blob: 78399b0b9bb4fc9ec2a547fd86bfc0ad2b3d26df [file] [log] [blame]
/* IPA function body analysis.
Copyright (C) 2003-2021 Free Software Foundation, Inc.
Contributed by Jan Hubicka
This file is part of GCC.
GCC is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free
Software Foundation; either version 3, or (at your option) any later
GCC is distributed in the hope that it will be useful, but WITHOUT ANY
WARRANTY; without even the implied warranty of MERCHANTABILITY or
for more details.
You should have received a copy of the GNU General Public License
along with GCC; see the file COPYING3. If not see
<>. */
#include "sreal.h"
#include "ipa-predicate.h"
/* Hints are reasons why IPA heuristics should prefer specializing given
function. They are represented as bitmap of the following values. */
enum ipa_hints_vals {
/* When specialization turns indirect call into a direct call,
it is good idea to do so. */
INLINE_HINT_indirect_call = 1,
/* Inlining may make loop iterations or loop stride known. It is good idea
to do so because it enables loop optimizations. */
INLINE_HINT_loop_iterations = 2,
INLINE_HINT_loop_stride = 4,
/* Inlining within same strongly connected component of callgraph is often
a loss due to increased stack frame usage and prologue setup costs. */
INLINE_HINT_same_scc = 8,
/* Inlining functions in strongly connected component is not such a great
win. */
INLINE_HINT_in_scc = 16,
/* If function is declared inline by user, it may be good idea to inline
it. Set by simple_edge_hints in ipa-inline-analysis.c. */
INLINE_HINT_declared_inline = 32,
/* Programs are usually still organized for non-LTO compilation and thus
if functions are in different modules, inlining may not be so important.
Set by simple_edge_hints in ipa-inline-analysis.c. */
INLINE_HINT_cross_module = 64,
/* We know that the callee is hot by profile. */
INLINE_HINT_known_hot = 128,
/* There is builtin_constant_p dependent on parameter which is usually
a strong hint to inline. */
INLINE_HINT_builtin_constant_p = 256
typedef int ipa_hints;
/* Simple description of whether a memory load or a condition refers to a load
from an aggregate and if so, how and where from in the aggregate.
Individual fields have the same meaning like fields with the same name in
struct condition. */
struct agg_position_info
bool agg_contents;
bool by_ref;
/* Representation of function body size and time depending on the call
context. We keep simple array of record, every containing of predicate
and time/size to account. */
class size_time_entry
/* Predicate for code to be executed. */
predicate exec_predicate;
/* Predicate for value to be constant and optimized out in a specialized copy.
When deciding on specialization this makes it possible to see how much
the executed code paths will simplify. */
predicate nonconst_predicate;
int size;
sreal time;
/* Summary about function and stack frame sizes. We keep this info
for inline clones and also for WPA streaming. For this reason this is not
part of ipa_fn_summary which exists only for offline functions. */
class ipa_size_summary
/* Estimated stack frame consumption by the function. */
HOST_WIDE_INT estimated_self_stack_size;
/* Size of the function body. */
int self_size;
/* Estimated size of the function after inlining. */
int size;
ipa_size_summary ()
: estimated_self_stack_size (0), self_size (0), size (0)
/* Structure to capture how frequently some interesting events occur given a
particular predicate. The structure is used to estimate how often we
encounter loops with known iteration count or stride in various
contexts. */
struct GTY(()) ipa_freqcounting_predicate
/* The described event happens with this frequency... */
sreal freq;
/* ...when this predicate evaluates to false. */
class predicate * GTY((skip)) predicate;
/* Function inlining information. */
class GTY(()) ipa_fn_summary
/* Keep all field empty so summary dumping works during its computation.
