diff --git a/src/webgpu/shader/execution/builtin/abs.spec.ts b/src/webgpu/shader/execution/builtin/abs.spec.ts
index 5c7f77ed03cb..706b4855a7f9 100644
--- a/src/webgpu/shader/execution/builtin/abs.spec.ts
+++ b/src/webgpu/shader/execution/builtin/abs.spec.ts
@@ -243,7 +243,7 @@ Component-wise when T is a vector. (GLSLstd450Fabs)
         {input: NumberRepr.fromF32(kValue.negPowTwo.toMinus30), expected: [NumberRepr.fromF32(kValue.powTwo.toMinus30)] },
         {input: NumberRepr.fromF32(kValue.negPowTwo.toMinus31), expected: [NumberRepr.fromF32(kValue.powTwo.toMinus31)] },
 
-        // Powers of 2.0: -2.0^i: 1 =< i =< 31
+        // Powers of 2.0: -2.0^i: 1 <= i <= 31
         {input: NumberRepr.fromF32(kValue.negPowTwo.to1), expected: [NumberRepr.fromF32(kValue.powTwo.to1)] },
         {input: NumberRepr.fromF32(kValue.negPowTwo.to2), expected: [NumberRepr.fromF32(kValue.powTwo.to2)] },
         {input: NumberRepr.fromF32(kValue.negPowTwo.to3), expected: [NumberRepr.fromF32(kValue.powTwo.to3)] },
diff --git a/src/webgpu/shader/execution/builtin/ceil.spec.ts b/src/webgpu/shader/execution/builtin/ceil.spec.ts
new file mode 100644
index 000000000000..bcd2d948e492
--- /dev/null
+++ b/src/webgpu/shader/execution/builtin/ceil.spec.ts
@@ -0,0 +1,141 @@
+export const description = `
+Execution Tests for the 'ceil' builtin function
+`;
+
+import { makeTestGroup } from '../../../../common/framework/test_group.js';
+import { assert } from '../../../../common/util/util.js';
+import { GPUTest } from '../../../gpu_test.js';
+import { NumberRepr } from '../../../util/conversion.js';
+import { generateTypes } from '../../types.js';
+
+import { kBit, kValue, runShaderTest } from './builtin.js';
+
+export const g = makeTestGroup(GPUTest);
+
+g.test('float_builtin_functions,ceil')
+  .uniqueId('38d65728ea728bc5')
+  .specURL('https://www.w3.org/TR/2021/WD-WGSL-20210929/#float-builtin-functions')
+  .desc(
+    `
+ceil:
+T is f32 or vecN<f32> ceil(e: T ) -> T Returns the ceiling of e. Component-wise when T is a vector. (GLSLstd450Ceil)
+
+Please read the following guidelines before contributing:
+https://github.com/gpuweb/cts/blob/main/docs/plan_autogen.md
+`
+  )
+  .params(u =>
+    u
+      .combineWithParams([
+        { storageClass: 'storage', storageMode: 'read_write', access: 'read' },
+      ] as const)
+      .combine('containerType', ['scalar', 'vector'] as const)
+      .combine('isAtomic', [false])
+      .combine('baseType', ['f32'] as const)
+      .expandWithParams(generateTypes)
+  )
+  .fn(async t => {
+    assert(t.params._kTypeInfo !== undefined, 'generated type is undefined');
+    runShaderTest(
+      t,
+      t.params.storageClass,
+      t.params.storageMode,
+      t.params.type,
+      t.params._kTypeInfo.arrayLength,
+      'ceil',
+      Float32Array,
+      /* prettier-ignore */ [
+        // Small positive numbers
+        {input: NumberRepr.fromF32(0.1), expected: [NumberRepr.fromF32(1.0)] },
+        {input: NumberRepr.fromF32(0.9), expected: [NumberRepr.fromF32(1.0)] },
+        {input: NumberRepr.fromF32(1.1), expected: [NumberRepr.fromF32(2.0)] },
+        {input: NumberRepr.fromF32(1.9), expected: [NumberRepr.fromF32(2.0)] },
+
+        // Small negative numbers
+        {input: NumberRepr.fromF32(-0.1), expected: [NumberRepr.fromF32(0.0)] },
+        {input: NumberRepr.fromF32(-0.9), expected: [NumberRepr.fromF32(0.0)] },
+        {input: NumberRepr.fromF32(-1.1), expected: [NumberRepr.fromF32(-1.0)] },
+        {input: NumberRepr.fromF32(-1.9), expected: [NumberRepr.fromF32(-1.0)] },
+
+        // Min and Max f32
+        {input: NumberRepr.fromF32Bits(kBit.f32.negative.max), expected: [NumberRepr.fromF32(0.0)] },
+        {input: NumberRepr.fromF32Bits(kBit.f32.negative.min), expected: [NumberRepr.fromF32Bits(kBit.f32.negative.min)] },
+        {input: NumberRepr.fromF32Bits(kBit.f32.positive.min), expected: [NumberRepr.fromF32(1.0)] },
+        {input: NumberRepr.fromF32Bits(kBit.f32.positive.max), expected: [NumberRepr.fromF32Bits(kBit.f32.positive.max)] },
+
+        // Subnormal f32
+        {input: NumberRepr.fromF32Bits(kBit.f32.subnormal.positive.max), expected: [NumberRepr.fromF32(1.0), NumberRepr.fromF32(0.0)] },
