rs6000: Add vector count under mask

Add support for new vclzdm and vctzdm vector instructions that
count leading and trailing zeros under control of a mask.

[gcc]

2020-05-11  Kelvin Nilsen  <kelvin@gcc.gnu.org>
	    Bill Schmidt  <wschmidt@linux.ibm.com>

	* config/rs6000/altivec.h (vec_clzm): New macro.
	(vec_ctzm): Likewise.
	* config/rs6000/altivec.md (UNSPEC_VCLZDM): New constant.
	(UNSPEC_VCTZDM): Likewise.
	(vclzdm): New insn.
	(vctzdm): Likewise.
	* config/rs6000/rs6000-builtin.def (BU_FUTURE_V_0): New macro.
	(BU_FUTURE_V_1): Likewise.
	(BU_FUTURE_V_2): Likewise.
	(BU_FUTURE_V_3): Likewise.
	(__builtin_altivec_vclzdm): New builtin definition.
	(__builtin_altivec_vctzdm): Likewise.
	* config/rs6000/rs6000-c.c (rs6000_target_modify_macros): Cause
	_ARCH_PWR_FUTURE macro to be defined if OPTION_MASK_FUTURE flag is
	set.
	* config/rs6000/rs6000-call.c (builtin_function_type): Set return
	value and parameter types to be unsigned for VCLZDM and VCTZDM.
	* config/rs6000/rs6000.c (rs6000_builtin_mask_calculate): Add
	support for TARGET_FUTURE flag.
	* config/rs6000/rs6000.h (RS6000_BTM_FUTURE): New macro constant.
	* doc/extend.texi (PowerPC Altivec Built-in Functions Available
	for a Future Architecture): New subsubsection.

[gcc/testsuite]

2020-05-11  Kelvin Nilsen  <kelvin@gcc.gnu.org>

	* gcc.target/powerpc/vec-clzm-0.c: New test.
	* gcc.target/powerpc/vec-clzm-1.c: New test.
	* gcc.target/powerpc/vec-ctzm-0.c: New test.
	* gcc.target/powerpc/vec-ctzm-1.c: New test.

5 files changed, 222 insertions, 0 deletions

diff --git a/gcc/testsuite/ChangeLog b/gcc/testsuite/ChangeLog
index a79cdb5..db274ee 100644
--- a/gcc/testsuite/ChangeLog
+++ b/gcc/testsuite/ChangeLog
@@ -25,6 +25,13 @@
 
