On Wed, 18 Oct 2023, liuhongt wrote:
> Also give up vectorization when niters_skip is negative which will be
> used for fully masked loop.
>
> Bootstrapped and regtested on x86_64-pc-linux-gnu{-m32,}.
> Ok for trunk?
>
> gcc/ChangeLog:
>
> PR tree-optimization/111820
> PR tree-optimization/111833
> * tree-vect-loop-manip.cc (vect_can_peel_nonlinear_iv_p): Give
> up vectorization for nonlinear iv vect_step_op_mul when
> step_expr is not exact_log2 and niters is greater than
> TYPE_PRECISION (TREE_TYPE (step_expr)). Also don't vectorize
> for nagative niters_skip which will be used by fully masked
> loop.
> (vect_can_advance_ivs_p): Pass whole phi_info to
> vect_can_peel_nonlinear_iv_p.
> * tree-vect-loop.cc (vect_peel_nonlinear_iv_init): Optimize
> init_expr * pow (step_expr, skipn) to init_expr
> << (log2 (step_expr) * skipn) when step_expr is exact_log2.
>
> gcc/testsuite/ChangeLog:
>
> * gcc.target/i386/pr111820-1.c: New test.
> * gcc.target/i386/pr111820-2.c: New test.
> * gcc.target/i386/pr103144-mul-1.c: Adjust testcase.
> ---
> .../gcc.target/i386/pr103144-mul-1.c | 6 ++--
> gcc/testsuite/gcc.target/i386/pr111820-1.c | 16 ++++++++++
> gcc/testsuite/gcc.target/i386/pr111820-2.c | 17 ++++++++++
> gcc/tree-vect-loop-manip.cc | 28 ++++++++++++++--
> gcc/tree-vect-loop.cc | 32 ++++++++++++++++---
> 5 files changed, 88 insertions(+), 11 deletions(-)
> create mode 100644 gcc/testsuite/gcc.target/i386/pr111820-1.c
> create mode 100644 gcc/testsuite/gcc.target/i386/pr111820-2.c
>
> diff --git a/gcc/testsuite/gcc.target/i386/pr103144-mul-1.c
> b/gcc/testsuite/gcc.target/i386/pr103144-mul-1.c
> index 640c34fd959..f80d1094097 100644
> --- a/gcc/testsuite/gcc.target/i386/pr103144-mul-1.c
> +++ b/gcc/testsuite/gcc.target/i386/pr103144-mul-1.c
> @@ -23,7 +23,7 @@ foo_mul_const (int* a)
> for (int i = 0; i != N; i++)
> {
> a[i] = b;
> - b *= 3;
> + b *= 4;
> }
> }
>
> @@ -34,7 +34,7 @@ foo_mul_peel (int* a, int b)
> for (int i = 0; i != 39; i++)
> {
> a[i] = b;
> - b *= 3;
> + b *= 4;
> }
> }
>
> @@ -46,6 +46,6 @@ foo_mul_peel_const (int* a)
> for (int i = 0; i != 39; i++)
> {
> a[i] = b;
> - b *= 3;
> + b *= 4;
> }
> }
> diff --git a/gcc/testsuite/gcc.target/i386/pr111820-1.c
> b/gcc/testsuite/gcc.target/i386/pr111820-1.c
> new file mode 100644
> index 00000000000..50e960c39d4
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/pr111820-1.c
> @@ -0,0 +1,16 @@
> +/* { dg-do compile } */
> +/* { dg-options "-O3 -mavx2 -fno-tree-vrp -Wno-aggressive-loop-optimizations
> -fdump-tree-vect-details" } */
> +/* { dg-final { scan-tree-dump "Avoid compile time hog on
> vect_peel_nonlinear_iv_init for nonlinear induction vec_step_op_mul when
> iteration count is too big" "vect" } } */
> +
> +int r;
> +int r_0;
> +
> +void f1 (void)
> +{
> + int n = 0;
> + while (-- n)
> + {
> + r_0 += r;
> + r *= 3;
> + }
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/pr111820-2.c
> b/gcc/testsuite/gcc.target/i386/pr111820-2.c
> new file mode 100644
> index 00000000000..bbdb40798c6
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/pr111820-2.c
> @@ -0,0 +1,17 @@
> +/* { dg-do compile } */
> +/* { dg-options "-O3 -mavx2 -fno-tree-vrp -fdump-tree-vect-details
> -Wno-aggressive-loop-optimizations" } */
> +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
> +
> +int r;
> +int r_0;
> +
> +void f (void)
> +{
> + int n = 0;
> + while (-- n)
> + {
> + r_0 += r ;
> + r *= 2;
> + }
> +}
> +
> diff --git a/gcc/tree-vect-loop-manip.cc b/gcc/tree-vect-loop-manip.cc
> index 2608c286e5d..a530088b61d 100644
> --- a/gcc/tree-vect-loop-manip.cc
> +++ b/gcc/tree-vect-loop-manip.cc
> @@ -1783,8 +1783,10 @@ iv_phi_p (stmt_vec_info stmt_info)
> /* Return true if vectorizer can peel for nonlinear iv. */
> static bool
> vect_can_peel_nonlinear_iv_p (loop_vec_info loop_vinfo,
> - enum vect_induction_op_type induction_type)
> + stmt_vec_info stmt_info)
> {
> + enum vect_induction_op_type induction_type
> + = STMT_VINFO_LOOP_PHI_EVOLUTION_TYPE (stmt_info);
> tree niters_skip;
> /* Init_expr will be update by vect_update_ivs_after_vectorizer,
> if niters or vf is unkown:
> @@ -1805,11 +1807,31 @@ vect_can_peel_nonlinear_iv_p (loop_vec_info
> loop_vinfo,
> return false;
> }
>
> + /* Avoid compile time hog on vect_peel_nonlinear_iv_init. */
> + if (induction_type == vect_step_op_mul)
> + {
> + tree step_expr = STMT_VINFO_LOOP_PHI_EVOLUTION_PART (stmt_info);
> + tree type = TREE_TYPE (step_expr);
> +
> + if (wi::exact_log2 (wi::to_wide (step_expr)) == -1
> + && LOOP_VINFO_INT_NITERS(loop_vinfo) >= TYPE_PRECISION (type))
> + {
> + if (dump_enabled_p ())
> + dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
> + "Avoid compile time hog on"
> + " vect_peel_nonlinear_iv_init"
> + " for nonlinear induction vec_step_op_mul"
> + " when iteration count is too big.\n");
> + return false;
> + }
> + }
> +
> /* Also doens't support peel for neg when niter is variable.
