remove useless

Author: Valentine233

test/prototype/inductor/test_int8_sdpa_fusion.py renamed to test/prototype/inductor/test_qsdpa_fusion.py

@@ -150,14 +150,14 @@ def _check_common(
     @skipIfRocm
     @unittest.skipIf(
         not torch_version_at_least("2.7.0"),
-        reason="int8 sdpa requires torch 2.7 or later",
+        reason="qsdpa requires torch 2.7 or later",
     )
     @unittest.skipIf(
         "CPU" not in torch._C._dispatch_dump("torchao::qscaled_dot_product"),
         reason="cpp kernels not built",
     )
     @config.patch({"freezing": True})
-    def _test_sdpa_int8_rewriter(self):
+    def _test_qsdpa_rewriter(self):
         import torchao.quantization.pt2e.quantizer.x86_inductor_quantizer as xiq
         from torchao.quantization.pt2e.quantize_pt2e import convert_pt2e, prepare_pt2e
         from torchao.quantization.pt2e.quantizer.x86_inductor_quantizer import (
@@ -212,9 +212,7 @@ def _test_sdpa_int8_rewriter(self):
 
     class SDPAPatternRewriterCpuTests(TestSDPAPatternRewriterTemplate):
         device = "cpu"
-        test_sdpa_int8_rewriter_cpu = (
-            TestSDPAPatternRewriterTemplate._test_sdpa_int8_rewriter
-        )
+        test_qsdpa_rewriter_cpu = TestSDPAPatternRewriterTemplate._test_qsdpa_rewriter
 
 
 if __name__ == "__main__":

torchao/csrc/cpu/quantized_sdpa.cpp

@@ -2509,7 +2509,6 @@ at::Tensor _qscaled_dot_product_cpu(
 #ifdef CPU_CAPABILITY_AVX512
       if (at::native::cpublas::could_pack(dtype)) {
           at::Tensor output = at::empty_like(query, query.options()).transpose(1, 2);
-          std::cout << "int8_sdpa_fused_kernel" << std::endl;
           int8_sdpa_fused_kernel(output, query, key, value,
               dropout_p, is_causal, attn_mask, scale,
               q_scale, q_zp,
@@ -2520,7 +2519,6 @@ at::Tensor _qscaled_dot_product_cpu(
           return output.transpose(1, 2);
       } else {
 #endif // CPU_CAPABILITY_AVX512
-          std::cout << "int8_sdpa_math_kernel" << std::endl;
           return int8_sdpa_math_kernel(query, key, value,
               dropout_p, is_causal, attn_mask, scale,
               q_scale, q_zp,
@@ -2536,7 +2534,6 @@ at::Tensor _qscaled_dot_product_cpu(
 // CPUBLAS_BRGEMM_F8F8F32 is defined if FP8 BRGEMM is supported in PyTorch CPUBlas.
       if (at::native::cpublas::could_pack(dtype)) {
           at::Tensor output = at::empty_like(query, query.options()).transpose(1, 2);
-          std::cout << "fp8_sdpa_fused_kernel" << std::endl;
           fp8_sdpa_fused_kernel(output, query, key, value,
               dropout_p, is_causal, attn_mask, scale,
               q_scale, k_scale,
@@ -2545,7 +2542,6 @@ at::Tensor _qscaled_dot_product_cpu(
           return output.transpose(1, 2);
       } else {
 #endif // CPU_CAPABILITY_AVX512 && CPUBLAS_BRGEMM_F8F8F32
-          std::cout << "fp8_sdpa_math_kernel" << std::endl;
           return fp8_sdpa_math_kernel(query, key, value,
               dropout_p, is_causal, attn_mask, scale,
               q_scale, k_scale,