Bug: skia:12302
Change-Id: I7ff7bae388c5991f2c23c8945355fea55c42095a
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/447436
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
Where DSLParser is #ifdef'ed out, it had experienced some code rot and
was no longer building. This fixes it to be able to successfully build
when SKSL_DSL_PARSER is set to 1.
Change-Id: I5846b05ba3342b5b1502663577663e332c55529b
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/442976
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
These objects are ~48 bytes apiece.
(Added bsalomon@ for public API review)
Change-Id: I41207929ec2c33caaf877d8530ad7c89b24b63cb
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/436823
Commit-Queue: Brian Salomon <bsalomon@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Previously, DSLVar represented local, global, and parameter variables.
This splits it into three separate subclasses.
In addition to just being a cleaner API in general, this also addresses
an issue we ran into with the upcoming DSLParser: previously, a global
DSLVar's storage was not set correctly until DeclareGlobal was called,
so an AddToSymbolTable call prior to DeclareGlobal would create the
SkSL variable with the wrong storage, causing spurious errors on
global-only modifiers. But holding off on the AddToSymbolTable tends to
break constructs like "int x = 0, y = x", so improving the API seemed
like the best way to address it.
Now that we have greater type safety around variables, we can
potentially avoid having to call AddToSymbolTable for DSLVar and
DSLGlobalVar altogether, since we know they are both supposed to end up
in the symbol table, but that isn't something I want to change in this
CL.
Change-Id: I5f390a7384ce0af6a2131d84f97fc5e5b318063f
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/428576
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reland works around Adreno issue with this formulation of sk_Clockwise:
(sk_RTFlip.y < 0.0 ? !gl_FrontFacing : gl_FrontFacing)
and instead adds this to the top of the function:
bool sk_Clockwise = gl_FrontFacing;
if (sk_RTFlip.y < 0.0) {
sk_Clockwise = !sk_Clockwise;
}
Original description:
SkSL language features that are origin sensitive now use a uniform
to conditionally flip their result rather than generating different
code.
Previously we would insert a "rt height" uniform if sk_FragCoord needed
to be flipped. sk_FragCoord,y was implemented as "realFragCoord.y" or
"rtHeight - realFragCoord.y" depending on SkSL::ProgramSettings::fFlipY.
Now we instead use a two component vector rtFlip and sk_FragCoord.y is
always "rtFlip.x + rtFlip.y*realFragCoord.y". We configure rtFlip as
either (0, 1) or (rtHeight, -1). sk_Clockwise and dFdy simiarly use
rtFlip.y to emit code that always works with either origin.
Bug: skia:12037
Change-Id: I3a2ad6f5667eb4dcd823b939abd5698f89b58929
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/425178
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
This reverts commit 943108b0b2.
Reason for revert: clockwise GM bad on android.
Original change's description:
> Don't key progams/pipelines on origin.
>
> SkSL language features that are origin sensitive now use a uniform
> to conditionally flip their result rather than generating different
> code.
>
> Previously we would insert a "rt height" uniform if sk_FragCoord needed
> to be flipped. sk_FragCoord,y was implemented as "realFragCoord.y" or
> "rtHeight - realFragCoord.y" depending on SkSL::ProgramSettings::fFlipY.
>
> Now we instead use a two component vector rtFlip and sk_FragCoord.y is
> always "rtFlip.x + rtFlip.y*realFragCoord.y". We configure rtFlip as
> either (0, 1) or (rtHeight, -1). sk_Clockwise and dFdy simiarly use
> rtFlip.y to emit code that always works with either origin.
>
> Bug: skia:12037
>
> Change-Id: I7a09d0caac60a58d72b76645ff31bcabde4086b6
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/414796
> Commit-Queue: Brian Salomon <bsalomon@google.com>
> Reviewed-by: Greg Daniel <egdaniel@google.com>
> Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
TBR=egdaniel@google.com,bsalomon@google.com,ethannicholas@google.com
Change-Id: I91cc0d86be216f6c32e453a231de088c991be4b2
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:12037
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/425056
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
SkSL language features that are origin sensitive now use a uniform
to conditionally flip their result rather than generating different
code.
