nvptx: Use cvt to perform sign-extension of truncation

This patch introduces some new define_insn rules to the nvptx backend,
to perform sign-extension of a truncation (from and to the same mode),
using a single cvt instruction.  As an example, the following function

int foo(int x) { return (char)x; }

with -O2 currently generates:

	mov.u32 %r24, %ar0;
	mov.u32 %r26, %r24;
	cvt.s32.s8      %value, %r26;

and with this patch, now generates:

	mov.u32 %r24, %ar0;
	cvt.s32.s8      %value, %r24;

This patch has been tested on nvptx-none hosted by x86_64-pc-linux-gnu
with a top-level "make" (including newlib) and a "make check" with no
new regressions.

gcc/ChangeLog:

	* config/nvptx/nvptx.md (*extend_trunc_<mode>2_qi,
	*extend_trunc_<mode>2_hi, *extend_trunc_di2_si): New insns.
	Use cvt to perform sign-extension of truncation in one step.

gcc/testsuite/ChangeLog:

	* gcc.target/nvptx/exttrunc-2.c: New test case.
	* gcc.target/nvptx/exttrunc-3.c: New test case.
	* gcc.target/nvptx/exttrunc-4.c: New test case.
	* gcc.target/nvptx/exttrunc-5.c: New test case.
	* gcc.target/nvptx/exttrunc-6.c: New test case.

diff --git a/gcc/config/nvptx/nvptx.md b/gcc/config/nvptx/nvptx.md
index 108de1c0c59c7a67005e11feeca87f30387f1292..b7a03935bb2e56fb25e9ca85522caa81b52adce0 100644 (file)
--- a/gcc/config/nvptx/nvptx.md
+++ b/gcc/config/nvptx/nvptx.md
@@ -401,6 +401,32 @@
    %.\\tst%A0.u%T0\\t%0, %1;"
   [(set_attr "subregs_ok" "true")])
 
+;; Sign-extensions of truncations
+
+(define_insn "*extend_trunc_<mode>2_qi"
+  [(set (match_operand:HSDIM 0 "nvptx_register_operand" "=R")
+       (sign_extend:HSDIM
+        (truncate:QI (match_operand:HSDIM 1 "nvptx_register_operand" "R"))))]
+  ""
+  "%.\\tcvt.s%T0.s8\\t%0, %1;"
+  [(set_attr "subregs_ok" "true")])
+
+(define_insn "*extend_trunc_<mode>2_hi"
+  [(set (match_operand:SDIM 0 "nvptx_register_operand" "=R")
+       (sign_extend:SDIM
+        (truncate:HI (match_operand:SDIM 1 "nvptx_register_operand" "R"))))]
+  ""
+  "%.\\tcvt.s%T0.s16\\t%0, %1;"
+  [(set_attr "subregs_ok" "true")])
+
+(define_insn "*extend_trunc_di2_si"
+  [(set (match_operand:DI 0 "nvptx_register_operand" "=R")
+       (sign_extend:DI
+        (truncate:SI (match_operand:DI 1 "nvptx_register_operand" "R"))))]
+  ""
+  "%.\\tcvt.s64.s32\\t%0, %1;"
+  [(set_attr "subregs_ok" "true")])
+
 ;; Integer arithmetic
 
 (define_insn "add<mode>3"

diff --git a/gcc/testsuite/gcc.target/nvptx/exttrunc-2.c b/gcc/testsuite/gcc.target/nvptx/exttrunc-2.c
new file mode 100644 (file)
index 0000000..b108b55
--- /dev/null
+++ b/gcc/testsuite/gcc.target/nvptx/exttrunc-2.c
@@ -0,0 +1,17 @@
+/* { dg-do compile } */
+/* { dg-options "-O2" } */
+
+int exttrunc_si2_qi(int x)
+{
+  return (char)x;
+}
+
+/* Match:
+   mov.u32 %r24, %ar0;
+   cvt.s32.s8 %value, %r24;  */
+
+/* { dg-final { scan-assembler-times "mov\.u32\t%r\[0-9\]*, %ar0" 1 } } */
+/* { dg-final { scan-assembler-times "mov\." 1 } } */
+
+/* { dg-final { scan-assembler-times "cvt\.s32\.s8" 1 } } */
+/* { dg-final { scan-assembler-times "cvt\." 1 } } */

