2011-11-17 Andrew MacLeod <amacleod@redhat.com>
* builtins.c (expand_builtin): Remove 4th parameter representing
weak/strong mode when __atomic_compare_exchange becomes a library call.
* gcc.dg/atomic-generic-aux.c (__atomic_compare_exchange): Fail if
memory model parameters don't match expected values.
* gcc.dg/atomic-generic.c: Pass specific memory model parameters to
__atomic_compare_exchange.
* gcc.dg/atomic-noinline.c: Pass specific memory model parameters to
__atomic_compare_exchange_n.
* gcc.dg/atomic-noinline-aux.c (__atomic_compare_exchange_2): Remove
weak/strong parameter and fail if memory models aren't correct.
From-SVN: r181453
* builtins.c (expand_builtin_mem_thread_fence): Remove.
(expand_builtin_mem_signal_fence): Remove.
(expand_builtin_atomic_thread_fence): Use expand_mem_thread_fence.
(expand_builtin_sync_synchronize): Likewise.
(expand_builtin_atomic_signal_fence): Use expand_mem_signal_fence.
* optabs.c (expand_asm_memory_barrier): Split out from
expand_builtin_mem_signal_fence.
(expand_mem_thread_fence): New, a combination of code from
expand_builtin_mem_thread_fence and expand_builtin_sync_synchronize.
(expand_mem_signal_fence): Moved and renamed from
expand_builtin_mem_signal_fence.
(expand_atomic_exchange): Use expand_mem_thread_fence.
(expand_atomic_load, expand_atomic_store): Likewise.
* expr.h, optabs.h: Update decls.
From-SVN: r181451
* builtins.c (expand_builtin_compare_and_swap): If target is const0,
don't pass the target to expand_atomic_compare_and_swap.
(expand_builtin_atomic_compare_exchange): Likewise.
From-SVN: r181323
PR rtl-optimization/51040
* optabs.c (expand_atomic_fetch_op): Patchup code for NAND should be AND
followed by NOT.
* builtins.c (expand_builtin_atomic_fetch_op): Patchup code for NAND
should be AND followed by NOT.
* testsuite/gcc.dg/atomic-noinline[-aux].c: Test no-inline NAND and
patchup code.
From-SVN: r181259
This allows a target which implements the __sync interfaces
in libgcc to continue to use them transparently with the
new __atomic builtins.
It is assumed that these libgcc routines DO NOT use spinlocks.
This is true of all extant libgcc instances.
* optabs.h (OTI_sync_compare_and_swap, OTI_sync_lock_test_and_set,
OTI_sync_old_add, OTI_sync_old_sub, OTI_sync_old_ior,
OTI_sync_old_and, OTI_sync_old_xor, OTI_sync_old_nand,
OTI_sync_new_add, OTI_sync_new_sub, OTI_sync_new_ior,
OTI_sync_new_and, OTI_sync_new_xor, OTI_sync_new_nand): Move and
rename from the direct_optab_index enum.
(sync_compare_and_swap_optab, sync_lock_test_and_set_optab,
sync_old_add_optab, sync_old_sub_optab, sync_old_ior_optab,
sync_old_and_optab, sync_old_xor_optab, sync_old_nand_optab,
sync_new_add_optab, sync_new_sub_optab, sync_new_ior_optab,
sync_new_and_optab, sync_new_xor_optab, sync_new_nand_optab): Read
from the optab_table, not the direct_optab_table.
(init_sync_libfuncs): Declare.
(can_compare_and_swap_p): Update parameters.
* optabs.c (init_sync_libfuncs_1, init_sync_libfuncs): New.
(can_compare_and_swap_p): Add allow_libcall parameter; if true,
test for the legacy compare-and-swap libcall.
(expand_atomic_exchange): Use the legacy test-and-set libcall.
(expand_atomic_compare_and_swap): Use the legacy CAS libcall.
(struct atomic_op_functions): Update for optab type changes.
(maybe_emit_op): Likewise.
(expand_atomic_fetch_op): Use the legacy fetch-op libcalls.
* builtins.c (fold_builtin_atomic_always_lock_free): Update call
to can_compare_and_swap_p.