+        {input: NumberRepr.fromF32Bits(kBit.f32.subnormal.positive.min), expected: [NumberRepr.fromF32(1.0), NumberRepr.fromF32(0.0)] },
+
+        // Infinity f32
+        {input: NumberRepr.fromF32Bits(kBit.f32.infinity.negative), expected: [NumberRepr.fromF32Bits(kBit.f32.infinity.negative)] },
+        {input: NumberRepr.fromF32Bits(kBit.f32.infinity.positive), expected: [NumberRepr.fromF32Bits(kBit.f32.infinity.positive)] },
+
+        // Powers of +2.0: 2.0^i: 1 <= i <= 31
+        {input: NumberRepr.fromF32(kValue.powTwo.to1), expected: [NumberRepr.fromF32(kValue.powTwo.to1)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to2), expected: [NumberRepr.fromF32(kValue.powTwo.to2)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to3), expected: [NumberRepr.fromF32(kValue.powTwo.to3)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to4), expected: [NumberRepr.fromF32(kValue.powTwo.to4)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to5), expected: [NumberRepr.fromF32(kValue.powTwo.to5)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to6), expected: [NumberRepr.fromF32(kValue.powTwo.to6)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to7), expected: [NumberRepr.fromF32(kValue.powTwo.to7)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to8), expected: [NumberRepr.fromF32(kValue.powTwo.to8)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to9), expected: [NumberRepr.fromF32(kValue.powTwo.to9)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to10), expected: [NumberRepr.fromF32(kValue.powTwo.to10)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to11), expected: [NumberRepr.fromF32(kValue.powTwo.to11)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to12), expected: [NumberRepr.fromF32(kValue.powTwo.to12)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to13), expected: [NumberRepr.fromF32(kValue.powTwo.to13)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to14), expected: [NumberRepr.fromF32(kValue.powTwo.to14)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to15), expected: [NumberRepr.fromF32(kValue.powTwo.to15)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to16), expected: [NumberRepr.fromF32(kValue.powTwo.to16)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to17), expected: [NumberRepr.fromF32(kValue.powTwo.to17)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to18), expected: [NumberRepr.fromF32(kValue.powTwo.to18)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to19), expected: [NumberRepr.fromF32(kValue.powTwo.to19)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to20), expected: [NumberRepr.fromF32(kValue.powTwo.to20)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to21), expected: [NumberRepr.fromF32(kValue.powTwo.to21)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to22), expected: [NumberRepr.fromF32(kValue.powTwo.to22)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to23), expected: [NumberRepr.fromF32(kValue.powTwo.to23)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to24), expected: [NumberRepr.fromF32(kValue.powTwo.to24)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to25), expected: [NumberRepr.fromF32(kValue.powTwo.to25)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to26), expected: [NumberRepr.fromF32(kValue.powTwo.to26)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to27), expected: [NumberRepr.fromF32(kValue.powTwo.to27)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to28), expected: [NumberRepr.fromF32(kValue.powTwo.to28)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to29), expected: [NumberRepr.fromF32(kValue.powTwo.to29)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to30), expected: [NumberRepr.fromF32(kValue.powTwo.to30)] },
+        {input: NumberRepr.fromF32(kValue.powTwo.to31), expected: [NumberRepr.fromF32(kValue.powTwo.to31)] },
+
+        // Powers of -2.0: -2.0^i: 1 <= i <= 31