diff --git a/gcc/testsuite/gcc.target/nvptx/exttrunc-3.c b/gcc/testsuite/gcc.target/nvptx/exttrunc-3.c
new file mode 100644 (file)
index 0000000..69e42fa
--- /dev/null
+++ b/gcc/testsuite/gcc.target/nvptx/exttrunc-3.c
@@ -0,0 +1,17 @@
+/* { dg-do compile } */
+/* { dg-options "-O2" } */
+
+long exttrunc_di2_qi(long x)
+{
+  return (char)x;
+}
+
+/* Match:
+   mov.u64 %r24, %ar0;
+   cvt.s64.s8      %value, %r24;  */
+
+/* { dg-final { scan-assembler-times "mov\.u64\t%r\[0-9\]*, %ar0" 1 } } */
+/* { dg-final { scan-assembler-times "mov\." 1 } } */
+
+/* { dg-final { scan-assembler-times "cvt\.s64\.s8" 1 } } */
+/* { dg-final { scan-assembler-times "cvt\." 1 } } */

diff --git a/gcc/testsuite/gcc.target/nvptx/exttrunc-4.c b/gcc/testsuite/gcc.target/nvptx/exttrunc-4.c
new file mode 100644 (file)
index 0000000..09009db
--- /dev/null
+++ b/gcc/testsuite/gcc.target/nvptx/exttrunc-4.c
@@ -0,0 +1,17 @@
+/* { dg-do compile } */
+/* { dg-options "-O2" } */
+
+int exttrunc_si2_hi(int x)
+{
+  return (short)x;
+}
+
+/* Match:
+   mov.u32 %r24, %ar0;
+   cvt.s32.s16     %value, %r24;
+
+/* { dg-final { scan-assembler-times "mov\.u32\t%r\[0-9\]*, %ar0" 1 } } */
+/* { dg-final { scan-assembler-times "mov\." 1 } } */
+
+/* { dg-final { scan-assembler-times "cvt\.s32\.s16" 1 } } */
+/* { dg-final { scan-assembler-times "cvt\." 1 } } */

diff --git a/gcc/testsuite/gcc.target/nvptx/exttrunc-5.c b/gcc/testsuite/gcc.target/nvptx/exttrunc-5.c
new file mode 100644 (file)
index 0000000..5445850
--- /dev/null
+++ b/gcc/testsuite/gcc.target/nvptx/exttrunc-5.c
@@ -0,0 +1,17 @@
+/* { dg-do compile } */
+/* { dg-options "-O2" } */
+
+long exttrunc_di2_hi(long x)
+{
+  return (short)x;
+}
+
+/* Match:
+   mov.u64 %r24, %ar0;
+   cvt.s64.s16 %value, %r24;  */
+
+/* { dg-final { scan-assembler-times "mov\.u64\t%r\[0-9\]*, %ar0" 1 } } */
+/* { dg-final { scan-assembler-times "mov\." 1 } } */
+
+/* { dg-final { scan-assembler-times "cvt\.s64\.s16" 1 } } */
+/* { dg-final { scan-assembler-times "cvt\." 1 } } */

diff --git a/gcc/testsuite/gcc.target/nvptx/exttrunc-6.c b/gcc/testsuite/gcc.target/nvptx/exttrunc-6.c
new file mode 100644 (file)
index 0000000..74f050f
--- /dev/null
+++ b/gcc/testsuite/gcc.target/nvptx/exttrunc-6.c
@@ -0,0 +1,17 @@
+/* { dg-do compile } */
+/* { dg-options "-O2" } */
+
+long exttrunc_di2_si(long x)
+{
+  return (int)x;
+}
+
+/* Match:
+   mov.u64 %r24, %ar0;
+   cvt.s64.s32     %value, %r24;  */
+
+/* { dg-final { scan-assembler-times "mov\.u64\t%r\[0-9\]*, %ar0" 1 } } */
+/* { dg-final { scan-assembler-times "mov\." 1 } } */
+
+/* { dg-final { scan-assembler-times "cvt\.s64\.s32" 1 } } */
+/* { dg-final { scan-assembler-times "cvt\." 1 } } */