* omp-low.c (expand_omp_atomic_fetch_op): Likewise.
(expand_omp_atomic_pipeline): Likewise.
* genopinit.c (optabs): Make sync_old_*_optab, sync_new_*_optab,
sync_compare_and_swap_optab, sync_lock_test_and_set_optab regular
optabs.
From-SVN: r181134
2011-11-07 Andrew MacLeod <amacleod@redhat.com>
libstdc++-v3
* include/bits/atomic_base.h (atomic_thread_fence): Call builtin.
(atomic_signal_fence): Call builtin.
(atomic_flag::test_and_set): Call __atomic_exchange when it is lockfree,
otherwise fall back to call __sync_lock_test_and_set.
(atomic_flag::clear): Call __atomic_store when it is lockfree,
otherwise fall back to call __sync_lock_release.
gcc
* doc/extend.texi: Docuemnt behaviour change for __atomic_exchange and
__atomic_store.
* optabs.c (expand_atomic_exchange): Expand to __sync_lock_test_and_set
only when originated from that builtin.
(expand_atomic_store): Expand to __sync_lock_release when originated
from that builtin.
* builtins.c (expand_builtin_sync_lock_test_and_set): Add flag that
expand_atomic_exchange call originated from here.
(expand_builtin_sync_lock_release): Add flag that expand_atomic_store
call originated from here.
(expand_builtin_atomic_exchange): Add origination flag.
(expand_builtin_atomic_store): Add origination flag.
* expr.h (expand_atomic_exchange, expand_atomic_store): Add boolean
parameters to indicate implementation fall back options.
From-SVN: r181111
* tree.h (avoid_folding_inline_builtin): New prototype.
* builtins.c (avoid_folding_inline_builtin): No longer static.
* gimple-fold.c (gimple_fold_builtin): Give up if
avoid_folding_inline_builtin returns true.
From-SVN: r179614
PR tree-optimization/50604
* builtins.c (fold_builtin_strcpy, fold_builtin_stpcpy,
fold_builtin_strncpy, fold_builtin_stxcpy_chk): Ensure
last argument to memcpy has size_type_node type instead of
ssizetype.
* tree-ssa-strlen.c (handle_builtin_memcpy): Use size_type_node
instead of TREE_TYPE (len) as type for newlen.
* gcc.dg/pr50604.c: New test.
From-SVN: r179508
2011-08-17 Richard Guenther <rguenther@suse.de>
* tree.h (convert_to_ptrofftype_loc): New function.
(convert_to_ptrofftype): Define.
* builtins.c (expand_builtin_bzero): Use size_type_node.
(fold_builtin_bzero): Likewise.
(std_gimplify_va_arg_expr): Build the BIT_AND_EXPR on the pointer.
* c-typeck.c (build_unary_op): Use convert_to_ptrofftype_loc.
* cgraphunit.c (thunk_adjust): Use fold_build_pointer_plus_loc.
(cgraph_redirect_edge_call_stmt_to_callee): Use size_int.
* expr.c (expand_expr_addr_expr_1): Use fold_build_pointer_plus.
* fold-const.c (build_range_check): Negate using the original
type.
(fold_unary_loc): Use fold_build_pointer_plus_loc.
* gimple-fold.c (gimple_adjust_this_by_delta): Use
convert_to_ptrofftype.
* gimplify.c (gimplify_self_mod_expr): Likewise.
* graphite-clast-to-gimple.c (clast_to_gcc_expression): Likewise.
(graphite_create_new_loop_guard): Likewise.
* graphite-sese-to-poly.c (my_long_long): Remove.
(scop_ivs_can_be_represented): Adjust.
* tree-cfg.c (verify_gimple_assign_unary): Use ptrofftype_p.
* tree-chrec.c (chrec_fold_plus_1): Use fold_build_pointer_plus.
* tree-loop-distribution.c (build_size_arg_loc): Use
size_type_node.
(generate_memset_zero): Simplify.
* tree-mudflap.c: Use fold_convert, not convert.
* tree-predcom.c (suitable_reference_p): Expand DR_OFFSET in
its own type.
(determine_offset): Likewise for DR_STEP.
(valid_initializer_p): Likewise.
