summaryrefslogtreecommitdiff
path: root/0248-Backport-SME-RISC-V-autovec-Verify-that-GET_MODE_NUN.patch
blob: df79d84ef2c506aa7adbfd02cbe90821668f1e43 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
From a3b4a0ac472415a52ce836e8997f7a69a06fad33 Mon Sep 17 00:00:00 2001
From: Michael Collison <collison@rivosinc.com>
Date: Sat, 6 May 2023 12:37:50 -0600
Subject: [PATCH 149/157] [Backport][SME] RISC-V: autovec: Verify that
 GET_MODE_NUNITS is a multiple of 2.

Reference: https://gcc.gnu.org/git/?p=gcc.git;a=commit;h=730909fa858bd691095bc23655077aa13b7941a9

While working on autovectorizing for the RISCV port I encountered an issue
where can_duplicate_and_interleave_p assumes that GET_MODE_NUNITS is a
evenly divisible by two. The RISC-V target has vector modes (e.g. VNx1DImode),
where GET_MODE_NUNITS is equal to one.

Tested on RISCV and x86_64-linux-gnu. Okay?

gcc/
	* tree-vect-slp.cc (can_duplicate_and_interleave_p):
	Check that GET_MODE_NUNITS is a multiple of 2.
---
 gcc/tree-vect-slp.cc | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/gcc/tree-vect-slp.cc b/gcc/tree-vect-slp.cc
index 6cbf8085f..d02f0ce37 100644
--- a/gcc/tree-vect-slp.cc
+++ b/gcc/tree-vect-slp.cc
@@ -401,10 +401,13 @@ can_duplicate_and_interleave_p (vec_info *vinfo, unsigned int count,
 	    (GET_MODE_BITSIZE (int_mode), 1);
 	  tree vector_type
 	    = get_vectype_for_scalar_type (vinfo, int_type, count);
+	  poly_int64 half_nelts;
 	  if (vector_type
 	      && VECTOR_MODE_P (TYPE_MODE (vector_type))
 	      && known_eq (GET_MODE_SIZE (TYPE_MODE (vector_type)),
-			   GET_MODE_SIZE (base_vector_mode)))
+			   GET_MODE_SIZE (base_vector_mode))
+	      && multiple_p (GET_MODE_NUNITS (TYPE_MODE (vector_type)),
+			     2, &half_nelts))
 	    {
 	      /* Try fusing consecutive sequences of COUNT / NVECTORS elements
 		 together into elements of type INT_TYPE and using the result
@@ -412,7 +415,7 @@ can_duplicate_and_interleave_p (vec_info *vinfo, unsigned int count,
 	      poly_uint64 nelts = GET_MODE_NUNITS (TYPE_MODE (vector_type));
 	      vec_perm_builder sel1 (nelts, 2, 3);
 	      vec_perm_builder sel2 (nelts, 2, 3);
-	      poly_int64 half_nelts = exact_div (nelts, 2);
+
 	      for (unsigned int i = 0; i < 3; ++i)
 		{
 		  sel1.quick_push (i);
-- 
2.33.0