> -----Original Message-----
> From: Richard Biener <[email protected]>
> Sent: Thursday, November 21, 2024 8:03 AM
> To: Tamar Christina <[email protected]>
> Cc: [email protected]; nd <[email protected]>
> Subject: Re: [PATCH]middle-end: Pass along SLP node when costing vector
> loads/stores
>
> On Wed, 20 Nov 2024, Tamar Christina wrote:
>
> > Hi All,
> >
> > With the support to SLP only we now pass the VMAT through the SLP node,
> however
> > the majority of the costing calls inside vectorizable_load and
> > vectorizable_store do no pass the SLP node along. Due to this the backend
> costing
> > never sees the VMAT for these cases anymore.
> >
> > Additionally the helper around record_stmt_cost when both SLP and stmt_vinfo
> are
> > passed would only pass the SLP node along. However the SLP node doesn't
> contain
> > all the info available in the stmt_vinfo and we'd have to go through the
> > SLP_TREE_REPRESENTATIVE anyway. As such I changed the function to just
> Always
> > pass both along. Unlike the VMAT changes, I don't believe there to be a
> > correctness issue here but would minimize the number of churn in the backend
> > costing until vectorizer costing as a whole is revisited in GCC 16.
>
> I agree this is the best way forward at this point - I originally
> thought to never pass both and treat the calls with SLP node as being
> the "future" way, but clearly we're not even close to a "future" costing
> API right now.
After trunk stabilizes a bit I plan to look at the other functions and move the
AArch64
backend to use only SLP_tree. One thing I was wondering about is what to do
with
costing that happens early during say DR analysis where we don't have an SLP
tree.
So I guess the ABI must always have the duality of getting both the stmt_vinfo
and
SLP tree but we could go back to your original design of not setting stmt_vinfo
if
An SLP tree is given which would distinguish between the two?
Thanks,
Tamar
>
> > These changes re-enable the cost model on AArch64 and also correctly find
> > the
> > VMATs on loads and stores fixing testcases such as sve_iters_low_2.c.
> >
> > Bootstrapped Regtested on aarch64-none-linux-gnu,
> > arm-none-linux-gnueabihf, x86_64-pc-linux-gnu -m32, -m64 and no issues.
> >
> > Ok for master?
>
> OK.
>
> Thanks,
> Richard.
>
> > Thanks,
> > Tamar
> >
> > gcc/ChangeLog:
> >
> > * tree-vect-data-refs.cc (vect_get_data_access_cost): Pass NULL for SLP
> > node.
> > * tree-vect-stmts.cc (record_stmt_cost): Expose.
> > (vect_get_store_cost, vect_get_load_cost): Extend with SLP node.
> > (vectorizable_store, vectorizable_load): Pass SLP node to all costing.
> > * tree-vectorizer.h (record_stmt_cost): Always pass both SLP node and
> > stmt_vinfo to costing.
> > (vect_get_load_cost, vect_get_store_cost): Extend with SLP node.
> >
> > ---
> > diff --git a/gcc/tree-vect-data-refs.cc b/gcc/tree-vect-data-refs.cc
> > index
> 3ea5fb883b1a5289195142171eb45fa422910a95..d87ca79b8e4c16d242e6743
> 1d1b527bdb8cb74e4 100644
> > --- a/gcc/tree-vect-data-refs.cc
> > +++ b/gcc/tree-vect-data-refs.cc
> > @@ -1729,12 +1729,14 @@ vect_get_data_access_cost (vec_info *vinfo,
> dr_vec_info *dr_info,
> > ncopies = vect_get_num_copies (loop_vinfo, STMT_VINFO_VECTYPE
> (stmt_info));
> >
> > if (DR_IS_READ (dr_info->dr))
> > - vect_get_load_cost (vinfo, stmt_info, ncopies,
> > alignment_support_scheme,
> > - misalignment, true, inside_cost,
> > - outside_cost, prologue_cost_vec, body_cost_vec, false);
> > + vect_get_load_cost (vinfo, stmt_info, NULL, ncopies,
> > + alignment_support_scheme, misalignment, true,
> > + inside_cost, outside_cost, prologue_cost_vec,
> > + body_cost_vec, false);
> > else
> > - vect_get_store_cost (vinfo,stmt_info, ncopies,
> > alignment_support_scheme,
> > - misalignment, inside_cost, body_cost_vec);
> > + vect_get_store_cost (vinfo,stmt_info, NULL, ncopies,
> > + alignment_support_scheme, misalignment, inside_cost,
> > + body_cost_vec);
> >
> > if (dump_enabled_p ())
> > dump_printf_loc (MSG_NOTE, vect_location,
> > diff --git a/gcc/tree-vect-stmts.cc b/gcc/tree-vect-stmts.cc
> > index
> 7a92da00f7ddcfdf146fa1c2511f609e8bc40e9e..46543c15c00f00e5127d06446f
> 58fce79951c3b0 100644
> > --- a/gcc/tree-vect-stmts.cc
> > +++ b/gcc/tree-vect-stmts.cc
> > @@ -93,7 +93,7 @@ stmt_in_inner_loop_p (vec_info *vinfo, class
> _stmt_vec_info *stmt_info)
> > target model or by saving it in a vector for later processing.
