diff --git a/docs/drivers/kosmickrisp/workarounds.rst b/docs/drivers/kosmickrisp/workarounds.rst
index 8b0910cbe65..960e4a1fced 100644
--- a/docs/drivers/kosmickrisp/workarounds.rst
+++ b/docs/drivers/kosmickrisp/workarounds.rst
@@ -49,6 +49,22 @@ info on what was updated.
 Workarounds
 ===========
 
+KK_WORKAROUND_4
+---------------
+| macOS version: 26.0.1
+| Metal ticket: FB21124215 (@aitor)
+| Metal ticket status: Waiting resolution
+| CTS test failure: ``dEQP-VK.draw.renderpass.shader_invocation.helper_invocation*`` and few others
+| Comments:
+
+``simd_is_helper_thread()`` will always return true if the shader was started
+as a non-helper thread, even after ``discard_fragment()`` is called. The
+workaround is to have a variable tracking this state and update it when the
+fragment is discarded. This issue is present in M1 and M2 chips.
+
+| Log:
+| 2025-11-22: Workaround implemented and reported to Apple
+
 KK_WORKAROUND_3
 ---------------
 | macOS version: 15.4.x
diff --git a/src/kosmickrisp/compiler/nir_to_msl.c b/src/kosmickrisp/compiler/nir_to_msl.c
index 57fc1f982ad..058af458e02 100644
--- a/src/kosmickrisp/compiler/nir_to_msl.c
+++ b/src/kosmickrisp/compiler/nir_to_msl.c
@@ -118,10 +118,6 @@ emit_local_vars(struct nir_to_msl_ctx *ctx, nir_shader *shader)
          P_IND(ctx, "uchar scratch[%d] = {0};\n", shader->scratch_size);
       }
    }
-   if (BITSET_TEST(shader->info.system_values_read,
-                   SYSTEM_VALUE_HELPER_INVOCATION)) {
-      P_IND(ctx, "bool gl_HelperInvocation = simd_is_helper_thread();\n");
-   }
 }
 
 static bool
@@ -969,9 +965,6 @@ intrinsic_to_msl(struct nir_to_msl_ctx *ctx, nir_intrinsic_instr *instr)
       P(ctx, "gl_BaseInstance;\n");
       break;
    case nir_intrinsic_load_helper_invocation:
-      P(ctx, "gl_HelperInvocation;\n");
-      break;
-   case nir_intrinsic_is_helper_invocation:
       P(ctx, "simd_is_helper_thread();\n");
       break;
    case nir_intrinsic_ddx:
@@ -1860,12 +1853,19 @@ msl_preprocess_nir(struct nir_shader *nir)
 
    NIR_PASS(_, nir, nir_lower_vars_to_ssa);
    NIR_PASS(_, nir, nir_remove_dead_variables, nir_var_function_temp, NULL);
+
+   /* lower_system_values needs to go before is_helper_invocation since it will
+    * generate discards. */
+   NIR_PASS(_, nir, nir_lower_system_values);
+
    if (nir->info.stage == MESA_SHADER_FRAGMENT) {
       nir_input_attachment_options input_attachment_options = {
          .use_fragcoord_sysval = true,
          .use_layer_id_sysval = true,
       };
       NIR_PASS(_, nir, nir_lower_input_attachments, &input_attachment_options);
+      /* KK_WORKAROUND_4 */
+      NIR_PASS(_, nir, nir_lower_is_helper_invocation);
    }
    NIR_PASS(_, nir, nir_opt_combine_barriers, NULL, NULL);
    NIR_PASS(_, nir, nir_lower_var_copies);
@@ -1881,8 +1881,6 @@ msl_preprocess_nir(struct nir_shader *nir)
             glsl_get_natural_size_align_bytes,
             glsl_get_natural_size_align_bytes);
 
-   NIR_PASS(_, nir, nir_lower_system_values);
-
    nir_lower_compute_system_values_options csv_options = {
       .has_base_global_invocation_id = 0,
       .has_base_workgroup_id = true,