This is useful for debugging. */
ipa_fn_summary ()
: min_size (0),
inlinable (false), single_caller (false),
fp_expressions (false),
estimated_stack_size (false),
time (0), conds (NULL),
size_time_table (), call_size_time_table (vNULL),
loop_iterations (NULL), loop_strides (NULL),
builtin_constant_p_parms (vNULL),
growth (0), scc_no (0)
/* Copy constructor. */
ipa_fn_summary (const ipa_fn_summary &s)
: min_size (s.min_size),
inlinable (s.inlinable), single_caller (s.single_caller),
fp_expressions (s.fp_expressions),
estimated_stack_size (s.estimated_stack_size),
time (s.time), conds (s.conds), size_time_table (),
call_size_time_table (vNULL),
loop_iterations (s.loop_iterations), loop_strides (s.loop_strides),
builtin_constant_p_parms (s.builtin_constant_p_parms),
growth (s.growth), scc_no (s.scc_no)
/* Default constructor. */
~ipa_fn_summary ();
/* Information about the function body itself. */
/* Minimal size increase after inlining. */
int min_size;
/* False when there something makes inlining impossible (such as va_arg). */
unsigned inlinable : 1;
/* True wen there is only one caller of the function before small function
inlining. */
unsigned int single_caller : 1;
/* True if function contains any floating point expressions. */
unsigned int fp_expressions : 1;
/* Information about function that will result after applying all the
inline decisions present in the callgraph. Generally kept up to
date only for functions that are not inline clones. */
/* Estimated stack frame consumption by the function. */
HOST_WIDE_INT estimated_stack_size;
/* Estimated runtime of function after inlining. */
sreal GTY((skip)) time;
/* Conditional size/time information. The summaries are being
merged during inlining. */
conditions conds;
/* Normal code is accounted in size_time_table, while calls are
accounted in call_size_time_table. This is because calls
are often adjusted by IPA optimizations and thus this summary
is generated from call summary information when needed. */
auto_vec<size_time_entry> GTY((skip)) size_time_table;
/* Unlike size_time_table that is initialized for all summaries
call_size_time_table is allocated only for functions with
many calls. Use effecient vl_ptr storage. */
vec<size_time_entry, va_heap, vl_ptr> GTY((skip)) call_size_time_table;
/* Predicates on when some loops in the function can have known bounds. */
vec<ipa_freqcounting_predicate, va_gc> *loop_iterations;
/* Predicates on when some loops in the function can have known strides. */
vec<ipa_freqcounting_predicate, va_gc> *loop_strides;
/* Parameters tested by builtin_constant_p. */
vec<int, va_heap, vl_ptr> GTY((skip)) builtin_constant_p_parms;
/* Estimated growth for inlining all copies of the function before start
of small functions inlining.
This value will get out of date as the callers are duplicated, but
using up-to-date value in the badness metric mean a lot of extra
expenses. */
int growth;
/* Number of SCC on the beginning of inlining process. */
int scc_no;
/* Record time and size under given predicates. */
void account_size_time (int, sreal, const predicate &, const predicate &,
bool call = false);
/* We keep values scaled up, so fractional sizes can be accounted. */
static const int size_scale = 2;
/* Maximal size of size_time_table before we start to be conservative. */
static const int max_size_time_table_size = 256;
class GTY((user)) ipa_fn_summary_t:
public fast_function_summary <ipa_fn_summary *, va_gc>
ipa_fn_summary_t (symbol_table *symtab):
fast_function_summary <ipa_fn_summary *, va_gc> (symtab) {}
static ipa_fn_summary_t *create_ggc (symbol_table *symtab)
class ipa_fn_summary_t *summary
= new (ggc_alloc_no_dtor<ipa_fn_summary_t> ()) ipa_fn_summary_t (symtab);
summary->disable_insertion_hook ();
return summary;
/* Remove ipa_fn_summary for all callees of NODE. */
void remove_callees (cgraph_node *node);
virtual void insert (cgraph_node *, ipa_fn_summary *);
virtual void remove (cgraph_node *node, ipa_fn_summary *)
remove_callees (node);
virtual void duplicate (cgraph_node *src, cgraph_node *dst,
ipa_fn_summary *src_data, ipa_fn_summary *dst_data);
extern GTY(()) fast_function_summary <ipa_fn_summary *, va_gc>
class ipa_size_summary_t:
public fast_function_summary <ipa_size_summary *, va_heap>
ipa_size_summary_t (symbol_table *symtab):
fast_function_summary <ipa_size_summary *, va_heap> (symtab)
disable_insertion_hook ();
virtual void duplicate (cgraph_node *, cgraph_node *,
ipa_size_summary *src_data,
ipa_size_summary *dst_data)
*dst_data = *src_data;
extern fast_function_summary <ipa_size_summary *, va_heap>
/* Information kept about callgraph edges. */
class ipa_call_summary
/* Keep all field empty so summary dumping works during its computation.
