rs6000: Add cntlzdm and cnttzdm

Add support for new scalar instructions for counting leading or
trailing zeros under control of a bitmask.

[gcc]

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

	* config/rs6000/rs6000-builtin.def (__builtin_cntlzdm): New
	built-in function definition.
	(__builtin_cnttzdm): Likewise.
	* config/rs6000/rs6000.md (UNSPEC_CNTLZDM): New constant.
	(UNSPEC_CNTTZDM): Likewise.
	(cntlzdm): New insn.
	(cnttzdm): Likewise.
	* doc/extend.texi (Basic PowerPC Built-in Functions available for
	a Future Architecture): Add descriptions of __builtin_cntlzdm and
	__builtin_cnttzdm functions.

[gcc/testsuite]

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

	* gcc.target/powerpc/cntlzdm-0.c: New test.
	* gcc.target/powerpc/cntlzdm-1.c: New test.
	* gcc.target/powerpc/cnttzdm-0.c: New test.
	* gcc.target/powerpc/cnttzdm-1.c: New test.

diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index d9f96a564af66847e45b9cd6b5ce4034a4c240d5..4dea211e5abc3bb678b397ddbd27a76f70030dae 100644 (file)
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -3,6 +3,19 @@
        PR target/95046
        * config/i386/mmx.md (sqrtv2sf2): New insn pattern.
 
+2020-05-11  Kelvin Nilsen  <kelvin@gcc.gnu.org>
+
+       * config/rs6000/rs6000-builtin.def (__builtin_cntlzdm): New
+       built-in function definition.
+       (__builtin_cnttzdm): Likewise.
+       * config/rs6000/rs6000.md (UNSPEC_CNTLZDM): New constant.
+       (UNSPEC_CNTTZDM): Likewise.
+       (cntlzdm): New insn.
+       (cnttzdm): Likewise.
+       * doc/extend.texi (Basic PowerPC Built-in Functions available for
+       a Future Architecture): Add descriptions of __builtin_cntlzdm and
+       __builtin_cnttzdm functions.
+
 2020-05-11  Kelvin Nilsen  <kelvin@gcc.gnu.org>
 
        * config/rs6000/altivec.h (vec_cfuge): New #define.

diff --git a/gcc/config/rs6000/rs6000-builtin.def b/gcc/config/rs6000/rs6000-builtin.def
index 9d80d03fe5047ae5eb30b068a670afcb857fa28b..c05d9f53d547c5c3676bb6106d8efcaac4b6ad0e 100644 (file)
--- a/gcc/config/rs6000/rs6000-builtin.def
+++ b/gcc/config/rs6000/rs6000-builtin.def
@@ -2575,6 +2575,8 @@ BU_P9_OVERLOAD_2 (CMPEQB, "byte_in_set")
 \f
 /* Future architecture scalar built-ins.  */
 BU_FUTURE_MISC_2 (CFUGED, "cfuged", CONST, cfuged)
+BU_FUTURE_MISC_2 (CNTLZDM, "cntlzdm", CONST, cntlzdm)
+BU_FUTURE_MISC_2 (CNTTZDM, "cnttzdm", CONST, cnttzdm)
 
 /* Future architecture vector built-ins.  */
 BU_FUTURE_V_2 (VCFUGED, "vcfuged", CONST, vcfuged)

diff --git a/gcc/config/rs6000/rs6000.md b/gcc/config/rs6000/rs6000.md
index 1ceb393efdeadf2965547ff5dcff86f94e0650df..6c9bae934d110126074b02eba0cfd72e75918a5c 100644 (file)
--- a/gcc/config/rs6000/rs6000.md
+++ b/gcc/config/rs6000/rs6000.md
@@ -149,6 +149,8 @@
    UNSPEC_PLTSEQ
    UNSPEC_PLT16_HA
    UNSPEC_CFUGED
+   UNSPEC_CNTLZDM
+   UNSPEC_CNTTZDM
   ])
 
 ;;
@@ -2463,6 +2465,24 @@
    "cfuged %0,%1,%2"
    [(set_attr "type" "integer")])
 