> > Return a preliminary estimate of the statement's cost. */
> >
> > -static unsigned
> > +unsigned
> > record_stmt_cost (stmt_vector_for_cost *body_cost_vec, int count,
> > enum vect_cost_for_stmt kind,
> > stmt_vec_info stmt_info, slp_tree node,
> > @@ -1008,8 +1008,8 @@ cfun_returns (tree decl)
> >
> > /* Calculate cost of DR's memory access. */
> > void
> > -vect_get_store_cost (vec_info *, stmt_vec_info stmt_info, int ncopies,
> > - dr_alignment_support alignment_support_scheme,
> > +vect_get_store_cost (vec_info *, stmt_vec_info stmt_info, slp_tree
> > slp_node,
> > + int ncopies, dr_alignment_support alignment_support_scheme,
> > int misalignment,
> > unsigned int *inside_cost,
> > stmt_vector_for_cost *body_cost_vec)
> > @@ -1019,7 +1019,7 @@ vect_get_store_cost (vec_info *, stmt_vec_info
> stmt_info, int ncopies,
> > case dr_aligned:
> > {
> > *inside_cost += record_stmt_cost (body_cost_vec, ncopies,
> > - vector_store, stmt_info, 0,
> > + vector_store, stmt_info, slp_node, 0,
> > vect_body);
> >
> > if (dump_enabled_p ())
> > @@ -1032,7 +1032,7 @@ vect_get_store_cost (vec_info *, stmt_vec_info
> stmt_info, int ncopies,
> > {
> > /* Here, we assign an additional cost for the unaligned store. */
> > *inside_cost += record_stmt_cost (body_cost_vec, ncopies,
> > - unaligned_store, stmt_info,
> > + unaligned_store, stmt_info, slp_node,
> > misalignment, vect_body);
> > if (dump_enabled_p ())
> > dump_printf_loc (MSG_NOTE, vect_location,
> > @@ -1058,8 +1058,8 @@ vect_get_store_cost (vec_info *, stmt_vec_info
> stmt_info, int ncopies,
> >
> > /* Calculate cost of DR's memory access. */
> > void
> > -vect_get_load_cost (vec_info *, stmt_vec_info stmt_info, int ncopies,
> > - dr_alignment_support alignment_support_scheme,
> > +vect_get_load_cost (vec_info *, stmt_vec_info stmt_info, slp_tree slp_node,
> > + int ncopies, dr_alignment_support alignment_support_scheme,
> > int misalignment,
> > bool add_realign_cost, unsigned int *inside_cost,
> > unsigned int *prologue_cost,
> > @@ -1072,7 +1072,7 @@ vect_get_load_cost (vec_info *, stmt_vec_info
> stmt_info, int ncopies,
> > case dr_aligned:
> > {
> > *inside_cost += record_stmt_cost (body_cost_vec, ncopies, vector_load,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0, vect_body);
> >
> > if (dump_enabled_p ())
> > dump_printf_loc (MSG_NOTE, vect_location,
> > @@ -1084,7 +1084,7 @@ vect_get_load_cost (vec_info *, stmt_vec_info
> stmt_info, int ncopies,
> > {
> > /* Here, we assign an additional cost for the unaligned load. */
> > *inside_cost += record_stmt_cost (body_cost_vec, ncopies,
> > - unaligned_load, stmt_info,
> > + unaligned_load, stmt_info, slp_node,
> > misalignment, vect_body);
> >
> > if (dump_enabled_p ())
> > @@ -1097,16 +1097,18 @@ vect_get_load_cost (vec_info *, stmt_vec_info
> stmt_info, int ncopies,
> > case dr_explicit_realign:
> > {
> > *inside_cost += record_stmt_cost (body_cost_vec, ncopies * 2,
> > - vector_load, stmt_info, 0, vect_body);
> > + vector_load, stmt_info, slp_node, 0,
> > + vect_body);
> > *inside_cost += record_stmt_cost (body_cost_vec, ncopies,
> > - vec_perm, stmt_info, 0, vect_body);
> > + vec_perm, stmt_info, slp_node, 0,
> > + vect_body);
> >
> > /* FIXME: If the misalignment remains fixed across the iterations
> > of
> > the containing loop, the following cost should be added to the
> > prologue costs. */
> > if (targetm.vectorize.builtin_mask_for_load)
