On Tue, Aug 3, 2021 at 6:56 AM H.J. Lu <hjl.to...@gmail.com> wrote:
>
> 1. Update x86 STORE_MAX_PIECES to use OImode and XImode only if inter-unit
> move is enabled since x86 uses vec_duplicate, which is enabled only when
> inter-unit move is enabled, to implement store_by_pieces.
> 2. Update op_by_pieces_d::op_by_pieces_d to set m_max_size to
> STORE_MAX_PIECES for store_by_pieces and to COMPARE_MAX_PIECES for
> compare_by_pieces.
>
> gcc/
>
> PR target/101742
> * expr.c (op_by_pieces_d::op_by_pieces_d): Set m_max_size to
> STORE_MAX_PIECES for store_by_pieces and to COMPARE_MAX_PIECES
> for compare_by_pieces.
> * config/i386/i386.h (STORE_MAX_PIECES): Use OImode and XImode
> only if TARGET_INTER_UNIT_MOVES_TO_VEC is true.
>
> gcc/testsuite/
>
> PR target/101742
> * gcc.target/i386/pr101742a.c: New test.
> * gcc.target/i386/pr101742b.c: Likewise.
> ---
> gcc/config/i386/i386.h | 20 +++++++++++---------
> gcc/expr.c | 6 +++++-
> gcc/testsuite/gcc.target/i386/pr101742a.c | 16 ++++++++++++++++
> gcc/testsuite/gcc.target/i386/pr101742b.c | 4 ++++
> 4 files changed, 36 insertions(+), 10 deletions(-)
> create mode 100644 gcc/testsuite/gcc.target/i386/pr101742a.c
> create mode 100644 gcc/testsuite/gcc.target/i386/pr101742b.c
>
> diff --git a/gcc/config/i386/i386.h b/gcc/config/i386/i386.h
> index bed9cd9da18..9b416abd5f4 100644
> --- a/gcc/config/i386/i386.h
> +++ b/gcc/config/i386/i386.h
> @@ -1783,15 +1783,17 @@ typedef struct ix86_args {
> /* STORE_MAX_PIECES is the number of bytes at a time that we can
> store efficiently. */
> #define STORE_MAX_PIECES \
> - ((TARGET_AVX512F && !TARGET_PREFER_AVX256) \
> - ? 64 \
> - : ((TARGET_AVX \
> - && !TARGET_PREFER_AVX128 \
> - && !TARGET_AVX256_SPLIT_UNALIGNED_STORE) \
> - ? 32 \
> - : ((TARGET_SSE2 \
> - && TARGET_SSE_UNALIGNED_STORE_OPTIMAL) \
> - ? 16 : UNITS_PER_WORD)))
> + (TARGET_INTER_UNIT_MOVES_TO_VEC \
> + ? ((TARGET_AVX512F && !TARGET_PREFER_AVX256) \
> + ? 64 \
> + : ((TARGET_AVX \
> + && !TARGET_PREFER_AVX128 \
> + && !TARGET_AVX256_SPLIT_UNALIGNED_STORE) \
> + ? 32 \
> + : ((TARGET_SSE2 \
> + && TARGET_SSE_UNALIGNED_STORE_OPTIMAL) \
> + ? 16 : UNITS_PER_WORD))) \
> + : UNITS_PER_WORD)
>
> /* If a memory-to-memory move would take MOVE_RATIO or more simple
> move-instruction pairs, we will do a cpymem or libcall instead.
> diff --git a/gcc/expr.c b/gcc/expr.c
> index b65cfcfdcd1..2964b38b9a5 100644
> --- a/gcc/expr.c
> +++ b/gcc/expr.c
> @@ -1131,7 +1131,11 @@ op_by_pieces_d::op_by_pieces_d (rtx to, bool to_load,
> bool qi_vector_mode)
> : m_to (to, to_load, NULL, NULL),
> m_from (from, from_load, from_cfn, from_cfn_data),
> - m_len (len), m_max_size (MOVE_MAX_PIECES + 1),
> + m_len (len),
> + m_max_size (((!to_load && from == nullptr)
> + ? STORE_MAX_PIECES
> + : (from_cfn != nullptr
> + ? COMPARE_MAX_PIECES : MOVE_MAX_PIECES)) + 1),
> m_push (push), m_qi_vector_mode (qi_vector_mode)
> {
> int toi = m_to.get_addr_inc ();
This larger expr.c patch passes the proper MAX_PIECES directly.