* tree-profile.c (prepare_instrumented_value): Convert the pointer
to an integer type of same size.
* tree-scalar-evolution.c (interpret_rhs_expr): Do not refer
to sizetype without need.
* tree-ssa-address.c (tree_mem_ref_addr): Likewise.
* tree-ssa-loop-ivopts.c (find_bivs): Use convert_to_ptrofftype.
* tree-ssa-loop-manip.c (create_iv): Likewise.
(determine_exit_conditions): Adjust comment.
* tree-ssa-pre.c (create_expression_by_pieces): Use
convert_to_ptrofftype.
* tree-ssa-structalias.c (get_constraint_for_1): Likewise.
* varasm.c (array_size_for_constructor): Compute using double_ints.
From-SVN: r177828
* builtins.c (expand_builtin_memcmp): Do not use cmpstrnsi
pattern.
* doc/md.texi (cmpstrn): Note that the comparison stops if both
fetched bytes are zero.
(cmpstr): Likewise.
(cmpmem): Note that the comparison does not stop if both of the
fetched bytes are zero.
From-SVN: r177701
gcc/
* doc/rtl.texi (MEM_OFFSET_KNOWN_P): Document.
(MEM_OFFSET): Change from returning an rtx to returning a
HOST_WIDE_INT.
* rtl.h (MEM_OFFSET_KNOWN_P): New macro.
(MEM_OFFSET): Return a HOST_WIDE_INT rather than an rtx.
* emit-rtl.h (set_mem_offset): Take a HOST_WIDE_INT rather than an rtx.
(clear_mem_offset): Declare.
* alias.c (ao_ref_from_mem): Adjust uses of MEM_OFFSET, using
MEM_OFFSET_KNOWN_P to test whether the offset is known, and
MEM_OFFSET to get a HOST_WIDE_INT offset.
(nonoverlapping_memrefs_p): Likewise. Adjust calls to...
(adjust_offset_for_component_ref): Take a bool "known_p"
parameter and a HOST_WIDE_INT "offset" parameter.
* builtins.c (get_memory_rtx): As for ao_ref_from_mem.
Adjust calls to set_mem_offset, passing a HOST_WIDE_INT rather
than an rtx. Use clear_mem_offset to clear the offset.
* cfgcleanup.c (merge_memattrs): Likewise.
* dwarf2out.c (tls_mem_loc_descriptor): Likewise.
* function.c (assign_parm_find_stack_rtl): Likewise.
(assign_parm_setup_stack): Likewise.
* print-rtl.c (print_rtx): Likewise.
* reload.c (find_reloads_subreg_address): Likewise.
* simplify-rtx.c (delegitimize_mem_from_attrs): Likewise.
* var-tracking.c (INT_MEM_OFFSET): Likewise.
* emit-rtl.c (set_reg_attrs_from_value): Likewise.
(get_mem_align_offset): Likewise.
(set_mem_offset): Take a HOST_WIDE_INT rather than an rtx.
(clear_mem_offset): New function.
* config/mips/mips.c (r10k_safe_mem_expr_p): Take a HOST_WIDE_INT
offset rather than an rtx. Assume both the expressio and offset
are available.
(r10k_needs_protection_p_1): Update accordingly, checking the
expression and offset availability here instead.
From-SVN: r176477
gcc/
PR tree-optimization/49545
* builtins.c (get_object_alignment_1): Update function comment.
Do not use DECL_ALIGN for functions, but test
TARGET_PTRMEMFUNC_VBIT_LOCATION instead.
* fold-const.c (get_pointer_modulus_and_residue): Don't check
for functions here.
* tree-ssa-ccp.c (get_value_from_alignment): Likewise.
gcc/testsuite/
* gcc.dg/torture/pr49169.c: Restrict to ARM and MIPS targets.
From-SVN: r175627
2011-06-21 Andrew MacLeod <amacleod@redhat.com>
* builtins.c: Add sync_ or SYNC__ to builtin names.
* sync-builtins.def: Add sync_ or SYNC__ to builtin names.
* omp-low.c: Add sync_ or SYNC__ to builtin names.
* c-family/c-common.c: Add sync_ or SYNC__ to builtin names.