+(define_insn "cntlzdm"
+  [(set (match_operand:DI 0 "gpc_reg_operand" "=r")
+       (unspec:DI [(match_operand:DI 1 "gpc_reg_operand" "r")
+                   (match_operand:DI 2 "gpc_reg_operand" "r")]
+        UNSPEC_CNTLZDM))]
+   "TARGET_FUTURE && TARGET_POWERPC64"
+   "cntlzdm %0,%1,%2"
+   [(set_attr "type" "integer")])
+
+(define_insn "cnttzdm"
+  [(set (match_operand:DI 0 "gpc_reg_operand" "=r")
+       (unspec:DI [(match_operand:DI 1 "gpc_reg_operand" "r")
+                   (match_operand:DI 2 "gpc_reg_operand" "r")]
+        UNSPEC_CNTTZDM))]
+   "TARGET_FUTURE && TARGET_POWERPC64"
+   "cnttzdm %0,%1,%2"
+   [(set_attr "type" "integer")])
+
 (define_insn "cmpb<mode>3"
   [(set (match_operand:GPR 0 "gpc_reg_operand" "=r")
        (unspec:GPR [(match_operand:GPR 1 "gpc_reg_operand" "r")

diff --git a/gcc/doc/extend.texi b/gcc/doc/extend.texi
index f7b30c7ead744469b425f2e5960c46fdc2ecd782..8c833d39feced794bb6b270348a0ca402930deba 100644 (file)
--- a/gcc/doc/extend.texi
+++ b/gcc/doc/extend.texi
@@ -17556,6 +17556,22 @@ Perform a 64-bit centrifuge operation, as if implemented by the Future
 @code{cfuged} instruction.
 @findex __builtin_cfuged
 
+@smallexample
+@exdent unsigned long long int
+@exdent __builtin_cntlzdm (unsigned long long int, unsigned long long int)
+@end smallexample
+Perform a 64-bit count leading zeros operation under mask, as if
+implemented by the future @code{cntlzdm} instruction.
+@findex __builtin_cntlzdm
+
+@smallexample
+@exdent unsigned long long int
+@exdent __builtin_cnttzdm (unsigned long long int, unsigned long long int)
+@end smallexample
+Perform a 64-bit count trailing zeros operation under mask, as if
+implemented by the future @code{cnttzdm} instruction.
+@findex __builtin_cnttzdm
+
 @node PowerPC AltiVec/VSX Built-in Functions
 @subsection PowerPC AltiVec/VSX Built-in Functions
 

diff --git a/gcc/testsuite/ChangeLog b/gcc/testsuite/ChangeLog
index 99d06830cc14d37bdeee53e1c0afec4ef8e1112c..333b803f4446c132b90e5f1e8f0028696c26c5fe 100644 (file)
--- a/gcc/testsuite/ChangeLog
+++ b/gcc/testsuite/ChangeLog
@@ -8,6 +8,13 @@
        PR target/95046
        * gcc.target/i386/pr95046-1.c (test_sqrt): Add.
 
+2020-05-11  Kelvin Nilsen  <kelvin@gcc.gnu.org>
+
+       * gcc.target/powerpc/cntlzdm-0.c: New test.
+       * gcc.target/powerpc/cntlzdm-1.c: New test.
+       * gcc.target/powerpc/cnttzdm-0.c: New test.
+       * gcc.target/powerpc/cnttzdm-1.c: New test.
+
 2020-05-11  Kelvin Nilsen  <kelvin@gcc.gnu.org>
 
        * gcc.target/powerpc/vec-cfuged-0.c: New test.

diff --git a/gcc/testsuite/gcc.target/powerpc/cntlzdm-0.c b/gcc/testsuite/gcc.target/powerpc/cntlzdm-0.c
new file mode 100644 (file)
index 0000000..8cdd261
--- /dev/null
+++ b/gcc/testsuite/gcc.target/powerpc/cntlzdm-0.c
@@ -0,0 +1,56 @@
+/* { dg-require-effective-target powerpc64 } */
+/* { dg-options "-mdejagnu-cpu=future" } */
+
+extern void abort (void);
+
+unsigned long long int
+do_cntlzdm (unsigned long long int source, unsigned long long int mask)
+{
+  return __builtin_cntlzdm (source, mask);
+}
+
+int main (int argc, char *argv [])
+{
+  unsigned long long int sources [4], masks [4];
+  unsigned long long int intermediates [4][4] = {
+    /* sources[0] with each of masks [0 .. 3] */
+    { 0x0000a5f0ll, 0x00007e3cll, 0x000050ecll, 0x0000af73ll },
+    /* sources[1] with each of masks [0 .. 3] */
+    { 0x00007e3cll, 0x0000a5f0ll, 0x0000ec50ll, 0x000073afll },
+    /* sources[2] with each of masks [0 .. 3] */
+    { 0x00003ca5ll, 0x0000f07ell, 0x0000c50ell, 0x00003af7ll },
+    /* sources[3] with each of masks [0 .. 3] */
+    { 0x00005a0fll, 0x0000e7c3ll, 0x0000af73ll, 0x000050ecll },
+  };
+  unsigned long long int results [4][4] = {
+    { 0, 1, 1, 0 },
+    { 1, 0, 0, 1 },
+    { 2, 0, 0, 2 },
+    { 1, 0, 0, 1 },
+  };
+
+  sources[0] = 0xa5f07e3cll;
+  sources[1] = 0x7e3ca5f0ll;
+  sources[2] = 0x3ca5f07ell;
+  sources[3] = 0x5a0fe7c3ll;
+
+  masks[0] = 0xffff0000ll;
+  masks[1] = 0x0000ffffll;
+  masks[2] = 0x0f0f0f0fll;
+  masks[3] = 0xf0f0f0f0ll;
+
+  unsigned long long int result;
+
+  for (int i = 0; i < 4; i++)
+    {
+      for (int j = 0; j < 4; j++)
+       {
+         if (do_cntlzdm (sources[i], masks[j]) != results [i][j])
+           abort ();
+       }
+    }
+
+  return 0;
+}
+
+/* { dg-final { scan-assembler {\mcntlzdm\M} } } */

