This is an automated email from the git hooks/post-receive script.

unknown user pushed a change to branch rsandifo/sve-rebase in repository gcc.

at 5b4cc1776f3 Use conditional internal functions in if-conversion

This branch includes the following new commits:

new bb7dfe3efef Add missing ECF_NOTHROW flags to internal.def new 40b69cff196 Improve ECF_NOTHROW flags for direct internal functions new 71f65b1e625 Add a mem_alias_size helper class new c25a25d1b67 Add a fixed_size_mode class new 953902585c4 Add a fixed_size_mode_pod class new c26e4c20f58 Widening optab cleanup new c2034dc61a3 Add a full_integral_type_p helper function new 59b2f8365a1 Add a partial_integral_type_p helper function new e03d86d45bb Make more use of paradoxical_subreg_p new 175387c7bfd Add a partial_subreg_p predicate new f662fc1f4ca Make more use of HWI_COMPUTABLE_MODE_P new 43ecd42adb4 Make more use of df_read_modify_subreg_p new 64b2f1d436f Make more use of subreg_offset_from_lsb new d1ae4048a50 Make more use of subreg_lowpart_offset new 89002777a1e Make more use of subreg_size_lowpart_offset new b4ac7958b40 Make more use of byte_lowpart_offset new eec23e50a03 Add subreg_memory_offset helper functions new 792ef1caf62 Add wider_subreg_mode helper functions new debbb7a286f Fix bogus CONST_WIDE_INT hash new 64a442ea773 Make more use of GET_MODE_UNIT_SIZE new a8131d3e69c Make more use of GET_MODE_UNIT_BITSIZE new 4ef2343e472 Make more use of GET_MODE_UNIT_PRECISION new 1d897d7eae4 Add helpers for shift count modes new 53974281653 Simplify pad_below implementation new 5407ba2baa4 Remove the frame size argument from function_prologue/epilogue new f522c3c8c51 Turn HARD_REGNO_CALL_PART_CLOBBERED into a target hook new 5f7383da3b5 Turn HARD_REGNO_MODE_OK into a target hook new 4614d7f6388 Turn MODES_TIEABLE_P into a target hook new 1a08aafee3b Turn SLOW_UNALIGNED_ACCESS into a target hook new 46fe110a43c Turn FUNCTION_ARG_PADDING into a target hook new 39895728dcf Use MACRO_MODE for HARD_REGNO_NREGS new 722833ab410 Use MACRO_MODE for SECONDARY_MEMORY_NEEDED_MODE new 16a67e21de4 Use MACRO_MODE for SECONDARY_MEMORY_NEEDED new eead50e23de Use MACRO_MODE for targhooks.c and address.h wrappers new 1544c45c5ad Use MACRO_MODE for CANNOT_CHANGE_MODE_CLASS new 029458d9574 Use MACRO_MODE for TRULY_NOOP_TRUNCATION_MODES_P new 3e5ad491e31 Use MACRO_MODE for FUNCTION_ARG_OFFSET new 03817ebe107 Pass rtx and index to read-md.c iterator routines new eca4e545ccb Factor out the mode handling in lower-subreg.c new 5864e8c8e6c Add a vect_get_num_copies helper routine new 56f47407d2e Add a vect_worthwhile_without_simd_p helper routine new fdb783eef05 Store VECTOR_CST_NELTS directly in tree node new 336cb967a84 Pass number of elements alongside tree* when constructing vectors new bb24a5f8f28 Add gimple_build_vector* helpers new 27629ec7bf0 Make more use of gimple-fold.h new b7a28e8d0e5 Add a vect_get_dr_size helper function new 9d349efd316 Let the target choose a vectorisation alignment new 02601d6761f Pass slp_index down to vect_analyze_stmt new e69da051066 Add rtx const vec_duplicate helpers new cd3fa6ea4d6 Allow vector CONSTs new 91a9d583e19 compare_values use in extract_range_from_multiplicative_op_1 new a9febfcdae3 Add missing int_cst_rangeN checks to tree-vrp.c new 14c61dc92de Tighten tree-ssa-ccp.c:get_value_for_expr condition new 171a671c219 Fix unguarded uses of tree_to_uhwi new 441245f3f75 Add a VEC_SERIES rtl code new 26d17444bf0 Add a VEC_DUPLICATE_EXPR tree code and associated optab new 24f5938a5b4 Add VEC_SERIES_EXPR and associated optab new 9ab0eca7db0 Treat VEC_{DUPLICATE,SERIES}_EXPR as gimple constants new 234d6d62227 [AArch64] Add an endian_lane_rtx helper routine new d8c3c0f7497 Fix for match.pd handling of three-constant bitops new c450f436a0c Fix type of bitstart in vectorizable_live_operation