> > *inside_cost += record_stmt_cost (body_cost_vec, 1, vector_stmt,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0, vect_body);
> >
> > if (dump_enabled_p ())
> > dump_printf_loc (MSG_NOTE, vect_location,
> > @@ -1132,17 +1134,17 @@ vect_get_load_cost (vec_info *, stmt_vec_info
> stmt_info, int ncopies,
> > {
> > *prologue_cost += record_stmt_cost (prologue_cost_vec, 2,
> > vector_stmt, stmt_info,
> > - 0, vect_prologue);
> > + slp_node, 0, vect_prologue);
> > if (targetm.vectorize.builtin_mask_for_load)
> > *prologue_cost += record_stmt_cost (prologue_cost_vec, 1,
> > vector_stmt, stmt_info,
> > - 0, vect_prologue);
> > + slp_node, 0, vect_prologue);
> > }
> >
> > *inside_cost += record_stmt_cost (body_cost_vec, ncopies, vector_load,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0, vect_body);
> > *inside_cost += record_stmt_cost (body_cost_vec, ncopies, vec_perm,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0, vect_body);
> >
> > if (dump_enabled_p ())
> > dump_printf_loc (MSG_NOTE, vect_location,
> > @@ -8551,7 +8553,7 @@ vectorizable_store (vec_info *vinfo,
> > if (vls_type == VLS_STORE_INVARIANT)
> > prologue_cost += record_stmt_cost (cost_vec, 1, scalar_to_vec,
> > stmt_info, 0, vect_prologue);
> > - vect_get_store_cost (vinfo, stmt_info, ncopies,
> > + vect_get_store_cost (vinfo, stmt_info, slp_node, ncopies,
> > alignment_support_scheme, misalignment,
> > &inside_cost, cost_vec);
> >
> > @@ -8623,7 +8625,7 @@ vectorizable_store (vec_info *vinfo,
> > else if (vls_type != VLS_STORE_INVARIANT)
> > return;
> > *prologue_cost += record_stmt_cost (cost_vec, 1, scalar_to_vec,
> > stmt_info,
> > - 0, vect_prologue);
> > + slp_node, 0, vect_prologue);
> > };
> >
> > if (memory_access_type == VMAT_ELEMENTWISE
> > @@ -8890,7 +8892,7 @@ vectorizable_store (vec_info *vinfo,
> > if (costing_p)
> > {
> > if (n_adjacent_stores > 0)
> > - vect_get_store_cost (vinfo, stmt_info, n_adjacent_stores,
> > + vect_get_store_cost (vinfo, stmt_info, slp_node, n_adjacent_stores,
> > alignment_support_scheme, misalignment,
> > &inside_cost, cost_vec);
> > if (dump_enabled_p ())
> > @@ -9202,7 +9204,7 @@ vectorizable_store (vec_info *vinfo,
> > if (costing_p)
> > {
> > if (n_adjacent_stores > 0)
> > - vect_get_store_cost (vinfo, stmt_info, n_adjacent_stores,
> > + vect_get_store_cost (vinfo, stmt_info, slp_node, n_adjacent_stores,
> > alignment_support_scheme, misalignment,
> > &inside_cost, cost_vec);
> > if (dump_enabled_p ())
> > @@ -9227,7 +9229,8 @@ vectorizable_store (vec_info *vinfo,
> > {
> > if (costing_p && vls_type == VLS_STORE_INVARIANT)
> > prologue_cost += record_stmt_cost (cost_vec, 1, scalar_to_vec,
> > - stmt_info, 0, vect_prologue);
> > + stmt_info, slp_node, 0,
> > + vect_prologue);
> > else if (!costing_p)
> > {
> > /* Since the store is not grouped, DR_GROUP_SIZE is 1, and
> > @@ -9304,7 +9307,8 @@ vectorizable_store (vec_info *vinfo,
> > unsigned int cnunits = vect_nunits_for_cost (vectype);
> > inside_cost
> > += record_stmt_cost (cost_vec, cnunits, scalar_store,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0,
> > + vect_body);
> > continue;
> > }
> >
> > @@ -9371,7 +9375,7 @@ vectorizable_store (vec_info *vinfo,
> > unsigned int cnunits = vect_nunits_for_cost (vectype);
> > inside_cost
> > += record_stmt_cost (cost_vec, cnunits, scalar_store,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0, vect_body);
