From bc23502b7f6590639be6473b4b8702187cac895d Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Wed, 11 Jan 2006 13:02:18 +0000 Subject: [PATCH] re PR tree-optimization/23109 (compiler generates wrong code leading to spurious division by zero with -funsafe-math-optimizations (instead of -ftrapping-math)) gcc: 2006-01-11 Paolo Bonzini PR tree-optimization/23109 PR tree-optimization/23948 PR tree-optimization/24123 * Makefile.in (tree-ssa-math-opts.o): Adjust dependencies. * tree-cfg.c (single_noncomplex_succ): New. * tree-flow.h (single_noncomplex_succ): Declare it. * tree-ssa-math-opts.c (enum place_reciprocal): Remove. * tree-ssa-math-opts.c (enum place_reciprocal): Remove. (struct occurrence, occ_head, occ_pool, is_divide_by, compute_merit, insert_bb, register_division_in, insert_reciprocals, replace_reciprocal, free_bb): New. (execute_cse_reciprocals_1): Rewritten. (execute_cse_reciprocals): Adjust calls to execute_cse_reciprocals_1. Do not commit any edge insertion. Always compute dominators and create the allocation pool. * target-def.h (TARGET_MIN_DIVISIONS_FOR_RECIP_MUL): New. * target.h (struct gcc_target): Add min_divistions_for_recip_mul. * targhooks.c (default_min_divistions_for_recip_mul): New. * targhooks.h (default_min_divistions_for_recip_mul): New prototype. * passes.c (init_optimization_passes): Run recip after tree loop optimizations. * doc/tm.texi (Misc): Document TARGET_MIN_DIVISIONS_FOR_RECIP_MUL. gcc/testsuite: 2006-01-11 Paolo Bonzini PR tree-optimization/23109 PR tree-optimization/23948 PR tree-optimization/24123 * gcc.dg/tree-ssa/recip-3.c, gcc.dg/tree-ssa/recip-4.c, gcc.dg/tree-ssa/recip-5.c, gcc.dg/tree-ssa/recip-6.c, gcc.dg/tree-ssa/recip-7.c, gcc.dg/tree-ssa/pr23109.c, g++.dg/tree-ssa/pr23948.C: New testcases. * gcc.dg/tree-ssa/recip-2.c, gcc.dg/tree-ssa/pr23234.c: Provide three divisions in order to do the optimization. From-SVN: r109578 --- gcc/ChangeLog | 52 ++- gcc/Makefile.in | 3 +- gcc/doc/tm.texi | 9 + gcc/passes.c | 2 +- gcc/target-def.h | 5 + gcc/target.h | 5 + gcc/targhooks.c | 8 + gcc/targhooks.h | 1 + gcc/testsuite/ChangeLog | 13 + gcc/testsuite/g++.dg/tree-ssa/pr23948.C | 19 + gcc/testsuite/gcc.dg/tree-ssa/pr23109.c | 34 ++ gcc/testsuite/gcc.dg/tree-ssa/pr23234.c | 12 +- gcc/testsuite/gcc.dg/tree-ssa/recip-2.c | 7 +- gcc/testsuite/gcc.dg/tree-ssa/recip-3.c | 27 ++ gcc/testsuite/gcc.dg/tree-ssa/recip-4.c | 45 +++ gcc/testsuite/gcc.dg/tree-ssa/recip-5.c | 31 ++ gcc/testsuite/gcc.dg/tree-ssa/recip-6.c | 26 ++ gcc/testsuite/gcc.dg/tree-ssa/recip-7.c | 27 ++ gcc/tree-cfg.c | 24 ++ gcc/tree-flow.h | 1 + gcc/tree-ssa-math-opts.c | 513 +++++++++++++++++++----- 21 files changed, 745 insertions(+), 119 deletions(-) create mode 100644 gcc/testsuite/g++.dg/tree-ssa/pr23948.C create mode 100644 gcc/testsuite/gcc.dg/tree-ssa/pr23109.c create mode 100644 gcc/testsuite/gcc.dg/tree-ssa/recip-3.c create mode 100644 gcc/testsuite/gcc.dg/tree-ssa/recip-4.c create mode 100644 gcc/testsuite/gcc.dg/tree-ssa/recip-5.c create mode 100644 gcc/testsuite/gcc.dg/tree-ssa/recip-6.c create mode 100644 gcc/testsuite/gcc.dg/tree-ssa/recip-7.c diff --git a/gcc/ChangeLog b/gcc/ChangeLog index 2f622a133a9..0701660bb13 100644 --- a/gcc/ChangeLog +++ b/gcc/ChangeLog @@ -1,3 +1,29 @@ +2006-01-11 Paolo Bonzini + + PR tree-optimization/23109 + PR tree-optimization/23948 + PR tree-optimization/24123 + + * Makefile.in (tree-ssa-math-opts.o): Adjust dependencies. + * tree-cfg.c (single_noncomplex_succ): New. + * tree-flow.h (single_noncomplex_succ): Declare it. + * tree-ssa-math-opts.c (enum place_reciprocal): Remove. + * tree-ssa-math-opts.c (enum place_reciprocal): Remove. + (struct occurrence, occ_head, occ_pool, is_divide_by, compute_merit, + insert_bb, register_division_in, insert_reciprocals, + replace_reciprocal, free_bb): New. + (execute_cse_reciprocals_1): Rewritten. + (execute_cse_reciprocals): Adjust calls to execute_cse_reciprocals_1. + Do not commit any edge insertion. Always compute dominators and + create the allocation pool. + * target-def.h (TARGET_MIN_DIVISIONS_FOR_RECIP_MUL): New. + * target.h (struct gcc_target): Add min_divistions_for_recip_mul. + * targhooks.c (default_min_divistions_for_recip_mul): New. + * targhooks.h (default_min_divistions_for_recip_mul): New prototype. + * passes.c (init_optimization_passes): Run recip after tree loop + optimizations. + * doc/tm.texi (Misc): Document TARGET_MIN_DIVISIONS_FOR_RECIP_MUL. + 2005-01-11 Danny Berlin Kenneth Zadeck @@ -151,31 +177,31 @@ 2006-01-10 John David Anglin PR target/20754 - * pa.md: Create separate 32 and 64-bit move patterns for SI, DI, SF - and DF modes. Add alternatives to copy between general and floating - point registers to the 32-bit patterns. - * pa-64.h (SECONDARY_MEMORY_NEEDED_RTX): Delete undefine. - * pa.h (SECONDARY_MEMORY_NEEDED_RTX): Delete define. + * config/pa/pa.md: Create separate 32 and 64-bit move patterns + for SI, DI, SF and DF modes. Add alternatives to copy between + general and floating point registers to the 32-bit patterns. + * config/pa/pa-64.h (SECONDARY_MEMORY_NEEDED_RTX): Delete undefine. + * config/pa/pa.h (SECONDARY_MEMORY_NEEDED_RTX): Delete define. (SECONDARY_MEMORY_NEEDED): Secondary memory is only needed when generating 64-bit code. - * pa.c (output_move_double): Handle copies between general and - floating registers. + * config/pa/pa.c (output_move_double): Handle copies between general + and floating registers. 