public inbox for gcc-cvs@sourceware.org
help / color / mirror / Atom feed
* [gcc r11-8149] aarch64: Restore bfxil optimization [PR100028]
@ 2021-04-13 10:44 Jakub Jelinek
0 siblings, 0 replies; only message in thread
From: Jakub Jelinek @ 2021-04-13 10:44 UTC (permalink / raw)
To: gcc-cvs
https://gcc.gnu.org/g:f6ba5d039f988babdd99b5cdfb4557c380e57d69
commit r11-8149-gf6ba5d039f988babdd99b5cdfb4557c380e57d69
Author: Jakub Jelinek <jakub@redhat.com>
Date: Tue Apr 13 12:43:39 2021 +0200
aarch64: Restore bfxil optimization [PR100028]
Similarly to PR87763 for bfi, the GCC 9 combiner changes to not combine
moves from hard registers regressed the following testcase where we no
longer recognize bfxil and emit 3 instructions instead.
The following patch adds define_insn patterns that match what the combiner
is trying to match in these cases. I haven't been able to see patterns
with the other order of the IOR operands, seems the IL is canonicalized this
way no matter what is written in the source.
2021-04-13 Jakub Jelinek <jakub@redhat.com>
PR target/100028
* config/aarch64/aarch64.md (*aarch64_bfxil<mode>_extr,
*aarch64_bfxilsi_extrdi): New define_insn patterns.
* gcc.target/aarch64/pr100028.c: New test.
Diff:
---
gcc/config/aarch64/aarch64.md | 32 +++++++++++++++++++++++++++++
gcc/testsuite/gcc.target/aarch64/pr100028.c | 22 ++++++++++++++++++++
2 files changed, 54 insertions(+)
diff --git a/gcc/config/aarch64/aarch64.md b/gcc/config/aarch64/aarch64.md
index a149748045f..9a7ed7866b1 100644
--- a/gcc/config/aarch64/aarch64.md
+++ b/gcc/config/aarch64/aarch64.md
@@ -5601,6 +5601,38 @@
[(set_attr "type" "bfm")]
)
+(define_insn "*aarch64_bfxil<mode>_extr"
+ [(set (match_operand:GPI 0 "register_operand" "=r")
+ (ior:GPI (and:GPI (match_operand:GPI 1 "register_operand" "0")
+ (match_operand:GPI 2 "const_int_operand" "n"))
+ (zero_extract:GPI
+ (match_operand:GPI 3 "register_operand" "r")
+ (match_operand:GPI 4 "aarch64_simd_shift_imm_<mode>" "n")
+ (match_operand:GPI 5 "aarch64_simd_shift_imm_<mode>" "n"))))]
+ "UINTVAL (operands[2]) == HOST_WIDE_INT_M1U << INTVAL (operands[4])
+ && INTVAL (operands[4])
+ && (UINTVAL (operands[4]) + UINTVAL (operands[5])
+ <= GET_MODE_BITSIZE (<MODE>mode))"
+ "bfxil\t%<GPI:w>0, %<GPI:w>3, %5, %4"
+ [(set_attr "type" "bfm")]
+)
+
+(define_insn "*aarch64_bfxilsi_extrdi"
+ [(set (match_operand:SI 0 "register_operand" "=r")
+ (ior:SI (and:SI (match_operand:SI 1 "register_operand" "0")
+ (match_operand:SI 2 "const_int_operand" "n"))
+ (match_operator:SI 6 "subreg_lowpart_operator"
+ [(zero_extract:DI
+ (match_operand:DI 3 "register_operand" "r")
+ (match_operand:SI 4 "aarch64_simd_shift_imm_si" "n")
+ (match_operand:SI 5 "aarch64_simd_shift_imm_si" "n"))])))]
+ "UINTVAL (operands[2]) == HOST_WIDE_INT_M1U << INTVAL (operands[4])
+ && INTVAL (operands[4])
+ && UINTVAL (operands[4]) + UINTVAL (operands[5]) <= 32"
+ "bfxil\t%w0, %w3, %5, %4"
+ [(set_attr "type" "bfm")]
+)
+
(define_insn "*extr_insv_lower_reg<mode>"
[(set (zero_extract:GPI (match_operand:GPI 0 "register_operand" "+r")
(match_operand 1 "const_int_operand" "n")
diff --git a/gcc/testsuite/gcc.target/aarch64/pr100028.c b/gcc/testsuite/gcc.target/aarch64/pr100028.c
new file mode 100644
index 00000000000..a25b92977b0
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/pr100028.c
@@ -0,0 +1,22 @@
+/* PR target/100028 */
+/* { dg-do compile } */
+/* { dg-options "-O2" } */
+
+#define W 3
+#define L 11
+
+int
+foo (int d, int s)
+{
+ int wmask = (1 << W) - 1;
+ return (d & ~wmask) | ((s >> L) & wmask);
+}
+
+long long int
+bar (long long int d, long long int s)
+{
+ long long int wmask = (1LL << W) - 1;
+ return (d & ~wmask) | ((s >> L) & wmask);
+}
+
+/* { dg-final { scan-assembler-times {\tbfxil\t} 2 } } */
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2021-04-13 10:44 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-04-13 10:44 [gcc r11-8149] aarch64: Restore bfxil optimization [PR100028] Jakub Jelinek
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).