diff --git a/gcc/testsuite/gcc.target/powerpc/cntlzdm-1.c b/gcc/testsuite/gcc.target/powerpc/cntlzdm-1.c
new file mode 100644 (file)
index 0000000..d5a087f
--- /dev/null
+++ b/gcc/testsuite/gcc.target/powerpc/cntlzdm-1.c
@@ -0,0 +1,56 @@
+/* { dg-do run } */
+/* { dg-require-effective-target powerpc_future_hw } */
+/* { dg-require-effective-target powerpc64 } */
+/* { dg-options "-mdejagnu-cpu=future" } */
+
+extern void abort (void);
+
+unsigned long long int
+do_cntlzdm (unsigned long long int source, unsigned long long int mask)
+{
+  return __builtin_cntlzdm (source, mask);
+}
+
+int main (int argc, char *argv [])
+{
+  unsigned long long int sources [4], masks [4];
+  unsigned long long int intermediates [4][4] = {
+    /* sources[0] with each of masks [0 .. 3] */
+    { 0x0000a5f0ll, 0x00007e3cll, 0x000050ecll, 0x0000af73ll },
+    /* sources[1] with each of masks [0 .. 3] */
+    { 0x00007e3cll, 0x0000a5f0ll, 0x0000ec50ll, 0x000073afll },
+    /* sources[2] with each of masks [0 .. 3] */
+    { 0x00003ca5ll, 0x0000f07ell, 0x0000c50ell, 0x00003af7ll },
+    /* sources[3] with each of masks [0 .. 3] */
+    { 0x00005a0fll, 0x0000e7c3ll, 0x0000af73ll, 0x000050ecll },
+  };
+  unsigned long long int results [4][4] = {
+    { 0, 1, 1, 0 },
+    { 1, 0, 0, 1 },
+    { 2, 0, 0, 2 },
+    { 1, 0, 0, 1 },
+  };
+
+  sources[0] = 0xa5f07e3cll;
+  sources[1] = 0x7e3ca5f0ll;
+  sources[2] = 0x3ca5f07ell;
+  sources[3] = 0x5a0fe7c3ll;
+
+  masks[0] = 0xffff0000ll;
+  masks[1] = 0x0000ffffll;
+  masks[2] = 0x0f0f0f0fll;
+  masks[3] = 0xf0f0f0f0ll;
+
+  unsigned long long int result;
+
+  for (int i = 0; i < 4; i++)
+    {
+      for (int j = 0; j < 4; j++)
+       {
+         if (do_cntlzdm (sources[i], masks[j]) != results [i][j])
+           abort ();
+       }
+    }
+
+  return 0;
+}

