Hi,
The attached patch adjusts vec_perm_const hook to accommodate the new parameter.
For rationale, please see:
https://gcc.gnu.org/pipermail/gcc-patches/2022-May/595128.html

Unfortunately, I am not sure how to cross test this patch.
make all-gcc stage-1 seems to build fine.
Is it OK to commit the patch ?

Thanks,
Prathamesh
diff --git a/gcc/config/s390/s390.cc b/gcc/config/s390/s390.cc
index 45bbb6c3d70..e46566c526c 100644
--- a/gcc/config/s390/s390.cc
+++ b/gcc/config/s390/s390.cc
@@ -17175,9 +17175,12 @@ vectorize_vec_perm_const_1 (const struct 
expand_vec_perm_d &d)
    hook is supposed to emit the required INSNs.  */
 
 bool
-s390_vectorize_vec_perm_const (machine_mode vmode, rtx target, rtx op0, rtx 
op1,
-                              const vec_perm_indices &sel)
+s390_vectorize_vec_perm_const (machine_mode vmode, machine_mode op_mode, rtx 
target,
+                              rtx op0, rtx op1, const vec_perm_indices &sel)
 {
+  if (vmode != op_mode)
+    return false;
+
   struct expand_vec_perm_d d;
   unsigned int i, nelt;
 

Reply via email to