[naga spv-out] bounds-check runtime-sized array access correctly.

[jimblandy]

Do not neglect to apply bounds checks to indexing operations on runtime-sized arrays, even when they are accessed via an AccessIndex instruction.

Before this commit, BlockContext::write_expression_pointer would not apply bounds checks to OpAccessChain indices provided by an AccessIndex instruction, apparently with the rationale that any out-of-bounds accesses should all have been reported by constant evaluation.

While it is true that the index operand of an AccessIndex expression is known at compile time, and that the WGSL constant evaluation rules require accesses that can be statically determined to be out-of-bounds to be shader creation or pipeline creation time errors, accesses to runtime-sized arrays don't follow this pattern: even if the index is known, the length with which it must be compared is not.

Fixes #4441.

• [naga] Extend snapshot tests for bounds checks.

  Extend the snapshot tests for bounds checking to cover the case where a runtime-sized array is indexed by a constant.

• [naga spv-out] Let BoundsCheckResult supply the index value.

  Let BoundsCheckResult::Conditional provide both the condition to check before carrying out the access, and the index to use for that access. The Conditional variant indicates that we generated a runtime bounds check, which implies we must have had a SPIR-V id for the index to pass to that check, so there's no reason not to provide that to the callers - especially if the bounds check code was able to reduce it to a known constant.

  At the moment, this is not much of a refactor, but later commits will use GuardedIndex in more places, at which point this will avoid a re-matching and assertion.

• [naga spv-out] Abstract out OpAccessChain index production.

  Abstract the code from write_expression_pointer to handle one indexing operation out into its own function, BlockContext::write_access_chain_index.

• [naga spv-out] Consolidate code to find index values.

  Let the SPIR-V backend use GuardedIndex::try_resolve_to_constant, rather than writing out its definition in write_restricted_index and write_index_comparison.

  Call try_resolve_to_constant in one place, in write_bounds_check, and simply pass the GuardedIndex into subroutines.

  Reduce write_restricted_index and write_index_comparison to case analysis and code generation.

  Note that this commit does have a benign effect on SPIR-V snapshot output for programs like this:

  let one_i = 1i;
  var vec0 = vec3<i32>();
  vec0[one_i] = 1;
  

  The value indexing vec0 here is an i32, but after this commit, the operand to OpAccessChain becomes a u32 constant (with the same value).

  This is because write_bounds_check now calls try_resolve_to_constant itself, rather than deferring this work to its callees, so it may return BoundsCheckResult::KnownInBounds even when the Unchecked policy is in force. This directs the caller, write_expression_pointer, to treat the OpAccessChain operand as a fresh u32 constant, rather than simply passing through the original i32 expression.

• [naga spv-out] Abstract extending a bounds check condition chain.

  Introduce a new function, BlockContext::extend_bounds_check_condition_chain, which adds a new boolean condition to the chain of bounds checks guarding an OpAccessChain instruction.

• [naga spv-out] Doc fix: typo

• [naga spv-out] Abstract out non-uniform binding array access test.

  Introduce a new function, BlockContext::is_nonuniform_binding_array_access, which determines whether a given array access expression means that the OpAccessChain instruction must have a NonUniform decoration.

[ErichDonGubler]

I got up to (but not including) the [naga spv-out] Consolidate code to find index values. So far, the review experience has been nice!

[teoxoy]

Looks great!

[jimblandy]

Hmm, this needs to be rebased across #6188.