New target check: vect_long_mult

We don't have a 64 bit vector integer multiply on z.  Add a specific
check for that.

gcc/ChangeLog:

2017-10-04  Andreas Krebbel  <krebbel@linux.vnet.ibm.com>

	* doc/sourcebuild.texi: Document vect_long_mult.

gcc/testsuite/ChangeLog:

2017-10-04  Andreas Krebbel  <krebbel@linux.vnet.ibm.com>

	* gcc.dg/vect/pr60656.c: Check vect_long_mult.
	* lib/target-supports.exp (check_effective_target_vect_long_mult):
	New proc.

From-SVN: r253405

gcc/ChangeLog

+2017-10-04  Andreas Krebbel  <krebbel@linux.vnet.ibm.com>
+
+	* doc/sourcebuild.texi: Document vect_long_mult.
+
 2017-10-04  Richard Sandiford  <richard.sandiford@linaro.org>
 
 	PR tree-optimization/82413

gcc/doc/sourcebuild.texi

@@ -1422,6 +1422,9 @@ Target supports @code{vector short} multiplication.
 @item vect_int_mult
 Target supports @code{vector int} multiplication.
 
+@item vect_long_mult
+Target supports 64 bit @code{vector long} multiplication.
+
 @item vect_extract_even_odd
 Target supports vector even/odd element extraction.
 

gcc/testsuite/ChangeLog

 2017-10-04  Andreas Krebbel  <krebbel@linux.vnet.ibm.com>
 
+	* gcc.dg/vect/pr60656.c: Check vect_long_mult.
+	* lib/target-supports.exp (check_effective_target_vect_long_mult):
+	New proc.
+
+2017-10-04  Andreas Krebbel  <krebbel@linux.vnet.ibm.com>
+
 	* gfortran.dg/vect/fast-math-mgrid-resid.f: Use -mzarch on S/390.
 	* gfortran.dg/vect/pr77848.f: Likewise.
 

gcc/testsuite/gcc.dg/vect/pr60656.c

@@ -43,4 +43,5 @@ int main()
   return 0;
 }
 
-/* { dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect" { target vect_widen_mult_si_to_di_pattern } } } */
+/* P * P * P requires a widening multiplication first as well as a longxlong->long after that.  */
+/* { dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect" { target { vect_widen_mult_si_to_di_pattern && vect_long_mult } } } } */

gcc/testsuite/lib/target-supports.exp

@@ -6299,6 +6299,30 @@ proc check_effective_target_vect_int_mult { } {
     return $et_vect_int_mult_saved($et_index)
 }
 
+# Return 1 if the target supports 64 bit hardware vector
+# multiplication of long operands with a long result, 0 otherwise.
+#
+# This can change for different subtargets so do not cache the result.
+
+proc check_effective_target_vect_long_mult { } {
+    if { [istarget i?86-*-*] || [istarget x86_64-*-*]
+	 || (([istarget powerpc*-*-*]
+              && ![istarget powerpc-*-linux*paired*])
+              && [check_effective_target_ilp32])
+	 || [is-effective-target arm_neon]
+	 || ([istarget sparc*-*-*] && [check_effective_target_ilp32])
+	 || [istarget aarch64*-*-*]
+	 || ([istarget mips*-*-*]
+	      && [et-is-effective-target mips_msa]) } {
+	set answer 1
+    } else {
+	set answer 0
+    }
+
+    verbose "check_effective_target_vect_long_mult: returning $answer" 2
+    return $answer
+}
+
 # Return 1 if the target supports vector even/odd elements extraction, 0 otherwise.
 
 proc check_effective_target_vect_extract_even_odd { } {