new 24c84b415ac Add LOOP_VINFO_MAX_VECT_FACTOR new f4066f5482b Add poly-int.h new 37f497895e3 Make mode query functions accept poly_ints new 7be8081c17f Add polynomial rtx constants new ca36f7f92bd Add polynomial tree constants new 83fe1bb8b50 Add poly_int dump routines new 20292972199 poly_int: compute_data_ref_aligment new 8070258d448 Make REG_OFFSET a poly_int64 new 265b7cb0d44 Add DWARF support for polynomial offsets new 8ec9da5287f poly_int: fold_ctor_reference new 3e2c2d16265 poly_int: same_addr_size_stores_p new 7e302542844 poly_int: ao_ref and vn_reference_op_t new 7987bc09dad poly_int: dse.c new fe60542cc03 poly_int: rtx_addr_can_trap_p_1 new ba9da4c2e26 poly_int: MEM_OFFSET and MEM_SIZE new 2c1628d84f2 poly_int: lra frame offsets new cb92e04c06b poly_int: store_bit_field bitrange new a58d2f800fc poly_int: extract_bit_field bitrange new 20efca25850 poly_int: C++ bitfield regions new 8437389ea2e poly_int: store_field & co new dbe9a01480d poly_int: SUBREG_BYTE new cf4593b2609 poly_int: operand_subword new de20cba09ff poly_int: DWARF CFA offsets new 2a3fcfab57a poly_int: ipa_parm_adjustment new c108ce200fd poly_int: get_ref_base_and_extent new b1abb145fa7 poly_int: get_addr_unit_base_and_extent new 6bb398048ec poly_int: aff_tree new 0ce73623d71 poly_int: symbolic_number new 8a70343b4e1 poly_int: pointer_may_wrap_p new 11cb22206f0 poly_int: get_inner_reference_aff new 234c565d5ba poly_int: get_inner_reference & co. new 3d8028db8c7 poly_int: reload.c new 29eb45fe30e poly_int: reload1.c new e2c537cffc1 poly_int: frame allocations new 15bb511e775 poly_int: push_block/emit_push_insn new 8c542123528 poly_int: REG_ARGS_SIZE new 00ae0b44f64 poly_int: argument sizes new 9fca576bdd2 poly_int: cfgexpand stack variables new 87c64cbc7aa poly_int: emit_inc new 85499c61e35 poly_int: reload<->ira interface new 40dbc27796f poly_int: emit_group_load/store new 3c4b9f6aeed poly_int: int_size_in_bytes new da7004e6444 poly_int: bit_field_size/offset new 2c8d1307522 poly_int: MEM_REF offsets new 9a81fa4e0ee poly_int: build_ref_for_offset new 167301dbf6f poly_int: get_binfo_at_offset new c733d2d6fa1 poly_int: tree-ssa-loop-ivopts.c:iv_use new a493d26cad3 poly_int: vectoriser vf and uf new 9e791265b87 poly_int: omp_max_vf new c9ba9673b0b poly_int: get_mask_mode new 9bf48826fc9 poly_int: current_vector_size and TARGET_AUTOVECTORIZE_VECT [...] new 824fcb9e2de poly_int: vect_no_alias_p new 88f3c042395 poly_int: GET_MODE_NUNITS new 4bd85055c37 poly_int: TYPE_VECTOR_SUBPARTS new b4267938646 poly_int: GET_MODE_PRECISION new f6f4bedfd02 poly_int: GET_MODE_BITSIZE new bae3890ae5e poly_int: GET_MODE_SIZE new 9c4477db868 Base subreg rules on REGMODE_NATURAL_SIZE rather than UNITS [...] new b5e6c63aa05 Use poly_int tree accessors instead of hwi accessors new 6540dd20bc6 Use poly_int rtx accessors instead of hwi accessors new 0fad5656c1e Make more use of simplify_gen_binary new ba0037bd728 Don't query the frontend for unsupported types new 2d93c2074d9 Improve spilling for variable-width slots new 53a3fcf568c Add support for MODE_VECTOR_BOOL new 03575106827 Add support for adjusting the number of units in a mode new 01f7bc66f82 Allow the target to set MAX_BITSIZE_MODE_ANY_MODE new 48b505c82fa Fix vectorizable_live_operation handling of vector booleans new 1492bb0ceab Use extract_bit_field_as_subreg for vectors new 9e975a5a6d4 Fix infinite loop in simplify_operand_subreg new 30aacebe296 Prevent invalid register mode changes in combine new 6ede44cbe60 Improve canonicalisation of TARGET_MEM_REFs new 0911cd7d4eb Improve vector mask pattern handling new 495271ac450 Improve ivopts handling of forced scales new 6db0d25a7f8 Fix a failure in the Fortran matmul* tests new e6fd6567469 SUBREG_PROMOTED_VAR_P handling in expand_direct_optab_fn new ffc24ba4eee [AArch64] Move code around