This is useful for debugging. */
ipa_call_summary ()
: predicate (NULL), param (vNULL), call_stmt_size (0), call_stmt_time (0),
loop_depth (0), is_return_callee_uncaptured (false)
/* Copy constructor. */
ipa_call_summary (const ipa_call_summary &s):
predicate (s.predicate), param (s.param), call_stmt_size (s.call_stmt_size),
call_stmt_time (s.call_stmt_time), loop_depth (s.loop_depth),
is_return_callee_uncaptured (s.is_return_callee_uncaptured)
/* Default destructor. */
~ipa_call_summary ();
class predicate *predicate;
/* Vector indexed by parameters. */
vec<inline_param_summary> param;
/* Estimated size and time of the call statement. */
int call_stmt_size;
int call_stmt_time;
/* Depth of loop nest, 0 means no nesting. */
unsigned int loop_depth;
/* Indicates whether the caller returns the value of it's callee. */
bool is_return_callee_uncaptured;
class ipa_call_summary_t: public fast_call_summary <ipa_call_summary *, va_heap>
ipa_call_summary_t (symbol_table *symtab):
fast_call_summary <ipa_call_summary *, va_heap> (symtab) {}
/* Hook that is called by summary when an edge is duplicated. */
virtual void duplicate (cgraph_edge *src, cgraph_edge *dst,
ipa_call_summary *src_data,
ipa_call_summary *dst_data);
/* Estimated execution times, code sizes and other information about the
code executing a call described by ipa_call_context. */
struct ipa_call_estimates
/* Estimated size needed to execute call in the given context. */
int size;
/* Minimal size needed for the call that is + independent on the call context
and can be used for fast estimates. */
int min_size;
/* Estimated time needed to execute call in the given context. */
sreal time;
/* Estimated time needed to execute the function when not ignoring
computations known to be constant in this context. */
sreal nonspecialized_time;
/* Further discovered reasons why to inline or specialize the give calls. */
ipa_hints hints;
/* Frequency how often a loop with known number of iterations is encountered.
Calculated with hints. */
sreal loops_with_known_iterations;
/* Frequency how often a loop with known strides is encountered. Calculated
with hints. */
sreal loops_with_known_strides;
class ipa_cached_call_context;
/* This object describe a context of call. That is a summary of known
information about its parameters. Main purpose of this context is
to give more realistic estimations of function runtime, size and
inline hints. */
class ipa_call_context
ipa_call_context (cgraph_node *node,
clause_t possible_truths,
clause_t nonspec_possible_truths,
vec<inline_param_summary> inline_param_summary,
ipa_auto_call_arg_values *arg_values);
ipa_call_context ()
: m_node(NULL)
void estimate_size_and_time (ipa_call_estimates *estimates,
bool est_times = true, bool est_hints = true);
bool equal_to (const ipa_call_context &);
bool exists_p ()
return m_node != NULL;
/* Called function. */
cgraph_node *m_node;
/* Clause describing what predicate conditionals can be satisfied
in this context if function is inlined/specialized. */
clause_t m_possible_truths;
/* Clause describing what predicate conditionals can be satisfied
in this context if function is kept offline. */
clause_t m_nonspec_possible_truths;
/* Inline summary maintains info about change probabilities. */
vec<inline_param_summary> m_inline_param_summary;
/* Even after having calculated clauses, the information about argument
values is used to resolve indirect calls. */
ipa_call_arg_values m_avals;
friend ipa_cached_call_context;
/* Variant of ipa_call_context that is stored in a cache over a longer period
of time. */
class ipa_cached_call_context : public ipa_call_context
void duplicate_from (const ipa_call_context &ctx);
void release ();
extern fast_call_summary <ipa_call_summary *, va_heap> *ipa_call_summaries;
/* In ipa-fnsummary.c */
void ipa_debug_fn_summary (struct cgraph_node *);
void ipa_dump_fn_summaries (FILE *f);
void ipa_dump_fn_summary (FILE *f, struct cgraph_node *node);
void ipa_dump_hints (FILE *f, ipa_hints);
void ipa_free_fn_summary (void);
void ipa_free_size_summary (void);
void inline_analyze_function (struct cgraph_node *node);
void estimate_ipcp_clone_size_and_time (struct cgraph_node *node,
ipa_auto_call_arg_values *avals,
ipa_call_estimates *estimates);
void ipa_merge_fn_summary_after_inlining (struct cgraph_edge *edge);
void ipa_update_overall_fn_summary (struct cgraph_node *node, bool reset = true);
void compute_fn_summary (struct cgraph_node *, bool);
bool refs_local_or_readonly_memory_p (tree);
bool points_to_local_or_readonly_memory_p (tree);
void evaluate_properties_for_edge (struct cgraph_edge *e,
bool inline_p,
clause_t *clause_ptr,
clause_t *nonspec_clause_ptr,
ipa_auto_call_arg_values *avals,
bool compute_contexts);
void ipa_fnsummary_c_finalize (void);
HOST_WIDE_INT ipa_get_stack_frame_offset (struct cgraph_node *node);
void ipa_remove_from_growth_caches (struct cgraph_edge *edge);
/* Return true if EDGE is a cross module call. */
static inline bool
cross_module_call_p (struct cgraph_edge *edge)
/* Here we do not want to walk to alias target becuase ICF may create
cross-unit aliases. */
if (edge->caller->unit_id == edge->callee->unit_id)
return false;
/* If the call is to a (former) comdat function or s symbol with mutiple
extern inline definitions then treat is as in-module call. */
if (edge->callee->merged_extern_inline || edge->callee->merged_comdat
|| DECL_COMDAT (edge->callee->decl))
return false;
return true;
#endif /* GCC_IPA_FNSUMMARY_H */