diff --git a/addon/aocl_gemm/aocl_gemm_bf16bf16f32of32.c b/addon/aocl_gemm/aocl_gemm_bf16bf16f32of32.c
index fe7b1d7cc..483f09e2c 100644
--- a/addon/aocl_gemm/aocl_gemm_bf16bf16f32of32.c
+++ b/addon/aocl_gemm/aocl_gemm_bf16bf16f32of32.c
@@ -117,7 +117,7 @@ AOCL_GEMM_MATMUL(bfloat16,bfloat16,float,float,bf16bf16f32of32)
 	{
 		goto err_hndl;
 	}
-	
+
 #ifdef LPGEMM_BF16_JIT
 	if( get_jit_kernels_generated() == FALSE )
 	{
@@ -176,6 +176,14 @@ AOCL_GEMM_MATMUL(bfloat16,bfloat16,float,float,bf16bf16f32of32)
 		goto err_hndl;
 	}
 
+	//For AVX-2 system support of BF16, handle the unsupported cases and return early
+	arch_t arch_id =  bli_arch_query_id();
+	if( (arch_id != BLIS_ARCH_ZEN4 ) && ( ( is_column_major == TRUE ) ||
+		( bli_is_trans(blis_transa ) ) || ( bli_is_trans(blis_transb ) ) ) )
+	{
+		bli_print_msg(" Transpose of A/B matrix or column major is not supported in AVX2.", __FILE__, __LINE__ );
+		goto err_hndl;
+	}
 	// From 5-loop function point of view
 	// B matrix needs to be packed in a certain format in order to be loaded
 	// and used in bf16 instrution. As such the mtag_b always needs to be either
@@ -226,7 +234,7 @@ AOCL_GEMM_MATMUL(bfloat16,bfloat16,float,float,bf16bf16f32of32)
 
 	lpgemm_cntx_t* lcntx_g = lpgemm_get_global_cntx_obj( BF16BF16F32OF32 );
 
-	if (( bli_arch_query_id() == BLIS_ARCH_ZEN4 ) && ( is_tiny_input_bf16of32( m, n, k, lcntx_g ) == TRUE ) &&
+	if ( ( arch_id == BLIS_ARCH_ZEN4 ) && ( is_tiny_input_bf16of32( m, n, k, lcntx_g ) == TRUE ) &&
 		 ( is_single_thread( &rntm_g ) == TRUE) &&
 	  	 ( is_row_major == TRUE ) )
 	{
diff --git a/addon/aocl_gemm/kernels/lpgemm_kernels.h b/addon/aocl_gemm/kernels/lpgemm_kernels.h
index 20d0b5894..24f145ad3 100644
--- a/addon/aocl_gemm/kernels/lpgemm_kernels.h
+++ b/addon/aocl_gemm/kernels/lpgemm_kernels.h
@@ -41,11 +41,10 @@
 // Disable BF16 kernel in cases where compilers support other avx 512
 // features except BF16 ISA.
 #if ( defined( BLIS_GCC ) && ( ( __GNUC__ < 11 ) || \
-	  ( ( __GNUC__ == 11 ) && ( __GNUC_MINOR__ < 2 ) ) ) )
-//Commenting the JIT definition, to enable the BF16 -> F32 path
-//#define LPGEMM_BF16_JIT
-//#define BPREFETCH_JIT
-//#define DUMP_JIT_CODE
+	  ( ( __GNUC__ == 11 ) && ( __GNUC_MINOR__ < 2 ) ) ) && defined(BLIS_KERNELS_ZEN4) )
+#define LPGEMM_BF16_JIT
+#define BPREFETCH_JIT
+#define DUMP_JIT_CODE
 #endif
 
 typedef void (*lpgemm_m_fringe_f32_ker_ft)