to be related to an issue
with int(floor(...)), since the equivalent tests inside the fragment
shader pass.

KNOWN ISSUE: If a switch is within a loop and one of the cases
             contains a "continue" statement, this will not be
             handled correctly at the moment. There are no dEQP
             tests for this at the moment, AFAIK.

Change-Id: I3ba34ab06a759d07e8520f6a87d75036a5cdaef5
Reviewed-on: https://swiftshader-review.googlesource.com/5272Tested-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

Added a few types:
- Half float
- 2_10_10_10_INT
- 2_10_10_10_UINT

Related dEQP tests pass.

Change-Id: I2fb04fee89c3d69367dc2e401ac53d9758add9bf
Reviewed-on: https://swiftshader-review.googlesource.com/5356Tested-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

Change-Id: I093be383c0c5b111592acad4cc9301711c75ab88
Reviewed-on: https://swiftshader-review.googlesource.com/5362Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

Change-Id: I3aaf4819fa629282ebeb9f7e70fbae53c261c25d
Reviewed-on: https://swiftshader-review.googlesource.com/5355Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

Change-Id: I1444a3d15977c2617866f293e0ac0ccbd253d878
Reviewed-on: https://swiftshader-review.googlesource.com/5353Tested-by: Nicolas Capens <capn@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

Change-Id: Ifb474e6ab12b254599b6d1713f9f2a01b869e09d
Reviewed-on: https://swiftshader-review.googlesource.com/5361Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

Change-Id: I7c854c0c357f480b0686b0ed12e371adca8a738f
Reviewed-on: https://swiftshader-review.googlesource.com/5354Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

Change-Id: Iea408eb9beb6aa3eec4d4e64a3bd34510411dfaa
Reviewed-on: https://swiftshader-review.googlesource.com/5360Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

This fixes issues on Android where applications use these signals in their
normal course of execution. The Mono C# framework uses SIGXCPU for garbage
collection, and ART uses SIGSEGV, SIGQUIT, and SIGUSR1.

Bug 27555932

Change-Id: I4977b8f0419da660a57a8eeef20a7fe747921a63
Reviewed-on: https://swiftshader-review.googlesource.com/5345Reviewed-by: Greg Hartman <ghartman@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

Change-Id: I8158c0bb2c19d6cdee12a970f24785fae609d5fc
Reviewed-on: https://swiftshader-review.googlesource.com/5352Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

https://msdn.microsoft.com/en-us/library/windows/desktop/bb172511(v=vs.85).aspx

Change-Id: I2bc8251bce77aacb55fbad1c677f865395f5a986
Reviewed-on: https://swiftshader-review.googlesource.com/5351Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

Change-Id: I192d8fe104c3734c9c37108312f1dc93470d8d9e
Reviewed-on: https://swiftshader-review.googlesource.com/5342Reviewed-by: Shannon Woods <shannonwoods@google.com>
Tested-by: Nicolas Capens <capn@google.com>

Change-Id: I9fe9bd339f2aacb04394259ee8753dbc8e041430
Reviewed-on: https://swiftshader-review.googlesource.com/5340Reviewed-by: Shannon Woods <shannonwoods@google.com>
Tested-by: Nicolas Capens <capn@google.com>

When depth and stencil are both present, they must be the same image,
or return GL_FRAMEBUFFER_UNSUPPORTED, according to the GLES 3.0 spec,
section 4.4.4, Framebuffer Completeness.

Change-Id: I607c05617fb55743635eda987da6de523385d92b
Reviewed-on: https://swiftshader-review.googlesource.com/5262Tested-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

Change-Id: I4a7aeefedcd2d891093520d5a10ebefadcddb5be
Reviewed-on: https://swiftshader-review.googlesource.com/5320Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

Change-Id: I72439a60850add0600face13c94cc481a7083792
Reviewed-on: https://swiftshader-review.googlesource.com/5221Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

Bug 28616401

Change-Id: Ia54134e6e1dc6b101c7f81d9047e9bcc9fab3061
Reviewed-on: https://swiftshader-review.googlesource.com/5311Tested-by: Nicolas Capens <capn@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

Bug 28616401

Change-Id: I888ac3dd3186206c7745c23a542b91510a461f45
Reviewed-on: https://swiftshader-review.googlesource.com/5310Tested-by: Nicolas Capens <capn@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

Bug 28474248

Change-Id: I61bf28687623437c6b9b383dc189ffa28ad24c22
Reviewed-on: https://swiftshader-review.googlesource.com/5330Tested-by: Nicolas Capens <capn@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

Change-Id: I7f62eae32d2122ec14f400841eaefa3712cf582d
Reviewed-on: https://swiftshader-review.googlesource.com/5291Tested-by: Nicolas Capens <capn@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

The spec states that "If <sync> is not a valid sync object for <dpy>,
EGL_FALSE is returned and an EGL_BAD_PARAMETER error is generated."