> > continue;
> > }
> > poly_uint64 offset_nunits
> > @@ -9478,14 +9482,14 @@ vectorizable_store (vec_info *vinfo,
> > consumed by the load). */
> > inside_cost
> > += record_stmt_cost (cost_vec, cnunits, vec_to_scalar,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0, vect_body);
> > /* N scalar stores plus extracting the elements. */
> > inside_cost
> > += record_stmt_cost (cost_vec, cnunits, vec_to_scalar,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0, vect_body);
> > inside_cost
> > += record_stmt_cost (cost_vec, cnunits, scalar_store,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0, vect_body);
> > continue;
> > }
> >
> > @@ -9679,7 +9683,8 @@ vectorizable_store (vec_info *vinfo,
> > int group_size = DR_GROUP_SIZE (first_stmt_info);
> > int nstmts = ceil_log2 (group_size) * group_size;
> > inside_cost += record_stmt_cost (cost_vec, nstmts, vec_perm,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0,
> > + vect_body);
> > if (dump_enabled_p ())
> > dump_printf_loc (MSG_NOTE, vect_location,
> > "vect_model_store_cost: "
> > @@ -9708,7 +9713,8 @@ vectorizable_store (vec_info *vinfo,
> > {
> > if (costing_p)
> > inside_cost += record_stmt_cost (cost_vec, 1, vec_perm,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0,
> > + vect_body);
> > else
> > {
> > tree perm_mask = perm_mask_for_reverse (vectype);
> > @@ -9901,7 +9907,7 @@ vectorizable_store (vec_info *vinfo,
> > if (costing_p)
> > {
> > if (n_adjacent_stores > 0)
> > - vect_get_store_cost (vinfo, stmt_info, n_adjacent_stores,
> > + vect_get_store_cost (vinfo, stmt_info, slp_node, n_adjacent_stores,
> > alignment_support_scheme, misalignment,
> > &inside_cost, cost_vec);
> >
> > @@ -9927,11 +9933,11 @@ vectorizable_store (vec_info *vinfo,
> > /* Spill. */
> > prologue_cost
> > += record_stmt_cost (cost_vec, ncopies, vector_store,
> > - stmt_info, 0, vect_epilogue);
> > + stmt_info, slp_node, 0, vect_epilogue);
> > /* Loads. */
> > prologue_cost
> > += record_stmt_cost (cost_vec, ncopies * nregs, scalar_load,
> > - stmt_info, 0, vect_epilogue);
> > + stmt_info, slp_node, 0, vect_epilogue);
> > }
> > }
> > }
> > @@ -10502,9 +10508,10 @@ vectorizable_load (vec_info *vinfo,
> > enum vect_cost_model_location cost_loc
> > = hoist_p ? vect_prologue : vect_body;
> > unsigned int cost = record_stmt_cost (cost_vec, 1, scalar_load,
> > - stmt_info, 0, cost_loc);
> > - cost += record_stmt_cost (cost_vec, 1, scalar_to_vec, stmt_info, 0,
> > - cost_loc);
> > + stmt_info, slp_node, 0,
> > + cost_loc);
> > + cost += record_stmt_cost (cost_vec, 1, scalar_to_vec, stmt_info,
> > + slp_node, 0, cost_loc);
> > unsigned int prologue_cost = hoist_p ? cost : 0;
> > unsigned int inside_cost = hoist_p ? 0 : cost;
> > if (dump_enabled_p ())
> > @@ -10725,7 +10732,8 @@ vectorizable_load (vec_info *vinfo,
> > n_adjacent_loads++;
> > else
> > inside_cost += record_stmt_cost (cost_vec, 1, scalar_load,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0,
> > + vect_body);
> > continue;
> > }
> > tree this_off = build_int_cst (TREE_TYPE (alias_off),
> > @@ -10763,7 +10771,8 @@ vectorizable_load (vec_info *vinfo,
> > {
> > if (costing_p)
> > inside_cost += record_stmt_cost (cost_vec, 1, vec_construct,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0,
> > + vect_body);
> > else
> > {
> > tree vec_inv = build_constructor (lvectype, v);