> ??? generate something like niter_expr & 1 ? init_expr : -init_expr? */
> niters_skip = LOOP_VINFO_MASK_SKIP_NITERS (loop_vinfo);
> if ((niters_skip != NULL_TREE
> - && TREE_CODE (niters_skip) != INTEGER_CST)
> + && (TREE_CODE (niters_skip) != INTEGER_CST
> + || (HOST_WIDE_INT) TREE_INT_CST_LOW (niters_skip) < 0))
So the bugs were not fixed without this hunk? IIRC in the audit
trail we concluded the value is always positive ... (but of course
a large unsigned value can appear negative if you test it this way?)
> || (!vect_use_loop_mask_for_alignment_p (loop_vinfo)
> && LOOP_VINFO_PEELING_FOR_ALIGNMENT (loop_vinfo) < 0))
> {
> @@ -1870,7 +1892,7 @@ vect_can_advance_ivs_p (loop_vec_info loop_vinfo)
> induction_type = STMT_VINFO_LOOP_PHI_EVOLUTION_TYPE (phi_info);
> if (induction_type != vect_step_op_add)
> {
> - if (!vect_can_peel_nonlinear_iv_p (loop_vinfo, induction_type))
> + if (!vect_can_peel_nonlinear_iv_p (loop_vinfo, phi_info))
> return false;
>
> continue;
> diff --git a/gcc/tree-vect-loop.cc b/gcc/tree-vect-loop.cc
> index 89bdcaa0910..6bb1f3dc462 100644
> --- a/gcc/tree-vect-loop.cc
> +++ b/gcc/tree-vect-loop.cc
> @@ -9134,11 +9134,33 @@ vect_peel_nonlinear_iv_init (gimple_seq* stmts, tree
> init_expr,
> init_expr = gimple_convert (stmts, utype, init_expr);
> unsigned skipn = TREE_INT_CST_LOW (skip_niters);
> wide_int begin = wi::to_wide (step_expr);
> - for (unsigned i = 0; i != skipn - 1; i++)
> - begin = wi::mul (begin, wi::to_wide (step_expr));
> - tree mult_expr = wide_int_to_tree (utype, begin);
> - init_expr = gimple_build (stmts, MULT_EXPR, utype, init_expr,
> mult_expr);
> - init_expr = gimple_convert (stmts, type, init_expr);
> + int pow2_step = wi::exact_log2 (begin);
> + /* Optimize init_expr * pow (step_expr, skipn) to
> + init_expr << (log2 (step_expr) * skipn). */
> + if (pow2_step != -1)
> + {
> + if (skipn >= TYPE_PRECISION (type)
> + || skipn > (UINT_MAX / (unsigned) pow2_step)
> + || skipn * (unsigned) pow2_step >= TYPE_PRECISION (type))
> + init_expr = build_zero_cst (type);
> + else
> + {
> + tree lshc = build_int_cst (utype, skipn * (unsigned) pow2_step);
> + init_expr = gimple_build (stmts, LSHIFT_EXPR, utype,
> + init_expr, lshc);
> + }
> + }
> + /* Any better way for init_expr * pow (step_expr, skipn)???. */
I think you can use one of the mpz_pow* functions and
wi::to_mpz/from_mpz for this. See tree-ssa-loop-niter.cc for the
most heavy user of mpz (but not pow I think).
Richard.
> + else
> + {
> + gcc_assert (skipn < TYPE_PRECISION (type));
> + for (unsigned i = 0; i != skipn - 1; i++)
> + begin = wi::mul (begin, wi::to_wide (step_expr));
> + tree mult_expr = wide_int_to_tree (utype, begin);
> + init_expr = gimple_build (stmts, MULT_EXPR, utype,
> + init_expr, mult_expr);
> + }
> + init_expr = gimple_convert (stmts, type, init_expr);
> }
> break;
>
>