Previously we would insert a "rt height" uniform if sk_FragCoord needed
to be flipped. sk_FragCoord,y was implemented as "realFragCoord.y" or
"rtHeight - realFragCoord.y" depending on SkSL::ProgramSettings::fFlipY.
Now we instead use a two component vector rtFlip and sk_FragCoord.y is
always "rtFlip.x + rtFlip.y*realFragCoord.y". We configure rtFlip as
either (0, 1) or (rtHeight, -1). sk_Clockwise and dFdy simiarly use
rtFlip.y to emit code that always works with either origin.
Bug: skia:12037
Change-Id: I7a09d0caac60a58d72b76645ff31bcabde4086b6
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/414796
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Change-Id: I87defa005fab8b1be92a700bf38c91b92c0ab8a7
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/420616
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This unifies various chunks of error handling code and makes the DSL
more gracefully handle failed object construction.
Change-Id: I4edc581c9eba276ce2ee2a17125ea7bd6057134a
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/420237
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
The DSL would not previously function within the compiler, because it
expected to be in charge of everything itself. The compiler and DSL also
disagreed about how to handle some things, such as the DSL not
respecting the compiler's override flags.
This CL moves responsibility for much of the setup process into
DSL::Start(), which the compiler now invokes. DSL::Start() now also
takes a ProgramSettings instead of DSL-specific flags, and the
externalFunctions vector has been moved into ProgramSettings.
Change-Id: I283ed8366e25d67f02c43833743c5f8afdedaefc
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/408136
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
This will be used by the upcoming DSLParser, which needs to be able to
put DSLExpression and DSLVar into containers such as std::vector and
std::optional.
Change-Id: I8d367cfd0b3a852a368c69a5b3be6c0eaa41d74a
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/404156
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
Previously, "x = 1.0" (as opposed to 1.0f) would fail with an ambiguous
operator resolution.
Change-Id: I9bcb4115d209a2aadb3fc4c237b61c345b25ca00
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/400619
Reviewed-by: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
This gives DSLVar the ability to be set up after initial construction.
To do this, we create a plain empty Var, then swap it with the actual
Var we want afterwards. This allows DSL to support Vars which are
`uniform half4` in some cases and `const half4` (or entirely unused) in
other cases.
This technique was adapted from similar code in Ethan's parser CL.
Change-Id: Ic54d037a0102fda77b25d4755caf77a291eaa8c6
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/400716
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Prior to this change, we were creating the SkSL variable and declaration
immediately on DSLVar creation. This causes problems with function
parameters, which are sometimes supposed to be tagged
SK_MAIN_COORDS_BUILTIN. If we have already created the variable and
inserted it into the symbol table, then by the time we determine the
variable is supposed to be SK_MAIN_COORDS_BUILTIN, it is too late to
modify the (now-const) variable.
We are not yet doing this tagging, but refactoring the creation in this
fashion paves the way to making it possible.
Change-Id: I031170502c5e7c1fff5ecfac01bea470ff4e61ce
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/389216
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
Declare
This solves several issues caused by the lack of ordering guarantees in
C++; it was possible for the SkSL backend to look for the value of a
variable before its Declare() call gets processed. Moving the initial
value out of Declare should fix this whole class of problems.
Change-Id: I428fe230f1c312a0128c1f00c2a36cb95f4590a6
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/380358
Reviewed-by: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
In addition to the unsurprising changes to eliminate references to
src/, we also had to tighten up some C++17-isms as they are not
permitted in public headers.
Change-Id: Ie5005a33d7a135e69fb66beca5e7a5f960dbd453
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/378496
Reviewed-by: Brian Osman <brianosman@google.com>