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to1), expected: [NumberRepr.fromF32(kValue.negPowTwo.to1)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to2), expected: [NumberRepr.fromF32(kValue.negPowTwo.to2)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to3), expected: [NumberRepr.fromF32(kValue.negPowTwo.to3)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to4), expected: [NumberRepr.fromF32(kValue.negPowTwo.to4)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to5), expected: [NumberRepr.fromF32(kValue.negPowTwo.to5)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to6), expected: [NumberRepr.fromF32(kValue.negPowTwo.to6)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to7), expected: [NumberRepr.fromF32(kValue.negPowTwo.to7)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to8), expected: [NumberRepr.fromF32(kValue.negPowTwo.to8)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to9), expected: [NumberRepr.fromF32(kValue.negPowTwo.to9)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to10), expected: [NumberRepr.fromF32(kValue.negPowTwo.to10)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to11), expected: [NumberRepr.fromF32(kValue.negPowTwo.to11)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to12), expected: [NumberRepr.fromF32(kValue.negPowTwo.to12)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to13), expected: [NumberRepr.fromF32(kValue.negPowTwo.to13)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to14), expected: [NumberRepr.fromF32(kValue.negPowTwo.to14)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to15), expected: [NumberRepr.fromF32(kValue.negPowTwo.to15)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to16), expected: [NumberRepr.fromF32(kValue.negPowTwo.to16)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to17), expected: [NumberRepr.fromF32(kValue.negPowTwo.to17)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to18), expected: [NumberRepr.fromF32(kValue.negPowTwo.to18)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to19), expected: [NumberRepr.fromF32(kValue.negPowTwo.to19)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to20), expected: [NumberRepr.fromF32(kValue.negPowTwo.to20)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to21), expected: [NumberRepr.fromF32(kValue.negPowTwo.to21)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to22), expected: [NumberRepr.fromF32(kValue.negPowTwo.to22)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to23), expected: [NumberRepr.fromF32(kValue.negPowTwo.to23)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to24), expected: [NumberRepr.fromF32(kValue.negPowTwo.to24)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to25), expected: [NumberRepr.fromF32(kValue.negPowTwo.to25)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to26), expected: [NumberRepr.fromF32(kValue.negPowTwo.to26)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to27), expected: [NumberRepr.fromF32(kValue.negPowTwo.to27)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to28), expected: [NumberRepr.fromF32(kValue.negPowTwo.to28)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to29), expected: [NumberRepr.fromF32(kValue.negPowTwo.to29)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to30), expected: [NumberRepr.fromF32(kValue.negPowTwo.to30)] },
+        {input: NumberRepr.fromF32(kValue.negPowTwo.to31), expected: [NumberRepr.fromF32(kValue.negPowTwo.to31)] },
+      ]
+    );
+  });
diff --git a/src/webgpu/shader/execution/builtin/float_built_functions.spce.ts b/src/webgpu/shader/execution/builtin/float_built_functions.spce.ts
index 79fa43f65793..5e026db288ce 100644
--- a/src/webgpu/shader/execution/builtin/float_built_functions.spce.ts
+++ b/src/webgpu/shader/execution/builtin/float_built_functions.spce.ts
@@ -58,21 +58,6 @@ g.test('float_builtin_functions,atan2')
 atan2:
 T is f32 or vecN<f32> atan2(e1: T ,e2: T ) -> T Returns the arc tangent of e1 over e2. Component-wise when T is a vector. (GLSLstd450Atan2)
 
-Please read the following guidelines before contributing:
-https://github.com/gpuweb/cts/blob/main/docs/plan_autogen.md
-`
-  )
-  .params(u => u.combine('placeHolder1', ['placeHolder2', 'placeHolder3']))
-  .unimplemented();
-
-g.test('float_builtin_functions,ceil')
-  .uniqueId('38d65728ea728bc5')
-  .specURL('https://www.w3.org/TR/2021/WD-WGSL-20210929/#float-builtin-functions')
-  .desc(
-    `
-ceil:
-T is f32 or vecN<f32> ceil(e: T ) -> T Returns the ceiling of e. Component-wise when T is a vector. (GLSLstd450Ceil)
-
 Please read the following guidelines before contributing:
 https://github.com/gpuweb/cts/blob/main/docs/plan_autogen.md
 `