 2020-05-11  Kelvin Nilsen  <kelvin@gcc.gnu.org>
 
+	* gcc.target/powerpc/vec-clzm-0.c: New test.
+	* gcc.target/powerpc/vec-clzm-1.c: New test.
+	* gcc.target/powerpc/vec-ctzm-0.c: New test.
+	* gcc.target/powerpc/vec-ctzm-1.c: New test.
+
+2020-05-11  Kelvin Nilsen  <kelvin@gcc.gnu.org>
+
 	* gcc.target/powerpc/dg-future-0.c: New.
 	* gcc.target/powerpc/dg-future-1.c: New.
 	* lib/target-supports.exp (check_powerpc_future_hw_available):
diff --git a/gcc/testsuite/gcc.target/powerpc/vec-clzm-0.c b/gcc/testsuite/gcc.target/powerpc/vec-clzm-0.c
new file mode 100644
index 0000000..099c5dc
--- /dev/null
+++ b/gcc/testsuite/gcc.target/powerpc/vec-clzm-0.c
@@ -0,0 +1,54 @@
+/* { dg-do compile } */
+/* { dg-options "-mdejagnu-cpu=future" } */
+
+#include <altivec.h>
+
+extern void abort (void);
+
+vector unsigned long long int
+do_vec_clzm (vector unsigned long long int source,
+	     vector unsigned long long int mask)
+{
+  return vec_clzm (source, mask);
+}
+
+int main (int argc, char *argv [])
+{
+  vector unsigned long long int source_a = { 0xa5f07e3cull, 0x7e3ca5f0ull };
+  vector unsigned long long int source_b = { 0x3ca5f07eull, 0x5a0fe7c3ull };
+
+  vector unsigned long long int mask_a = { 0xffff0000ull, 0x0000ffffull };
+  vector unsigned long long int mask_b = { 0x0f0f0f0full, 0xf0f0f0f0ull };
+
+  /* See cntlzdm-0.c for derivation of expected results.
+
+     result_aa [0] is compute (source [0], mask [0];
+     result_aa [1] is compute (source [1], mask [1].
+
+     result_ab [0] is compute (source [0], mask [2];
+     result_ab [1] is compute (source [1], mask [3].
+
+     result_ba [0] is compute (source [2], mask [0];
+     result_ba [1] is compute (source [3], mask [1].
+
+     result_bb [0] is compute (source [2], mask [2];
+     result_bb [1] is compute (source [3], mask [3].  */
+
+  vector unsigned long long int result_aa = { 0, 0 };
+  vector unsigned long long int result_ab = { 1, 1 };
+  vector unsigned long long int result_ba = { 2, 0 };
+  vector unsigned long long int result_bb = { 0, 1 };
+
+  if (!vec_all_eq (do_vec_clzm (source_a, mask_a), result_aa))
+    abort ();
+  if (!vec_all_eq (do_vec_clzm (source_a, mask_b), result_ab))
+    abort ();
+  if (!vec_all_eq (do_vec_clzm (source_b, mask_a), result_ba))
+    abort ();
+  if (!vec_all_eq (do_vec_clzm (source_b, mask_b), result_bb))
+    abort ();
+
+  return 0;
+}
+
+/* { dg-final { scan-assembler {\mvclzdm\M} } } */
diff --git a/gcc/testsuite/gcc.target/powerpc/vec-clzm-1.c b/gcc/testsuite/gcc.target/powerpc/vec-clzm-1.c
new file mode 100644
index 0000000..43b8611
--- /dev/null
+++ b/gcc/testsuite/gcc.target/powerpc/vec-clzm-1.c
@@ -0,0 +1,54 @@
+/* { dg-do run } */
+/* { dg-require-effective-target powerpc_future_hw } */
+/* { dg-options "-mdejagnu-cpu=future" } */
+
+#include <altivec.h>
+
+extern void abort (void);
+
+vector unsigned long long int
+do_vec_clzm (vector unsigned long long int source,
+	     vector unsigned long long int mask)
+{
+  return vec_clzm (source, mask);
+}
+
+int main (int argc, char *argv [])
+{
+  vector unsigned long long int source_a = { 0xa5f07e3cull, 0x7e3ca5f0ull };
+  vector unsigned long long int source_b = { 0x3ca5f07eull, 0x5a0fe7c3ull };
+
+  vector unsigned long long int mask_a = { 0xffff0000ull, 0x0000ffffull };
+  vector unsigned long long int mask_b = { 0x0f0f0f0full, 0xf0f0f0f0ull };
+
+  /* See cntlzdm-0.c for derivation of expected results.
+
+     result_aa [0] is compute (source [0], mask [0];
+     result_aa [1] is compute (source [1], mask [1].
+
+     result_ab [0] is compute (source [0], mask [2];
+     result_ab [1] is compute (source [1], mask [3].
+
+     result_ba [0] is compute (source [2], mask [0];
+     result_ba [1] is compute (source [3], mask [1].
+
+     result_bb [0] is compute (source [2], mask [2];
+     result_bb [1] is compute (source [3], mask [3].  */
+
+  vector unsigned long long int result_aa = { 0, 0 };
+  vector unsigned long long int result_ab = { 1, 1 };
+  vector unsigned long long int result_ba = { 2, 0 };
+  vector unsigned long long int result_bb = { 0, 1 };
+
+  if (!vec_all_eq (do_vec_clzm (source_a, mask_a), result_aa))
+    abort ();
+  if (!vec_all_eq (do_vec_clzm (source_a, mask_b), result_ab))