> diff --git a/gcc/testsuite/gcc.target/i386/pr101742a.c
> b/gcc/testsuite/gcc.target/i386/pr101742a.c
> new file mode 100644
> index 00000000000..67ea40587dd
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/pr101742a.c
> @@ -0,0 +1,16 @@
> +/* { dg-do compile } */
> +/* { dg-additional-options "-O3 -mtune=nano-x2" } */
> +
> +int n2;
> +
> +__attribute__ ((simd)) char
> +w7 (void)
> +{
> + short int xb = n2;
> + int qp;
> +
> + for (qp = 0; qp < 2; ++qp)
> + xb = xb < 1;
> +
> + return xb;
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/pr101742b.c
> b/gcc/testsuite/gcc.target/i386/pr101742b.c
> new file mode 100644
> index 00000000000..ba19064077b
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/pr101742b.c
> @@ -0,0 +1,4 @@
> +/* { dg-do compile } */
> +/* { dg-additional-options "-O3 -mtune=nano-x2
> -mtune-ctrl=sse_unaligned_store_optimal" } */
> +
> +#include "pr101742a.c"
> --
> 2.31.1
>
--
H.J.
diff --git a/gcc/expr.c b/gcc/expr.c
index b65cfcfdcd1..66ac1986f02 100644
--- a/gcc/expr.c
+++ b/gcc/expr.c
@@ -1110,8 +1110,8 @@ class op_by_pieces_d
}
public:
- op_by_pieces_d (rtx, bool, rtx, bool, by_pieces_constfn, void *,
- unsigned HOST_WIDE_INT, unsigned int, bool,
+ op_by_pieces_d (unsigned int, rtx, bool, rtx, bool, by_pieces_constfn,
+ void *, unsigned HOST_WIDE_INT, unsigned int, bool,
bool = false);
void run ();
};
@@ -1122,8 +1122,8 @@ class op_by_pieces_d
and its associated FROM_CFN_DATA can be used to replace loads with
constant values. LEN describes the length of the operation. */
-op_by_pieces_d::op_by_pieces_d (rtx to, bool to_load,
- rtx from, bool from_load,
+op_by_pieces_d::op_by_pieces_d (unsigned int max_pieces, rtx to,
+ bool to_load, rtx from, bool from_load,
by_pieces_constfn from_cfn,
void *from_cfn_data,
unsigned HOST_WIDE_INT len,
@@ -1131,7 +1131,7 @@ op_by_pieces_d::op_by_pieces_d (rtx to, bool to_load,
bool qi_vector_mode)
: m_to (to, to_load, NULL, NULL),
m_from (from, from_load, from_cfn, from_cfn_data),
- m_len (len), m_max_size (MOVE_MAX_PIECES + 1),
+ m_len (len), m_max_size (max_pieces + 1),
m_push (push), m_qi_vector_mode (qi_vector_mode)
{
int toi = m_to.get_addr_inc ();
@@ -1324,8 +1324,8 @@ class move_by_pieces_d : public op_by_pieces_d
public:
move_by_pieces_d (rtx to, rtx from, unsigned HOST_WIDE_INT len,
unsigned int align)
- : op_by_pieces_d (to, false, from, true, NULL, NULL, len, align,
- PUSHG_P (to))
+ : op_by_pieces_d (MOVE_MAX_PIECES, to, false, from, true, NULL,
+ NULL, len, align, PUSHG_P (to))
{
}
rtx finish_retmode (memop_ret);
@@ -1421,8 +1421,8 @@ class store_by_pieces_d : public op_by_pieces_d
store_by_pieces_d (rtx to, by_pieces_constfn cfn, void *cfn_data,
unsigned HOST_WIDE_INT len, unsigned int align,
bool qi_vector_mode)
- : op_by_pieces_d (to, false, NULL_RTX, true, cfn, cfn_data, len,
- align, false, qi_vector_mode)
+ : op_by_pieces_d (STORE_MAX_PIECES, to, false, NULL_RTX, true, cfn,
+ cfn_data, len, align, false, qi_vector_mode)
{
}
rtx finish_retmode (memop_ret);
@@ -1618,8 +1618,8 @@ class compare_by_pieces_d : public op_by_pieces_d
compare_by_pieces_d (rtx op0, rtx op1, by_pieces_constfn op1_cfn,
void *op1_cfn_data, HOST_WIDE_INT len, int align,
rtx_code_label *fail_label)
- : op_by_pieces_d (op0, true, op1, true, op1_cfn, op1_cfn_data, len,
- align, false)
+ : op_by_pieces_d (COMPARE_MAX_PIECES, op0, true, op1, true, op1_cfn,
+ op1_cfn_data, len, align, false)
{
m_fail_label = fail_label;
}