> > @@ -10809,7 +10818,8 @@ vectorizable_load (vec_info *vinfo,
> > vect_transform_slp_perm_load (vinfo, slp_node, vNULL, NULL, vf,
> > true, &n_perms, &n_loads);
> > inside_cost += record_stmt_cost (cost_vec, n_perms, vec_perm,
> > - first_stmt_info, 0, vect_body);
> > + first_stmt_info, slp_node, 0,
> > + vect_body);
> > }
> > else
> > vect_transform_slp_perm_load (vinfo, slp_node, dr_chain, gsi, vf,
> > @@ -10819,7 +10829,7 @@ vectorizable_load (vec_info *vinfo,
> > if (costing_p)
> > {
> > if (n_adjacent_loads > 0)
> > - vect_get_load_cost (vinfo, stmt_info, n_adjacent_loads,
> > + vect_get_load_cost (vinfo, stmt_info, slp_node, n_adjacent_loads,
> > alignment_support_scheme, misalignment, false,
> > &inside_cost, nullptr, cost_vec, cost_vec,
> > true);
> > @@ -11173,7 +11183,7 @@ vectorizable_load (vec_info *vinfo,
> > "vect_model_load_cost: %d "
> > "unused vectors.\n",
> > gaps);
> > - vect_get_load_cost (vinfo, stmt_info, gaps,
> > + vect_get_load_cost (vinfo, stmt_info, slp_node, gaps,
> > alignment_support_scheme,
> > misalignment, false, &inside_cost,
> > &prologue_cost, cost_vec, cost_vec,
> > @@ -11302,7 +11312,7 @@ vectorizable_load (vec_info *vinfo,
> > if (costing_p)
> > {
> > if (n_adjacent_loads > 0)
> > - vect_get_load_cost (vinfo, stmt_info, n_adjacent_loads,
> > + vect_get_load_cost (vinfo, stmt_info, slp_node, n_adjacent_loads,
> > alignment_support_scheme, misalignment, false,
> > &inside_cost, &prologue_cost, cost_vec,
> > cost_vec, true);
> > @@ -11379,7 +11389,7 @@ vectorizable_load (vec_info *vinfo,
> > unsigned int cnunits = vect_nunits_for_cost (vectype);
> > inside_cost
> > = record_stmt_cost (cost_vec, cnunits, scalar_load,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0, vect_body);
> > continue;
> > }
> > if (STMT_VINFO_GATHER_SCATTER_P (stmt_info))
> > @@ -11455,7 +11465,7 @@ vectorizable_load (vec_info *vinfo,
> > unsigned int cnunits = vect_nunits_for_cost (vectype);
> > inside_cost
> > = record_stmt_cost (cost_vec, cnunits, scalar_load,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0, vect_body);
> > continue;
> > }
> > poly_uint64 offset_nunits
> > @@ -11590,7 +11600,7 @@ vectorizable_load (vec_info *vinfo,
> > vector. */
> > inside_cost
> > = record_stmt_cost (cost_vec, const_nunits, scalar_load,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0, vect_body);
> > inside_cost
> > = record_stmt_cost (cost_vec, 1, vec_construct,
> > stmt_info, slp_node, 0, vect_body);
> > @@ -12177,7 +12187,7 @@ vectorizable_load (vec_info *vinfo,
> > /* Leave realign cases alone to keep them simple. */
> > if (alignment_support_scheme == dr_explicit_realign_optimized
> > || alignment_support_scheme == dr_explicit_realign)
> > - vect_get_load_cost (vinfo, stmt_info, 1,
> > + vect_get_load_cost (vinfo, stmt_info, slp_node, 1,
> > alignment_support_scheme,
> misalignment,
> > add_realign_cost, &inside_cost,
> > &prologue_cost, cost_vec, cost_vec,
> > @@ -12250,7 +12260,8 @@ vectorizable_load (vec_info *vinfo,
> > {
> > if (costing_p)
> > inside_cost = record_stmt_cost (cost_vec, 1, vec_perm,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0,
> > + vect_body);
> > else
> > {
> > tree perm_mask = perm_mask_for_reverse (vectype);
> > @@ -12319,7 +12330,8 @@ vectorizable_load (vec_info *vinfo,
> > vect_transform_slp_perm_load (vinfo, slp_node, vNULL, nullptr, vf,
> > true, &n_perms, nullptr);