diff --git a/gcc/testsuite/gcc.target/powerpc/cnttzdm-0.c b/gcc/testsuite/gcc.target/powerpc/cnttzdm-0.c
new file mode 100644 (file)
index 0000000..36acc0b
--- /dev/null
+++ b/gcc/testsuite/gcc.target/powerpc/cnttzdm-0.c
@@ -0,0 +1,52 @@
+/* { dg-options "-mdejagnu-cpu=future" } */
+
+extern void abort (void);
+
+unsigned long long int
+do_cnttzdm (unsigned long long int source, unsigned long long int mask) {
+  return __builtin_cnttzdm (source, mask);
+}
+
+int main (int argc, char *argv [])
+{
+  unsigned long long int sources [4], masks [4];
+  unsigned long long int intermediates [4][4] = {
+    /* sources[0] with each of masks [0 .. 3] */
+    { 0x0000a5f0ll, 0x00007e3cll, 0x000050ecll, 0x0000af73ll },
+    /* sources[1] with each of masks [0 .. 3] */
+    { 0x00007e3cll, 0x0000a5f0ll, 0x0000ec50ll, 0x000073afll },
+    /* sources[2] with each of masks [0 .. 3] */
+    { 0x00003ca5ll, 0x0000f07ell, 0x0000c50ell, 0x00003af7ll },
+    /* sources[3] with each of masks [0 .. 3] */
+    { 0x00005a0fll, 0x0000e7c3ll, 0x0000af73ll, 0x000050ecll },
+  };
+  unsigned long long int results [4][4] = {
+    { 4, 2, 2, 0 },
+    { 2, 4, 4, 0 },
+    { 0, 1, 1, 0 },
+    { 0, 0, 0, 2 },
+  };
+
+  sources[0] = 0xa5f07e3cll;
+  sources[1] = 0x7e3ca5f0ll;
+  sources[2] = 0x3ca5f07ell;
+  sources[3] = 0x5a0fe7c3ll;
+
+  masks[0] = 0xffff0000ll;
+  masks[1] = 0x0000ffffll;
+  masks[2] = 0x0f0f0f0fll;
+  masks[3] = 0xf0f0f0f0ll;
+
+  for (int i = 0; i < 4; i++)
+    {
+      for (int j = 0; j < 4; j++)
+       {
+         if (do_cnttzdm (sources[i], masks[j]) != results [i][j])
+           abort ();
+       }
+    }
+
+  return 0;
+}
+
+/* { dg-final { scan-assembler {\mcnttzdm\M} } } */

diff --git a/gcc/testsuite/gcc.target/powerpc/cnttzdm-1.c b/gcc/testsuite/gcc.target/powerpc/cnttzdm-1.c
new file mode 100644 (file)
index 0000000..cb11976
--- /dev/null
+++ b/gcc/testsuite/gcc.target/powerpc/cnttzdm-1.c
@@ -0,0 +1,53 @@
+/* { dg-do run } */
+/* { dg-require-effective-target powerpc_future_hw } */
+/* { dg-require-effective-target powerpc64 } */
+/* { dg-options "-mdejagnu-cpu=future" } */
+
+extern void abort (void);
+
+unsigned long long int
+do_cnttzdm (unsigned long long int source, unsigned long long int mask) {
+  return __builtin_cnttzdm (source, mask);
+}
+
+int main (int argc, char *argv [])
+{
+  unsigned long long int sources [4], masks [4];
+  unsigned long long int intermediates [4][4] = {
+    /* sources[0] with each of masks [0 .. 3] */
+    { 0x0000a5f0ll, 0x00007e3cll, 0x000050ecll, 0x0000af73ll },
+    /* sources[1] with each of masks [0 .. 3] */
+    { 0x00007e3cll, 0x0000a5f0ll, 0x0000ec50ll, 0x000073afll },
+    /* sources[2] with each of masks [0 .. 3] */
+    { 0x00003ca5ll, 0x0000f07ell, 0x0000c50ell, 0x00003af7ll },
+    /* sources[3] with each of masks [0 .. 3] */
+    { 0x00005a0fll, 0x0000e7c3ll, 0x0000af73ll, 0x000050ecll },
+  };
+  unsigned long long int results [4][4] = {
+    { 4, 2, 2, 0 },
+    { 2, 4, 4, 0 },
+    { 0, 1, 1, 0 },
+    { 0, 0, 0, 2 },
+  };
+
+  sources[0] = 0xa5f07e3cll;
+  sources[1] = 0x7e3ca5f0ll;
+  sources[2] = 0x3ca5f07ell;
+  sources[3] = 0x5a0fe7c3ll;
+
+  masks[0] = 0xffff0000ll;
+  masks[1] = 0x0000ffffll;
+  masks[2] = 0x0f0f0f0fll;
+  masks[3] = 0xf0f0f0f0ll;
+
+  for (int i = 0; i < 4; i++)
+    {
+      for (int j = 0; j < 4; j++)
+       {
+         if (do_cnttzdm (sources[i], masks[j]) != results [i][j])
+           abort ();
+       }
+    }
+
+  return 0;
+}