* c-family/c-omp.c: Add sync_ or SYNC__ to builtin names.
* java/builtins.c: Add sync_ or SYNC__ to builtin names.
* java/expr.c: Add sync_ or SYNC__ to builtin names.
* cp/semantics.c: Add sync_ or SYNC__ to builtin names.
* fortran/trans-openmp.c: Add sync_ or SYNC__ to builtin names.
* fortran/trans-stmt.c: Add sync_ or SYNC__ to builtin names.
* fortran/trans-decl.c: Add sync_ or SYNC__ to builtin names.
From-SVN: r175270
PR middle-end/49489
* builtins.c (expand_builtin_unop): Call expand_unop with 0 as
unsignedp argument instead of 1 for clrsb_optab.
(fold_builtin_bitop): Fix masking for width > HOST_BITS_PER_WIDE_INT
and < 2 * HOST_BITS_PER_WIDE_INT. Optimize BUILT_IN_CLRSB*.
(fold_builtin_1): Call fold_builtin_binop for BUILT_IN_CLRSB*.
* optabs.c (widen_leading): Call widen_operand and expand_unop
with 0 as unsignedp argument instead of 1 for clrsb_optab.
(expand_unop): Subtract difference of mode sizes also for
clrsb_optab.
From-SVN: r175265
2011-06-14 H.J. Lu <hongjiu.lu@intel.com>
PR middle-end/47364
* builtins.c (expand_builtin_strlen): Expand strlen to Pmode
and properly handle result not in Pmode.
From-SVN: r175034
2011-05-23 Richard Guenther <rguenther@suse.de>
PR middle-end/15419
* builtins.c (fold_builtin_memory_op): Be less restrictive about
what pointer types we accept for folding.
* gcc.dg/memcpy-3.c: New testcase.
From-SVN: r174065
2011-04-29 Richard Guenther <rguenther@suse.de>
* builtins.c (fold_builtin_classify_type): Use integer_type_node
for the type of the result.
(fold_builtin_isascii): Likewise.
(fold_builtin_toascii): Use integer_type_node where appropriate.
(fold_builtin_logb): Likewise.
(fold_builtin_frexp): Likewise.
(fold_builtin_strstr): Likewise.
(fold_builtin_strpbrk): Likewise.
(fold_builtin_fputs): Likewise.
(fold_builtin_sprintf): Likewise.
(fold_builtin_snprintf): Likewise.
(fold_builtin_printf): Likewise.
(do_mpfr_remquo): Use a proper type for the assigned constant.
(do_mpfr_lgamma_r): Likewise.
* dwarf2out.c (resolve_one_addr): Use size_int.
* except.c (init_eh): Likewise.
(assign_filter_values): Use integer_type_node for filter values.
(sjlj_emit_dispatch_table): Use integer_type_node for dispatch
indices.
* tree-cfg.c (move_stmt_eh_region_tree_nr): Use integer_type_node
for EH region numbers.
* tree-vrp.c (simplify_div_or_mod_using_ranges): Use integer_type_node
for the shift amount.
From-SVN: r173167
gcc/
2011-03-28 Richard Sandiford <richard.sandiford@linaro.org>
* builtins.c (expand_builtin_memset_args): Use gen_int_mode
instead of GEN_INT.
From-SVN: r171632
PR tree-optimization/48129
* builtins.c (fold_builtin_snprintf): Convert to type of
built_in_decls[BUILT_IN_SNPRINTF] retval instead of
implicit_built_in_decls[BUILT_IN_SNPRINTF] retval.
From-SVN: r170985
PR middle-end/47917
* builtins.c (fold_builtin_snprintf): New function.
(fold_builtin_3): Call it for BUILT_IN_SNPRINTF.
(fold_builtin_4): Likewise.
* gcc.c-torture/execute/pr47917.c: New test.
* gcc.dg/pr47917.c: New test.
From-SVN: r170959
2011-02-02 Richard Guenther <rguenther@suse.de>
PR tree-optimization/47566
* builtins.c (builtin_save_expr): No SAVE_EXPR for SSA_NAMEs.
* gcc.dg/lto/20110201-1_0.c: New testcase.
From-SVN: r169518