2006-01-10 Stuart Hastings - * gcc/config/i386/i386.md (set_got): Update. + * config/i386/i386.md (set_got): Update. (set_got_labelled): New. (UNSPEC_LD_MPIC): New. (builtin_setjmp_receiver): Mach-O support. - * gcc/config/i386/darwin.h (TARGET_ASM_FILE_END) Define. + * config/i386/darwin.h (TARGET_ASM_FILE_END) Define. (GOT_SYMBOL_NAME): Define. (FORCE_PREFERRED_STACK_BOUNDARY_IN_MAIN): New. (TARGET_DEEP_BRANCH_PREDICTION): Remove. - * gcc/config/i386/i386.c (override_options): Revise for Darwin. + * config/i386/i386.c (override_options): Revise for Darwin. (USE_HIDDEN_LINKONCE): Enable for Mach-O. (ix86_file_end): Mach-O support. (darwin_x86_file_end): New. (output_set_got): Add label parameter, revise for Mach-O. (x86_output_mi_thunk): Likewise. - * gcc/config/i386/i386-protos.h (output_set_got): Likewise. - * gcc/config/darwin.c (machopic_legitimize_pic_address): Update + * config/i386/i386-protos.h (output_set_got): Likewise. + * config/darwin.c (machopic_legitimize_pic_address): Update regs_ever_live[]. 2006-01-10 Kaz Kojima @@ -604,7 +630,7 @@ 2006-01-03 Adrian Straetling - * gcc/builtins.c (get_builtin_sync_mem): New function. + * builtins.c (get_builtin_sync_mem): New function. (expand_builtin_sync_operation, expand_builtin_compare_and_swap, expand_builtin_lock_test_and_set, expand_builtin_lock_release): Call get_builtin_sync_mem to generate mem rtx. diff --git a/gcc/Makefile.in b/gcc/Makefile.in index 8318090645a..895532158a1 100644 --- a/gcc/Makefile.in +++ b/gcc/Makefile.in @@ -1970,7 +1970,8 @@ tree-ssa-loop-im.o : tree-ssa-loop-im.c $(TREE_FLOW_H) $(CONFIG_H) \ $(TREE_DUMP_H) tree-pass.h $(FLAGS_H) real.h $(BASIC_BLOCK_H) \ hard-reg-set.h tree-ssa-math-opts.o : tree-ssa-math-opts.c $(TREE_FLOW_H) $(CONFIG_H) \ - $(SYSTEM_H) $(TREE_H) $(TIMEVAR_H) tree-pass.h $(TM_H) $(FLAGS_H) + $(SYSTEM_H) $(TREE_H) $(TIMEVAR_H) tree-pass.h $(TM_H) $(FLAGS_H) \ + alloc-pool.h $(BASIC_BLOCK_H) $(TARGET_H) tree-ssa-alias.o : tree-ssa-alias.c $(TREE_FLOW_H) $(CONFIG_H) $(SYSTEM_H) \ $(RTL_H) $(TREE_H) $(TM_P_H) $(EXPR_H) $(GGC_H) tree-inline.h $(FLAGS_H) \ function.h $(TIMEVAR_H) convert.h $(TM_H) coretypes.h langhooks.h \ diff --git a/gcc/doc/tm.texi b/gcc/doc/tm.texi index 61b22e42ccf..146b475b96e 100644 --- a/gcc/doc/tm.texi +++ b/gcc/doc/tm.texi @@ -8893,6 +8893,15 @@ point number to a signed fixed point number also convert validly to an unsigned one. @end defmac +@deftypefn {Target Hook} int TARGET_MIN_DIVISIONS_FOR_RECIP_MUL (enum machine_mode @var{mode}) +When @option{-ffast-math} is in effect, GCC tries to optimize +divisions by the same divisor, by turning them into multiplications by +the reciprocal. This target hook specifies the minimum number of divisions +that should be there for GCC to perform the optimization for a variable +of mode @var{mode}. The default implementation returns 3 if the machine +has an instruction for the division, and 2 if it does not. +@end deftypefn + @defmac MOVE_MAX The maximum number of bytes that a single instruction can move quickly between memory and registers or between two memory locations. diff --git a/gcc/passes.c b/gcc/passes.c index 903a9fb2e9d..7d43bedd847 100644 --- a/gcc/passes.c +++ b/gcc/passes.c @@ -551,12 +551,12 @@ init_optimization_passes (void) we add may_alias right after fold builtins which can create arbitrary GIMPLE. */ NEXT_PASS (pass_may_alias); - NEXT_PASS (pass_cse_reciprocals); NEXT_PASS (pass_split_crit_edges); NEXT_PASS (pass_pre); NEXT_PASS (pass_may_alias); NEXT_PASS (pass_sink_code); NEXT_PASS (pass_tree_loop); + NEXT_PASS (pass_cse_reciprocals); NEXT_PASS (pass_reassoc); NEXT_PASS (pass_dominator); diff --git a/gcc/target-def.h b/gcc/target-def.h index 204be4e6030..642e9701040 100644 --- a/gcc/target-def.h +++ b/gcc/target-def.h @@ -336,6 +336,10 @@ Foundation, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. #define TARGET_SHIFT_TRUNCATION_MASK default_shift_truncation_mask #endif +#ifndef TARGET_MIN_DIVISIONS_FOR_RECIP_MUL +#define TARGET_MIN_DIVISIONS_FOR_RECIP_MUL default_min_divisions_for_recip_mul +#endif + #ifndef TARGET_VALID_POINTER_MODE #define TARGET_VALID_POINTER_MODE default_valid_pointer_mode #endif @@ -588,6 +592,7 @@ Foundation, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. TARGET_ENCODE_SECTION_INFO, \ TARGET_STRIP_NAME_ENCODING, \ TARGET_SHIFT_TRUNCATION_MASK, \ + TARGET_MIN_DIVISIONS_FOR_RECIP_MUL, \ TARGET_VALID_POINTER_MODE, \ TARGET_SCALAR_MODE_SUPPORTED_P, \ TARGET_VECTOR_MODE_SUPPORTED_P, \ diff --git a/gcc/target.h b/gcc/target.h index 8c61bfe9692..b32be830120 100644 --- a/gcc/target.h +++ b/gcc/target.h @@ -440,6 +440,11 @@ struct gcc_target return the mask that they apply. Return 0 otherwise. */ unsigned HOST_WIDE_INT (* shift_truncation_mask) (enum machine_mode mode); + /* Return the number of divisions in the given MODE that should be present, + so that it is profitable to turn the division into a multiplication by + the reciprocal. */ + unsigned int (* min_divisions_for_recip_mul) (enum machine_mode mode); + /* True if MODE is valid for a pointer in __attribute__((mode("MODE"))). */ bool (* valid_pointer_mode) (enum machine_mode mode); diff --git a/gcc/targhooks.c b/gcc/targhooks.c index 2ff2b9b5a44..786b1522afb 100644 --- a/gcc/targhooks.c +++ b/gcc/targhooks.c @@ -148,6 +148,14 @@ default_shift_truncation_mask (enum machine_mode mode) return SHIFT_COUNT_TRUNCATED ? GET_MODE_BITSIZE (mode) - 1 : 0; } +/* The default implementation of TARGET_MIN_DIVISIONS_FOR_RECIP_MUL. */ + +unsigned int +default_min_divisions_for_recip_mul (enum machine_mode mode ATTRIBUTE_UNUSED) +{ + return have_insn_for (DIV, mode) ? 3 : 2; +} + /* Generic hook that takes a CUMULATIVE_ARGS pointer and returns true. */ bool diff --git a/gcc/targhooks.h b/gcc/targhooks.h index 96bff20aebe..44396480b7f 100644 --- a/gcc/targhooks.h +++ b/gcc/targhooks.h @@ -33,6 +33,7 @@ extern bool default_pretend_outgoing_varargs_named (CUMULATIVE_ARGS *); extern enum machine_mode default_eh_return_filter_mode (void); extern unsigned HOST_WIDE_INT default_shift_truncation_mask (enum machine_mode); +extern unsigned int default_min_divisions_for_recip_mul (enum machine_mode); extern tree default_stack_protect_guard (void); extern tree default_external_stack_protect_fail (void); diff --git a/gcc/testsuite/ChangeLog b/gcc/testsuite/ChangeLog index d2aa0aa3610..a91cd82c6d1 100644 --- a/gcc/testsuite/ChangeLog +++ b/gcc/testsuite/ChangeLog @@ -1,3 +1,16 @@ +2006-01-11 Paolo Bonzini + + PR tree-optimization/23109 + PR tree-optimization/23948 + PR tree-optimization/24123 + + * gcc.dg/tree-ssa/recip-3.c, gcc.dg/tree-ssa/recip-4.c, + gcc.dg/tree-ssa/recip-5.c, gcc.dg/tree-ssa/recip-6.c, + gcc.dg/tree-ssa/recip-7.c, gcc.dg/tree-ssa/pr23109.c, + g++.dg/tree-ssa/pr23948.C: New testcases. + * gcc.dg/tree-ssa/recip-2.c, gcc.dg/tree-ssa/pr23234.c: Provide + three divisions in order to do the optimization. + 2005-01-11 Zdenek Dvorak PR c++/25632 diff --git a/gcc/testsuite/g++.dg/tree-ssa/pr23948.C b/gcc/testsuite/g++.dg/tree-ssa/pr23948.C new file mode 100644 index 00000000000..c34161409c1 --- /dev/null +++ b/gcc/testsuite/g++.dg/tree-ssa/pr23948.C @@ -0,0 +1,19 @@ +/* { dg-options "-O1 -ffast-math -fdump-tree-recip" } */ +/* { dg-do compile } */ + +struct MIOFILE { + ~MIOFILE(); +}; +double potentially_runnable_resource_share(); +void f1(double); +int make_scheduler_request(double a, double b) +{ + MIOFILE mf; + double prrs = potentially_runnable_resource_share(); + f1(a/prrs); + f1(1/prrs); + f1(b/prrs); +} + +/* { dg-final { scan-tree-dump-times " / " 1 "recip" } } */ +/* { dg-final { cleanup-tree-dump "recip" } } */ diff --git a/gcc/testsuite/gcc.dg/tree-ssa/pr23109.c b/gcc/testsuite/gcc.dg/tree-ssa/pr23109.c new file mode 100644 index 00000000000..e121a553833 --- /dev/null +++ b/gcc/testsuite/gcc.dg/tree-ssa/pr23109.c @@ -0,0 +1,34 @@ +/* { dg-do compile } */ +/* { dg-options "-O2 -funsafe-math-optimizations -fdump-tree-recip -fdump-tree-lim" } */ + +double F[2] = { 0., 0. }, e = 0.; + +int main() +{ + int i; + double E, W, P, d; + + /* make sure the program crashes on FP exception */ + unsigned short int Mask; + + W = 1.; + d = 2.*e; + E = 1. - d; + + for( i=0; i < 2; i++ ) + if( d > 0.01 ) + { + P = ( W < E ) ? (W - E)/d : (E - W)/d; + F[i] += P; + } + + return 0; +} + +/* LIM only performs the transformation in the no-trapping-math case. In + the future we will do it for trapping-math as well in recip, check that + this is not wrongly optimized. */ +/* { dg-final { scan-tree-dump-not "reciptmp" "lim" } } */ +/* { dg-final { scan-tree-dump-not "reciptmp" "recip" } } */ +/* { dg-final { cleanup-tree-dump "recip" } } */ + diff --git a/gcc/testsuite/gcc.dg/tree-ssa/pr23234.c b/gcc/testsuite/gcc.dg/tree-ssa/pr23234.c index bd0b62b558f..3a3869815c8 100644 --- a/gcc/testsuite/gcc.dg/tree-ssa/pr23234.c +++ b/gcc/testsuite/gcc.dg/tree-ssa/pr23234.c @@ -9,6 +9,7 @@ double f1 (double a, double b, double c) { double y0; + double y1; if (a == 0.0) { @@ -16,7 +17,8 @@ f1 (double a, double b, double c) return y0; } y0 = c / b; - return y0; + y1 = a / b; + return y0 * y1; } /* Labels may end up in the middle of a block. Also bad. */ @@ -24,6 +26,7 @@ double f2 (double a, double b, double c) { double y0; + double y1; a_label: another_label: @@ -33,7 +36,8 @@ another_label: return y0; } y0 = c / b; - return y0; + y1 = a / b; + return y0 * y1; } /* Uses must still be dominated by their defs. */ @@ -41,6 +45,7 @@ double f3 (double a, double b, double c) { double y0; + double y1; y0 = -c / b; if (a == 0.0) @@ -48,5 +53,6 @@ f3 (double a, double b, double c) return y0; } y0 = c / b; - return y0; + y1 = a / b; + return y0 * y1; } diff --git a/gcc/testsuite/gcc.dg/tree-ssa/recip-2.c b/gcc/testsuite/gcc.dg/tree-ssa/recip-2.c index 7d0e97ae371..af628053ad5 100644 --- a/gcc/testsuite/gcc.dg/tree-ssa/recip-2.c +++ b/gcc/testsuite/gcc.dg/tree-ssa/recip-2.c @@ -10,14 +10,19 @@ float e(float a, float b, float c, float d, float e, float f) } /* The PHI nodes for these divisions should be combined. */ + d = d / a; e = e / a; f = f / a; a = a / c; b = b / c; - return a + b + e + f; + /* This should not be left as a multiplication. */ + c = 1 / c; + + return a + b + c + d + e + f; } /* { dg-final { scan-tree-dump-times " / " 2 "recip" } } */ +/* { dg-final { scan-tree-dump-times " \\* " 5 "recip" } } */ /* { dg-final { cleanup-tree-dump "recip" } } */ diff --git a/gcc/testsuite/gcc.dg/tree-ssa/recip-3.c b/gcc/testsuite/gcc.dg/tree-ssa/recip-3.c new file mode 100644 index 00000000000..45c8db148ac --- /dev/null +++ b/gcc/testsuite/gcc.dg/tree-ssa/recip-3.c @@ -0,0 +1,27 @@ +/* { dg-do compile } */ +/* { dg-options "-O1 -fno-trapping-math -funsafe-math-optimizations -fdump-tree-recip" } */ + +double F[2] = { 0.0, 0.0 }, e; + +/* In this case the optimization is interesting. */ +float h () +{ + int i; + double E, W, P, d; + + W = 1.; + d = 2.*e; + E = 1. - d; + + for( i=0; i < 2; i++ ) + if( d > 0.01 ) + { + P = ( W < E ) ? (W - E)/d : (E - W)/d; + F[i] += P; + } + + F[0] += E / d; +} + +/* { dg-final { scan-tree-dump-times " / " 1 "recip" } } */ +/* { dg-final { cleanup-tree-dump "recip" } } */ diff --git a/gcc/testsuite/gcc.dg/tree-ssa/recip-4.c b/gcc/testsuite/gcc.dg/tree-ssa/recip-4.c new file mode 100644 index 00000000000..20d7681d28d --- /dev/null +++ b/gcc/testsuite/gcc.dg/tree-ssa/recip-4.c @@ -0,0 +1,45 @@ +/* { dg-do compile } */ +/* { dg-options "-O1 -fno-trapping-math -funsafe-math-optimizations -fdump-tree-recip" } */ + +/* based on the test case in pr23109 */ + +double F[2] = { 0., 0. }, e = 0.; + +/* Nope, we cannot prove the optimization is worthwhile in this case. */ +void f () +{ + int i; + double E, W, P, d; + + W = 1.; + d = 2.*e; + E = 1. - d; + + if( d > 0.01 ) + { + P = ( W < E ) ? (W - E)/d : (E - W)/d; + F[i] += P; + } +} + +/* We also cannot prove the optimization is worthwhile in this case. */ +float g () +{ + int i; + double E, W, P, d; + + W = 1.; + d = 2.*e; + E = 1. - d; + + if( d > 0.01 ) + { + P = ( W < E ) ? (W - E)/d : (E - W)/d; + F[i] += P; + } + + return 1.0 / d; +} + +/* { dg-final { scan-tree-dump-not "reciptmp" "recip" } } */ +/* { dg-final { cleanup-tree-dump "recip" } } */ diff --git a/gcc/testsuite/gcc.dg/tree-ssa/recip-5.c b/gcc/testsuite/gcc.dg/tree-ssa/recip-5.c new file mode 100644 index 00000000000..bcbd1839cbc --- /dev/null +++ b/gcc/testsuite/gcc.dg/tree-ssa/recip-5.c @@ -0,0 +1,31 @@ +/* { dg-options "-O1 -funsafe-math-optimizations -ftrapping-math -fdump-tree-recip -fdump-tree-optimized" } */ +/* { dg-do compile } */ + +/* Test the reciprocal optimizations together with trapping math. */ + +extern int f2(); + +double f1(double y, double z, double w, double j, double k) +{ + double b, c, d, e, f, g; + + if (f2 ()) + /* inserts one division here */ + b = 1 / y, c = z / y, d = j / y; + else + /* one division here */ + b = 3 / y, c = w / y, d = k / y; + + /* and one here, that should be removed afterwards but is not right now */ + e = b / y; + f = c / y; + g = d / y; + + return e + f + g; +} + +/* { dg-final { scan-tree-dump-times " / " 3 "recip" } } */ +/* { dg-final { scan-tree-dump-times " / " 2 "optimized" { xfail *-*-* } } } */ +/* { dg-final { cleanup-tree-dump "recip" } } */ +/* { dg-final { cleanup-tree-dump "optimized" } } */ + diff --git a/gcc/testsuite/gcc.dg/tree-ssa/recip-6.c b/gcc/testsuite/gcc.dg/tree-ssa/recip-6.c new file mode 100644 index 00000000000..60fefd01da5 --- /dev/null +++ b/gcc/testsuite/gcc.dg/tree-ssa/recip-6.c @@ -0,0 +1,26 @@ +/* { dg-options "-O1 -funsafe-math-optimizations -fno-trapping-math -fdump-tree-recip" } */ +/* { dg-do compile } */ + +/* Test inserting in a block that does not contain a division. */ + +extern int f2(); + +double f1(double y, double z, double w) +{ + double b, c, d, e, f; + + if (g ()) + b = 1 / y, c = z / y; + else + b = 3 / y, c = w / y; + + d = b / y; + e = c / y; + f = 1 / y; + + return d + e + f; +} + +/* { dg-final { scan-tree-dump-times " / " 1 "recip" } } */ +/* { dg-final { cleanup-tree-dump "recip" } } */ + diff --git a/gcc/testsuite/gcc.dg/tree-ssa/recip-7.c b/gcc/testsuite/gcc.dg/tree-ssa/recip-7.c new file mode 100644 index 00000000000..af1bf3c008b --- /dev/null +++ b/gcc/testsuite/gcc.dg/tree-ssa/recip-7.c @@ -0,0 +1,27 @@ +/* { dg-options "-O1 -funsafe-math-optimizations -fno-trapping-math -fdump-tree-recip" } */ +/* { dg-do compile } */ + +/* Test inserting in a block that does not contain a division. */ + +extern double h(); + +double f(int x, double z, double w) +{ + double b, c, d, e, f; + double y = h (); + + if (x) + b = 1 / y, c = z / y; + else + b = 3 / y, c = w / y; + + d = b / y; + e = c / y; + f = 1 / y; + + return d + e + f; +} + +/* { dg-final { scan-tree-dump-times " / " 1 "recip" } } */ +/* { dg-final { cleanup-tree-dump "recip" } } */ + diff --git a/gcc/tree-cfg.c b/gcc/tree-cfg.c index 342b9d2fdb1..40a29bc7c2e 100644 --- a/gcc/tree-cfg.c +++ b/gcc/tree-cfg.c @@ -1389,6 +1389,30 @@ tree_merge_blocks (basic_block a, basic_block b) } +/* Return the one of two successors of BB that is not reachable by a + reached by a complex edge, if there is one. Else, return BB. We use + this in optimizations that use post-dominators for their heuristics, + to catch the cases in C++ where function calls are involved. */ + +basic_block +single_noncomplex_succ (basic_block bb) +{ + edge e0, e1; + if (EDGE_COUNT (bb->succs) != 2) + return bb; + + e0 = EDGE_SUCC (bb, 0); + e1 = EDGE_SUCC (bb, 1); + if (e0->flags & EDGE_COMPLEX) + return e1->dest; + if (e1->flags & EDGE_COMPLEX) + return e0->dest; + + return bb; +} + + + /* Walk the function tree removing unnecessary statements. * Empty statement nodes are removed diff --git a/gcc/tree-flow.h b/gcc/tree-flow.h index afdd1437913..0c685522bd9 100644 --- a/gcc/tree-flow.h +++ b/gcc/tree-flow.h @@ -487,6 +487,7 @@ extern bool is_ctrl_stmt (tree); extern bool is_ctrl_altering_stmt (tree); extern bool computed_goto_p (tree); extern bool simple_goto_p (tree); +extern basic_block single_noncomplex_succ (basic_block bb); extern void tree_dump_bb (basic_block, FILE *, int); extern void debug_tree_bb (basic_block); extern basic_block debug_tree_bb_n (int); diff --git a/gcc/tree-ssa-math-opts.c b/gcc/tree-ssa-math-opts.c index 456043fc3f4..825c0bdec61 100644 --- a/gcc/tree-ssa-math-opts.c +++ b/gcc/tree-ssa-math-opts.c @@ -35,7 +35,55 @@ Software Foundation, 51 Franklin Street, Fifth Floor, Boston, MA z = z * rmodulus; We do this for loop invariant divisors, and with this pass whenever - we notice that a division has the same divisor multiple times. */ + we notice that a division has the same divisor multiple times. + + Of course, like in PRE, we don't insert a division if a dominator + already has one. However, this cannot be done as an extension of + PRE for several reasons. + + First of all, with some experiments it was found out that the + transformation is not always useful if there are only two divisions + hy the same divisor. This is probably because modern processors + can pipeline the divisions; on older, in-order processors it should + still be effective to optimize two divisions by the same number. + We make this a param, and it shall be called N in the remainder of + this comment. + + Second, if trapping math is active, we have less freedom on where + to insert divisions: we can only do so in basic blocks that already + contain one. (If divisions don't trap, instead, we can insert + divisions elsewhere, which will be in blocks that are common dominators + of those that have the division). + + We really don't want to compute the reciprocal unless a division will + be found. To do this, we won't insert the division in a basic block + that has less than N divisions *post-dominating* it. + + The algorithm constructs a subset of the dominator tree, holding the + blocks containing the divisions and the common dominators to them, + and walk it twice. The first walk is in post-order, and it annotates + each block with the number of divisions that post-dominate it: this + gives information on where divisions can be inserted profitably. + The second walk is in pre-order, and it inserts divisions as explained + above, and replaces divisions by multiplications. + + In the best case, the cost of the pass is O(n_statements). In the + worst-case, the cost is due to creating the dominator tree subset, + with a cost of O(n_basic_blocks ^ 2); however this can only happen + for n_statements / n_basic_blocks statements. So, the amortized cost + of creating the dominator tree subset is O(n_basic_blocks) and the + worst-case cost of the pass is O(n_statements * n_basic_blocks). + + More practically, the cost will be small because there are few + divisions, and they tend to be in the same basic block, so