new ef3485e6e0b [AArch64] Rework interface to add constant/offset routines new 29c9d8f2f46 [AArch64] Set NUM_POLY_INT_COEFFS to 2 new 0a27b347f49 [AArch64] Remove use of wider vector modes new 5e84c429ddc [AArch64] Rename cmp_result iterator new 498327fd12e [AArch64] Rename the internal "Upl" constraint new a3dd42cccae [AArch64] Add const_offset field to aarch64_address_info new 635d0fa1be8 [AArch64] Tweak aarch64_classify_address interface new 57f9e8068b2 [AArch64] Fix label mode new 2deb65ad226 [AArch64] Rewrite aarch64_simd_valid_immediate new 37eb639398b [AArch64] Generate permute patterns using rtx builders new 1c71de48343 [AArch64] Tighten address register subreg checks new 204c415e0c3 Add copy_rtx call to RTL loop unroller new ff67a1b2094 Use asm volatile ("" ::: "memory") new f5187922eca Add VECTOR_BITS to tree-vect.h new d30eb455de7 available_vector_sizes new 0eda3d12b5e vect_permN new d7523fa38e3 vect_element_align_preferred new 1849c19c85b target_vect_unaligned_possible new 417b8e89c37 vect_variable_length new e1671cd5ad6 vect_align_stack_vars new 3cffbeadc1a vect_masked_store new 26ac9f1af7d Revert DECL_USER_ALIGN patch new b8fd5de83d7 [AArch64] Add SVE support new 4bc25f69d4d [AArch64] Testsuite markup for SVE new dd03cc91692 Fix folding of vector mask EQ/NE expressions new fcbf7b99b53 [AArch64] SVE load/store_lanes support new 7b4572047a8 Add support for masked load/store_lanes new 71b41c785be Add support for bitwise reductions new d5ab9c7ee29 Add support for fully-predicated loops new 9114d45abe3 Make ivopts handle calls to internal functions new 355d8934f96 Allow the number of iterations to be smaller than VF new ca28eb79390 Handle peeling for alignment with masking new 3589f213d07 Add optabs for common types of permutation new c4feca5ea41 Handle more SLP constant and extern definitions for variable VF new 01ba1017671 Improve ivopts handling of offset multiples new 05a40262f14 Rework the legitimize_address_displacement hook new b00acb2bd1a Allow combine to reorder statements new 2c5d5405ffa SLP reductions with variable-length vectors new 022118c695b Predicated comparison folds new dac252557a4 Predicated arithmetic folds new adec056105a Add an empty_mask_is_expensive hook new 9d2fca2e530 Reuse results of vect_create_addr_base_for_vector_ref new e3a02277039 Allow capped vectorisation factors new 7be36d32083 Add support for BRKA and LASTB new 9b4010e1304 Add support for CLASTB new e3281a1b9b5 Add support for FADDA new bab1a5e1e5a Add support for gather loads and scatter stores new 5ac2211dbb5 Avoid pessimistic check for overlapping groups new 765c55fc8b8 Fix for big field stores new 5d92e45c05b Support for aliasing with variable strides new 4d4ff19f64f Add support for speculative loads new c193e765edc Add support for first-faulting loads new 2b78f45beb3 Make tree-ssa-strlen.c handle partial unterminated strings new 213f1294adc Allow single-element interleaving for non-power-of-2 strides new a7e11ea91b6 Use single-iteration epilogues when peeling for gaps new 58896e5b50e Replace FMA_EXPR with one internal fn per optab new 456a12b9380 Support fused multiply-adds in fully-masked reductions new 1d102ca4727 Record equivalences for spill registers new 42de6c12f1b Add early rematerialisation pass new 9459fcd9d50 Don't require an integer mode for PARALLELs new 42c09e02c88 Remove unnecessary temporary in tree-if-conv.c new 3f0e0191212 Handle vector boolean types when calculating the SLP unroll factor new 231ad6081b6 Implement SLP of internal functions new 5b4cc1776f3 Use conditional internal functions in if-conversion

The 201 revisions listed above as "new" are entirely new to this repository and will be described in separate emails. The revisions listed as "adds" were already present in the repository and have only been added to this reference.