> > inside_cost = record_stmt_cost (cost_vec, n_perms, vec_perm,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0,
> > + vect_body);
> > }
> > else
> > {
> > @@ -12346,7 +12358,8 @@ vectorizable_load (vec_info *vinfo,
> > int group_size = DR_GROUP_SIZE (first_stmt_info);
> > int nstmts = ceil_log2 (group_size) * group_size;
> > inside_cost += record_stmt_cost (cost_vec, nstmts, vec_perm,
> > - stmt_info, 0, vect_body);
> > + stmt_info, slp_node, 0,
> > + vect_body);
> >
> > if (dump_enabled_p ())
> > dump_printf_loc (MSG_NOTE, vect_location,
> > @@ -12375,7 +12388,7 @@ vectorizable_load (vec_info *vinfo,
> > || memory_access_type == VMAT_CONTIGUOUS_REVERSE
> > || memory_access_type == VMAT_CONTIGUOUS_PERMUTE);
> > if (n_adjacent_loads > 0)
> > - vect_get_load_cost (vinfo, stmt_info, n_adjacent_loads,
> > + vect_get_load_cost (vinfo, stmt_info, slp_node, n_adjacent_loads,
> > alignment_support_scheme, misalignment, false,
> > &inside_cost, &prologue_cost, cost_vec, cost_vec,
> > true);
> > diff --git a/gcc/tree-vectorizer.h b/gcc/tree-vectorizer.h
> > index
> dcad41dcf182045e868a83276e39ca71a82738d5..7f69a3f57b492ad9ecbd63ec
> dea27e9abe386ac5 100644
> > --- a/gcc/tree-vectorizer.h
> > +++ b/gcc/tree-vectorizer.h
> > @@ -2354,6 +2354,10 @@ extern unsigned record_stmt_cost
> (stmt_vector_for_cost *, int,
> > extern unsigned record_stmt_cost (stmt_vector_for_cost *, int,
> > enum vect_cost_for_stmt,
> > enum vect_cost_model_location);
> > +extern unsigned record_stmt_cost (stmt_vector_for_cost *, int,
> > + enum vect_cost_for_stmt, stmt_vec_info,
> > + slp_tree, tree, int,
> > + enum vect_cost_model_location);
> >
> > /* Overload of record_stmt_cost with VECTYPE derived from STMT_INFO. */
> >
> > @@ -2375,12 +2379,8 @@ record_stmt_cost (stmt_vector_for_cost
> *body_cost_vec, int count,
> > slp_tree node,
> > int misalign, enum vect_cost_model_location where)
> > {
> > - if (node)
> > - return record_stmt_cost (body_cost_vec, count, kind, node,
> > - STMT_VINFO_VECTYPE (stmt_info), misalign, where);
> > - else
> > - return record_stmt_cost (body_cost_vec, count, kind, stmt_info,
> > - STMT_VINFO_VECTYPE (stmt_info), misalign, where);
> > + return record_stmt_cost (body_cost_vec, count, kind, stmt_info, node,
> > + STMT_VINFO_VECTYPE (stmt_info), misalign, where);
> > }
> >
> > extern void vect_finish_replace_stmt (vec_info *, stmt_vec_info, gimple *);
> > @@ -2411,12 +2411,12 @@ extern bool vect_nop_conversion_p
> (stmt_vec_info);
> > extern opt_result vect_analyze_stmt (vec_info *, stmt_vec_info, bool *,
> > slp_tree,
> > slp_instance, stmt_vector_for_cost *);
> > -extern void vect_get_load_cost (vec_info *, stmt_vec_info, int,
> > +extern void vect_get_load_cost (vec_info *, stmt_vec_info, slp_tree, int,
> > dr_alignment_support, int, bool,
> > unsigned int *, unsigned int *,
> > stmt_vector_for_cost *,
> > stmt_vector_for_cost *, bool);
> > -extern void vect_get_store_cost (vec_info *, stmt_vec_info, int,
> > +extern void vect_get_store_cost (vec_info *, stmt_vec_info, slp_tree, int,
> > dr_alignment_support, int,
> > unsigned int *, stmt_vector_for_cost *);
> > extern bool vect_supportable_shift (vec_info *, enum tree_code, tree);
> >
> >
> >
> >
> >
>
> --
> Richard Biener <[email protected]>
> SUSE Software Solutions Germany GmbH,
> Frankenstrasse 146, 90461 Nuernberg, Germany;
> GF: Ivo Totev, Andrew McDonald, Werner Knoblich; (HRB 36809, AG Nuernberg)