Change-Id: I4bbd2ddfefd7baba9c301decb644bfb545e66f01
Reviewed-on: https://swiftshader-review.googlesource.com/5290Tested-by: Nicolas Capens <capn@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

Implemented modf as trunc + sub
Passes all related dEQP tests

Change-Id: I43656c51a670d235153e5fac390a8db311b14f8d
Reviewed-on: https://swiftshader-review.googlesource.com/5280Tested-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

Bug 28474248

Change-Id: Id5b93f3494c918cb344e5da7010dfa48a25cdde1
Reviewed-on: https://swiftshader-review.googlesource.com/5263Tested-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

This reverts commit 0637eac3.
It's considered a spec bug:
https://cvs.khronos.org/bugzilla/show_bug.cgi?id=8656

Change-Id: Id49121d5adf91ecab3d261c3c12e2f879f47e856
Reviewed-on: https://swiftshader-review.googlesource.com/5192Reviewed-by: Shannon Woods <shannonwoods@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

Change-Id: I1d1a716bd9bd5b636c5b263a7094ff13ac9aa549
Reviewed-on: https://swiftshader-review.googlesource.com/5260Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

Change-Id: Iddd9acf8496abe4dfa147f8a59b4cfb2f23df3a7
Reviewed-on: https://swiftshader-review.googlesource.com/5261Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

This cl adds a utility function that assigns the correct buffer
for each linked transform feedback variable. All the information
about location and size is sent to the VertexProcessor for use
during rendering.

Change-Id: I942805250804f56805de1fc117024c20976e83a0
Reviewed-on: https://swiftshader-review.googlesource.com/5174Tested-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

Previously we would mask instruction execution after hitting a LEAVE,
but still jump back on every loop iteration (even if infinite). This
change applies the enableLeave mask to the loop test condition so we
break out of the loop when all strands have hit LEAVE.

Change-Id: Ia331a91fb363d20743d5b87a468c3cfcaa7b481e
Reviewed-on: https://swiftshader-review.googlesource.com/5250Tested-by: Nicolas Capens <capn@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

The not equal folding logic was wrong. It would end the loop
as soon as it found equal components, which was checking if
ALL components are not equal, but that's not how not equal
works, it should return true if ANY component is not equal.

I also refactored it because it was hard to follow, with the
boolNodeFlag variable being used upside down of the result.

Change-Id: I704f26f2fd31a1cc637f5d7601409de8161c5b80
Reviewed-on: https://swiftshader-review.googlesource.com/5211Tested-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

The SRGB conversion code was already available, but wasn't used
specifically for the SRGB type framebuffers. Also, the SRGB
conversion should always be applied after blending.

According to the GLES 3.0 spec, section 4.1.8 - sRGB Conversion:
"the R, G, and B values after blending are converted
 into the non-linear sRGB color space by computing."

All related dEQP tests pass.

Change-Id: I9342d2f74aa650f28835a951bdfa8bd371bc6924
Reviewed-on: https://swiftshader-review.googlesource.com/5189Tested-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

Change-Id: I0bde678865902d549d4260bf8f6c1a368a46f4d4
Reviewed-on: https://swiftshader-review.googlesource.com/5191Tested-by: Nicolas Capens <capn@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

Change-Id: If90d45b5f0c83620486dd7b6decc2f0f5fc7e558
Reviewed-on: https://swiftshader-review.googlesource.com/5220Tested-by: Nicolas Capens <capn@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

Whenever texture projection is used with a constant texture
coordinate, folding allows to use the full precision reciprocal
and not generate the lower precision reciprocal operation.

Change-Id: I6cab6567d63ecd9abe1cedbd7e46e1fd9099a3d3
Reviewed-on: https://swiftshader-review.googlesource.com/5210Tested-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

Added constant folding for:
rad, deg, sin, cos, tan, asin, acos, atan, sinh, cosh, tanh,
asinh, acosh, atanh, log, log2, exp, exp2, sqrt, inverstsqrt

Change-Id: Ifc6b089e5981da9a3739d4ed7636953a682ce220
Reviewed-on: https://swiftshader-review.googlesource.com/5200Tested-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>

There's an LLVM bug currently that prevents us from using ICMP_EQ,
but ICMP_NE works fine, so the "equality" test has been changed to
"!inequality" to fix the issue. The fix should be reverted once
LLVM is updated to a version where the ICMP_EQ issue is fixed.

Change-Id: I79d6ca99554317cc64ffa2905ae2804bd4805e2a
Reviewed-on: https://swiftshader-review.googlesource.com/5193Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Alexis Hétu <sugoi@google.com>

Bug 27690616

Change-Id: I28b62fc56b0b572c1c9e94499d8a0c4c56737f3c
Reviewed-on: https://swiftshader-review.googlesource.com/5190Tested-by: Nicolas Capens <capn@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>

Addition evaluates before the ternary operator, which resulted in the wrong
sizes being computed.

Bug 28346104

Change-Id: Id28843132097a4fb1599eed19c05f479f16f1315
Reviewed-on: https://swiftshader-review.googlesource.com/5187Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>
Tested-by: Nicolas Capens <capn@google.com>