insert_bb + is called very few times. + + If we did this using domwalk.c, an efficient implementation would have + to work on all the variables in a single pass, because we could not + work on just a subset of the dominator tree, as we do now, and the + cost would also be something like O(n_statements * n_basic_blocks). + The data structures would be more complex in order to work on all the + variables in a single pass. */ #include "config.h" #include "system.h" @@ -47,6 +95,348 @@ Software Foundation, 51 Franklin Street, Fifth Floor, Boston, MA #include "real.h" #include "timevar.h" #include "tree-pass.h" +#include "alloc-pool.h" +#include "basic-block.h" +#include "target.h" + + +/* This structure represents one basic block that either computes a + division, or is a common dominator for basic block that compute a + division. */ +struct occurrence { + /* The basic block represented by this structure. */ + basic_block bb; + + /* If non-NULL, the SSA_NAME holding the definition for a reciprocal + inserted in BB. */ + tree recip_def; + + /* If non-NULL, the MODIFY_EXPR for a reciprocal computation that + was inserted in BB. */ + tree recip_def_stmt; + + /* Pointer to a list of "struct occurrence"s for blocks dominated + by BB. */ + struct occurrence *children; + + /* Pointer to the next "struct occurrence"s in the list of blocks + sharing a common dominator. */ + struct occurrence *next; + + /* The number of divisions that are in BB before compute_merit. The + number of divisions that are in BB or post-dominate it after + compute_merit. */ + int num_divisions; + + /* True if the basic block has a division, false if it is a common + dominator for basic blocks that do. If it is false and trapping + math is active, BB is not a candidate for inserting a reciprocal. */ + bool bb_has_division; +}; + + +/* The instance of "struct occurrence" representing the highest + interesting block in the dominator tree. */ +static struct occurrence *occ_head; + +/* Allocation pool for getting instances of "struct occurrence". */ +static alloc_pool occ_pool; + + + +/* Allocate and return a new struct occurrence for basic block BB, and + whose children list is headed by CHILDREN. */ +static struct occurrence * +occ_new (basic_block bb, struct occurrence *children) +{ + struct occurrence *occ; + + occ = bb->aux = pool_alloc (occ_pool); + memset (occ, 0, sizeof (struct occurrence)); + + occ->bb = bb; + occ->children = children; + return occ; +} + + +/* Insert NEW_OCC into our subset of the dominator tree. P_HEAD points to a + list of "struct occurrence"s, one per basic block, having IDOM as + their common dominator. + + We try to insert NEW_OCC as deep as possible in the tree, and we also + insert any other block that is a common dominator for BB and one + block already in the tree. */ + +static void +insert_bb (struct occurrence *new_occ, basic_block idom, + struct occurrence **p_head) +{ + struct occurrence *occ, **p_occ; + + for (p_occ = p_head; (occ = *p_occ) != NULL; ) + { + basic_block bb = new_occ->bb, occ_bb = occ->bb; + basic_block dom = nearest_common_dominator (CDI_DOMINATORS, occ_bb, bb); + if (dom == bb) + { + /* BB dominates OCC_BB. OCC becomes NEW_OCC's child: remove OCC + from its list. */ + *p_occ = occ->next; + occ->next = new_occ->children; + new_occ->children = occ; + + /* Try the next block (it may as well be dominated by BB). */ + } + + else if (dom == occ_bb) + { + /* OCC_BB dominates BB. Tail recurse to look deeper. */ + insert_bb (new_occ, dom, &occ->children); + return; + } + + else if (dom != idom) + { + gcc_assert (!dom->aux); + + /* There is a dominator between IDOM and BB, add it and make + two children out of NEW_OCC and OCC. First, remove OCC from + its list. */ + *p_occ = occ->next; + new_occ->next = occ; + occ->next = NULL; + + /* None of the previous blocks has DOM as a dominator: if we tail + recursed, we would reexamine them uselessly. Just switch BB with + DOM, and go on looking for blocks dominated by DOM. */ + new_occ = occ_new (dom, new_occ); + } + + else + { + /* Nothing special, go on with the next element. */ + p_occ = &occ->next; + } + } + + /* No place was found as a child of IDOM. Make BB a sibling of IDOM. */ + new_occ->next = *p_head; + *p_head = new_occ; +} + +/* Register that we found a division in BB. */ + +static inline void +register_division_in (basic_block bb) +{ + struct occurrence *occ; + + occ = (struct occurrence *) bb->aux; + if (!occ) + { + occ = occ_new (bb, NULL); + insert_bb (occ, ENTRY_BLOCK_PTR, &occ_head); + } + + occ->bb_has_division = true; + occ->num_divisions++; +} + + +/* Compute the number of divisions that postdominate each block in OCC and + its children. */ + +static void +compute_merit (struct occurrence *occ) +{ + struct occurrence *occ_child; + basic_block dom = occ->bb; + + for (occ_child = occ->children; occ_child; occ_child = occ_child->next) + { + basic_block bb; + if (occ_child->children) + compute_merit (occ_child); + + if (flag_exceptions) + bb = single_noncomplex_succ (dom); + else + bb = dom; + + if (dominated_by_p (CDI_POST_DOMINATORS, bb, occ_child->bb)) + occ->num_divisions += occ_child->num_divisions; + } +} + + +/* Return whether USE_STMT is a floating-point division by DEF. */ +static inline bool +is_division_by (tree use_stmt, tree def) +{ + return TREE_CODE (use_stmt) == MODIFY_EXPR + && TREE_CODE (TREE_OPERAND (use_stmt, 1)) == RDIV_EXPR + && TREE_OPERAND (TREE_OPERAND (use_stmt, 1), 1) == def; +} + +/* Walk the subset of the dominator tree rooted at OCC, setting the + RECIP_DEF field to a definition of 1.0 / DEF that can be used in + the given basic block. The field may be left NULL, of course, + if it is not possible or profitable to do the optimization. + + DEF_BSI is an iterator pointing at the statement defining DEF. + If RECIP_DEF is set, a dominator already has a computation that can + be used. */ + +static void +insert_reciprocals (block_stmt_iterator *def_bsi, struct occurrence *occ, + tree def, tree recip_def, int threshold) +{ + tree type, new_stmt; + block_stmt_iterator bsi; + struct occurrence *occ_child; + + if (!recip_def + && (occ->bb_has_division || !flag_trapping_math) + && occ->num_divisions >= threshold) + { + /* Make a variable with the replacement and substitute it. */ + type = TREE_TYPE (def); + recip_def = make_rename_temp (type, "reciptmp"); + new_stmt = build2 (MODIFY_EXPR, void_type_node, recip_def, + fold_build2 (RDIV_EXPR, type, + build_real (type, dconst1), def)); + + + if (occ->bb_has_division) + { + /* Case 1: insert before an existing division. */ + bsi = bsi_after_labels (occ->bb); + while (!bsi_end_p (bsi) && !is_division_by (bsi_stmt (bsi), def)) + bsi_next (&bsi); + + bsi_insert_before (&bsi, new_stmt, BSI_SAME_STMT); + } + else if (def_bsi && occ->bb == def_bsi->bb) + { + /* Case 2: insert right after the definition. Note that this will + never happen if the definition statement can throw, because in + that case the sole successor of the statement's basic block will + dominate all the uses as well. */ + bsi_insert_after (def_bsi, new_stmt, BSI_NEW_STMT); + } + else + { + /* Case 3: insert in a basic block not containing defs/uses. */ + bsi = bsi_after_labels (occ->bb); + bsi_insert_before (&bsi, new_stmt, BSI_SAME_STMT); + } + + occ->recip_def_stmt = new_stmt; + } + + occ->recip_def = recip_def; + for (occ_child = occ->children; occ_child; occ_child = occ_child->next) + insert_reciprocals (def_bsi, occ_child, def, recip_def, threshold); +} + + +/* Replace the division at USE_P with a multiplication by the reciprocal, if + possible. */ + +static inline void +replace_reciprocal (use_operand_p use_p) +{ + tree use_stmt = USE_STMT (use_p); + basic_block bb = bb_for_stmt (use_stmt); + struct occurrence *occ = (struct occurrence *) bb->aux; + + if (occ->recip_def && use_stmt != occ->recip_def_stmt) + { + TREE_SET_CODE (TREE_OPERAND (use_stmt, 1), MULT_EXPR); + SET_USE (use_p, occ->recip_def); + fold_stmt_inplace (use_stmt); + update_stmt (use_stmt); + } +} + + +/* Free OCC and return one more "struct occurrence" to be freed. */ + +static struct occurrence * +free_bb (struct occurrence *occ) +{ + struct occurrence *child, *next; + + /* First get the two pointers hanging off OCC. */ + next = occ->next; + child = occ->children; + occ->bb->aux = NULL; + pool_free (occ_pool, occ); + + /* Now ensure that we don't recurse unless it is necessary. */ + if (!child) + return next; + else + { + while (next) + next = free_bb (next); + + return child; + } +} + + +/* Look for floating-point divisions among DEF's uses, and try to + replace them by multiplications with the reciprocal. Add + as many statements computing the reciprocal as needed. + + DEF must be a GIMPLE register of a floating-point type. */ + +static void +execute_cse_reciprocals_1 (block_stmt_iterator *def_bsi, tree def) +{ + use_operand_p use_p; + imm_use_iterator use_iter; + struct occurrence *occ; + int count = 0, threshold; + + gcc_assert (FLOAT_TYPE_P (TREE_TYPE (def)) && is_gimple_reg (def)); + + FOR_EACH_IMM_USE_FAST (use_p, use_iter, def) + { + tree use_stmt = USE_STMT (use_p); + if (is_division_by (use_stmt, def)) + { + register_division_in (bb_for_stmt (use_stmt)); + count++; + } + } + + /* Do the expensive part only if we can hope to optimize something. */ + threshold = targetm.min_divisions_for_recip_mul (TYPE_MODE (TREE_TYPE (def))); + if (count >= threshold) + { + for (occ = occ_head; occ; occ = occ->next) + { + compute_merit (occ); + insert_reciprocals (def_bsi, occ, def, NULL, threshold); + } + + FOR_EACH_IMM_USE_SAFE (use_p, use_iter, def) + { + tree use_stmt = USE_STMT (use_p); + if (is_division_by (use_stmt, def)) + replace_reciprocal (use_p); + } + } + + for (occ = occ_head; occ; ) + occ = free_bb (occ); + + occ_head = NULL; +} + static bool gate_cse_reciprocals (void) @@ -54,135 +444,58 @@ gate_cse_reciprocals (void) return optimize && !optimize_size && flag_unsafe_math_optimizations; } -/* Where to put the statement computing a reciprocal. */ -enum place_reciprocal -{ - PR_BEFORE_BSI, /* Put it using bsi_insert_before. */ - PR_AFTER_BSI, /* Put it using bsi_insert_after. */ - PR_ON_ENTRY_EDGE /* Put it on the edge between the entry - and the first basic block. */ -}; - -/* Check if DEF's uses include more than one floating-point division, - and if so replace them by multiplications with the reciprocal. Add - the statement computing the reciprocal according to WHERE. - - Does not check the type of DEF, nor that DEF is a GIMPLE register. - This is done in the caller for speed, because otherwise this routine - would be called for every definition and phi node. */ -static void -execute_cse_reciprocals_1 (block_stmt_iterator *bsi, tree def, - enum place_reciprocal where) -{ - use_operand_p use_p; - imm_use_iterator use_iter; - tree t, new_stmt, type; - int count = 0; - bool ok = !flag_trapping_math; - - /* Find uses. */ - FOR_EACH_IMM_USE_FAST (use_p, use_iter, def) - { - tree use_stmt = USE_STMT (use_p); - if (TREE_CODE (use_stmt) == MODIFY_EXPR - && TREE_CODE (TREE_OPERAND (use_stmt, 1)) == RDIV_EXPR - && TREE_OPERAND (TREE_OPERAND (use_stmt, 1), 1) == def) - { - ++count; - /* Check if this use post-dominates the insertion point. */ - if (ok || dominated_by_p (CDI_POST_DOMINATORS, bsi->bb, - bb_for_stmt (use_stmt))) - ok = true; - } - if (count >= 2 && ok) - break; - } - - if (count < 2 || !ok) - return; - - /* Make a variable with the replacement and substitute it. */ - type = TREE_TYPE (def); - t = make_rename_temp (type, "reciptmp"); - new_stmt = build2 (MODIFY_EXPR, void_type_node, t, - fold_build2 (RDIV_EXPR, type, build_real (type, dconst1), - def)); - - if (where == PR_BEFORE_BSI) - bsi_insert_before (bsi, new_stmt, BSI_SAME_STMT); - else if (where == PR_AFTER_BSI) - bsi_insert_after (bsi, new_stmt, BSI_NEW_STMT); - else if (where == PR_ON_ENTRY_EDGE) - bsi_insert_on_edge (single_succ_edge (ENTRY_BLOCK_PTR), new_stmt); - else - gcc_unreachable (); - - FOR_EACH_IMM_USE_SAFE (use_p, use_iter, def) - { - tree use_stmt = USE_STMT (use_p); - if (use_stmt != new_stmt - && TREE_CODE (use_stmt) == MODIFY_EXPR - && TREE_CODE (TREE_OPERAND (use_stmt, 1)) == RDIV_EXPR - && TREE_OPERAND (TREE_OPERAND (use_stmt, 1), 1) == def) - { - TREE_SET_CODE (TREE_OPERAND (use_stmt, 1), MULT_EXPR); - SET_USE (use_p, t); - } - } -} +/* Go through all the floating-point SSA_NAMEs, and call + execute_cse_reciprocals_1 on each of them. */ static void execute_cse_reciprocals (void) { basic_block bb; tree arg; - if (flag_trapping_math) - calculate_dominance_info (CDI_POST_DOMINATORS); + occ_pool = create_alloc_pool ("dominators for recip", + sizeof (struct occurrence), + n_basic_blocks / 3 + 1); - if (single_succ_p (ENTRY_BLOCK_PTR)) - for (arg = DECL_ARGUMENTS (cfun->decl); arg; arg = TREE_CHAIN (arg)) - if (default_def (arg)) - { - block_stmt_iterator bsi; - bsi = bsi_start (single_succ (ENTRY_BLOCK_PTR)); - execute_cse_reciprocals_1 (&bsi, default_def (arg), - PR_ON_ENTRY_EDGE); - } + calculate_dominance_info (CDI_DOMINATORS | CDI_POST_DOMINATORS); + +#ifdef ENABLE_CHECKING + FOR_EACH_BB (bb) + gcc_assert (!bb->aux); +#endif + + for (arg = DECL_ARGUMENTS (cfun->decl); arg; arg = TREE_CHAIN (arg)) + if (default_def (arg) + && FLOAT_TYPE_P (TREE_TYPE (arg)) + && is_gimple_reg (arg)) + execute_cse_reciprocals_1 (NULL, default_def (arg)); FOR_EACH_BB (bb) { block_stmt_iterator bsi; tree phi, def; - for (bsi = bsi_start (bb); - !bsi_end_p (bsi) && TREE_CODE (bsi_stmt (bsi)) == LABEL_EXPR; - bsi_next (&bsi)) - ; for (phi = phi_nodes (bb); phi; phi = PHI_CHAIN (phi)) { def = PHI_RESULT (phi); if (FLOAT_TYPE_P (TREE_TYPE (def)) && is_gimple_reg (def)) - execute_cse_reciprocals_1 (&bsi, def, PR_BEFORE_BSI); + execute_cse_reciprocals_1 (NULL, def); } - for (; !bsi_end_p (bsi); bsi_next (&bsi)) + for (bsi = bsi_after_labels (bb); !bsi_end_p (bsi); bsi_next (&bsi)) { tree stmt = bsi_stmt (bsi); if (TREE_CODE (stmt) == MODIFY_EXPR && (def = SINGLE_SSA_TREE_OPERAND (stmt, SSA_OP_DEF)) != NULL && FLOAT_TYPE_P (TREE_TYPE (def)) && TREE_CODE (def) == SSA_NAME) - execute_cse_reciprocals_1 (&bsi, def, PR_AFTER_BSI); + execute_cse_reciprocals_1 (&bsi, def); } } - if (flag_trapping_math) - free_dominance_info (CDI_POST_DOMINATORS); - - if (single_succ_p (ENTRY_BLOCK_PTR)) - bsi_commit_one_edge_insert (single_succ_edge (ENTRY_BLOCK_PTR), NULL); + free_dominance_info (CDI_DOMINATORS | CDI_POST_DOMINATORS); + free_alloc_pool (occ_pool); } struct tree_opt_pass pass_cse_reciprocals =