+    abort ();
+  if (!vec_all_eq (do_vec_clzm (source_b, mask_a), result_ba))
+    abort ();
+  if (!vec_all_eq (do_vec_clzm (source_b, mask_b), result_bb))
+    abort ();
+
+  return 0;
+}
+
diff --git a/gcc/testsuite/gcc.target/powerpc/vec-ctzm-0.c b/gcc/testsuite/gcc.target/powerpc/vec-ctzm-0.c
new file mode 100644
index 0000000..315edf4
--- /dev/null
+++ b/gcc/testsuite/gcc.target/powerpc/vec-ctzm-0.c
@@ -0,0 +1,54 @@
+/* { dg-do compile } */
+/* { dg-options "-mdejagnu-cpu=future" } */
+
+#include <altivec.h>
+
+extern void abort (void);
+
+vector unsigned long long int
+do_vec_ctzm (vector unsigned long long int source,
+	     vector unsigned long long int mask)
+{
+  return vec_ctzm (source, mask);
+}
+
+int main (int argc, char *argv [])
+{
+  vector unsigned long long int source_a = { 0xa5f07e3cull, 0x7e3ca5f0ull };
+  vector unsigned long long int source_b = { 0x3ca5f07eull, 0x5a0fe7c3ull };
+
+  vector unsigned long long int mask_a = { 0xffff0000ull, 0x0000ffffull };
+  vector unsigned long long int mask_b = { 0x0f0f0f0full, 0xf0f0f0f0ull };
+
+  /* See cnttzdm-0.c for derivation of expected results.
+
+     result_aa [0] is compute (source [0], mask [0];
+     result_aa [1] is compute (source [1], mask [1].
+
+     result_ab [0] is compute (source [0], mask [2];
+     result_ab [1] is compute (source [1], mask [3].
+
+     result_ba [0] is compute (source [2], mask [0];
+     result_ba [1] is compute (source [3], mask [1].
+
+     result_bb [0] is compute (source [2], mask [2];
+     result_bb [1] is compute (source [3], mask [3].  */
+
+  vector unsigned long long int result_aa = { 4, 4 };
+  vector unsigned long long int result_ab = { 2, 0 };
+  vector unsigned long long int result_ba = { 0, 0 };
+  vector unsigned long long int result_bb = { 1, 2 };
+
+  if (!vec_all_eq (do_vec_ctzm (source_a, mask_a), result_aa))
+    abort ();
+  if (!vec_all_eq (do_vec_ctzm (source_a, mask_b), result_ab))
+    abort ();
+  if (!vec_all_eq (do_vec_ctzm (source_b, mask_a),result_ba))
+    abort ();
+  if (!vec_all_eq (do_vec_ctzm (source_b, mask_b), result_bb))
+    abort ();
+
+  return 0;
+}
+
+/* { dg-final { scan-assembler {\mvctzdm\M} } } */
diff --git a/gcc/testsuite/gcc.target/powerpc/vec-ctzm-1.c b/gcc/testsuite/gcc.target/powerpc/vec-ctzm-1.c
new file mode 100644
index 0000000..3dc4a03
--- /dev/null
+++ b/gcc/testsuite/gcc.target/powerpc/vec-ctzm-1.c
@@ -0,0 +1,53 @@
+/* { dg-do compile } */
+/* { dg-require-effective-target powerpc_future_hw } */
+/* { dg-options "-mdejagnu-cpu=future" } */
+
+#include <altivec.h>
+
+extern void abort (void);
+
+vector unsigned long long int
+do_vec_ctzm (vector unsigned long long int source,
+	     vector unsigned long long int mask)
+{
+  return vec_ctzm (source, mask);
+}
+
+int main (int argc, char *argv [])
+{
+  vector unsigned long long int source_a = { 0xa5f07e3cull, 0x7e3ca5f0ull };
+  vector unsigned long long int source_b = { 0x3ca5f07eull, 0x5a0fe7c3ull };
+
+  vector unsigned long long int mask_a = { 0xffff0000ull, 0x0000ffffull };
+  vector unsigned long long int mask_b = { 0x0f0f0f0full, 0xf0f0f0f0ull };
+
+  /* See cnttzdm-0.c for derivation of expected results.
+
+     result_aa [0] is compute (source [0], mask [0];
+     result_aa [1] is compute (source [1], mask [1].
+
+     result_ab [0] is compute (source [0], mask [2];
+     result_ab [1] is compute (source [1], mask [3].
+
+     result_ba [0] is compute (source [2], mask [0];
+     result_ba [1] is compute (source [3], mask [1].
+
+     result_bb [0] is compute (source [2], mask [2];
+     result_bb [1] is compute (source [3], mask [3].  */
+
+  vector unsigned long long int result_aa = { 4, 4 };
+  vector unsigned long long int result_ab = { 2, 0 };
+  vector unsigned long long int result_ba = { 0, 0 };
+  vector unsigned long long int result_bb = { 1, 2 };
+
+  if (!vec_all_eq (do_vec_ctzm (source_a, mask_a), result_aa))
+    abort ();
+  if (!vec_all_eq (do_vec_ctzm (source_a, mask_b), result_ab))
+    abort ();
+  if (!vec_all_eq (do_vec_ctzm (source_b, mask_a),result_ba))
+    abort ();
+  if (!vec_all_eq (do_vec_ctzm (source_b, mask_b), result_bb))
+    abort ();
+
+  return 0;
+}