diff --git a/src/webgpu/api/validation/encoding/programmable/pipeline_immediate.spec.ts b/src/webgpu/api/validation/encoding/programmable/pipeline_immediate.spec.ts
index b3ff821ea7f2..b5ba96b25c8a 100644
--- a/src/webgpu/api/validation/encoding/programmable/pipeline_immediate.spec.ts
+++ b/src/webgpu/api/validation/encoding/programmable/pipeline_immediate.spec.ts
@@ -99,7 +99,8 @@ g.test('required_slots_set')
     const { encoderType, scenario, usage, stage } = t.params;
 
     let code = '';
-    let kRequiredSize = 0;
+    let layoutImmediateSize = 0;
+    let trailingPaddingBytes = 0;
 
     const use_vertex = stage === 'vertex' || stage === 'both';
     const use_fragment = stage === 'fragment' || stage === 'both';
@@ -117,17 +118,18 @@ g.test('required_slots_set')
 
     switch (scenario) {
       case 'scalar':
-        kRequiredSize = 4;
+        layoutImmediateSize = 4;
         declarations = 'var<immediate> data: u32;';
         helpers = 'fn use_data() { _ = data; }';
         break;
       case 'vector':
-        kRequiredSize = 16;
+        layoutImmediateSize = 16;
         declarations = 'var<immediate> data: vec4<u32>;';
         helpers = 'fn use_data() { _ = data; }';
         break;
       case 'struct_padding':
-        kRequiredSize = 40;
+        layoutImmediateSize = 64;
+        trailingPaddingBytes = 24;
         declarations = `
           struct A { v: vec3<u32> }
           struct B { v: vec2<u32> }
@@ -141,7 +143,7 @@ g.test('required_slots_set')
         callFragment = 'use_b();';
         break;
       case 'mixed_types':
-        kRequiredSize = 32;
+        layoutImmediateSize = 32;
         declarations = `
           struct Mixed { v: u32, f: vec4<u32> }
           var<immediate> data: Mixed;`;
@@ -153,7 +155,7 @@ g.test('required_slots_set')
         callFragment = 'use_f();';
         break;
       case 'dynamic_indexing':
-        kRequiredSize = 16;
+        layoutImmediateSize = 16;
         declarations = 'var<immediate> data: array<u32, 4>;';
         helpers = 'fn use_data(i: u32) { _ = data[i]; }';
         computeArgs = '@builtin(local_invocation_index) i: u32';
@@ -165,7 +167,7 @@ g.test('required_slots_set')
         callFragment = 'use_data(i);';
         break;
       case 'multiple_variables':
-        kRequiredSize = 32;
+        layoutImmediateSize = 32;
         declarations = `
           struct S1 { a: u32, x: u32 }
           struct S2 { a: u32, y: vec4<u32> }
@@ -197,7 +199,14 @@ g.test('required_slots_set')
       }
     `;
 
-    const layoutSize = usage === 'overprovision' ? kRequiredSize + 4 : kRequiredSize;
+    const kRequiredSize = layoutImmediateSize - trailingPaddingBytes;
+    // When overprovisioning: if the struct already has trailing padding, the layout
+    // size is already larger than what the shader uses, so no extra bytes are needed.
+    // Only when there's no trailing padding do we need to increase the layout size.
+    const layoutSize =
+      usage === 'overprovision' && trailingPaddingBytes === 0
+        ? layoutImmediateSize + 4
+        : layoutImmediateSize;
     if (layoutSize > t.device.limits.maxImmediateSize!) {
       t.skip('maxImmediateSize not large enough for overprovision test');
     }
@@ -210,7 +219,10 @@ g.test('required_slots_set')
     };
 
     if (usage === 'overprovision') {
-      setImmediates(0, kRequiredSize + 4);
+      // When trailingPaddingBytes > 0, the layout already overprovisions beyond
+      // shader usage, so set the full layoutImmediateSize. When trailingPaddingBytes
+      // is 0, we added +4 to the layout, so set kRequiredSize + 4.
+      setImmediates(0, trailingPaddingBytes > 0 ? layoutImmediateSize : kRequiredSize + 4);
     } else if (usage === 'full') {
       setImmediates(0, kRequiredSize);
     } else if (usage === 'partial') {